Cun-Bao LIU, Bin SHAN, Hong-Mei BAI, Jing TANG, Long-Zong YAN, Yan-Bing MA,*

?

Hydrophilic/hydrophobic characters of antimicrobial peptides derived from animals and their effects on multidrug resistant clinical isolates

Cun-Bao LIU1, Bin SHAN2, Hong-Mei BAI1, Jing TANG3, Long-Zong YAN4, Yan-Bing MA1,*

1Yunnan Key Laboratory of Vaccine Research and Development on Severe Infectious Diseases, Institute of Medical Biology, Chinese Academy of Medical Sciences and Peking Union Medical College, Kunming Yunnan 650118, China2Department of Clinical Lab, the First Affiliated Hospital of Kunming Medical University, Kunming Yunnan 650032, China3Department of Biochemistry and Molecular Biology, Kunming Medical University, Kunming Yunnan 650500, China4Department of Burn, the Second Affiliated Hospital of Kunming Medical University, Kunming Yunnan 650101, China

Multidrug resistant (MDR) pathogen infections are serious threats to hospitalized patients because of the limited therapeutic options. A novel group of antibiotic candidates, antimicrobial peptides (AMPs), have recently shown powerful activities against both Gram-negative and Gram-positive bacteria. Unfortunately, the viability of using these AMPs in clinical settings remains to be seen, since most still need to be evaluated prior to clinical trials and not all of AMPs are potent against MDR clinical isolates. To find a connection between the characteristics of several of these AMPs and their effects against MDR pathogens, we selected 14 AMPs of animal origin with typical structures and evaluated theiractivities against clinical strains of extensive drug-resistant, methicillin-resistant, extended spectrum β-lactamase-producingand extended spectrum β-lactamase-producingOur results showed that these peptides’ hydrophilic/hydrophobic characteristics, rather than their secondary structures, may explain their antibacterial effects on these clinical isolates. Peptides that are amphipathic along the longitudinal direction seemed to be effective against Gram-negative pathogens, while peptides with hydrophilic terminals separated by a hydrophobic intermediate section appeared to be effective against both Gram-negative and Gram-positive pathogens. Among these, cathelicidin-BF was found to inhibit all of the Gram-negative pathogens tested at dosages of no more than 16 mg/L, killing a pandrug-resistantstrain within 2 h at 4×MICs and 4 h at 2×MICs. Tachyplesin III was also found capable of inhibiting all Gram-negative and Gram-positive pathogens tested at no more than 16 mg/L, and similarly killed the samestrain within 4 h at 4×MICs and 2×MICs. These results suggest that both cathelicidin-BF and tachyplesin III are likely viable targets for the development of AMPs for clinical uses.

Hydrophilic/hydrophobic character; Multidrug resistant clinical isolate; Cathelicidin-BF; Tachyplesin III

INTRODUCTION

Multidrug resistant (MDR) pathogens continue to pose serious threats to hospitalized patients because there are limited effective therapies capable of combatting the infection. Among these pathogens,is particularly intractable because several of its clinical isolates have gained resistance to nearly all currently available antibiotics, leading it to be described as “extensive drug resistant” (XDR) (Falagas & Karageorgopoulos, 2008). Though work on developing novel cocktails of antibiotics that can be used in tandem may help in the short-run, a long-term solution is urgently needed.

One promising candidate source of novel antibiotic treatments, antimicrobial peptides (AMPs), have gained increased attention.[1]These AMPs, which are innate immune molecules that are widely distributed among animals, have previously exhibited powerful killing effects to both Gram-negative and Gram-positive bacteria. Unfortunately, clinical applications of these AMPs have made relatively little progress over the last decade and have been rarely reported (Hancock & Sahl, 2006; Lipsky et al, 2008; Vaara, 2009). The reasons for this lack of progress are multifaceted; aside from the instability of these peptides, three further reasons warrant some explanation. First, not all of the currently AMPs reported are potent AMPs. Traditionally, AMPs are purified directly from tissues following their activities, which may assure their antimicrobial activities. Currently, a growing number AMPs are simply predicted via bioinformatics, and then sometimes synthesized as peptides to test for antimicrobial activities. Typically, these synthetic AMPs are found to exhibit comparatively weak antimicrobial effects. Second, the structures of AMPs are highly diversified. It is nearly impossible to test the antimicrobial activities of these AMPs one by one, largely due to the intensive time and financial requirements. To get around this obstacle, it is usually necessary to select typical AMPs and then estimate the antibacterial activities of their corresponding groups, of which four predominate: helix, beta-sheet formed with 2-3 disulfide bridges, linear peptides rich in special amino acids and loop peptides formed by one disulfide bridge (Vaara, 2009). Third, the primary laboratory screening procedure of AMPs usually involves testing their antibacterial activities on several type culture strains (e.g., American Type Culture Collection, or ATCC strains) or several randomly selected clinical isolates. Though occasionally insightful, these findings are rarely trans-latable to combatting clinical MDR strains.

To date, more than 2300 AMPs have been discovered, and more are constantly being reported (http://aps.unmc.edu/AP/main.php). This growing pool of usable materials is quite timely (Wang et al, 2009). In the present study, we have attempted and elucidate some relationship between the characters of several discovered AMPs and their activities against MDR clinical pathogens, so as to provide clues to the screening procedure of clinically valuable candidates. Here, we selected 14 AMPs of animal origin with typical structures to test theireffects on XDR, methicillin-resistant(MRSA), extended spectrum β-lactamase (ESBL)-producingand ESBL-producing.

MATERIALS AND METHODS

Ethics

The protocols of this study were approved by the ethics committee of Kunming Medical University and ethics committee of Institute of Medical Biology, Chinese Academy of Medical Sciences and Peking Union Medical College. Isolation of multidrug resistant clinical strains were undertaken with the informed and written consent of each patient. The study methodologies conformed to the standards set by the Declaration of Helsinki and all other relevant national and international regulations.

Bacterial strains

XDRstrains Ab5753 and Ab5755 were isolated from sputum samples of two different ICU patients in the First Affiliated Hospital of Kunming Medical University. All other strains were isolated from patients at the Second Affiliated Hospital of Kunming Medical University. These strains tested in this study include the following: (I) XDRstrain Ab1408 isolated from the sputum sample of one oncology department patient; (II) MRSA strain Sa1390 isolated from the sputum sample of one surgery intensive care unit (SICU) patient; (III) ESBLstrains Pa1409 and Pa4216 isolated from the sputum samples of two different oncology department patients; and (IV) ESBLstrain Ec513 isolated from the blood sample of one general surgery department patient. Bacterial identification was performed using a Vitek 32 system (bioMérieux, France).

Antimicrobial peptides and antibiotics

All tested AMPs were supplied by GL Biochem Ltd (Shanghai, China), and had a purity ≥95%. The tested AMPs were dissolved in water at 2 mg/mL and stored at ?20 °C before use. Determining whether the AMPs are amidated on the C-terminus or not was made through the previously published literature (Table 1). Antibiotics used as references were supplied by Sigma-Aldrich, and dissolved in water at 2 mg/mL prior use, and included colistin sulfate salt (colistin), vancomycin hydrochloride hydrate (vancomycin), and tigecycline hydrate (tigecycline).

Susceptibility testing

The minimal inhibitory concentrations (MICs) of all AMPs and antibiotic references to clinical isolates were determined by broth dilution method in MHII following Clinical and Laboratory Standards Institute (CLSI) recommendations (Clinical and Laboratory Standards Institute, 2013).

Time-killing curves

Time-kill studies of tachyplesinIII, cathelicidin-BF, colistin, and tigecycline on XDRstrain Ab1408 with initial inocula between 1×106and 1×107CFU/mL were performed using 4×MICs, 2×MICs, 1×MIC, and 0.5×MIC concentrations. Samples were taken at 0, 0.5, 2, 4, 8, and 24 h after incub-ation. The effects of drug carryover were addressed via three dilution steps. Only plates with between 30 and 300 colonies were counted. An antibiotic was considered bactericidal when a reduction of 3 log10CFU/mL was achieved, as compared with the initial inocula (Isenberg, 2004). These tests were each performed in duplicate.

Antimicrobial peptide structure prediction

Human LL-37 and beta-defensin-2—both of which are longer than 30 amino acids—were sourced directly from NCBI PDB database PDB 2K6O and PDB 1FD3. The other 12 AMPs were predicted by the structure prediction software PEP-FOLD at http://bioserv.rpbs.univ-paris-diderot.fr/PEP-FOLD/ (Thévenet et al, 2012), and model 1 of each peptide with the best conformation was selected for further analysis.

Table 1 Peptides tested for antimicrobial effects against drug resistant bacteria

a: For antimicrobial peptides with disulfide bridges, cysteine (C) with the same type font or underlined formed one disulfide.

RESULTS

Minimal inhibitory concentrations (MICs)

All MICs are listed in Table 2. We found that all threestrains were resistant to colistin according to CLSI standards (i.e., ≥4 mg/L). According to the susceptibility/resistance breakpoints of tigecycline, as interpreted by the European Committee on Antimicrobial Susceptibility Testing (EUCAST) (susceptible, ≤1 mg/L; resistant, ≥4 mg/L) (The European Committee on Antimicrobial Susceptibility Testing (EUCAST) Steering Committee, 2006), Ab5753 has intermediate resistance to tigecycline (XDR) whereas Ab5755 and Ab1408 are resistant to tigecycline (Pandrug resistance, PDR). Four AMPs, including CA(1-7)M(2-9), [D4K]B2RP, cathelicidin-BF, and Tachyplesin III, showed MICs between 4 and 16 mg/L to all of the threestrains. Given that the molecular weights of these peptides (from 1770 to 3638) are about three times to six times that of tigecycline (586), these peptides are considered as effective as tigecycline in inhibiting XDR.

ESBLandare still sensitive to colistin, but the twostrains were resistant to tigecycline. Four AMPs, namely, CA(1-7)M(2-9), [D4K]B2RP, cathelicidin-BF, and tachyplesinIII, showed MICs of between 4 and 8 mg/L to Ec513. By contrast, only two AMPs, namely, cathelicidin-BF and tachyplesin III, showed MICs between 4 and 8 mg/L to Pa4216 and Pa1409. This phenomenon is consistent with the results on tigecycline; that is, ESBLis more resistant than ESBLis

MRSA strain Sal1390 is sensitive to vancomycin, but intermediately resistant to tigecycline. Three AMPs, namely, CA(1-7)M(2-9), [D4K]B2RP, and tachyplesin III, showed MICs of 16 mg/L to this MRSA strain. Notably, [E4K]Alyteserin-1c and cathelicidin-BF, which showed a generally effective inhibitory activity on all MDR Gram-negative bacteria, had no effect on Gram-positive MRSA.

Time-killing kinetics

Figure 1 shows the time-killing kinetics. Cathelicidin-BF kills XDRstrain Ab1408 within 4 h at 2×MICs and 2 h at 4×MICs respectively. This is similar to the action of colistin, i.e., Cathelicidin-BF was also able to rapidly kill bacteria. By comparison, tachyplesin III kills XDRafter 4 h of incubation at 2×MICs and 4×MICs. Compared with colistin, tigecycline showed a markedly slower killing effect (after 8 h of incubation at 2×MICs and 4×MICs), but this effect lasted longer (24 h at 4×MICs). The optimum killing effects of cathelicidin-BF and tachyp-lesin III appeared to be 4 h of incubation, and afterward gradually declined.

Figure 1 Time-killing curves of colistin, tigecycline, cathelicidin-BF, and tachyplesin III on one XDRstrain (Ab1408)

Initial inocula between 1×106and 1×107CFU/mL were performed using 4×MICs, 2×MICs, 1×MIC, and 0.5×MIC concentrations. Samples were obtained at 0, 0.5, 2, 4, 8, and 24 h after incubation.

Amphipathic structure distributions of AMPs

Although CA(1-7)M(2-9), [D4K]B2RP, [E4K]Alyteserin-1c, cathelicidin-BF, LL-37, and cramp are all classified as helical peptides by their secondary struct-ures, these AMPs showed different antibacterial effects. Since the amphipathic structure distributions are generally considered crucial for AMPs to kill bacteria (Brogden, 2005), we studied the three-dimensional structures of the eight AMPs that were found to be effective on our tested MDR clinical isolates (Figure 2). [E4K]Alyte-serin-1c and cathelicidin-BF, which were effective on all three Gram-negative bacteria, both have classical long linear amphipathic structures, withhydrophobic regions on one side and hydrophilic regions on the other side along the linear peptides. Similar structures exist in LL-37 and cramp, but these AMPs only showed slight activity on a portion of these Gram-negative strains. The helical peptides of CA(1-7)M(2-9) and [D4K]B2RP, which could inhibit not only Gram-negative but also Gram-positive bacteria, tend to manifest a more contractive style, with hydrophilic terminals separated by a hydrophobic intermediate section. A similar amphipathic structure distribution can be seen in tachyplesin III, which is also effective on both Gram-negative and Gram-positive clinical isolates, but is classified as a beta-sheet formed by two disulfide bridges according to its secondary structure.Interestingly, a similar structure also exists in Ranalexin-1Ca, which only showed slight activity to Gram-positive strains, but was ineffective on Gram-negative strains.

AMPs (mainly beta-sheet AMPs like beta-defensin-2 and alpha-defensin-2 that are formed with 3 disulfide bridges, and linear AMPs like oncocin, indolicidin, histatin and thanatin, which are rich in special amino acids) with neither classical long linear amphipathic structures nor hydrophilic terminals separated by hydrophobic intermediate sections, showed no effects on the tested MDR clinical isolates (Table 2, Figure 3).

Figure 2 Structures and three-dimensional hydrophilic/hydrophobic arrangements of effective antimicrobial peptides

The backbones of these AMPs are shown in flat ribbons, and the three-dimensional surfaces of these AMPs are shown in solid. Blue: hydrophilic regions; Red: hydrophobic regions. A: CA(1-7)M(2-9); B: [D4K]B2RP; C: Tachyplesin III; D: Ranalexin-1Ca; E: cathelicidin-BF; F: [E4K]Alyteserin-1c; G: LL-37; H: cramp. Dotted lines indicate different hydrophilic/hydrophobic region arrangement styles between A-D (both Gram-negative and Gram-positive effective, except for D, i.e., Ranalexin-1Ca, which is only Gram-positive effective) and E-H (Gram-negative effective).

Figure 3 Structures and three-dimensional hydrophilic/hydrophobic arrangements of ineffective antimicrobial peptides

The backbones of these AMPs are shown in flat ribbons, and the three-dimensional surfaces of these AMPs are shown in solid. Blue: hydrophilic regions; Red: hydrophobic regions. A: Indolicidin; B: Histatin; C: Thanatin; D: Oncocin; E: beta-Defensin-2; F: alpha-defensin-2.

Table 2 MICs (mg/L) of AMPs and antibiotics against drug resistant bacteria

Abbreviations: AMP: antimicrobial peptides; MIC: minimum inhibitory concentration; XDR: extensive drug resistance; ESBLs: extended spectrum beta-lactamase producing; MRSA: methicillin-resistant

DISCUSSION

In total, more than 2300 antimicrobial peptides have been reported to date, but as we mentioned earlier, clinical applications of these AMPs are held up by a number of roadblocks. In this study, we collected a set of AMPs with different structures to evaluate theireffects on typical MDR clinical isolates, which included XDR, MRSA, ESBL-producingandOur results showed that helical AMPs appear to be more effective at killing MDR bacteria than other types of AMPs. By contrast, linear AMPs rich in special amino acids and AMPs with disulfide bridges (with the exception of tachyplesin III) are generally less effective in combatting MDR bacteria as compared with helical AMPs. Three-dimensional analysis showed that the hydrophilic/hydrophobic arrangements of these peptides, and not their secondary structures, seem to contribute more to their efficacy and position on the antimicrobial spectra. Peptides with long linear amphipathic structures were found to be effective only on Gram-negative pathogens, whereas peptides with more contractive styles with hydrophilic terminals separated by a hydrophobic intermediate section appeared to be effective on both Gram-negative and Gram-positive pathogens. Even though all AMPs are reported as antimicrobial peptides, only 2 of the 14 tested AMPs (tachyplesin III and cathelicidin-BF) were effective against all the tested MDR pathogens within their antimicrobial spectrum and with low MICs (≤16 mg/L). Since the net charge these AMPs carried does not correlate with their antibacterial activities (data not shown), the underlying mechanisms that lead to different MICs to kill bacteria remains to be further determined in more targeted follow-up studies.

Our results also showed that Cathelicidin-BF kills bacteria as quickly as colistin (within 4 h at 2×MICs and 2 h at 4×MICs), while Tachyplesin III is slower (after 4 h of incubation at 2×MICs and 4×MICs) than colistin, but faster than tigecycline. Considering the difference of the amphipathic structure distributions between these two peptides, the different rate of killing bacteria may imply a difference in the mechanism underlying their activities. Notably, the killing effects of these two potent peptides all declined after 4 h of incubation, which may, in part, be due to the nature of the peptides. Again, further targ-eted studies are needed to shed more light into the differences of these AMPs.

In conclusion, in the present study not all of the reported AMPs were effective against several tested MDR clinical isolates. Of the 14 potential AMPs, only two, Tachyplesin III and cathelicidin-BF, which differ in both their secondary structures and three-dimensional hydrophilic/hydrophobic arrangements, showed potent activities (≤16 mg/L) against nearly all of the MDR clinical isolates within corresponding antimicrobial spectrum. What differentiates these twoAMPs in terms of their efficacy to kill bacteria compared with other AMPs with similar hydrophilic/hydrophobic arran-gements remains to be determined.

REFERENCES

Brogden KA. 2005. Antimicrobial peptides: pore formers or metabolic inhibitors in bacteria?3(3): 238-250.

Chromek M, Slamová Z, Bergman P, Kovács L, Podracká L, Ehrén I, H?kfelt T, Gudmundsson GH, Gallo RL, Agerberth B, Brauner A. 2006. The antimicrobial peptide cathelicidin protects the urinary tract against invasive bacterial infection.12(6): 636-641.

Cirioni O, Ghiselli R, Silvestri C, Kamysz W, Orlando F, Mocchegiani F, Di Matteo F, Riva A, ?ukasiak J, Scalise G, Saba V, Giacometti A. 2007. Efficacy of tachyplesin III, colistin, and imipenem against a multiresistantstrain.51(6): 2005-2010.

Clinical and Laboratory Standards Institute. 2013. Performance standards for antimicrobial susceptibility testing; 23rd informational supplement. M100-S23. Wayne, PA.

Conlon JM, Ahmed E, Condamine E. 2009. Antimicrobial properties of brevinin-2-related peptide and its analogs: Efficacy against multidrug-resistant.74(5): 488-493.

Conlon JM, Ahmed E, Pal T, Sonnevend A. 2010. Potent and rapid bactericidal action of alyteserin-1c and its [E4K] analog against multidrug-resistant strains of.31(10): 1806-1810.

Falagas ME, Karageorgopoulos DE. 2008. Pandrug resistance (PDR), extensive drug resistance (XDR), and multidrug resistance (MDR) among Gram-negative bacilli: need for international harmonization in terminology.46(7): 1121-1122.

Giacometti A, Cirioni O, Kamysz W, D'Amato G, Silvestri C, Del Prete MS, ?ukasiak J, Scalise G. 2003. Comparative activities of cecropin A, melittin, and cecropin A-melittin peptide CA(1-7)M(2-9)NH2against multidrug-resistant nosocomial isolates of.24(9): 1315-1318.

Giacometti A, Cirioni O, Kamysz W, D'Amato G, Silvestri C, Del Prete MS, Licci A, Riva A, ?ukasiak J, Scalise G. 2005. In vitro activity of the histatin derivative P-113 against multidrug-resistant pathogens responsible for pneumonia in immunocompromised patients.49(3): 1249-1252.

Halverson T, Basir YJ, Knoop FC, Conlon JM. 2000. Purification and characterization of antimicrobial peptides from the skin of the North American green frog.21(4): 469-476.

Hancock RE, Sahl HG. 2006. Antimicrobial and host-defense peptides as new anti-infective therapeutic strategies.24(12): 1551-1557.

Isenberg HD. 2004. Clinical Microbiology Procedures Handbook. Washington DC: ASM Press.

Knappe D, Piantavigna S, Hansen A, Mechler A, Binas A, Nolte O, Martin LL, Hoffmann R. 2010. Oncocin (VDKPPYLPRPRPPRRI YNR-NH2): a novel antibacterial peptide optimized against gram-negative human pathogens.53(14): 5240-5247.

Lipsky BA, Holroyd KJ, Zasloff M. 2008. Topical versus systemic antimicrobial therapy for treating mildly infected diabetic foot ulcers: a randomized, controlled, double-blinded, multicenter trial of pexiganan cream.47(12): 1537-1545.

Ouellette AJ, Miller SI, Henschen AH, Selsted ME. 1992. Purification and primary structure of murine cryptdin-1, a Paneth cell defensin.304(2-3): 146-148.

Pagès JM, Dimarcq JL, Quenin S, Hetru C. 2003. Thanatin activity on multidrug resistant clinical isolates ofand.22(3): 265-269.

Routsias JG, Karagounis P, Parvulesku G, Legakis NJ, Tsakris A. 2010.bactericidal activity of human β-defensin 2 against nosocomial strains.31(9): 1654-1660.

Selsted ME, Novotny MJ, Morris WL, Tang YQ, Smith W, Cullor JS. 1992. Indolicidin, a novel bactericidal tridecapeptide amide from neutrophils.267(7): 4292-4295.

The European Committee on Antimicrobial Susceptibility Testing (EUCAST) Steering Committee. 2006. EUCAST technical note on tigecycline.12(11): 1147-1149.

Thévenet P, Shen YM, Maupetit J, Guyon F, Derreumaux P, Tufféry P. 2012. PEP-FOLD: an updatedstructure prediction server for both linear and disulfide bonded cyclic peptides.40(W1): W288-293.

Thomas-Virnig CL, Centanni JM, Johnston CE, He LK, Schlosser SJ, Van Winkle KF, Chen RB, Gibson AL, Szilagyi A, Li LJ, Shankar R, Allen-Hoffmann BL. 2009. Inhibition of multidrug-resistantby nonviral expression of hCAP-18 in a bioengi-neered human skin tissue.17(3): 562-569.

Vaara M. 2009. New approaches in peptide antibiotics.9(5): 571-576.

Wang G, Li X, Wang Z. 2009. APD2: the updated antimicrobial peptide database and its application in peptide design.37(Database issue): D933- D937.

Wang YP, Hong J, Liu XH, Yang HL, Liu R, Wu J, Wang AL, Lin DH, Lai R. 2008. Snake cathelicidin fromis a potent peptide antibiotics.3(9): e3127.

28 July 2014; Accepted: 06 November 2014

This study was supported by the Peking Union Medical College (PUMC) Youth Fund and the Fundamental Research Funds for the Central Universities, China (333203084)

, E-mail: yanbingma@126.com

10.13918/j.issn.2095-8137.2015.1.41