Applied Microbiology and Biotechnology

, Volume 95, Issue 4, pp 939–945

Bactericidal activity of Musca domestica cecropin (Mdc) on multidrug-resistant clinical isolate of Escherichia coli

Authors

  • X. Lu
    • Guangdong Provincial Key Laboratory of Pharmaceutical Bioactive SubstancesGuangdong Pharmaceutical University
    • School of Public Health and Tropical MedicineSouthern Medical University
  • J. Shen
    • Guangdong Provincial Key Laboratory of Pharmaceutical Bioactive SubstancesGuangdong Pharmaceutical University
    • School of Public Health and Tropical MedicineSouthern Medical University
  • X. Jin
    • Guangdong Provincial Key Laboratory of Pharmaceutical Bioactive SubstancesGuangdong Pharmaceutical University
  • Y. Ma
    • Guangdong Provincial Key Laboratory of Pharmaceutical Bioactive SubstancesGuangdong Pharmaceutical University
    • School of Public Health and Tropical MedicineSouthern Medical University
  • Y. Huang
    • Guangdong Provincial Key Laboratory of Pharmaceutical Bioactive SubstancesGuangdong Pharmaceutical University
  • H. Mei
    • Guangdong Provincial Key Laboratory of Pharmaceutical Bioactive SubstancesGuangdong Pharmaceutical University
  • F. Chu
    • Guangdong Provincial Key Laboratory of Pharmaceutical Bioactive SubstancesGuangdong Pharmaceutical University
    • School of Public Health and Tropical MedicineSouthern Medical University
    • Guangdong Provincial Key Laboratory of Pharmaceutical Bioactive SubstancesGuangdong Pharmaceutical University
Biotechnologically relevant enzymes and proteins

DOI: 10.1007/s00253-011-3793-2

Cite this article as:
Lu, X., Shen, J., Jin, X. et al. Appl Microbiol Biotechnol (2012) 95: 939. doi:10.1007/s00253-011-3793-2

Abstract

The housefly (Musca domestica) larvae have been used clinically to cure osteomyelitis, decubital necrosis, lip boil, ecthyma and malnutritional stagnation ever since the Ming/Qing Dynasty (1368 Anno Domini) till now, in China. In prior research, we have cloned and characterized a new gene of antimicrobial peptide cecropin from M. domestica larvae. This peptide was potently active against Gram-positive and Gram-negative bacteria standard strain. In the present study, we evaluated the possibility of Mdc to be a potential bactericidal agent against clinical isolates of multidrug-resistant (MDR) Escherichia coli and to elucidate the related antimicrobial mechanisms. Antimicrobial activity assays indicated a minimal inhibitory concentration (MIC) of 1.56 μM. Bactericidal kinetics at MIC showed that Mdc rapid killing of MDR E. coli. Lipopolysaccharide (LPS) dose-dependently suppressed Mdc antibacterial potency indicates that LPS is the initial binding site of Mdc in E. coli. Propidium iodide-based flow cytometry revealed that Mdc causes E. coli membrane permeabilization. Transmission electron micrographs further indicated that a remarkable damage in the bacteria’s outer and inner membrane, even the leakage of cytoplasmic contents induced by Mdc. DNA binding experimental result implies that DNA is one of the possible intracellular targets of Mdc. Of note, Mdc did not show a perceptible cytotoxic effect on human red blood cells. Altogether, these results suggest that Mdc could be an excellent candidate for the development of more efficacious bactericidal agents.

Keywords

Musca domestica cecropinMultidrug resistantFlow cytometryTransmission electronLipopolysaccharide

Introduction

Escherichia coli is one of the major bacterial pathogens responsible for nosocomial infections especially in immunocompromised patients (Ashour and El-Sharif 2009; Bodmann 2005; Kuntaman et al. 2005). Chemotherapy is often used against E. coli infections; however, the emergence of multidrug-resistant (MDR) E. coli such as resistance for β-lactam antibiotics and other chemotherapeutic agents, is a serious clinical problem (Lim et al. 2009; Song et al. 2009). In various MDR clinical isolates, an altered profile of non-specific porins associated with the presence of mutation of drug-binding proteins on the membrane, production of a β-lactamase and active drug efflux, contributes to a high resistance level for structurally unrelated molecules such as aminoglycosides, β-lactam antibiotics and quinolones (Wu et al. 2009). Different mechanisms in conjunction maintain a very low intracellular concentration of drug(s) resulting in MDR pathogens. Therefore, a new strategy is required for the development of new chemotherapeutic agents with antibacterial mechanisms that differ from traditional antibiotics.

Antimicrobial peptides (AMPs) are evolutionarily conserved molecules involved in the defensive mechanisms of a wide range of organisms. Produced in bacteria, insects, plants, and vertebrates, AMPs are small (3–5 kDa) helical peptides with broad activities against bacteria, fungi, viruses, and certain parasites (Huang et al. 2009; Guani-Guerra et al. 2010). These molecules are presently gaining increasing importance as a consequence of their remarkable resistance to microorganism adaptation (Guerreiro et al. 2008). Several hundreds of AMPs have been identified and characterized thus far, in which cecropin is one of the most extensively studied (Strauss et al. 2010). Previous studies in cecropin and its related peptides demonstrate that permeabilization of the membrane seems to be the most likely mechanism (Schmitt et al. 2008; Arcidiacono et al. 2009). However, how cecropin damage and kill microorganisms still needs to be clarified.

Housefly (Musca domestica) has been implicated in the spread of over 30 bacterial and protozoan diseases; however, they can thrive without causing infection. The housefly larvae have been used clinically to cure osteomyelitis, decubital necrosis, lip boil, ecthyma and malnutritional stagnation ever since the Ming/Qing Danysty (1368 Anno Domini) till now in China and were also used to treat coma and gastric cancer when combined with other drugs (Hou et al. 2007). Previously, we have cloned and characterized a new gene belonging to the cecropin family from housefly larvae. The peptide has a molecular weight of 4 kDa with a significant potency against Gram-positive and Gram-negative bacteria standard strain (Lu et al. 2010). However, the precise antibacterial mechanism(s) of M. domestica cecropin (Mdc) remain unknown. Moreover, it is unclear whether Mdc has potent antibacterial activity against multidrug-resistant bacterial strain.

In the present study, we evaluated the potency of Mdc against clinical isolates of E. coli that produce extended spectrum β-lactamases (ESBLs) and are multidrug-resistant. We also attempted to acquire the insight for the underlying antibacterial mechanisms, used competitive assay of LPS versus Mdc, flow cytometry, transmission electron microscopy, and DNA-binding analysis.

Materials and methods

Microorganisms and medium

MDR clinical isolate of E. coli GIM1.457 was obtained from the Center of Medical Laboratory of the first affiliated hospital of Guangdong Pharmaceutical University, Guangzhou, China. This strain that produces ESBLs are resistant to most tested antibiotics, including ampicillin, cefazolin, ceftriaxone, ceftazidime, piperacillin, cefepime, aztreonam, ampicillin/sulbactam, amoxicillin/clavulanic acid, piperacillin/tazobactam, chloramphenicol, gentamicin, ciprofloxacin, and levofloxacin, with the exception of imipenem/cilastatin. E. coli ATCC 35218 (β-lactamase producing strain) was employed as quality control strain. Luria–Bertani (LB) medium [0.5% yeast extract (w/v), 1% tryptone, and 1% NaCl] and Mueller–Hutton (MH) medium [0.5% beef extract (w/v), 1.75% casamino acid, and 0.15% starch] were employed for pre-incubation of the test bacteria and antimicrobial assays. The bacteria were grown overnight to an optical density at 600 nm (OD600) of 0.8–1.0 prior to appropriate dilution and antimicrobial testing.

Peptide and material

M. domestica cecropin (Mdc) was chemically synthesized by conventional Fmoc solid-phase synthetic method with a 431-Å peptide synthesizer (Applied Biosystems Inc., Foster City, CA). The synthesized peptides were purified by preparative reversed phase-HPLC (RP-HPLC) (Waters Delta-PakTM C18, 15 μm, 300 Å, 25 × 100 mm) and eluted with a 2% to 90% gradient of acetonitrile in 0.05% trifluoroacetic acid in water. The concentration and purity for the above prepared peptides were determined by analytical RP-HPLC (Waters Symmetry C18, 3.5 μm, 100 Å, 4.6 × 150 mm) followed by mass determination of the eluate with an API electrospray ionization mass spectrometer (Perkin Elmer SCIEX) (data not shown). LPS from E. coli serotype O111:B4 was purchased from Sigma-Aldrich (St. Louis, MO) and was prepared in sterile saline, aliquoted and stored at −80 °C until use. All other chemicals were standard commercial products with analytical reagent grade.

Antimicrobial activity assay

The minimal inhibition concentration (MIC) was determined in 96-well polypropylene microtitre plates by concentration-dependent reducing Mdc binding. A series of twofold dilutions in phosphate-buffered saline (PBS) solution was prepared from a stock solution of the peptide (200 μM). The strain was grown to OD600 of 0.8 in MH broth and diluted to 2 × 106 colony forming units (CFUs) per milliliter with the same medium. The suspension (100 μl) was then mixed with 100 μl of peptide dilutions. PBS was used as positive controls, while wells containing peptide without bacterial suspension were served as negative controls. After incubation of 18 h at 37 °C, the MIC was read as the lowest concentration of antimicrobial agent resulting in the complete inhibition of visible growth, and the results were presented as mean values of three independent experiments. Minimal bactericidal concentration (MBC) was estimated from the same test by viable counting assay and defined as the lowest concentration of peptide that killed 99.9% of the test inoculum.

Bactericidal kinetics was determined using the assay described by Arcidiacono et al. with minor modifications (Arcidiacono et al. 2009). Mdc at the MIC concentration of 1.56 μM was incubated with 106 CFU per milliliter of E. coli GIM1.457 strains for 120 min. Aliquots of the mix were serially diluted tenfold in PBS, plated on LB agar and incubated overnight at 37 °C. CFU was counted 24 h later to determine cell viability. Peptide-free controls were also run to determine the concentrations of control cells.

Competitive assay between Mdc and LPS in E. coli

To evaluate whether E. coli LPS was the target for the Mdc, we developed a competitive assay by mixing 1.56 μM of Mdc with 10 and 100 times LPS, respectively. The Mdc–LPS mixture and bacteria were incubated at 37 °C with shaking for 60 min. After that, aliquots of Mdc–LPS mixture were serially diluted tenfold in PBS and plated on LB agar and incubated overnight at 37 °C. CFUs were counted after 24 h of culture to estimate the inhibitory effect on growth. All experiments were performed three times in triplicate.

Membrane permeabilization by FACScan analysis

Membrane permeabilization after peptide treatment was estimated by flow cytometry using the propidium iodide (PI) (Invitrogen Ltd., Paisley, UK) inclusion criterium. E. coli GIM1.457 strains were cultured in LB as described before and were harvested at log phase. A suspension of approximately 2 × 106 cells per milliliter were incubated with 3.12 μM of Mdc for 60 min at 37 °C. The samples were centrifuged at 6,600 rpm for 7 min, washed three times with PBS, and incubated with PI (final concentration, 5 mg/ml) for 30 min at room temperature in the dark, followed by removal of the unbound dye through extensive washes with PBS. Flow cytometry was performed using a FACScan (BD Biosciences, NJ, USA). Control experiments were carried out in bacterial cells either without the addition of peptide (viability control) or with 0.1% SDS (which disrupts lipids in the bacterial membrane). All experiments were conducted three times in triplicate.

Transmission electron microscopy

Transmission electron microscopy (TEM) was used to examine the morphology of bacterial cells after treatment with the peptide. Log phase cells of E. coli GIM1.457 were collected by centrifugation (3,000 rpm for 15 min), and the cell pellet was washed twice and resuspended with phosphate-buffered saline solution. E. coli strains at a concentration of 1.5 × 108 cells per milliliter were incubated for 60 min at 37 °C with Mdc at a concentration of 3.12 μM. After centrifugation, the pellet was resuspended in 0.1 M phosphate buffer containing 3% of glutaraldehyde and 1.5% paraformaldehyde. After 2.5 h of incubation at 4 °C, the cells were washed with 0.1 M phosphate buffer three times. The pellet was fixed with 1% osmium tetroxide and 1.5% potassium ferrocyanide for 1.5 h at 4 °C, and then treated with 2% uranyl acetate for 1 h before the dehydration in an ethanol series and embedded in epoxy resin. Finally, the pellet was sectioned and examined in a JEM1400 (Jeol, Tokyo, Japan) transmission electron microscopy. Microphotographs were taken with a digital camera Bioscan 792 (Gatan Inc., Pleasanton, CA, USA).

DNA-binding assay

The DNA-binding assays were as previously described with minor modifications (Song et al. 2005). Briefly, 100 ng of the plasmid DNA (EcoRI digested pBluescript II SK-) were mixed with increasing amounts of peptide in 20 μL of binding buffer (5% glycerol, 10 mM Tris–HCl at pH 8.0, 1 mM EDTA, 1 mM dithiothreitol, 20 mM KCl, and 50 μg/mL bovine serum albumin). The reaction mixtures were incubated at room temperature for 1 h. Subsequently, 4 μL of native loading buffer was added, and an aliquot of 20 μL was applied to a 1% agarose gel electrophoresis in 0.5× Tris–borate–EDTA buffer.

RBC preparation and hemolytic activity

The potential effect of hemolytic activity for Mdc was evaluated by determining the hemoglobin release from 8% (v/v) suspensions of fresh human erythrocytes at 414 nm. The peptide was solubilized in PBS to final concentrations of 200, 100, 50, 25, 12.5, 6.25 and 3.12 μM. Fifty microliters of human suspended red blood cells (8%) were mixed with 50 μl of each above prepared peptide solution followed by incubation at 37 °C for 60 min with shaking. After centrifugation at 3,500 × g for 10 min, 80 μl of the resulting supernatant were employed for analysis of hemolysis by measuring the absorbance at 414 nm with an ELISA plate reader (Molecular Devices Emax, Sunnyvale, CA, USA). PBS (0%) and 0.1% Triton-X 100 (100%) were used as negative and positive controls, respectively. The hemolysis percentage was calculated using the following equation: hemolysis (%) = [(Abs414 nm in the peptide solution − Abs414 nm in PBS)/(Abs414 nm in 0.1% Triton-X 100 − Abs414 nm in PBS)] × 100.

Results

Antimicrobial activity

The MIC and MBC values of Mdc against E. coli GIM1.457 were 1.56 and 3.12 μM, respectively. Similar results were found for quality control strain E. coli ATCC 35218. To further explore its antimicrobial potency, bactericidal kinetic assays were conducted to determine the rate of lysis. As compared to negative controls, the addition of Mdc (1.56 μM) resulted in a 1-log reduction after 10 min of culture, and around 90% bacteria were undergoing lysis evidenced by a 5-log reduction after 30 min of culture (Fig. 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs00253-011-3793-2/MediaObjects/253_2011_3793_Fig1_HTML.gif
Fig. 1

Bactericidal kinetics of Mdc versus E. coli GIM1.457. A 5-log reduction occurs after 30 min at the MIC (1.56 μM). (SD, ≤10%, n = 3)

LPS dose-dependently suppressed Mdc bactericidal activity

To explore the underlying mechanisms of bactericidal activity for Mdc, we examined whether Mdc interacts with LPS and through which it suppresses E. coli growth. To this end, we first mixed Mdc (1.56 μM) with 10 and 100 times of LPS, and then cultured E. coli in the presence of either Mdc alone or Mdc/LPS mixtures. As shown in Fig. 2 Mdc alone suppressed E. coli GIM1.457 growth by 90.4%. On the contrary, the suppressive effect was reduced by 1.6-fold when E. coli cells were cultured with the Mdc/LPS mixture at 1:10 ratio. Remarkably, the suppressive effect was almost completely abolished when 100 times of LPS were mixed with Mdc as evidenced by that E. coli cells showed a 96.3% growth rate (Fig. 2). Together, these data suggest that LPS dose-dependently suppresses Mdc bactericidal activity, and therefore, LPS could be the primary target for Mdc.
https://static-content.springer.com/image/art%3A10.1007%2Fs00253-011-3793-2/MediaObjects/253_2011_3793_Fig2_HTML.gif
Fig. 2

Inhibition of Mdc bactericidal activity with LPS. Of the Mdc, 1.56 μM was incubated with 10 and 100 times the concentration of LPS and then were incubated with E. coli GIM1.457. The percentage of inhibition on the growth was determined by counting the CFU from serial tenfold dilutions prepared in LB agar of the E. coli strain after the treatment with peptide. All experiments were made three times in triplicate. The mean and deviation standard are indicated

Membrane permeabilization by FACScan analysis

The effect of Mdc on the integrity of bacterial membrane was examined by PI-based flow cytometry analysis. The negative controls in which E. coli GIM1.457 were cultured with normal medium showed 99% viability (Fig. 3a). E. coli cells cultured in the presence of detergent SDS were used as positive controls (Fig. 3c). It was interestingly found that the addition of Mdc led to a significant E. coli membrane disruption as manifested by that around 86.8 ± 5.6% of E. coli cells were PI-positive cells (Fig. 3b, d).
https://static-content.springer.com/image/art%3A10.1007%2Fs00253-011-3793-2/MediaObjects/253_2011_3793_Fig3_HTML.gif
Fig. 3

Membrane permeability of E. coli GIM1.457 analyzed by the FACScan flow cytometer. The relative fluorescence intensities within the P2 regions were taken as dead cells. a Untreated cells control, b cells treated with 3.12 μM of Mdc at 37 °C for 60 min, c dead control (incubated with 0.1% SDS), d percentage of cells stained with PI

Transmission electron microscopy

To demonstrate the morphological changes of E. coli GIM1.457 cells after culture with Mdc, we first treated E. coli cells with Mdc at a concentration of 3.12 μM for 60 min. TEM was then performed on thin sections of the bacteria. Microphotographs corresponding to the intact (control) cells of E. coli GIM1.457 were shown in Fig. 4a and b, in which the cytoplasmic content was evenly distributed and filled the whole space encapsulated by the bacterial wall. The outer and inner membranes of the bacterial envelope were smooth, and the cell aggregation was not detected. In sharp contrast, significant morphological changes were observed in Mdc-treated E. coli GIM1.457 cells (Fig. 4c, d). The bacteria’s outer membrane showed a remarkable damage and appeared distended and displaced in relation to the cytoplasmic membrane (Fig. 4d). Coagulated material was seen inside the treated cells, especially close to the envelope of the cell. For some bacteria, the envelope seemed to be almost completely destroyed, and the cytoplasm content was released to the surrounding medium (Fig. 4c). In line with these results, a high percentage of cell aggregation was detected, and the cell shape was significantly swollen.
https://static-content.springer.com/image/art%3A10.1007%2Fs00253-011-3793-2/MediaObjects/253_2011_3793_Fig4_HTML.gif
Fig. 4

Ultrastructural damages in bacteria treated with peptide. Bacteria were untreated (a, b) or treated with Mdc (c, d) at a concentration of 3.12 μM. The cells were processed to be analyzed under TEM

DNA-binding activity

To determine the molecular mechanisms of action, we examined the DNA-binding properties of Mdc. The DNA-binding affinity of Mdc was examined by analyzing the electrophoretic mobility of DNA bands at various peptide/DNA weight ratios. MDC inhibited the migration of DNA at concentrations above 25 μM (Fig. 5).
https://static-content.springer.com/image/art%3A10.1007%2Fs00253-011-3793-2/MediaObjects/253_2011_3793_Fig5_HTML.gif
Fig. 5

Interaction of Mdc with plasmid DNA. Binding was assayed by the inhibition effect of Mdc on plasmid DNA (100 ng; pBluescript II SK-) migration. Lane 1 plasmid DNA alone, lane 2 0.78 μM Mdc, lane 3 1.56 μM Mdc, lane 4 3.13 μM Mdc, lane 5 6.25 μM Mdc, lane 6 12.5 μM Mdc, lane 7 25 μM Mdc, lane 8 50 μM Mdc, lane 9 100 μM Mdc. DNA and Mdc were co-incubated for 1 h at room temperature before electrophoresis on a 1.0% agarose gel

Hemolytic activity

We finally examined the potential cytotoxicity of Mdc in mammalian cells. For this purpose, we checked the effect of Mdc to lyse human erythrocytes, and by which, we cultured freshly isolated human red blood cells with different doses of Mdc. Excitingly, we failed to detect any hemolytic activity for all doses tested for Mdc (data not shown), indicating that Mdc is not toxic to human erythrocytes.

Discussion

E. coli is the most frequent pathogen involved in some of the most common bacterial infections, ranging from pneumonia to urinary tract infections and bacteremia. The emergence of ESBL producers along with multiple-resistant isolates poses a serious problem in the hospital setting making these pathogens important targets to evaluate the bactericidal activity of Mdc (Lim et al. 2009; Song et al. 2009). In the present study, we show that Mdc has a potent antibacterial activity against the clinical isolates of E. coli which produces ESBLs and are multidrug resistant, including aminoglycosides, quinolones, and β-lactam antibiotics. The results imply that Mdc could be a unique agent with antibacterial mechanisms that differ from traditional antibiotics. Killing kinetic analysis revealed that Mdc can mediate the rapid killing of MDR E. coli within 30 min at a concentration as low as 1.56 μM. Interestingly, these observations are consistent with recent reports that showed that peptide can mediate the rapid killing of E. coli and resulted in log orders of cell death within minutes of peptide addition (Arcidiacono et al. 2009).

The envelope of Gram-negative bacteria consists of two membranes, the inner cytoplasmic membrane, whose lipid matrix is composed of phospholipids, and the outer leaflet of the outer membrane, which is composed of LPS. Given the fact that LPS shields the bacteria and provides an effective barrier against different compounds (King et al. 2009), it would be logical to examine whether LPS is the initial binding site for Mdc. LPS dose-dependent suppressed Mdc bactericidal activity (Fig. 2) indicates that Mdc could be able to interact initially with the negatively charged LPS molecules in E. coli. Moreover, the initial binding of the Mdc to the outer LPS leaflet would cause its depolarization and expansion, which results in transient “cracks” in the outer membrane, permitting the passage of a variety of molecules, including the uptake of Mdc itself (Papo and Shai 2005). These results are in agreement with the previous report in which LPS binding is part of a self-promoted uptake for cationic peptides (Hammer et al. 2010). Once Mdc have gained access to the inner cytoplasmic membrane, they can interact with the lipid bilayer. To explore the correlation of bactericidal action and membrane permeability, PI-based flow cytometry were undertaken. PI is a specific fluorescent dye that only penetrates bacteria with compromised membranes, to selectively label nonviable cells. The results indicate that Mdc caused bacterial membrane permeabilization (Fig. 3). Transmission electron micrographs further confirmed this result evidenced by the observations that the cell shape was significantly swollen, the outer membrane became detached from the plasma membrane, and the plasma membrane was damaged. The leakage of cytoplasmic contents was also demonstrated by the electron micrographs (Fig. 4c, d). Collectively, these results suggest that Mdc could be interacting with LPS, altering the permeability properties to mediate the disruption of the E. coli membrane, which then leads to the death of MDR E. coli cells.

Although the extensive disruption of membrane eventually leads to the death of bacterial cells, there is increasing speculation that these effects are not the only bactericidal mechanisms (Brogden 2005). The change in cytoplasmic electron density upon incubation with Mdc (Fig. 4c, d) might indicate an intracellular interaction of Mdc with additional targets such as negatively charged DNA. Thus, gel retardation assay was undertaken. Our result showed that Mdc binds to DNA and alters their electrophoretic mobilities in 1% agarose gels. The result suggests that one possible mechanism of antibacterial action of Mdc is related to the inhibition of metabolic pathways by blocking or reducing DNA replication and/or transcription through binding DNA. This study increases our knowledge regarding the molecular mechanism of DNA, which is considered as the intracellular target for Mdc.

A major limitation for the transition of cecropin peptides to therapeutic purpose is the cytotoxicity or hemolytic properties (Mihajlovic and Lazaridis 2010). Of important note, we failed to detect a discernable cytotoxic effect for Mdc on human red blood cells at diverse doses. In summary, data resulted from current studies demonstrated convincing evidence that Mdc could be a unique candidate for the development of novel and effective antibiotics against multidrug-resistant E. coli without hemolytic activity.

Acknowledgment

This work was supported by the National Natural Science Foundation of China (no. 30671832).

Conflict of interest

None declared.

Copyright information

© Springer-Verlag 2011