Perfluorocarbon Nanoemulsions for Simultaneous Delivery of Oxygen and Antioxidants During Machine Perfusion Supported Organ Preservation

Abstract

Background: Solid organ transplantation (SOT) is a life-saving treatment for patients with end-stage diseases and/or organ failure. However, access to healthy organs is often limited by challenges in organ preservation. Furthermore, upon transplantation, ischemia–reperfusion injury (IRI) can lead to increased organ rejection or graft failures. The work presented aims to address both challenges using an innovative nanomedicine platform for simultaneous drug and oxygen delivery. In recent studies, resveratrol (RSV), a natural antioxidant, anti-inflammatory, and reactive oxygen species (ROS) scavenging agent, has been reported to protect against IRI by inhibiting ferroptosis. Here, we report the design, development, and scalable manufacturing of the first-in-class dual-function perfluorocarbon-nanoemulsion (PFC-NE) perfusate for simultaneous oxygen and antioxidant delivery, equipped with a near-infrared fluorescence (NIRF) reporter, longitudinal, non-invasive NIRF imaging of perfusate flow through organs/tissues during machine perfusion. Methods: A Quality-by-Design (QbD)-guided optimization was used to formulate a triphasic PFC-NE with 30% w/v perfluorooctyl bromide (PFOB). Drug-free perfluorocarbon nanoemulsions (DF-NEs) and RSV-loaded nanoemulsions (RSV-NEs) were produced at 250–1000 mL scales using M110S, LM20, and M110P microfluidizers. Colloidal attributes, fluorescence stability, drug loading, and RSV release were evaluated using DLS, NIRF imaging, and HPLC, respectively. PFC-NE oxygen loading and release kinetics were evaluated during perfusion through the BMI OrganBank^® machine with the MEDOS HILITE^® oxygenator and by controlled flow of oxygen. The in vitro antioxidant activity of RSV-NE was measured using the oxygen radical scavenging antioxidant capacity (ORAC) assay. The cytotoxicity and ferroptosis inhibition of RSV-NE were evaluated in RAW 264.7 macrophages. Results: PFC-NE batches maintained a consistent droplet size (90–110 nm) and low polydispersity index (<0.3) across all scales, with high reproducibility and >80% PFOB loading. Both DF-NE and RSV-NE maintained colloidal and fluorescence stability under centrifugation, serum exposure at body temperature, filtration, 3-month storage, and oxygenation. Furthermore, RSV-NE showed high drug loading and sustained release (63.37 ± 2.48% at day 5) compared with the rapid release observed in free RSV solution. In perfusion studies, the oxygenation capacity of PFC-NE consistently exceeded that of University of Wisconsin (UW) solution and demonstrated stable, linear gas responsiveness across flow rates and FiO₂ (fraction of inspired oxygen) inputs. RSV-NE displayed strong antioxidant activity and concentration-dependent inhibition of free radicals. RSV-NE maintained higher cell viability and prevented RAS-selective lethal compound 3 (RSL3)-induced ferroptosis in murine macrophages (macrophage cell line RAW 264.7), compared to the free RSV solution. Morphological and functional protection against RSL3-induced ferroptosis was confirmed microscopically. Conclusions: This study establishes a robust and scalable PFC-NE platform integrating antioxidant and oxygen delivery, along with NIRF-based non-invasive live monitoring of organ perfusion during machine-supported preservation. These combined features position PFC-NE as a promising next-generation acellular perfusate for preventing IRI and improving graft viability during ex vivo machine perfusion.

1. Introduction

According to the Global Observatory on Donation and Transplantation, nearly 172,409 solid organs were transplanted worldwide in 2023, a 9.5% increase from 2022 (data produced by the WHO–ONT collaboration). Additionally, there were around 399,802 patients on the waiting list, and 13,639 died while waiting for transplantation [1]. These numbers confirm an increasing demand for SOT, but a shortage of organs to meet this demand. The organs from deceased donors (DBD: donation after brain death; DCD: donation after circulatory death) have been recently included to expand the pool of donated organs. However, the Scientific Registry of Transplant Recipients (SRTR) reported 10,085 organs discarded in 2023 (organs procured but not transplanted) due to poor quality and function [2]. There is a strong consensus that ischemia–reperfusion injury (IRI) is the major contributor to poor organ quality and dysfunction [3], and the current preservation methods are ineffective in preventing such a pathological condition [4,5]. IRI occurs due to initial organ ischemia after procurement, followed by restoration of blood supply and reoxygenation during transplantation. IRI is accountable for morbidity and mortality in multiple health conditions, such as myocardial infarction, ischemic stroke, and following organ transplantation [6]. Hypoxia-induced mitochondrial dysfunction, excessive reactive oxygen species, and ferroptotic lipid peroxidation drive IRI pathology in transplanted organs [7,8,9].

Currently, solid organs are often preserved using either conventional static cold storage (SCS) or ex vivo machine perfusion (MP) [10,11]. Ex vivo MP is a technique in which the organ is continuously perfused with preservation solution at normothermic (37 °C), sub-normothermic (25 °C), or hypothermic (4 °C) temperatures. Ex vivo MP has greater potential than SCS to improve organ preservation and reduce the number of discarded organs, as it mimics normal blood flow, provides oxygen and nutrients, removes metabolic waste, and may prolong preservation time [12,13,14]. The preservation solution or perfusate can be supplemented with oxygen, nutrients, or drugs to protect against IRI during transplantation. Oxygen is a key metabolic precursor for the formation of adenosine triphosphate (ATP). A lack of oxygen supply during ischemia reduces ATP production, leading to cell death. The standard method for supplying oxygen involves cellular perfusates (whole blood or blood substitutes). However, the cellular perfusates pose significant risks and limitations, including immune modulations, acute transfusion reactions, volume overload, and hemolytic reactions. Furthermore, the oxygen dissociation from hemoglobin is dependent on pH and temperature [15,16].

Conventional perfusates, including the University of Wisconsin (UW) solution, lack an oxygen-carrying component. Although hemoglobin-based oxygen carriers (HBOCs) offer reliable oxygen carriage, they can lead to immunological challenges, vasoconstriction, and oxidation upon storage [17,18]. In contrast, perfluorocarbon nanoemulsions (PFC-NEs) offer stable, temperature- and pH-independent oxygen loading and offloading and are devoid of immunological liabilities compared to standard perfusates because they are chemically and biologically inert [19,20,21].

The UW solution did not show any clinical improvement in oxygenated hypothermic MP of kidneys compared to SCS, probably due to its poor oxygen-carrying capacity [22,23,24]. On the other hand, HBOCs have reported serious adverse effects, such as neurotoxicity, nephrotoxicity, antigenicity, oxidation of hemoglobin to inactive methemoglobin, gastrointestinal distress, and neurotoxicity, which have limited their FDA approval as blood substitutes [25]. Perfluorocarbons (PFCs) have been reported as excellent oxygen carriers [26,27,28,29]. PFCs consist of the strongest C-F covalent bond, where the highly electronegative fluorine atom forms a dense shell around the carbon skeleton, making it metabolically and chemically inert. Additionally, PFCs have low intermolecular forces and thus high vapor pressure, which influences clearance from the body. PFCs are also reported to be biologically inert [30], mainly due to their chemical inertness, as well as their hydrophobic and lipophobic character [16,28]. PFCs demonstrate hydrophobicity due to their low polarizability and high interfacial tension with water. Conversely, lipophobicity comes from the extreme polarity of the C-F bonds [27,31]. They have a high capacity to dissolve gases (oxygen, carbon dioxide) because of the low intermolecular forces. The dissolution of oxygen in PFCs follows Henry’s law, in which the gas solubility is directly proportional to the partial pressure of the gas. Unlike HBOCs, the dissociation of oxygen from PFC is not affected by temperature and pH, making them safer alternatives [31,32].

The pathophysiology of IRI is multifaceted, with reactive oxygen species (ROS) as the main cause of oxidative cell death following reperfusion. IRI progresses through two phases: ischemia, characterized by reduced blood and oxygen supply, anaerobic metabolism, and mitochondrial dysfunction; and reperfusion, marked by the overproduction of ROS and lipid peroxidation. This causes sterile inflammation, associated with signaling events from Toll-like receptors, as well as the recruitment and activation of neutrophils and macrophages, thereby exacerbating tissue damage. Because the mechanisms of IRI remain unclear, effective therapies are limited [33]. Regulated cell death (RCD), including apoptosis, necroptosis, autophagy, and ferroptosis, has been explored as a therapeutic target. Among these pathways, ferroptosis, an iron-dependent RCD, has emerged as particularly relevant, as its molecular mechanism closely mirrors the pathological hallmarks of IRI. Post-transplantation, the reperfusion event leads to increased intracellular iron concentration and mitochondrial dysfunction, promoting iron-catalyzed Fenton reaction, which triggers excess ROS and phospholipid peroxidation, ultimately leading to cell death. Concurrently, downregulation of antioxidant enzyme glutathione peroxidase 4 (GPX4) and upregulation of acyl-CoA synthetase long-chain family member 4 (ACSL4), which generates peroxidation-prone polyunsaturated fatty acids, together promote ferroptosis. Characteristic mitochondrial shrinkage, cristae loss, and increased membrane density observed in IRI further support ferroptosis involvement in IRI [34,35]. Ferroptosis-targeted interventions, such as iron chelators, peroxidation inhibitors, and antioxidants (e.g., Ferrostatin-1, Deferoxamine, resveratrol, quercetin), have shown protective effects in myocardial, intestinal, and cerebral IRI models [7,8,9]. Particularly, antioxidants mitigate ferroptosis by scavenging ROS, reducing lipid peroxidation, and upregulating antioxidant enzymes (Catalase, Superoxide dismutase, GPX4) [36,37]. Natural compounds like polyphenols and flavonoids have shown significant efficacy against oxidative stress disorders [38]. Collectively, these findings establish ferroptosis as a mechanistic link between oxidative stress and tissue injury in IRI and support ferroptosis inhibition as a rational strategy for preserving organ viability. Resveratrol (RSV) is a natural polyphenolic compound predominantly found in grapes, mulberries, peanuts, and other vegetables. It was first isolated from the plant, Polygonum cuspidatum, used in traditional Chinese and Japanese medicine. RSV exhibits antioxidant, anti-inflammatory, anti-tumor, and cardioprotective effects, and recent studies have shown its protective effects against IRI by inhibiting ferroptosis. Ting Li et al. demonstrated that RSV reduced Fe²⁺ levels, oxidative stress, and myocardial IRI by suppressing ferroptosis [39]. In another study, Xuqi Li et al. reported that RSV activated Sirt3, upregulating antioxidant enzymes (SOD, CAT, GPX4), and decreasing ROS, lipid peroxidation, and iron in intestinal I/R models [40]. However, RSV’s clinical application is limited by poor pharmacokinetics—its lipophilic, poorly soluble structure (Biopharmaceutical Classification System Class II) leads to rapid metabolism, low bioavailability, and short half-life. Moreover, RSV in solution degrades under UV light, converting from its active trans (E) form to inactive cis (Z) form [41].

Several nano-encapsulation strategies have been implemented to improve resveratrol’s pharmacokinetic properties [42]. Among them, nanoemulsions are highly effective for delivering via various routes, such as oral, intravenous, and topical [43,44,45,46,47]. Nanoemulsions are kinetically stable oil and water dispersions, stabilized by surfactants with a mean droplet diameter of <500 nm [21]. They have high encapsulation efficiency for RSV due to the lipophilic nature. The high surface area of the nanoemulsion droplet provides better absorption and improved bioavailability. Perfluorocarbon nanoemulsions are uniquely suited for dual oxygen–drug delivery because the PFC core provides superior oxygen solubility with a high extraction ratio (90%) [15], while the surrounding hydrocarbon shell offers stable encapsulation and sustained release of RSV. Additionally, the chemical stability of RSV is improved by preserving its active trans (Z)-isomer. Previously, we adopted a Quality-by-Design (QbD) methodology to develop a triphasic PFC-NE for oxygen delivery [48]. However, earlier formulations showed limited shelf-life stability and suboptimal PFC concentrations (up to 15% w/v). In this work, we further optimized the composition of the PFC-based artificial oxygen carrier (AOC) to improve long-term stability and increase oxygen-carrying potential by increasing the PFC content to 30% w/v, and validated the manufacturing process for scalability, reproducibility, and robustness.

The efficacy of oxygenation during machine perfusion and drug delivery for the prevention of IRI is directly related to the distribution of the perfusate through the blood vessels and capillaries of the machine perfused tissue and/or organ. To date, to the best of our knowledge, no perfusates allow non-invasive monitoring of perfusion during machine-supported organ preservation. Therefore, a clinical-grade near-infrared fluorescent (NIRF) dye, indocyanine green, was also incorporated into the hydrocarbon shell of the PFC-NE triphasic nanoemulsion droplet using our previously reported methods [21,49]. The NIRF dye in the perfusate PFC-NEs serves two purposes: (1) non-invasive live monitoring of perfusate distribution in the organ; (2) assessing PFC-NE stability under machine perfusion-supported organ preservation. Therefore, we extensively evaluated the presented PFC-NEs for long-term colloidal and fluorescence stability, sustained RSV release, oxygen-carrying capacity, radical-scavenging ability, in vitro cytotoxicity, and ferroptosis inhibition in cultured cells. We also report stability under machine perfusion conditions, confirming that perfusate PFC-NEs remain stable and functional throughout the process of machine perfusion-supported organ preservation.

2. Materials and Methods

2.1. Materials

Chemicals were purchased from commercially available sources and were used without modifications. The surfactants (Pluronic^® P105 and Cremophor^® EL (polyethoxylated castor oil)) were purchased from US Biological Life Sciences (Salem, MA, USA) and Spectrum (New Brunswick, NJ, USA). Miglyol^® 812N was purchased from CREMER Oleo Product Division (Hamburg, Germany), Perfluoro-15-crown-ether and Perfluorooctyl bromide were purchased from Exfluor Corporation (RoundRock, TX, USA), Transcutol^® (2-(2-ethoxyethoxy)-ethanol) was purchased from Spectrum (New Brunswick, NJ, USA), and indocyanine green (ICG) dye was purchased from Sigma-Aldrich (St. Louis, MO, USA). Resveratrol was purchased from TCI (Portland, OR, USA), Cell Titer-Glo 2.0^® Assay Kit was purchased from Promega Corporation (Madison, WI, USA), and OxiSelect™ Assay Kit (STA-345) was purchased from Cell Biolabs Inc. (San Diego, CA, USA). Three microfluidizers: Microfluidizer M110S, Microfluidizer LM20, and Microfluidizer M110P from Microfluidics Corporation (Westwood, MA, USA) were used for nanoemulsion preparation depending on manufacturing scale.

2.2. Methods

2.2.1. Perfluorocarbon Nanoemulsion Perfusate Formulation and Manufacturing

Nanoemulsions were manufactured following previously reported procedures [48,49,50,51]. The PFC concentration (30% w/v) was selected to fall within the range reported for commercial PFC-based oxygen carriers (15–90% w/v) to maximize oxygen-carrying capacity. Non-ionic surfactants (Pluronic P105 and Cremophor EL) were selected to match the required hydrophilic–lipophilic balance (HLB) of the hydrocarbon oil, Miglyol 812N, thereby improving shelf-life stability. The internal phase (Miglyol, transcutol, PFOB) to external phase (surfactant-blended micelle) ratio was 1.84:1 w/w, consistent with our previously optimized formulations following the Quality-by-Design methodology [48]. Briefly, a micelle solution of blended non-ionic surfactants (2% w/v Pluronic P-105, 3% w/v Cremophor EL) was formulated following the previously reported protocol [52]. For RSV and/or NIRF dye loading, RSV was dissolved in transcutol 24 h in advance, and ICG was pre-mixed with stearylamine before being added to the hydrocarbon oil, PFC and micelle for magnetic stirring and coarse emulsification with hand mixing. Coarse emulsions were then charged into the Microfluidizer (Microfluidics Corp., Westwood, MA, USA) and processed at a specified pressure on an iced interaction chamber for a specified number of passes. Samples were collected after discrete passes for monitoring size reduction during processing by dynamic light scattering (DLS). The final product was collected into a sterile polypropylene bottle and stored at 4 °C. The sequential workflow of the manufacturing process and comprehensive characterization of PFC-NE are shown in Figure 1.

2.2.2. Colloidal and Fluorescence Assessments of Nanoemulsions

Colloidal properties of nanoemulsion samples were determined by DLS on a Zetasizer Nano (Malvern Instruments, Worcestershire, UK), which measures the hydrodynamic diameter (nm) and polydispersity index (PDI). The nanoemulsions were diluted 1:80 v/v in deionized water and mixed well in disposable cuvettes before triplicate measurements. The operating parameters of DLS were as follows: refractive indices of material and dispersant were 1.59 and 1.33, respectively; count rate between 250 and 280 kps; temperature, 25 °C; and 173° back scatter angle. For NIRF fluorescence measurements, nanoemulsion samples were serially diluted with deionized water from 1:10 v/v to 1:160 v/v. The diluted samples were then transferred to a clear 96-well plate in triplicate, and the plates were imaged on a Li-COR Odyssey for fluorescence scanning at 800 nm, with a focus of 3.75. Collected images were processed using ImageStudio 5.0 Software to quantify fluorescence in each well. Each measurement was performed in triplicate.

2.2.3. Colloidal Stability Under Stress Conditions

Nanoemulsion samples were centrifuged at 1100 or 3000 rpm for 5 min (Eppendorf 5804R, Framingham, MA, USA). For the filtration test, the nanoemulsions were filtered through a 0.45 µm pore size mixed cellulose ester syringe filter (Merck Millipore Ltd., Burlington, MA, USA). The particle size and polydispersity index of the centrifuged and filtered samples were measured and compared with the control samples. All samples were diluted 1:80 v/v in deionized water before characterization. For serum stability, nanoemulsion samples were diluted 1:80 v/v in water, Dulbecco’s Modified Eagle’s Medium (DMEM), or 20% Fetal Bovine Serum (FBS) in DMEM and stored at 37 °C for 72 h. The particle size and size distribution of all samples were measured using DLS, before and after 72 h incubation. For accelerated stability studies, 500 µL of undiluted nanoemulsion samples were added to a 1.5 mL Eppendorf tube covered in aluminum foil and stored at 37 °C. The particle size, near-infrared fluorescence, and drug loading of the nanoemulsion samples were measured before and after incubation for 24 and 72 h.

2.2.4. Resveratrol Quantification and In Vitro Release Study

RSV loading in nanoemulsions was quantified using an optimized reverse-phase HPLC method on Dionex Ultimate 3000. The method involved an isocratic elution of RSV using methanol–water (75:25) mobile phase in a C18 Hypersil Gold 150 × 4.6 mm column (Thermo Scientific, Waltham, MA, USA). The retention time of RSV was 4.3 min when the flow rate was set to 0.5 mL/min, column temperature of 30 °C, and UV detection at 305 nm. For RSV quantification, the nanoemulsion samples were first diluted with methanol and agitated using a vortex mixer to break the nanoemulsion. Then, it was diluted with deionized water to match the mobile phase ratio. The methanol–water mixture was mixed again using a vortex mixer and centrifuged for ~30 s to allow PFOB to settle at the bottom. Finally, 1 mL of the supernatant was transferred to an HPLC vial for analysis. A standard curve was constructed based on the area under the curve (AUC) of the RSV reference standards, ranging from 2.5 to 50 μg/mL. The final concentration of RSV in the nanoemulsion sample was in the linear range of detection in UV-Vis. All analyses were performed in triplicate.

The in vitro RSV release study was performed using the dialysis bag technique with molecular cutoff 3000 Dalton (Thermo Fisher Scientific, Waltham, MA, USA snakeskin dialysis tubing REF 6803), as reported in a previous procedure from our lab [49,52]. Briefly, 1 mL of RSV-NE and an equivalent amount of drug-containing Free-RSV solution were injected into the dialysis bag (n = 3). The bags were then placed inside a 50 mL Falcon tube containing 15 mL of release media (1% w/v Tween 80 in phosphate-buffered saline; pH 7.4). The Falcon tubes containing dialysis bags were kept on a shaking incubator at 50 rpm and 37 ± 0.5 °C. To maintain sink conditions, the dialysis bag was transferred into a new Falcon tube containing fresh release medium at regular intervals (1, 2, 3, 4, 6, 8, 10, 24, 34, 48, 56, 72, and 120 h). For RSV quantification, 1 mL of an aliquot was collected at each time point and measured using HPLC. The cumulative percentage of drug release and the cumulative amount of drug release were plotted against time to compare the drug release from RSV-NE and Free-RSV solution.

2.2.5. Nanoemulsion Stability Under Oxygenation Conditions

Oxygenation of the perfluorocarbon-based nanoemulsion was performed using both a machine perfusion setup and the bubbling method. For the machine perfusion setup, a MEDOS HILITE^® 7000 Oxygenator (MEDOS Medizintechnik AG, Stolberg, Germany) equipped with a microporous hollow fiber membrane was used. Firstly, the oxygenator was primed with 1.68× phosphate-buffered saline (PBS), and then the PFC-NE was perfused using a pump with a flow rate of 10 mL/min at 37 °C. A mixture of oxygen gas and room air (55% FiO₂) was supplied to the oxygenator using an oximeter at a total flow rate of 3.2 L/min. In the oxygenator, the oxygen gas diffuses across the hollow fiber membrane and is carried by DF-NE. The perfused oxygenated DF-NE is collected in a beaker and periodically pumped back to the oxygenator. The partial pressure of oxygen (pO₂) in the perfused DF-NE was measured using the PRESENS optical oxygen sensor. A baseline pO₂ measurement was performed for 15 min without oxygen supply, and then the pO₂ level was monitored for 6 h. Sampling was performed at 0, 1, 2, 3, 4, 5, and 6 h for DLS measurements and NIRF imaging.

In case of oxygen bubbling, the bulk RSV-NE was aliquoted into 3 × 300 mL volumes for triplicate measurements. The oxygen gas (100% FiO₂) was bubbled into the RSV-NE sample through a glass Pasteur pipette for about 3 h with continuous measurement of pO₂ level using the PRESENS oxygen probe. Sampling of RSV-NE was performed before and after oxygen bubbling for DLS measurements, NIRF imaging, and drug loading.

2.2.6. Oxygen Saturation and Off-Loading in an Ex Vivo Machine Perfusion Circuit

BMI OrganBank^® develops ex vivo machine perfusion devices for multiple tissue types and applications. One of BMI OrganBank’s systems was utilized in this study to determine the compatibility of the PFC-NEs within a perfusion circuit. Their system combines a proprietary pulsatile perfusion circuit with advanced flow, pressure, and temperature control settings that simulate the cardiac waveform to maintain physiological mean arterial pressures (MAPs) in the vasculature of tissue. The pulse-wave flow of the system facilitates uniform distribution of the PFC-NEs and prevents damage to the capillary beds while eliminating edema even with prolonged perfusion. The system also contains a sensor suite to monitor perfusion metrics. Within the sensor suite, a Pre-Sens^© flow-through oxygen cell was integrated to monitor oxygen saturation of the perfusates under different conditions. The flow-through cell was placed in-line in the perfusion circuit after the oxygenator. First, we tested the oxygen loading and release kinetics of the PFC-NEs during machine perfusion conditions. Approximately 250 mL of PFC-NEs were added to the perfusion circuit, and the circuit was primed and de-bubbled. The system was run briefly to check for leaks, and carbogen supply lines were connected to the oxygenator. The PreSens^© software (v4.0.0.2353) was then set up for data collection in the flow-through O₂ cell, and the sensors were calibrated before use. The PFC-NEs were tested under varying flow conditions as well as varying gas input conditions. For all perfusion experiments with the PFC-NEs, carbogen gas was used, which has a composition of 95% O₂ and 5% CO₂. The performance of the PFC-NEs was compared against Belzer UW^® Solution as a control, which is an FDA-approved organ preservation solution typically used for the flushing and cold storage of organs.

To determine the effect of gas input on oxygen saturation of the perfusates, the PFC-NEs and UW Solution were perfused at varying gas inputs. While holding the flow rate constant at 30 mL/min, the gas supply was set at 100% before decreasing to 80%, 60%, 40%, 20%, and 0% and n = 3 samples were collected at each gas input condition for subsequent stability analysis. For the flow-rate study, gas input was held at 100% carbogen with a delivery rate of 0.5 L/min. The PFC-NEs and UW solution were perfused independently at 10 mL/min, 20 mL/min, and 30 mL/min through the circuit, and oxygen saturation was measured throughout the duration of the study. We collected three samples at each flow condition for subsequent stability analysis of the PFC-NEs.

2.2.7. In-Line Perfusion NIRF Imaging of Nanoemulsion and UW Solution

NIRF imaging was performed to monitor the in-line perfusion of PFC-NE and UW solution through a hollow fiber membrane oxygenator using the Fluobeam LX Imager (FLUOPTICS, Grenoble, France). The UW solution was used as a control. Images were analyzed using Image Studio (V.5.2) to show the fluorescence signal displayed using the “heat map” feature.

2.2.8. Oxygen Radical Antioxidant Capacity (ORAC) of Nanoemulsions

The Oxygen Radical Antioxidant Capacity (ORAC) of nanoemulsions were determined using the OxiSelect™ Assay Kit (STA-345, Cell Biolabs Inc., San Diego, CA, USA), as per the manufacturer’s instructions [53]. The antioxidants in the samples can block the peroxyl radical (AAPH)-mediated oxidation of fluorescein over 60 min. The final assay values are expressed as the relative fluorescence units (RFU), which is the ratio of Ex480 nm/Em520 nm. The area under each curve of the RFU without or with resveratrol or Trolox additions for 60 min was calculated as the AUC. The Trolox (5, 10, 20, 30, 40, 50, and 100 μM) were used to plot a standard curve between AUC and Trolox concentrations. The ORAC activity of DF-NE, RSV-NE, and Free-RSV was calculated and expressed as μM Trolox equivalents (μM TE).

2.2.9. Macrophage Cell Viability and Ferroptosis Inhibition Assay

Cell viability assay was performed using the RAW 264.7 macrophage cell line (ATCC TIB 71) to assess the cytotoxicity of DF-NE and RSV-NE. A previously reported protocol was followed for this study [49,52]. Briefly, cells were seeded at 10,000 cells per well in 96-well plates and incubated at 37 °C with 5% CO₂ for 24 h, followed by treatment with DF-NE and RSV-NE diluted in cell culture medium from 0 to 80 μL/mL (n = 6) for 24 h. Free RSV in DMSO containing equivalent drug concentration of RSV-NE was used as a free drug control, and only DMSO as a vehicle control. The cell viability was measured using the Luminescent Cell Titer-Glo 2.0^® assay kit (Promega, Madison, WI, USA). After 24 h of treatment, the cell-culture media was removed and replaced with 100 µL of fresh cell culture media, followed by the addition of 40 µL of Cell Titer-Glo reagent. The 96-well plate was then placed on a plate shaker at 50 rpm for 15 min. The cell lysates were transferred from the culture plate to an opaque white 96-well plate for luminescence measurement in a BioTek Synergy HTX Multi-Mode plate reader (Agilent, Santa Clara, CA, USA).

For the ferroptosis inhibition assay, RAW 264.7 macrophages were seeded in 96-well plates at 10,000 cells/well in cell culture media and incubated for 24 h at 37 °C, 5% CO₂, and 95% humidity. The next day, cells were exposed to RSV-NE, volume-matched DF-NE, drug concentration-matched FD-RSV, and Ferrostatin-1 with five doses, respectively. RSV-NE and FD-RSV were diluted in media and added to cells in 6 replicates to achieve a final RSV concentration range of 2.5–40 μM. DF-NE was volume-matched to each of the tested concentrations. Ferrostatin-1 was used as a positive control, diluted in media and added to 6 replicates to achieve a final concentration range of 2.5–40 μM. For each plate, we left one column (n = 6) as a no-treatment control, and another column (n = 6) as a ferroptosis induction with RSL3, methyl (1S,3R)-2-(2-chloroacetyl)-1-(4-methoxycarbonylphenyl)-1,3,4,4a,9,9a-hexahydropyrido [3,4-b]indole-3-carboxylate. After 24 h of treatment, RSL3 (2 μM) was added diluted in the complete medium at 50 μL/well treatment volume to induce ferroptosis, and the plates were incubated for an additional 6 h. Following incubation, cell culture media (100 µL) were removed from each well and replaced with fresh media (100 µL/well), and Cell Titer-Glo^® (40 µL/well) was added, and placed in a shaking incubator for 5 min. The luminescence mixture was then transferred to solid flat-bottom white plates, 90 µL/well, and read on a BioTek Synergy HTX multi-mode luminescence plate reader. Before the Cell Titer Glo assay, microscopic images of the cells were captured using an EVOS microscope at 20× magnification.

2.2.10. Statistical Analyses

The data were analyzed using GraphPad Prism v10. The difference between the compared groups for analysis of droplet size, %RSV loading, and %PFOB loading was assessed using either a two-tailed unpaired t-test or a one-way analysis of variance (ANOVA). For the drug release study, cytotoxicity, and ferroptosis inhibition, a two-way repeated measures analysis of variance (ANOVA) was performed. The statistical test details are also described in the legends of each figure, with a significance level of p < 0.05.

3. Results

3.1. Nanoemulsions Composition Optimization for Increased Oxygen Loading and Stability

In our previous reports, we have adopted a Quality-by-Design (QbD) methodology to develop triphasic PFC-NE for oxygen delivery [48]. In this study, the multilinear regression analysis established four composition and one processing variables (volume fraction, HC/PFC ratio, solubilizer fraction, and number of passes) that had a significant impact on colloidal stability and in vitro oxygen release. However, the overall PFC concentration was only as high as 14.9% w/v, so the present objective was to increase the PFC concentration beyond 15% w/v, which is in the range of the reported PFC concentrations of commercially developed PFC oxygen carriers (15–90% w/v) [28]. In present work, we adjusted the overall volume fraction. We maintained the hydrocarbon oil phase volume at approximately the same level as in the previous formulation, resulting in a 30% w/v PFC-NE. Moreover, the hydrophilic lipophilic balance (HLB) value of the surfactant blend (2% w/v Pluronic P105 and 3% w/v Pluronic P123) in the previous NEs was 10.8, which was lower than the required HLB (13) of Miglyol, resulting in phase separation after 2–3 months. So, we replaced the Pluronic P123 (HLB = 8) with Cremophor EL (HLB = 13), which raised the overall HLB by three units, matching the required HLB of Miglyol^® 812N. To identify the optimal combination of hydrocarbon oil and PFC, Miglyol^® 812N, a blend of saturated medium chain triglycerides (MCTs), or olive oil, composed of a mixture of saturated and unsaturated long-chain triglycerides, at 6% w/v, was paired with perfluoro-15-crown-ether (PCE), a cyclic molecule, or perfluorooctyl bromide (PFOB), a linear molecule, at 30% w/v. The PFOB and Miglyol^® 812N combination was selected as the internal phase, as the resulting NEs exhibited a droplet diameter of approximately 100 nm, which sustained over 2 months (Figure 2A).

Once the key composition was established, we proceeded to incorporate a near-infrared fluorescent (NIRF) dye and/or a lipophilic small molecule drug (resveratrol), and adjusted the external phase to 1.68× phosphate-buffered saline (PBS; pH 7.4) from 1× PBS to improve osmotic and oncotic properties of the perfusate. Versatile fluorescence imaging capacity of the PFC-NE was demonstrated by successfully incorporating multiple fluorescent dyes (See

Table 1. Composition and process variables of all NE batches.

Batch No.	RSV (mg/mL)	Dye	Dispersant	Batch Size (mL)	Processor	N Passes	Pressure (psi)	Size (nm)	PDI	Mean ± SD (nm)	Coefficient of Variation (%)
1	-	DiI	DI water	25	M-110S	6	18,640	99	0.2	102 ± 4.0	3.7
2	1.23	DiI and DiR						97	0.239
3	-	DiI and DiR						105	0.237
4	0.685	DiI and DiR						105	0.193
5	-	-						104	0.231
6	0.685	-	1× PBS	125	LM20	5	16,500	93	0.225	96 ± 2.5	2.6
7	-	-						96	0.228
8	0.685	-						98	0.223
9	-	ICG	1.68× PBS	250	LM20	4	20,000	83	0.179	80 ± 2.1	2.6
10	0.69							80	0.175
11	-	ICG	1.68× PBS	600	LM20	5	18,500	79	0.182	94 ± 3.5	3.7
12	-							97	0.222
13	-							81	0.216
14	-	ICG	1.68× PBS	600	M-110P	6	18,500	94	0.131	90 ± 8.4	9.4
15	-							80	0.208
16	-							98	0.123
17	-	ICG	1.68× PBS	1000	M-110P	4	18,500	90	0.217	90 ± 2.0	2.2
18	0.69							88	0.189
19	-							92	0.214

): 1,1′-Dioctadecyl-3,3,3′,3′-Tetramethylindocarbocyanine Perchlorate [‘DiI’; DiIC18(3)], 1,1′-Dioctadecyl-3,3,3′,3′-Tetramethylindotricarbocyanine Iodide, [‘DiR’; DiIC18(7)), and sodium 4-[(2Z)-2-[(2E,4E,6E)-7-[1,1-dimethyl-3-(4-sulfonatobutyl)benzo[e]indol-3-ium-2-yl]hepta-2,4,6-trienylidene]-1,1 dimethylbenzo[e]indol-3-yl]butane-1-sulfonate (Indocyanine Green, ICG) dye by adapting earlier reported methods [49,50]. Indocyanine green (ICG) is a clinical-grade near-infrared dye used as a diagnostic imaging contrast agent in surgical settings [54,55].

Lipophilic dyes (DiI and DiR) were incorporated into the hydrocarbon shell of the triphasic PFC-NE by pre-dissolving them in a co-solvent. As well-established lipid tracers, these dyes remain stably confined within the nanoemulsion droplets, enabling robust incorporation without the need for additional modification [56,57,58,59,60,61]. However, the ICG dye is hydrophilic and requires to be ion-paired with a cationic lipid, stearylamine (SA), to increase the lipophilicity before addition to the hydrocarbon phase [21,49]. During machine perfusion of organs in pre-clinical research, introducing fluorescent dyes at varied wavelengths supports extensive post-perfusion tissue analyses, including microscopy and flow cytometry, which allows for in-depth assessments of how the PFC-NEs interact with preserved tissues following 24 h or longer machine perfusion. Introducing a clinical NIRF dye, ICG, however, provides a unique approach to directly image the PFC-NE perfusion through the organs and tissues. This is a non-invasive technique that uses optimized clinical-grade near-infrared fluorescence imagers for tracking ICG in living tissues/organs during surgery [62,63]. Consequently, the ability of NIRF live imaging of PFC-NE perfusion can further accelerate clinical translation as it allows for in-line quality control of the perfusate itself and machine perfusion organ preservation by adapting imaging procedures already established in clinical settings.

Finally, we optimized the aqueous phase of the PFC-NEs to better control the osmotic and metabolic balance of the PFC-NEs as perfusates. In the presented examples, we chose PBS due to its physiological pH, high buffering capacity, and an osmolality that closely matches plasma. The PBS concentration increased to 1.68× to adjust the dilution factor from introduced PFCs, oils, and surfactants and balance the osmotic and oncotic properties of the PFC-NE perfusate. Key processing and composition variables and quality control attributes of the PFC-NEs produced are summarized in Table 1.

Table 1 summarizes multiple important findings as follows: (1) batch size changes and manufacturing equipment had no major influence on final droplet size; (2) presence of dye and/or drug also had no impact on droplet size; (3) changes in aqueous phase composition, including an increase in the presence of buffering salts, had no major impact on droplet size. These results strongly suggest that the presented PFC-NE triphasic nanoemulsion perfusate is a robust oxygen and drug delivery platform that is scalable and can be easily adapted to the specific needs of machine perfusion-supported organ preservation settings.

3.2. Comparison of Production Scale and High Shear Processing Equipment

All batches in the previous study were manufactured on a small scale, i.e., 25 mL (NEs 1–5), which was convenient for preliminary optimization. Here, we have scaled up the formulation systematically using different Microfluidizer processors. These microfluidizers have the same interaction chamber and processing mechanism but differ in their processing capacity. Figure 2C,D illustrate a comparison of batch sizes based on colloidal attributes and %PFOB loading. Before comparing all the batches, we established that the drug (RSV) or NIRF dye has a negligible impact on the particle size of NEs, as shown in Figure 2B. The 25 mL (1×) batches produced in the M110S processor demonstrated an average particle size of 102 nm and PDI of 0.2, which was within specification. We then scaled the formulation 5× (125 mL; NEs 6–8) and 24× (600 mL; NEs 11–13) using a larger-volume capacity LM20 processor. The particle sizes of 125 mL and 600 mL batches were in good agreement with 25 mL batches (Figure 2D). In the product lifecycle, the manufacturing process should be robust and transferable to a new manufacturing unit. Hence, we introduced a second large benchtop microfluidizer (M-110P) to transfer the 600 mL scale NEs process. The 600 mL scale batches produced in LM20 (NEs 11–13) and M-110P (NEs 14–16) were uniform and homogenous with comparable colloidal and fluorescence attributes, and no difference in %PFOB loading, confirming successful process transfer (Figure 2C,E). The size of discrete pass samples was measured during processing, shown in Figure S1 for a representative batch from each production scale. Reproducibility is a key concern in pharmaceutical manufacturing processes. Figure 2F–I depict the size distributions of all 25, 125, and 600 mL-scale batches. In this layout, batches are grouped by batch size and equipment. Similarly, Table 1 presents reproducibility statistics comparing the particle sizes of replicate batches produced at four different scales and across three different Microfluidizer processors. The 25 mL batches produced on the M-110S exhibited a coefficient of variation of 3.7%. Batches made with the LM20 had a coefficient of variation of 2.6% (125 mL) and 3.7% (600 mL). The 24×-scale batches produced in M-110P had a coefficient of variation of 9.4%. Moreover, all batch sizes exhibited higher and comparable %PFOB loading, ranging from 80% to 95% (Figure 2C).

3.3. Evaluation of Drug-Free and Resveratrol-Loaded PFC-NEs for Oxygen Carrier Capacity

Using the optimized formulation and manufacturing process, we prepared two batches each of DF-NE and RSV-NE at different production scales. Figure 3A illustrates the triphasic PFC-NE structure incorporated with RSV and NIRF dye in the hydrocarbon shell surrounded by a micelle layer. The 250 mL small-scale batches (NEs 9–10) were processed using the LM20 microfluidizer, while the 1000 mL large-scale batches (NEs 17–19) were produced using the M110P microfluidizer. These batches are a subset of the NEs presented in Table 1. Despite the same interaction chamber and processing mechanism in both microfluidizers, they differ in their processing capacity. We successfully transferred the process from small-scale to large-scale with consistent colloidal and fluorescence attributes across all batches. No significant difference in the particle size was observed for DF-NE and RSV-NE in both 250 and 1000 mL batches, with a mean droplet diameter of ~105 nm and a polydispersity index of less than 0.3 (Figure 3B). Additionally, they showed similar fluorescence intensity (Figure 3C) at all tested dilutions and no significant difference in %PFOB loading (Figure 3I). To evaluate the stability of nanoemulsions in high-stress conditions during storage or transport, the high-speed centrifugation test was performed. RSV-NE showed no more than 10% change in particle size and overlapping size distribution after centrifugation at 1100 and 3000 rpm (Figure 3E). Since these nanoemulsions are designed for organ perfusion, it is essential to assess their colloidal stability in the presence of biological fluid. Figure 3F shows a <10% change in particle size of RSV-NE after 72 h incubation in water, Dulbecco’s Modified Eagle Medium (DMEM), and 20% Fetal Bovine Serum (FBS) in DMEM, which is consistent with our previously reported PFC-NEs [43,48,49,50,52].

Before exposure to serum, the nanoemulsions were sterilized by filtering through a 0.22 µm membrane filter. This filtration process did not change the particle size and distribution of the nanoemulsions (Figure 3D). The colloidal stability of DF-NE under all stress conditions is shown in Figure S2. To ensure that the nanoemulsions retain their critical quality attributes for the desired performance during shelf life, the colloidal and fluorescence stability were evaluated over a 3-month storage period. As shown in Figure 3G,H, RSV-NE requires ~7 days after manufacturing to equilibrate, after which consistent particle size and fluorescence intensity are achieved under the storage conditions. The uniform particle size distribution of the developed NEs helps to avoid Ostwald ripening and provide a consistent drug concentration and release profile. Moreover, the ability of both nanoemulsions to maintain a high fluorescence intensity throughout their shelf-life facilitates stable in vivo NIRF tracking. Furthermore, accelerated stability studies at elevated temperature (37 °C) demonstrated a non-significant change in particle size, fluorescence intensity, and drug loading of nanoemulsions (Figure S3).

3.4. Drug Quantification and In Vitro Release Study of Resveratrol from PFC-NE

An HPLC calibration curve was developed by quantification of standard resveratrol in pure methanol. The standards samples of resveratrol from concentrations of 2.5–50 µg/mL were prepared in the mobile phase (75:25 methanol and water) to construct a calibration curve of AUC against concentration (Figure 4A). The limit of detection (LOD) and limit of quantification (LOQ) computed based on ICH Q2 (R2) were 3.07 µg/mL and 9.32 µg/mL. In Figure 4B, RSV-NE showed a high and comparable % drug loading across 250 mL and 1000 mL batches processed in two different microfluidizers. Figure 4C shows <10% drug loss in the 1000 mL batch over 2 months.

The representative figure of the resveratrol HPLC chromatogram is provided in Figure S4. The in vitro release of RSV from RSV-NE and Free RSV was performed using a dialysis bag method in a release medium (1% w/v Tween 80 in phosphate-buffered saline; pH 7.4) at 37 °C. The cumulative amount of RSV release and % cumulative RSV release from RSV-NE and free RSV are shown in Figure 4D,E. A burst release of RSV was observed after 3 h in the free RSV group, which had a % RSV cumulative release of 58.03 ± 1.95% compared to 3.56 ± 1.11% in the RSV-NE group (Figure 4F). The free RSV showed a decrease in the % RSV release at later time points, probably due to degradation of RSV in aqueous solution. In contrast, RSV-NE exhibited slow and sustained release of RSV throughout the study period. By day 5, the cumulative release of RSV from free RSV solution decreased to 36.54 ± 1.8%, whereas RSV-NE achieved a significantly higher cumulative release of 63.37 ± 2.48%. Importantly, this sustained release behavior correlated with enhanced protection against ferroptosis-induced cell death in RAW 264.7 macrophages compared to free RSV, where rapid release and transient high drug concentrations contributed to increased cytotoxicity (Section 3.8. The drug-release data for RSV-NE showed a good fit to the Korsmeyer–Peppas model, with a correlation coefficient of R² = 0.97 (Figure S4). Drug release was described using the equation Mt/Mα = k·tⁿ, where Mt/Mα represents the fraction of drug released at time t, k is the release-rate constant, and n is the release exponent that reflects the underlying release mechanism [53]. The calculated value n = 0.57 (within the range 0.5 < n < 1) indicates an anomalous, non-Fickian transport behavior, suggesting that drug release is governed by a combination of mechanisms, including diffusion. This release behavior ensures prolonged drug availability from RSV-NE over a clinically relevant 24–48 h ex vivo organ preservation period, where continuous antioxidant protection is essential to counter ongoing oxidative stress.

3.5. Oxygen-Loading and In-Line Perfusion NIRF Imaging of PFC-NEs

The oxygen-carrying capacity of the PFC-NE was evaluated using two methods: machine perfusion setup and oxygen bubbling. In the machine perfusion setup, the PFC-NE is pumped to an oxygenator through which the oxygen gas (55% FiO₂) is diffused and carried by the PFC-NE. UW solution is the standard choice of organ preservation solution, which was used as a positive control. As shown in Figure 5A, the partial pressure of oxygen (pO₂) levels of both oxygenated perfusates increased with time and plateaued after 100 min, in which the pO₂ of PFC-NE (340 Torr) was found to be significantly higher than that of UW solution (250 Torr) at 6 h. In Figure 5B,C, PFC-NE showed no significant change in particle size and fluorescence signal throughout the 6 h, suggesting its excellent colloidal and fluorescence stability in a machine perfusion setup.

In the bubbling method, 100% FiO₂ was bubbled into a sealed beaker containing the RSV-loaded PFC-NE or UW solution. The pO₂ in the sample was measured until the plateau phase was achieved. Triplicate measurements were performed for RSV-NE and UW solutions by monitoring pO₂ levels in three independent batches from the bulk volume. The average pO₂ level achieved at the plateau phase in RSV-NE was ~600 Torr, and ~570 Torr in the UW solution (Figure 5D). To evaluate the stability of the formulation in the presence of oxygen, the colloidal attributes and drug loading were measured before and after oxygen bubbling. As shown in Figure 5E,F, there is a tight overlap of size distribution curves and no significant change in the drug content of the RSV-NE samples before and after oxygen bubbling.

To demonstrate the utility of NIRF fluorescence PFC-NEs for non-invasive monitoring of perfusate flow by NIRF imaging, we used a clinical-grade NIRF imager, Fluobeam LX (Fluoptics). The imager head was positioned over the membrane oxygenator, and the PFC-NE was pumped through by a controlled-pressure circuit. Figure 5G shows an NIRF image of the PFC-NE, which shows a uniform fluorescence signal across the perfusion tubing and the membrane oxygenator. In contrast, the UW solution showed only a low autofluorescence signal from plastic materials under identical imaging conditions (Figure 5H). NIRF imaging can help evaluate the colloidal and fluorescence integrity of the PFC-NE as it runs through the circuits and membrane oxygenators. Under organ perfusion conditions, NIRF imaging would support evaluating PFC-NE distribution through the organ under machine perfusion preservation. NIRF signal relates to the presence of PFC-NE droplets in tissues, which directly relates to the presence of oxygen carried by the PFC-NE. Therefore, NIRF imaging can inform the perfusionist of perfusion effectiveness and serve as an indirect measure of oxygenation during machine perfusion-based organ preservation.

3.6. Oxygen Saturation and Off-Loading Under Machine Perfusion

Both the PFC-NEs and Belzer UW^® solution demonstrated a linear oxygen loading capability in response to the different gas inputs. From the gas-input study, Belzer UW^® solution had oxygen saturations of 642.9 ± 3.2, 541.2 ± 2.6, 422.6 ± 5.1, 288.7 ± 4.5, and 168.4 ± 3.5 mmHg at 100%, 80%, 60%, 40%, and 20% carbogen inputs, respectively. The PFC-NEs had oxygen saturations of 671.6 ± 0.9, 592.4 ± 7.5, 493.6 ± 21.4, 380.7 ± 25.0, and 277.7 ± 20.1 mmHg at 100%, 80%, 60%, 40%, and 20% carbogen inputs, respectively. At 100% gas input, there was no significant difference between the two perfusates, although the PFC-NEs had statistically significant (p < 0.05) higher oxygen saturations for each of the subsequent inputs from 80% to 20% (Figure 6A).

Oxygen saturation was comparable throughout all three flow rates, with no significant decrease in oxygen loading at any of the conditions for either Belzer UW^® Solution or PFC-NEs. Belzer UW^® Solution had oxygen saturations of 630.0 ± 7.2, 636.7 ± 5.6, and 642.1 ± 3.1 mmHg at flow rates of 30 mL/min, 20 mL/min, and 10 mL/min, respectively. The PFC-NEs had oxygen saturations of 635.5 ± 5.9, 646.3 ± 6.1, and 665.6 ± 6.5 mmHg at flow rates of 30 mL/min, 20 mL/min, and 10 mL/min, respectively. After comparison between the groups, the PFC-NEs had statistically significant (p < 0.05) higher oxygen saturation at 30 mL/min, while the lower flow rates were not statistically significant (Figure 6B).

Gas responsiveness of both perfusates was characterized by measuring the oxygen saturation with a PreSens© flow-through oxygen cell. During the experiment, the time of gas input change was recorded, and the time it took for the perfusate to begin to respond to the change in input was also recorded. The Belzer UW^® Solution had a slower response time to the change in input, while exhibiting a faster rate of oxygen loss following the change in input. In contrast, the PFC-NEs demonstrated a slower overall rate of oxygen loss despite having a faster response time to register the change in input (Figure 6C). It was noted that both perfusates had reasonably quick gas responsiveness, with changes being seen within 2.8 min and 1.9 min for Belzer UW^® and PFC-NE, respectively (Figure 6D).

3.7. Oxygen Radical Antioxidant Capacity (ORAC) of Resveratrol Loaded PFC-NE

The in vitro antioxidant activity of FD-RSV and RSV-NE was evaluated using the ORAC assay. The ORAC assay is based on the inhibition of the peroxyl-radical-induced oxidation of a fluorescent probe (fluorescein) by an antioxidant. 2,2′-Azobis(2-aminopropyl) dihydrochloride (AAPH) acts as a peroxyl radical initiator, quenching the fluorescence of fluorescein probe over time. The fluorescence intensity was followed to monitor the decay of the fluorescence curve. The sample antioxidant capacity correlates with the fluorescence decay curve, which is represented as the area under the curve (AUC). We performed three independent assays to assess reproducibility.

As shown in Figure 7B, the net AUC of each Trolox concentration (100, 50, 40, 30, 20, 10, 5, and 0 μM) was used to plot the standard curve for determining the ORAC of the antioxidant samples. We compared the fluorescence decay curves of all treatments at the highest concentrations. Figure 7A shows that Trolox, FD-RSV, and RSV-NE delay the fluorescence decay compared to DF-NE and Blank solutions, in which the fluorescence signal decays readily. We were able to reproduce these results in the other two assays (Figure S5A,B). A dose-dependent increase in net AUC was observed for both FD-RSV and RSV-NE (Figure 7C), which means that an increase in RSV concentrations leads to increased delay in fluorescence decay. The dose-dependent effects are also evident in the delayed fluorescence decay curves of Trolox, RSV-NE, and Free-RSV, as shown in Figure S5D,E. The ORAC values, expressed as Trolox equivalents (TE) for FD-RSV and RSV-NE at 1.25 μM each, were 58.34 μM TE and 103.45 μM TE. This ORAC assay suggests that RSV-NE is a potent free-radical scavenger and exhibits dose-dependent antioxidant activity. In oxidative stress conditions, such as ferroptosis and IRI, RSV released from NE can effectively prevent radical-induced lipid peroxidation and cellular damage.

3.8. Cell Viability and Ferroptosis Inhibition Study In Vitro

The cytotoxicity of the nanoemulsions was evaluated before performing any functional assays. An ATP-based CellTiter-Glo^® luminescent assay was conducted to assess the viability of RAW 264.7 macrophages following exposure to DF-NE and RSV-NE for 24 h. The % viability was measured as the luminescence signal generated by ATP released by viable cells, normalized to that of untreated cells. Figure 8A shows no significant drop in the ATP levels in either of the DF-NE or RSV-NE group until a dose of 40 μL/mL nanoemulsion. However, the free RSV dissolved in DMSO exhibited a dose-dependent decrease in % macrophage cell viability, with only ~15% viable cells at the highest drug concentration (182.5 μM). In the RSV-NE group, there is no dramatic drop in % macrophage cell viability, and it shows a 3-fold higher number of viable cells at 182.5 μM drug concentration. The toxicity of the free RSV dissolved in DMSO was due to the drug alone (RSV), as the volume-matched DMSO-treated group did not show significant toxic effects (Figure 8B). The toxicity of RSV at higher concentrations is consistent with our previous findings [43]. Next, to evaluate the ferroptosis inhibitory activity of RSV-NE, RAW 264.7 macrophage cells were simultaneously exposed to RSV-NE, volume-matched DF-NE, and FD-RSV solution, and Ferrostatin-1 (a ferroptosis inhibitor), along with RSL3 (a ferroptosis inducer). The macrophage cell viability data in Figure 8C show that the radical trapping antioxidant, Ferrostatin-1, prevented ferroptosis compared to the untreated group. Similarly, RSV-NE prevented ferroptosis and showed similar cell viability to Ferrostatin-1, even at low concentrations. The % viability in the RSV-NE group reduced at higher concentrations, however, significantly higher than the untreated and the FD-RSV groups. FD-RSV was unable to prevent ferroptosis and showed less than 50% viability at all tested concentrations. At the highest RSV concentration, RSV-NE exhibited approximately 7.1% higher cell viability than FD-RSV. Interestingly, DF-NE also increased cell viability, slightly lower than the RSV-NE group. This effect could be the result of metabolic adaptation of macrophages upon nanoparticle uptake or oxygen-mediated protective effects of PFC-NE. Wan Chen et al. reported that hyperbaric oxygen therapy (100% FiO₂) protected against cerebral IRI in rats through ferroptosis inhibition [64]. As shown in Figure 8A,D, the baseline pO₂ levels (~120 Torr) indicate that PFC-NE carries nascent oxygen that may protect against ferroptotic cell death. We were able to replicate these results in another independent ferroptosis inhibition assay (Figure S6). Moreover, Microscopy images were captured for the treated and untreated cells after incubation with RAS-selective lethal compound 3, (1S,3R)-RSL3, (1S,3R)-Methyl 2-(2-chloroacetyl)-2,3,4,9-tetrahydro-1-[4-(methoxycarbonyl)phenyl]-1H-pyrido [3,4-b]indole-3-carboxylate (RSL3, ferroptosis inducer) using microscopy. At the highest concentration, the RSV-NE and Ferrostatin-1 treated groups retained the macrophage cell morphology (Figure 8E,F) compared to the untreated group, where the cells are shrunk and damaged (Figure 8D). The free RSV in DMSO at the highest concentration also showed shrunken cell size and damaged structural integrity. The microscopy images of other treatment groups are in Figure S6.

4. Discussion

In this study, we developed a scalable, stable, and multifunctional perfluorocarbon nanoemulsion (PFC-NE) platform that serves as both an artificial oxygen carrier (AOC) and a controlled-release delivery system for resveratrol (RSV). The PFC-NEs are designed to serve as perfusate for ex vivo organ preservation in a machine perfusion system, where oxygenation and the prevention of ischemia–reperfusion injury (IRI) remain major challenges. This optimized PFC-based AOC was developed by leveraging conclusions from a previous comprehensive study that focused on QbD-driven development of triphasic PFC-NE for oxygen delivery in ex vivo machine perfusion. The triphasic nanoemulsion structure included an internal phase of PFC and hydrocarbon oil, dispersed as nanosized droplets in a surfactant-stabilized external aqueous phase. The hydrocarbon oil phase consisted of Miglyol 812N, which is a medium-chain triglyceride, and a GRAS (generally regarded as safe) excipient widely used in parenteral emulsions. However, this formulation was limited by suboptimal PFC concentrations (<15% w/v) and poor shelf-life storage stability, suggesting further improvements. In the present work, compositional modifications successfully increased the PFC concentration to 30% w/v, within the range reported for commercially developed PFC-based AOC. The surfactant system was optimized by replacing Pluronic P123 (HLB 8) with Cremophor EL (HLB 13) to match the required HLB of the hydrocarbon oil, Miglyol 812N^®. Furthermore, Miglyol 812N^® and PFOB (Perfluorooctyl bromide) were selected as the ideal combination of HC oil and PFC based on particle size minimization over a 2-month follow-up. We incorporated resveratrol (RSV) in the formulation as a model antioxidant drug that can scavenge excess ROS and protect against ferroptosis-induced IRI during organ preservation. ICG, a clinical-grade NIRF dye, was chosen for real-time monitoring of perfused organs, as well as to compare different NE batches for reproducibility. The batch size of the NEs in the previous study was 25 mL, which is an insufficient volume for organ perfusion on a commercial scale. To evaluate the scalability, robustness, and reproducibility of our manufacturing process, we developed a series of NE batches varying the batch size and Microfluidizer processing unit. After finding no significant impact of the presence of drug or dye on NE particle size, we compared all batches grouped according to their batch size and processing equipment. The particle size and size distribution are critical quality attributes of NEs that govern colloidal stability, while the PFOB loading correlates with the oxygen-carrying capacity of NEs [21]. A consistent particle size range of 90–110 nm and high PFOB loading (>70%) was achieved across small- to large-scale batches processed in different Microfluidizers. We scaled up our formulation systematically and successfully transferred our manufacturing process to new equipment, maintaining the critical quality attributes of the product. The International Council for Harmonization (ICH) Q10 Pharmaceutical Quality System Guidance document states that the pharmaceutical quality system should assure that the desired product quality is routinely met. A less than 10% coefficient of variation in particle size and an overlapping size distribution were obtained between batches produced in the same scale or equipment, demonstrating a reproducible manufacturing process [65].

A pair of triphasic nanoemulsions: drug-free nanoemulsion (DF-NE) and resveratrol-loaded nanoemulsion (RSV-NE) produced on the largest scale (1000 mL) in the M11OP processor was evaluated under physical stress conditions and long-term shelf life storage stability at 4 °C. Less than 10% change in particle size was observed in these batches after being subjected to high-speed centrifugation, filtration, and serum exposure at 37 °C for 72 h or 3-month shelf-life storage. Additionally, RSV-NE exhibited a <10% change in drug loss after 2-month shelf-life storage or 72 h incubation at 37 °C. These attributes are consistent with other well-characterized nanoemulsion systems intended for intravenous or perfusate applications. This data confirms the stability of our formulation in a machine perfusion setting for hypothermic (4 °C), sub-normothermic (25 °C), or normothermic (37 °C) organ preservation. The in vitro drug release study revealed a slower release of RSV from NE compared to the burst release seen in the RSV solution. The slower drug release profile may help to maintain the minimum effective therapeutic concentration throughout the period of organ preservation and avoid toxic effects due to sudden accumulation of the drug [66,67]. RSV is prone to chemical degradation and isomerization in solution, which can be a reason for the decrease in the cumulative amount of RSV release from RSV solution at later time points (after 10 h). In organ transplantation, the organ undergoes hypoxic stress during procurement and preservation, followed by reperfusion injury after being transplanted to the recipient’s body. Supply of oxygen is essential to maintain the integrity of the organ during machine perfusion, and currently available perfusates lack an oxygen-carrying component. PFC-NE are known to be effective oxygen carriers with a tunable oxygen release profile.

The oxygen-carrying capacity of PFC-NE was evaluated using two independent methods: an ex vivo machine perfusion system and direct oxygen bubbling. Both oxygenation methods are clinically relevant as machine perfusion is one of the most widely used dynamic organ preservation techniques, whereas direct oxygen bubbling is considered when a cheap, easy, and transport-compatible method is needed [68,69]. We developed a machine perfusion setup, including an oxygenator equipped with a microporous hollow fiber membrane through which the oxygen is diffused and carried by the PFC-NE. In this setup, we found that PFC-NE achieved consistently higher pO₂ levels compared to the standard UW solution after oxygenation for 6 h. The oxygen bubbling method provided complementary evidence of consistent and slightly higher pO₂ levels of PFC-NE compared to the UW solution after triplicate measurements from three independent batches. This enhanced oxygenation capability stems from the physicochemical properties of PFCs, which dissolve large quantities of gases in accordance with Henry’s law, regardless of temperature and pH [20,27,29]. PFC-NE was also able to maintain its colloidal, fluorescence, and drug loading stability during oxygenation cycles and perfusion conditions, validating its potential as a high-performance acellular oxygen carrier. These results were further confirmed using the BMI OrganBank^® machine perfusion circuit, in which both PFC-NE and UW showed a proportional increase in pO₂ level at increasing gas input concentrations (20–100% FiO₂). However, PFC-NE exhibited significantly higher oxygenation at every level, even at reduced FiO₂. These findings suggest that the PFC-NE facilitates efficient oxygen loading even under moderate oxygen supply, which is essential in clinical situations where oxygen delivery may be limited or variable. Importantly, oxygen loading in both perfusates remained linear with respect to the FiO₂ inputs, confirming predictable, controllable oxygen-carrying behavior, which is essential for safe, precise oxygenation during organ preservation. Moreover, in-line perfusion NIRF imaging demonstrated a strong and uniform fluorescence signal for NIRF-labeled PFC-NE, whereas the UW showed a negligible signal. This enables real-time, non-invasive monitoring of perfusate distribution during organ preservation, a capability not feasible with conventional preservation solutions, such as UW. Even when ICG dye is added to standard clinical perfusates, it cannot maintain long-term signal stability due to rapid fluorescence loss in aqueous media. In contrast, the ICG incorporated into the nanoemulsion droplet sustains fluorescence signal under mechanical stress, yielding a stable signal throughout perfusion and providing a reliable measure of perfusate distribution through the machine circuitry and the potential perfused organ.

Excess accumulation of ROS and lipid peroxides during the reperfusion phase has been linked to ferroptosis [35]. Hydroxyl radicals are the most common ROS that oxidize membrane lipids, forming lipid peroxides, which together damage the cell [70]. In the ORAC assay, RSV-NE exhibited potent radical scavenging activity by inhibiting the AAPH radical-induced fluorescence decay of the fluorescein probe and showed dose-dependent antioxidant activity like that of the free-RSV solution. The in vitro cytotoxicity study demonstrated that DF-NE and RSV-NE did not cause a significant decrease in RAW 264.7 macrophage cell viability at doses up to 40 μL/mL. However, free RSV in DMSO showed a dose-dependent reduction in cell viability and significant toxicity at higher concentrations. This toxic effect could be attributed to the rapid release of RSV from the solution compared to RSV-NE. Lastly, two independent in vitro ferroptosis inhibition assays were performed to evaluate the protective effects of RSV-NE in RSL3-treated macrophages. RSV-NE prevented ferroptosis and maintained cell viability similar to the control Ferrostatin-1 at low concentrations, but decreased viability at higher concentrations. Notably, RSV-NE exhibited significantly higher cell viability than free RSV in DMSO at all tested concentrations. The RSV-NE-treated group at the highest concentration retained the morphological characteristics of RSL3-activated macrophage cells, unlike the untreated group, where the structural integrity of cells was lost. These findings support the hypothesis that RSV-NE has the potential to counteract ROS accumulation and lipid peroxides, thereby preventing ferroptosis and IRI. The presented PFC-NE offers several advantages over existing perfusates and artificial oxygen carriers. Unlike hemoglobin-based oxygen carriers (HBOCs), which exhibit pH- and temperature-dependent oxygen dissociation and carry risks of oxidative toxicity and immunogenicity, PFC-based AOCs are chemically and biologically inert with stable oxygen loading and dissociation. Although we compared our optimized PFC-NE with the UW solution, comparison with HBOCs and other PFC-based formulations was not within the scope of this study. Furthermore, integrating RSV within the PFC-NE uniquely positions this formulation as a dual-functional perfusate additive—delivering oxygen while actively mitigating ferroptosis and IRI. From a clinical perspective, the dual-functionality of the PFC-NE platform has direct implications for organ preservation, particularly for extended-criteria and marginal organs that are highly susceptible to ischemia–reperfusion injury. By enhancing oxygen delivery while providing sustained antioxidant protection against ferroptosis-induced IRI during machine perfusion, this approach has the potential to improve organ viability, extend preservation windows, reduce IRI-associated functional decline, and lower organ discard rates. Despite this potential, several challenges remain that need to be addressed for clinical translation. This includes evaluation of long-term storage stability, assessing immunotoxicity and safety, and overcoming regulatory compliance associated with acellular PFC-NE perfusates for clinical use. Batch-to-batch variability remains a key challenge when scaling up the manufacture of PFC-NEs to large volumes (≈1000 mL), particularly for formulations containing ICG, a temperature-sensitive dye. Ensuring sterility through aseptic manufacturing and tightly controlling pyrogen levels are therefore critical. In addition, prior unsuccessful clinical trials of perfluorocarbon oxygen carriers are likely to increase regulatory scrutiny [16,28,71,72]. Nevertheless, implementing a Quality-by-Design (QbD) framework with clearly defined critical quality attributes (CQAs) for the perfusate can help address both manufacturing and regulatory challenges. Future studies will focus on long-term stability studies under clinically relevant conditions, as well as mechanistic in vitro and in vivo studies. For in vitro studies, we will assess the antioxidant and ferroptosis-inhibitory activity of RSV-NE in a hypoxia-reperfusion model using relevant cell types to closely mimic in vivo organ IRI through the measurement of oxidative stress markers (ROS, lipid peroxides). For in vivo studies, PFC-NE perfusate will be integrated into ex vivo normothermic machine perfusion models, which have gained increasing attention for safety and functional assessment of multiple organs in large animals [73,74,75,76]. Together, these approaches will confirm safety and efficacy across cellular and organ-level, advancing the PFC-NE platform towards clinical application.

5. Conclusions

In this work, we developed a robust, scalable, and dual-functional PFC-based nanoemulsion platform designed to address oxygen deficiency and oxidative stress during ex vivo organ perfusion. Through formulation optimization, we increased the PFC content to 30% w/v and achieved reproducible manufacturing at up to 1000 mL scale, maintaining high PFOB loading, and colloidal and fluorescence stability. RSV-NE retained colloidal stability under physical and biological stress conditions and provided sustained antioxidant delivery with minimal loss during storage or oxygenation cycles. Across multiple oxygenation methods—including bubbling, oxygenator, and BMI OrganBank^® machine perfusion—the PFC-NEs demonstrated higher and stable oxygen loading relative to UW solution, even under reduced FiO₂. RSV-NE exhibited dose-dependent ORAC antioxidant activity and provided significant protection against RSL3-induced ferroptosis, with RSV-NE outperforming free RSV and matching the efficacy of Ferrostatin-1 at lower concentrations. Collectively, these findings demonstrate that the PFC-NE platform integrates oxygen delivery with antioxidant protection, offering a promising next-generation acellular perfusate for mitigating IRI during organ perfusion. This technology has strong translational potential for protecting the structural and functional integrity of organs after transplantation. Future work will include organ-relevant in vitro models and large-animal ex vivo normothermic machine perfusion studies (pig kidney or liver) using RSV-loaded PFC-NE to evaluate cellular and organ viability, functional recovery, and protection against IRI.