Escherichia coli and Pseudomonas aeruginosa eradication by nano-penicillin G

    Escherichia coli and Pseudomonas aeruginosa Eradication by Nano-Penicillin G Margarida M. Fernandes, Kristina Ivanova, Antonio Francesko, Diana Rivera, Juan Torrent-Burgu´es, Aharon Gedanken, Ernest Mendonza, Tzanko Tzanov PhD PII: DOI: Reference:

S1549-9634(16)30065-X doi: 10.1016/j.nano.2016.05.018 NANO 1354

To appear in:

Nanomedicine: Nanotechnology, Biology, and Medicine

Received date: Revised date: Accepted date:

13 October 2015 28 April 2016 26 May 2016

Please cite this article as: Fernandes Margarida M., Ivanova Kristina, Francesko Antonio, Rivera Diana, Torrent-Burgu´es Juan, Gedanken Aharon, Mendonza Ernest, Tzanov Tzanko, Escherichia coli and Pseudomonas aeruginosa Eradication by NanoPenicillin G, Nanomedicine: Nanotechnology, Biology, and Medicine (2016), doi: 10.1016/j.nano.2016.05.018

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Escherichia coli and Pseudomonas aeruginosa Eradication by Nano-Penicillin G

T

Margarida M Fernandesa, Kristina Ivanovaa, Antonio Franceskoa, Diana Riveraa, Juan

Grup de Biotecnologia Molecular i Industrial, Department d’Enginyeria Química,

SC

a

RI P

Torrent-Burgués a, Aharon Gedankenb, Ernest Mendonzac, Tzanko Tzanova*

Universitat Politècnica de Catalunya, Rambla Sant Nebridi 22, 08222, España Department of Chemistry, Institute of Nanotechnology and Advanced Materials, BarIlan

MA NU

b

University, Israel

PT

ED

c

*Correspondence: Dr. Tzanko Tzanov, tel.: +34 93 739 85 70, fax: +34 93 739 82 25, e-mail:

AC

CE

[email protected]

Word count for abstract: 149 Complete manuscript word count: 4382 Number of figures: 8 Number of tables: 2 Number of references: 39

The research leading to these results has received funding from the European Community’s Seventh Framework Program FP7 under the Marie Curie Intra-European Fellowship (IEF) project NanoQuench (Grant agreement No. 331416).

1

ACCEPTED MANUSCRIPT Abstract The transformation of penicillin G into nano/micro-sized spheres (nanopenicillin) using

T

sonochemical technology was explored as a novel tool for the eradication of Gram-negative

SC

RI P

bacteria and their biofilms. Known by its

induced 1.2 log-reduction of

MA NU

The efficient penetration of the spheres within a Langmuir monolayer

reach the periplasmic space in Gram-negative bacteria where they inhibit the β-lactam targets: the transferases that build the bacteria cell wall. Moreover, it

ED

considerably suppressed the growth of both bacterial biofilms on a medically relevant

PT

polystyrene surface, leaving majority of the adhered cells dead compared to the treatment with the non-processed penicillin G. Importantly, nanopenicillin was found innocuous

AC

CE

towards human fibroblasts at the antibacterial-effective concentrations.

Key words: penicillin G; sonochemistry; nano/microspheres; nano-bio interactions; antibacterial; antibiofilm

2

ACCEPTED MANUSCRIPT Introduction Until the discovery of penicillin and its introduction in medical practice back in the

T

middle of the last century, bacterial infections were a major cause of death worldwide. This

RI P

“wonder drug” marked the beginning of the antibiotic era and had a profound impact on the quality of human life, providing relief from pain and suffering, and decreased mortality due

SC

to microbial infections.1 Penicillins belong to the broad class of -lactam antibiotics. The four-membered -lactam ring binds and inhibits the transpeptidase enzymes involved in the

MA NU

synthesis of the peptidoglycan layer, thereby hindering the synthesis of the cell wall. An imbalance between cell wall production and degradation is thus created, leading to cell death.2

ED

Transpeptidases are located in the periplasmic space and thus directly accessible by antibiotics in Gram-positive bacteria, which explains the good killing effect of β-lactams

PT

towards these microorganisms.3 In Gram-negative bacteria, however, β-lactams have to cross the outer membrane of the bacteria either by passive diffusion or via porin channels, which

CE

sometimes is hindered by the resistance mechanism bacteria develop in which a permeability

AC

barrier to the antibiotic by porins modification is created, explaining the poor effect of penicillin against these pathogens.1,3 To improve its efficacy and provide an extended spectrum of activity the chemical modification of pristine penicillin has been commonly applied. For example, the introduction of an amino group on the structure of penicillin to generate aminopenicillins (e.g. ampicillin), enables the penetration of the antibiotic in the outer membrane of Gram-negative bacteria, where it finally acts as a transpeptidase inhibitor, exerting its antibacterial activity.4,5 Despite the advances in the development of newer synthetic analogues, the chemical modification of β-lactams does not change their interaction with the target, and although it improves the antibiotic efficacy, the underlying resistance mechanisms are still present in the environment.6 In fact, most clinically relevant antibacterial

3

ACCEPTED MANUSCRIPT agents inhibit a very short list of cellular targets, which make bacteria prone to resistance development.7,8

T

Apart from the mechanisms used by planktonic bacteria to resist the action of

RI P

antibiotics, these single-cell individuals are also able to gather into sedentary but dynamic communities - biofilms - grown within a complex structure and acting as a barrier to the

SC

penetration of antibiotics.9 This mode of growth is considered an effective resistance mechanism that bacteria use to survive in harsh environments. Bacteria encased in a biofilm

MA NU

matrix are over 1000 times more resistance to antibiotics and host immune system than in culture and highly contributes to the chronicity of infections such as those associated with implanted medical devices.10,11 The biofilm polysaccharide matrix that strongly adheres to

ED

surfaces serves as a perfect location for bacteria growing and establishing infections to the host. These difficult-to-treat biofilm-associated infections account in a large scale for the

PT

morbidity and mortality in hospitals.12 Naturally, there is a need to fight efficiently both antibiotic resistance and biofilm-associated infections, and one approach with great

CE

perspective is the nanotechnology.

AC

The use of nanotechnology as a therapeutic tool to enhance clinical effectiveness of drugs and fighting microbial resistance is not new,13,14 but mostly comprises the strategy to use drug carriers, such as liposomes, nano/microcapsules and nano/microparticles, to deliver the antibacterial agents at the infection site.7 The transformation of the active agents themselves into nanosize entities has been only sparsely reported.15 This can be done using an ultrasonic emulsification method that allows obtaining oil-filled nano/microspheres (NMSs), in which the antibiotic is located at the interface of the droplet, i.e. not encapsulated but the shell itself.16 Such nano-transformation holds significant promise because it drastically changes the physical properties of molecules allowing for novel biological modes of action.14 The assemblies of the molecules at nanometer dimensions have unique properties, differing

4

ACCEPTED MANUSCRIPT from those of the free molecules and the bulk materials with the same composition. In fact, previous work performed in our group has shown that biopolymers processed sonochemically

T

into nano-sized spheres were more efficient antibacterial agents than their non-processed

RI P

counterparts due to the effective disintegration of the bacterial cell wall through a detergent model mechanism.17

SC

This work aims at developing sonochemically-engineered penicillin G nanospheres with extended efficiency towards clinically relevant Gram-negative bacteria and biofilms.

MA NU

We opted for altering the form and enhancing the interaction with the bacterial cell wall of an existing and already FDA approved antibiotic as a facile method to create a broad-spectrum antibiotic, rather than undertaking a time- and resource-consuming screening for new drug

ED

entities. Assuming a timeline of 3-5 years to deliver a new antimicrobial candidate and roughly 6 years to complete the clinical testing, it is expected that the introduction of an

PT

antibiotic with a novel activity mechanism against Gram-negative infections could take 10-15 years. Therefore, the strategy we propose would shorten the drug development process,

CE

because it does not alter the drug target, but simply adds a supporting mechanism to the main

AC

activity of penicillin G. Moreover, mostly Gram-positive-spectrum agents are currently in clinical trials,6 while the hereby produced nano-antibiotic will extend its activity also to Gram-negative species, relying on a mechanism of action that comprises additionally bacterial membrane disturbance and easy penetration into biofilms.

Methods

Sonochemical preparation of penicillin nanospheres Penicillin (Sigma-Aldrich (Spain)) and control (NMSs) were prepared by an adaptation of Suslick method.18 Briefly, a two-phase solution containing 70 % of 0.250 g/L antibiotic

5

ACCEPTED MANUSCRIPT aqueous solution containing 0.1 % (w/v) of surfactant (Poloxamer 407, Sigma-Aldrich (Spain)) (for control NMSs the antibiotic is omitted) and 30 % of commercial sun-flower oil

T

(organic phase) was prepared and placed into a thermostated (8 ± 1 ºC) sonicator cell using a

RI P

high-intensity Vibra-Cell VCX 750 ultrasonic processor (Sonics & Materials, Inc., USA) with a 20 kHz Ti horn at 35 % amplitude. The bottom of the probe was positioned at the

SC

aqueous-organic interface for 3 min using an ice-cooling bath to maintain the low temperature. The nanospheres were further characterized as reported in the Supplementary

MA NU

Material section.

Interaction with cell membrane models

ED

The interaction of penicillin and nanopenicillin with cell membrane models was assessed using a Langmuir monolayer technique

19,20

by measuring the effect of the antimicrobial

PT

agents in contact with L-α-Phosphatidylethanolamine phospholipid extracted from

AC

Material section.

CE

Escherichia coli (Sigma-Aldrich (Spain)). More details are reported in the Supplementary

Antibacterial and antibiofilm activity The Gram-negative Escherichia coli (E. coli, ATCC 25922), Pseudomonas aeruginosa (P. aeruginosa, ATCC 10145), Staphylococcus aureus (S. aureus, ATCC 25923) purchased from American Type Culture Collection (LGC Standards S.L.U, Spain), were used in all antimicrobial assays. For the preparation of bacterial inoculum, a single colony from the corresponding stock bacterial cultures was used and allowed to grow overnight at 37 ºC and 230 rpm. Nutrient Broth (NB) was employed to test the bacteria susceptibility and antibacterial activity, while Triptic Soy Broth (TSB) was used to grow bacteria for biofilm inhibition tests. Details on the methodology used for determining the P. aeruginosa, E. coli

6

ACCEPTED MANUSCRIPT and S. aureus susceptibility to Penicillin G, the antibacterial activity of the NMSs to the Gram-negative bacteria and the bacteria viability in the biofilm structure 21 are reported in the

RI P

T

Supplementary Material section.

Cytotoxicity evaluation

SC

Cell culture. To determine the potential toxicity of the NMSs towards mammalian cells, human foreskin fibroblast cells line BJ-5ta (ATCC CRL-4001) were used.

Different

MA NU

concentrations of nano/microspheres were placed in contact with cultured cells and after 24 h the number of viable cells were quantified using Alamar Blue assay, as reported in the

ED

Supplementary Material section.

Results

PT

Penicillin G is a beta-lactam antibiotic mainly used in the treatment of bacterial infections caused by Gram-positive organisms by targeting the transpeptidase enzyme that

CE

catalyzes the formation of peptidoglycan cross-links as the cell grows and divides. Due to the

AC

fact that its main target is the peptidoglycan cell wall, it is implicit why the effect of βlactams is less prominent or inexistent when it comes to Gram-negative bacteria, in which the peptidoglycan is located in the periplasmic space, “protected” between the outer and inner membrane layer. In fact, gram-negative bacterial strains such as E. coli and P. aeruginosa used in this work were found to be resistant to the action of Penicillin G while the Grampositive S. aureus were sensitive at concentrations above 15 g/mL (Abs600 ≈ 0) (Fig. 1). In order to extend the activity spectrum of penicillin G also to Gram-negative bacteria, a sonochemical nanospherization of the free antibiotic was carried out. This method has already been applied to boost the antibacterial activity of biopolymers such as aminocellulose and thiolated chitosan.17

7

ACCEPTED MANUSCRIPT The generated nanospheres were spherical in shape (Fig. 2B) with a mean size of 192  6 nm, a polidispersity of 0.12  0.04 (Fig. 2A) and a zeta potential of - 12.8  1.2 mV. This

T

transformation induced some changes on the chemical structure of the antibiotic, seen by

RI P

infrared spectroscopy. The most noticeable ones were the disappearance of the β-lactam carbonyl group band at 1790 cm-1 (responsible for the binding and inhibition of

SC

transpeptidase enzymes involved in the synthesis of the peptidoglycan layer) and the increase of the broad band at 3500 cm-1, which correspond the to hydroxyl (-OH) from the carboxylic

MA NU

group (-CO2H) due to the hydrogen bonding effect (Fig. 3). These changes do not to compromise the mechanism of action of Penicillin G since transpeptidase compete with the existing hydrogen bonding, inducing a reaction that is chemically more favourable. In this

ED

case, the NMSs would be destabilised, however, only after eliciting the effect. Therefore, we may conclude that the sonochemical-induced formation of nanopenicillin was accompanied

PT

by the creation of a hydrogen bond network between the hydroxyl group and the β-lactam carbonyl that contributed for the stabilization of the NMSs. This effect has already been

CE

observed when tetracycline was formulated into nanoparticles using the same sonochemical

AC

method.15 Moreover, the peaks assigned to penicillin found at 1600 cm-1 and corresponding to amide I band were not significantly changed, suggesting that no other major changes were induced in the structure of Penicillin G upon spherization. The remaining peaks were assigned to the surfactant and oil as it could be seen in their IR spectra. On the other hand, the IR spectra of control NMSs presented a combination of the surfactant, oil and hydrogen bonding peaks. The hydrogen bonding is present either in the spectra of nanopenicillin and control NMSs, suggesting their important role in the formation of the spheres (Fig. 3). Penicillin and nanopenicillin were further evaluated for their antibacterial and antibiofilm activity against the Gram-negative P. aeruginosa and E. coli. Control NMSs were also produced, i.e. spheres without penicillin, comprised solely of surfactant and oil, in order

8

ACCEPTED MANUSCRIPT to determine both the effect of the surfactant and the effect of the particle spherical shape on the interaction with the bacteria and the biofilms.

T

Penicillin in solution, tested at the same concentration and under the same assay

RI P

conditions, was less effective in killing both Gram-negative bacteria than nanopenicillin. The last induced a 3 log reduction of P. aeruginosa and 1.2 log reduction of E. coli while the

SC

penicillin in solution induced less than 0.3 log reduction of both microorganisms (Fig. 4). An

negative bacteria was attained.

MA NU

improvement of 10- and 4-fold, respectively, in the inhibition of these planktonic Gram-

The effect of the nanopenicillin on biofilms was also evaluated. A decrease of the total biomass was observed under fluorescence microscopy after Live/Dead staining when

ED

either penicillin and nanopenicillin was applied at a concentration of 125 μg/mL on previously formed and well-established biofilms (Control in Fig. 5). Nevertheless, the effect

PT

was more pronounced when nanopenicillin was applied (Fig. 5) and in the case of P. aeruginosa, red cells could also be observed which is indicative of a killing effect. The

CE

control NMSs, on the other hand, did not produce a decrease on the biomass neither a killing

AC

effect, confirming that NMSs lacking the antibiotic were not able to eradicate bacteria and prevent the biofilm formation. It has been reported that extracellular DNA (eDNA), an important component of the biofilm matrix of P. aeruginosa, is also stained by the propidium iodide39, one of the dyes that incorporates the live/dead kit. Therefore, it was important to quantify the viable microorganisms within biofilms using the Calgary Biofilm Device (CBD- MBEC™ assay), in order to sustain the theory that nanopenicillin is indeed able to induce the P. aeruginosa killing effect, as visualised through fluorescence microscopy. This technique allows microorganisms to grow on 96 identical pegs protruding down from a plastic lid to form a biofilm. Putting the grown biofilm in contact with the antimicrobial compounds and further

9

ACCEPTED MANUSCRIPT plating them in specific agar, is possible to quantify the viable bacteria within the biofilms. Nanopenicillin was also able to kill more viable bacteria within P. aeruginosa and E. coli

T

biofilms than Penicillin G. P. aeruginosa was more susceptible to the action of nanopenicillin

RI P

being reduced in 2.5 log while less effect was observed in E. coli biofilm bacteria, in which a 1.65 log reduction was detected. Penicillin in solution was not effective against P. aeruginosa

SC

but it was able to induce a 0.8 log reduction of E. coli in biofilms, corroborating the results found by fluorescence microscopy and antibacterial assays (Table 1). The red dots observed

MA NU

on fluorescence microscopy images could thus be assigned to a combination of both dead cells and eDNA.

To understand the mechanism of action of the novel NMSs, a Langmuir monolayer

ED

technique, used for mimicking bacterial membranes, was applied. This technique is commonly used as a supportive model to explain the interactions between different molecules

PT

and nanoparticulate systems with bacterial membranes even without considering the other constituents of the bacterial membrane. In a first assessment, both penicillin in solution and

CE

nanopenicillin were found to be surface active when analyzed by Langmuir technique in the

AC

absence of phospholipid upon compression (Fig. 6A). The movement of the spheres to the air-water interface indicated that they might be capable to interact with a membrane model and possibly with a bacterial membrane (Fig. 6A right panel). To prove this concept, the interaction and internalization of nanopenicillin and penicillin in solution with a membrane model comprised of L-α-phosphatidylethanolamine (PE), the principal phospholipid extracted from E. coli22 was studied. This technique has been commonly applied in our group to elucidate the interactions of antimicrobial agents with bacterial cell membranes at a molecular level, thus indirectly assessing their antimicrobial potential.17,23 The isotherms of E. coli PE with only water in the subphase presented a lift-off at molecular areas of around 300 Å2/molecule, while the others lifted-off at significantly

10

ACCEPTED MANUSCRIPT higher molecular areas indicating a membrane disturbing effect (Fig. 6B). In fact, the same concentration of penicillin and nanopenicillin were able to efficiently interact with the PE

T

membrane of E. coli, although at different extent, inducing a significant expansion of the

RI P

monolayer at large areas per molecule of phospholipid (Fig. 6B). This is mainly due to the penetration of the spheres and penicillin molecules within the monolayer when the molecules

SC

of phospholipids are not highly compressed as depicted in the right panel of Fig. 6B (upper image). Regarding the control NMSs, despite the fact that the effect was not as pronounced as

MA NU

the one observed for penicillin and nanopenicillin, a disturbing effect on the membrane model was also detected, indicating that the spherical shape and the surfactant indeed play an important role in the interaction with the membrane model.

ED

After compression, at lower phospholipid molecular areas, the isotherm of the control NMSs merged with the isotherm of the pure monolayer (water in the subphase), which means

PT

that these spheres were expelled from the monolayer after full Langmuir compression,24 indicating that they produce an insignificant effect on the established compressed monolayer.

CE

Nevertheless, this did not occur with penicillin and specially with nanopenicillin, as the

AC

isotherms collapsed before reaching the monolayer highly compressed stage, demonstrating that even at smaller areas per phospholipid the nanopenicillin is inserted in between the phospholipid molecules, as depicted in the right panel of Fig. 6B. This effect is called penetration and occurs when the compounds present in the subphase emerge to the air-water interface and penetrates in between the phospholipid molecules.24 Moreover, by determining the maximum compression modulus of the isotherms it could be observed that they all experienced a unique phase transition, from gaseous to liquid-expanded state, represented by the Cs-1, max around 50 mN/m (Table 2). The inability of the monolayer to reach a condensed state, commonly represented by higher values of compression modulus (Cs−1), is due to the fact that the E. coli PE phospholipid is a non-pure compound that

11

ACCEPTED MANUSCRIPT The presence of nanopenicillin, in particular, induced a significant i

RI P

T

Cs−1,max

By analyzing the behavior of the isotherms on Fig. 6B at a fixed surface pressure of

SC

30 mN/m, which corresponds to the lateral pressure of naturally occurring cell membrane, 25 we could also infer about how these spheres interact with the bacterial membrane. We

MA NU

observed that at this surface pressure, nanopenicillin highly increased the expansion (membrane fluidity) of the monolayer as one molecule of phospholipid occupied ≈ 190 Å (see straight line on Fig. 6B). On the other hand, the phospholipid molecules only occupied ≈ 120 Å when either penicillin in solution or the control NMSs were present in the subphase.

ED

These results demonstrated that by engineering penicillin into nanospheres we are indeed able

PT

to improve the interaction of penicillin with a bacterial membrane. To further elucidate this behavior, all tested compounds were also injected into the

CE

subphase beneath the E. coli PE monolayer, previously formed at 30 mN/m, and the variation

AC

in the surface tension of the monolayer was monitored as a function of time (Fig. 6C). Nanopenicillin efficiently penetrated the membrane model, indicated by the higher increase of the surface tension in comparison to the penicillin molecules and control NMSs inserted in the subphase. Nevertheless, the control NMSs also induced a significant increase on the surface tension indicating that the spherical shape and nanosize play an important role in the interaction and penetration of nanopenicillin into the bacterial membrane (Fig. 6C). Since nanoparticles and nanomaterials become highly active at nanometer scale, illustrated here by the enhanced bactericidal capacity of nanopenicillin, nanotoxicology studies should be performed to determine whether these properties might pose a threat to humans. We used skin fibroblasts as a model for human cells and analyzed the cytotoxicity of

12

ACCEPTED MANUSCRIPT nanopenicillin and penicillin at the antimicrobial-effective concentrations. Nanopenicillin was cytotoxic at concentration of 250 μg/mL, but at the antibacterial-effective concentration

RI P

T

of 125 μg/mL the antibiotic NMSs were fully biocompatible (Fig. 7).

Discussion

SC

The use of nanotechnology to overcome bacteria resistance to β-lactams appears to hold significant promise as different biological modes of action towards bacteria are created,

MA NU

thus making the bacteria more susceptible. In fact, the nano-transformation has been considered as a novel tool to boost the antibacterial activity of materials. One common example is silver, which once transformed into nanoparticles acquires a high antimicrobial activity due to the large surface area of the nanoparticles and better contact with the

ED

microorganisms.26

PT

In this work, we sonochemically transformed the penicillin G into NMSs as a novel tool for overcoming resistance in Gram-negative bacteria, aiming the improved interactions

CE

and penetration of the NMSs with the bacterial membrane. The generation of penicillin

AC

nanoparticles using ultrasound has also been reported by Yariv et al. (although in absence of an organic phase or surfactant), as a way of enhancing the antimicrobial activity against a S. aureus clinical isolate.27 However, the reported penicillin nanoparticles did not kill Gramnegative bacteria. Herein, we obtained nano/micro-sized, oil-filled, surfactant-containing spheres able to efficiently interact with the membrane of Gram-negative bacteria. The presence of the surfactant together with penicillin in the spheres played a key role on the interaction with the bacterial membrane, as suggested by the insertion capacity of the control NMSs (solely comprised of surfactant) on the Langmuir monolayer. Indeed, surfactants such as poloxamer have been reported to be biomembrane active due to their relative ratio of hydrophobic versus hydrophilic building blocks.28

13

ACCEPTED MANUSCRIPT Apart from assisting the efficient interaction with the membrane, poloxamer was also used for stabilization purposes. The generated NMSs were stable in aqueous solution for

T

more than 6 months, i.e. did not aggregate in solution, despite their relatively low zeta-

RI P

potential (< - 20mV). This hydrophilic surfactant, comprised of polyethylene oxide (PEO) and polypropylene oxide (PPO), is able to induce steric stabilization of the spheres. The

SC

hydrophobic PPO chains adsorb on the particle surfaces as the “anchor chain”, while the hydrophilic PEO chains pull out from the surface to the aqueous medium, producing a

MA NU

stabilizer layer.29 This effect shifts the plane of shear to a farer distance from the particle surface thus reducing the measured zeta potential.30 Poloxamers have been frequently applied for stabilization of nanoparticles, regardless of the method used for their synthesis. Pezeshiki et al. have synthesized nanostructured lipid carriers with similar size as the one obtained in

ED

this study using 6 % Poloxamer29 while we used only 0.1 % (w/v). Vineeth et al. also

PT

suggested that not only the surfactant influences the size of particle, but also the choice of the organic solvent.31 However, most of the organic solvents used in the cited studies are

CE

considered cytotoxic, while our NMSs encapsulate edible sunflower oil as an organic phase

AC

in order to avoid harmful effects in humans. Besides Poloxamers, other anionic, cationic and/or amphoteric surfactants have also been reported to efficiently stabilize nanoparticulate systems. The choice of the best surfactant for each formulation depends on the final application and several factors such as the type of stabilization required (electrostatic versus steric stabilization) and the route of administration (for example, steric stabilization is recommended because it is less susceptible to electrolytes in the gut or blood). Several studies have been carried out to determine the effect of different surfactants and organic solvents in nanoemulsions.32–35 Mainly cationic nanoparticles have been reported to efficiently kill bacteria by disrupting the cell membrane due to electrostatic interaction between the positively charged

14

ACCEPTED MANUSCRIPT particles and the negatively charge bacterial membrane, leading to cell leakage and death - an effect that we have already observed for other cationic nanospheres.17 The herein generated

T

NMSs possessed negatively charged properties and it was therefore expected that they would

RI P

not affect significantly the bacterial membrane. Nevertheless, these particles were surface active and able to penetrate the membrane model, providing the boosting effect compared to

SC

Penicillin G in solution for killing gram-negative bacteria. In fact, negatively charged particles have been reported to efficiently penetrate the cells, without disrupting the

MA NU

membrane, through nonspecific binding and clustering of the particles on the cationic sites of the membrane followed by internalization via endocytosis.36 The results obtained from Langmuir monolayer studies suggested that by engineering

ED

penicillin G into NMSs the interaction and penetration of otherwise less penetrable antibiotic molecules into the bacterial membrane is favored. Since the resistance of Gram-negative

PT

bacteria to β-lactams is mainly due to the incapability of the penicillin molecules to penetrate the outer membrane and inhibit the transpeptidases in the periplasmic space, it is expected

CE

that nanopenicillin, in contrast, would be able to insert into the membrane and overcome this

AC

antibacterial-resistance mechanism (Fig. 8A). Combining nanoparticles with antibiotics has already been proved to generate a synergy against bacterial resistance by reducing the toxicity of both agents towards human cells due to the decrease on the requirement for high dosages, but also to enhance their bactericidal properties.37 Herein, the same enhanced killing effect was observed, however, in our case the antibiotic itself was formulated into the nanoform, instead of being encapsulated or combined with other particles, thereby combining both bactericidal and nanoscale effects. The spherical shape and large surface area of the antibiotic NMSs was further designated as the main reason for the successful eradication of bacterial biofilm by penetrating them more easily than the Penicillin G in solution and by providing better contact

15

ACCEPTED MANUSCRIPT with microorganisms encased, thus allowing the eradication of Gram-negative biofilms and even induce a killing effect, mainly on P. aeruginosa (Fig. 8B). Treating biofilm infections

T

remains a major challenge due to the fact that high doses of antimicrobial agents are required.

RI P

It has also been reported that by packing the antibiotic into nanoparticles increased killing of biofilm-encased bacteria could be achieved compared to the free antibiotic due to the

SC

protection effect that the nanoparticles provide against enzymatic degradation of the

MA NU

antibiotic and its binding to matrix components.38

particularly important

due to the fact that this highly prevalent opportunistic pathogen is one of the most worrisome bacteria in clinical settings owing to its low antibiotic susceptibility, especially in the form of

AC

CE

PT

ED

biofilms.

16

ACCEPTED MANUSCRIPT References 1.

Miller EL. The penicillins: a review and update. J Midwifery Wom Heal 2002; 47:

T

426–434. Chain B. Penicillin and beyond. Nature. 1991; 353: 492–494.

3.

Silhavy TJ, Kahne D, Walker S. The bacterial cell envelope. Cold Spring Harb.

RI P

2.

SC

Perspect. Biol. 2010; 2: a000414.

Cunha BA. Aminopenicillins in urology. Urology. 1992; 40: 186–190.

5.

Barber M, Waterworth PM. Antibacterial Activity of the Penicillins. Br. Med. J. 1962;

MA NU

4.

1: 1159–1164. 6.

Payne DJ, Gwynn MN, Holmes DJ, Pompliano DL. Drugs for bad bugs: confronting the challenges of antibacterial discovery. Nat. Rev. Drug Discov. 2007; 6: 29–40. Abeylath SC, Turos E. Drug delivery approaches to overcome bacterial resistance to β-

ED

7.

8.

PT

lactam antibiotics. Expert Opin. Drug Deliv. 2008; 5: 931–949. Zapun A, Contreras-Martel C, Vernet T. Penicillin-binding proteins and beta-lactam

Singh R, Ray P, Das A, Sharma M. Penetration of antibiotics through Staphylococcus

AC

9.

CE

resistance. FEMS Microbiol. Rev. 2008; 32: 361–385.

aureus and Staphylococcus epidermidis biofilms. J. Antimicrob. Chemother. 2010; 65: 1955–1958 10.

Høiby N, Bjarnsholt T, Givskov M, Molin S, Ciofu O. Antibiotic resistance of bacterial biofilms. Int. J. Antimicrob. Agents. 2015; 35: 322–332.

11.

Lewis K. Riddle of biofilm resistance. Antimicrob. Agents Chemother. 2001; 45: 999– 1007

12.

Del Pozo JL, Patel R. The challenge of treating biofilm-associated bacterial infections. Clin. Pharmacol. Ther. 2007; 82: 204–209.

17

ACCEPTED MANUSCRIPT 13.

Pelgrift RY, Friedman AJ. Nanotechnology as a therapeutic tool to combat microbial resistance. Adv. Drug Deliv. Rev. 2013; 65: 1803–1815. Huh AJ, Kwon YJ. “Nanoantibiotics”: A new paradigm for treating infectious diseases

T

14.

RI P

using nanomaterials in the antibiotics resistant era. J Control Release. 2011; 156:128– 145.

Shimanovich U, Lipovsky A, Eliaz D, Zigdon S, Knowles TP, Nitzan Y, Michaeli S,

SC

15.

Gedanken A. Tetracycline Nanoparticles as Antibacterial and Gene-Silencing Agents.

16.

MA NU

Adv. Healthc. Mater. 2015; 4: 723–728.

Gedanken A. Preparation and Properties of Proteinaceous Microspheres Made Sonochemically. Chem-Eur J. 2008; 14: 3840–3853. Fernandes MM, Francesko A, Torrent-Burgués J, Carrión-Fité FJ, Heinze T, Tzanov

ED

17.

T. Sonochemically processed cationic nanocapsules: Efficient antimicrobials with

18.

PT

membrane disturbing capacity. Biomacromolecules. 2014; 15, 1365–1374. Suslick KS, Grinstaff MW. Protein microencapsulation of nonaqueous liquids. J Am

Hoyo

J,

Torrent-Burgués

J,

Guaus

E.

Biomimetic

monolayer

films

of

AC

19

CE

Chem Soc. 1990; 112: 7807–7809.

monogalactosyldiacylglycerol incorporating ubiquinone. J Colloid Interf Sci. 2012; 384: 189–197. 20.

Torrent-Burgués J, Vocanson F, Pérez-González JJ, Errachid A. Synthesis, Langmuir and Langmuir–Blodgett films of a calix[7]arene ethyl ester. Colloid Surf. A. 2012; 401: 137–147.

21.

Ivanova K, Fernandes M, Mendoza E, Tzanov T. Enzyme multilayer coatings inhibit Pseudomonas aeruginosa biofilm formation on urinary catheters. Appl. Microbiol. Biotechnol. 2015; 99: 4373–4385.

18

ACCEPTED MANUSCRIPT 22.

Golovastov

V

V,

Mikhaleva

NI,

Nesmeyanova

MA.

Depletion

of

phosphatidylethanolamine- the major membrane phospholipid of Escherichia coli-

T

depresses posttranslocational modification of alkaline phosphatase in the periplasm.

23.

RI P

Biochemistry (Mosc). 2002; 67(7); 765–769.

Fernandes MM, Francesko A, Torrent-Burgués J, Tzanov T. Effect of thiol-

SC

functionalisation on chitosan antibacterial activity: Interaction with a bacterial membrane model. React. Funct. Polym. 2013; 73(10): 1384–1390. Dynarowicz-Ła̧tka P, Dhanabalan A, Oliveira ON. Modern physicochemical research

MA NU

24.

on Langmuir monolayers. Adv. Colloid Interface Sci. 2001; 91(2): 221–293. 25.

Dennison S, Kim Y, Cha H, Phoenix D. Investigations into the ability of the peptide,

26.

ED

HAL18, to interact with bacterial membranes. Eur. Biophys. J. 2008; 38(1): 7–43. Ravishankar Rai V, Jamuna Bai A. Nanoparticles and their potential application as

PT

antimicrobials: In Méndez-Vilas A, editor. Science against microbial pathogens: communicating current research and technological advances. Volume 1, Formatex

Yariv I, Lipovsky A, Gedanken A, Lubart R, Fixler D. Enhanced pharmacological

AC

27.

CE

Research Center; 2011, p. 197-209.

activity of Vitamin B12 and Penicillin as nanoparticles. Int. J. Nanomedicine. 2015; 10: 3593-3601. 28.

Cheng CY, Wang J, Kausik R, Lee KYC, Han S. Nature of interactions between PEOPPO-PEO triblock copolymers and lipid membranes: (II) role of hydration dynamics revealed by dynamic nuclear polarization. Biomacromolecules. 2012; 13:2624–33.

29.

Pezeshki A, Ghanbarzadeh B, Mohammadi M, Fathollahi I, Hamishehkar H. Encapsulation of Vitamin A Palmitate in Nanostructured Lipid Carrier (NLC)-Effect of Surfactant Concentration on the Formulation Properties. Adv. Pharm. Bull. 2014; 4(Suppl 2): 563-568.

19

ACCEPTED MANUSCRIPT Quaglia F, Ostacolo L, Mazzaglia A, Villari V, Zaccaria D, Sciortino MT. The intracellular effects of non- ionic amphiphilic cyclodextrin nanoparticles in the

T

delivery of anticancer drugs. Biomaterials 2009; 30: 374-382.

RI P

Vineeth P, Vadaparthi P R R, Kumar K, Babu D, Rao A V, Babu K S. Influence of organic solvents on nanoparticle formation and surfactants on release behaviour in-

SC

vitro using costunolide as mode. Int. J. Pharm. Pharm. Sci. 2014; 6: 638–645. Feczkó T, Tóth J, Gyenis J. Comparison of the preparation of PLGA–BSA nano- and

Asp. 2008; 319: 188–195.

MA NU

microparticles by PVA, poloxamer and PVP. Colloids Surfaces A Physicochem. Eng.

Zhou F, Zhou R, Hao X, Wu X, Rao W, Chen Y, Gao D. Influences of surfactant

ED

(PVA) concentration and pH on the preparation of copper nanoparticles by electron beam irradiation. Radiat. Phys. Chem. 2008; 77: 169–173.

PT

Mu L, Seow P H. Application of TPGS in polymeric nanoparticulate drug delivery system. Colloids Surfaces B Biointerfaces 2006; 47: 90–97.

CE

Cohen-Sela E, Chorny M, Koroukhov N, Danenberg H D, Golomb G, New Double A. Emulsion Solvent Diffusion Technique for Encapsulating Hydrophilic Molecules in

AC

30.

PLGA

Verma A, Stellacci F. Effect of surface properties on nanoparticle-cell interactions. Small. 2010; 6: 12–21.

20

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

39.

21

ACCEPTED MANUSCRIPT

T

Figure 1. Optical density at 600 nm of P. aeruginosa, E. coli and S. aureus after 24 h

RI P

incubation with Penicillin G.

Figure 2. A) Dynamic light scattering, and B) scanning electron microscopy images of

SC

nanopenicillin.

MA NU

Figure 3. FTIR spectra of lyophilized penicillin, nanopenicillin, surfactant (poloxamer) and oil.

Figure 4. Antimicrobial activity of 125 μg/mL penicillin and nanopenicillin on E. coli and P. aeruginosa after 24 h contact. **P  0.01 when compared with each other.

ED

Figure 5. Fluorescence microscopy live/dead images of 24 h grown P. aeruginosa and E. coli biofilms incubated with 125 μg/mL penicillin, nanopenicillin and control NSs (live cells are

PT

represented in green and dead cells in red). Scale bars denote 100 micron.

CE

Figure 6. A) Surface pressure-area isotherms (or surface activity) and schematic representation of the nanospheres surface activity upon Langmuir compression; B) Surface

AC

pressure-area isotherm of E. coli PE monolayer (the area per molecule in X-axe refers to E. coli PE) in water with 0.5 g/mL penicillin in solution, nanopenicillin and control NSs and schematic representation of the NSs interaction with the a Langmuir monolayer upon compression; and C) kinetic adsorption process resulting from the incorporation of 0.2 g/mL (final concentration) penicillin, nanopenicillin and control NSs onto the air/water interface of E. coli PE monolayer at a surface pressure of 30 mN/m and schematic representation of the incorporation of the spheres in the subphase beneath the phospholipid monolayer built at 30 mN/m.

22

ACCEPTED MANUSCRIPT Figure 7. Relative viabilities (%) of human fibroblasts exposed to 250 μg/mL and 125 μg/mL

T

penicillin and nanopenicillin for 24 h. *P

RI P

Figure 8. Schematic representation of the interaction of nanopenicillin and penicillin G onto:

AC

CE

PT

ED

MA NU

SC

A) the gram-negative bacterial cell wall and B) the biofilm structure.

23

Fig. 1

AC

CE

PT

ED

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

24

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 2

25

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 3

26

Fig. 4

AC

CE

PT

ED

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

27

Fig. 5

AC

CE

PT

ED

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

28

Fig. 6

AC

CE

PT

ED

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

29

Fig. 7

AC

CE

PT

ED

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

30

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Fig. 8

31

T

ACCEPTED MANUSCRIPT

RI P

Table 1. Viable cells within P. aeruginosa and E. coli biofilms before (control) and after applying 125 μg/mL of penicillin, nanopenicillin and control NSs.

Escherichia coli

SC

Log10 of CFU/mL

Pseudomonas

6.60 ± 1.2

Control NSs

6.70 ± 0.9

Penicillin

5.78 ± 1.7

Nanopenicillin

4.95 ± 0.9

7.52 ± 1.5 7.50 ± 1.1 7.51 ± 0.9 5.26 ± 1.6

AC

CE

PT

ED

Control

MA NU

aeruginosa

32

ACCEPTED MANUSCRIPT

T

Table 2. Maximum compression modulus of E. coli PE monolayers with water, penicillin,

RI P

nanopenicillin and control NMSs in the subphase E. coli PE Cs-1, max (mN/m)

Phase transitions

Control (H2O)

55.8

Gaseous-liquid expanded

Penicillin

51.9

Gaseous-liquid expanded

Nanopenicillin

26.7

Control NMSs

48.1

MA NU

SC

Sample

Gaseous-liquid expanded

AC

CE

PT

ED

Gaseous-liquid expanded

33

ACCEPTED MANUSCRIPT Graphical Abstract The transformation of a known antibiotic (penicillin G) into nano-sized spheres for the

T

creation of a novel antimicrobial agent with extended-spectrum activity is reported. Penicillin

RI P

nanospheres (nanopenicillin) were able to better eradicate Gram-negative E. coli and P. aeruginosa in culture and in the form of biofilm while their molecular counterparts were

SC

insufficient to do so. The capability of the spheres to penetrate both bacterial membrane and biofilm structure was found to be the reason for such successful outcome. By penetrating in

MA NU

the outer membrane of Gram-negative bacteria, nanopenicillin is able to reach the periplasmic space where it inhibits the β-lactam targets: the transferases that build the

AC

CE

PT

ED

bacteria cell wall.

34