Beta-defensin derived cationic antimicrobial peptides with potent killing activity against gram negative and gram positive bacteria

Beta-defensin derived cationic antimicrobial peptides with potent killing activity against gram... Background: Avian β-defensins (AvBD) are cationic antimicrobial peptides (CAMP) with broad-spectrum antimicrobial activity, chemotactic property, and low host cytotoxicity. However, their bactericidal activity is greatly compromised under physiological salt concentrations which limits the use of these peptides as therapeutic agents. The length and the complex structure involving three conserved disulfide bridges are additional drawbacks associated with high production cost. In the present study, short linear CAMPs (11 to 25 a.a. residues) were developed based on the key functional components of AvBDs with additional modifications. Their biological functions were characterized. Results: CAMP-t1 contained the CCR2 binding domain (N-terminal loop and adjacent α-helix) of AvBD-12 whereas CAMP-t2 comprised the key a.a. residues responsible for the concentrated positive surface charge and hydrophobicity of AvBD-6. Both CAMP-t1 and CAMP-t2 demonstrated strong antimicrobial activity against Pseudomonas aeruginosa, Staphylococcus aureus and Staphylococcus pseudintermedius. However, CAMP-t1 failed to show chemotactic activity and CAMP-t2, although superior in killing Staphylococcus spp., remained sensitive to salts. Using an integrated design approach, CAMP-t2 was further modified to yield CAMP-A and CAMP-B which possessed the following characteristics: α-helical structure with positively and negatively charged residues aligned on the opposite side of the helix, lack of protease cutting sites, C-terminal poly-Trp tail, N-terminal acetylation, and C-terminal amidation. Both CAMP-A and CAMP-B demonstrated strong antimicrobial activity against multidrug- resistant P. aeruginosa and methicillin-resistant S. pseudintermedius (MRSP) strains. These peptides were resistant to major proteases and fully active at physiological concentrations of NaCl and CaCl . The peptides were minimally cytotoxic to avian and murine cells and their therapeutic index was moderate (≥ 4.5). Conclusions: An integrated design approach can be used to develop short and potent antimicrobial peptides, such as CAMP-A and CAMP-B. The advantageous characteristics, including structural simplicity, resistance to salts and proteases, potent antimicrobial activity, rapid membrane attacking mode, and moderate therapeutic index, suggest that CAMP-A and CAMP-B are excellent candidates for development as therapeutic agents against multidrug-resistant P. aeruginosa and methicillin-resistant staphylococci. Keywords: Cationic antimicrobial peptides, Peptide design, Salt resistance, Antimicrobial activity, Multidrug-resistant Pseudomonas aeruginosa, Methicillin-resistant Staphylococcus pseudintermedius * Correspondence: zhangshup@missouri.edu Department of Veterinary Pathobiology, College of Veterinary Medicine, University of Missouri, Columbia, MO 65211, USA Veterinary Medical Diagnostic Laboratory, College of Veterinary Medicine, University of Missouri, Columbia, MO 65211, USA Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Yang et al. BMC Microbiology (2018) 18:54 Page 2 of 14 Background the wild-type AvBD-12 [13]. However, the linear AvBDs The rapid emergence and spread of antimicrobial resist- designed in our previous study are still sensitive to ance, particularly those associated with Pseudomonas physiological salt conditions and the length of the pep- aeruginosa and Staphylococcus spp., have become a ser- tides (45 amino acid residues) remain to be shortened to ious threat to public health [1, 2]. The Centers for Dis- control the manufacturing cost. In addition, a previous ease Control and Prevention (CDC) estimated that each study has indicated that linear peptides are more suscep- year in the United States there are approximately 88,000 tible to protease degradation due to lack of complex ter- cases and 11,000 deaths due to infections with tiary structure stabilized by disulfide bridges found in methicillin-resistant Staphylococcus aureus (MRSA) [3]. natural defensin peptides [16]. Various studies have been conducted to search for new To increase salt- and protease-resistance, several solu- classes of antimicrobial therapeutic agents or antibiotic tions have been proposed, including: incorporating non- alternatives with novel targets and modes of action [4]. proteinogenic amino acids (e.g. D-amino acid Host cationic antimicrobial peptides (CAMPs), including substitutions and bulky amino acid β-naphthylalanin) linear peptides, α-helical peptides, circular and complex [17, 18] or LPS binding peptide motif (β-boomerang structures with loops and β-sheets constitute the first motif GWKRKRFG) [19], modifying the terminal regions line of innate defense against microbial pathogens [5]. via covalent linkage of a hydrophobic moiety (e. g. a The features shared by these CAMPs are net positive sterol or a fatty acid) [20, 21], peptidomimetic [22], alter- charge and amphipathicity [6]. The cationic property of ing the structure, charge, hydrophobicity, and shortening CAMP allows for the initial interaction of the peptide the length of the peptide [23, 24]. These strategies suc- with the anionic surface groups of the microbial mem- cessfully improved the antimicrobial function of CAMPs, brane and the hydrophobicity enables the peptide to in- but resulted in elevated hemolytic activity and increased tegrate into the hydrophobic core of the membrane. The manufacturing cost [18, 25]. In the present study, an inte- mechanism of action of CAMPs is complex, achieved grated approach was utilized to design short and compos- primarily through membrane damage and possibly sub- itionally simple CAMPs with potent antimicrobial activity, sequent interactions with cellular machineries, and the improved resistance to salts and proteases and minimal potential for development of microbial resistance is low cytotoxicity to host cells. The antibacterial property of the [6]. A major group of CAMPs with broad-spectrum anti- newly designed CAMPs against P. aeruginosa and Staphylo- microbial activity is β-defensins which contain three coccus spp., including clinical isolates of multidrug-resistant cysteine-cysteine disulfide bridges [7]. In addition to P. aeruginosa and methicillin-resistant S. pseudintermedius their antimicrobial activity and low potential for the de- (MRSP) was assessed under various conditions. velopment of resistance by bacteria, β-defensins have several other beneficial characteristics, such as modulat- Methods ing host immune response (e.g. chemo-attracting im- Bacterial strains and cultures mune cells) [8–11]. Our previous studies show that Pseudomonas aeruginosa (P. aeruginosa, ATCC 27853) avian β-defensins (AvBDs) such as AvBD-6 and and Staphylococcus aureus (S. aureus, ATCC 29213) AvBD-12 possess the following biological properties: were used to evaluate the novel CAMPs’ antimicrobial broad-spectrum antimicrobial activity, LPS-neutralizing activity, salt- and protease-resistance, and membrane per- ability, chemotactic activity, and minimal cell cytotox- meability. Ten multiple-drug resistant P. aeruginosa and icity [12–14]. Although β-defensins represent potentially ten methicillin-resistant S. pseudintermedius (MRSP) clin- a novel class of antimicrobial therapeutic agents, several ical isolates (Table 2) were used to evaluate the antimicro- obstacles must be overcome, including host cell cytotox- bial efficacy of CAMPs. The clinical isolates were cultured icity, degradation by proteases, loss of antimicrobial ac- from diagnostic specimens by the microbiology staff at the tivity in the presence of a physiological concentration of University of Missouri Veterinary Medical Diagnostic La- salts, and high production cost due to their complex boratory as part of standard service. The isolates were do- structure [15]. nated to the present project with appropriate permission. Via the characterization of the structure-function rela- All bacterial strains were maintained and grown in tionship of AvBDs and various analogues, it has been Luria-Bertani broth or agar (LB, BD Difco™)at 37 °C as identified that the concentrated surface net positive described previously [13, 14]. charge and the N-terminal α-helix and the β2-β3 loop structure are essential functional domains for antimicro- Peptide synthesis and characteristics bial and chemotactic properties [13]. Linear AvBD ana- All peptides were custom synthesized using the standard logues with a high net positive charge (+ 9) and an solid phase 9-fluorenylmethoxycarbonyl (Fmoc) method N-terminal helix-loop possess improved antimicrobial as previously synthesizing wild-type AvBDs [14] and potency and partial chemotactic activity, compared to purified by reverse phase high-performance liquid Yang et al. BMC Microbiology (2018) 18:54 Page 3 of 14 chromatography (RP-HPLC) (Lifetein, Hillsborough, NJ). Antimicrobial activity assay The purity of the synthetic CAMPs was greater than Minimum inhibitory concentrations (MICs) were deter- 98.5% as verified by liquid chromatography-mass spec- mined primarily based on the guidelines of the Clinical trometry (LC-MS) (Lifetein, Hillsborough, NJ). The and Laboratory Standards Institute (CLSI) [27, 28]. The charge and hydrophobicity of the newly designed Muller Hinton (MH) II broth used in MIC assay con- CAMPs at neutral pH were calculated using online Pep- tained 20–25 mg/L of calcium and 10–12.5 mg/L of tide property calculator (PepCalc.com). Protease cutting magnesium. The procedures were described in previous sites were predicted by using PROSPER (https://pros- studies [13, 14]. In brief, two-fold serially diluted per.erc.monash.edu.au) and SignalP 4.1 server (http:// CAMPs (2 to 256 μg/ml) were mixed with appropriate www.cbs.dtu.dk/ services/SignalP/). The bacterial strains at a final concentration of 5 × 10 CFU/ three-dimensional structures of AvBDs and the newly ml in a 96-well microtiter plate (Nunc™, Thermo Fisher designed templates were analyzed by using the Scientific). Following incubation at 37 °C for 24 h, MIC I-TASSER (Iterative Threading Assembly Refinement) was recorded. All assays were conducted in triplicate. protein structure and function prediction program (http://zhanglab.ccmb.med.umich.edu/I-TASSER). The Salt resistance assay distribution of selected amino acid residues was evalu- A major hindrance to clinical application of defensin ated using PyMOL, a user-sponsored molecular peptides is the interference of function by cationic salts, visualization system (https://www.pymol.org/). The hel- often referred to as salt sensitivity [29]. The effect of salt ical wheel projection was calculated using the Helical on antimicrobial activity of novel CAMPs against P. aer- Wheel Projections program (http://rzlab.ucr.edu/scripts/ uginosa ATCC 27853 and S. aureus ATCC 29213 was wheel/wheel.cgi). determined by a colony count assay as described previ- ously [14], in the presence of either 0, 50, 100, and Circular dichroism spectrum analysis 150 mM NaCl or 0, 0.5, 1, and 2 mM CaCl . Two pep- Peptide structures were examined by far-UV circular di- tide concentrations, 0.5 × MIC and 1 × MIC, were in- chroism (CD) spectroscopy with an Aviv Model 62DS cluded. Medium without CAMP served as a negative spectrometer (Lakewood, NJ), in the wavelengths ran- control. Percent of killing was calculated using the fol- ging from 190 to 250 nm using a path length of 1 mm. lowing formula: (CFU -CFU ) / CFU × control treated control The spectra of peptides were measured at a concentra- 100% [14]. All assays were performed in triplicate. tion of 0.15 mg/ml in water. Spectra were baseline cor- rected by subtracting a blank spectrum containing only Hemolytic assay 2 − buffer and expressed as molar ellipticity θ (deg·cm ·mol The hemolytic assay was performed as described previ- ). ously [25]. Briefly, mouse red blood cells (RBCs, Innova- tive Research, Novi, MI) were washed three times with Membrane permeabilizing assay phosphate-buffered saline (PBS, pH 7.4), centrifuged at The membrane permeabilizing ability was determined 1000×g for 10 min, and resuspended in PBS to 10% (v/v). using the propidium iodide (PI) uptake assay [26]. PI is a The RBCs were treated with CAMPs at various concentra- fluorescent molecule that can only penetrate the im- tions ranging from 4 to 512 μg/ml (2-fold serial dilutions) paired microbial membrane and intercalate at 37 °C for 1 h. PBS and 0.2% Triton X-100 were used as double-stranded DNA. PI staining was done according negative and positive controls, respectively. The super- to the manufacturer’s instruction (Sigma Aldrich). In natant was transferred to a 96-well flat-bottomed polystyr- brief, the mid-logarithmic culture of P. aeruginosa ene plate (Thermo Fisher Scientific), and the amount of (ATCC 27853) was harvested by centrifugation at hemoglobin released into the supernatant was determined 1000×g for 10 min and resuspended in PBS (1 × by measuring the absorbance with a spectrophotometer at 10 CFU/ml). The bacteria were treated with each 540 nm. Hemolytic activity was expressed as the percent- CAMP at a concentration of 1 × MIC for 15, 30, 60, and age of hemolysis and calculated using the following equa- 90 min, respectively. After the addition of PI, the sus- tion: hemolysis (%) = (A -A )/(A -A ) × 100, where A is s 0 100 0 s pension was further incubated for 5 min at room the absorbance of the sample, A is the absorbance of temperature and shielded from light. The bacterial mix- completely lysed RBCs in 0.2% Triton X-100, and A is ture was coated on a microscope slide for analysis of red the absorbance in the complete absence of hemolysis (PBS fluorescence. Images were captured using a Nikon fluor- treatment). The assay was performed in triplicate. The escent microscope with Olympus DP2-BSE software therapeutic index (T.I.) was calculated according to a pre- (ECLIPSE E600, Japan) and the number of fluorescent viously published formula: T.I. = MHC / MIC [30]. GM cells per field was counted by the ImageJ software (NIH, MHC was the minimum hemolytic concentration that Bethesda, MD). The assay was performed in triplicate. caused 5% hemolysis of mouse RBCs. The MIC was GM Yang et al. BMC Microbiology (2018) 18:54 Page 4 of 14 the minimum inhibitory concentration of the peptide con- Statistical analysis centrations against bacterial growth after the geometric Differences between treatment groups were analyzed mean was calculated. MIC was the MIC for using the one-way analysis of variance (ANOVA) G- GM Gram-negative bacteria; MIC was the MIC for followed by Duncan’s test for multiple comparisons G+ GM Gram-positive bacteria. (SPSS 19.0, IBM Corp., Armonk, NY). Statistical signifi- cance was indicated by p < 0.05. Cell cytotoxicity assay The cytotoxicity of CAMPs to JASWII (ATCC CRL-11904) Results and CHO-K1 (ATCC CCL-61) cells was determined using Peptide design MTT (3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium Initially, two CAMP templates were designed to retain the bromide, Thermo Fisher Scientific) a cell proliferation assay antimicrobial and chemotactic properties of wild-type as described previously [14]. The following peptide concen- AvBD-6 and AvBD-12, respectively. The first CAMP tem- trations were included in the present study: 64, 128, 256, plate (CAMP-t1) possessed the structural domain (N-ter- and 512 μg/ml. Following CAMP treatment, percent of vi- minal α-helix and β2-β3 loop) of AvBD-12aswellasits able cells, relative to the untreated control, was recorded. analogues A2 and A3 which appeared to be essential to the The assays were performed in triplicate. broad chemotactic activity of AvBD-12 [13]. To increase net positive charge, the negatively charged amino acid resi- Chemotaxis assay dues Asp (D) and Glu (E) were substituted with positively CAMP-induced migration of JAWSII (ATCC charged amino acid residues Lys (K) or Arg (R). To en- CRL-11904) and CHO-K1 (ATCC CCL-61) transfected hance membrane permeabilization and salt resistance, a with CCR-2 was determined using a microchemotaxis poly-Trp tail was incorporated to the C-terminus of the assay described previously [14, 31]. Chemotactic indexes peptide, as it was previously shown that coating antimicro- (C.I.) was calculated as the number of migrated cells in- bial peptides with 3 Trp residues significantly increased salt duced by CAMPs divided by the number of migrated resistance [18, 29, 35]. The resulting peptide CAMP-t1 con- cells in the control wells without CAMPs. The assays sisted of 25 amino acid residues: RKFLRRRGE- were repeated five times. VAHFSQKSLGLYCWWW. The predicted three-dimensional structure of CAMP-t1 mimics that of Protease resistance assay AvBD-12 (Fig. 1a). The second CAMP template Protease resistance was evaluated by using sodium dodecyl (CAMP-t2) consisted of the key amino acid residues of sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) AvBD-6: PIHRRIPPRWPRLKRRW, responsible for the followed by antimicrobial assays. CAMP peptides (10 μg) concentrated surface charge and hydrophobicity of AvBD-6 were treated with various proteases at various concentra- [13]. In silico analysis indicated that CAMP-t2 assumed a tions comparable to or higher than that found in host or coil and α-helical structure (Fig. 1b). To optimize anti- bacterial culture, including 0.12, 0.6, and 1.2 μg/ml of microbial activity, CAMP-t2 was subjected to further modi- α-chymotrypsin (Thermo Fisher Scientific) [25], 0.4, 4.4, fications to create CAMP-A and CAMP-B (Fig. 1c and d) and 20 μg/ml of matrilysin (metalloproteinase-7, Sigma Al- using the following criteria: 1) short in length (≤ 20 a.a. resi- drich) [32], 0.2, 2, and 20 μg/ml elastase (Thermo Fisher dues), 2) α-helical structure, 3) hydrophobic and hydro- Scientific) [33], or 0.2, 2, and 20 μg/ml of cathepsin B philic residues on opposite sides of the helical surface to (Thermo Fisher Scientific) [34]. Protease digestion assay facilitate pore formation in bacterial membrane by multiple was carried out in 20 μl of digestion buffer (25 mM Tris peptides [36], 4) lack of cutting sites for major proteases: and 150 mM NaCl, pH 7.8) for 1 h at 37 °C. After treat- aspartic protease, cysteine protease, metalloprotease, serine ment, the digestion mixture was analyzed by SDS-PAGE on proteases, 5) N-terminal acetylation and C-terminal amida- 16.5% polyacrylamide gel. To determine the effect of prote- tion to increase the metabolic stability of CAMPs [37], and ase digestion on the antimicrobial activity of two most ac- 6) addition of C-terminal poly-Trp tail to enhance mem- tive CAMPs, each peptide was first treated with a protease brane permeabilization and salt resistance [29, 35]. for 1 h at 37 °C. The mixture was then diluted to 1 × MIC of the peptide and subjected to colony count assay as de- Structural features scribed above. Assay buffer containing protease but not The amino acid sequences and relevant biochemical peptide and buffer containing untreated peptide were in- characteristics of all CAMPs were presented in Table 1. cluded as controls. Protease inhibition of the antimicrobial The structures of CAMPs was analyzed by a far-UV activity of CAMP peptide was expressed as a percentage of spectrometer. In general, the CD spectrum of α-helical killing by protease treated peptide in relevance to the un- structures presents two negative bands at 208 and treated peptide. The experiment was repeated three times 222 nm along with a positive band at 192 nm whereas in triplicate in each assay. random coil is characterized by a single band below Yang et al. BMC Microbiology (2018) 18:54 Page 5 of 14 Fig. 1 The predicted structures of newly designed templates CAMP-t1 and CAMP-t2. a The three-dimensional structure of CAMP-t1 derived from AvBD-12. Red: α-helix; Green: loop (β2-β3 loop in AvBD-12, β1-β2 loop in template CAMP-t1). b The three-dimensional structure of CAMP-t2 derived from AvBD-6. CAMP-t2 were further optimized to CAMP-A (c) and CAMP-B (d). CAMP-A and CAMP-B were coated with poly-Trp tails. Red: positively charged amino acid residues; Blue: hydrophobic amino acid residues; Green: prolines negative 200 nm [38]. As shown in Fig. 2a, CAMP-t1 Antimicrobial activity showed two weak bands at 202 nm and 225 nm, indicat- The minimum inhibitory concentrations of the newly ing partial α-helical structure of the peptide. CAMP-t2 designed CAMPs against P. aeruginosa and S. aureus displayed a strong single band around 200 nm, confirm- ATCC reference strains and multidrug-resistant P. aeru- ing the random coil structure as predicted in silico (Fig. ginosa and methicillin-resistant S. pseudinetermedius 1b). CAMP-A had two negative bands at 200 nm and (MASP) strains were compared to that of AvBD-6, a nat- 225 nm along with a positive band at 190 nm which ural host CAMP with potent antimicrobial activity under verified the predicted α-helical structure. CAMP-B low salt conditions (Table 2). CAMP-t1 showed signifi- showed two weak bands at 202 nm and 225 nm which cantly improved anti-Pseudomonas activity with MIC was consistent with the predicted partial α-helix (Fig. values 4-fold (for ATCC reference strain) and 2-fold 1d). The CAMPs were subjected to helical wheels pro- (clinical isolates) lower than that of AvBD-6. CAMP-t1 jection analysis (Fig. 2b). CAMP-t1 showed random dis- showed improved antimicrobial activity against S. aureus tribution of amino acids and CAMP-t2 showed a large ATCC reference strain (MIC was 4-fold lower than that distribution angle of hydrophilic residues (proline and of AvBD-6), but not MRSP clinical isolates as evidenced positively charged amino acids) and a small distribution by the high MIC values similar to that of AvBD-6. angle of hydrophobic residues. CAMP-A displayed an CAMP-t2, the shorter template, demonstrated similar amphipathic structure with hydrophobic and hydrophilic antimicrobial activity against Pseudomonas strains and residues on the opposite site of the helix, which sup- significantly enhanced anti-Staphylococcus activity, com- ported the designing feature (Figs. 1c and 2b). CAMP-B pared to CAP-t1. The MICs of CAMP-t2 against S. aur- exhibited a small helical wheel with hydrophobic residue eus and MRSP strains were up to 4-fold lower than that W4 in the middle of positively charged residues (Fig. 1d) of CAMP-t1 and AvBD-6. CAMP-A, a derivative of and the W9W10W11-tail folding along with the helix. CAMP-t2, established further improved antimicrobial Table 1 The characteristics of newly designed CAMPs Peptide Amino acid sequence Length (aa) Molecular Weight Charge Hydrophobicity CAMP-t1 RKFLRRRGEVAHFSQKSLGLYCWWW 25 3251.84 + 4 44% CAMP-t2 PIHRRIPPRWPRLKRRW 17 2361.90 + 7 29% CAMP-A LRRLKPLIRPWLRPLRRWWW 20 2839.55 + 7 50% CAMP-B RRRWRKRRWWW 11 1869.24 + 7 36% CAMP-t1, CAMP-t2, and CAMP-B are coated with a Trp-tail. N-terminal acetylation and C-terminal amidation were introduced to all peptides Yang et al. BMC Microbiology (2018) 18:54 Page 6 of 14 Fig. 2 The far-UV CD spectra and helical wheels projections of CAMPs. a The far-UV CD spectra of CAMPs in H2O recorded at room temperature. 2 − 1 Spectra were baseline corrected and expressed as molar ellipticity θ (deg·cm ·mol ). Gray solid line: CAMP-t1; dash line: CAMP-t2; dotted line: CAMP-A; dark solid line: CAMP-B. b Helical wheels projections of CAMPs. Relevant features of amino acid residues were coded by various shapes and colors. Hydrophilic residues: circles, hydrophobic residues: diamonds, positively charged residues: pentagons. Hydrophobic residues: greento yellow, as the hydrophobicity decreased to zero, color changed gradually from dark green to yellow. Hydrophilic residues: red, the red tone decreased proportionally to the decrease in hydrophilicity. Charged residues: light blue activity against P. aeruginosa and S. aureus reference Salt resistance strains, multidrug-resistant P. aeruginosa strains and MRSP The impact of cationic salts on the bactericidal activity clinical isolates. The MICs of CAMP-A against P. aerugi- of CAMPs was assessed using an assay system contain- nosa and MRSP isolates were 4- to 32- fold lower than that ing various concentrations of NaCl (0 to 150 mM) or of AvBD-6. CAMP-B, the second derivative of CAMP-t2, CaCl (0 to 2 mM) at two peptide concentrations (1 × also showed improved antimicrobial activity against P. aer- MIC and 0.5 × MIC). As shown in Fig. 4, increasing uginosa and similar potency against Staphylococcus spp. NaCl and CaCl concentrations had no impact on the However, CAMP-B was less effective than CAMP-A in kill- bactericidal activity of CAMP-t1 which possesses a ing both Pseudomonas spp. and methicillin-resistant S. C-terminal poly-Trp tail. In contrast, the bactericidal ac- pseudintermedius (p <0.05, Table 2). tivity of CAMP-B without the Trp tail was negatively af- fected by increased salt concentrations. At the Membrane permeabilizing activity physiological conditions of NaCl (100 to 150 mM) and A propidium iodide (PI) uptake assay was carried out to CaCl (1 to 2 mM), CAMP-t2 retained approximately determine the membrane permeabilizing activity of newly 40% of its killing activity against P. aeruginosa and S. designed CAMPs (Fig. 3). As shown in Fig. 3a, aureus, compared to the results obtained at salt-free CAMP-treated P. aeruginosa were stained red, indicating condition (Fig. 4). The salt-resistance pattern of that bacterial membranes were damaged by CAMPs. In CAMP-t2 was similar at two different peptide concen- contrast, untreated bacteria did not show red fluores- trations. Similar to CAMP-t1, CAMP-A and CAMP-B cence. When the permeabilizing ability was assessed based with a Trp tail exhibited strong tolerance to NaCl and on the numbers of red cells per field, a time-dependent in- CaCl (Fig. 4). crease was observed from 15 to 90 min (Fig. 3b and c). However, the number of red bacteria did not increase sig- Hemolytic activity and cytotoxicity nificantly after 30 min, indicating a fast-action mode of The hemolytic activity of CAMPs to mouse RBCs was CAMPs. Similar results were obtained for all CAMPs and analyzed (Fig. 5). At high concentrations, CAMP-A lysed both bacterial pathogens (Fig. 3). approximately 3.6% of mRBCs at 128 μg/ml, 10.1% of Yang et al. BMC Microbiology (2018) 18:54 Page 7 of 14 Table 2 The minimum inhibitory concentration (MIC) of CAMPs CAMPs CAMP-t1 CAMP-t2 CAMP-A CAMP-B AvBD-6 Bacteria MIC (μg/ml) MIC (μg/ml) MIC (μg/ml) MIC (μg/ml) MIC (μg/ml) (strain identification) P. aeruginosa (ATCC 27853) 64 64 16 32 > 256 P. aeruginosa (1704173) 128 64 16 32 > 256 P. aeruginosa (1703357) 128 128 16 64 > 256 P. aeruginosa (1703511) 128 128 16 64 > 256 P. aeruginosa (1703000) 128 64 8 32 > 256 P. aeruginosa (1703002) 128 256 16 64 > 256 P. aeruginosa (1703451) 128 64 16 64 > 256 P. aeruginosa (1703949) 128 128 16 64 > 256 P. aeruginosa (1703290) 128 128 16 64 > 256 P. aeruginosa (1703983) 128 256 16 64 > 256 P. aeruginosa (1704175) 128 64 16 32 > 256 a a c b MIC 122.18 ± 19.29 122.18 ± 72.71 15.27 ± 2.41 52.36 ± 16.14 > 256 Average G- S. aureus (ATCC 29213) 64 64 32 32 256 S. pseudintermedius(13164006) 256 32 16 32 > 256 S. pseudintermedius(13203008) 256 32 16 64 256 S. pseudintermedius(13178007) 256 64 16 64 256 S. pseudintermedius(13267017) 128 32 16 32 > 256 S. pseudintermedius(13252001) 64 32 16 32 256 S. pseudintermedius(13269013) 256 32 32 32 > 256 S. pseudintermedius(13193006) 256 128 32 32 > 256 S. pseudintermedius(13228005) 256 32 16 64 > 256 S. pseudintermedius(13207007) 256 32 16 64 > 256 S. pseudintermedius(13250111) 256 32 32 64 256 a b c b MIC 209.45 ± 81.41 46.55 ± 29.89 21.82 ± 8.07 46.55 ± 16.71 ≥256 Average G+ Multidrug resistant clinical isolates of P. aeruginosa were resistant to chloramphenicol, tetracycline, sulfamethoxazole, and β-lactam antibiotics: amoxicillin, ampicillin and cefazolin. Methicillin-resistant S. pseudintermedius (MRSP) Superscripts a, b, and c mean significant difference (p < 0.05) between MICs of four CAMPs against either Gram-negative or Gram-positive bacteria mRBCs at 256 μg/ml, and 17.5% of mRBCs at 512 μg/ treatment with CAMP-A at a concentration equal to or ml. CAMP-t1, CAMP-t2, and CAMP-B did not cause greater than 128 μg/ml. After treatment with CAMP-A, more than 5% of mRBCs at the concentration of 512 μg/ no significant difference in the percentage of cell viabil- ml. The minimum hemolytic concentration (MHC), geo- ity was observed between 4 h treatment and 48 h metric means of the MIC (MIC ), and therapeutic treatment. GM index (T.I.) were determined for each CAMP (Table 3). The average T.I. of CAMP-A and CAMP-B against P. Chemotactic activity aeruginosa were 8.72 ± 2.41 and > 10.90 ± 4.0, respect- The chemotactic activity of the CAMPs for JAWSII and ively. The T.I. of CAMP-A and CAMP-B against S. aur- CCR2-transfected CHO-K1 cells were determined eus were 6.54 ± 2.01 and > 12.36 ± 4.18, respectively. (Fig. 7). The results indicated that CAMP-t1 with the Cell cytotoxicity of the newly designed CAMPs was N-terminal helix-loop structure of AvBD-12 did not also determined by MTT cell viability assay (Fig. 6). Ex- show expected chemotactic activity to either cell line posure of murine immature dendritic cell line JAWSII (Fig. 7a and b). Interestingly, CAMP-A with the highest (Fig. 6a) and CHO-K1 (Fig. 6b) cells to CAMP-t1, antimicrobial activity showed mild chemotactic activity CAMP-t2, and CAMP-B at concentrations of 64, 128, at a concentration of 64 μg/ml (C.I. = 5.13; 77.5% of 256, and 512 μg/ml for 4 to 48 h did not significantly wild-type AvBD-12, C.I. = 6.62) for JAWSII cells. As affect cell viability. However, the viability of both JAW- shown in Fig. 7c, more JAWSII cell migration was in- SII and CHO-K1 cells was significantly decreased by duced by CAMP-A with increasing peptide Yang et al. BMC Microbiology (2018) 18:54 Page 8 of 14 Fig. 3 Membrane permeabilizing activity of CAMPs. a Representative fluorescence microscopy images of CAMP-treated and control bacteria stained with membrane-impermeable DNA dye propidium iodide (PI). Right panels are enlarged focal areas (a and b) from the left panels. b The number per field of positively stained P. aeruginosa at various times post-CAMP-treatment. c The number per field of S. aureus at various times post-CAMP-treatment. Data are expressed as the means ± SD of three independent experiments. An asterisk indicates a significant difference between different time points (*p < 0.05). Bar: 100 μm concentrations, ranging from 1 to 64 μg/ml. No signifi- and cathepsin B at concentrations up to 20 μg/ml. cant chemotactic activity for CCR-2 transfected Treatment with these proteases did not affect the anti- CHO-K1 cells was detected (Fig. 7b). CAMP-t2 and microbial activity of CAMP-A and CAMP-B (data not CAMP-B did not show any chemotactic activity for ei- shown). ther JAWSII or CCR2-CHO-K1 cells. Discussion Protease resistance Host cationic antimicrobial peptides, such as defensins, The resistance of CAMP-A and CAMP-B, two peptides have been a subject of research interest because of their exhibiting strong antimicrobial activity and broad-spectrum antimicrobial activity and low potential salt-resistance, to various proteases was evaluated by for resistance development. For instance, human subjecting protease-treated peptides to SDS-PAGE α-defensin HD5 and HDP4 showed strong killing activity (Fig. 8). The results indicated that CAMP-A and against S. aureus ATCC 25923 and ATCC 29213 with CAMP-B were partially digested by α-chymotrypsin at the lethal doses [39]. Our previous structure-function 0.12 to 1.2 μg/ml, as indicated by the presence of pep- analysis of AvBDs and their analogues indicates that the tide bands with lower molecular weight than that of the highly concentrated surface positive charge plays a pre- untreated peptides (Fig. 8a and b). The antimicrobial ac- dominant role in the antimicrobial potency of the pep- tivity of CAMP-A and CAMP-B against P. aeruginosa tides whereas the CCR2-binding domain (N-terminal decreased significantly post-digestion (Fig. 8c). The α-helix along with an adjacent loop) is responsible for digested peptides retained approximately 90% (at the broad-spectrum chemotactic activity for both avian 0.12 μg/ml), 80% (at 0.66 μg/ml), and 30% (at 1.2 μg/ml) and mammalian dendritic cells [13]. For example, linear of the killing activity of the untreated CAMP-A or AvBD analogues with a high net positive charge (+ 9), CAMP-B. In contrast, the anti-Staphylococcus activity modest hydrophobicity (40%), and a predicted CCR2 was mildly affected only when the CAMPs were treated binding domain exhibit strong antimicrobial and mild with the highest concentration (1.2 μg/ml) of chemotactic activities. However, the linear peptides de- α-chymotrypsin (Fig. 8d). CAMP-A and CAMP-B were signed in our previous study are still lengthy (45 amino not cleaved by metalloproteinases matrilysin, elastase, acid residues) and susceptible to physiological Yang et al. BMC Microbiology (2018) 18:54 Page 9 of 14 Fig. 4 Effect of salts on the antibacterial activity of CAMPs against P. aeruginosa and S. aureus. The effect of salts on the antibacterial activity was determined using two peptide concentrations: 1 × MIC and 0.5 × MIC. a Percent of killing against P. aeruginosa at 0, 50, 100, and 150 mM NaCl; (b) Percent of killing against S. aureus at 0, 50, 100, and 150 mM NaCl; (c) Percent of killing against S. aureus at 0, 0.5, 1, and 2 mM CaCl . d Percent of killing against S. aureus at 0, 0.5, 1, and 2 mM CaCl . Data represent the means ± SD of three independent experiments. An asterisk indicates the statistically significant difference between antimicrobial activity in the presence and absence of salts (*p < 0.05 and **p < 0.01) concentrations of NaCl. Although the sensitivity of achieve the following goals: structurally simple (linear, AvBD analogues to proteases was not evaluated in our short and all natural amino acids), resistant to proteases previous studies, investigations conducted by others sug- and cationic salts, non-cytotoxic, strong and gest that linear peptides are susceptible to bacterial broad-spectrum antimicrobial activity, and potentially metalloprotease, cysteine protease and human neutro- chemotactic for immune cells. phil elastase [38, 39]. To develop antimicrobial peptides We first designed two CAMP templates, CAMP-t1 suitable for therapeutic or preventive use, we utilized an and CAMP-t2, by extrapolating the CCR2 binding do- integrated approach to modify AvBD analogues to main of AvBD-12 (CAMP-t1) and key amino acid Fig. 5 Hemolytic activity of CAMPs. CAMP-induced hemolysis (%) of mouse red blood cells at various peptide concentrations is defined as a percentage of complete hemolysis caused by 0.2% Triton X-100. Data are expressed as the means ± SD of three independent experiments Yang et al. BMC Microbiology (2018) 18:54 Page 10 of 14 Table 3 Therapeutic index of new cationic antimicrobial peptides (CAMPs) Peptides CAMP-t1 CAMP-t2 CAMP-A CAMP-B MHC > 512 > 512 128 > 512 MIC 122.18 ± 19.29 122.18 ± 72.71 15.27 ± 2.41 52.36 ± 16.14 Average G- MIC 209.45 ± 81.41 46.55 ± 29.89 21.82 ± 8.07 46.55 ± 16.71 Average G+ T.I. of G- > 4.36 ± 1.21 > 5.45 ± 2.54 8.72 ± 2.41 > 10.90 ± 4.0 T.I. of G+ > 3.27 ± 2.41 > 13.45 ± 4.48 6.54 ± 2.01 > 12.36 ± 4.18 Therapeutic index, (T.I.) is defined as the ratio of, MHC to, MIC . MHC (μg/ml) is the minimum hemolytic concentration that caused 5% hemolysis of mouse red GM blood cells, (mRBCs). The MIC (μg/ml) means the geometric mean, (GM) of the, MIC values of the peptides against bacteria. MIC is the, MIC for Gram- GM Average G- GM negative bacteria, MIC is the MIC for Gram-positive bacteria Average G+ GM residues contributing to the concentrated surface posi- degradation [40]. To evaluate the antimicrobial proper- tive charge and hydrophobicity of AvBD-6 (CAMP-t2), ties of these peptides, we determined their MICs against respectively. For CAMP-A, the negatively charged amino P. aeruginosa and Staphylococcus according to the acid residues (D and E) in AvBD-12 were replaced by guidelines of CLSI [27, 28]. Both CAMP-t1 and positively charged residues (K and R). Because the net CAMP-t2 demonstrated improved antimicrobial activity, positive charge of CAMP-t1 was still relatively low (+ 4), compared to AvBD-6 and AvBD-12 as well as previously a poly-Trp tail was incorporated into its C-terminus to designed AvBD analogues [13, 14]. Although CAMP-t1 boost the antimicrobial activity. Trp is known for its ten- retained the N-terminal α-helix and an adjacent loop dency to insert into membrane lipid bilayer and structure of AvBD-12, it lost the desired chemotactic Trp-rich peptides exhibit enhanced antimicrobial activity property [13], suggesting that either the amino acid and salt resistance [18]. N-terminal acetylation and composition was not optimal or additional structural C-terminal amidation (mimicking native proteins) were components were required for CCR2 binding. We then also incorporated to increase the metabolic stability of focused on CAMP-t2, a shorter template with a peptides as well as their resistance to enzymatic coil-helix structure. This peptide showed stronger Fig. 6 Cytotoxicity of CAMPs to JAWSII and CHO-K1 cells. Effect of CAMPs on the viability of mouse immature dendritic JAWSII cells (a) and hamster ovary CHO-K1 cells (b) at 4 and 48 h of incubation with peptide at the concentration of 64 to 512 μg/ml. Results are percentages of viable cells relative to the untreated control cells. The data are expressed as the mean ± SD of three independent experiments. An asterisk indicates a statistically significant difference in the viability of CAMP-treated cells and untreated cells (*p < 0.05 and **p < 0.01) Yang et al. BMC Microbiology (2018) 18:54 Page 11 of 14 Fig. 7 Chemotactic activity of CAMPs. CAMP-induced migration of mouse immature dendritic JAWSII cells (a) and CHO-K1 cells expressing avian CCR2 (b) was measured at the following peptide concentrations: 1, 4, 16, and 64 μg/ml. c Migrated JAWSII cells on the membrane induced by CAMP-A at peptide concentrations of 1, 4, 16, and 64 μg/ml. White pores are membrane pores (diameter: 8 μm) and blue ones are stained JAWSII cells. Bar: 50 μm. Chemotactic index (C.I.) was expressed as the number of migrated cells induced by CAMP / the number of migrated cells in response to chemotactic buffer. Data represent the means of five independent experiments ± SD. An asterisk indicates significant difference (*p < 0.05 and **p < 0.01) antimicrobial activity against methicillin-resistant S. methicillin-resistant S. pseudintermedius clinical isolates. pseudintermedius than CAMP-t1 and AvBD-6, but was With a poly-Trp tail, α-helical structure and increased sensitive to high concentrations of cationic salts. The surface positive charge, CAMP-A and CAMP-B were rest of the study was concentrated primarily on improv- fully functional at physiological concentrations of NaCl ing the antimicrobial property and salt resistance of and CaCl . These peptides were also resistant to metal- CAMP-t2. loproteinases, matrilysin and elastase, and cathepsin B at Studies have indicated that amphipathicity is a key concentrations higher than that in bacterial protein se- characteristic required for membrane permeabilization cretion [42] or in mammalian host cells [34]. Although in which hydrophobic residues interact with membrane they were still cleaved by α-chymotrypsin, the antimicro- lipid components while hydrophilic regions either bind bial activity was minimally affected at the concentration with the phospholipid head groups or form the lumen of of 0.12 μg/ml, about 3 to 40 times higher than the con- a membrane pore [5, 14]. Alpha-helical peptides with centration tested in human samples using different hydrophobic and hydrophilic residues on opposite sides methods, 4 ng/ml [43] and 37.5 ng/ml [44]. At a high of the peptide molecule have antimicrobial property [8]. concentration (1.2 μg/ml), α-chymotrypsin treatment re- The shorter template, CAMP-t2 with coil-helix structure duced the killing activity against P. aeruginosa (p < 0.05) (only 35% of residues form α-helix), was further modi- but not S. aureus (p > 0.05). The discrepancy could be fied to form an α-helix structure which confers struc- associated with structural difference between tural stability [41]. To maximize the antimicrobial Gram-negative and Gram-positive bacterial membranes activity and minimize the damaging effect on host cell which warrants further investigation into the mechanism membrane, Trp and Pro residues were incorporated and of antimicrobial action of these peptides. the amino acid residues were strategically arranged to Our previous studies have shown that AvBDs could avoid protease cutting sites predicted using online disrupt bacterial membrane resulting in cell deform- PROSPER and SignalP 4.1 servers. The resulting pep- ation, increased membrane permeabilization, and mem- tides, CAMP-A and CAMP-B, demonstrated strong anti- brane damage [13, 14]. In the present study, data from microbial activity against ATCC bacterial reference propidium iodide (PI) staining assay suggested that the strains as well as multi-drug resistant P. aeruginosa and primary mode of action of the newly designed CAMPs Yang et al. BMC Microbiology (2018) 18:54 Page 12 of 14 Fig. 8 Effects of protease treatment on the antimicrobial activity of CAMP-A and CAMP-B against P. aeruginosa and S. aureus. Peptides were digested with α-chymotrypsin, elastase, matrilysin, or cathepsin B for 1 h at 37 °C and subjected to SDS-PAGE (16.5% polyacrylamide gel) analysis. a CAMP-(a). b CAMP-(b). c Antimicrobial activity of CAMP-A post-digestion by α-chymotrypsin at the concentration of 0.12, 0.66, and 1.2 μg/ml. d Antimicrobial activity of CAMP-B post-digestion by α-chymotrypsin at the concentration of 0.12, 0.66, and 1.2 μg/ml. Results are expressed as percent killing by digested peptides over untreated peptides. Data are expressed as the means ± SD of three independent experiments. An asterisk indicates a statistically significant difference between antimicrobial activity with and without protease (*p < 0.05 and **p < 0.01) was membrane attacking, which is considered a mechanism to show chemotactic activity. Interestingly, CAMP-A less likely to trigger bacterial resistance. The CAMPs did not with high positive charge and modest hydrophobicity in- show any detectable cytotoxicity and hemolytic activity at duced chemotactic migration of JAWSII cells which oc- the doses required for effective bacterial killing. CAMP-t1, curred possibly via the formyl-peptide receptors like CAMP-t2, and CAMP-B had minimal cytotoxic and mechanism such reported for human cathelicidin LL-37 hemolytic activities at a relatively high peptide concentration [50] and cathelicidin-like pleurocidins [51]. (512 μM/ml). CAMP-A, the most potent antimicrobial pep- tide, exhibited hemolytic and cytotoxic activities at concen- Conclusion trations equal to or greater than 128 μg/ml which, however, CAMP-t1 and CAMP-t2 were designed as templates was 6-fold higher than the MIC against P. aeruginosa and based on key structural and functional components of 4-fold higher than the MIC against Staphylococcus spp. The AvBD-12 and AvBD-6. CAMP-t1 with a predicted CCR cytotoxic property of CAMP-A was not surprising because binding domain of AvBD-12 demonstrated improved the peptide had a relatively high hydrophobicity (50%), antimicrobial activity but lost the original chemotactic which is known hydrophobicity is associated with their cyto- function. CAMP-t2 with key amino acid residues of toxic effect [45]. The undesired hemolytic activity was still AvBD-6 showed strong antimicrobial activity, but sensi- mild compared to other antimicrobial peptides including tivity to high concentrations of cationic salts. CAMP-t2 magainin isolated from the skin of African frog Xenopus lae- was further modified using an integrated design ap- vis and melittin from bee venom [46]. proach. CAMP-A and CAMP-B possess the following It has been suggested that the three conserved disul- advantageous characteristics: structural simplicity (short fide bridges were required for the chemotactic function and linear), resistance to salts and proteases, potent anti- of β-defensins [14, 47–49]. Data from our previous study microbial activity against multidrug-resistant P. aerugi- indicated that a predicted CCR2 binding domain (N-ter- nosa and methicillin-resistant Staphylococcus, rapid minal α-helix and an adjacent β2-β3 loop) in membrane attacking mode, and moderate therapeutic AvBD-12A3 (a linear peptide) without disulfide bridges index. Our data suggest that CAMP-A and CAMP-B are was chemotactic to JAWSII cells [13]. In the present excellent candidates for development as antimicrobial study, CAMP-t1 with a similar helix-loop domain failed therapeutic agents. Yang et al. BMC Microbiology (2018) 18:54 Page 13 of 14 Abbreviations 8. Bahar A, Ren D. Antimicrobial peptides. Pharmaceuticals. 2013;6(12):1543. ANOVA: One-way analysis of variance; AvBD: Avian beta-defensin; 9. Zhang G, Sunkara LT. Avian antimicrobial host defense peptides: from BSA: Bovine serum albumin; CFU: Colony-forming unit; CI: Chemotaxis index; biology to therapeutic applications. Pharmaceuticals (Basel). 2014;7:220–47. DCs: Dendritic cells; ESI-MS: Electrospray ionization mass spectrometry; 10. Greber KE, Dawgul M. Antimicrobial peptides under clinical trials. Curr Top FBS: Fetal bovine serum; fMLF: N-Formyl-methionyl-leucyl-phenylalanine; Med Chem. 2017;17(5):620–8. Fmoc: 9-fluorenyl-methoxycarbonyl; GM-CSF: Granulocyte macrophage 11. Mahlapuu M, Håkansson J, Ringstad L, Björn C. Antimicrobial peptides: an colony-stimulating factor; hBD6: Human β-defensin 6; LB: Luria-Bertani; emerging category of therapeutic agents. Front Cell Infect Microbiol. 2016;6:194. LPS: Lipopolysaccharides; LTA: Lipoteichoic acid; MIC: Minimum inhibitory 12. Zhao L, Yang M, Zhang M, Zhang S. Expression, purification, and in vitro concentration; PBS: Phosphate buffered saline; PI: Propidium iodide; RP- comparative characterization of avian beta-defensin-2, −6, and −12. Avian HPLC: Reversed-phase high-performance liquid chromatography; Dis. 2014;58(4):541–9. SD: Standard deviation 13. Yang M, Zhang C, Zhang MZ, Zhang S. Novel synthetic analogues of avian β-defensin-12: the role of charge, hydrophobicity, and disulfide bridges in Acknowledgments biological functions. BMC Microbiol. 2017;17(1):43. The authors would like to thank the bacteriology laboratory of Veterinary 14. Yang M, Zhang C, Zhang X, Zhang MZ, Rottinghaus GE, Zhang S. Structure- Medical Diagnostic Laboratory of University of Missouri for sharing the multi- function analysis of avian beta-defensin-6 and beta-defensin-12: role of drug resistant bacterial isolates. The authors thank Dr. Michael T. Henzl for charge and disulfide bridges. BMC Microbiol. 2016;16:210. the help with CD spectra analysis. The authors also wish to thank Ms. Chris- 15. Mohamed MF, Abdelkhalek A, Seleem MN. Evaluation of short synthetic tine Yan Zhang for editorial assistance. antimicrobial peptides for treatment of drug-resistant and intracellular Staphylococcus aureus. Sci Rep. 2016;6:29707. Funding 16. Rozek A, Powers JP, Friedrich CL, Hancock RE. Structure-based design of an The present study was funded by the College of Veterinary Medicine of the indolicidin peptide analogue with increased protease stability. Biochemistry. University of Missouri for the design of the study and collection, analysis, and 2003;42(48):14130–8. interpretation of data and in writing the manuscript. 17. Giangaspero A, Sandri L, Tossi A. Amphipathic alpha helical antimicrobial peptides. Eur J Biochem. 2001;268(21):5589–600. Availability of data and materials 18. Yu HY, Tu CH, Yip BS, Chen HL, Cheng HT, Huang KC, Lo HJ, Cheng JW. The datasets used and/or analysed during the current study are available Easy strategy to increase salt resistance of antimicrobial peptides. from the corresponding author on reasonable request. Antimicrob Agents Chemother. 2011;55(10):4918–21. 19. Mohanram H, Bhattacharjya S. Resurrecting inactive antimicrobial peptides from Authors’ contributions the lipopolysaccharide trap. Antimicrob Agents Chemother. 2014;58(4):1987–96. Conceived and designed the experiments, MY, CZ, MZ, SZ performed the 20. Jerala R. Synthetic lipopeptides: a novel class of anti-infectives. Expert Opin experiments, MY, CZ. Analyzed the data, MY, CZ, MZ, SZ contributed Investig Drugs. 2007;16(8):1159–69. reagents/materials/analysis tools, MY, SZ drafted the manuscript, MY, CZ, MZ, 21. Chu-Kung AF, Nguyen R, Bozzelli KN, Tirrell M. Chain length dependence of SZ. All authors read and approved the manuscript. antimicrobial peptide-fatty acid conjugate activity. J Colloid Interface Sci. 2010;345(2):160–7. Ethics approval and consent to participate 22. Vagner J, Qu H, Hruby VJ. Peptidomimetics, a synthetic tool of drug Not applicable. discovery. Curr Opin Chem Biol. 2008;12(3):292–6. 23. Ringstad L, Kacprzyk L, Schmidtchen A, Malmsten M. Effects of topology, Competing interests length, and charge on the activity of a kininogen-derived peptide on lipid The authors declare that they have no competing interests. membranes and bacteria. Biochim Biophys Acta. 2007;1768(3):715–27. 24. Saravanan R, Li X, Lim K, Mohanram H, Peng L, Mishra B, Basu A, Lee JM, Bhattacharjya S, Leong SS. Design of short membrane selective Publisher’sNote antimicrobial peptides containing tryptophan and arginine residues for Springer Nature remains neutral with regard to jurisdictional claims in improved activity, salt-resistance, and biocompatibility. Biotechnol Bioeng. published maps and institutional affiliations. 2014;111(1):37–49. 25. Kim H, Jang JH, Kim SC, Cho JH. De novo generation of short antimicrobial Author details peptides with enhanced stability and cell specificity. J Antimicrob Department of Veterinary Pathobiology, College of Veterinary Medicine, Chemother. 2014;69(1):121–32. University of Missouri, Columbia, MO 65211, USA. Department of Biomedical 26. Tyagi P, Singh M, Kumari H, Kumari A, Mukhopadhyay K. Bactericidal activity Science, College of Veterinary Medicine, University of Missouri, Columbia, MO of curcumin I is associated with damaging of bacterial membrane. PLoS 65211, USA. Veterinary Medical Diagnostic Laboratory, College of Veterinary One. 2015;10(3):e0121313. Medicine, University of Missouri, Columbia, MO 65211, USA. 27. CLSI. M07-A8 methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically; approved standard. 8th ed; 2009. Received: 26 September 2017 Accepted: 18 May 2018 28. CLSI: M100-S22 performance standards for antimicrobial susceptibility testing; twenty-second informational supplement. 2012. 29. Chu HL, Yu HY, Yip BS, Chih YH, Liang CW, Cheng HT, Cheng JW. Boosting References salt resistance of short antimicrobial peptides. Antimicrob Agents 1. Donoghue DJ. Antibiotic residues in poultry tissues and eggs: human health Chemother. 2013;57(8):4050–2. concerns? Poult Sci. 2003;82(4):618–21. 30. Zhang SK, Song JW, Gong F, Li SB, Chang HY, Xie HM, Gao HW, Tan YX, Ji 2. Ventola CL. The antibiotic resistance crisis: part 1: causes and threats. P T. SP. Design of an α-helical antimicrobial peptide with improved cell-selective 2015;40(4):277–83. and potent anti-biofilm activity. Sci Rep. 2016;6:27394. 3. Mohanram H, Bhattacharjya S. Salt-resistant short antimicrobial peptides. 31. Falk W, Goodwin RH Jr, Leonard EJ. A 48-well micro chemotaxis assembly Biopolymers. 2016;106(3):345–56. 4. Straus SK, Hancock REW. Mode of action of the new antibiotic for gram- for rapid and accurate measurement of leukocyte migration. J Immunol Methods. 1980;33(3):239–47. positive pathogens daptomycin: comparison with cationic antimicrobial peptides and lipopeptides. Biochim Biophys Acta Biomembr. 2006;1758(9): 32. Kuhlmann KF, van Till JW, Boermeester MA, de Reuver PR, Tzvetanova ID, Offerhaus 1215–23. GJ,Ten Kate FJ, Busch OR, van Gulik TM,Gouma DJ, CrawfordHC. Evaluation of 5. Jenssen H, Hamill P, Hancock RE. Peptide antimicrobial agents. Clin matrix metalloproteinase 7 in plasma and pancreatic juice as a biomarker for Microbiol Rev. 2006;19(3):491–511. pancreatic cancer. Cancer Epidemiol Biomark Prev. 2007;16(5):886–91. 6. Yeaman MR, Yount NY. Mechanisms of antimicrobial peptide action and 33. Forde E, Humphreys H, Greene CM, Fitzgerald-Hughes D, Devocelle M. resistance. Pharmacol Rev. 2003;55(1):27–55. Potential of host defense peptide prodrugs as neutrophil elastase- 7. Ganz T. Defensins: antimicrobial peptides of innate immunity. Nat Rev dependent anti-infective agents for cystic fibrosis. Antimicrob Agents Immunol. 2003;3(9):710–20. Chemother. 2014;58(2):978–85. Yang et al. BMC Microbiology (2018) 18:54 Page 14 of 14 34. Doxakis A, Maria A, Savvas P, Zafiroula I. Assessment of the roles of cathepsins B, H and L in the progression of colorectal cancer. Journal Cancer Therapy. 2013;4:1–7. 35. Hansen A, Schafer I, Knappe D, Seibel P, Hoffmann R. Intracellular toxicity of proline-rich antimicrobial peptides shuttled into mammalian cells by the cell-penetrating peptide penetratin. Antimicrob Agents Chemother. 2012; 56(10):5194–201. 36. Brogden KA. Antimicrobial peptides: pore formers or metabolic inhibitors in bacteria? Nat Rev Microbiol. 2005;3(3):238–50. 37. Nguyen LT, Chau JK, Perry NA, de Boer L, Zaat SA, Vogel HJ. Serum stabilities of short tryptophan- and arginine-rich antimicrobial peptide analogs. PLoS One. 2010;5(9):e12684. 38. Gopal R, Park JS, Seo CH, Park Y. Applications of circular dichroism for structural analysis of gelatin and antimicrobial peptides. Int J Mol Sci. 2012; 13(3):3229–44. 39. Ericksen B, Wu Z, Lu W, Lehrer RI. Antibacterial activity and specificity of the six human {alpha}-defensins. Antimicrob Agents Chemother. 2005;49(1):269–75. 40. Bulet P, Stocklin R, Menin L. Anti-microbial peptides: from invertebrates to vertebrates. Immunol Rev. 2004;198:169–84. 41. Park IY, Cho JH, Kim KS, Kim YB, Kim MS, Kim SC. Helix stability confers salt resistance upon helical antimicrobial peptides. J Biol Chem. 2004;279(14): 13896–901. 42. Shinnar AE, Butler KL, Park HJ. Cathelicidin family of antimicrobial peptides: proteolytic processing and protease resistance. Bioorg Chem. 2003;31(6): 425–36. 43. Lefkowitz RB, Schmid-Schonbein GW, Heller MJ. Whole blood assay for elastase, chymotrypsin, matrix metalloproteinase-2, and matrix metalloproteinase-9 activity. Anal Chem. 2010;82(19):8251–8. 44. Iwaki K, Ogawa M, Tanaka S, Kosaki G. Radioimmunoassay for human pancreatic chymotrypsin and measurement of serum immunoreactive chymotrypsin contents in various diseases. Res Commun Chem Pathol Pharmacol. 1983;40(3):489–96. 45. Mora-Navarro C, Méndez-Vega J, Caraballo-León J, Lee M-r, Palecek S, Torres-Lugo M, Ortiz-Bermúdez P. Hydrophobicity of antifungal β-peptides is associated with their cytotoxic effect on in vitro human colon Caco-2 and liver HepG2 cells. PLoS One. 2016;11(3):e0149271. 46. Matsuzaki K, Sugishita K-i, Harada M, Fujii N, Miyajima K. Interactions of an antimicrobial peptide, magainin 2, with outer and inner membranes of gram- negative bacteria. Biochim Biophys Acta Biomembr. 1997;1327(1):119–30. 47. Klüver E, Schulz-Maronde S, Scheid S, Meyer B, Forssmann WG, Adermann K. Structure-activity relation of human beta-defensin 3: influence of disulfide bonds and cysteine substitution on antimicrobial activity and cytotoxicity. Biochemistry. 2005;44(28):9804–16. 48. Hoover DM, Wu Z, Tucker K, Lu W, Lubkowski J. Antimicrobial characterization of human beta-defensin 3 derivatives. Antimicrob Agents Chemother. 2003;47(9):2804–9. 49. Schroeder BO, Wu Z, Nuding S, Groscurth S, Marcinowski M, Beisner J, Buchner J, Schaller M, Stange EF, Wehkamp J. Reduction of disulphide bonds unmasks potent antimicrobial activity of human beta-defensin 1. Nature. 2011;469(7330):419–23. 50. Tjabringa GS, Ninaber DK, Drijfhout JW, Rabe KF, Hiemstra PS. Human cathelicidin LL-37 is a chemoattractant for eosinophils and neutrophils that acts via formyl-peptide receptors. Int Arch Allergy Immunol. 2006;140(2): 103–12. 51. Pundir P, Catalli A, Leggiadro C, Douglas SE, Kulka M. Pleurocidin, a novel antimicrobial peptide, induces human mast cell activation through the FPRL1 receptor. Mucosal Immunol. 2014;7(1):177–87. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Microbiology Springer Journals

Beta-defensin derived cationic antimicrobial peptides with potent killing activity against gram negative and gram positive bacteria

Free
14 pages

Loading next page...
 
/lp/springer_journal/beta-defensin-derived-cationic-antimicrobial-peptides-with-potent-trgujXUu7C
Publisher
BioMed Central
Copyright
Copyright © 2018 by The Author(s).
Subject
Life Sciences; Microbiology; Biological Microscopy; Mycology; Parasitology; Virology; Life Sciences, general
eISSN
1471-2180
D.O.I.
10.1186/s12866-018-1190-z
Publisher site
See Article on Publisher Site

Abstract

Background: Avian β-defensins (AvBD) are cationic antimicrobial peptides (CAMP) with broad-spectrum antimicrobial activity, chemotactic property, and low host cytotoxicity. However, their bactericidal activity is greatly compromised under physiological salt concentrations which limits the use of these peptides as therapeutic agents. The length and the complex structure involving three conserved disulfide bridges are additional drawbacks associated with high production cost. In the present study, short linear CAMPs (11 to 25 a.a. residues) were developed based on the key functional components of AvBDs with additional modifications. Their biological functions were characterized. Results: CAMP-t1 contained the CCR2 binding domain (N-terminal loop and adjacent α-helix) of AvBD-12 whereas CAMP-t2 comprised the key a.a. residues responsible for the concentrated positive surface charge and hydrophobicity of AvBD-6. Both CAMP-t1 and CAMP-t2 demonstrated strong antimicrobial activity against Pseudomonas aeruginosa, Staphylococcus aureus and Staphylococcus pseudintermedius. However, CAMP-t1 failed to show chemotactic activity and CAMP-t2, although superior in killing Staphylococcus spp., remained sensitive to salts. Using an integrated design approach, CAMP-t2 was further modified to yield CAMP-A and CAMP-B which possessed the following characteristics: α-helical structure with positively and negatively charged residues aligned on the opposite side of the helix, lack of protease cutting sites, C-terminal poly-Trp tail, N-terminal acetylation, and C-terminal amidation. Both CAMP-A and CAMP-B demonstrated strong antimicrobial activity against multidrug- resistant P. aeruginosa and methicillin-resistant S. pseudintermedius (MRSP) strains. These peptides were resistant to major proteases and fully active at physiological concentrations of NaCl and CaCl . The peptides were minimally cytotoxic to avian and murine cells and their therapeutic index was moderate (≥ 4.5). Conclusions: An integrated design approach can be used to develop short and potent antimicrobial peptides, such as CAMP-A and CAMP-B. The advantageous characteristics, including structural simplicity, resistance to salts and proteases, potent antimicrobial activity, rapid membrane attacking mode, and moderate therapeutic index, suggest that CAMP-A and CAMP-B are excellent candidates for development as therapeutic agents against multidrug-resistant P. aeruginosa and methicillin-resistant staphylococci. Keywords: Cationic antimicrobial peptides, Peptide design, Salt resistance, Antimicrobial activity, Multidrug-resistant Pseudomonas aeruginosa, Methicillin-resistant Staphylococcus pseudintermedius * Correspondence: zhangshup@missouri.edu Department of Veterinary Pathobiology, College of Veterinary Medicine, University of Missouri, Columbia, MO 65211, USA Veterinary Medical Diagnostic Laboratory, College of Veterinary Medicine, University of Missouri, Columbia, MO 65211, USA Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Yang et al. BMC Microbiology (2018) 18:54 Page 2 of 14 Background the wild-type AvBD-12 [13]. However, the linear AvBDs The rapid emergence and spread of antimicrobial resist- designed in our previous study are still sensitive to ance, particularly those associated with Pseudomonas physiological salt conditions and the length of the pep- aeruginosa and Staphylococcus spp., have become a ser- tides (45 amino acid residues) remain to be shortened to ious threat to public health [1, 2]. The Centers for Dis- control the manufacturing cost. In addition, a previous ease Control and Prevention (CDC) estimated that each study has indicated that linear peptides are more suscep- year in the United States there are approximately 88,000 tible to protease degradation due to lack of complex ter- cases and 11,000 deaths due to infections with tiary structure stabilized by disulfide bridges found in methicillin-resistant Staphylococcus aureus (MRSA) [3]. natural defensin peptides [16]. Various studies have been conducted to search for new To increase salt- and protease-resistance, several solu- classes of antimicrobial therapeutic agents or antibiotic tions have been proposed, including: incorporating non- alternatives with novel targets and modes of action [4]. proteinogenic amino acids (e.g. D-amino acid Host cationic antimicrobial peptides (CAMPs), including substitutions and bulky amino acid β-naphthylalanin) linear peptides, α-helical peptides, circular and complex [17, 18] or LPS binding peptide motif (β-boomerang structures with loops and β-sheets constitute the first motif GWKRKRFG) [19], modifying the terminal regions line of innate defense against microbial pathogens [5]. via covalent linkage of a hydrophobic moiety (e. g. a The features shared by these CAMPs are net positive sterol or a fatty acid) [20, 21], peptidomimetic [22], alter- charge and amphipathicity [6]. The cationic property of ing the structure, charge, hydrophobicity, and shortening CAMP allows for the initial interaction of the peptide the length of the peptide [23, 24]. These strategies suc- with the anionic surface groups of the microbial mem- cessfully improved the antimicrobial function of CAMPs, brane and the hydrophobicity enables the peptide to in- but resulted in elevated hemolytic activity and increased tegrate into the hydrophobic core of the membrane. The manufacturing cost [18, 25]. In the present study, an inte- mechanism of action of CAMPs is complex, achieved grated approach was utilized to design short and compos- primarily through membrane damage and possibly sub- itionally simple CAMPs with potent antimicrobial activity, sequent interactions with cellular machineries, and the improved resistance to salts and proteases and minimal potential for development of microbial resistance is low cytotoxicity to host cells. The antibacterial property of the [6]. A major group of CAMPs with broad-spectrum anti- newly designed CAMPs against P. aeruginosa and Staphylo- microbial activity is β-defensins which contain three coccus spp., including clinical isolates of multidrug-resistant cysteine-cysteine disulfide bridges [7]. In addition to P. aeruginosa and methicillin-resistant S. pseudintermedius their antimicrobial activity and low potential for the de- (MRSP) was assessed under various conditions. velopment of resistance by bacteria, β-defensins have several other beneficial characteristics, such as modulat- Methods ing host immune response (e.g. chemo-attracting im- Bacterial strains and cultures mune cells) [8–11]. Our previous studies show that Pseudomonas aeruginosa (P. aeruginosa, ATCC 27853) avian β-defensins (AvBDs) such as AvBD-6 and and Staphylococcus aureus (S. aureus, ATCC 29213) AvBD-12 possess the following biological properties: were used to evaluate the novel CAMPs’ antimicrobial broad-spectrum antimicrobial activity, LPS-neutralizing activity, salt- and protease-resistance, and membrane per- ability, chemotactic activity, and minimal cell cytotox- meability. Ten multiple-drug resistant P. aeruginosa and icity [12–14]. Although β-defensins represent potentially ten methicillin-resistant S. pseudintermedius (MRSP) clin- a novel class of antimicrobial therapeutic agents, several ical isolates (Table 2) were used to evaluate the antimicro- obstacles must be overcome, including host cell cytotox- bial efficacy of CAMPs. The clinical isolates were cultured icity, degradation by proteases, loss of antimicrobial ac- from diagnostic specimens by the microbiology staff at the tivity in the presence of a physiological concentration of University of Missouri Veterinary Medical Diagnostic La- salts, and high production cost due to their complex boratory as part of standard service. The isolates were do- structure [15]. nated to the present project with appropriate permission. Via the characterization of the structure-function rela- All bacterial strains were maintained and grown in tionship of AvBDs and various analogues, it has been Luria-Bertani broth or agar (LB, BD Difco™)at 37 °C as identified that the concentrated surface net positive described previously [13, 14]. charge and the N-terminal α-helix and the β2-β3 loop structure are essential functional domains for antimicro- Peptide synthesis and characteristics bial and chemotactic properties [13]. Linear AvBD ana- All peptides were custom synthesized using the standard logues with a high net positive charge (+ 9) and an solid phase 9-fluorenylmethoxycarbonyl (Fmoc) method N-terminal helix-loop possess improved antimicrobial as previously synthesizing wild-type AvBDs [14] and potency and partial chemotactic activity, compared to purified by reverse phase high-performance liquid Yang et al. BMC Microbiology (2018) 18:54 Page 3 of 14 chromatography (RP-HPLC) (Lifetein, Hillsborough, NJ). Antimicrobial activity assay The purity of the synthetic CAMPs was greater than Minimum inhibitory concentrations (MICs) were deter- 98.5% as verified by liquid chromatography-mass spec- mined primarily based on the guidelines of the Clinical trometry (LC-MS) (Lifetein, Hillsborough, NJ). The and Laboratory Standards Institute (CLSI) [27, 28]. The charge and hydrophobicity of the newly designed Muller Hinton (MH) II broth used in MIC assay con- CAMPs at neutral pH were calculated using online Pep- tained 20–25 mg/L of calcium and 10–12.5 mg/L of tide property calculator (PepCalc.com). Protease cutting magnesium. The procedures were described in previous sites were predicted by using PROSPER (https://pros- studies [13, 14]. In brief, two-fold serially diluted per.erc.monash.edu.au) and SignalP 4.1 server (http:// CAMPs (2 to 256 μg/ml) were mixed with appropriate www.cbs.dtu.dk/ services/SignalP/). The bacterial strains at a final concentration of 5 × 10 CFU/ three-dimensional structures of AvBDs and the newly ml in a 96-well microtiter plate (Nunc™, Thermo Fisher designed templates were analyzed by using the Scientific). Following incubation at 37 °C for 24 h, MIC I-TASSER (Iterative Threading Assembly Refinement) was recorded. All assays were conducted in triplicate. protein structure and function prediction program (http://zhanglab.ccmb.med.umich.edu/I-TASSER). The Salt resistance assay distribution of selected amino acid residues was evalu- A major hindrance to clinical application of defensin ated using PyMOL, a user-sponsored molecular peptides is the interference of function by cationic salts, visualization system (https://www.pymol.org/). The hel- often referred to as salt sensitivity [29]. The effect of salt ical wheel projection was calculated using the Helical on antimicrobial activity of novel CAMPs against P. aer- Wheel Projections program (http://rzlab.ucr.edu/scripts/ uginosa ATCC 27853 and S. aureus ATCC 29213 was wheel/wheel.cgi). determined by a colony count assay as described previ- ously [14], in the presence of either 0, 50, 100, and Circular dichroism spectrum analysis 150 mM NaCl or 0, 0.5, 1, and 2 mM CaCl . Two pep- Peptide structures were examined by far-UV circular di- tide concentrations, 0.5 × MIC and 1 × MIC, were in- chroism (CD) spectroscopy with an Aviv Model 62DS cluded. Medium without CAMP served as a negative spectrometer (Lakewood, NJ), in the wavelengths ran- control. Percent of killing was calculated using the fol- ging from 190 to 250 nm using a path length of 1 mm. lowing formula: (CFU -CFU ) / CFU × control treated control The spectra of peptides were measured at a concentra- 100% [14]. All assays were performed in triplicate. tion of 0.15 mg/ml in water. Spectra were baseline cor- rected by subtracting a blank spectrum containing only Hemolytic assay 2 − buffer and expressed as molar ellipticity θ (deg·cm ·mol The hemolytic assay was performed as described previ- ). ously [25]. Briefly, mouse red blood cells (RBCs, Innova- tive Research, Novi, MI) were washed three times with Membrane permeabilizing assay phosphate-buffered saline (PBS, pH 7.4), centrifuged at The membrane permeabilizing ability was determined 1000×g for 10 min, and resuspended in PBS to 10% (v/v). using the propidium iodide (PI) uptake assay [26]. PI is a The RBCs were treated with CAMPs at various concentra- fluorescent molecule that can only penetrate the im- tions ranging from 4 to 512 μg/ml (2-fold serial dilutions) paired microbial membrane and intercalate at 37 °C for 1 h. PBS and 0.2% Triton X-100 were used as double-stranded DNA. PI staining was done according negative and positive controls, respectively. The super- to the manufacturer’s instruction (Sigma Aldrich). In natant was transferred to a 96-well flat-bottomed polystyr- brief, the mid-logarithmic culture of P. aeruginosa ene plate (Thermo Fisher Scientific), and the amount of (ATCC 27853) was harvested by centrifugation at hemoglobin released into the supernatant was determined 1000×g for 10 min and resuspended in PBS (1 × by measuring the absorbance with a spectrophotometer at 10 CFU/ml). The bacteria were treated with each 540 nm. Hemolytic activity was expressed as the percent- CAMP at a concentration of 1 × MIC for 15, 30, 60, and age of hemolysis and calculated using the following equa- 90 min, respectively. After the addition of PI, the sus- tion: hemolysis (%) = (A -A )/(A -A ) × 100, where A is s 0 100 0 s pension was further incubated for 5 min at room the absorbance of the sample, A is the absorbance of temperature and shielded from light. The bacterial mix- completely lysed RBCs in 0.2% Triton X-100, and A is ture was coated on a microscope slide for analysis of red the absorbance in the complete absence of hemolysis (PBS fluorescence. Images were captured using a Nikon fluor- treatment). The assay was performed in triplicate. The escent microscope with Olympus DP2-BSE software therapeutic index (T.I.) was calculated according to a pre- (ECLIPSE E600, Japan) and the number of fluorescent viously published formula: T.I. = MHC / MIC [30]. GM cells per field was counted by the ImageJ software (NIH, MHC was the minimum hemolytic concentration that Bethesda, MD). The assay was performed in triplicate. caused 5% hemolysis of mouse RBCs. The MIC was GM Yang et al. BMC Microbiology (2018) 18:54 Page 4 of 14 the minimum inhibitory concentration of the peptide con- Statistical analysis centrations against bacterial growth after the geometric Differences between treatment groups were analyzed mean was calculated. MIC was the MIC for using the one-way analysis of variance (ANOVA) G- GM Gram-negative bacteria; MIC was the MIC for followed by Duncan’s test for multiple comparisons G+ GM Gram-positive bacteria. (SPSS 19.0, IBM Corp., Armonk, NY). Statistical signifi- cance was indicated by p < 0.05. Cell cytotoxicity assay The cytotoxicity of CAMPs to JASWII (ATCC CRL-11904) Results and CHO-K1 (ATCC CCL-61) cells was determined using Peptide design MTT (3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium Initially, two CAMP templates were designed to retain the bromide, Thermo Fisher Scientific) a cell proliferation assay antimicrobial and chemotactic properties of wild-type as described previously [14]. The following peptide concen- AvBD-6 and AvBD-12, respectively. The first CAMP tem- trations were included in the present study: 64, 128, 256, plate (CAMP-t1) possessed the structural domain (N-ter- and 512 μg/ml. Following CAMP treatment, percent of vi- minal α-helix and β2-β3 loop) of AvBD-12aswellasits able cells, relative to the untreated control, was recorded. analogues A2 and A3 which appeared to be essential to the The assays were performed in triplicate. broad chemotactic activity of AvBD-12 [13]. To increase net positive charge, the negatively charged amino acid resi- Chemotaxis assay dues Asp (D) and Glu (E) were substituted with positively CAMP-induced migration of JAWSII (ATCC charged amino acid residues Lys (K) or Arg (R). To en- CRL-11904) and CHO-K1 (ATCC CCL-61) transfected hance membrane permeabilization and salt resistance, a with CCR-2 was determined using a microchemotaxis poly-Trp tail was incorporated to the C-terminus of the assay described previously [14, 31]. Chemotactic indexes peptide, as it was previously shown that coating antimicro- (C.I.) was calculated as the number of migrated cells in- bial peptides with 3 Trp residues significantly increased salt duced by CAMPs divided by the number of migrated resistance [18, 29, 35]. The resulting peptide CAMP-t1 con- cells in the control wells without CAMPs. The assays sisted of 25 amino acid residues: RKFLRRRGE- were repeated five times. VAHFSQKSLGLYCWWW. The predicted three-dimensional structure of CAMP-t1 mimics that of Protease resistance assay AvBD-12 (Fig. 1a). The second CAMP template Protease resistance was evaluated by using sodium dodecyl (CAMP-t2) consisted of the key amino acid residues of sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) AvBD-6: PIHRRIPPRWPRLKRRW, responsible for the followed by antimicrobial assays. CAMP peptides (10 μg) concentrated surface charge and hydrophobicity of AvBD-6 were treated with various proteases at various concentra- [13]. In silico analysis indicated that CAMP-t2 assumed a tions comparable to or higher than that found in host or coil and α-helical structure (Fig. 1b). To optimize anti- bacterial culture, including 0.12, 0.6, and 1.2 μg/ml of microbial activity, CAMP-t2 was subjected to further modi- α-chymotrypsin (Thermo Fisher Scientific) [25], 0.4, 4.4, fications to create CAMP-A and CAMP-B (Fig. 1c and d) and 20 μg/ml of matrilysin (metalloproteinase-7, Sigma Al- using the following criteria: 1) short in length (≤ 20 a.a. resi- drich) [32], 0.2, 2, and 20 μg/ml elastase (Thermo Fisher dues), 2) α-helical structure, 3) hydrophobic and hydro- Scientific) [33], or 0.2, 2, and 20 μg/ml of cathepsin B philic residues on opposite sides of the helical surface to (Thermo Fisher Scientific) [34]. Protease digestion assay facilitate pore formation in bacterial membrane by multiple was carried out in 20 μl of digestion buffer (25 mM Tris peptides [36], 4) lack of cutting sites for major proteases: and 150 mM NaCl, pH 7.8) for 1 h at 37 °C. After treat- aspartic protease, cysteine protease, metalloprotease, serine ment, the digestion mixture was analyzed by SDS-PAGE on proteases, 5) N-terminal acetylation and C-terminal amida- 16.5% polyacrylamide gel. To determine the effect of prote- tion to increase the metabolic stability of CAMPs [37], and ase digestion on the antimicrobial activity of two most ac- 6) addition of C-terminal poly-Trp tail to enhance mem- tive CAMPs, each peptide was first treated with a protease brane permeabilization and salt resistance [29, 35]. for 1 h at 37 °C. The mixture was then diluted to 1 × MIC of the peptide and subjected to colony count assay as de- Structural features scribed above. Assay buffer containing protease but not The amino acid sequences and relevant biochemical peptide and buffer containing untreated peptide were in- characteristics of all CAMPs were presented in Table 1. cluded as controls. Protease inhibition of the antimicrobial The structures of CAMPs was analyzed by a far-UV activity of CAMP peptide was expressed as a percentage of spectrometer. In general, the CD spectrum of α-helical killing by protease treated peptide in relevance to the un- structures presents two negative bands at 208 and treated peptide. The experiment was repeated three times 222 nm along with a positive band at 192 nm whereas in triplicate in each assay. random coil is characterized by a single band below Yang et al. BMC Microbiology (2018) 18:54 Page 5 of 14 Fig. 1 The predicted structures of newly designed templates CAMP-t1 and CAMP-t2. a The three-dimensional structure of CAMP-t1 derived from AvBD-12. Red: α-helix; Green: loop (β2-β3 loop in AvBD-12, β1-β2 loop in template CAMP-t1). b The three-dimensional structure of CAMP-t2 derived from AvBD-6. CAMP-t2 were further optimized to CAMP-A (c) and CAMP-B (d). CAMP-A and CAMP-B were coated with poly-Trp tails. Red: positively charged amino acid residues; Blue: hydrophobic amino acid residues; Green: prolines negative 200 nm [38]. As shown in Fig. 2a, CAMP-t1 Antimicrobial activity showed two weak bands at 202 nm and 225 nm, indicat- The minimum inhibitory concentrations of the newly ing partial α-helical structure of the peptide. CAMP-t2 designed CAMPs against P. aeruginosa and S. aureus displayed a strong single band around 200 nm, confirm- ATCC reference strains and multidrug-resistant P. aeru- ing the random coil structure as predicted in silico (Fig. ginosa and methicillin-resistant S. pseudinetermedius 1b). CAMP-A had two negative bands at 200 nm and (MASP) strains were compared to that of AvBD-6, a nat- 225 nm along with a positive band at 190 nm which ural host CAMP with potent antimicrobial activity under verified the predicted α-helical structure. CAMP-B low salt conditions (Table 2). CAMP-t1 showed signifi- showed two weak bands at 202 nm and 225 nm which cantly improved anti-Pseudomonas activity with MIC was consistent with the predicted partial α-helix (Fig. values 4-fold (for ATCC reference strain) and 2-fold 1d). The CAMPs were subjected to helical wheels pro- (clinical isolates) lower than that of AvBD-6. CAMP-t1 jection analysis (Fig. 2b). CAMP-t1 showed random dis- showed improved antimicrobial activity against S. aureus tribution of amino acids and CAMP-t2 showed a large ATCC reference strain (MIC was 4-fold lower than that distribution angle of hydrophilic residues (proline and of AvBD-6), but not MRSP clinical isolates as evidenced positively charged amino acids) and a small distribution by the high MIC values similar to that of AvBD-6. angle of hydrophobic residues. CAMP-A displayed an CAMP-t2, the shorter template, demonstrated similar amphipathic structure with hydrophobic and hydrophilic antimicrobial activity against Pseudomonas strains and residues on the opposite site of the helix, which sup- significantly enhanced anti-Staphylococcus activity, com- ported the designing feature (Figs. 1c and 2b). CAMP-B pared to CAP-t1. The MICs of CAMP-t2 against S. aur- exhibited a small helical wheel with hydrophobic residue eus and MRSP strains were up to 4-fold lower than that W4 in the middle of positively charged residues (Fig. 1d) of CAMP-t1 and AvBD-6. CAMP-A, a derivative of and the W9W10W11-tail folding along with the helix. CAMP-t2, established further improved antimicrobial Table 1 The characteristics of newly designed CAMPs Peptide Amino acid sequence Length (aa) Molecular Weight Charge Hydrophobicity CAMP-t1 RKFLRRRGEVAHFSQKSLGLYCWWW 25 3251.84 + 4 44% CAMP-t2 PIHRRIPPRWPRLKRRW 17 2361.90 + 7 29% CAMP-A LRRLKPLIRPWLRPLRRWWW 20 2839.55 + 7 50% CAMP-B RRRWRKRRWWW 11 1869.24 + 7 36% CAMP-t1, CAMP-t2, and CAMP-B are coated with a Trp-tail. N-terminal acetylation and C-terminal amidation were introduced to all peptides Yang et al. BMC Microbiology (2018) 18:54 Page 6 of 14 Fig. 2 The far-UV CD spectra and helical wheels projections of CAMPs. a The far-UV CD spectra of CAMPs in H2O recorded at room temperature. 2 − 1 Spectra were baseline corrected and expressed as molar ellipticity θ (deg·cm ·mol ). Gray solid line: CAMP-t1; dash line: CAMP-t2; dotted line: CAMP-A; dark solid line: CAMP-B. b Helical wheels projections of CAMPs. Relevant features of amino acid residues were coded by various shapes and colors. Hydrophilic residues: circles, hydrophobic residues: diamonds, positively charged residues: pentagons. Hydrophobic residues: greento yellow, as the hydrophobicity decreased to zero, color changed gradually from dark green to yellow. Hydrophilic residues: red, the red tone decreased proportionally to the decrease in hydrophilicity. Charged residues: light blue activity against P. aeruginosa and S. aureus reference Salt resistance strains, multidrug-resistant P. aeruginosa strains and MRSP The impact of cationic salts on the bactericidal activity clinical isolates. The MICs of CAMP-A against P. aerugi- of CAMPs was assessed using an assay system contain- nosa and MRSP isolates were 4- to 32- fold lower than that ing various concentrations of NaCl (0 to 150 mM) or of AvBD-6. CAMP-B, the second derivative of CAMP-t2, CaCl (0 to 2 mM) at two peptide concentrations (1 × also showed improved antimicrobial activity against P. aer- MIC and 0.5 × MIC). As shown in Fig. 4, increasing uginosa and similar potency against Staphylococcus spp. NaCl and CaCl concentrations had no impact on the However, CAMP-B was less effective than CAMP-A in kill- bactericidal activity of CAMP-t1 which possesses a ing both Pseudomonas spp. and methicillin-resistant S. C-terminal poly-Trp tail. In contrast, the bactericidal ac- pseudintermedius (p <0.05, Table 2). tivity of CAMP-B without the Trp tail was negatively af- fected by increased salt concentrations. At the Membrane permeabilizing activity physiological conditions of NaCl (100 to 150 mM) and A propidium iodide (PI) uptake assay was carried out to CaCl (1 to 2 mM), CAMP-t2 retained approximately determine the membrane permeabilizing activity of newly 40% of its killing activity against P. aeruginosa and S. designed CAMPs (Fig. 3). As shown in Fig. 3a, aureus, compared to the results obtained at salt-free CAMP-treated P. aeruginosa were stained red, indicating condition (Fig. 4). The salt-resistance pattern of that bacterial membranes were damaged by CAMPs. In CAMP-t2 was similar at two different peptide concen- contrast, untreated bacteria did not show red fluores- trations. Similar to CAMP-t1, CAMP-A and CAMP-B cence. When the permeabilizing ability was assessed based with a Trp tail exhibited strong tolerance to NaCl and on the numbers of red cells per field, a time-dependent in- CaCl (Fig. 4). crease was observed from 15 to 90 min (Fig. 3b and c). However, the number of red bacteria did not increase sig- Hemolytic activity and cytotoxicity nificantly after 30 min, indicating a fast-action mode of The hemolytic activity of CAMPs to mouse RBCs was CAMPs. Similar results were obtained for all CAMPs and analyzed (Fig. 5). At high concentrations, CAMP-A lysed both bacterial pathogens (Fig. 3). approximately 3.6% of mRBCs at 128 μg/ml, 10.1% of Yang et al. BMC Microbiology (2018) 18:54 Page 7 of 14 Table 2 The minimum inhibitory concentration (MIC) of CAMPs CAMPs CAMP-t1 CAMP-t2 CAMP-A CAMP-B AvBD-6 Bacteria MIC (μg/ml) MIC (μg/ml) MIC (μg/ml) MIC (μg/ml) MIC (μg/ml) (strain identification) P. aeruginosa (ATCC 27853) 64 64 16 32 > 256 P. aeruginosa (1704173) 128 64 16 32 > 256 P. aeruginosa (1703357) 128 128 16 64 > 256 P. aeruginosa (1703511) 128 128 16 64 > 256 P. aeruginosa (1703000) 128 64 8 32 > 256 P. aeruginosa (1703002) 128 256 16 64 > 256 P. aeruginosa (1703451) 128 64 16 64 > 256 P. aeruginosa (1703949) 128 128 16 64 > 256 P. aeruginosa (1703290) 128 128 16 64 > 256 P. aeruginosa (1703983) 128 256 16 64 > 256 P. aeruginosa (1704175) 128 64 16 32 > 256 a a c b MIC 122.18 ± 19.29 122.18 ± 72.71 15.27 ± 2.41 52.36 ± 16.14 > 256 Average G- S. aureus (ATCC 29213) 64 64 32 32 256 S. pseudintermedius(13164006) 256 32 16 32 > 256 S. pseudintermedius(13203008) 256 32 16 64 256 S. pseudintermedius(13178007) 256 64 16 64 256 S. pseudintermedius(13267017) 128 32 16 32 > 256 S. pseudintermedius(13252001) 64 32 16 32 256 S. pseudintermedius(13269013) 256 32 32 32 > 256 S. pseudintermedius(13193006) 256 128 32 32 > 256 S. pseudintermedius(13228005) 256 32 16 64 > 256 S. pseudintermedius(13207007) 256 32 16 64 > 256 S. pseudintermedius(13250111) 256 32 32 64 256 a b c b MIC 209.45 ± 81.41 46.55 ± 29.89 21.82 ± 8.07 46.55 ± 16.71 ≥256 Average G+ Multidrug resistant clinical isolates of P. aeruginosa were resistant to chloramphenicol, tetracycline, sulfamethoxazole, and β-lactam antibiotics: amoxicillin, ampicillin and cefazolin. Methicillin-resistant S. pseudintermedius (MRSP) Superscripts a, b, and c mean significant difference (p < 0.05) between MICs of four CAMPs against either Gram-negative or Gram-positive bacteria mRBCs at 256 μg/ml, and 17.5% of mRBCs at 512 μg/ treatment with CAMP-A at a concentration equal to or ml. CAMP-t1, CAMP-t2, and CAMP-B did not cause greater than 128 μg/ml. After treatment with CAMP-A, more than 5% of mRBCs at the concentration of 512 μg/ no significant difference in the percentage of cell viabil- ml. The minimum hemolytic concentration (MHC), geo- ity was observed between 4 h treatment and 48 h metric means of the MIC (MIC ), and therapeutic treatment. GM index (T.I.) were determined for each CAMP (Table 3). The average T.I. of CAMP-A and CAMP-B against P. Chemotactic activity aeruginosa were 8.72 ± 2.41 and > 10.90 ± 4.0, respect- The chemotactic activity of the CAMPs for JAWSII and ively. The T.I. of CAMP-A and CAMP-B against S. aur- CCR2-transfected CHO-K1 cells were determined eus were 6.54 ± 2.01 and > 12.36 ± 4.18, respectively. (Fig. 7). The results indicated that CAMP-t1 with the Cell cytotoxicity of the newly designed CAMPs was N-terminal helix-loop structure of AvBD-12 did not also determined by MTT cell viability assay (Fig. 6). Ex- show expected chemotactic activity to either cell line posure of murine immature dendritic cell line JAWSII (Fig. 7a and b). Interestingly, CAMP-A with the highest (Fig. 6a) and CHO-K1 (Fig. 6b) cells to CAMP-t1, antimicrobial activity showed mild chemotactic activity CAMP-t2, and CAMP-B at concentrations of 64, 128, at a concentration of 64 μg/ml (C.I. = 5.13; 77.5% of 256, and 512 μg/ml for 4 to 48 h did not significantly wild-type AvBD-12, C.I. = 6.62) for JAWSII cells. As affect cell viability. However, the viability of both JAW- shown in Fig. 7c, more JAWSII cell migration was in- SII and CHO-K1 cells was significantly decreased by duced by CAMP-A with increasing peptide Yang et al. BMC Microbiology (2018) 18:54 Page 8 of 14 Fig. 3 Membrane permeabilizing activity of CAMPs. a Representative fluorescence microscopy images of CAMP-treated and control bacteria stained with membrane-impermeable DNA dye propidium iodide (PI). Right panels are enlarged focal areas (a and b) from the left panels. b The number per field of positively stained P. aeruginosa at various times post-CAMP-treatment. c The number per field of S. aureus at various times post-CAMP-treatment. Data are expressed as the means ± SD of three independent experiments. An asterisk indicates a significant difference between different time points (*p < 0.05). Bar: 100 μm concentrations, ranging from 1 to 64 μg/ml. No signifi- and cathepsin B at concentrations up to 20 μg/ml. cant chemotactic activity for CCR-2 transfected Treatment with these proteases did not affect the anti- CHO-K1 cells was detected (Fig. 7b). CAMP-t2 and microbial activity of CAMP-A and CAMP-B (data not CAMP-B did not show any chemotactic activity for ei- shown). ther JAWSII or CCR2-CHO-K1 cells. Discussion Protease resistance Host cationic antimicrobial peptides, such as defensins, The resistance of CAMP-A and CAMP-B, two peptides have been a subject of research interest because of their exhibiting strong antimicrobial activity and broad-spectrum antimicrobial activity and low potential salt-resistance, to various proteases was evaluated by for resistance development. For instance, human subjecting protease-treated peptides to SDS-PAGE α-defensin HD5 and HDP4 showed strong killing activity (Fig. 8). The results indicated that CAMP-A and against S. aureus ATCC 25923 and ATCC 29213 with CAMP-B were partially digested by α-chymotrypsin at the lethal doses [39]. Our previous structure-function 0.12 to 1.2 μg/ml, as indicated by the presence of pep- analysis of AvBDs and their analogues indicates that the tide bands with lower molecular weight than that of the highly concentrated surface positive charge plays a pre- untreated peptides (Fig. 8a and b). The antimicrobial ac- dominant role in the antimicrobial potency of the pep- tivity of CAMP-A and CAMP-B against P. aeruginosa tides whereas the CCR2-binding domain (N-terminal decreased significantly post-digestion (Fig. 8c). The α-helix along with an adjacent loop) is responsible for digested peptides retained approximately 90% (at the broad-spectrum chemotactic activity for both avian 0.12 μg/ml), 80% (at 0.66 μg/ml), and 30% (at 1.2 μg/ml) and mammalian dendritic cells [13]. For example, linear of the killing activity of the untreated CAMP-A or AvBD analogues with a high net positive charge (+ 9), CAMP-B. In contrast, the anti-Staphylococcus activity modest hydrophobicity (40%), and a predicted CCR2 was mildly affected only when the CAMPs were treated binding domain exhibit strong antimicrobial and mild with the highest concentration (1.2 μg/ml) of chemotactic activities. However, the linear peptides de- α-chymotrypsin (Fig. 8d). CAMP-A and CAMP-B were signed in our previous study are still lengthy (45 amino not cleaved by metalloproteinases matrilysin, elastase, acid residues) and susceptible to physiological Yang et al. BMC Microbiology (2018) 18:54 Page 9 of 14 Fig. 4 Effect of salts on the antibacterial activity of CAMPs against P. aeruginosa and S. aureus. The effect of salts on the antibacterial activity was determined using two peptide concentrations: 1 × MIC and 0.5 × MIC. a Percent of killing against P. aeruginosa at 0, 50, 100, and 150 mM NaCl; (b) Percent of killing against S. aureus at 0, 50, 100, and 150 mM NaCl; (c) Percent of killing against S. aureus at 0, 0.5, 1, and 2 mM CaCl . d Percent of killing against S. aureus at 0, 0.5, 1, and 2 mM CaCl . Data represent the means ± SD of three independent experiments. An asterisk indicates the statistically significant difference between antimicrobial activity in the presence and absence of salts (*p < 0.05 and **p < 0.01) concentrations of NaCl. Although the sensitivity of achieve the following goals: structurally simple (linear, AvBD analogues to proteases was not evaluated in our short and all natural amino acids), resistant to proteases previous studies, investigations conducted by others sug- and cationic salts, non-cytotoxic, strong and gest that linear peptides are susceptible to bacterial broad-spectrum antimicrobial activity, and potentially metalloprotease, cysteine protease and human neutro- chemotactic for immune cells. phil elastase [38, 39]. To develop antimicrobial peptides We first designed two CAMP templates, CAMP-t1 suitable for therapeutic or preventive use, we utilized an and CAMP-t2, by extrapolating the CCR2 binding do- integrated approach to modify AvBD analogues to main of AvBD-12 (CAMP-t1) and key amino acid Fig. 5 Hemolytic activity of CAMPs. CAMP-induced hemolysis (%) of mouse red blood cells at various peptide concentrations is defined as a percentage of complete hemolysis caused by 0.2% Triton X-100. Data are expressed as the means ± SD of three independent experiments Yang et al. BMC Microbiology (2018) 18:54 Page 10 of 14 Table 3 Therapeutic index of new cationic antimicrobial peptides (CAMPs) Peptides CAMP-t1 CAMP-t2 CAMP-A CAMP-B MHC > 512 > 512 128 > 512 MIC 122.18 ± 19.29 122.18 ± 72.71 15.27 ± 2.41 52.36 ± 16.14 Average G- MIC 209.45 ± 81.41 46.55 ± 29.89 21.82 ± 8.07 46.55 ± 16.71 Average G+ T.I. of G- > 4.36 ± 1.21 > 5.45 ± 2.54 8.72 ± 2.41 > 10.90 ± 4.0 T.I. of G+ > 3.27 ± 2.41 > 13.45 ± 4.48 6.54 ± 2.01 > 12.36 ± 4.18 Therapeutic index, (T.I.) is defined as the ratio of, MHC to, MIC . MHC (μg/ml) is the minimum hemolytic concentration that caused 5% hemolysis of mouse red GM blood cells, (mRBCs). The MIC (μg/ml) means the geometric mean, (GM) of the, MIC values of the peptides against bacteria. MIC is the, MIC for Gram- GM Average G- GM negative bacteria, MIC is the MIC for Gram-positive bacteria Average G+ GM residues contributing to the concentrated surface posi- degradation [40]. To evaluate the antimicrobial proper- tive charge and hydrophobicity of AvBD-6 (CAMP-t2), ties of these peptides, we determined their MICs against respectively. For CAMP-A, the negatively charged amino P. aeruginosa and Staphylococcus according to the acid residues (D and E) in AvBD-12 were replaced by guidelines of CLSI [27, 28]. Both CAMP-t1 and positively charged residues (K and R). Because the net CAMP-t2 demonstrated improved antimicrobial activity, positive charge of CAMP-t1 was still relatively low (+ 4), compared to AvBD-6 and AvBD-12 as well as previously a poly-Trp tail was incorporated into its C-terminus to designed AvBD analogues [13, 14]. Although CAMP-t1 boost the antimicrobial activity. Trp is known for its ten- retained the N-terminal α-helix and an adjacent loop dency to insert into membrane lipid bilayer and structure of AvBD-12, it lost the desired chemotactic Trp-rich peptides exhibit enhanced antimicrobial activity property [13], suggesting that either the amino acid and salt resistance [18]. N-terminal acetylation and composition was not optimal or additional structural C-terminal amidation (mimicking native proteins) were components were required for CCR2 binding. We then also incorporated to increase the metabolic stability of focused on CAMP-t2, a shorter template with a peptides as well as their resistance to enzymatic coil-helix structure. This peptide showed stronger Fig. 6 Cytotoxicity of CAMPs to JAWSII and CHO-K1 cells. Effect of CAMPs on the viability of mouse immature dendritic JAWSII cells (a) and hamster ovary CHO-K1 cells (b) at 4 and 48 h of incubation with peptide at the concentration of 64 to 512 μg/ml. Results are percentages of viable cells relative to the untreated control cells. The data are expressed as the mean ± SD of three independent experiments. An asterisk indicates a statistically significant difference in the viability of CAMP-treated cells and untreated cells (*p < 0.05 and **p < 0.01) Yang et al. BMC Microbiology (2018) 18:54 Page 11 of 14 Fig. 7 Chemotactic activity of CAMPs. CAMP-induced migration of mouse immature dendritic JAWSII cells (a) and CHO-K1 cells expressing avian CCR2 (b) was measured at the following peptide concentrations: 1, 4, 16, and 64 μg/ml. c Migrated JAWSII cells on the membrane induced by CAMP-A at peptide concentrations of 1, 4, 16, and 64 μg/ml. White pores are membrane pores (diameter: 8 μm) and blue ones are stained JAWSII cells. Bar: 50 μm. Chemotactic index (C.I.) was expressed as the number of migrated cells induced by CAMP / the number of migrated cells in response to chemotactic buffer. Data represent the means of five independent experiments ± SD. An asterisk indicates significant difference (*p < 0.05 and **p < 0.01) antimicrobial activity against methicillin-resistant S. methicillin-resistant S. pseudintermedius clinical isolates. pseudintermedius than CAMP-t1 and AvBD-6, but was With a poly-Trp tail, α-helical structure and increased sensitive to high concentrations of cationic salts. The surface positive charge, CAMP-A and CAMP-B were rest of the study was concentrated primarily on improv- fully functional at physiological concentrations of NaCl ing the antimicrobial property and salt resistance of and CaCl . These peptides were also resistant to metal- CAMP-t2. loproteinases, matrilysin and elastase, and cathepsin B at Studies have indicated that amphipathicity is a key concentrations higher than that in bacterial protein se- characteristic required for membrane permeabilization cretion [42] or in mammalian host cells [34]. Although in which hydrophobic residues interact with membrane they were still cleaved by α-chymotrypsin, the antimicro- lipid components while hydrophilic regions either bind bial activity was minimally affected at the concentration with the phospholipid head groups or form the lumen of of 0.12 μg/ml, about 3 to 40 times higher than the con- a membrane pore [5, 14]. Alpha-helical peptides with centration tested in human samples using different hydrophobic and hydrophilic residues on opposite sides methods, 4 ng/ml [43] and 37.5 ng/ml [44]. At a high of the peptide molecule have antimicrobial property [8]. concentration (1.2 μg/ml), α-chymotrypsin treatment re- The shorter template, CAMP-t2 with coil-helix structure duced the killing activity against P. aeruginosa (p < 0.05) (only 35% of residues form α-helix), was further modi- but not S. aureus (p > 0.05). The discrepancy could be fied to form an α-helix structure which confers struc- associated with structural difference between tural stability [41]. To maximize the antimicrobial Gram-negative and Gram-positive bacterial membranes activity and minimize the damaging effect on host cell which warrants further investigation into the mechanism membrane, Trp and Pro residues were incorporated and of antimicrobial action of these peptides. the amino acid residues were strategically arranged to Our previous studies have shown that AvBDs could avoid protease cutting sites predicted using online disrupt bacterial membrane resulting in cell deform- PROSPER and SignalP 4.1 servers. The resulting pep- ation, increased membrane permeabilization, and mem- tides, CAMP-A and CAMP-B, demonstrated strong anti- brane damage [13, 14]. In the present study, data from microbial activity against ATCC bacterial reference propidium iodide (PI) staining assay suggested that the strains as well as multi-drug resistant P. aeruginosa and primary mode of action of the newly designed CAMPs Yang et al. BMC Microbiology (2018) 18:54 Page 12 of 14 Fig. 8 Effects of protease treatment on the antimicrobial activity of CAMP-A and CAMP-B against P. aeruginosa and S. aureus. Peptides were digested with α-chymotrypsin, elastase, matrilysin, or cathepsin B for 1 h at 37 °C and subjected to SDS-PAGE (16.5% polyacrylamide gel) analysis. a CAMP-(a). b CAMP-(b). c Antimicrobial activity of CAMP-A post-digestion by α-chymotrypsin at the concentration of 0.12, 0.66, and 1.2 μg/ml. d Antimicrobial activity of CAMP-B post-digestion by α-chymotrypsin at the concentration of 0.12, 0.66, and 1.2 μg/ml. Results are expressed as percent killing by digested peptides over untreated peptides. Data are expressed as the means ± SD of three independent experiments. An asterisk indicates a statistically significant difference between antimicrobial activity with and without protease (*p < 0.05 and **p < 0.01) was membrane attacking, which is considered a mechanism to show chemotactic activity. Interestingly, CAMP-A less likely to trigger bacterial resistance. The CAMPs did not with high positive charge and modest hydrophobicity in- show any detectable cytotoxicity and hemolytic activity at duced chemotactic migration of JAWSII cells which oc- the doses required for effective bacterial killing. CAMP-t1, curred possibly via the formyl-peptide receptors like CAMP-t2, and CAMP-B had minimal cytotoxic and mechanism such reported for human cathelicidin LL-37 hemolytic activities at a relatively high peptide concentration [50] and cathelicidin-like pleurocidins [51]. (512 μM/ml). CAMP-A, the most potent antimicrobial pep- tide, exhibited hemolytic and cytotoxic activities at concen- Conclusion trations equal to or greater than 128 μg/ml which, however, CAMP-t1 and CAMP-t2 were designed as templates was 6-fold higher than the MIC against P. aeruginosa and based on key structural and functional components of 4-fold higher than the MIC against Staphylococcus spp. The AvBD-12 and AvBD-6. CAMP-t1 with a predicted CCR cytotoxic property of CAMP-A was not surprising because binding domain of AvBD-12 demonstrated improved the peptide had a relatively high hydrophobicity (50%), antimicrobial activity but lost the original chemotactic which is known hydrophobicity is associated with their cyto- function. CAMP-t2 with key amino acid residues of toxic effect [45]. The undesired hemolytic activity was still AvBD-6 showed strong antimicrobial activity, but sensi- mild compared to other antimicrobial peptides including tivity to high concentrations of cationic salts. CAMP-t2 magainin isolated from the skin of African frog Xenopus lae- was further modified using an integrated design ap- vis and melittin from bee venom [46]. proach. CAMP-A and CAMP-B possess the following It has been suggested that the three conserved disul- advantageous characteristics: structural simplicity (short fide bridges were required for the chemotactic function and linear), resistance to salts and proteases, potent anti- of β-defensins [14, 47–49]. Data from our previous study microbial activity against multidrug-resistant P. aerugi- indicated that a predicted CCR2 binding domain (N-ter- nosa and methicillin-resistant Staphylococcus, rapid minal α-helix and an adjacent β2-β3 loop) in membrane attacking mode, and moderate therapeutic AvBD-12A3 (a linear peptide) without disulfide bridges index. Our data suggest that CAMP-A and CAMP-B are was chemotactic to JAWSII cells [13]. In the present excellent candidates for development as antimicrobial study, CAMP-t1 with a similar helix-loop domain failed therapeutic agents. Yang et al. BMC Microbiology (2018) 18:54 Page 13 of 14 Abbreviations 8. Bahar A, Ren D. Antimicrobial peptides. Pharmaceuticals. 2013;6(12):1543. ANOVA: One-way analysis of variance; AvBD: Avian beta-defensin; 9. Zhang G, Sunkara LT. Avian antimicrobial host defense peptides: from BSA: Bovine serum albumin; CFU: Colony-forming unit; CI: Chemotaxis index; biology to therapeutic applications. Pharmaceuticals (Basel). 2014;7:220–47. DCs: Dendritic cells; ESI-MS: Electrospray ionization mass spectrometry; 10. Greber KE, Dawgul M. Antimicrobial peptides under clinical trials. Curr Top FBS: Fetal bovine serum; fMLF: N-Formyl-methionyl-leucyl-phenylalanine; Med Chem. 2017;17(5):620–8. Fmoc: 9-fluorenyl-methoxycarbonyl; GM-CSF: Granulocyte macrophage 11. Mahlapuu M, Håkansson J, Ringstad L, Björn C. Antimicrobial peptides: an colony-stimulating factor; hBD6: Human β-defensin 6; LB: Luria-Bertani; emerging category of therapeutic agents. Front Cell Infect Microbiol. 2016;6:194. LPS: Lipopolysaccharides; LTA: Lipoteichoic acid; MIC: Minimum inhibitory 12. Zhao L, Yang M, Zhang M, Zhang S. Expression, purification, and in vitro concentration; PBS: Phosphate buffered saline; PI: Propidium iodide; RP- comparative characterization of avian beta-defensin-2, −6, and −12. Avian HPLC: Reversed-phase high-performance liquid chromatography; Dis. 2014;58(4):541–9. SD: Standard deviation 13. Yang M, Zhang C, Zhang MZ, Zhang S. Novel synthetic analogues of avian β-defensin-12: the role of charge, hydrophobicity, and disulfide bridges in Acknowledgments biological functions. BMC Microbiol. 2017;17(1):43. The authors would like to thank the bacteriology laboratory of Veterinary 14. Yang M, Zhang C, Zhang X, Zhang MZ, Rottinghaus GE, Zhang S. Structure- Medical Diagnostic Laboratory of University of Missouri for sharing the multi- function analysis of avian beta-defensin-6 and beta-defensin-12: role of drug resistant bacterial isolates. The authors thank Dr. Michael T. Henzl for charge and disulfide bridges. BMC Microbiol. 2016;16:210. the help with CD spectra analysis. The authors also wish to thank Ms. Chris- 15. Mohamed MF, Abdelkhalek A, Seleem MN. Evaluation of short synthetic tine Yan Zhang for editorial assistance. antimicrobial peptides for treatment of drug-resistant and intracellular Staphylococcus aureus. Sci Rep. 2016;6:29707. Funding 16. Rozek A, Powers JP, Friedrich CL, Hancock RE. Structure-based design of an The present study was funded by the College of Veterinary Medicine of the indolicidin peptide analogue with increased protease stability. Biochemistry. University of Missouri for the design of the study and collection, analysis, and 2003;42(48):14130–8. interpretation of data and in writing the manuscript. 17. Giangaspero A, Sandri L, Tossi A. Amphipathic alpha helical antimicrobial peptides. Eur J Biochem. 2001;268(21):5589–600. Availability of data and materials 18. Yu HY, Tu CH, Yip BS, Chen HL, Cheng HT, Huang KC, Lo HJ, Cheng JW. The datasets used and/or analysed during the current study are available Easy strategy to increase salt resistance of antimicrobial peptides. from the corresponding author on reasonable request. Antimicrob Agents Chemother. 2011;55(10):4918–21. 19. Mohanram H, Bhattacharjya S. Resurrecting inactive antimicrobial peptides from Authors’ contributions the lipopolysaccharide trap. Antimicrob Agents Chemother. 2014;58(4):1987–96. Conceived and designed the experiments, MY, CZ, MZ, SZ performed the 20. Jerala R. Synthetic lipopeptides: a novel class of anti-infectives. Expert Opin experiments, MY, CZ. Analyzed the data, MY, CZ, MZ, SZ contributed Investig Drugs. 2007;16(8):1159–69. reagents/materials/analysis tools, MY, SZ drafted the manuscript, MY, CZ, MZ, 21. Chu-Kung AF, Nguyen R, Bozzelli KN, Tirrell M. Chain length dependence of SZ. All authors read and approved the manuscript. antimicrobial peptide-fatty acid conjugate activity. J Colloid Interface Sci. 2010;345(2):160–7. Ethics approval and consent to participate 22. Vagner J, Qu H, Hruby VJ. Peptidomimetics, a synthetic tool of drug Not applicable. discovery. Curr Opin Chem Biol. 2008;12(3):292–6. 23. Ringstad L, Kacprzyk L, Schmidtchen A, Malmsten M. Effects of topology, Competing interests length, and charge on the activity of a kininogen-derived peptide on lipid The authors declare that they have no competing interests. membranes and bacteria. Biochim Biophys Acta. 2007;1768(3):715–27. 24. Saravanan R, Li X, Lim K, Mohanram H, Peng L, Mishra B, Basu A, Lee JM, Bhattacharjya S, Leong SS. Design of short membrane selective Publisher’sNote antimicrobial peptides containing tryptophan and arginine residues for Springer Nature remains neutral with regard to jurisdictional claims in improved activity, salt-resistance, and biocompatibility. Biotechnol Bioeng. published maps and institutional affiliations. 2014;111(1):37–49. 25. Kim H, Jang JH, Kim SC, Cho JH. De novo generation of short antimicrobial Author details peptides with enhanced stability and cell specificity. J Antimicrob Department of Veterinary Pathobiology, College of Veterinary Medicine, Chemother. 2014;69(1):121–32. University of Missouri, Columbia, MO 65211, USA. Department of Biomedical 26. Tyagi P, Singh M, Kumari H, Kumari A, Mukhopadhyay K. Bactericidal activity Science, College of Veterinary Medicine, University of Missouri, Columbia, MO of curcumin I is associated with damaging of bacterial membrane. PLoS 65211, USA. Veterinary Medical Diagnostic Laboratory, College of Veterinary One. 2015;10(3):e0121313. Medicine, University of Missouri, Columbia, MO 65211, USA. 27. CLSI. M07-A8 methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically; approved standard. 8th ed; 2009. Received: 26 September 2017 Accepted: 18 May 2018 28. CLSI: M100-S22 performance standards for antimicrobial susceptibility testing; twenty-second informational supplement. 2012. 29. Chu HL, Yu HY, Yip BS, Chih YH, Liang CW, Cheng HT, Cheng JW. Boosting References salt resistance of short antimicrobial peptides. Antimicrob Agents 1. Donoghue DJ. Antibiotic residues in poultry tissues and eggs: human health Chemother. 2013;57(8):4050–2. concerns? Poult Sci. 2003;82(4):618–21. 30. Zhang SK, Song JW, Gong F, Li SB, Chang HY, Xie HM, Gao HW, Tan YX, Ji 2. Ventola CL. The antibiotic resistance crisis: part 1: causes and threats. P T. SP. Design of an α-helical antimicrobial peptide with improved cell-selective 2015;40(4):277–83. and potent anti-biofilm activity. Sci Rep. 2016;6:27394. 3. Mohanram H, Bhattacharjya S. Salt-resistant short antimicrobial peptides. 31. Falk W, Goodwin RH Jr, Leonard EJ. A 48-well micro chemotaxis assembly Biopolymers. 2016;106(3):345–56. 4. Straus SK, Hancock REW. Mode of action of the new antibiotic for gram- for rapid and accurate measurement of leukocyte migration. J Immunol Methods. 1980;33(3):239–47. positive pathogens daptomycin: comparison with cationic antimicrobial peptides and lipopeptides. Biochim Biophys Acta Biomembr. 2006;1758(9): 32. Kuhlmann KF, van Till JW, Boermeester MA, de Reuver PR, Tzvetanova ID, Offerhaus 1215–23. GJ,Ten Kate FJ, Busch OR, van Gulik TM,Gouma DJ, CrawfordHC. Evaluation of 5. Jenssen H, Hamill P, Hancock RE. Peptide antimicrobial agents. Clin matrix metalloproteinase 7 in plasma and pancreatic juice as a biomarker for Microbiol Rev. 2006;19(3):491–511. pancreatic cancer. Cancer Epidemiol Biomark Prev. 2007;16(5):886–91. 6. Yeaman MR, Yount NY. Mechanisms of antimicrobial peptide action and 33. Forde E, Humphreys H, Greene CM, Fitzgerald-Hughes D, Devocelle M. resistance. Pharmacol Rev. 2003;55(1):27–55. Potential of host defense peptide prodrugs as neutrophil elastase- 7. Ganz T. Defensins: antimicrobial peptides of innate immunity. Nat Rev dependent anti-infective agents for cystic fibrosis. Antimicrob Agents Immunol. 2003;3(9):710–20. Chemother. 2014;58(2):978–85. Yang et al. BMC Microbiology (2018) 18:54 Page 14 of 14 34. Doxakis A, Maria A, Savvas P, Zafiroula I. Assessment of the roles of cathepsins B, H and L in the progression of colorectal cancer. Journal Cancer Therapy. 2013;4:1–7. 35. Hansen A, Schafer I, Knappe D, Seibel P, Hoffmann R. Intracellular toxicity of proline-rich antimicrobial peptides shuttled into mammalian cells by the cell-penetrating peptide penetratin. Antimicrob Agents Chemother. 2012; 56(10):5194–201. 36. Brogden KA. Antimicrobial peptides: pore formers or metabolic inhibitors in bacteria? Nat Rev Microbiol. 2005;3(3):238–50. 37. Nguyen LT, Chau JK, Perry NA, de Boer L, Zaat SA, Vogel HJ. Serum stabilities of short tryptophan- and arginine-rich antimicrobial peptide analogs. PLoS One. 2010;5(9):e12684. 38. Gopal R, Park JS, Seo CH, Park Y. Applications of circular dichroism for structural analysis of gelatin and antimicrobial peptides. Int J Mol Sci. 2012; 13(3):3229–44. 39. Ericksen B, Wu Z, Lu W, Lehrer RI. Antibacterial activity and specificity of the six human {alpha}-defensins. Antimicrob Agents Chemother. 2005;49(1):269–75. 40. Bulet P, Stocklin R, Menin L. Anti-microbial peptides: from invertebrates to vertebrates. Immunol Rev. 2004;198:169–84. 41. Park IY, Cho JH, Kim KS, Kim YB, Kim MS, Kim SC. Helix stability confers salt resistance upon helical antimicrobial peptides. J Biol Chem. 2004;279(14): 13896–901. 42. Shinnar AE, Butler KL, Park HJ. Cathelicidin family of antimicrobial peptides: proteolytic processing and protease resistance. Bioorg Chem. 2003;31(6): 425–36. 43. Lefkowitz RB, Schmid-Schonbein GW, Heller MJ. Whole blood assay for elastase, chymotrypsin, matrix metalloproteinase-2, and matrix metalloproteinase-9 activity. Anal Chem. 2010;82(19):8251–8. 44. Iwaki K, Ogawa M, Tanaka S, Kosaki G. Radioimmunoassay for human pancreatic chymotrypsin and measurement of serum immunoreactive chymotrypsin contents in various diseases. Res Commun Chem Pathol Pharmacol. 1983;40(3):489–96. 45. Mora-Navarro C, Méndez-Vega J, Caraballo-León J, Lee M-r, Palecek S, Torres-Lugo M, Ortiz-Bermúdez P. Hydrophobicity of antifungal β-peptides is associated with their cytotoxic effect on in vitro human colon Caco-2 and liver HepG2 cells. PLoS One. 2016;11(3):e0149271. 46. Matsuzaki K, Sugishita K-i, Harada M, Fujii N, Miyajima K. Interactions of an antimicrobial peptide, magainin 2, with outer and inner membranes of gram- negative bacteria. Biochim Biophys Acta Biomembr. 1997;1327(1):119–30. 47. Klüver E, Schulz-Maronde S, Scheid S, Meyer B, Forssmann WG, Adermann K. Structure-activity relation of human beta-defensin 3: influence of disulfide bonds and cysteine substitution on antimicrobial activity and cytotoxicity. Biochemistry. 2005;44(28):9804–16. 48. Hoover DM, Wu Z, Tucker K, Lu W, Lubkowski J. Antimicrobial characterization of human beta-defensin 3 derivatives. Antimicrob Agents Chemother. 2003;47(9):2804–9. 49. Schroeder BO, Wu Z, Nuding S, Groscurth S, Marcinowski M, Beisner J, Buchner J, Schaller M, Stange EF, Wehkamp J. Reduction of disulphide bonds unmasks potent antimicrobial activity of human beta-defensin 1. Nature. 2011;469(7330):419–23. 50. Tjabringa GS, Ninaber DK, Drijfhout JW, Rabe KF, Hiemstra PS. Human cathelicidin LL-37 is a chemoattractant for eosinophils and neutrophils that acts via formyl-peptide receptors. Int Arch Allergy Immunol. 2006;140(2): 103–12. 51. Pundir P, Catalli A, Leggiadro C, Douglas SE, Kulka M. Pleurocidin, a novel antimicrobial peptide, induces human mast cell activation through the FPRL1 receptor. Mucosal Immunol. 2014;7(1):177–87.

Journal

BMC MicrobiologySpringer Journals

Published: Jun 5, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off