Entry of a Six-Residue Antimicrobial Peptide Derived from

Subscriber access provided by UNIVERSITY OF THE SUNSHINE COAST

Article

Entry of 6-Residue Antimicrobial Peptide Derived from Lactoferricin B into Single Vesicles and E. coli Cells without Damaging their Membranes Md. Moniruzzaman, Md. Zahidul Islam, Sabrina Sharmin, Hideo Dohra, and Masahito Yamazaki Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.6b01274 • Publication Date (Web): 28 Jul 2017 Downloaded from http://pubs.acs.org on July 31, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Entry of 6-Residue Antimicrobial Peptide Derived from Lactoferricin B into Single Vesicles and E. coli Cells without Damaging their Membranes Md. Moniruzzaman,a Md. Zahidul Islam,a,# Sabrina Sharmin,a Hideo Dohra,b and Masahito Yamazakia, c, d,*

a

Integrated Bioscience Section, Graduate School of Science and Technology, b Instrumental Research Support Office,

Research Institute of Green Science and Technology,

c

Nanomaterials Research Division, Research Institute of

Electronics, d Department of Physics, Graduate School of Science, Shizuoka University, Shizuoka 422-8529, Japan

#

Present address: Department of Biotechnology and Genetic Engineering, Jahangirnagar University, Savar,

Dhaka-1342, Bangladesh

Funding This work was supported in part by a Grant-in-Aid for Scientific Research (B) (No.15H04361) from the Japan Society for the Promotion of Science (JSPS) to M.Y.

*Correspondence should be addressed to: Dr. Masahito Yamazaki Nanomaterials Research Division, Research Institute of Electronics, Shizuoka University 836 Oya, Suruga-ku, Shizuoka 422-8529, Japan Tel/Fax: 81-54-238-4741 E-mail: [email protected]

1 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 33

ABSTRACT Lactoferricin B (LfcinB) and shorter versions of this peptide have antimicrobial activity. However, the elementary processes of interactions of these peptides with lipid membranes and bacteria are still not well understood. To elucidate the mechanism of their antimicrobial activity, we investigated the interactions of LfcinB (4-9) (its sequence of RRWQWR) with E. coli cells and giant unilamellar vesicles (GUVs). LfcinB (4-9) and lissamine rhodamine B redlabeled LfcinB (4-9) (Rh-LfcinB (4-9)) did not induce an influx of a membrane-impermeant fluorescent probe, SYTOX green, from the outside of E. coli cells into their cytoplasm, indicating that no damage occurred in their plasma membrane. To examine the activity of LfcinB (4-9) to enter E. coli cytoplasm, we investigated the interaction of Rh-LfcinB (4-9) with single cells of E. coli containing calcein using confocal microscopy. We found that RhLfcinB (4-9) entered the cytoplasm without leakage of calcein. Next, we investigated the interactions of Rh-LfcinB (4-9) with single GUVs of dioleoylphosphatidylglycerol (DOPG) and dioleoylphosphatidylcholine (DOPC) mixtures containing a fluorescent probe, Alexa Fluor 647 hydrazide (AF647), using the single GUV method. The results indicate that Rh-LfcinB (4-9) outside the GUV translocated through the GUV membrane and entered its lumen without leakage of AF647. Interaction of Rh-LfcinB (4-9) with DNA increased its fluorescence intensity greatly. Therefore, we can conclude that Rh-LfcinB (4-9) can translocate across lipid membrane regions of the plasma membrane of E. coli cells to enter their cytoplasm without leakage of calcein and its antimicrobial activity is not due to damage of their plasma membranes.


Page 3 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Antimicrobial peptides (AMPs) with an activity to kill bacteria and fungi are produced by various organisms to defend themselves against microbes.1-3 The target of most AMPs such as -defensin, magainin 2, and LL-37 is bacterial and fungal plasma membranes; these AMPs induce damages the membranes such as pore formation to increase membrane permeability.1-3 There are some experimental results to support that the target of these AMPs is the lipid regions of bacterial and fungal plasma membranes. For example, all-D amino acid magainin 2 has the same antibacterial activity as that of natural, all-L amino acid magainin 2,4 indicating that specific proteins are not required for its antibacterial activity. AMPs can induce membrane permeation (or leakage) of the internal contents of vesicles of lipid membranes such as large unilamellar vesicles (LUVs) and giant unilamellar vesicles (GUVs),5,6 indicating damages of the lipid membranes or pore formation in the lipid membrane. It is also reported the LL-37 induced membrane permeation from living bacteria.7 In contrast, buforin II can translocate across the plasma membrane of bacteria to enter their cytoplasm and bind with DNA.8 The mechanisms of the AMP-induced damage of plasma membrane and the translocation of AMP across it are not well understood. Lactoferricin B (LfcinB) is one of the AMPs, which is derived from bovine lactoferrin (LF), a transferrin-like glycoprotein in exocrine secretions such as milk.9-11 LF in milk is hydrolyzed to various peptides by pepsin digestion under acidic conditions (e.g., in mammal stomach); some of the resulting peptides exhibit antimicrobial activity and among them LfcinB has the highest activity. LfcinB is composed of 25 amino acids, with an amino acid sequence of FKCRRWQWRMKKLGAPSITCVRRAF, and has a disulfide bond, although this bond is not important for bactericidal activity.12 Interactions of LfcinB with lipid membranes have been investigated using fluorescence spectroscopy.13 Our recent studies indicate that the main mechanism of bactericidal activity of LfcinB is a damage of plasma membrane of bacteria.14 LfcinB induced the entry of a membrane-impermeant fluorescent probe, SYTOX green, from the outside of E. coli into its cytoplasm. Moreover, LfcinB induced rapid leakage of a fluorescent probe, calcein, from single GUVs stochastically, which was due to local rupture of the GUV membrane.14 On the other hand, shorter versions of LfcinB also have an antimicrobial activity and the interactions of these peptides with lipid membranes have been investigated using various biophysical techniques.15-24 However, the elementary processes of the interaction of the shorter versions of LfcinB with lipid membranes and bacteria and also the mechanisms of antimicrobial activity of these shorter peptides are still not well understood.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 33

Here we examined the mechanism of the antimicrobial activity of one of the shorter versions of LfcinB, i.e., LfcinB (4-9) composed of 6 amino acid residues with the sequence of RRWQWR, which is one of the highest antimicrobial activity among the shorter versions of LfcinB.15 First we examined the interactions of LfcinB (4-9) and fluorescent probe, lissamine rhodamine B red-labeled LfcinB (4-9) (i.e., Rh-LfcinB (4-9)) with E. coli cells. The results indicated that Rh-LfcinB (4-9) entered the cytoplasm of E. coli cells without damage in their plasma membranes. To elucidate the elementary processes of the entry of Rh-LfcinB (4-9), we investigated the interactions of Rh-LfcinB (4-9) with single GUVs composed of negatively charged dioleoylphosphatidylglycerol (DOPG) and electrically neutral dioleoylphosphatidylcholine (DOPC) mixtures (1/1; molar ratio) (i.e., PG/PC (1/1)-GUVs) using the single GUV method for cell-penetrating peptides (CPPs).25-27 Based on the results, we discussed the mechanism of the antimicrobial activity of LfcinB (4-9) and its translocation across the lipid membranes. ■ MATERIALS AND METHODS Materials DOPC and DOPG were purchased from Avanti Polar Lipids Inc. (Alabaster, AL). Bovine serum albumin (BSA) and Nutrient Broth medium contained Bacto peptone, beef extract, and agar were bought from Wako Pure Chemical Industry Ltd. (Osaka, Japan). Alexa Fluor 647 hydrazide (AF647), acetoxymethyl (AM)-calcein, and sulforhodamine B were bought from ThermoFisher Scientific (Waltham, MA). Lissamine rhodamine B Red (LRB Red) succinimidylester was purchased from AAT Bioquest Inc. (Sunnyvale, CA). DNA from salmon testes was purchased from Sigma Co. (St. Lois, MO). E. coli (JM-109) was gifted from Prof. Taketomo Fujiwara, Department of Biological Science, Shizuoka University, Shizuoka, Japan. LfcinB (4-9) was prepared by the FastMoc method using a 433A peptide synthesizer (PE Applied Biosystems, Foster City, CA). The sequence of LfcinB (4-9) (6-mer) is RRWQWR, which has an amide-blocked C terminus. The fluorescence probe LRB red-labeled LfcinB (4-9) (Rh-LfcinB (4-9)), which has one fluorophore Rh at the N-terminus of the peptide, was synthesized using a standard method25 by the reaction of LRB Red succinimidylester (20 mg) with LfcinB (4-9)-peptide resin (80 mg) (molar ratio of reagent to peptide: 1.1 to 1) in dimethylformamide for 24 h at room temperature. Rh-LfcinB (4-9) was cleaved from the resin using trifluoroacetic acid (TFA), 1, 2-ethanedithiol, thioanisole, and MilliQ water (10/0.25/0.5/0.5, volume ratio) and phenol (0.15 g per 2 mL TFA). The method for purification of the peptide was described previously.28 Mass spectra of LfcinB (4-9) and Rh-LfcinB (4-9) were


Page 5 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

acquired by LC−MS analysis using the linear ion trap time-of-flight mass spectrometer (LIT–TOF MS), NanoFrontier eLD (Hitachi High-Technologies Corporation, Tokyo, Japan) as previously described.14 The samples were separated by a MonoCap C18 Fast-flow column (0.05 mm ID x 150 mm long; GL Sciences, Inc., Tokyo, Japan) and ionized with a nano-electrospray ionization source, and then mass spectra were obtained in the positive ion mode at scan mass range m/z 200–2000. The measured masses of LfcinB (4-9) and Rh-LfcinB (4-9) were 985.56  0.01 and 1636.78  0.01 Da, respectively, which correspond to the molecular masses calculated from the monoisotopic mass of all the atoms in these molecules. LfcinB (4-9) concentrations in buffer were determined by absorbance using the molar extinction coefficient of Trp at 278 nm (i.e., 5,500 M1cm1). LfcinB (4-9) has two Trp residues and hence the molar extinction coefficient of LfcinB (4-9) is 11,000 M1cm1. On the other hand, Rh-LfcinB (4-9) concentrations in buffer were determined by absorbance using the molar extinction coefficient of Rh at 568 nm (i.e., 95,000 M1cm1). Measurement of minimum inhibitory concentration (MIC) We measured the MIC of LfcinB (4-9) and Rh-LfcinB (4-9) against E. coli (JM-109) using the standard method.29,30 Briefly, a suspension of E. coli in Nutrient Broth medium was mixed with various concentrations of LfcinB (4-9) solution in the individual wells of a 96-well plate. The final density of bacteria in the wells was 1105 CFU (colony forming unit) /mL, and the final peptide concentration in the wells ranged from 0.25 to 80 M. After incubation at 37 C for 18-20 h, the absorbance at 600 nm for LfcinB (4-9) and the absorbance at 450 nm for RhLfcinB (4-9) were measured using a microplate reader (Infinite M200, TECAN, Grödig, Austria), and then the value of absorbance of LfcinB (4-9) and Rh-LfcinB (4-9) were subtracted. The MIC was defined as the lowest concentration of peptide at which there was no change in the absorbance, i.e., no growth of E. coli. Measurement of hemolysis Hemolytic activities of peptides were measured according to Yamaji et al.31 Briefly, erythrocytes were isolated from chicken blood (Cosmo Bio Co., Ltd., Tokyo, Japan) and washed three times using phosphate-buffered saline (PBS; 8.1 mM Na2HPO4, 1.5 mM KH2PO4, 137 mM NaCl, 2.7 mM KCl) by centrifugation. Erythrocyte suspensions were incubated with various concentrations of LfcinB (4-9) and Rh-LfcinB (4-9) at 37 C for 30 min (final cell concentration was 1  107 cells/mL). After centrifugation of suspensions at 800  g for 10 min at 4 C, the absorbance


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 33

values of their supernatants were measured at 415 nm. 100 % hemolysis was determined by addition of Triton X-100 (its final concentration was 1.0 %) to an erythrocyte suspension described above. The absorbance values for PBS buffer and the supernatant of the erythrocyte suspension mixed with Triton X-100 solution were used to determine 0 % and 100 % hemolysis, respectively. In the case of Rh-LfcinB (4-9), the absorbance due to Rh-LfcinB (4-9) at 415 nm was subtracted from total absorbance of the supernatants. Peptides-induced entry of SYTOX green into cytoplasm of E. coli cells A suspension of E. coli (JM-109) cells in Nutrient Broth medium was centrifuged (350  g, 10 min) and the pellet was resuspended in buffer A (10 mM PIPES, pH 7.0, 150 mM NaCl, and 1.0 mM EGTA). Aliquots of this E. coli suspension were mixed with SYTOX green in DMSO solution, and then mixed with various concentrations of LfcinB (4-9) and Rh-LfcinB (4-9) solution, yielding mixtures with final bacterial densities of 1106 CFU/mL, final SYTOX green concentrations of 5.0 μM, final LfcinB (4-9) concentrations of 0, 25, 50, and 100 μM, and final DMSO concentrations of 0.25% (v/v).14 In the case of Rh-LfcinB (4-9), final peptide concentrations were 0, 5, 9, and 20 μM. Immediately after the mixing, we started to measure the fluorescence intensity (FI) of the mixture. The time courses of FI of these suspensions were measured using a Hitachi F7000 spectrofluorometer (Hitachi, Tokyo, Japan). FI values of samples were measured at the excitation wavelength 480 nm, the emission wavelength 550 nm, and the excitation and emission band-pass 2.5 nm. The temperature of the cell was held at 25 C with a water bath circulator (Cool-Bit circulator, ACE-05AN, KELK (old name: Komatsu), Ltd., Tokyo, Japan). For 100% permeabilization of E. coli, 1.5 mL of the same suspension of E. coli without LfcinB (4-9) was pelleted by centrifugation (350  g, 10 min, H-18F, Kokusan Co. Ltd., Tokyo, Japan) and the pellet was resuspended in 2.0 mL 70% 2-propanol and allowed to stand at room temperature for 2.0 h. This E. coli suspension was pelleted by centrifugation (14000  g, 20 min, CF15R, Hitachi Koki Co. Ltd.) at 20 C and resuspended in 2.0 mL buffer A. To calculate SYTOX influx (%) quantitatively, the FI of this suspension under the above conditions was taken as 100% SYTOX influx, and the FI of the suspension in the absence of LfcinB (4-9) at t = 0 was taken as 0% SYTOX influx. In the case of Rh-LfcinB (49), the FI due to Rh-LfcinB (4-9) under the above conditions was subtracted from total FI of the suspensions. Investigation of the interactions of Rh-LfcinB (4-9) with single cells of E. coli containing calcein using confocal microscopy


Page 7 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

An E. coli suspension was subcultured on nutrient agar plates at 37 C for overnight to get single colonies. A single colony of bacteria was then grown in Nutrient Broth medium (0.5 colony/mL medium) for 6 to 7 h (exponential phase cells). The bacterial suspension was diluted to obtain a final bacterial concentration of 1106 CFU/mL, which was pelleted by centrifugation (350  g, 10 min, H-18F) and the pellet was washed using fresh media. For loading calcein in E. coli, the method of Dubey et al.32 was used. 50 µL of AM-calcein (1.0 µg/µL in DMSO solution) was added into 1.0 mL of the bacteria suspension, which was shook using a rotary shaker for 2.3 h at room temperature under dark condition. Then the suspension was centrifuged at 350  g for 10 min, and the pellet was resuspended in fresh, dye-free Nutrient Broth medium. This procedure was repeated twice, and finally the suspension was resuspended in buffer A. This E. coli suspension was transferred into a hand-made chamber,33 and after 15 min almost all E. coli cells were settled down onto the glass surface to be adsorbed strongly on it, and therefore they did not move. Interactions of peptides with single cells of E. coli containing calcein were observed using a confocal laser scanning microscope (CLSM) (FV-1000, Olympus, Tokyo, Japan) at 25 ± 1 °C with a stage thermocontrol system (Thermoplate, Tokai Hit, Shizuoka, Japan). For CLSM measurements, fluorescence images of calcein (473 nm laser), Rh-LfcinB (4-9) (559 nm laser), and differential interference contrast (DIC) images were obtained using a 60×objective (UPLSAPO060X0, Olympus) (NA = 1.35).25,37 During the interaction of peptides with single cells of E. coli, various concentrations of Rh-LfcinB (4-9) in buffer A were added continuously to the neighborhood of the single cells of E. coli through a 20-μm-diameter glass micropipette positioned by a micromanipulator.33,37 The distance between the single cells and the tip of the micropipette was 50 μm, and the applied pressure, P (= Pin  Pout, where Pin and Pout were the pressure of the inside and the outside of a micropipette, respectively) was 30 Pa.33,37 For the experiments of interaction of peptide with E. coli cells, we used buffer A instead of medium. The advantage of buffer is that it does not contain any substances interacting with the peptide, but the disadvantage is that it does not contain any nutrients for E coli cells. It is well known that bacteria such as E. coli can survive in buffer or water without nutrients.34 As a control experiment, we made a same experiment using buffer A containing 0.1 M glucose, and found that all the E coli cells moved rapidly in the buffer and thus did not settle down to the glass surface. Glucose is one of nutrients for E. coli cells.35,36 Therefore, we can consider that if there is no nutrient E. coli cells cannot move rapidly and hence they settle down to the glass surface, which induces their strong adsorption on it. To


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 33

examine whether E. coli cells adsorbed on the glass surface in buffer A were live or dead cells, we investigated the interaction of SYTOX green with these E. coli cells (Figure S1 in Supporting Information (SI)), because the SYTOX green staining is the standard method to distinguish dead cells from live cells. No fluorescence increase was detected in these cells, indicating that these cells were live cells. GUV preparations GUVs of DOPG/DOPC (1/1; molar ratio) membranes (i.e., PG/PC (1/1)-GUVs) were prepared by incubation of buffer A containing 0.10 M sucrose and 1.0 mM calcein or 6.0 M AF647 with dry lipid films by the natural swelling method at 37 C for 23 h.25 We prepared GUVs containing small vesicles inside the GUV lumen according to the method described in our previous paper.25 The membrane filtering method was used to remove untrapped fluorescent probes.38 Purified GUV suspension (0.10 M sucrose and 0.10 M glucose in buffer A as the internal solution and the external solution, respectively) was transferred into a hand-made microchamber.33 To prevent strong interaction between the glass surface and GUVs, the inside of the microchamber was coated with 0.10 %(w/v) BSA in the same buffer for the experiments.33 Investigation of the interactions of LfcinB (4-9) with single GUVs containing calcein We observed GUVs using an inverted fluorescence phase contrast microscope (IX-70, Olympus) at 25  1 C using the stage thermocontrol system (Thermoplate, Tokai Hit). Various concentrations of LfcinB (4-9) solution in buffer A containing 0.10 M glucose were continuously added in the vicinity of a GUV through a ~20-m-diameter glass micropipette positioned using a micromanipulator. The distance between the GUV and the tip of the micropipette was ~70 m, and the applied pressure, P, was 30 Pa.33,37 Phase-contrast and fluorescence images of GUVs were recorded using a high-sensitivity EM-CCD camera (C9100-12, Hamamatsu Photonics K.K., Hamamatsu, Japan) with a hard disk. Three ND filters were used to decrease the intensity of the incident light, resulting in conditions where almost no photobleaching of fluorescent probes in a GUV occurred during the interaction of the LfcinB (4-9) solution with single GUVs. Thus, under these conditions, the decrease in FI inside a GUV corresponded to leakage of the fluorescent probes from the inside to the outside of the GUV. The FI inside the GUVs was determined using the AquaCosmos software (Hamamatsu Photonics K.K.), and the average intensity per GUV was estimated. The details of this method were described in our previous reports.33,37


Page 9 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Investigation of the interactions of Rh-LfcinB (4-9) with single GUVs containing AF647 using CLSM For detection of entry of Rh-LfcinB (4-9) into single GUVs, we used PG/PC (1/1)-GUVs containing small vesicles of PG/PC (1/1) and AF647 inside the GUV lumen.25-27 On the other hand, for analysis of time course of rim intensity, we used PG/PC (1/1)-GUVs containing AF647, but not containing small vesicles.25-27 The GUVs were observed using the CLSM at 25  1 C using the stage thermocontrol system. Fluorescence images of GUVs due to AF647 (635 nm laser), Rh-LfcinB (4-9) (559 nm laser), and DIC images were obtained using the 60 objective.25 Interactions of Rh-LfcinB (4-9) with a single GUV were investigated by adding various concentrations of Rh-LfcinB (4-9) in buffer A containing 0.1 M glucose continuously to the neighborhood of the GUV using a 20-m-diameter glass micropipette. The distance between the GUV and the tip of the micropipette was ~50 m, and the applied pressure, P, was 30 Pa. The details of this experimental method and the analysis method to obtain the time course of the FI of GUVs were described in our previous paper.25 Interaction of Rh-LfcinB (4-9) and LfcinB (4-9) with DNA Appropriate amounts of DNA was dissolved in buffer A at 4 C overnight, and sonicated for 30 s intermittently 10 times at 4 C using a probe-type ultrasonic homogenizer (MISONIX, XL-2000, Newtown, CT), which produced fragments with length of less than1000 base pairs (confirmed by agarose gel electrophoresis). DNA concentration in buffer A was determined by the absorbance at 260 nm (Abs (260 nm) = 1.0 for 50 g/mL). A Hitachi F7000 spectrofluorometer was used for fluorescence measurements, and the temperature of the cell was held at 25 C with a water bath circulator. For the interactions of Rh-LfcinB (4-9) with DNA, emission spectra of samples were measured at the excitation wavelength of 573 nm and the emission wavelength from 580 to 620 nm, and the excitation and emission band-pass were 1.0 nm and 5.0 nm, respectively. First, 5.0 M Rh-LfcinB (4-9) in buffer A was measured. Then, after appropriate volumes of the DNA solutions were added with this Rh-LfcinB (4-9) solution, the mixtures were mixed with a stirrer in the cell for 3 min at 25 C, and then wavelength-scan measurements were performed. The dilution effect on the FI was corrected. For the interactions of LfcinB (4-9) with DNA, emission spectra of samples were measured at the excitation wavelength of 280 nm and the emission wavelength from 300 to 500 nm, and the excitation and emission band-pass


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 33

were 5.0 nm and 5.0 nm, respectively. First, 5.0 M LfcinB (4-9) in buffer A were measured. Then, after appropriate volumes of the DNA solutions were added into this LfcinB (4-9) solution, the mixtures were mixed with a stirrer in the cell for 3 min at 25 C, and then wavelength-scan measurements were performed. The dilution effect on the FI was corrected. For the interactions of sulforhodamine B with DNA, emission spectra of samples were measured at the excitation wavelength of 554 nm and the emission wavelength from 570 to 620 nm, and the excitation and emission band-pass were 5.0 nm and 5.0 nm, respectively. First, 5.0 M sulforhodamine B in buffer A were measured. Then, after appropriate volumes of the DNA solutions were added into this sulforhodamine B solution, the mixtures were mixed with a stirrer in the cell for 3 min at 25 C, and then wavelength-scan measurements were performed. The dilution effect on the FI was corrected.

■ RESULTS AND DISCUSSION Antimicrobial and hemolytic activity of LfcinB (4-9) and Rh-LfcinB (4-9) First we investigated the antimicrobial activity of the LfcinB (4-9) and Rh-LfcinB (4-9). Using the standard method, we determined the MIC of LfcinB (4-9) and Rh-LfcinB (4-9) against E. coli (JM-109) as 25  10 M (= 25 g/mL) and 5  1 M (= 8 g/mL), respectively. This result indicates that the antimicrobial activity of Rh-LfcinB (4-9) is greater than that of LfcinB (4-9), suggesting that the labelling of a fluorescent probe, rhodamine, increased the antimicrobial activity. It is reported that the MIC values against E. coli of LfcinB,14 LfcinB (4-14),23 and LfcinB (17-31)17 were 3 M (= 9 g/mL), 32 M (= 50 g/mL), and 20 M (= 40 g/mL), respectively, although the strains of E. coli are different. Hence, the antimicrobial activity of LfcinB (4-9) is larger than that of other shorter versions of LfcinB, but smaller than that of LfcinB, and among all the peptides Rh-LfcinB (4-9) has the largest antimicrobial activity. We also investigated the hemolytic activity of LfcinB (4-9) and Rh-LfcinB (4-9). Using the standard method, we observed 0 % and 3 % hemolysis for 25 M LfcinB (4-9) (= MIC) and 100 M LfcinB (4-9) (= 4MIC), respectively, indicating that hemolytic activity of LfcinB (4-9) was low. In the case of Rh-LfcinB (4-9), we observed 5 % and 7 % hemolysis for 5 M Rh-LfcinB (4-9) (= MIC) and 9 M Rh-LfcinB (4-9) (= 1.8MIC), respectively, indicating that


Page 11 of 33

hemolytic activity of Rh-LfcinB (4-9) was also low. However, at higher concentration of Rh-LfcinB (4-9), i.e., 20 M LfcinB (4-9) (= 4MIC), 20 % hemolysis was observed, which was higher than that of LfcinB (4-9). We consider that the relative high activity of hemolysis is due to the attachment of hydrophobic rhodamine to LfcinB (4-9). Influx of SYTOX green into cytoplasm of E. coli cells induced by LfcinB (4-9) and Rh-LfcinB (4-9) To clarify the target of LfcinB (4-9) and Rh-LfcinB (4-9) for its bactericidal activity against E. coli, we investigated permeabilization of the plasma membrane of E. coli by measuring the entry of the membraneimpermeant fluorescent probe, SYTOX green, into the bacterial cytoplasm. Notably, the FI of SYTOX green is elevated upon binding to nucleic acids; thus enhanced FI indicates influx of the probe into the cytoplasm.39,40 As shown in Figure 1, the influx of SYTOX green did not increase significantly during the interaction of 100 M LfcinB (4-9) with E. coli in the presence of SYTOX green. The SYTOX green influx values at 10 min were 0 %, 2 %, and 2 % at 25, 50, and 100 M (= 4MIC) LfcinB (4-9), respectively. In the case of Rh-LfcinB (4-9), we obtained similar results. Figure 1 shows that the influx of SYTOX green did not increase significantly during the interaction of 20 M Rh-LfcinB (4-9) with E. coli and that the SYTOX green influx values at 10 min were 3 %, 5 %, and 6 % at 5, 9, and 20 M (= 4MIC) Rh-LfcinB (4-9), respectively. In contrast, as we indicated previously,14 50 M LfcinB increased the FI of SYTOX green and 80 % SYTOX green influx was observed at 10 min (Figure 1).

Figure 1 100 SYTOX influx (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

80 60 40 20 0 0

200

400 Time (s)

600

Figure 1: Influx of SYTOX green into cytoplasms of E. coli cells. Time courses of influx of SYTOX green (%) are shown during the interaction of various concentrations of LfcinB (4-9) and Rh-LfcinB (4-9) with E. coli suspension at 25 C. (blue ▲) 100 M LfcinB (4-9) and (red ●) 20 M Rh-LfcinB (4-9). Bacterial density was 1.0  106 CFU/mL. For comparison, the data for 50 M LfcinB (○) is shown, which is reprinted from Ref. [14] with permission from the American Chemical Society.


Biochemistry

Figure 2 (A)

(B) Fluorescence Intensity

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 33

1.0 0.8 0.6 0.4 0.2 0.0 0

100 200 300 Time (s)

Figure 2: Membrane permeation of calcein from single PG/PC (1/1)-GUVs induced by LfcinB (4-9). (A) Leakage of calcein from a single PG/PC (1/1)-GUV was induced by 50.0 M LfcinB (4-9) in buffer A at 25 C. Fluorescence images (2) show that the calcein concentration inside the GUV remained constant after the addition of LfcinB (4-9). The numbers below each image show the time in seconds after the LfcinB (4-9) addition was started. Phase contrast images of the GUV at time 0 (1) and 370 s (3) are shown. The bar is 20 m. (B) The change in the normalized fluorescence intensity of the GUV over time shown in (A) (green line). We defined the normalized fluorescence intensity of the intact GUV before the initiation of the membrane permeation as 1.0. For comparison, the data for 5 M LfcinB (black line) is shown, which is reprinted from Ref. [14] with permission from the American Chemical Society.

The results of Figure 1 indicate that during the interaction of E. coli cells with LfcinB (4-9) and Rh-LfcinB (4-9) their plasma membranes were not significantly damaged and as a result SYTOX green did not enter their cytoplasm. These results contrast well with those of LfcinB that the plasma membranes of E. coli cells were rapidly damaged and as a result SYTOX green permeabilized through the plasma membranes to enter their cytoplasm. These results indicate that bactericidality due to LfcinB (4-9) and Rh-LfcinB (4-9) is not caused by the damage of E. coli plasma membrane. Calcein leakage from PG/PC (1/1)-GUVs induced by LfcinB (4-9) To elucidate the interaction of LfcinB (4-9) with plasma membranes of E. coli cells, we examined the interaction of LfcinB (4-9) with single GUVs of PG/PC (1/1) membranes containing a water-soluble fluorescent probe, calcein, and 0.10 M sucrose in their lumens using the single GUV method.26,33 During observation of a single GUV in buffer A containing 0.10 M glucose at 25 C using phase-contrast fluorescence microscopy, the tip of a micropipette was


Page 13 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

approached to the GUV and the LfcinB (4-9) solution was continuously provided from the micropipette into the neighborhood of the GUV from t = 0. First we investigated the interaction of single GUVs with 50.0 M LfcinB (49). Prior to LfcinB (4-9) addition, a GUV had a high phase contrast (Figure 2A (1)) due to the difference in the saccharide concentrations between the inside (0.10 M sucrose) and the outside (0.10 M glucose) of the GUV, and the GUV lumen also has a high FI (Figure 2A (2)), indicating that it contained a high concentration of calcein. After starting the addition of the 50.0 M solution of LfcinB (4-9), the FI of the GUV lumen did not change during the interaction (up to 6 min) (Figure 2A (2), B). After 370 s a phase-contrast image of the same GUV was the same as before the interaction (Figure 2A (3)). When the same experiments were carried out using 20 single GUVs, we obtained the same results. These results show that LfcinB (4-9) could not induce any leakage of calcein from GUV lumen, indicating that LfcinB (4-9) could not induce pore formation in PG/PC (1/1)-GUVs nor rupture of these GUVs. This strongly contrasts with the result of LfcinB; as indicated previously,14 low concentrations of LfcinB induced a rapid leakage from GUV lumens (see an example of 5.0 M LfcinB in Figure 2B). These results are in agreement with the experimental results obtained using SYTOX green (Figure 1); LfcinB (49) did not induce any entry of SYTOX green into the cytoplasm of E. coli cells, which suggests that this peptide does not induce appreciable damages in the bacterial plasma membrane, but LfcinB induced entry of SYTOX green into their cytoplasm. Entry of Rh-LfcinB (4-9) into single cells of E. coli without leakage of calcein The above results indicate that bactericidality due to LfcinB (4-9) is not caused by the damage of E. coli plasma membrane. Hence, one possibility is that this peptide enters the cytoplasm of E. coli and binds to DNA, or perhaps to other bacterial proteins, causing suppression of growth or cell death of E. coli. To examine the activity of LfcinB (4-9) to enter cytoplasm of E. coli, we investigated the interaction of Rh-LfcinB (4-9) with single cells of E. coli containing calcein using CLSM. First we investigated the interaction of 5.0 M Rh-LfcinB (4-9) with single cells of E. coli in buffer A at 25 C. The Rh-LfcinB (4-9) solution was continuously provided to the neighborhood of a cell of E. coli through a micropipette, so the Rh-LfcinB (4-9) concentration near the cell of E. coli became constant at the steady state, which was almost the same as that in the micropipette.37 There were two kinds of E. coli cells, nonseptating and septating cells. Figure 3A shows a typical result for non-septating cells. During the interaction of 5.0


Biochemistry

Figure 3 (C)

(A)

(B)

(D) 2000

2000 FI of total cell

FI of total cell

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 33

1000

0 0

200 400 Time (s)

1000

0

600

0

200 400 Time (s)

600

Figure 3: Interactions of Rh-LfcinB (4-9) with single cells of E. coli containing calcein. (A) (C) CLSM images due to calcein (left images with green color) and Rh-LfcinB (4-9) (right images with red color). The number on the left side of each image shows the time in seconds after the addition of 5.0 M Rh-LfcinB (4-9) was started. The bar is 2.0 m. The figure on the right side of each CLSM image shows the FI profile along the white line in the calcein image at 0 s, where green and red curves correspond to FI of calcein and of Rh-LfcinB (4-9), respectively. Superimposed images at final images ((A) 520 s and (C) 540 s) are also shown; (1) calcein and DIC images, (2) RhLfcinB (4-9) and DIC images, (3) only DIC image. (B) and (D) show change in the FI of the total cell of E. coli during the interaction of Rh-LfcinB (4-9) over time shown in (A) and (C), respectively. Green and red lines correspond to the FI of calcein and of Rh-LfcinB (4-9) in the cell of E. coli, respectively.


Page 15 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

M solution of Rh-LfcinB (4-9), the FI of E. coli cell due to calcein did not change during the experiment (up to 10 min) (Figure 3A and a green line in Figure 3B), indicating no pore formation in its plasma membrane through which calcein leaked. On the other hand, the FI of the total E. coli cell due to Rh-LfcinB (4-9) gradually increased with time to reach a steady value at ~70 s, which remained almost constant for 60 s, then rapidly increased from 120 s, and then after 164 s the FI gradually increased with time (Figure 3A and a red line in Figure 3B). The right figures in Figure 3A show the FI profiles along the white line in the image of Figure 3A (0 s of calcein image). At the beginning less than 187 s, the FI due to Rh-LfcinB (4-9) at the rim of E. coli cell corresponding to the E. coli membranes (i.e., the rim intensity) was larger than the FI at the central region of E. coli cell due to Rh-LfcinB (4-9) corresponding to the cytoplasm of E. coli, but after 187 s the FI at the central region became larger than the rim intensity. We consider that the initial weak intensity at the rim is due to the binding of Rh-LfcinB (4-9) to the bacterial membrane and the rapid increase in FI from 120 s is due to the entry of Rh-LfcinB (4-9) into the periplasm of E. coli and the following gradual increase in FI from 164 s is due to the entry of Rh-LfcinB (4-9) into the cytoplasm of E. coli. As the criterion of the entry of Rh-LfcinB (4-9) into the cytoplasm, we used here the condition that the FI at the central region of E. coli cell is larger than the rim intensity due to Rh-LfcinB (4-9). It is noted that the areas of the CLSM images of RhLfcinB (4-9) were larger than those of calcein. This fact shows that the FI due to the membranes of cells at their rim regions in the images of Rh-LfcinB (4-9) was much larger than that in the images of calcein, indicating that high concentrations of Rh-LfcinB (4-9) located at the membranes whereas calcein molecules did not locate near the membranes. The same experiments were performed using 20 cells of E. coli (i.e., 10 non-septating and 10 septating cells), and the entry of Rh-LfcinB (4-9) into the cytoplasm was observed in 9 cells (i.e., 4 non-septating and 5 septating cells). These results indicate that Rh-LfcinB (4-9) outside the E. coli cell translocated through the membranes of E. coli and entered its cytoplasm without leakage of calcein, i.e., Rh-LfcinB (4-9) has a cell-penetrating activity against E. coli. In contrast, for other cells of E. coli, the FI of the central region of the cells did not become larger than the rim intensity of the cells within 10 min (Figure 3C). The FI of total cell (Figure 3D) was lower than that of Figure 3B. This result suggests that Rh-LfcinB (4-9) outside the cell could not enter its cytoplasm. Figure 3D shows that the FI due to Rh-LfcinB (4-9) increased after 420 s, suggesting that Rh-LfcinB (4-9) entered into its periplasm. Therefore, these results indicate that for the interaction of 5.0 M Rh-LfcinB (4-9) with E. coli cells, the entry of Rh-LfcinB (4-


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 33

9) into its cytoplasm was observed in 45 % of total examined cells. Figure S2 in SI shows the images of all the examined E. coli cells and FI profiles along a line perpendicular to the long axis of the cells after the interaction with Rh-LfcinB (4-9) for 10 min. We also examined the concentration dependence of the entry of Rh-LfcinB (4-9) into single cells of E. coli without leakage of calcein. For 9.0 M Rh-LfcinB (4-9), among the examined cells (i.e., 20 cells of E. coli; 10 nonseptating and 10 septating cells), the entry of Rh-LfcinB (4-9) into its cytoplasm was observed in 14 cells (i.e., 7 nonseptating and 7 septating cells) (Figure S3 and S4 in SI). In contrast, for 2.0 M Rh-LfcinB (4-9), among the examined cells (i.e., 20 cells of E. coli ; 10 non-septating cells and 10 septating cells), the FI of the cytoplasm of E. coli cells was lower than that of the rim of the cells in 16 cells (7 non-septating and 9 septating cells), indicating that in most cells Rh-LfcinB (4-9) could not enter into their cytoplasm. Therefore, we conclude that the cell-penetrating activity of Rh-LfcinB (4-9) against E. coli increased with increasing its concentration. Entry of Rh-LfcinB (4-9) into single PG/PC (1/1)-GUVs without leakage of AF647 To examine the activity of LfcinB (4-9) to translocate across lipid membranes, we investigated whether Rh-LfcinB (4-9) can enter single GUVs of lipid membranes using the single GUV method for investigation of CPPs developed by us recently.25-27 For this aim, we observed using CLSM the interaction of Rh-LfcinB (4-9) with single PG/PC (1/1)-GUVs containing small vesicles composed of PG/PC (1/1) and AF647 in the GUV lumen. First we investigated the interaction of 5.0 M Rh-LfcinB (4-9) with single GUVs in buffer A containing 0.1 M glucose at 25 C. The RhLfcinB (4-9) solution was continuously provided to the neighborhood of the GUV through a micropipette. After starting the addition of the 5.0 M Rh-LfcinB (4-9), the FI of AF647 in the GUV lumen did not change during the experiment (Figure 4A (1) and red solid square in Figure 4B), indicating no pore formation in the lipid membrane through which AF647 leaked. On the other hand, the FI of the GUV membrane (i.e., the rim intensity) due to RhLfcinB (4-9) gradually increased and at 100 s it became almost steady, which remained constant during the interaction (Figure 4A (2) and green open squares in Figure 4B). At initial time of the interaction, there was no FI inside the GUV, but after 143 s, the membranes of the small vesicles inside the GUV emitted fluorescence (t = 143600 s in Figure 4A (2)). These results indicate that Rh-LfcinB (4-9) in the aqueous solution outside the GUV translocated through the GUV membrane and entered the GUV lumen without leakage of AF647, then bound to the membrane of


Page 17 of 33

the small vesicles. The same experiments were performed using 12 GUVs. The entry of Rh-LfcinB (4-9) into the GUV lumen within 10 min interaction was observed in 4 GUVs, and hence the fraction of entry of Rh-LfcinB (4-9) at 10 min without leakage of AF647, Pentry (10 min), was 0.25. To confirm reproducibility, we performed two independent experiments, and obtained similar results. The mean value and the standard error of Pentry (10 min) was 0.31  0.06. We also examined the concentration dependence of the entry of Rh-LfcinB (4-9) into a GUV without the leakage of AF647. Figure 4C shows the Rh-LfcinB (4-9) concentration dependence of Pentry (10 min). At Rh-LfcinB (4-9) concentrations of  2.0 M, the entry of Rh-LfcinB (4-9) into a GUV was not observed during 10 min interaction.

Figure 4 (A)

(B)

(C) 1.2

1.0

1.0

0.8

Fraction of Entry

Fluorescence Intensity

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

0.8 0.6 0.4 0.2 0.0

0.6 0.4 0.2 0.0

0

100 200 300 400 500 600 Time(s)

0

2 4 6 8 10 Rh-LfcinB (4-9) Conc. (M)

Figure 4: Entry of Rh-LfcinB (4-9) into single PG/PC (1/1)-GUVs containing small vesicles. (A) CLSM images of (1) AF647, (2) Rh-LfcinB (4-9), and (3) DIC. The numbers above each image show the time in seconds after the addition of 5.0 M Rh-LfcinB (4-9) was started. The bar is 20 m. (B) Change in the normalized fluorescence intensity of the GUV during the interaction of Rh-LfcinB (4-9) over time shown in (A). Red and green points correspond to the fluorescence intensity of AF647 in the GUV lumen and of Rh-LfcinB (4-9) in the rim of the GUV, respectively. (C) Dependence of the fraction of entry of Rh-LfcinB (4-9) without pore formation at 10 min on the Rh-LfcinB (4-9) concentration.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 33

At and above 5.0 M Rh-LfcinB (4-9), the entry of Rh-LfcinB (4-9) was observed in some examined GUVs, and Pentry (10 min) increased with an increase in Rh-LfcinB (4-9) concentration. At 9.0 M Rh-LfcinB (4-9), Pentry (10 min) was 0.77  0.08. As we described above, we interpreted the results of Figure 4 as follows: Rh-LfcinB (4-9) in the aqueous solution outside the GUV translocated through the GUV membrane and entered the GUV lumen without leakage of AF647, then bound to the membrane of the small vesicles.25-27 Someone may have other interpretations. One interpretation is that small vesicles inside a GUV lumen are connected to the mother GUV (i.e., the membranes of small vesicles are connected to the membrane of the mother GUV, the GUV rim) and hence first Rh-LfcinB (4-9) binds to the membrane of the mother GUV and diffuses into the membranes of small vesicles. However, the diffusion coefficient of lipid molecules in the liquid-crystalline phase is large and therefore Rh-LfcinB (4-9) diffuses into the membranes of small vesicles rapidly, which cannot explain a large lag time between the time when the FI at the mother GUV membrane reached a steady value (100 s) and the time when the membranes of the small vesicles started to emit fluorescence (143 s). Moreover, when we prepared PG/PC (1/1)-GUVs using the standard method (i.e., GUVs not containing small vesicles), we did not observe fluorescence due to small vesicles inside the GUVs (see e.g., Figure 5A) because most GUVs do not contain small vesicles in their lumens. The other interpretation is that Rh-LfcinB (4-9) induces endocytosis in the mother GUV to produce small vesicles inside the GUV lumen, which can emit fluorescence. However, if a large size vesicle (such as the largest vesicle with a diameter of ~10 m in the GUV lumen in Figure 4A(2)) would be produced via endocytosis from a mother GUV, then the area of the mother GUV membrane would decrease, inducing a decrease in diameter of the mother GUV. However, we did not observe a significant change in diameter of the mother GUVs (see Figure 4A as an example). Moreover, in the DIC image of the GUV before the interaction of Rh-LfcinB (4-9), small vesicles were observed inside the GUV lumen (e.g., Figure 4A (3) t = 0). As described later, when we investigated the interaction of Rh-LfcinB (4-9) with single GUVs not containing small vesicles, we could not observe fluorescence of small vesicles inside the GUV lumen. These results clearly deny the interpretation using the concept of the endocytosis. Analysis of kinetics of the entry of Rh-LfcinB (4-9) into single PG/PC (1/1)-GUVs


Page 19 of 33

To elucidate the kinetics of entry of Rh-LfcinB (4-9), we next examined and analyzed the time course of RhLfcinB (4-9) concentration in the GUV membrane, CM (t) [M], using the method reported in our previous papers.25,27 For this purpose, we examined the interaction of Rh-LfcinB (4-9) with single GUVs not containing small vesicles using the same method described in the above section, because in the experiments using the GUVs containing the small vesicles in the GUV lumen shown in Figure 4A some fluctuations of the intensity of the GUV membrane occurred due to the small vesicles near the GUV membrane.25 Since we cannot analyze the rapid increase in rim intensity shown in Figure 4B,25 here we used lower concentrations of Rh-LfcinB (4-9). During the interaction of 1.0 M Rh-LfcinB (4-9) with a single PG/PC (1/1)-GUV containing AF647 (Figure 5A (1)), a high concentration of AF647 inside the GUV remained essentially constant during the experiment (up to 10 min) (red circles in Figure 5B), indicating no leakage of AF647. The rim intensity gradually increased with time and at 350 s was almost steady (Figure 5A (2) and green squares in Figure 5B).

Figure 5 (A)

(B)

(C)

1.2

0.020

1.0 k app (s-1)


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

0.8 0.6 0.4

0.015 0.010 0.005

0.2 0

0.000

0

150

300 450 Time (s)

600

0.0

0.5 1.0 1.5 2.0 Rh-LfcinB (4-9) conc. (M)

2.5

Figure 5: Analysis of kinetics of the entry of Rh-LfcinB (4-9) into single PG/PC (1/1)-GUVs not containing small vesicles. (A) CLSM images of a GUV induced by 1.0 M Rh-LfcinB (4-9) due to (1) AF647 and (2) Rh-LfcinB (49). The numbers above each image show the time in seconds after the addition of Rh-LfcinB (4-9) was started. The bar corresponds to 20 m. (B) Change in normalized fluorescence intensity of the GUV over time shown in (A). Red circles and green squares correspond to the fluorescence intensity of AF647 in the GUV lumen and of Rh-LfcinB (49) in the rim of the GUV, respectively. The solid red line is the best fit curve using eq. 1. (C) Dependence of kapp on Rh-LfcinB (4-9) concentration. The time course of the rim intensity of 21-22 GUVs was measured (two independent experiments). The mean values and standard errors of kapp are shown. The solid red line is the best fit curve using eq. 2.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 33

We previously proposed elementary processes of the entry of CPPs such as TP10 and R9 into single GUVs,25,27 and here we applied the same elementary processes to the entry of Rh-LfcinB (4-9) into a GUV lumen (Figure S5 in SI). First, Rh-LfcinB (4-9) in the solution outside a GUV binds to the external monolayer of the GUV at its membrane interface, whose rate constant is kON. Next, Rh-LfcinB (4-9) translocates from the external to the internal monolayer, whose rate constant is kFF, and then unbinds from the internal monolayer into aqueous solution near the GUV membrane (where the Rh-LfcinB (4-9) concentration is Cin), whose rate constant is kOFF. Finally, Rh-LfcinB (4-9) diffuses into the GUV bulk lumen, whose rate constant is kdiff. We have to take into account for the backward reactions. After starting the interaction of Rh-LfcinB (4-9) with a GUV, the Rh-LfcinB (4-9) concentration in the neighborhood of the GUV elevated from zero to a constant, steady value, C eq [M], for a short time, and remained constant during out the interaction of Rh-LfcinB (4-9) with the single GUV.25,27 If the transfer of Rh-LfcinB (4-9) between the external and internal monolayers is fast, i.e., the rate of the transfer is faster than that of binding of Rh-LfcinB (4-9) and faster than that of unbinding from the membrane to aqueous solution, we can use a following equation of Rh-LfcinB (4-9) concentration in the GUV membrane, CM (t), which is held for the initial time (i.e., when the Rh-LfcinB (4-9) concentration in the GUV lumen is low):25

CM (t )  A [1  exp kappt ]

(1)

eq where kapp  kONCout / 2  kOFF

(2)

where kapp is the apparent rate constant of the elevation in CM (t) and A is a constant. The increase in the rim intensity of PG/PC (1/1)-GUVs (Figure 5B) over time was fit well by eq. 1 (the red line in Figure 5B) and gave a value for

kapp of 8.1  103 s1. We made the same experiments using different concentrations of Rh-LfcinB (4-9) and obtained the values of kapp. Figure 5C shows that kapp increased linearly with increasing C eq and the best fit of this out relationship by eq. 2 provided values of kON = (1.4  0.1)  104 M1s1 and kOFF = (2.3  0.4)  103 s1. Therefore, the binding constant of Rh-LfcinB (4-9) to the membrane, KB, is determined by KB =kON/kOFF: hence KB = (6  1)  106 M1 for PG/PC (1/1)-GUVs. Interaction of Rh-LfcinB (4-9) and LfcinB (4-9) with DNA After the entry of Rh-LfcinB (4-9), the FI of the E. coli cells became higher than that of outside of the cells. We can consider many possibilities of this increase in FI. Here we investigated the effect of the interaction of Rh-LfcinB


Page 21 of 33

(4-9) with DNA on its FI due to Rh. Figure 6A shows the fluorescence spectra of the mixtures of 5.0 M Rh-LfcinB (4-9) and various concentrations of DNA. The FI of Rh-LfcinB (4-9) increased with an increase in DNA concentration (Figure 6B). For example, the maximum FI of Rh-LfcinB (4-9) in the presence of 4.0 mg/mL DNA was 8.3 times larger than that in the absence of DNA. A blue shift of the peak was observed with an increase in DNA concentration. This result indicates that the interaction of Rh-LfcinB (4-9) with DNA, probably the binding of this peptide to DNA, increases greatly the FI of Rh-LfcinB (4-9), which can be explained by the decrease in solvent polarity of the environment of fluorescent probes.41 This result also indicates that the rate of binding of this peptide to DNA was large. Next, we examined the effect of the interaction of LfcinB (4-9) with DNA on its FI due to tryptophan (W). Figure 6C shows the fluorescence spectra of the mixtures of 5.0 M LfcinB (4-9) and various concentrations of DNA. The FI of LfcinB (4-9) greatly decreased with an increase in DNA concentration. At 0.20 mg/mL DNA, the FI of

Figure 6 (A)

(B) Fluorescence Intensity

30

25 20 15 10 5 0 580

590

600 610 Wavelength (nm)

160 120

40 0 300

20 15 10 5 0

1 2 3 4 DNA conc. (mg/mL)

0.0

0.2 0.4 DNAconc. (mg/mL)

(D)

(C)

80

25

0

620

DNA conc. (mg/mL) 0.00 0.02 0.06 0.17 0.58

350 400 Wavelength (nm)



30

Fluorescence intensity (a.u.)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

5

150 100 50 0 0.6

Figure 6: Interaction of Rh-LfcinB (4-9) and LfcinB (4-9) with DNA detected by the measurements using fluorescence spectroscopy. (A) Fluorescence spectra of Rh-LfcinB (4-9) in buffer A after mixing with various concentrations of DNA solution are shown. Effects of dilution were corrected; final Rh-LfcinB (4-9) concentration was 5.0 M, and final concentrations of DNA (mg/mL) were 0, 1.1, 2.0, 2.6, 3.2, 4.0, 4.6 mg/mL (from lower to upper). (B) Dependence of FI of Rh-LfcinB (4-9) on DNA concentration. (C) Fluorescence spectra of LfcinB (4-9) in buffer A after mixing with various concentrations of DNA solution are shown. Effects of dilution were corrected; LfcinB (4-9) concentration was 5.0 M, and final concentrations of DNA (mg/mL) were indicated at the right of each curve. (D) Dependence of FI of LfcinB (4-9) on DNA concentration.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 33

LfcinB (4-9) became 10 % of the initial intensity, and at 0.40 mg/mL it became almost zero (Figure 6D). This result can be explained by quenching of fluorescence of Trp by DNA base, which have been reported in the interaction between DNA-binding proteins with DNA.41-44 Figure S4 in SI shows the quenching of Trp fluorescence as a function of the ratio of nucleotides of DNA to LfcinB (4-9), which was obtained by replotting Figure 6D. This result indicates that LfcinB (4-9) strongly binds to DNA. If we compare the results of the interactions of Rh-LfcinB (4-9) and LfcinB (4-9) with DNA molecules, it is evident that LfcinB (4-9) completely bound with lower concentration of DNA, indicating that the binding constant of LfcinB (4-9) with DNA is larger than that of Rh-LfcinB (4-9) with DNA. The bulky group of Rh may hinder the strong binding of the peptide with DNA. We also examined the effect of the interaction of sulforhodamin B, which has a similar structure of LRB Red, with DNA. The FI of 5.0 M sulforhodamin B did not increase significantly with an increase in DNA concentration. For example, 4.0 mg/mL DNA increased the FI of sulforhodamin B by only 25 %. Hence, this result indicates that the interaction of the fluorescence probe of Rh-LfcinB (4-9) with DNA is small. Therefore, we can conclude that the binding of Rh-LfcinB (4-9) with DNA occurs due to the interaction of the peptide part of Rh-LfcinB (4-9) with DNA.

■ GENERAL DISCUSSION The above results show that Rh-LfcinB (4-9) entered cytoplasm of single cells of E. coli and lumen of single GUVs without damages of these membranes, indicating that Rh-LfcinB (4-9) is one of CPPs.45-49 Therefore, we can conclude that its antimicrobial activity is not due to damage of plasma membrane of E. coli cells. The cell-penetrating activity of Rh-LfcinB (4-9) increased with an increase in its concentration, and it became large at and above 5.0 M for 10 min interactions. This concentration range is a little larger than the MIC value of Rh-LfcinB (4-9) against E.

coli (i.e., 5  1 M). Therefore, this result suggests that entry of Rh-LfcinB (4-9) into the cytoplasm of E. coli cells might induce suppression of growth or cell death of E. coli, although there are other possibilities of the causes of suppression of growth or cell death of E. coli; such as the binding of the peptide with various components in its outer membranes.50 One of the targets of Rh-LfcinB (4-9) in the E. coli cytoplasm is DNA. The data in this report clearly indicate that Rh-LfcinB (4-9) can bind to DNA and this binding increases the FI of Rh-LfcinB (4-9) greatly. The previous report indicated that polycationic CPPs such as TAT and R9 form complexes with DNA,51 supporting our


Page 23 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

results in this report. Therefore, the binding of Rh-LfcinB (4-9) to DNA is a candidate of suppression of growth or cell death of E. coli, which was suggested in the case of buforin II.8 However, we can consider that the bindings of Rh-LfcinB (4-9) to ribosomes and proteins such as DNA gyrase inside the cytoplasm are also these candidates. In contrast, the results in this report did not directly indicate that LfcinB (4-9) can enter the cytoplasm of E. coli cells. However, the data in this report clearly indicate that LfcinB (4-9) can bind to DNA. If the suppression of growth and the cell death of E. coli is related to the entry of LfcinB (4-9) into its cytoplasm, the fact that the MIC of LfcinB (4-9) is larger than that of Rh-LfcinB (4-9) can be explained by the labeling of fluorescent probe (Rh) to the peptide: the attachment of a hydrophobic fluorescent probe affects the interaction of LfcinB (4-9) with lipid bilayers and increases the cell penetrating activity, which has been suggested in the case of CPPs.52 Many studies on shorter versions of LfcinB have been reported. Interactions of LfcinB (4-9) with lipid membranes were investigated using the characteristics of Trp fluorescence and its quenching by fluorescence spectroscopy, and their results indicate that Trp side chain of this peptide locates at near membrane interface.20 11-residue linear fragment of LfcinB (i.e., LfcinB (4-14)) had a strong antimicrobial activity, but it did not induce any leakage of calcein from LUVs of various lipid compositions (egg PE/egg PG (1/1), and egg PE/egg PC (1/1)).23 The structure of LfcinB (4-14) in SDS micelles indicates that it does not form an -helix but it has an amphipathic characters,23 which is similar to LfcinB.53 15-residue linear fragment of LfcinB (i.e., LfcinB (17-31) ) has also a strong antimicrobial activity, although mutants of LfcinB (4-14) whose Trp6 or Trp8 was replaced with Ala did not have an antimicrobial activity.17 LfcinB (17-31) and its mutants did not induce significant leakage of calcein from LUVs of various lipid compositions (palmitoyl oleoyl PG (POPG), palmitoyl oleoyl PC (POPC), POPG/POPC (1/1)), but differential scanning calorimetry (DSC) studies indicate that these peptides strongly bind to negatively charged PG membranes.21 Most results indicate that these shorter fragments of LfcinB did not damage lipid membranes to induce leakage of internal contents of vesicles, which support the conclusion in this report. Very recently, it has been reported that 5-residue linear fragment of LfcinB (i.e., LfcinB (4-8)) can enter human lung cancer A549 cells,54 which supports our conclusion that the short fragments of LfcinB are one of CPPs. Next we consider the mechanism of entry of Rh-LfcinB (4-9) into vesicles of lipid membranes. The results in this report clearly indicate that this peptide entered single GUVs without leakage of AF647. For the mechanisms of the translocation of CPPs across lipid membranes, several models have been proposed. One model is the entry of CPPs


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 33

through pores in lipid membranes (model A),55-57 where CPPs induce the formation of pores through which watersoluble fluorescent probe such as AF633 leaks from vesicles and then these CPPs enter the vesicles through the pores. Model B is the translocation of CPPs through the formation of inverted micelles in the membrane, where CPPs form a neutral complex with negatively charged lipids.58,59 However, the internalization of CPPs (i.e., translocation) occurred only in lipid membranes containing very high concentrations of negatively charged lipids ( 90 mol%).58,59 Recently, another model of the translocation of CPPs across lipid membranes has been proposed: translocation of CPPs occurs through transient hydrophilic prepores in the membrane (model C).27,60 Entry of transportan 10 and oligoarginine, such as R9, into vesicles of most lipid membranes belongs to the model C. Generally, thermal fluctuations of lipid membranes in the liquid-crystalline phase are large, producing fluctuation of their local lateral density such as rarefaction (i.e., area of decreased density) and condensation (i.e., area of increased density). Each area of rarefaction can be considered as a prepore with an effective radius.61 When the radius of a rarefaction becomes above a threshold value, a hydrophilic prepore is formed, which has a toroidal structure where hydrophilic segments of lipid molecules contact with water at the wall of the prepore.62-64 Such prepores are unstable due to large line tension at the edge of prepores, and hence immediately close. According to the theory of tension-induced pore formation, if the radius of hydrophilic prepore reaches a critical radius the prepore transforms into a pore, which can induce leakage of water-soluble fluorescent probes.57-60 However, the energy barrier or the activation energy of the transformation of a prepore to a pore is so large that pore formation does not occur without large tension.64-69 In the model C, because the structure of a prepore is a toroidal-like structure, CPP molecules bound to negatively charged lipids in the external monolayer can diffuse to the wall of a toroidal prepore, and then into the internal monolayer. The diameter of this prepore is smaller than that of the fluorescent probe AF647, and hence no leakage of AF647 occurs during the translocation of CPP molecules across the bilayer. It is reported that application of voltage induced fluctuation of small electric currents in a planar bilayer interacting R9 molecules.70 This type of currents can be considered to occur through transient prepores because ions, which are much smaller than the fluorescent probes, can pass through the prepores. The results using molecular dynamics simulation indicated that R8 can translocate across the bilayer via pores formed spontaneously due to thermal fluctuations of the membrane.71 These pores correspond to the prepores mentioned above. Based on the results in this report, we have a hypothesis that the entry of Rh-LfcinB (4-9) into PG/PC (1/1)-GUV lumens can be explained by the model C: the translocation of Rh-LfcinB (4-9) across a


Page 25 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

lipid membrane occurs through transient hydrophilic prepores in the membrane. However, to verify this hypothesis, we need more experimental data to elucidate the mechanism. After the entry of Rh-LfcinB (4-9), the FI of the E. coli cells became higher than that of outside of the cells. Similar increase in FI of fluorescent probe-labeled CPPs and AMPs were reported.7,8 However, its mechanism is unknown. The result of this report indicates that the binding of Rh-LfcinB (4-9) to DNA increased its FI greatly, which can be explained by the decrease in solvent polarity of the environment of fluorescent probes.41 In the cytoplasm of E. coli, there are a circular genome DNA and plasmids, and therefore the interaction of Rh-LfcinB (49) with these DNA molecules can increase its FI greatly. Moreover, the binding of Rh-LfcinB (4-9) to proteins and ribosomes may increase its FI due to the same reason described above. On the basis of the present study, it may be necessary to reconsider the interaction of LfcinB with lipid membranes and E. coli cells. Because LfcinB (4-9) is a part of LfcinB, LfcinB may have an activity of translocation of lipid membranes through transient hydrophilic prepores and thus an activity of entry into the cytoplasm of E. coli. It is reported that TP10, one of CPPs, can induce pore formation after it translocates across lipid membranes through transient hydrophilic prepores.25 We need further investigation on the interaction of fluorescent probe-labeled LfcinB with single GUVs and E. coli cells.

■ CONCLUSION In this report, we found that Rh-LfcinB (4-9) translocated continuously across lipid bilayers of single GUVs of pure lipid membranes (i.e., PG/PC (1/1)) and entered their lumens without leakage of AF647 for the first time, to the best of our knowledge. This phenomena is the same as that of oligoarginine such as R9, one of CPPs.27 The elementary processes of the entry of Rh-LfcinB (4-9) into single GUVs were revealed; the rate constant of the binding of RhLfcinB (4-9) with PG/PC (1/1) membrane and that of its unbinding from the membrane to aqueous solution were determined. We also found that Rh-LfcinB (4-9) entered cytoplasm of single cells of E. coli without damages of their plasma membranes, indicating that Rh-LfcinB (4-9) has a cell-penetrating activity against E. coli cells. On the other hand, we found that the interaction of Rh-LfcinB (4-9) with DNA increased its fluorescence intensity greatly, which may explain the result that the FI due to Rh-LfcinB (4-9) in the E. coli cytoplasm was higher than that of outside of the E. coli cell after the entry of Rh-LfcinB (4-9). Based on these results, we can reasonably consider that Rh-LfcinB


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 33

(4-9) can translocate through lipid membrane regions of the plasma membrane of E. coli cells and that its antimicrobial activity is not due to damage of plasma membrane of E. coli.

Supporting Information (SI) Data of the interaction of SYTOX green with single cells of E. coli adsorbed on a glass surface in buffer, the interactions of 5.0 and 9.0 M Rh-LfcinB (4-9) with single cells (non-septating and septating cells ) of E. coli containing calcein, and a scheme showing the elementary processes for the entry of CPPs into a GUV.

REFERENCES (1) Zasloff, M. (2002) Antimicrobial peptides of multicellular organisms. Nature 415, 389-395. (2) Melo, M. N., Ferre, R., and Castanho, A. R. B. (2009) Antimicrobial peptides: linking partition, activity and high membrane-bound concentrations. Nat. Rev. Microbiol. 8, 1-5. (3) Hwang, P. M., and Vogel, H. J. (1998) Structure-function relationships of antimicrobial peptides. Biochem. Cell

Biol. 76, 235-246. (4) Wade, D., Boman, A., Wahlin, B., Drain, C. M., Andreu, A., Boman, H. G., and Merrifield, R. B. (1990) All-D amino acid-containing channel-forming antibiotic peptides. Proc. Natl. Acad. Sci. USA. 87, 4761-4765. (5) Matsuzaki, K., Murase, K., Fujii, N., and Miyajima, K. (1995) Translocation of a channel-forming antimicrobial peptide, magainin 2, across lipid bilayers by forming a pore. Biochemistry 34, 6521-6526. (6) Tamba, Y., and Yamazaki, M. (2005) Single giant unilamellar vesicle method reveals effect of antimicrobial peptide Magainin 2 on membrane permeability. Biochemistry 44, 15823-15833. (7) Sochacki, K. A., Barns, K. J., Bucki, R., and Weisshaar, J. C. (2011) Real-time attack on single Escherichia coli cells by the human antimicrobial peptide LL-37. Proc. Natl. Acad. Sci. USA. 108, E77-E81. (8) Park, C. B., Yi, K.-S., Matsuzaki, K., Kim, M. S., and Kim, S. C. (2000) Structure-activity analysis of buforin II, a histone H2A-derived antimicrobial peptide: The proline hinge is responsible for the cell-penetrating ability of buforin II. Proc. Natl. Acad. Sci. USA. 97, 8245-8250.


Page 27 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

(9) Bellamy, W., Takase, M., Yamauchi, K., Wakabayashi, H., Kawase, K., and Tomita M. (1992) Identification of the bacterial domain of lactoferrin, Biochim. Biophys. Acta 1121, 130-136. (10)

Masson, P. L., Heremans, J. F., and Dive, C. H. (1966) An iron-binding protein common to many external

secretions, Clin. Chim. Acta 13, 735-739. (11)

Tomita, M., Wakabayashi, H., Shin, K., Yamauchi, K., Yaeshima, T., and Iwatsuki, K. (2009) Twenty-five

years of research on bovine lactoferrin applications. Biochimie, 91, 52-57. (12)

Gifford, J. L., Hunter, H. N., and Vogel, H. J. (2005) Lactoferricin: a lactoferrin-derived peptide with

antimicrobial, antiviral, antitumor, and immunological properties. Cell. Mol. Life. Sci. 62, 2588-2598. (13)

Vogel, H. J., Schibli, D. J., Jing, W., Lohmeirer-Vogel, E. M. Epand, R. F., and Epand, R. M. (2002)

Towards a structure-function analysis of bovine lactoferricin and related tryptophan- and arginine-containing peptides. Biochem. Cell. Biol. 80, 49-63. (14)

Moniruzzaman, M., Alam, J. M., Dohra, H., and Yamazaki, M. (2015) Antimicrobial peptide lactoferricin

B-induced rapid leakage of internal contents from single giant unilamellar vesicles. Biochemistry, 54, 5802-5814. (15)

Tomita, M., Takase, M., Bellamy, W., and Shimamura, S. (1994) A review: The active peptide of lactoferrin,

Acta Paediatr. J. 36, 585-591. (16)

Kang, J. H., Lee, M. K., Kim, K. L., and Hahm, K. –S. (1996) Structure-biological activity relationships of

11-residue highly basic peptide segment of bovine lactoferrin, Int. J. Pept. Protein Res., 48, 357-363. (17)

Rekdal, Ǿ., Andersen, J., Vorland, L. H., and Svendsen, J. S. (1999) Construction and synthesis of

lactoferricin derivatives with enhanced antibacterial activity, J. Pept. Sci. 5, 32-45. (18)

Wakabayashi, H., Matsumoto, H., Hashimoto, K., Teraguchi, S., Takase, M., and Hayasawa, H. (1999) N-

acylated and D enantiomer derivatives of a nonamer core peptide of lactoferricin B showing improved antimicrobial activity, Antimicrob. Agents Chemothera. 43, 1267-1269. (19)

Strǿm, M. B., Rekdal, Ǿ., and Svendsen, J. S. (2000) Antibacterial activity of 15-residue lactoferricin

derivatives, J. Pept. Res. 56, 265-274. (20)

Schibli, D. J., Epand, R. F., Vogel, H. J., and Epand, R. M. (2002) Tryptophan-rich antimicrobial peptides:

comparative properties and membrane interactions, Biochem. Cell Biol. 80, 667-677.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(21)

Page 28 of 33

Strǿm, M. B., Haug, B. E., Rekdal, Ǿ., Skar, M. L., Strensen, W., and Svendsen, J. S. (2002) Important

structural features of 15-residue lactoferricin derivatives and methods for improvement of antimicrobial activity,

Biochem. Cell Biol. 80, 65-74. (22)

Wakabayashi, H., Takase, M., and Tomita, M. (2003) Lactoferricin derived from milk protein lactoferrin,

Cur. Pharm. Design, 9, 1277-1287. (23)

Nguyen, L. T., Schibli, D., and Vogel, H. J. (2005) Structural studies and model membrane interactions of

two peptides derived from lactoferricin, J. Peptide Sci., 11, 379-389. (24)

Jing, W., Svendsen J. S., and Vogel, H. J. (2006) Comparison of NMR structures and model-membrane

interactions of 15-residue antimicrobial peptide derived from bovine lactoferricin, Biochem. Cell Biol. 84, 312326. (25)

Islam, M. Z., Ariyama, H., Alam, J. M., and Yamazaki, M. (2014) Entry of cell-penetrating peptide

transportan 10 into a single vesicle by translocating across lipid membrane and its induced pores. Biochemistry,

53, 386-396. (26)

Islam, M. Z., Alam, J. M., Tamba, Y., Karal, M. A. S., and Yamazaki, M. (2014) The single GUV method

for revealing the functions of antimicrobial, pore-forming toxin, and cell-penetrating peptides or proteins. Phys.

Chem. Chem. Phys., 16, 15752-15767. (27)

Sharmin, S., Islam, M. Z., Karal, M. A. S., Shibly, S. U. A., Dohra, H., and Yamazaki, M. (2016) Effects

of lipid composition on the entry of cell-penetrating peptide oligoarginine into single vesicles. Biochemistry, 55, 4154-4165. (28)

Masum, S. M., Li, S. J., Tamba, Y., Yamashita, Y., Tanaka, T., and Yamazaki, M. (2003) Effect of de Novo

Designed Peptides Interacting with the Lipid-Membrane Interface on the Stability of the Cubic Phases of the Monoolein Membrane. Langmuir, 19, 4745-4753. (29)

Balouiri, M., Sadiki, M., and Ibnsouda, S. K. (2016) Methods for in vitro evaluating antimicrobial activity:

A review. J. Pharm. Anal. 6, 71-79. (30)

Andrews, J. M. (2001) Determination of minimum inhibitory concentrations. J. Antimicro. Chemo. 48, S1,

5-16.


Page 29 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(31)

Biochemistry

Yamaji, A., Sekizawa, Y., Emoto, K., Sakuraba, H., Inoue, K., Kobayashi, H., and Umeda, M. (1996)

Lysenin, a novel sphingomyelin-specific binding proteins. J. Biol. Chem. 273, 5300-5306. (32)

Dubey, G. P. and Ben-Yehuda, S. (2011) Intercellular nanotubes mediate bacterial communication. Cell,

144, 590-600. (33)

Yamazaki, M. (2008) The single GUV method to reveal elementary processes of leakage of internal

contents from liposomes induced by antimicrobial substances, Adv. Planar Lipid Bilayers and Liposomes, 7, 121-142. (34)

Liao, C. –H., Shollenberger, L. M. (2003) Survivability and long-term preservation of bacteria in water and

in phosphate-buffered saline, Lett. Applied. Microbiol. 37, 45-50. (35)

Groat, R. G., Schultz, J. E., Zychlinsky, E., Bockman, A., Matin, A. (1986) Starvation proteins in

Escherichia coli: kinetics of synthesis and role in starvation survival. J. Bacteriol. 168, 486-493. (36)

Sezonov, G., Joseleau-Petit, D., D’Ari, R. (2007) Escherichia coli physiology in Luria-Bertani broth. J.

Bacteriol. 189, 8746-8749. (37)

Karal, M. A. S., Alam, J. M., Takahashi, T., Levadny, V., and Yamazaki, M. (2015) Stretch-Activated Pore

of Antimicrobial Peptide Magainin 2, Langmuir, 31, 3391-3401. (38)

Tamba, Y., Terashima, H., and Yamazaki, M. (2011) A membrane filtering method for the purification of

giant unilamellar vesicles. Chem. Phys. Lipids 164, 351-358. (39)

Roth, B. L., Poot, M., Yue, S. T., and Millard, P. J. (1997) Bacterial viability and antibiotic susceptibility

testing with SYTOX green nucleic acid stain. App. Env. Microbiol. 63, 2421-2431. (40)

Mukherjee, S., Zheng, H., Derebe, M. G., Callenberg, K. M., Patch, C. L., Rollins, D., Propheter, D., Rizo,

J., Grabe, M., Jiang, Q.-X., and Hooper, L. V. (2014) Antibacterial membrane attack by a pore-forming intestinal C-type lectin, Nature 305, 103-107. (41)

Lakowicz, J. R. (2006) Principles of fluorescence spectroscopy, 3rd ed., Springer, New York.

(42)

Toulmé, J.-J., and Hélène, C. (1977) Specific recognition of single-stranded nucleic acids. J. Biol. Chem.

252, 244-249. (43)

Kim, C., Paulus B., and Wold, M. S. (1994) Interactions of human replication protein A with

oligonucleotides. Biochemistry, 33, 14197-14206.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(44)

Page 30 of 33

Kozlov, A. G., Galletto, R., and Lohman, T. M. (2012) SSB-DNA binding monitored by fluorescence

intensity and anisotropy. Methods Mol Biol. 922, 55-83. (45)

Magzoub, M., and Gräslund, A. (2004) Cell-penetrating peptides: small from inception to application.

Quart. Rev. Biophys. 37, 147-195. (46)

Zorko, M., and Langel, Ű. (2005) Cell-penetrating peptides: mechanism and kinetics of cargo delivery. Adv.

Drug. Deliv. Rev. 57, 529-545. (47)

Koren, E., and Torchilin, V. P. (2012) Cell-penetrating peptides: breaking through to the other side. Trends.

Mol. Med. 18, 385-393. (48)

Stanzl, E. G., Trantow, B. M., Vargas, J. R., and Wender P. A. (2013) Fifteen years of cell-penetrating,

guanidinium-rich molecular transporters: basic science, research tools, and clinical applications. Acc. Chem. Res. 46, 2944-2954. (49)

Pisa, M. D., Chassaing, G., and Swiecicki, J.-M. (2015) Translocation mechanism(s) of cell-penetrating

peptides: biophysical studies using artificial membrane bilayers. Biochemistry, 54, 194-207. (50)

Farnaud, S., Spiller, C., Moriarty, L. C., Patel, A., Gant, V., Odell, E. W., and Evans, R. W. (2004)

Interactions of lactoferricin-derived peptides with LPS and antimicrobial activity. FEMS Micro. Lett. 233, 193199. (51)

Schmidt, N. W., Jin, F., Lande, R., Curk, T., Xian, W., Lee, C., Frasca, L., Frenkel, D., Dobnikar, J., Gilliet,

M., and Wong, G. C. L. (2015) Liquid-crystalline ordering of antimicrobial peptide-DNA complexes controls TLR9 activation. Nature Mat., 14, 696-700. (52)

Bechara, C., and Sagan, S. (2013) Cell-penetrating peptides: 20 years later, where do we stand? FEBS Lett.

587, 1693-1702. (53)

Hwang, P. M., Zhou, N., Shan, X., Arrosmith, C. H., and Vogel, H. J. (1998) Three-dimensional solution

structure of lactoferricin B, an antimicrobial peptide derived from bovine lactoferrin, Biochemistry 37, 4288-4298. (54)

Liu, B. R.; Huang, Y.-W., Aronstam, R. S., and Lee. H. –J. (2016) Identification of a short cell-penetrating

peptide from Bovine lactoferricin for intracellular delivery of DNA in human A 549 cells, Plos one, 11, e0150439.


Page 31 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(55)

Biochemistry

Mishra, A., Gordon, V. D., Yang, L., Coridan, R., and Wong, G. C. L. (2008) HIV TAT forms pores in

membranes by inducing saddle-spray curvature: potential role of bidentate hydrogen bonding. Angew. Chem. Int.

Ed., 47, 2986-2989. (56)

Ciobanasu, C., Siebrasse, J. P., and Kubitscheck, U. (2010) Cell-penetrating HIV1 TAT peptides can

generate pores in model membranes. Biophys. J. 99, 153-162. (57)

Mishra, A., Lai, G. H., Schmidt, N. W., Sun, V. Z., Rodriguez, A. R., Tong, R.;, Tang, L., Cheng, J., Deming,

T. J., Kamei, D. T., and Wong, G. C. L. (2011) Translocation of HIV TAT peptide and analogues induced by multiplexed membrane and cytoskeletal interactions, Proc. Natl. Acad. Sci. USA, 108, 16883-16888. (58)

Berlose, J.-P.; Convert, O.; Derossi, D.; Brunissen, A.; Chassaing, G. Conformational and associative

behaviours of the third helix of antennapedia homeodomain in membrane-mimetic environments. Eur. J. Biochem., 1996, 242, 372-386. (59)

Swiecicki, J.-M.; Bartsch, A.; Tailhades, J.; Di Pisa, M.; Heller, B.; Chassaing, G.; Mansuy, C.; Burlina,

F.; Lavielle, S. The efficacies of cell-penetrating peptides in accumulating in large unilamellar vesicles depend on their ability to form inverted micelles. ChemBioChem. 2014, 15, 884-891. (60)

Islam, M. Z., Sharmin, S., Levadnyy, V., Shibly, S. U. A., and Yamazaki, M. (2017) Effects of Mechanical

Properties of Lipid Bilayers on Entry of Cell-Penetrating Peptides into Single Vesicles, Langmuir, 33, 2433-2443. (61)

Levadny, V., Tsuboi, T., Belaya, M.;, and Yamazaki, M. (2013) Rate Constant of Tension-Induced Pore

Formation in Lipid Membranes. Langmuir 29, 3848-3852. (62)

Glaser, R. W., Leikin, S. L., Chernomordik, L. V., Pastushenko, V. F., and Sokirko, A. I. (1988) Reversible

electrical breakdown of lipid bilayers: formation and evolution of pores. Biochim. Biophys. Acta, 940, 275-287. (63)

Wohlert, J., den Otter, W. K., Edholm, O., and Briels, W. J. (2006) Free energy of a trans-membrane pore

calculated from atomistic molecular dynamics simulations. J. Chem. Phys. 124, 154905. (64)

Karal, M. A. S., Levadnyy, V., and Yamazaki, (2016) M. Analysis of Constant Tension-Induced Rupture

of Lipid Membranes Using Activation Energy. Phys. Chem. Chem. Phys. 18, 13487-13495. (65)

Fuertes, G. , Giménez, D., Esteban-Martín, S., Sánchez-Muñoz, O. L., and Salgado, J. (2011) A lipocentric

view of peptide-induced pores. Eur. Biophys. J. 40, 399-415.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(66)

Page 32 of 33

Deryagin, B. V., and Gutop, Y. V. (1962) Theory of the breakdown (rupture) of free films. Kolloidn. Zh.

(rus), 24, 370- 374. (67)

Lister, J. D. (1975) Stability of lipid bilayers and red blood cell membranes, Phys. Lett. A 53, 193-194.

(68)

Evans, E., Heinrich, V., Ludwig, F., and Rawicz, W. (2003) Dynamic tension spectroscopy and strength of

biomembranes. Biophys. J. 85, 2342-2350. (69)

Evans, E., and Smith, B. A. (2011) Kinetics of hole nucleation in biomembrane rupture. New J. Phys. 13,

095010. (70)

Herce, H. D., Garcia, A. E., Litt, J., Kane, R. S., Martin, P., Enrique, N., Rebolledo, A., and Milesi, V.

(2009) Arginine-rich peptides destabilizes the plasma membrane, consistent with a pore formation translocation mechanism of cell-penetrating peptides. Biophys. J. 97, 1917-1925. (71)

Sun, D., Forsman, J., and Woodward, C. E. (2015) Atomistic molecular simulations suggest a kinetic model

for membrane translocation by arginine-rich peptides. J. Phys. Chem. B 119, 14413-14420.


Page 33 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

For Table of Contents Use Only

Entry of 6-Residue Antimicrobial Peptide Derived from Lactoferricin B into Single Vesicles and E. coli Cells without Damaging their Membranes Md. Moniruzzaman, Md. Zahidul Islam, Sabrina Sharmin, Hideo Dohra, and Masahito Yamazaki


Entry of a Six-Residue Antimicrobial Peptide Derived from

Recommend Documents