Designer Liposomic Nanocarriers Are Effective Biofilm Eradicators

Abstract

Drug delivery via nanovehicles is successfully employed in several clinical settings, yet bacterial infections, forming microbial communities in the form of biofilms, present a strong challenge to therapeutic treatment due to resistance to conventional antimicrobial therapies. Liposomes can provide a versatile drug-vector strategy for biofilm treatment, but are limited by the need to balance colloidal stability with biofilm penetration. We have discovered a liposomic functionalization strategy, using membrane-embedded moieties of poly[2-(methacryloyloxy)ethyl phosphorylcholine], pMPC, that overcomes this limitation. Such pMPCylation results in liposomic stability equivalent to current functionalization strategies (mostly PEGylation, the present gold-standard), but with strikingly improved cellular uptake and cargo conveyance. Fluorimetry, cryo-electron, and fluorescence microscopies reveal a far-enhanced antibiotic delivery to model Pseudomonas aeruginosa biofilms by pMPC-liposomes, followed by faster cytosolic cargo release, resulting in significantly greater biofilm eradication than either PEGylation or free drug. Moreover, this combination of techniques uncovers the molecular mechanism underlying the enhanced interaction with bacteria, indicating it arises from bridging by divalent ions of the zwitterionic groups on the pMPC moieties to the negatively charged lipopolysaccharide chains emanating from the bacterial membranes. Our results point to pMPCylation as a transformative strategy for liposomal functionalization, leading to next-generation delivery systems for biofilm treatment.

Keywords: antibiotic resistance; biofilm; drug delivery; liposome functionalization; zwitterionic polymer−bacteria membrane interactions.

Figure 1

(A) Schematic illustration of the pMPC-functionalized liposomes, depicting pMPC polymer conjugated to phosphatidylethanolamine (DSPE) lipid present at the liposome surface at 5% (mol/mol). The pMPCylated liposomes are composed of HSPC with 40% (mol/mol) cholesterol and 5% (mol/mol) SA. Liposomes were loaded with antimicrobial agents: either SMX on their own or together with EA. (B) Representative cryo-TEM images of unloaded pMPCylated LUVs after 1 h incubation in (a) 10 mM and (b) 40 mM Ca(Ac)₂ solution and (c) PEG-LUVs in 40 mM solution of Ca(Ac)₂. White arrows indicate adhesion points, black arrow indicates no adhesion. Scale bar, 50 nm. (C) Confocal microscopy images of time-lapse acquisition of pMPC-GUVs stained with DiI dye (red), interacting upon addition of 40 mM Ca(Ac)₂. Scale bar, 20 μm.

Figure 2

pMPC-LUVs interacting with bacteria-mimicking membranes in differing calcium ions and LPS conditions. (A) Graphical illustration of the bacteria-LUVs fusion with pMPC-liposomes measured with calcein dequenching assay. (B) Profile of calcein content mixing assay between bacteria-liposomes (w or w/o LPS) and calcein-free pMPC-liposomes in the presence and absence of calcium ions (final concentration: 0.72 mM). Results are an average of three independent experiments.

Figure 3

Representative super-resolution microscopy images of PA14 cells after 4 h incubation with (A) pMPCylated and (B) PEGylated-liposomes at 37 °C. Inset shows a zoomed-in detail of bacteria cells with liposomes adhering to the membrane. Images are presented as overlays of bacteria membrane stained with FM1–43 dye (green) with liposomes labeled with DiR dye (red) visualized using stochastic optical reconstruction microscopy (STORM). Scale bar, 1 μm; insert scale bar 1 μm. (C) Quantification of bacteria cells containing liposomal fluorescence from STORM images. Each fluorescent locus on the cells’ membrane with a minimum size of 180 nm was counted as one individual cell-attached liposomal unit. Differences between groups shown in the box plot were tested with a one-way ANOVA. Boxes represent the 25–75 percentiles of the sample distribution, with black vertical lines representing 1.5 × IQR (interquartile range). Black horizontal line represents the median.

Figure 4

Internalization and cargo release by calcein-encapsulating PEG- and pMPC-functionalized vesicles incubated with P. aeruginosa cells and biofilms. Representative microscopy images of PA14 cells following 4 h (A) and 24 h (B) incubation, followed by thorough washing, with PEG- (upper panels) and pMPC- (lower panels) calcein-loaded liposomes. Cells were grown at 37 °C for 24 h prior to the incubation with liposomes. Images are presented as a fluorescent intensity of the calcein signal (green) and overlay between calcein-fluorescence and brightfield. Insert in lower right panel (A) shows a zoomed-in detail of bacteria cells displaying a weak but recognizable fluorescent intensity. Scale bars: main, 10 μm; insert, 1 μm. (C) Quantification of single cell microscopy images of the number of cells displaying luminal calcein signal following 24 h incubation of cells with either PEG-LUVs or pMPC-LUVs. Differences between groups shown in the box plot were tested with a one-way ANOVA. Boxes represent the 25–75 percentiles of the sample distribution, with black vertical lines representing the 1.5 × IQR (interquartile range). Black horizontal line represents the median. (D) and (E) Kinetic profile of calcein release from liposomes upon interaction with bacterial biofilms of strain PA14 (D) or LESB58 (E) at 37 °C incubated for 17 h. Liposomes were either nonfunctionalized (brown circles) or functionalized with 5% pMPC (red squares) and 5% PEG (green triangles) polymer. Calcein-loaded pMPC- LUVs were incubated with naive BM2G media as a negative control (blue circles). Results are an average of a minimum of three experiments.

Figure 5

Representative cryo-TEM images of pMPCylated vesicles: (A) unloaded liposomes, (B) liposomes loaded with SMX, and (C) coloaded with SMX/EA. White arrows indicate a dark structure inside liposomes. Scale bar, 50 nm. (D) Release profiles of SMX, EA, and SMX/EA from pMPC-functionalized liposomes at physiological conditions (37 °C, pH 7.2). Results are shown as an average and standard deviation of three independent experiments. Dashed lines represent a trend of the data points.

Figure 6

Effect of SMX and SMX/EA-LUVs on biofilm viability. (A) PA14-cell viability after two-dose treatment (4h each) with SMX- and SMX/EA- LUVs and free drugs quantified by MBEC-based resazurin assay. Blue boxes (SMX [1.5 mg/mL], SMX/EA [0.7/1.5 mg/mL]) represent treatment with free drugs. Data obtained from 8 or more biological repeats. (B(a–g)) Representative confocal microscopy images of the antibacterial effect of drug-loaded LUVs and free drugs on 24 h biofilm evaluated by a live (green)/dead (red) assay. Images show representative areas from chamber slides. (a) nontreated (NT) biofilm; (b) free SMX [1.5 mg/mL]; (c) SMX-loaded PEG-LUVs; (d) SMX/EA-loaded PEG-LUVs; (e) SMX/EA-loaded pMPC-LUVs; (f) SMX-loaded pMPC-LUVs. Scale bar, 50 μm. (g) Percentage of dead bacteria quantified from at least four different microscopic images. Data represent a minimum of two biological repeats with two technical repeats each. (C(a–d)): Confocal microscopy of cross sections of paraffin-embedded PA14-colony after 4h treatment with pMPC-LUVs loaded with SMX. (a) Preparation of a 10 μm-thin section of the colony upon treatment with (b) 5 μL unloaded pMPC-LUVs or (c) 5 μL pMPC-LUVs loaded with SMX. Live (green)/dead (red) staining was applied prior to fixation. Scale bar, 100 μm. (d) Spatial profiles for biofilm sections displaying variation in dead bacteria cells fraction upon injection with either SMX-loaded pMPC-LUVs (red) or PEG-LUVs (green), or nonloaded liposomes (blue). Details of the quantification in the Methods sections and Figure S15. Differences between groups shown in box plots were tested with a one-way ANOVA. Boxes represent the 25–75 percentiles of the sample distribution, with black vertical lines representing the 1.5 × IQR. Black horizontal line represents the median.

Figure 7

Schematic of proposed two-stage mechanism for calcium-mediated adhesion and fusion between pMPC- LUVs and P. aeruginosa membrane. (1) Liposomes functionalized with pMPC reach the bacterial membrane, where the presence of divalent cations (yellow points) at the cell membrane’s surface bridges pMPC with LPS and pulls them toward the cell surface. (2) Fusion between liposome and bacterial membrane occurs due to charge–charge interaction (see text).