Research Article Received: 19 July 2014

Revised: 5 December 2014

Accepted: 12 December 2014

Published online in Wiley Online Library

(wileyonlinelibrary.com) DOI 10.1002/psc.2739

A novel chimeric peptide with antimicrobial activity Begum Alaybeyoglu,a Berna Sariyar Akbulutb and Elif Ozkirimlia* Beta-lactamase-mediated bacterial drug resistance exacerbates the prognosis of infectious diseases, which are sometimes treated with co-administration of beta-lactam type antibiotics and beta-lactamase inhibitors. Antimicrobial peptides are promising broadspectrum alternatives to conventional antibiotics in this era of evolving bacterial resistance. Peptides based on the Ala46–Tyr51 beta-hairpin loop of beta-lactamase inhibitory protein (BLIP) have been previously shown to inhibit beta-lactamase. Here, our goal was to modify this peptide for improved beta-lactamase inhibition and cellular uptake. Motivated by the cell-penetrating pVEC sequence, which includes a hydrophobic stretch at its N-terminus, our approach involved the addition of LLIIL residues to the inhibitory peptide N-terminus to facilitate uptake. Activity measurements of the peptide based on the 45–53 loop of BLIP for enhanced inhibition verified that the peptide was a competitive beta-lactamase inhibitor with a Ki value of 58 μM. Incubation of beta-lactam-resistant cells with peptide decreased the number of viable cells, while it had no effect on beta-lactamase-free cells, indicating that this peptide had antimicrobial activity via beta-lactamase inhibition. To elucidate the molecular mechanism by which this peptide moves across the membrane, steered molecular dynamics simulations were carried out. We propose that addition of hydrophobic residues to the N-terminus of the peptide affords a promising strategy in the design of novel antimicrobial peptides not only against beta-lactamase but also for other intracellular targets. Copyright © 2015 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; beta-lactamase; antibiotic resistance; delivery; cell-penetrating peptide

Introduction Bacterial resistance to antibiotics is a worldwide health problem in the treatment of infectious diseases [1]. One of the most important mechanisms of resistance bacteria have acquired to avoid the lethal effects of beta-lactam antibiotics is the production of the antibioticinactivating beta-lactamases. TEM-1 class A beta-lactamase is a plasmid-encoded enzyme of 263 amino acids found in the periplasmic space of gram-negative bacteria. In order to combat betalactamase-mediated resistance, commercial combination therapies of beta-lactam antibiotics with beta-lactamase inhibitors such as sulbactam, tazobactam and clavulanic acid have been clinically used. Despite the success of these combination therapies, betalactamases have evolved inhibitor-resistant mutations [2]. Therefore, design of novel inhibitors is an area of intense research. Guiding this field is the interaction between beta-lactamase and the 165 residue long beta-lactamase inhibitory protein (BLIP) of Streptomyces clavuligerus. BLIP inhibits TEM-1 with an inhibition constant (Ki) of 0.6 nM [3]. The three dimensional structure of TEM-1 betalactamase [4,5] has two main domains with the substrate-binding site located at the interface of the two domains. The Ala46–Tyr51 loop of BLIP [3,6] blocks the beta-lactamase active site with the Asp49 residue of BLIP mimicking the beta-lactam carboxyl group. Mutation of this Asp residue was therefore found to reduce inhibition by BLIP about 40-fold [7]. By using a sliding window of ten amino acids that move through the BLIP sequence, it was found that BLIP 41 to 50 residues (N-His-Cys-Arg-Gly-His-Ala-Ala-Gly-AspTyr-COOH) comprise the highest percentage of contact residues [8]. Indeed, mutation of His41, Asp49 and Tyr53 resulted in more than a 40-fold decrease in binding affinity in a kinetic assay of inhibition of beta-lactam hydrolysis, and these residues comprised the

J. Pept. Sci. 2015

functional epitope of BLIP for binding to class A beta-lactamase [7]. Mutation of Tyr53 to alanine was found to be as critical as D49A mutation resulting in an ~40-fold decrease in binding affinity, substantially increasing the free energy of binding [7,9,10]. Motivated by these findings on the TEM-1–BLIP interface, BLIP-based peptide inhibitor design has been an ongoing research effort. Peptides based on the Ala46–Tyr51 region of BLIP inhibit TEM-1 beta-lactamase with Ki values around 500 μM [8]. Phage display studies showed that peptides such as RRHGHYY-NH2 or Pep90 (YHFLWGP) have higher affinity at 136 and 333 μM, respectively [11–14]. In addition to having high affinity, the peptide inhibitor needs to cross the cell wall and reach its intracellular target, in this case betalactamase. However, hydrophilic character of peptides prevents their cellular uptake due to the hydrophobic nature of the cell wall and membranes [15]. Several strategies have been developed in order to circumvent the limitations in peptide uptake. An alternative in the delivery of peptide inhibitors is the use of cell-penetrating peptides (CPPs), short cationic and amphipathic peptides, which can penetrate the cell membrane and carry cargos into the cell [16]. An example for this type of peptide is the 18-residue pVEC

* Correspondence to: Elif Ozkirimli, Chemical Engineering Department, Bogazici University, 34342 Istanbul, Turkey. E-mail: [email protected] a Chemical Engineering Department, Bogazici University, Bebek, 34342, Istanbul, Turkey b Bioengineering Department, Marmara University, Kadikoy, 34722, Istanbul, Turkey Abbreviations: SMD, steered molecular dynamics; BLIP, beta-lactamase inhibitor protein; CPP, cell-penetrating peptide.

Copyright © 2015 European Peptide Society and John Wiley & Sons, Ltd.

ALAYBEYOGLU ET AL. peptide (LLIILRRRIRKQAHAHSK) derived from the adhesion molecule vascular endothelial cadherin [17]. The N-terminal hydrophobic region (LLIIL) of this CPP was shown to be crucial for transport [18,19]. A similar group of membrane active peptides is called antimicrobial peptides (AMPs) because they kill bacteria, virus and fungi either by disrupting their membrane integrity or by inhibiting some cellular functions [20]. Recent studies have shown that the design of chimeric peptides opens a new way of rational engineering of novel antimicrobial peptides with membrane penetration or membrane-disruptive properties. Fusing peptides to Tat11 argininerich motif [21], to a stretch of cationic cell-penetrating nona (D-arginine) residues [22], to the antimicrobial peptide ri-melittin [23] or to pVEC [24] results in peptides with high uptake potential. In the present work, a peptide based on the 45–53 loop of BLIP was modified, and the potential of these peptides to pass the cell wall and inhibit periplasmic beta-lactamase was examined. The modification approach involved the addition of the five N-terminal residues (LLIIL) of pVEC to the N-terminus of the peptide to design a chimeric peptide that combines key fragments of an inhibitor protein and a cell-penetrating peptide. Our results showed that the addition of the LLIIL residues facilitated peptide interactions with the membrane, and the modified peptide had antimicrobial effect on antibiotic-resistant bacterial cells. We propose that this novel design of a chimeric peptide that combines properties of a cellpenetrating peptide and an inhibitory peptide is a powerful strategy for the design of new antimicrobial peptides.

Growth Media and Conditions The pre-culture was prepared from a single colony of E. coli K12 without pUC18 in 5 ml Luria–Bertani (LB) medium. Cells were grown overnight at 37 °C and 180 rpm. This pre-culture was used to inoculate 20 ml fresh LB medium at 1 : 100 dilution. Cells growing at 37 °C and 180 rpm were treated with peptides at a final concentration of 100 μM in early logarithmic growth phase (OD600nm ~ 0.2). Pre-culture and fresh growth media are supplemented with 100 μg/ml ampicillin at every step for the experiments with E. coli K12 pUC18 cells for the selectivity of the resistant cells. Equal volume of buffer was added to the control set of P1, and equal volume of 20% (v/v) DMSO–buffer mixture was added to the control set of P2 in order to eliminate any possible DMSO and/or buffer effect on the cell growth.

Measurement of Cell Growth and Viable Cell Count Cell growth was monitored by measuring optical density at 600 nm. In order to measure viable cells, bacterial samples taken at different time intervals were plated on solid LB media with serial dilution. Solid LB was supplemented with 100 μg/ml ampicillin for the experiments with E. coli K12 pUC18 cells. The plates were incubated overnight at 37 °C, and colonies formed on the plates were counted to find the number of colony-forming units (CFUs). All of the experiments were repeated three times, and the average of duplicate plate counts at each time point for three repetitions of each experiment was plotted with standard error.

Materials and Methods Peptides Used

Measurement of Beta-lactamase Activity

Peptides used in the study are given in Table 1. Peptides were ordered with given sequences from Thermo Fisher Scientific Inc. and were delivered as powders. Prior to aliquoting, peptides were dissolved in PBS buffer (50 mM, pH = 7.0) to a final concentration of 1 mM. Although P1 is soluble in PBS, P2 forms aggregates in buffer solution because of its hydrophobic character. To overcome P2’s solubility issues, 20% (v/v) DMSO was added to P2 stock solution. One hundred microliters of peptide aliquots was prepared and stored at 20 °C.

Periplasmic R-TEM-1 beta-lactamase was extracted from E. coli K12 cells by osmotic shock procedure. Activity of R-TEM-1 betalactamase was measured by monitoring the hydrolysis of the chromogenic substrate CENTA (Calbiochem, Cat. No. 219475) by betalactamase. CENTA stock was prepared in PBS buffer (50 mM, pH = 7.0) at a concentration of 4.7 mM and stored at 20 °C in 50 μl aliquots. For the determination of Km value for CENTA with R-TEM-1 beta-lactamase, five assays with different substrate concentrations (in the final concentration range of 35–120 μM) were carried out in a total volume of 1 ml in 50 mM potassium phosphate buffer (pH 7.0) with final enzyme concentration of 36.33 nM at room temperature. For inhibition studies, peptide at a final concentration of 100 μM was added to the enzyme–buffer mixture. Activity was measured upon addition of substrate CENTA, similarly at five different concentrations, to the assay mixture. One unit (U) of betalactamase activity was defined as the amount of product formed in micromoles per liter of enzyme used per minute. The hydrolysis of CENTA was monitored by continuous recording of the absorbance variation at 405 nm. The extinction coefficient of CENTA at 405 nm was previously determined as 6400 M 1 cm 1 [26]. The Km and vmax (and Ki for the inhibition) values were calculated using the initial rate measurements with application of linear least squares algorithm (R2 > 0.9) on the Hanes–Woolf plot, which is a linear transformation and a graphical representation of the Michaelis– Menten equation. The choice of Hanes–Woolf plot as the linear transformation is favorable because it can handle a wider range of initial substrate concentrations and the data are not lumped even at low concentrations. Therefore, it prevents the slope of the line fitted to the data to change at a large extent, and it provides a better estimation of vmax.

Bacterial Strains and Plasmids The experiments were performed with the Escherichia coli K12 strain. Beta-lactamase expression was achieved from pUC18 plasmid carrying the gene for R-TEM-1. There is a two-residue difference between TEM-1 beta-lactamase and R-TEM-1 beta-lactamase (I84V and V184A), but these residues are located away from the binding interface; hence, they are not expected to alter binding and inhibition properties [25]. Cells and pUC18 plasmid were obtained from our laboratory stocks. Table 1. Peptides used in this study Peptide name P1 P2

Description

Sequence

BLIP 45–53 region BLIP 45–53 region with Nterminal pVEC residues

H-HAAGDYYAY-CONH2 H-LLIILHAAGDYYAY-CONH2

wileyonlinelibrary.com/journal/jpepsci

Copyright © 2015 European Peptide Society and John Wiley & Sons, Ltd.

J. Pept. Sci. 2015

A NOVEL CHIMERIC PEPTIDE WITH ANTIMICROBIAL ACTIVITY Steered Molecular Dynamics Simulations Details of the SMD simulations and analysis methods can be found in our previous report on pVEC [19]. The initial coordinates of residues HAAGDYYAY were based on the reference structure (45–53 residues of BLIP, PDB ID: 1jtg), which forms a beta-hairpin, or they were assigned as an extended chain (ϕ = 180° and ψ = 180°, or random coil) using visual molecular dynamics (VMD) [27]. The coordinates for the LLIIL residues in P2 were assigned as an extended chain. The time step was 2 fs, and the coordinates were stored every 0.5 ps to create the simulation trajectories. Peptides placed into a water box (10 Å × 10 Å × 10 Å) were energy minimized for 1000 steps and equilibrated for 2 ns with constraints on peptide atoms and for 2 ns without any constraints. The 128-lipid membrane was selected as 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE), which has been used to mimic the inner leaflet of the outer membrane of E. coli [28]. The membrane was solvated with TIP3P water molecules 60 Å in positive and negative z directions and then equilibrated for 0.5 ns with constraints and for 0.5 ns without any constrains. The previously equilibrated peptide was placed into the water layer above the membrane manually to a minimum distance of 10 Å. SMD simulations [29] were performed by applying force on the N-terminal Cα atom (SMD atom) of the peptide in the negative z direction. Constraints were applied on the phosphorous heads of the lipid bilayer only in the z direction and left free to move in the x and y directions to allow the formation of a pore with minimal membrane disruption. The spring constant was 10 kcal/mol/Å2, and the velocity of pulling was 2.50 Å/ns. Simulations were performed at constant temperature and pressure, 300 K and 1 atm, using NAMD [29] molecular dynamics package with CHARMM27 [30] force-field parameters. All the simulations (for each peptide and each initial structure) were repeated three times (for 40 ns each), and the results were averaged for analysis. The interaction energy (including van der Waals and electrostatics energy terms) between the peptide and lipids was calculated using the NAMD Energy plugin of the VMD software. Lipid tail order parameters, which provide information about the order and configuration of the fatty acyl chains, were computed with respect to the membrane normal axis using order parameter = ½ , where θ is the angle between a vector along the acyl chain methylene/methyl hydrogens and the membrane normal axis. Angular brackets indicate averaging over all lipids and all time frames. For examining the time-dependent change in order parameters, the averaging is performed over all lipids [31].

Results and Discussion The strong interaction of the 46–51 loop of BLIP with betalactamase has guided the design and discovery of BLIP-based beta-lactamase inhibitor peptides [13]. Here, we examine the inhibition and uptake potential of the peptide P1 with the sequence corresponding to residues 45 to 53 of BLIP (HAAGDYYAY), to be able to include His45, Asp49 and Tyr53 that were shown to contribute to BLIP binding [7]. P1 was then modified to include LLIIL residues at its N-terminus to obtain P2 (LLIILHAAGDYYAY) for enhanced cellular uptake. In-Vitro Beta-lactamase Inhibition TEM-1 beta-lactamase Km and kcat values were previously determined to be 70 μM and 110 s 1, respectively, for a final enzyme concentration of 11.2 nM with the initial rate measurements method

J. Pept. Sci. 2015

Table 2. Kinetic parameters for beta-lactamase inhibition with P1

[Enz] (nM) [CENTA] (μM) [P1] (μM) vmax (μM/min) Km (μM) 1 kcat (s ) Ki (μM)

CENTA

P1

36.33 35–120 – 5125 55.02 2351 –

36.33 35–120 100 5231 93.88 929 58.6

[26]. The Ki value for the peptide based on the 46–51 loop of BLIP was 500 μM [8]. Here, our aim was to enhance the previously identified peptide by including His45 and Tyr53 residues that were shown to be important for beta-lactamase–BLIP interaction. Beta-lactamase activity and inhibition by P1 were measured using the chromogenic substrate CENTA. Table 2 summarizes the kinetic constants obtained with CENTA and R-TEM-1 beta-lactamase in the presence and absence of P1. The kinetic constants obtained for CENTA hydrolysis with R-TEM-1 beta-lactamase are consistent with the literature values. The Km and vmax were calculated as 55.02 μM and 5125 μM/min, respectively, for a final enzyme concentration of 36.33 nM. Upon addition of P1, vmax did not change (5231 μM/min), while Km value increased to 93.88 μM, showing that P1 is a competitive inhibitor of R-TEM-1 beta-lactamase with a Ki value of 58.6 μM. In comparison, the Ki for the peptide based on the 46–51 loop of BLIP was 500 μM [8], suggesting that extension of this previously identified peptide to include His45 and Tyr53 results in tighter binding and improved inhibition, as was the aim of this step. In-Vivo Beta-lactamase Inhibition with Peptides E. coli K12 cells harboring the pUC18 plasmid synthesize betalactamase that confers resistance against beta-lactam antibiotics. These antibiotic-resistant E. coli K12 pUC18 cells were incubated with the designed peptides, and growth of the cells was monitored by spectrophotometry as well as by counting the number of viable cells. Based on the in-vitro kinetics measured, these peptides were expected to inhibit beta-lactamase, resulting in cells susceptible to beta-lactam antibiotics. Spectrophotometric analysis of cell growth showed that incubation of P1 or P2 had no effect on the growth of cells (Figure 1a and b). P2, which is highly hydrophobic, was dissolved in buffer by addition of 20% (v/v) DMSO into the P2 stock solution. Also, buffer and 20% (v/v) DMSO–buffer mixture was added to the control sets of P1 and P2, respectively, to eliminate any possible effects of chemicals on cell growth. The growth curves of control sets of both peptides are similar (Figure 1a and b), showing that the effect of DMSO and/or buffer is negligible. Incubation with P1 had no effect on the number of viable cells (Figure 1a), either suggesting that P1 had no antimicrobial activity or it was unable to disrupt or penetrate the cell wall to reach and inhibit beta-lactamase. Interestingly, however, even though spectrophotometric growth profiles showed no significant effect of P2 on cell density, the number of viable cells decreased by 30% upon incubation with P2. This result may suggest that P2 behaves as a beta-lactamase inhibitor, but it is also possible that it is acting as an antimicrobial peptide with a cell wall/membrane-disrupting capability. In order to ascertain that retardation in cell growth is due to uptake of the peptide and inhibition of intracellular beta-lactamase, wild-type E. coli K12 cells with

Copyright © 2015 European Peptide Society and John Wiley & Sons, Ltd.

wileyonlinelibrary.com/journal/jpepsci

ALAYBEYOGLU ET AL.

a.

2.5

3

x 10

9

pUC18 + P1

2

CFU

OD600

2 1.5 1

1

0.5 0

0 0

1

2

3

4

0

5

1

2

Time (h)

4

5

2.5

3

x 10

4

5

4

5

9

pUC18 + P2

b.

3

Time (h)

2

1.5

CFU

OD600

2

1

1

0.5 0

0 0

1

2

3

4

5

0

1

2

Time (h) 5

2

4

1.5

3

1 0.5 0

x 10

9

wild-type + P2

2.5

CFU

OD600

c.

3

Time (h)

2 1

0

1

2

3

4

5

Time (h)

0

0

1

2

3

Time (h)

Figure 1. Growth profiles (left) and colony-forming units (right) in the presence of 100 μM peptide (o) or in the absence of peptide (*) for (a) E. coli K12 pUC18 cells treated with P1, (b) E. coli K12 pUC18 cells treated with P2 and (c) E. coli K12 cells treated with P2. 0 h indicates the time of peptide addition. Average of triplicate measurements is plotted with standard error.

no beta-lactamase production were also treated with P2. If disruption of the cell wall/membrane is the reason for cell death, then these cells would die upon treatment with P2. On the other hand, if the antimicrobial action of P2 is due to inhibition of betalactamase, then these wild-type cells would be unaffected. Consistent with the second hypothesis, P2 did not have any effect on wildtype E. coli K12 cells (Figure 1c). These results showed that P2 acts as a beta-lactamase inhibitor, while P1 has no effect on bacterial cells, suggesting that addition of the N-terminus hydrophobic tail facilitates delivery into the cell to access an intracellular enzyme. Translocation of the Peptides through the Membrane The molecular mechanism by which the peptides move through the membrane and the contribution of the hydrophobic Nterminus, if there is any, to facilitating transport were examined by SMD simulations on P1 and P2 through a POPE lipid bilayer, which mimics the bacterial membrane. In order to investigate the effect of initial peptide conformation on the translocation mechanism, the coordinates for the peptides P1 and P2 were either based on the structure of 45–53 residues of the reference BLIP structure or assigned as an extended chain. For P1, the beta-hairpin (formed by H bonding between residues Gly4, Asp5 and Tyr6) is maintained during equilibration in water. On the other hand, when the initial structure is an extended chain, the peptide also tends to become more compact in water by bringing the N- and C-terminals toward each other, and the peptide structure that makes the initial contact with the phosphate heads assumes a beta-hairpin formed by residues Gly4, Asp5 and Tyr6,

wileyonlinelibrary.com/journal/jpepsci

resembling the structure observed with the beta-hairpin structure of P1. Similarly, for P2, the peptide assumes a more compact form whether the starting structure is an extended chain or a betahairpin. The hairpin formed by Gly9, Asp10 and Tyr11 residues of P2 in the initial beta-hairpin structure is maintained after equilibration of the peptide with the beta-hairpin initial form, and the same hairpin is formed after equilibration of the extended chain structure. Interestingly, the peptide conformations in the lipid bilayer during the SMD simulations are very similar for P1 and P2. These observations of the calculated SMD trajectories suggest that the initial structure assignment does not change the translocation pathway and interactions with the membrane. For all cases, the peptide moves through the bilayer as a single chain, penetrating the bilayer like a needle. Peptide–lipid interactions The modification of P1 (addition of LLIIL) increases the hydrophobicity of the peptide; hence, P2 is expected to have increased interactions with the membrane lipids. Even though the averaged interaction energy profiles for both peptides are similar in trend and magnitude and do not portray any discernible stages in the transport process, the electrostatic interaction energy profiles suggest that the transport occurs in four stages at electrostatic energy values of 40, 100, 40 and 60 kcal/mol. For P1, electrostatic energy value is 40 kcal/mol only between z values of 40 to 30 Å, but for P2, the electrostatic energy value is 40 kcal/mol between z values of 40 to 10 Å (Figure 2A). This stage (i) corresponds to the initial interactions of the peptides with the membrane, and P2

Copyright © 2015 European Peptide Society and John Wiley & Sons, Ltd.

J. Pept. Sci. 2015

A NOVEL CHIMERIC PEPTIDE WITH ANTIMICROBIAL ACTIVITY

Figure 2. (A) Averaged interaction energy, electrostatic energy and van der Waals energy profiles between the lipid bilayer and P1 (left) or P2 (right) with respect to the SMD atom position. z coordinate is 0 when the SMD atom is at the middle of the membrane. The curves are the averages of three independent simulations for each peptide. (B) Snapshots of simulations trajectories taken at corresponding points for P1 (top) and P2 (bottom).

has increased interactions due to its hydrophobic N-terminus. The electrostatic energy increases to a value of 60 kcal/mol, and it is conserved until the SMD atom reaches the membrane core. At this stage, P2 is observed to be in contact with the membrane closely and reaches a better folded state compared with its initial structure. This behavior of P2 resembles that of antimicrobial peptides, which are usually found to adapt horizontal orientation during their initial contact with the membranes. In fact, small cationic peptides that target negatively charged bacterial membranes are found to be bound to the membrane surface horizontally until a peptide concentration threshold is reached, after which they form a permeation pathway [32]. The initial interactions of P1, on the other hand, increase almost linearly as the peptide penetrates into the bilayer with no plateau in interaction energy (Figure 2A). The maximum values of electrostatic energies reached by P1 and P2 are similar in magnitude, but the driving forces of the interactions are different. Electrostatic energy of P1 reaches its maximum values at stage (ii) as the peptide N-terminus penetrates and starts embedding

J. Pept. Sci. 2015

itself in the lipid bilayer (Figure 2A and B). On the other hand, the electrostatic energy of P2 reaches its maximum when the hydrophobic N-terminus reaches the bilayer core, and this value is conserved until the N-terminus exits the bilayer. Indeed, when the electrostatic interaction energy values are calculated on a residue basis, the strongest interactions occur between Asp, C-terminus Tyr (Asp5 and Tyr9 for P1, Asp10 and Tyr14 for P2) residues and phosphate groups for both P1 and P2. His1 residue of P1 is one of the major contributors to electrostatic interactions, whereas in the case of P2, His6 residue is no longer of importance and Leu1 has become a major electrostatic interaction contributor. Stage (iii) corresponds to the conformations of both peptides fully embedded in the membrane (Figure 2A and B) where the van der Waals and interaction energy between the peptides and the membrane reach maximum values. For both peptides, the electrostatic interaction energy decreases upon exit of N-terminus from the bilayer at stage (iv) (Figure 2A). The final increase in the electrostatic energy at stage (v) is observed as the C-terminus (YYAY) of both peptides leave the

Copyright © 2015 European Peptide Society and John Wiley & Sons, Ltd.

wileyonlinelibrary.com/journal/jpepsci

ALAYBEYOGLU ET AL. membrane (Figure 2A and B). The van der Waals energy between the lipids and P2 reaches higher values than that with P1 when the peptide is fully embedded in the membrane (between z values of 10 to 30 Å) (Figure 2A). Membrane properties The order parameters of lipid tail carbon atoms measured using deuterium solid-state NMR of deuterium-labeled lipids provide information on the physical properties of the membrane. An order parameter of 1 indicates full order and that chains reside along the bilayer normal, whereas an order parameter of 0.5 indicates that chains are perpendicular to the normal [31]. A larger order parameter for a carbon atom correlates with lower fluctuations throughout the simulation and hence more order. Solid-state NMR data have shown that cationic amphipathic peptides that align parallel to the surface of phospholipid membranes disorder the fatty acyl chains strongly and the disordering caused by these peptides are inversely proportional to their penetration depth. On the other hand, transmembrane proteins and hydrophobic peptides have modest disordering effect on membrane because they either are embedded into the membrane or reside close to the bilayer core, aligned parallel to the lipid tails [33]. Here, the order of the lipid tails in the presence of the peptides was compared with that in a 20 ns membrane–water MD simulation to determine the effect of peptide penetration (Figure 3). The calculated order parameters are in good agreement with experimentally measured C–D bond order parameters of sn-2 acyl chain at 30 °C for POPE membrane [34]. As expected, the order parameter for the membrane was higher in the absence of the peptides than in the presence of the peptides. The decrease in order parameter in the Palmitoyl tail

Order Parameter

0.3

0.2

0.1

0

2

P1 P2 Membrane 4

6

8

10

12

14

16

Carbon Number

presence of the peptides is most significant at the bilayer surface, while it can be considered negligible at the bilayer core. A decrease in order parameters and hence a peptide-induced disorder have been observed for arginine-rich peptides [35] and for NK-2 [36] using simulations and PGLa peptide using NMR [37]. This decrease in order was ascribed to the membrane thinning effect induced by binding of these antimicrobial peptides parallel to the membrane surface similar to the effect produced by binding of detergents [38]. It is known that the order parameters provide information on the area per lipid for a bilayer [39], and an increase in the area can be related to a decrease in membrane thickness. On the other hand, for transmembrane proteins and peptides that insert into the lipid region with a perpendicular orientation to the membrane surface, the effect on order parameters is more modest or negligible [33]. Throughout the SMD simulations, harmonic constraints are applied on the phosphate heads of lipid bilayers in the z direction, and as a result, the thickness of the bilayer stays almost constant. If the phosphate heads are not constrained during SMD simulations, the whole membrane will start to move with the peptide because of the force applied on the SMD atom and the translocation across the membrane cannot be observed. As the peptide moves through the bilayer, a pore forms (because the phosphate heads are free to move in the x–y direction) in the membrane plane. Therefore, the increase in disorder in the presence of peptide is not due to membrane thinning, which cannot be observed with the current simulations, but it is due to the disorder induced by transport of the peptides across the bilayer and the interactions of the peptide with the lipid acyl chains. On average, P2 creates slightly higher disorder especially near the lipid head groups (Figure 3), possibly suggesting that the higher hydrophobicity of P2 results in increased interactions with the membrane lipids during initial penetration causing fluctuations of the lipid tails that are closer to the lipid–water interface. Indeed, while the change in order parameters for the C8 or C16 carbons of the palmitoyl tail as a function of time is negligible, the order parameter of C2 in the palmitoyl tail fluctuates significantly during the SMD simulations (Supplementary Figure 1). Such an increase in the disorder of carbons near the membrane surface suggests that the strong interaction with the membrane surface may induce pore formation. The results are in agreement with interaction energy analysis results, which show that the initial penetration of the phosphate head layer is an important energy barrier in the uptake. Our lab is currently investigating the effect of increasing peptide concentration on the transport properties of the peptides.

Oleyl tail

Conclusion

Order Parameter

0.3 0.25 0.2 0.15 0.1

P1 P2 Membrane

0.05 0

2

4

6

8

10

12

14

16

18

Carbon Number Figure 3. Order parameters of the 128-lipid POPE bilayer palmitoyl (top) and oleyl (bottom) carbon atoms in the presence of P1 (triangles) or P2 (circles) and in the absence of the peptides (squares). Carbon number increases when it is away from the lipid head group and closer to the bilayer core.

wileyonlinelibrary.com/journal/jpepsci

The production of the beta-lactamase enzymes that inactivate beta-lactam antibiotics is the leading cause of antibiotic resistance, which has created a worldwide health problem in the treatment of infectious diseases over the past years. Although some combination therapies have been successful, beta-lactamase enzymes are able to mutate and evolve such that they are no longer sensitive to the inhibitors. Therefore, design of novel inhibitors is necessary to overcome the resistance to known therapies. Beta-lactamase inhibitory protein structures in complex with various types of beta-lactamases have provided important clues in the search for novel beta-lactamase inhibitors. Peptides based on one of the key regions of BLIP, Ala46–Tyr51 beta-hairpin loop, have shown micromolar inhibition of beta-lactamase activity [11–14,35,40]. In light of the foregoing work, our aim was to

Copyright © 2015 European Peptide Society and John Wiley & Sons, Ltd.

J. Pept. Sci. 2015

A NOVEL CHIMERIC PEPTIDE WITH ANTIMICROBIAL ACTIVITY design a peptide-based beta-lactamase inhibitor that not only binds and inhibits the protein, but that can also enter the bacterial cells to access intracellular beta-lactamase. To this aim, the peptides HAAGDYYAY (P1), and LLIILHAAGDYYAY (P2) based on the 45–53 loop region of BLIP, were tested in vivo on E. coli K12 cells for cellular uptake, and P1 was tested in vitro for beta-lactamase inhibition. Enzyme inhibition assays showed that P1 is a better beta-lactamase inhibitor compared with BLIP 46–51 loop, because of extension of the peptide to include Tyr53. However, P1 had no effect on antibiotic-resistant cells, whereas addition of P2 led to a 30% (less than one log reduction) decrease in the number of CFUs. While it is possible that P2 reduces the number or culturable cells without killing them [41], we suggest that P2 acts on the cells by inhibiting beta-lactamase and not by disrupting the cell membrane or wall because P2 treatment does not affect viability of wild-type cells. When SMD simulations were applied to transport the peptides from one side of the membrane to the other, the effect of LLIIL tail was apparent in the interactions of P2 with the lipid bilayer and electrostatic interactions during the initial penetration of the peptide are altered. The translocation of peptides though the membrane caused modest disordering effects on the lipid bilayer like many other antimicrobial peptides. The disordering effect of P2 was slightly higher than P1 as expected because of its increased interactions with the membrane during penetration. We introduce a novel antimicrobial peptide that targets an intracellular enzyme. This novel inhibitor peptide is conjugated such that it incorporates characteristics of a cell-penetrating peptide. The results reported in this work on a chimeric peptide promise potential toward the design of novel antimicrobial peptides as well as peptides that target other intracellular proteins. Acknowledgements This work was supported by TUBITAK Research Grant 109M229, Bogazici University Research Grant 09HA504P, DPT (State Planning Organization) Project 2009K120520 (EO) and Marmara University Research Foundation (FEN-C-YLP-181208-0290 and FEN-C-YLP070211-0010) (BSA). The authors thank Deniz Meneksedag-Erol for her help in the initial stages of manuscript preparation.

References 1 Spellberg B. Antibiotic resistance and antibiotic development. Lancet Infect. Dis. 2008; 8(4): 211–2. 2 Williams JD. Beta-lactamases and beta-lactamase inhibitors. Int. J. Antimicrob. Agents 1999; 12 Suppl 1:S3–7; discussion S26–7. 3 Strynadka NC, Jensen SE, Johns K, Blanchard H, Page M, Matagne A, Frere JM, James MN. Structural and kinetic characterization of a betalactamase-inhibitor protein. Nature 1994; 368(6472): 657–60. 4 Jelsch C, Mourey L, Masson JM, Samama JP. Crystal structure of Escherichia coli TEM1 beta-lactamase at 1.8 A resolution. Proteins 1993; 16(4): 364–83. 5 Strynadka NC, Adachi H, Jensen SE, Johns K, Sielecki A, Betzel C, Sutoh K, James MN. Molecular structure of the acyl-enzyme intermediate in betalactam hydrolysis at 1.7 A resolution. Nature 1992; 359(6397): 700–5. 6 Doran JL, Leskiw BK, Aippersbach S, Jensen SE. Isolation and characterization of a beta-lactamase-inhibitory protein from streptomyces clavuligerus and cloning and analysis of the corresponding gene. J. Bacteriol. 1990; 172(9): 4909–18. 7 Zhang Z, Palzkill T. Determinants of binding affinity and specificity for the interaction of TEM-1 and SME-1 betalactamase with beta-lactamase inhibitory protein. J. Biol. Chem. 2003; 278(46): 45706–12.

J. Pept. Sci. 2015

8 Rudgers GW, Huang W, Palzkill T. Binding properties of a peptide derived from beta-lactamase inhibitory protein. Antimicrob. Agents Chemother. 2001; 45(12): 3279–3286. 9 Reichmann D, Cohen M, Abramovich R, Dym O, Lim D, Strynadka NC, Schreiber G. Binding hot spots in the TEM1–BLIP interface in light of its modular architecture. J. Mol. Biol. 2007; 365(3): 663–79. 10 Yuan J, Huang W, Chow DC, Palzkill T. Fine mapping of the sequence requirements for binding of beta-lactamase inhibitory protein (BLIP) to TEM-1 beta-lactamase using a genetic screen for BLIP function. J. Mol. Biol. 2009; 389(2): 401–412. 11 Yribarren AS, Thomas D, Friboulet A, Avalle B. Selection of peptides inhibiting a beta-lactamase-like activity. Eur. J. Biochem. 2003; 270(13): 2789–95. 12 Phichith D, Bun S, Padiolleau-Lefevre S, Guellier A, Banh S, Galleni M, Frere JM, Thomas D, Friboulet A, Avalle B. Novel peptide inhibiting both TEM-1 beta-lactamase and penicillin-binding proteins. FEBS J. 2010; 277(23): 4965–72. 13 Huang W, Beharry Z, Zhang Z, Palzkill T. A broad-spectrum peptide inhibitor of beta-lactamase identified using phage display and peptide arrays. Protein Eng. 2003; 16(11): 853–860. 14 Huang W, Zhang Z, Palzkill T. Design of potent beta-lactamase inhibitors by phage display of beta-lactamase inhibitory protein. J. Biol. Chem. 2000; 275(20): 14964–8. 15 Henriques ST, Melo MN, Castanho MA. Cell-penetrating peptides and antimicrobial peptides: how different are they? Biochem. J. 2006; 399(1): 1–7. 16 Heitz F, Morris MC, Divita G. Twenty years of cell-penetrating peptides: from molecular mechanism to therapeutics. Br. J. Pharmacol. 2009; 157: 195–206. 17 Elmquist A, Lindgren M, Bartfai T, Langel U. VE-cadherin-derived cellpenetrating peptide, pVEC, with carrier functions. Exp. Cell Res. 2001; 269(2): 237–244. 18 Elmquist A, Hansen M, Langel U. Structure–activity relationship study of the cell-penetrating peptide pVEC. Biochim. Biophys. Acta, Biomembr. 2006; 1758(6): 721–729. 19 Akdag IO, Ozkirimli E. The uptake mechanism of the cell-penetrating pVEC peptide. J. Chem. 2013; 2013. 20 Brogden KA. Antimicrobial peptides: pore formers or metabolic inhibitors in bacteria? Nat. Rev. Microbiol. 2005; 3(3): 238–50. 21 Salomone F, Cardarelli F, Di Luca M, Boccardi C, Nifosi R, Bardi G, Di Bari L, Serresi M, Beltram F. A novel chimeric cell-penetrating peptide with membrane-disruptive properties for efficient endosomal escape. J. Control. Release 2012; 163(3): 293–303. 22 Cantini L, Attaway CC, Butler B, Andino LM, Sokolosky ML, Jakymiw A. Fusogenic-oligoarginine peptide-mediated delivery of sIRNAs targeting the CIP2A oncogene into oral cancer cells. PLoS One 2013; 8(9): e73348. 23 Quebatte G, Kitas E, Seelig J. Ridom, a cell-penetrating peptide. Interaction with DNA and heparan sulfate. J. Phys. Chem. B 2013; 117: 10807–817. 24 Myrberg H, Zhang L, Mae M, Langel U. Design of a tumor-homing cell-penetrating peptide. Bioconjug. Chem. 2008; 19(1): 70–5. 25 Lim D, Park HU, De Castro L, Kang SG, Lee HS, Jensen S, Lee KJ, Strynadka NC. Crystal structure and kinetic analysis of beta-lactamase inhibitor protein-II in complex with TEM-1 beta-lactamase. Nat. Struct. Biol. 2001; 8(10): 848–852. 26 Bebrone C, Moali C, Mahy F, Rival S, Docquier JD, Rossolini GM, Fastrez J, Pratt RF, Frere JM, Galleni M. CENTA as a chromogenic substrate for studying beta-lactamases. Antimicrob. Agents Chemother. 2001; 45(6): 1868–71. 27 Humphrey W, Dalke A, Schulten K. VMD: visual molecular dynamics. J. Mol. Graph. 1996; 14(1): 33–38. 28 Murzyn K, Rog T, Pasenkiewicz-Gierula M. Phosphatidylethanolaminephosphatidylglycerol bilayer as a model of the inner bacterial membrane. Biophys. J. 2005; 88(2): 1091–103. 29 Phillips JC, Braun R, Wang W, Gumbart J, Tajkhorshid E, Villa E, Chipot C, Skeel RD, Kale L, Schulten K. Scalable molecular dynamics with NAMD. J. Comput. Chem. 2005; 26(16): 1781–1802. 30 MacKerell AD, Bashford D, Bellott M, Dunbrack RL, Evanseck JD, Field MJ, Fischer S, Gao J, Guo H, Ha S, Joseph-McCarthy D, Kuchnir L, Kuczera K, Lau FT, Mattos C, Michnick S, Ngo T, Nguyen DT, Prodhom B, Reiher WE, Roux B, Schlenkrich M, Smith JC, Stote R, Straub J, Watanabe M, Wiorkiewicz-Kuczera J, Yin D, Karplus M. All-atom empirical potential for molecular modeling and dynamics studies of proteins. J. Phys. Chem. B 1998; 102(18): 3586–616.

Copyright © 2015 European Peptide Society and John Wiley & Sons, Ltd.

wileyonlinelibrary.com/journal/jpepsci

ALAYBEYOGLU ET AL. 31 Egberts E, Marrink SJ, Berendsen HJ. Molecular dynamics simulation of a phospholipid membrane. Eur. Biophys. J. 1994; 22(6): 423–36. 32 Sengupta D, Leontiadou H, Mark AE, Marrink SJ. Toroidal pores formed by antimicrobial peptides show significant disorder. Biochim. Biophys. Acta, Biomembr. 2008; 1778(10): 2308–17. 33 Salnikov ES, Mason AJ, Bechinger B. Membrane order perturbation in the presence of antimicrobial peptides by (2)H solid-state NMR spectroscopy. Biochimie 2009; 91(6): 734–43. 34 Perly B, Smith IC, Jarrell HC. Acyl chain dynamics of phosphatidylethanolamines containing oleic acid and dihydrosterculic acid: 2H NMR relaxation studies. Biochemistry 1985; 24(17): 4659–65. 35 Sun D, Forsman J, Lund M, Woodward CE. Effect of arginine-rich cell penetrating peptides on membrane pore formation and life-times: a molecular simulation study. Phys. Chem. Chem. Phys. 2014; 16(38): 20785–95. 36 Pimthon J, Willumeit R, Lendlein A, Hofmann D. Membrane association and selectivity of the antimicrobial peptide NK-2: a molecular dynamics simulation study. J. Pept. Sci. 2009; 15(10): 654–67. 37 Bechinger B, Salnikov ES. The membrane interactions of antimicrobial peptides revealed by solid-state NMR spectroscopy. Chem. Phys. Lipids 2012; 165(3): 282–301.

wileyonlinelibrary.com/journal/jpepsci

38 Bechinger B. Insights into the mechanisms of action of host defence peptides from biophysical and structural investigations. J. Pept. Sci. 2011; 17(5): 306–14. 39 Nagle JF. Area/lipid of bilayers from NMR. Biophys. J. 1993; 64(5): 1476–81. 40 Xie L, Xu M, Yang T, Zhu C, Zhu B, Hu Y. Studies on amino acid replacement and inhibitory activity of a beta-lactamase inhibitory peptide. Biochemistry (Mosc.) 2010; 75(3): 336–41. 41 Kell DB, Kaprelyants AS, Weichart DH, Harwood CR, Barer MR. Viability and activity in readily culturable bacteria: a review and discussion of the practical issues. Antonie Van Leeuwenhoek 1998; 73(2): 169–87.

Supporting Information Additional supporting information may be found in the online version of this article at the publisher’s web site.

Copyright © 2015 European Peptide Society and John Wiley & Sons, Ltd.

J. Pept. Sci. 2015

A novel chimeric peptide with antimicrobial activity.

Beta-lactamase-mediated bacterial drug resistance exacerbates the prognosis of infectious diseases, which are sometimes treated with co-administration...
381KB Sizes 3 Downloads 12 Views