In Vivo Biodistribution and Ototoxicity Assessment of Cationic Liposomal-Ceftriaxone via Noninvasive Trans-Tympanic Delivery in Chinchilla Models: Implications for Otitis Media Therapy

Abstract

Objectives:

We report the in vivo biodistribution and ototoxicity of cationic liposomal-ceftriaxone (CFX) delivered via ear drop formulation in adult chinchilla.

Methods:

CFX was encapsulated in liposomes with size of ~100 nm and surface charge of +20 mV. 100μl liposomes or free drug was applied twice daily in both external ear canals of adult chinchillas for either 3 or 10 days. Study groups included free ceftriaxone (CFX, Day 3: n=4, Day 10: n=8), liposomal ceftriaxone (CFX-Lipo, Day 3: n=4, Day 10: n=8), and a systemic control group (Day 3: n=4, Day 10: n=4). Ceftriaxone delivery to the middle ear and systemic circulation was quantified by HPLC assays. Liposome transport was visualized via confocal microscopy. Auditory brainstem response (ABR) tests and cochlear histology were used to assess ototoxicity.

Results:

Liposomal ceftriaxone (CFX-Lipo) displayed a ~658-fold increase in drug delivery efficiency in the middle ear relative to the free CFX (8.548 ± 0.4638% vs. 0.013 ± 0.0009%, %Injected dose, Mean ± SEM). CFX measured in blood serum (48.2 ± 7.78 ng/ml) following CFX-Lipo treatment in ear was 41-fold lower compared to systemic free-CFX treatment (1990.7 ± 617.34 ng/ml). ABR tests and histological analysis indicated no ototoxicity due to the treatment.

Conclusion:

Cationic liposomal encapsulation results in potent drug delivery across the tympanic membrane to the middle ear with minimal systemic exposure and no ototoxicity.

1. Introduction

Otitis media (OM), an inflammatory condition of the middle ear (ME) affecting millions of people worldwide, poses a significant health challenge, especially in pediatric patients [1], [2]. This disease is a leading cause of hearing impairment, surgery, and antibiotic prescription, demonstrating its critical health impact [3], [4]. Multi-day oral antibiotic therapy is the primary treatment for acute OM. However, the pervasive issue of antibiotic resistance, marginal delivery of the drug to the ME, and gastrointestinal (GI) and other systemic side effects limit the efficacy of oral antibiotic administration [3], [5]–[9]. To date, there have not been successful methods to deliver antimicrobial therapy across an intact tympanic membrane (TM) to treat OM. Therefore, relatively invasive methods such as surgical implantation of a tympanostomy tube to alleviate OM and allow transtympanic antimicrobial delivery have been utilized, posing administration challenges, particularly in young children [10]–[15]. Consequently, there is a pressing need for more effective, less invasive, and safer treatment options for OM.

The challenge of developing effective non-invasive drug delivery systems for OM lies in overcoming the TM barrier. The TM is comprised of up to four layers of cells bound by tight junctions; these layers restrict solute and water diffusion through paracellular routes, limiting effective drug delivery via non-invasive external ear application [16], [17]. To achieve therapeutic efficacy against common OM pathogens such as Nontypeable Haemophilus influenzae (NTHi) and Streptococcus Pneumoniae (SP), the drug must reach the minimum inhibitory concentration (MIC) of approximately 1 μg/mL in the middle ear, which is currently only possible with invasive trans-TM methods [18]–[20]. Previously reported non-invasive trans-TM delivery methods leverage drug carriers and enhancers to facilitate drug penetration through the TM barrier. Antibiotic-loaded hydrogels with chemical permeation enhancers such as limonene, sodium dodecyl sulfate, and bupivacaine have been shown to effectively deliver ciprofloxacin across the TM for treating OM [21]–[23]. The benefits of this method include the controlled release of the drug, which allows for sustained antimicrobial activity across 7+ days. Similarly, TM crossing-peptides identified by phage-display methods conjugated to drugs such as amoxicillin or neomycin have been used to enhance drug delivery to the middle ear [24], [25]. Despite some promising prospects for these methods, they are not without potential pitfalls. The employment of chemical permeation enhancers can potentially raise the risk of toxicity and inflammation[26], while the attachment of targeting peptides to drugs or drug carriers could possibly impact the stability or effectiveness of the correlating drug [27]. Challenges such as poor stability, rapid renal clearance, unwanted immune reactions, and challenges translating favorable features into actual clinical results exist within peptide-drug conjugates (PDCs) [28]. This is evident as PDCs frequently come up short in clinical trials owing to complications in converting well-thought-out drug conjugate designs into effective treatments. To date, none of these techniques have resulted in a clinically available treatment modality. Therefore, more research is necessary to refine these techniques and lessen their potential shortcomings[29]–[33].

To address these challenges of trans-TM delivery, nanoparticle-based drug-delivery systems for the middle and inner ears have also been proposed recently. These nanomaterials offer a unique potential for overcoming biological barriers, such as the TM, and for delivering therapeutics effectively to targeted areas [34]–[38]. Herein, we demonstrate, for the first time, the utility of cationic liposomes for trans-TM antibiotic delivery. Among the various nanoparticle-based delivery systems, cationic liposomes have been recognized for their potential in treating bacterial infections, including those affecting the lungs, skin, and gastrointestinal tract [39]–[43]. Liposomes, vesicular structures composed of lipid bilayers, can encapsulate both hydrophobic and hydrophilic drugs, thereby offering a versatile platform for antibiotic delivery [39], [44]. Charged liposomes have shown promise in enhancing drug concentration within the lungs, skin, and gastrointestinal tract, encapsulating multiple drugs, and reducing systemic side effects [42], [45]–[47]. Charged nanoparticles, particularly cationic liposomes, are known to cross cell layers using various mechanisms like phagocytosis, micropinocytosis, endocytosis, direct diffusion, and adhesive interactions [48]. This complex process of cellular internalization is significantly influenced by the physical and chemical attributes of the nanoparticles, including their size, shape, surface charge, and composition [48], [49]. These nanoparticles utilize passive pathways like diffusion significantly driven by their high surface area to volume ratio, and adhesive interactions for traversing cellular membranes and the body’s critical outer layers such as skin or gastrointestinal lining, thereby enhancing their penetration and cellular uptake [50], [51]. In the specific context of ME treatment, this ability could potentially amplify the penetration of cationic liposomes across the TM, thereby optimizing their efficacy in delivering encapsulated drugs. The surface charge and size of these liposomes can be fine-tuned through lipid combinations, enabling them to interact selectively with cell membranes, improve muco-adhesion, and optimize TM crossing [52].

However, the development and application of these nanoparticle systems, including cationic liposomes, come with their own set of challenges. These encompass issues such as cellular toxicity, instability, and drug delivery efficacy for in vivo applications [53]–[57]. Therefore, gaining a deeper understanding of the factors that influence the safety and efficacy of these delivery systems is critical and has been the focus of our laboratory. In this manuscript, we describe the liposomal formulation of third generation antibiotic Ceftriaxone, evaluate the stability and release of the drug from cationic liposome formulations, and employ a clinically relevant chinchilla-based animal model of ME anatomy and evaluate drug delivery efficacy, biodistribution, and ototoxicity of liposomal antibiotics formulated for twice daily eardrop style application via ex vivo and in vivo assays.

Materials and Methods

All methods employed in this study were carried out in accordance with relevant guidelines and regulations, following approved protocols by the Medical College of Wisconsin Institutional Biosafety Committee (IBC) and Institutional Animal Care and Use Committee (IACUC). The Medical College of Wisconsin (MCW) holds an Animal Welfare Assurance (Assurance number D16-00064 (A3102-01)) on file with the Office of Laboratory Animal Welfare, National Institutes of Health (NIH). Animal experiments were performed in compliance with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication no. 85–23, revised 1996). [58]

2.1. Study design

The primary goal of this study was to evaluate the in vivo biodistribution and ototoxicity of noninvasive trans-tympanic delivered liposomal ceftriaxone for OM treatment. Chinchillas were chosen as the animal model due to their anatomical similarities to humans, particularly in terms of ME structure. The study was divided into several stages, including liposome preparation and characterization, ceftriaxone loading assessment, in vitro and ex vivo drug release, antimicrobial efficacy evaluation, biodistribution, and ototoxicity tests. Additionally, cochlear histological analysis and ceftriaxone quantification using high-performance liquid chromatography (HPLC) were performed. Confocal microscopy was employed to visualize the biodistribution of the liposomes in the harvested tympanic membranes and cochleae.

Experiments were conducted in a controlled, randomized manner, with each group receiving distinct treatment regimens. The chinchillas were divided into three experimental groups and two timepoints. The groups included free ceftriaxone (CFX), liposomal-ceftriaxone (CFX-Lipo), and a systemic control group. Data collection was terminated after 3 and 10 days for the biodistribution experiments. The number of animals per group for each time point was as follows: free ceftriaxone (CFX) (Day 3: n=4, Day 10: n=8), liposomal-ceftriaxone (CFX-Lipo) (Day 3: n=4, Day 10: n=8), and a systemic control group (Day 3: n=4, Day 10: n=4).

A power analysis was conducted to determine the appropriate sample size for the study. The analysis was focused on detecting differences between the CFX and CFX-Lipo groups, as the main objective of the study was to evaluate the efficiency of trans-tympanic delivery of liposomal-CFX. The power analysis was performed using the “sampsizepwr” function from the MATLAB Statistics and Machine Learning Toolbox. (Mathworks Inc. Natick, MA) This function calculates the required sample size for a two-sample t-test to achieve the desired power. The analysis indicated that a sample size of 8 for each treatment group would provide 80% power to detect a 50% difference in drug concentration in the middle ear, assuming a standard deviation of 20%, a significance level of 0.05, and a two-sided test.

In vivo biodistribution and ototoxicity experiments were performed in a blinded manner. Ototoxicity tests employed ABR measurements, which were conducted at baseline, early post-treatment (day 7), and late post-treatment (day 30) time points to comprehensively evaluate potential ototoxic effects (n=4 for all groups, Lipo, CFX, CFX-Lipo, Control).

Data collection and analysis were conducted utilizing appropriate controls and statistical tests, ensuring the validity and reliability of the results.

2.2. Animals

Chinchillas of mixed genders, aged between 4-6 months, with an average weight of 512 grams, were obtained from Hickory Way, LLC (Richland, MI). The animals were housed in the temperature-controlled vivarium with a 12-hour light/dark cycle and had commercial food and water ad libitum. All procedures and animal handling described herein were conducted according to the protocol approved by the Institutional Animal Care and Use Committee at the Medical College of Wisconsin.

2.3. Preparation and Characterization of Charged Liposomes

Charged liposomes were synthesized utilizing a modified version of the standard thin film hydration method with a total lipid content of 25 mg, as previously reported, with minor alterations [58]. Briefly, an optimized formulation consisting of cationic DOTAP (1,2-dioleoyl-3-trimethylammonium-propane), DOPC (1,2-dioleoyl-sn-glycero-3-phosphocholine), and DOPE (1,2-dioleoyl-sn-glycero-3-phosphoethanolamine) (Figure 1A) liposomes was loaded with Ceftriaxone (CFX). The cationic liposomes were synthesized using DOTAP: DOPC: DOPE (50:46:4 molar ratio) in 2 ml of chloroform (Avanti Polar Lipids, Inc., Alabaster, AL). Following rotary evaporation, the resulting lipid film was dried under a vacuum overnight to ensure complete solvent removal. The lipid film was then rehydrated with 2 ml of a 3 mg/ml CFX solution dissolved in 1X HBS (pH = 7.2). The lipid film was thoroughly mixed with the solution through vortexing, producing large multilamellar non-extruded liposomes.

Figure 1: Composition, Imaging, and Characterization of Liposomal Formulations.

(A) Illustrates the structural organization of the DOPE, DOPC, and DOTAP lipids, highlighting the hydrophilic head groups in red and the hydrophobic chains in blue. (B) Exhibits the Transmission Electron Microscopy (TEM) images of liposomes encapsulated in both a Graphene Liquid Cell (GLC) and graphene layers. The left panel displays liposomes at low magnification, and the right panel presents hydrated liposomes at high magnification within a Graphene Liquid Cell (GLC), demonstrating the formation and morphology of Ceftriaxone-loaded liposomes. The GLC corners are marked by red dashed lines, and an air bubble is indicated by a black dotted line. (C) Showcases the zeta potential readings for unloaded liposomes (Lipo), Ceftriaxone-loaded liposomes (CFX-Lipo), and fluorescently labeled liposomes (RhB-CFX-Lipo), providing insights into their surface charge properties. (D) Details the hydrodynamic size data of liposomal formulations, giving an understanding of their size range in a colloidal system.

To generate small, extruded liposomes, the non-extruded liposomes were subjected to a 10-minute sonication process using an ultrasonic bath (2800 ultrasonic cleaner, Branson, MO, United States) at 37°C, followed by freezing at −80°C for 30 minutes. The liposomes were thawed in a 37°C water bath for 10 minutes and vortexed for 1 minute. Freeze-thaw cycle was conducted ten times. The liposomes were then extruded 11 times through a 100 nm polycarbonate membrane (Avanti Polar Lipids, Inc., Alabaster, AL) using an Avanti MiniExtruder^® for size homogenization at room temperature.

The removal of non-entrapped CFX was performed using Amicon Ultra-50^® centrifugal filtration devices (Millipore, Billerica, MA) with a 30 kDa molecular weight cut-off. Briefly, 15 ml of the 20 times diluted liposome suspension was added to the upper chamber of the ultrafiltration tube and centrifuged at 4000 rpm at 4°C using an Eppendorf^® 5810R centrifuge (Eppendorf Corp, Germany) until the volume was reduced to 1000 μl. The non-entrapped CFX was collected in the lower chamber of the ultrafiltration tube, while the purified liposome suspension was obtained from the upper chamber.

For fluorescent labeling of the liposomal formulation, a similar technique was employed, with the addition of Lissamine rhodamine-B (ThermoFisher Scientific, MA, USA) dye to the lipid mixture in chloroform. The resulting fluorescent lipid film was hydrated with CFX solution, followed by the same freeze-thaw and extrusion processes to form fluorescently labeled liposomes, referred to as RhB-CFX-Lipo. The final product was stored at 4 °C until further use.

2.4. Assessment of Hydrodynamic Size and Zeta Potential of Liposomes

The hydrodynamic size and zeta potential of unloaded liposomes (Lipo), Ceftriaxone-loaded liposomes (CFX-Lipo), and fluorescently labeled liposomes (RhB-CFX-Lipo) were assessed after extrusion through a 100 nm polycarbonate membrane using a Malvern Zetasizer^® 3600 (Malvern Instruments Ltd., UK).

2.5. Electron Microscopy and Morphological Assessment

The preparation of liposomal formulations encapsulated in Graphene Liquid Cells (GLCs) and graphene layers followed a two-step process, drawing upon techniques similar to those previously reported [59]. Dry samples were directly deposited onto a 2000 mesh graphene-coated TEM grid (Graphene Supermarket Inc., NY, USA) and placed in a desiccator for 2 hours. Then, they were sandwiched between a second graphene-coated TEM grid. On the other hand, hydrated samples were immediately sandwiched between two graphene-coated TEM grids to maintain their hydrated state for encapsulation within GLCs.

The morphology and physical properties of the liposomes were assessed using a JEOL 2100 Transmission Electron Microscope (TEM) operated at a low electron voltage (60-80 eV), minimizing potential radiation damage. The GLCs contained various areas of liposomes in solution and dried liposomes sandwiched between graphene layers. An increased electron dose rate confirmed the presence of liquid in hydrated samples via gas bubbles generation inside the GLCs (Figure 1B).

2.6. Ceftriaxone Loading Assessment

The loading efficiency and capacity of ceftriaxone into liposomes were assessed using a modified method, as previously reported [60]. Briefly, 5 mL of CFX-Lipo suspension was centrifuged at 4700 × g for 2 hours using an Allegra^® X-15R centrifuge (Beckman Coulter, Fullerton, California) to yield a pellet, and the supernatant was subsequently discarded. The pellets were resuspended in 6 mL of 0.5% (w/v) SDS solution in HEPES buffer via vortexing and sonication. Ceftriaxone concentration analysis was performed using a high-performance liquid chromatography (HPLC) described in section 2.13. The encapsulation efficiency was calculated using the following formula:

$$\text{encapsulation efficiency}(\%)=(\text{amount of ceftriaxone in liposomes}/\text{total amount of ceftriaxone used})\times 100.$$

2.7. In Vitro Drug Release from Liposomes (Fractional Release)

The assessment of ceftriaxone release from liposomes in vitro was performed utilizing a modified dialysis sac technique, as outlined by Chidambaram and Burgess [61]. A 1 mL sample of a freshly prepared 25 mg (lipid)/mL liposomal-ceftriaxone (CFX-Lipo) solution in phosphate buffer saline, PBS (pH 7.4) was introduced into a 100 kDa MWCO Spectra/Por^® Float-ALyzer^® (Repligen, MA, USA) Dialysis sacs loaded with liposome suspensions were placed in glass tubes (Kimax glass culture tubes; 25 mm × 200 mm) containing 50 mL of PBS, maintained at a stable temperature of 37/4°C in a shaking water bath (New Brunswick, Edison, NJ), and stirred at 50 rpm. The temperature of the buffer was kept at 37°C or 4°C for the entire study. At specific time intervals (0.5, 1, 2, 4, 8, 16, 32, 48, and 60 hours), 1 mL aliquots were withdrawn for release analysis and promptly replaced with fresh buffer. Ceftriaxone concentration was quantified using HPLC analysis at λ_max = 275 nm. The entire experiment was carried out in triplicate, with results presented as mean ± standard error of the mean (SEM) (n = 3).

2.8. Ex Vivo Drug Release from Liposomes Through TM

An ex vivo study was conducted to evaluate liposomal ceftriaxone’s drug delivery capability in OM treatment adapted from the method reported previously [62]. Chinchillas were euthanized using an intracardiac injection of a pentobarbital-based euthanasia solution following an intramuscular injection of a ketamine-xylazine mixture. Temporal bones containing the ear canal, tympanic membrane, and intact middle ear cavity were aseptically collected from the chinchillas and positioned upright.

An aqueous suspension of liposomal-ceftriaxone (200 μl) was applied to the harvested chinchilla auditory bullae’s external auditory canal (EAC), which was preserved in phosphate-buffered saline (PBS). Simultaneously, under the consistent experimental condition of 37°C, a comparative 200 μl dose of ceftriaxone was applied to control samples. Through a one-centimeter hole at the dorsal-most part of the bulla, 800 μl of sterile PBS was introduced into the ME cavity to collect the permeated drug from the TM. The fractions of PBS in the ME cavity (200 μl) were collected at various time intervals (5, 15, 30, and 60 minutes) to monitor the diffusion of liposomal-ceftriaxone and evaluate its drug delivery efficacy.

The permeation of ceftriaxone across the TM into the receiving chamber was quantified using HPLC. The results were expressed as mean ± SEM (n=3).

2.9. Ex Vivo Evaluation of TM Penetration and Confocal Microscopy

To evaluate the biodistribution and ME delivery of the fluorescently labeled Liposomes, RhB-CFX-Lipo, the TM and cochlea from euthanized chinchillas were harvested post-treatment. Three formulations were used in this study: PBS as a control, RhB-CFX-Lipo, and Rhodamine B (RhB) alone, which served as a control to ensure that the fluorescence observed was not solely due to the dye penetrating the ear.

The formulations were left for 1 hour in the EAC of the intact bulla, followed by a thorough washing procedure, repeated five times, to remove any unabsorbed formulation. The TM and cochlea were then dissected and fixed in 4% paraformaldehyde at 4°C for 24 hours. The TM was embedded in an optimal cutting temperature (OCT) solution and serially sectioned at a thickness of 10 microns. The cochlea underwent a decalcification process in 10% Ethylenediaminetetraacetic acid (EDTA) at 40°C for 48 hours, after which a basal turn of the cochlea was dissected and prepared for confocal microscopy.

Confocal microscopy was performed using a custom-built confocal microscope with an ECLIPSE Ti2 (Nikon Corporation, Japan). Images were captured using 10×, 40×, and 60× magnification. Tissue autofluorescence was utilized to visualize the cell layers of the TM and cochlea sections, with the excitation/emission wavelengths for tissue autofluorescence set at 488 nm/520 nm. The Rhodamine B fluorescence was captured using the appropriate excitation/emission wavelengths set at 555 nm/580 nm.

2.10. Evaluation of Antimicrobial Efficacy of Liposomal-Ceftriaxone

Non-typeable Haemophilus influenzae (NTHi) strain 86-028NP was cultured overnight on brain heart infusion agar supplemented (sBHI) with hemin (ICN Biochemicals) and nicotinamide adenine dinucleotide (NAD) (Sigma), and then harvested and adjusted to a concentration of approximately 10⁶ CFU/mL using sterile PBS. To assess the bactericidal efficacy, different treatment groups consisting of Lipo, CFX-Lipo, and CFX solution were administered. The CFX-Lipo treatment contained a concentration of 12.5μg/ml, which was equivalent to the final concentration of the CFX solution. The Lipo control group contained the same quantity of liposomes as the CFX-Lipo group. The mixtures of bacteria and treatments were incubated at 37°C in a 5% CO₂ atmosphere for 12-16 hours. The viable bacteria were enumerated by performing serial dilutions of the cultures, followed by plating on sBHI agar plates and incubating at 37°C with 5% CO₂ for 24-48 hours and reported as mean ± SEM (adapted from [62]).

2.11. Biodistribution of Ceftriaxone-Loaded Liposomes

The study was conducted using chinchillas, which were randomly assigned to two time points and three groups: CFX (Day 3: n=4, Day 10: n=8), CFX-Lipo (Day 3: n=4, Day 10: n=8 and systemic control (Day 3: n=4, Day 10: n=4). The experimental drug formulations were applied to the chinchilla TM using a speculum, and a successful application was verified by inspecting the TM under a surgical microscope (Prescott’s, Inc.).

Animals in the CFX and CFX-Lipo groups received bilateral trans-tympanic applications of 200μ1 per ear of either free ceftriaxone (3mg/ml) or CFX-Lipo (6.25mg/ml of lipid), administered twice daily for three or ten days. The systemic control group was given injectable ceftriaxone intramuscularly at a 4mg/kg dose, administered once daily for three or ten days. Following the final treatment, middle ear lavage samples were collected using 2ml of sterile PBS (pH 7.4) for drug concentration analysis.

The drug concentration in the middle ear lavage samples was assessed using a validated HPLC method (Section 2.13) and reported as mean ± SEM.

2.12. Ototoxicity Test

To evaluate the ototoxicity of liposomal-ceftriaxone (CFX-Lipo), an in vivo study was conducted using three groups of four chinchillas each: Lipo, CFX, and CFX-Lipo. The trans-tympanic administration procedure and treatment regimen were as previously described in section 2.10. Non-treated animals served as a normal control group (n=4). ABRs were measured for each animal at three different time points: baseline (prior to treatment application), early post-treatment (day 7), and late post-treatment (day 30). The ABR tests were carried out in an acoustic chamber with the animals under ketamine/xylazine anesthesia. Acoustic stimuli consisting of 2, 4, 8, and 16 kHz tone pips were presented in a free-field condition and ABRs recorded using a TDT BioSig Rx6 system (Tucker Davis Technologies, FL, USA), as described previously[63]. ABRs were measured in 5 dB sound pressure level (SPL) increments, starting at the maximum amplitude of 90 dB with decreasing intensity until threshold was reached. Threshold was determined as the lowest stimulus level at which the response waveforms were visually present and repeatable. The ABR procedure was conducted by a trained technician, PK. The data obtained was further analyzed by and blinded to our senior auditory scientist, CR, who verified and confirmed the hearing threshold. Statistical analysis was performed by KY. Following the tests, cochleae were collected for histological analysis.

2.13. Histologic Analysis

Eight cochleae were collected, with two from each experimental group, and subjected to fixation in a solution containing 2.5% glutaraldehyde and 0.1M sodium cacodylate for a 24-hour period. Following fixation, the cochleae were post-fixed using a 1% osmium tetroxide solution, rinsed thoroughly, and then decalcified. The specimens were dehydrated through graded methanol before embedding in EMBed 812 epoxy resin. Serial sections of 5 microns thickness were obtained from the embedded cochleae, which were then stained using toluidine blue and mounted onto microscope slides. Structural analysis of the cochlea was performed by CR using an Axiocam microscope (Zeiss, Gottingen, Germany) to further assess potential ototoxic effects of the noninvasive trans-tympanic delivery of liposomal-ceftriaxone in the treatment of otitis media.

2.14. Ceftriaxone Quantification

To evaluate Ceftriaxone levels in all samples, including CFX-Lipo, the ME lavage fluid, and sera, a high-performance liquid chromatography (HPLC) technique was employed, utilizing UV absorbance detection at a wavelength of 270 nm. The analysis was carried out using an SPD-M20A detector (Shimadzu, Kyoto, Japan) and an Eclipse XDB C8 column (4.6mm × 150mm, 3μm particle size). The mobile phase consisted of an 80:20 ratio of acetonitrile to 0.1% trifluoroacetic acid (pH=2.0), with a flow rate of 1 mL/min. Prior to the measurement, samples were filtered through a 0.4μm filter. A standard curve for HPLC was developed based on the area under the curve reported by the integrator and demonstrated linearity over the studied range (0.001 to 100 μg/ml).

2.15. Statistical Analysis

All collected data were expressed as mean ± standard error of the mean (SEM) values. We performed one-way Analyses of Variance (ANOVA) to compare the differences among groups. In cases where significant differences were identified, we conducted pairwise post hoc analysis by employing the Tukey-Kramer method to adjust for multiple comparisons. Throughout the analysis, a p-value of less than 0.05 was considered to indicate statistical significance.

In the context of our findings, we have represented the degrees of statistical significance with specific notations: * for P < 5e-2, ** for P < 1e-3, *** for P < 1e-4, and **** for P < 1e-6. This approach allows for an intuitive understanding of the statistical implications of our results.

3. Results

3.1. Characterization of Liposome Formulations

The liposomal formulations were first prepared, characterized, and evaluated based on their zeta potential, as depicted in Figure 1C. All liposomal formulations displayed closely ranged zeta potential values.; unloaded liposomes (Lipo) demonstrated a potential of 21.7 ± 0.62 mV. After loading with Ceftriaxone, the charged liposomes, or CFX-Lipo, exhibited a reduced zeta potential of 16.8 ± 1.05 mV. Upon further incorporation of Rhodamine B (RhB), the zeta potential slightly increased to 18.3 ± 0.74 mV for RhB-CFX-Lipo, signifying the alteration of surface charge characteristics post-labeling and drug-loading.

The hydrodynamic size (HDS) for the liposomal formulations was assessed to determine the effects of drug loading and fluorescent labeling on the size of the liposomes, as depicted in Figure 1D. The unloaded Lipo demonstrated a peak HDS of 91.83 nm. With Ceftriaxone loading, CFX-Lipo showed an increase in size to 103.04 nm. Labeling of RhB expanded the size of RhB-CFX-Lipo to 126.80 nm, indicating that the addition of RhB influences the size of the liposomes.

Simultaneously, the polydispersity index (PDI) of the samples was measured, indicating homogenous size distribution with a uniform PDI value of 0.161 ± 0.008, 0.160 ± 0.01, and 0.083 ± 0.012 across all three liposomal samples, Lipo, CFX-Lipo and RhB-CFX-Lipo.

Transmission Electron Microscopy (TEM) and Graphene Liquid Cells (GLCs) were utilized to visually substantiate the characterization of liposomal formations under hydrated conditions. As shown in Figure 1B, the TEM images confirmed liposome size and morphology consistency. Notably, the hydrated state revealed visible regions demonstrating the lipid bilayer of the liposomes. Additionally, in the dried state, there was no visible aggregation of the liposomes.

The drug encapsulation efficiency and loading capacity of the liposome formulation were evaluated to ensure efficient drug delivery. The Ceftriaxone concentration was determined to be equivalent to 17.5 μg/mL, extrapolating to a loading capacity of 17.5 μg of Ceftriaxone per 12.5 mg of lipid. The encapsulation efficiency was calculated to be 1.16%.

3.2. Antibacterial Efficacy Study

The antibacterial efficacy of ceftriaxone-loaded liposomes (CFX-Lipo) was evaluated against the NTHi strain 86-028NP bacterial cultures. Our analysis revealed that CFX-Lipo maintained antibacterial activity on par with the free drug, CFX. In contrast, the unloaded liposomes (control) demonstrated a dose-dependent antibacterial efficacy (Figure 2).

Figure 2: Antimicrobial Efficacy of Ceftriaxone-Loaded Liposomes.

The graph represents antibacterial activity across different treatments and concentrations. Unloaded liposomes manifested a dose-dependent reduction in bacterial counts compared to the untreated control. Both ceftriaxone (CFX) and ceftriaxone-loaded liposomes (CFX-Lipo) exhibited a significant decrease in bacterial presence across all concentrations, with CFX-Lipo demonstrating equivalent antibacterial potency to free CFX. Error bars denote the standard deviation for unloaded liposomes (n = 4).

The collected data from four independent experiments showed that untreated samples (control with no treatment) exhibited a mean bacterial count of 8.59 ± 0.06 log 10 (CFU/ml). When exposed to blank liposomes at concentrations of 25, 50, 100, and 200 μl, a notable reduction in bacterial count was observed, ranging from 8.68 ± 0.03 to 2.57 ± 0.02 log 10 (CFU/ml), illustrating a dose-dependent antibacterial effect.

The CFX at equivalent volumes eliminated bacterial presence, with a consistent count of 1 log 10 (CFU/ml) across all tested volumes, signifying a near-total bacterial eradication. Significantly, the CFX-Lipo mirrored the potent antibacterial performance of the free CFX. Regardless of CFX-Lipo volume (25, 50, 100, and 200 μl), a minimal bacterial count of 1 log 10 (CFU/ml) was consistently observed, indicating that the liposomal delivery did not compromise the antibacterial potency of CFX.

3.3. Ceftriaxone Release from Liposomes and Permeation through the TM

To evaluate the effectiveness of our noninvasive trans-tympanic delivery system, we assessed the release of ceftriaxone from liposomes both in vitro and ex vivo, and its transport through the TM.

The in vitro fractional release of ceftriaxone from liposomes exhibited a temperature-dependent pattern. At 37°C, the drug release was rapid, achieving an average release of 0.997 ± 0.002 at 60 hours, with over 90% of ceftriaxone released within the 8-16 hour window (Figure 3A). Conversely, under storage conditions at 4°C, drug release substantially slowed, showcasing an average release of 0.199 ± 0.020 at 60 hours, indicating the stability of our liposomal formulation.

Figure 3: Comparative Analysis of Ceftriaxone Release and Transport through the Tympanic Membrane.

(A) The graph illustrates the temperature-dependent fractional release of ceftriaxone from liposomes at 37°C and 4°C over a period of 60 hours. (B) The graph shows a comparative ex vivo analysis of the cumulative release of free ceftriaxone and liposomal ceftriaxone (CFX-Lipo) over a 60-minute period. (C) confocal images of the TM treated with Rohodamin B (RhB) alone and RhB-liposomal ceftriaxone (RhB-CFX-Lipo). Separate channels for Whitefield (WF), autofluorescence (AF), and RhB are displayed, illustrating the distribution of RhB across the thickness of the TM in the liposomal ceftriaxone-treated sample.

In the ex vivo analysis, the liposomal ceftriaxone showed a more efficient drug release than free ceftriaxone. Our liposomal formulation exhibited an increasing release rate, achieving an average cumulative drug concentration of 1.247 ± 0.016 μg/mL at 60 minutes. In contrast, free ceftriaxone plateaued early, reaching an average cumulative concentration of 0.433 ±0.031 μg/mL at 60 minutes (Figure 3B). The experiment was concluded at 60 minutes due to potential tissue degradation and loss of biological function.

3.4. Visualization of Liposomal Penetration through the TM using Confocal Microscopy

Confocal microscopy was utilized on ex vivo TMs to visually affirm the trans-tympanic transport of our liposomal formulation. The fluorescent signal from Rhodamine B (RhB) was used as an integral part of the liposomal formulation to track its distribution. The images distinctly showed uniform penetration of the RhB-labelled formulation across the TM. However, when RhB was applied alone, it was found to reside only on the outer surface of the TM. This clear distinction establishes that the observed fluorescence is primarily due to the penetration of the liposomal formulation rather than the RhB dye itself (Figure 3C).

3.5. Quantitative Assessment of Liposomal-Ceftriaxone Biodistribution and Concentration in Chinchillas

The biodistribution of CFX and CFX-Lipo in the ME and blood serum was evaluated at two time points, day 3 and day 10, post-application. The results were determined by investigating the concentration of CFX in the ME and blood serum using HPLC analysis (Figure 4).

Figure 4: Comparative Analysis of Ceftriaxone Biodistribution and Concentration in Middle Ear and Serum Samples.

The six panels illustrate the biodistribution of ceftriaxone following different delivery methods—free ceftriaxone (CFX), liposomal ceftriaxone (CFX-Lipo), and systemic control (CFX-Sys-Control)—at two-time points (Day 3 and Day 10). The rows represent (from top to bottom) the absolute concentration of ceftriaxone in the middle ear, the percentage of detected ceftriaxone relative to the total administered dose in the middle ear and, finally, the absolute ceftriaxone concentration in the serum. Columns represent the two time points—Day 3 and Day 10. Box plots represent the first quartile, median, and third quartile for each dataset.

In the ME lavage samples, on day 3, the average concentration of CFX detected was 35.1 ± 9.45 ng/ml (Mean ± SEM) for the CFX group (n=4), while the CFX-Lipo group (n=4) displayed a higher concentration of 170.1 ± 29.78 ng/ml. The systemic control group (n=4) showed the highest CFX concentration, with 85.6 ± 20.98 ng/ml. Similarly, on day 10, the concentration of CFX increased in all groups with the highest levels detected in the CFX-Lipo group (1596.0 ± 98.65 ng/ml, n=8), followed by the systemic control group (1429.2 ± 292.19 ng/ml, n=4) and the free CFX group (772.9 ± 49.40 ng/ml, n=8).

When analyzing the CFX biodistribution in terms of the percentage of total administered dose detected, on day 3, an average of 0.002 ± 0.0005% of the administered dose was detected in the CFX group, whereas a substantially higher percentage was detected in the CFX-Lipo group (3.240 ± 0.567%). The systemic control group displayed the lowest proportion with 0.005 ± 0.001% of the administered dose detected. On day 10, the CFX-Lipo again showed the highest percentage (8.548 ± 0.4638%), followed by the CFX group (0.013 ± 0.0009%) and the systemic control group (0.024 ± 0.0048%).

In the blood serum samples, at day 3, the CFX-Lipo group demonstrated the lowest CFX concentration (0.39 ± 0.16 ng/ml, n=4), significantly less than the free CFX (6.87 ± 2.07 ng/ml, n=4) and systemic control groups (13603.0 ± 4141.10 ng/ml, n=4). The pattern was consistent on day 10, where the concentration of CFX in the CFX-Lipo group remained the lowest (48.2 ± 7.78 ng/ml, n=8) compared to the values obtained in the CFX group (125.0 ± 42.81 ng/ml, n=8) and the systemic control group (1990.7 ± 617.34 ng/ml, n=4).

3.6. Ototoxicity Evaluation of Liposomes

In understanding the ototoxic potential of our treatment, investigations were carried out concerning histologic changes, ABR, and the penetration of the liposomes into the inner ear.

The chinchilla cochlear histologic analysis was performed in eight samples. Each sample was examined using 8-10 serial sections. The cochlea structure was examined in all the basal, middle and apical turns. The best representation of each sample groups was showed (Figure 5A). The spiral organ of corti demonstrated three intact OHC with well-defined cilia and one IHC, particularly the basal turn where the effects of ototoxicity may present. The comparative evaluation of the Lipo, CFX-Lipo, CFX cochleas showed minimal difference from that of the Control group. The ABR threshold (Figure 5B) corroborated this histologic evaluation indicating that cochlear integrity remained unaffected post trans-tympanic administration, underscoring the otological safety of the treatment.

Figure 5: Ototoxicity and Auditory Function Assessment After Liposomal-Ceftriaxone Treatment.

(A) Histological examinations of cochleae from all groups (Lipo, CFX-Lipo, CFX, and control) showed no ototoxic changes with clear representations of Inner Hair Cells (IHC) and Outer Hair Cells (OHC) (n = 4 for each group)(scale bars= 50 μm). (B) ABR tests at baseline, 7 and 30 days post-treatment for both ears found no significant intergroup differences, demonstrating preserved auditory function across all groups (n = 4 for each group).

To assess potential changes in hearing due to the treatment, we utilized ABR tests, administered at baseline, 7 days post-treatment, and 30 days post-treatment across all groups. Consistent with the results of the histologic analysis, ABR thresholds indicated no significant differences in the hearing levels among the different groups (Figure 5B). This indicates that the trans tympanic delivery of noninvasive liposomal-ceftriaxone did not adversely affect the chinchillas’ hearing.

We also explored the ability of liposomal penetration beyond the ME into the cochlea, despite not being our primary objective. Based on the confocal images of the inner ear, minimal penetration of liposomes into the cochlea was noted (Figure 6A). A more detailed analysis of these images demonstrated small traces of liposomes present in the inner and outer hair cells (Figure 6B). Despite the signal extraction being minimal, requiring amplification to distinguish between treated and control samples, preliminary evidence was provided indicating the liposomes’ potential ability to traverse into the inner ear. These results, however, should be interpreted with caution due to the study was not primarily designed to precisely identify the levels of liposome penetration into the inner ear.

Figure 6: Confocal Microscopy Images Demonstrating the Penetration of Liposomes into the Inner Ear.

(A) The first row of images displays the overall inner ear treated with liposomal-ceftriaxone at a 10X magnification (scale bar: 200 μm). The region of interest inside the yellow box in the middle image is further magnified at 40X in the second row, with marked areas showing the Organ of Corti (OC) and the Scala Vestibuli (SV). (B) Panel B provides a higher resolution view of both Inner Hair Cells (IHC) and Outer Hair Cells (OHC), magnified at 60X and revealing the presence of the liposomal-ceftriaxone (scale bar: 100 μm). Each image is broken down into different channels: the whitefield channel (WF) depicts the general morphology, tissue autofluorescence channel (AF) represents the inherent tissue emissions, and the rhodamine channel (RhB) pinpoints the location of the liposomal-ceftriaxone.

4. Discussion

This study aimed to develop and evaluate a cationic liposomal delivery system for the noninvasive trans-tympanic administration of ceftriaxone in the treatment of OM. Our research involved a thorough characterization of the liposomal formulations, an assessment of their antibacterial efficacy, and both in vivo and ex vivo investigations into the drug’s release. The liposomal formulations were designed with an emphasis on enhancing therapeutic potential. DOTAP was selected for its cationic properties, which facilitated adhesion to the cell membrane and enhanced drug release [64], [65]. DOPC was included to provide stability and prevent aggregation [66], while DOPE increased liposome flexibility and enabled penetration through barriers [67], [68].

The liposomal formulations displayed consistent sizes and uniform polydispersity, with changes in size and charge post-drug-loading and labeling. This size consistency and narrow range of values are desirable, suggesting that the liposomal formulations are suitable for optimal delivery and treatment of otitis media following trans-tympanic administration [21].

Regarding drug encapsulation efficiency and loading capacity, our results compare favorably with previous studies, specifically those exploring liposome-based nanocarriers’ effectiveness for encapsulating antibiotics for targeted delivery [44], [69], [70]. A significant outcome was the demonstration of potent dose-dependent antibacterial activity against the NTHi strain 86-028NP by our CFX-Lipo. This performance aligns positively with that of free ceftriaxone, which mirrors similar findings in other research on liposomal antibiotic formulations [62]. Uniquely, empty liposomes also exhibited dose-dependent antibacterial properties due to the electrostatic interactions between the positive lipid charges and negative bacterial cell charges. These interactions disrupt bacterial cell integrity, contributing to distinct antimicrobial effects. This observation aligns with the outcomes of similar studies on the effects of electrostatic interactions on bacterial cell integrity [41], [71].

The TM presents a significant obstacle to delivering therapeutic agents into the ME due to its tri-layered structure and dense accumulation of keratinized dead cells. Despite this, confocal microscopy results have confirmed the successful penetration of our liposomal formulation across the TM. This is likely attributable to two key factors: the cationic characteristics of the liposomal elements interacting with the TM’s negatively charged surface and the relatively small size of our liposomes, measuring approximately 100 nm in diameter. Smaller nanoparticles, in this size range, have been demonstrated in numerous studies to have a higher propensity to cross biological barriers, including the tightly packed cells of the TM [21], [72], [73]. This combination of size and charge suggests a promising ability to transverse the TM successfully. However, it should be noted that human TMs are significantly thicker than those of chinchillas [22], which might limit drug penetration in humans, indicating a necessity for further investigation into these specific factors regarding trans-tympanic drug delivery in humans.

Our biodistribution study results underscore the exceptional performance of the liposomal-ceftriaxone delivery system in comparison to both systemic administration and local delivery of free ceftriaxone. This superiority is evident in the quantifiable concentration of ceftriaxone in both the ME and blood serum on the third- and tenth-days following application.

Notably, the CFX-Lipo demonstrated a considerable concentration of the drug in the middle ear, revealing a measurement of 56.1 ± 29.78 ng/ml by day 3, compared to 35.1 ± 9.45 ng/ml of free ceftriaxone. This disparity widened by day 10, as the CFX-Lipo group displayed a significant increase in ceftriaxone concentration (1596.0 ± 98.65 ng/ml) in comparison to not just the free ceftriaxone group (772.9 ± 49.40 ng/ml) but also to the systemic administration.

An analysis of the ceftriaxone biodistribution in terms of the percentage of the total detected dose showcased a substantially elevated percentage (3.240 ± 0.567% on day 3 and 8.548 ± 0.463% on day 10) for the CFX-Lipo group, in stark contrast to the free ceftriaxone group (0.002 ± 0.001% on day 3 and 0.013 ± 0.001% on day 10) and to the systemic administration rates.

Further, the CFX-Lipo group demonstrated substantially lower ceftriaxone concentrations in blood serum samples, 0.39 ± 0.16 ng/ml on day 3 and 48.2 ± 7.78 ng/ml on day 10, in comparison to systemic administration and the free ceftriaxone group. This notable reduction in systemic antibiotic concentrations is vital as it alludes to the decreased potential for adverse effects commonly associated with systemic antibiotic distribution, such as toxicity and antibiotic resistance [3], [13], [74], [75]. As such, these results highlight not only the superior drug delivery and biodistribution capabilities of our liposomal-ceftriaxone delivery system but also its potential safety attributes in reducing systemic antibiotic burden.

The minimal cochlear penetration observed suggests that our liposomal system primarily targets the middle ear. However, by altering the stability and composition of the liposomes, this system could potentially be tailored to target the inner ear. This could involve optimizing the absorption of drugs through the round window membrane (RWM), the most accessible entryway for inner ear drug delivery, as suggested in previous studies [76]. This, potentially, opens up substantial additional benefits of this methodology of drug delivery through an intact TM. Currently, there do not exist systems to treat sudden sensorineural hearing loss, balance disorders and other conditions impacting the inner ear. With optimization of the current techniques, and our knowledge that this liposomal system can carry other drug cargoes that would be optimal to treat these disorders, we anticipate further extension of this technology for the treatment of inner ear disorders. Experiments along these lines are currently being conducted in our laboratory.

Finally, histologic and ABR tests revealed no discernable structural or functional ototoxicity in the cochlea post-treatment. This highlights a significant benefit for the use of this system as we explore clinical and human applications.

4.1. Limitations, Strengths, and Future Directions

While our study has yielded promising results, we acknowledge several limitations. The use of chinchilla models, although informative, may not fully recapitulate human TM physiology. The absence of an actual OM condition in our models may affect the generalizability of our findings, as the inflamed state of the TM in OM could significantly alter study conditions such as TM thickness, permeability, and the inflammatory environment, potentially affecting the rate of drug penetration and distribution. Additionally, the presence of middle ear effusion in Otitis media can influence the pharmacokinetics of drugs within the middle ear cavity, potentially necessitating adjustments in dosage and delivery frequency. The inflammatory milieu of Otitis media could also affect the stability and interaction of liposomes with the TM and middle ear tissues. Furthermore, the thin-film method employed for liposome preparation, while effective, presents challenges for scaling up production, which is a necessary consideration for clinical application. The stability and drug release rate of liposomes also warrant further optimization to reduce the frequency of application and enhance patient compliance potentially.

On the other hand, our study’s strengths are notable. The noninvasive nature of the liposomal delivery system circumvents the need for invasive TM injections, reducing patient discomfort and the risk of complications. The ability to encapsulate multiple drugs within the liposomes opens avenues for combination therapies. The cationic liposomes demonstrate improved adhesion and permeation through the TM, which, along with the potential for controlled drug release, minimizes systemic exposure and the risk of antibiotic resistance.

Our findings lay the groundwork for future research, which should include the exploration of microfluidic methods for mass production of liposomes, and investigations into the long-term stability of liposomes for extended shelf life. Additionally, the versatility of the liposomal platform could be harnessed to load various therapeutic agents, potentially broadening the scope of treatment for a range of ear pathologies. Clinical trials will be paramount in validating the efficacy and safety of this system in humans, with particular attention to pharmacoeconomic considerations and the integrity of the TM upon repeated applications.

In light of these considerations, our study presents a significant step forward in noninvasive trans-tympanic drug delivery, with the potential to revolutionize the management of OM and other ear-related conditions.

5. Conclusion

In conclusion, our multi-part study has effectively and successfully demonstrated the development of a noninvasive trans-tympanic liposomal delivery system for administering ceftriaxone in the treatment of otitis media. The liposomal formulations were meticulously characterized, with similar sizes and uniform polydispersity evident, increasing the potential for optimal otitis media treatment via trans-tympanic administration. Despite the tympanic membrane presenting an obstacle for the delivery of treatment agents into the middle ear, our study shows a successful permeation of our liposomal formulation across the membrane. This confirms the potential for a novel, noninvasive and non-toxic method of treatment for otitis media, which can significantly benefit patient care and outcomes. This research offers a foundation for further investigation and development of noninvasive drug delivery systems for otitis media and potentially other middle- and inner-ear diseases.

7. Data Availability Statement:

The data supporting this study’s findings are available upon reasonable request from the corresponding author.