Theranostic nanoemulsions suppress macrophage-mediated acute inflammation in rats

Abstract

In inflammatory diseases or following an injury, dysregulated inflammation is a common driver of pain and tissue damage. Macrophages are immune cells that contribute to the initiation, maintenance, and resolution of inflammation due to their phenotypic plasticity in response to signals from inflammatory microenvironments. Macrophages infiltrate and polarize toward a pro-inflammatory phenotype (M₁-like), thereby increasing the severity of inflammation. Therefore, we aimed to suppress the pro-inflammatory activity of M₁-like macrophages and decrease their infiltration at the site of inflammatory insult to resolve tissue inflammation. To achieve this, we developed a theranostic curcumin-loaded nanoemulsion platform that delivers a low dose of curcumin, a known anti-inflammatory phytochemical, to macrophages and allows in vivo tracking of macrophages by near-infrared fluorescence (NIRF) imaging technique. In vitro, we showed that curcumin-loaded nanoemulsion suppressed polarization of macrophages towards M₁-like phenotype, consequently decreasing the release of pro-inflammatory cytokines and mediators like IL-6, IL-$\:\beta\:,$ TNF-$\:\alpha\:$, and nitric oxide (NO). Furthermore, curcumin-loaded nanoemulsion increased the level of IL-10, an anti-inflammatory cytokine, and protected macrophages against ferroptosis compared to drug-free nanoemulsion. In a rodent model of Complete Freund’s adjuvant (CFA)-induced inflammation, we demonstrated that infiltrating macrophages sequestered curcumin-loaded nanoemulsion droplets and acted as cellular drug depots at the site of inflammation. This consequently decreased macrophage infiltration at the CFA-induced inflammation site in both sexes compared to drug-free nanoemulsion, as demonstrated by NIRF imaging, H&E staining, and immunofluorescence. Taken together, our results indicated that the anti-inflammatory efficacy of curcumin was significantly improved when directly delivered to pro-inflammatory macrophages via theranostic nanoemulsion. This work opens an avenue for exploring theranostic nanoemulsions as a platform for delivering natural anti-inflammatory products for immune modulation.

Graphical abstract

Introduction

Inflammation is a complex sequence of events involving both the innate and adaptive immune systems, triggered in response to noxious stimuli, infection, or injury [1]. Macrophages are mononuclear immune cells that initiate and resolve both chronic and acute inflammatory responses [2]. Macrophages participate in three major functions during inflammation: antigen presentation, phagocytosis, and tissue immunomodulation [2, 3]. During inflammation, both the tissue-resident and infiltrating macrophages undergo proliferation and differentiation to restore the lost homeostatic balance. In response to the encountered cellular microenvironments, macrophages activate their metabolic plasticity and rewire their cellular metabolism, nutrient usage, generation of metabolites, expression of cell surface markers, and secretion of cytokines [4, 5]. The resting macrophages (M₀) undergo stimulus-responsive differentiation, leading to the development of two distinct polarization states referred to as M₁-like (pro-inflammatory phenotype) and M₂-like (anti-inflammatory phenotype) [4]. The M₁-like macrophage subset is particularly involved in guiding and initiating acute inflammatory responses and is present in inflammatory environments dominated by toll-like receptor (TLR) and interferon (IFN) signaling pathways. This subset actively produces proinflammatory cytokines and mediators such as IL-6, IL-$\:\beta\:,$ TNF-$\:\alpha\:$, and nitric oxide (NO), respectively [3, 6]. The involvement of M₁-like macrophages in the pathogenesis of various inflammatory diseases implicates them as key targets for developing anti-inflammatory therapeutics [7]. Therefore, in the proposed work, we aim to engage these pro-inflammatory macrophages as a strategic means to resolve acute inflammation.

In this study, we selected a rodent model of acute inflammation triggered by Complete Freund’s adjuvant (CFA). CFA is composed of inactivated mycobacteria in an oil-in-water emulsion, that primarily triggers a Type IV cell-mediated hypersensitivity immune response by activating macrophages, dendritic cells, and T-helper 1 (Th1) cells. Liu et al. [8] showed that subcutaneous administration of CFA in rodent foot pads resulted in severe local inflammation, primarily due to excessive infiltration of macrophages at the site of inflammation. Lin et al. [9] reported that these infiltrating macrophages showed an increased expression of iNOS, Arg1, TNF-$\:\alpha\:$, and IL-$\:\beta\:$, indicators of the M₁-like phenotype. Furthermore, research studies have linked the release of pro-inflammatory cytokines, such as IL-6 and IL-$\:\beta\:$, to the occurrence of ferroptosis [10, 11]. Ferroptosis is a non-apoptotic, caspase-independent cell death characterized by an iron-dependent accumulation of lipid peroxides and is known to aggravate inflammatory responses [12]. However, the occurrence of ferroptosis is contingent upon the type of cells or tissue involved. In macrophages, inhibiting ferroptosis can decrease inflammatory disease progression and promote wound healing [13,14,15,16]. A recent study showed that macrophage phenotypes have distinct susceptibilities to ferroptosis, where pro-inflammatory macrophages are more resistant while anti-inflammatory macrophages are more susceptible to ferroptosis [17]. Therefore, a promising therapeutic approach would be to prevent these anti-inflammatory macrophages from undergoing ferroptosis, which could lead to sustained anti-inflammatory effects at the site of inflammation.

Curcumin is a natural polyphenolic compound that regulates multiple inflammatory signaling pathways in different inflammatory disease models [18]. Curcumin is an NLRP3 inflammasome inhibitor and thus results in potent anti-inflammatory action in vitro and in vivo [19, 20]. Curcumin also inhibits NF-κB, a key up-regulator of proinflammatory cytokines (TNF-α, IL-1β, IL-6, etc.) in macrophages [21]. It is reported to induce polarization of M₁-like phenotypic macrophages towards the M₂-like phenotype by increasing the production of IL-4 and/or IL-13 in a STAT6-dependent manner [22,23,24]. Curcumin has also been shown to protect chondrocytes [25], hippocampal cells [26], and bone marrow stromal cells [27] from ferroptosis. However, its effect on macrophages is not well established. Randomized clinical trials have demonstrated that high doses of curcumin (80 mg/day to 1500 mg/day for weeks or months) are required to achieve anti-inflammatory or protective benefits [28, 29]. This could be attributed to factors like the low systemic bioavailability of curcumin post-oral administration or its inability to specifically engage with inflammatory cells upon systemic administration.

In our previous work, we have established that drug delivery to the circulating monocytes and macrophages via nanoemulsions (NEs) reduces macrophage infiltration at the site of inflammation compared to free drug solution, which is not encapsulated in a nanocarrier system [8, 30]. NEs are colloidal dispersions of two immiscible liquids, generally oil and water, with a mean droplet diameter less than 500 nm [31]. The reduced macrophage infiltration and polarization towards the M₂-like phenotype led to attenuation of neuropathic pain and mechanical hypersensitivity in a chronic nerve constriction injury (CCI) rat model of peripheral nerve injury [30, 32, 33].

In this research, we hypothesized that curcumin-loaded nanoemulsion (CUR-NE) would deliver curcumin to activated macrophages, inhibit their M₁-like action, and decrease their infiltration at the site of inflammation. This would consequently reduce acute inflammation in both male and female rats. Importantly, these anti-inflammatory benefits would be achieved at a low dose of curcumin. To test the hypothesis, we first formulated CUR-NE with two imaging modalities, a ¹⁹F Magnetic Resonance Imaging (MRI) contrast agent and a Near-Infrared Fluorescence (NIRF) imaging dye. The designed CUR-NE allows for real-time monitoring of infiltrating macrophages at inflamed sites. In vitro, we evaluated the anti-inflammatory and protective effects of CUR-NE on macrophages. In vivo, we assessed the effect of CUR-NE to reduce macrophage infiltration and inflammation in CFA-induced rodent footpads using both rodent sexes. Taken together, this study confirms that curcumin delivery via theranostic NE to macrophages attenuates inflammation in vivo via suppression of macrophage infiltration and associated inflammatory cytokines.

Materials and methods

Formulation of nanoemulsions (NEs)

Nanoemulsions were manufactured using M110S Microfluidizer (Microfluidics Corporation, Westwood, MA, USA) following the previous published protocol with some modifications [31, 34]. Briefly, for CUR-NE, CUR (Alfa Aesar, MA, USA) was dissolved by shaking in Transcutol (2-(2-ethoxyethoxy)-ethanol, catalog number E1022) overnight. A pre-emulsion was made by mixing solubilized CUR, Miglyol 812 (CREMER Oleo, Hamburg, Germany), Perfluoro-15-crown-5-ether (Exfluor, TX, USA), pre-dissolved fluorescent dyes: 1,1’-Dioctadecyl-3,3,3’,3’-Tetramethylindotricarbocyanine Iodide (DiR dye) and/or 1,1′-Dioctadecyl-3,3,3′,3′-Tetramethylindocarbocyanine Perchlorate [‘DiI’; DiIC18(3)] in DMSO, and 0.22 μm filtered 5% w/v surfactant micelle solution (3% Kolliphor EL and 2% Pluronic P105) in 1X PBS. The pre-emulsion was gently vortexed to avoid excess foaming and poured into the Microfluidizer M110S inlet reservoir. The pre-emulsion was processed at ~ 15,000 psi liquid pressure for 20 pulses. The same procedure was followed to manufacture drug-free nanoemulsion (DF-NE).

Colloidal characterization and TEM of the NEs

Dynamic light scattering (DLS) was used to measure size distribution, polydispersity index (PDI), and zeta potential (mV). This technique was performed on a Zetasizer Nano (Malvern Instruments, Worcestershire, UK). CUR-NE and DF-NE were diluted 1:40 v/v in deionized water prior to any characterization by DLS. The operating parameters were set as follows: refractive indices of material, 1.59; refractive index of dispersant, 1.33; temperature, 25 °C; viscosity of the dispersant, 0.8872 cP; and backscatter angle, 173 degrees. DLS data is plotted as the mean ± standard deviation of three measurements. Long-term stability follow-ups were performed for three months, and droplet diameter was measured using the DLS.TEM (FEI Tecnai F20) was used to study the morphology of CUR-NE. A copper grid (catalog no. 01700-F, Ted Pella) was activated using plasma cleaner for three seconds before use. The grid was stained with CUR-NE for one minute and the excess was removed using a filter paper. The grid was stained using solution of phosphotungstic acid (0.5% w/v, pH adjusted to 7) for 15 s before analyzing through TEM.

Stability Testing of NEs: Centrifugation and serum Stability

For the centrifugation test, NEs were centrifuged at 3000 rpm for 30 min at ambient temperature (Labnet Prism R Refrigerated Micro-Centrifuge). NEs were diluted 1:40 v/v in deionized water and characterized for droplet diameter and PDI. For serum stability, NEs were diluted 1:40 v/v in high (20%) FBS-containing cell culture medium. The diluted NEs were incubated at 37 °C for 72 h. Droplet diameter and PDI were recorded using DLS at the beginning and end of 72 h of incubation.

Near-Infrared fluorescence (NIRF) imaging of NEs

NIRF imaging of NEs was performed on the Li-COR Odyssey M. Serial dilutions of NEs in deionized water were prepared such that the DiR concentration was in the range of 5 µM to 0.156 µM. Dilutions were transferred to a clear 96-well plate in triplicate, and the fluorescence intensity was measured. The wavelength was set to 800 nm for the DiR channel, and imaging parameters such as intensity and focus were kept constant between all analyses.

Reverse phase high-performance liquid chromatography (RP-HPLC)

CUR loading in NE was measured using a validated RP-HPLC method on a Dionex Ultimate 3000 using a C18 column (Hypersil Gold™ C18 150 mm × 4.6 mm, 5 μm pore size, Thermo Fisher Scientific, USA) with UV detection at 448 nm. The mobile phase for CUR was acetonitrile: 2% v/v acetic acid in water (75:25). The retention time for CUR was around 3.1 min when the flow rate was maintained at 0.75 mL/min and the column temperature was 33°C. For CUR content analysis, NEs were first diluted in pure acetonitrile. Further, acetonitrile supernatants were diluted with 2% v/v acetic acid in water to match the mobile phase. All drug content analyses were performed in triplicate. The HPLC standard curve for CUR was produced in triplicate and the limit of detection (LOD) and limit of quantification (LOQ) were 0.18 $\:\mu\:$g/mL and 0.55 $\:\mu\:$g/mL, respectively.

In Vitro release of CUR from CUR-NE

The quantitative in vitro release test was performed at 37 ± 0.5 °C using the dialysis bag technique (molecular cutoff of 3000 Dalton). CUR-NE and CUR free drug (CUR FD) solution, drug equivalent to 500$\:\mu\:$g, were placed in the snakeskin dialysis bag (Thermo Fischer Scientific, REF 68035, USA). The receptor compartment consisted of a 15 mL mixture of 0.5% Tween 80 in phosphate-buffered saline). CUR release was also evaluated at physiological pH (pH = 7.4) and low pH (pH = 6), generally observed in inflamed tissue. In the drug release method, Tween 80 was used to increase the solubility of CUR and for maintaining appropriate sink conditions. The dialysis bag was switched into fresh medium at regular intervals as reported previously [35]. The quantitative analysis of CUR was performed using the HPLC method, as described in the “RP-HPLC analysis” section. The cumulative percentage of drug release versus time was plotted to evaluate the drug release pattern from a CUR NE and CUR solution.

Macrophage viability assay

RAW 264.7 macrophages (ATCC TIB-71, MD, USA, passage numbers 10–15) were seeded at 5,000 cells per well in a 96-well plate and incubated overnight. Macrophages were treated with NEs for 24 h. CellTiter Glo^® 2.0 (Promega Corporation, WI, USA) Luminescent Cell Viability Assay Kit was used to assess cell viability by quantifying ATP in metabolically active and viable cells using the manufacturer’s protocol. Luminescence was recorded using the Synergy HTX multi-mode reader (BioTek Instruments, Winooski, VT, USA).

Fluorescent microscopy

RAW 264.7 cells were seeded at 20,000 cells per well in an 8-well chamber slide (Lab-TekII) and left for attachment overnight. After 24 h, the existing media was removed, and the cells were treated with 40 µM of CUR-NE and volume-matched DF-NE for 6 h. Then cells were thoroughly washed with 1X PBS and mounted with mounting solution containing DAPI at ambient temperature before being imaged on the Keyence BZX microscope using CY7 channel.

Flow Cytometry

For studying the uptake of CUR-NE and DF-NE, RAW 264.7 macrophages (passage number 8–10) were plated at a density of 0.2 million cells per well in 12-well plates and left overnight for attachment. Macrophages were activated with LPS (500 ng/mL) for 18 h and then treated with 40 $\:\mu\:$M of CUR-NE and volume-matched DF-NE for 180 min. Macrophages were collected through trypsinization and fixed at room temperature with 2% PFA in 1X PBS for 20 min before flow cytometry analysis. For studying CD86 expression changes, LPS-activated macrophages treated with varying doses of CUR-NE (40 − 5 $\:\mu\:$M), volume-matched DF-NE, and CUR FD (40 $\:\mu\:$M) were collected through trypsinization, CD16/CD32 Fc block (catalog no. 14-0161-81 eBioscience™, MA, USA) was performed for 15 min on ice before staining with CD86 (B7-2) Monoclonal Antibody (catalog no. 46-0862-82 eBioscience™, MA, USA) for 30 min on ice. All experiments were performed in triplicate, and samples were analyzed using Attune Nxt and recording 40,000 events per sample. The NE was detected in the RL3 (DiR) channel and CD86 + macrophages were detected in BL3 channel. Gating of macrophages was applied based on forward scatter (FSC) and side scatter (SSC), as reported previously [35]. Data was analyzed using Attune NXT software.

Enzyme-linked immunosorbent assay (ELISA)

RAW 264.7 macrophages (passage numbers between 8 and 10) were plated into 6-well plates (0.3 million cells per well) and incubated for 48 h. The existing medium was replaced, and then cells were treated for 24 h with either CUR-NE (40 µM, 20 µM, 10 µM, and 5 µM), DF-NE (volume-matched to the highest concentration of CUR-NE), and CUR FD (40 $\:\mu\:$M). The existing medium was replaced with LPS (500 ng/mL) for 18 h. The supernatants were collected and spun at 4 °C and 1100 rpm for 5 min. The supernatants were stored at − 80 °C for ELISA analysis. IL-6 (catalog no. 88-7064-22, Invitrogen), IL-$\:\beta\:$ (catalog no. DY401, R&D Biosystems), TNF-$\:\alpha\:$ ELISA (catalog no. DY410, R&D Biosystems), and IL-10 (catalog no. DY417, R&D Biosystems) plates were developed following the manufacturer’s protocols.

Nitric oxide assay

The levels of NO in the presence or absence of NEs were measured based on measuring the levels of nitrite and nitrate using a colorimetric assay (catalog no. 23479, Millipore Sigma, MO, USA). Briefly, RAW 264.7 macrophages (1,500 cells per well) were seeded into 96-well plates. Post 24 h, cells were treated with varying doses of CUR-NE, DF-NE (volume-matched to the highest concentration), and CUR FD (40 $\:\mu\:$M) for 24 h. The next day medium was replaced with an LPS-containing medium (200 ng/mL) for 18 h. The Nitrite/Nitrate colorimetric assay was performed according to the manufacturer’s protocol.

Ferroptosis inhibition assay

Ferroptosis inhibition assay was performed as previously reported [36]. Briefly, RAW 264.7 macrophages were plated in a 96-well plate at 10,000 cells/well in cell culture media containing 40 ng/mL interleukin 4 (IL-4, catalog no. AF-214-14-1) and incubated for 24 h at 37 °C, 5% CO₂, and 95% humidity. CUR-NE and CUR FD were diluted in media and added to cells in 6 replicates to achieve a final curcumin concentration range of 2.5–20 µM. DF-NE was volume-matched to each of the tested concentration. Ferrostatin-1 was used as positive control and complete media was used as a negative control. Following 24 h of treatment, ferroptosis was induced by the addition of the GPX4 inhibitor RSL3 to culture media at 50 µL/well to achieve a final concentration of 2 µM and the plates were incubated for an additional 6 h. After ferroptosis induction was completed, cell culture medium (100 µL) was removed from each well and replaced with fresh media (100 µL/well), followed by CellTiter-Glo^Ⓡ (40 µL/well) assay.

Complete Freund’s adjuvant (CFA) induced inflammation in rats

All experiments and procedures involving the use of rats were approved by the Institutional Animal Care and Use Committee (#2012-11, Duquesne University, USA) in accordance with the National Research Council Guide for the Care and Use of Laboratory Animals. 6-week-old male (n = 9) and female (n = 9) Sprague Dawley rats were obtained from Hiltop Lab Animals Inc, PA, USA for experiment. Rats were maintained at the Research Animal Care Facility of Duquesne University in a 12 h light-dark cycle, with access to food and water ad libitum. Macrophage infiltration was evaluated in a Complete Freund’s adjuvant (CFA) inflammation rat model. Rats of both sexes were administered two 300 µL doses of CUR-NE or DF-NE tail vein injection at t = 0 h and t = 24 h, n = 3 rats per treatment group. To evaluate whether CUR FD has any effect on macrophage infiltration, rats were administered two 300 µL doses of CUR FD immediately followed by two 300 µL doses DF-NE (to track changes in macrophage infiltration pattern) at t = 0 h and t = 24 h via tail vein injection, n = 3. The formulated CUR-NE had a concentration of approximately 0.87 mg/mL. Thus, a single 300 µL dose delivered 0.261 mg curcumin, which corresponds to 1.04 mg of CUR/kg for an average 250 g rat. Inflammation was induced via a 50 µL intraplantar injection of CFA (catalog no. 77140, ThermoFisher Scientific) into the right hind paw at t = 0 h.

Whole-body NIRF imaging

NIRF images were acquired at t = 24, 48, 96, 120, 144, and 168 h (t = 1, 2, 4, 5, 6, 7 days) relative to CFA injection on a LiCor Pearl Trilogy small animal imaging system at the 700 nm channel, resolution 175 μm, focus offset of -2. Images from the Li-COR PEARL were imported into Image Studio Lite software. The fluorescence signal in the right hind paw (inflamed) was quantified using Image Studio Lite software by circling a region of interest (ROI) around the right hind paw from toe to heel. After final day 7 imaging, the rats were euthanized, and organs and hindfeet were collected for histology. Organs and hindfeet were placed on an organ imaging bed and imaged on the LiCor Pearl using a resolution of 170 μm and + 2 focus.

Immunofluorescence

Rat footpads were sectioned at 30 μm and placed on superfrost plus slides (Fisher Scientific; Pittsburgh, PA). The slides were kept at -20 °C until they were stained. Transverse sections of both the CFA injected and uninjected hindpaws from male and female rats (n = 1) were used to identify DiI signal from our NIRF-labeled NEs and CD68⁺CD40⁺ was used to stain M₁-like macrophage populations. For immunofluorescent staining, sections were removed from − 20 °C and allowed to warm to RT. An ImmEdgeTM Pen (Vector Laboratories; Burlingame, CA) was used to draw a hydrophobic barrier on each slide and the slides were dipped 10x in PBS to remove re-sidual OCT before the adding PBS containing 10% goat serum (Sigma)/1% Bovine serum albumin (Fisher Scientific)/ 2% DMSO (Sigma)/ 1 mg/ml digitonin (Millipore; Burlington, MA) (blocking solution). The slides were incubated in blocking solution for 1 h at RT. The blocking solution was replaced with antibodies diluted in blocking solution: rat anti-CD68 (1:1000, Clone: ED1; Bio-Rad) and rabbit anti-CD40 (1:50, Polyclonal; ThermoFisher). Primary antibodies were allowed to incubate overnight for at 4 °C in a humidified slide staining chamber. The next day, the slides were rinsed by dipping 10x in PBS, and goat anti-rabbit Alexa Fluor 488 (1:500; Invitrogen), goat anti-rat Alexa Fluor 647(1:500) and DAPI (500 ng/ml; Sigma) were added to each slide in PBS. The secondary antibodies and DAPI were incubated for 3 h at RT. Slides were dipped 10x in PBS, and 10% neutral buffered formalin was added to the sides for 10 min to help maintain DiI signal within the tissues. The sections were rinsed one final time by dipping 10x in PBS, and coverslipped with Prolong Gold Mounting Medium (Cell Signaling Technology, Danvers, MA). Images were obtained with a 20x (Numerical aperture: 0.75) objective on a Nikon A1R Confocal Microscope (Nikon Instruments Inc., Melville, NY), and image processing was performed using Nikon NIS-Elements Advanced Research (Nikon Instruments Inc.)

Histopathology

Male and female rats were sacrificed after 7 days post CFA injection. After whole body perfusion fixation performed following earlier established protocols [37], both the right (CFA injection) and left (internal/contralateral control) paws were disarticulated at the ankle joint and decalcified by transferring to Leica Decalcifier I solution (10% formic acid) for 8 h followed by rinsing for 20 min under water. The excised paw tissue samples were then fixed in 4% PFA solution, embedded in paraffin, and cut into 10 μm sections using microtome (Leica, Buffalo Grove, IL). Tissues were then prepared for histology with hematoxylin and eosin (H&E) and immunostaining. For immunostaining, non-specific binding was blocked with Dako serum-free protein for 15 min at room temperature. H&E and immunostained samples were imaged using the Keyence BZ-X800 all-in-one fluorescence microscope system (Keyence, Itasca, IL).

Immunostaining protocol

For immunostaining, non-specific binding was blocked with Dako serum-free protein for 15 min at room temperature. Rat anti-rat CD68 (Bio-Rad, Hercules, CA) (1/200 dilution in DPBS/0.1% BSA/0.05% Tween-20) was added overnight at 4 °C followed by a secondary antibody, anti-rat-Alexa 488 (Invitrogen, Grand Island, NY) (1/300 dilution in PBS/0.1% BSA/0.05% Tween-20) for 1 h at RT. After washing in DPBS/0.05% Tween-20, coverslips were mounted using Diamond anti-fading medium (Invitrogen, Grand Island, NY). Immunofluorescence-stained specimens were imaged using a Keyence fluorescence microscope (BZ-X800, Osaka, Japan) with a GFP filter, [ET - EGFP (FITC/Cy2) (Chroma, Vermont, US) (excitation Wavelength: 470/40x; emission wavelength 525/50m). An mOrange filter [mKO/mOrange (Chroma, Vermont, US) (excitation Wavelength: 530/30x; emission wavelength 575/40m)] was used to image DF-NE and CUR-NE labeled cells in tissue samples and co-localize the immunofluorescent staining with CD68 stain.

Statistical information

We generated all dose curves and response graphs using GraphPad Prism 9. The differences between treatment groups were performed with a two-tail unpaired t-test, one-way, or two-way repeated measures analysis of variance (ANOVA) with an appropriate multiple comparisons test, as described in the legends of each figure. *p < 0.05 was considered statistically significant.

Results

Development and characterization of theranostic curcumin nanoemulsion (CUR-NE)

Our group has previously shown the development of fluorescently labeled PFC NEs using the quality by design (QbD) approach [34, 38, 39]. QbD is a systematic, risk-based approach recommended by the US FDA to be adopted during pharmaceutical product development [40]. The developed NEs are triphasic systems, wherein the oil and PFC are simultaneously emulsified using non-ionic surfactants on a small-scale (M110S) microfluidizer. Miglyol 812, a medium-chain triglyceride, was selected as an oil phase based on the reported high solubility of CUR (2.4 mg/mL) [41]. Transcutol was used as a co-solubilizer to increase the solubility of CUR. By using the multiple regression modeling (MLR) approach in our prior study [34], we demonstrated that among the suitable PFCs selected as ¹⁹F MRI probes, PFCE was an ideal candidate as compared to perfluorooctyl bromide (PFOB) and perfluorodecalin (PFD) for developing triphasic NEs with increased colloidal stability. Additionally, PFCE has 20 magnetically equivalent fluorine atoms that result in a single peak on ¹⁹F NMR spectra. This increases the signal-to-noise ratio and produces high-contrast ¹⁹F “hot spots” that can be placed within the anatomical context using ¹H MRI during clinical MRI. Therefore, the fluorocarbon phase consisted of 30% w/v of PFCE, which supports successful labeling of macrophages in clinical settings [42]. NIRF imaging can be used as a simple and cost-effective alternative to ¹⁹F MRI during preclinical studies involving cell cultures and small rodents [43]. Therefore, we showed that the fluorescence signal intensity of DiR NIRF dye was not quenched due to the presence of CUR, unlike observed with DiD or DiO NIRF imaging dyes [34]. Thus, the theranostic NEs used in this study consist of bimodal imaging agents, a NIRF imaging dye (DiR), and a ¹⁹F MRI contrast agent (PFCE).

The size distributions of CUR-NE and DF-NE used in the presented in vivo study tightly overlapped each other, with a mean particle size of approximately 130 nm (Fig. 1A) and a polydispersity index of less than 0.2 (Figure S1 A). The zeta potentials for both the CUR-NE (-3.6 $\:\pm\:\:0.6\:$mV) and DF-NE (-3.9 $\:\pm\:\:$1.3 mV) were near the electroneutral range as the NEs were stabilized using non-ionic surfactants (Figure S1B). TEM was used to characterize the morphology of CUR-NE. TEM images showed spherical morphology with an average droplet size less than 130 nm (Fig. 1B).

Fig. 1

Characterization of NEs (CUR-NE and DF-NE). (A) Overlay of average size distribution by intensity measured on Day 0. (B) TEM image of CUR-NE. Magnification = 71kX, Scale bar = 100 nm. (C) Size of NEs stored for 72 h at 37 °C in high (20%) bovine serum-containing culture media. (D) Size of NEs before and after centrifugation at 3000 rpm for 30 min. (E) NIRF signal intensity comparison performed at the same settings between CUR-NE and DF-NE. (F) Long-term size stability monitoring of NEs was performed over three months. Each column in bar graph represents the mean ± SD (n = 3). unpaired t-test, ns not significant

The CUR loading in the NE was 79.23 ± 0.6%, quantified through RP-HPLC. To model the mechanical and cell culture conditions, CUR-NE and DF-NE were exposed to high-speed centrifugation and high serum-containing cell culture medium for 72 h at 37 °C, respectively. No significant increase in the droplet diameter was observed post-exposure to the stress conditions (Fig. 1C and D). The fluorescence signal intensities for DiR-labeled CUR-NE and DF-NE tightly overlapped each other, ensuring reliable comparison during in vitro macrophage uptake studies and in vivo NIRF imaging (Fig. 1E). Both the NEs showed less than a 10% increase in droplet diameter after 3 months of storage at 4 °C (Fig. 1F). Lipophilic dye leakage into the aqueous environment due to destabilization of NE can reduce fluorescence signal intensity of NEs over time. However, we did not observe a significant decrease in DiR fluorescence signal intensity from the developed theranostic NE over the three-month storage period (Figure S1C).

In vitro release of curcumin from CUR-NE at neutral and acidic pH

The in vitro drug release of CUR from NE was determined at two different pH values. The percentage cumulative release of CUR from NE at pH = 7.4, encountered post i.v administration, was only 3.03 ± 0.86% in 2 h. However, CUR FD solution showed a burst release, as 13.04 ± 1.6% of CUR was released within 2 h at 37 ± 0.5 °C. The release of CUR from NE was more sustained, as only 17.75 ± 2.16% of CUR was released in a day. A cumulative release of 64.09 ± 0.93% was observed from the CUR FD solution during the same duration. At the end of the study (t = 168 h), the percent cumulative release of CUR from NE was 48.94 ± 4.6% as compared to 85.00 ± 2.26% released from CUR solution (Fig. 2A). We also evaluated the release of CUR from NE at acidic pH (pH = 6), observed in inflamed tissue. The release of CUR from NE was sustained without any burst effect compared to CUR solution (Fig. 2B). At t = 168 h, 72.6 ± 2.3% of CUR was released from CUR-NE compared to 93.17 ± 6.7% released from CUR FD. Therefore, the release of CUR was significantly improved at acidic pH compared to neutral pH, especially at later time points (Fig. 2C). The release data was fitted into mathematical models for quantitative interpretation of release kinetics. The kinetics of dissolution were fitted to the zero-order, first-order, Higuchi, Hixson–Crowell, and Korsmeyer-Peppas models. The best model was selected based on the higher value of the correlation coefficient (R²) (Table S1). The CUR-NE followed the Higuchi release kinetics irrespective of pH, as the R² value was the highest compared to other evaluated models.

Fig. 2

In vitro release data from CUR-NE at pH = 7.4 and pH = 6. (A) Comparison of in vitro release between CUR-NE and CUR solution in PBS with 0.5% Tween 80 (pH = 7.4). (B) Comparison of in vitro release between CUR-NE and CUR solution in PBS with 0.5% Tween 80 (pH = 6). (C) Comparison of in vitro release of CUR from CUR-NE at pH = 6 and pH = 7.4. The data are expressed as mean values ± SD (n = 3). two-way ANOVA with Holm-Šídák’s multiple comparisons, **p < 0.005, ****p < 0.00005

Macrophages change their CD86 expression in response to CUR-NE uptake

Before proceeding to the in vitro and in vivo experiments, we performed an ATP-based CellTiter-Glo^Ⓡ luminescent assay to evaluate the viability of RAW 264.7 macrophages following exposure to CUR-NE, DF-NE, free CUR, and free CUR vehicle (DMSO) for 24 h. We did not observe cytotoxicity in RAW 264.7 macrophages when exposed to varying concentrations of NEs with or without CUR. However, a dose-dependent decrease in macrophage viability was observed in the presence of the CUR solution. The resulting toxicity in macrophages was due to CUR, as volume-matched DMSO, which was selected as the vehicle for dissolving CUR, did not contribute to any cytotoxic effects (Figure S2). We further qualitatively observed the uptake of both CUR-NE and DF-NE in macrophages using fluorescence microscopy. A strong green fluorescence signal was observed in macrophages. The DiR-labeled NEs appeared to be distributed around the DAPI-stained nucleus in the cytoplasm in punctate patterns (Fig. 3A). We quantified the uptake of NEs in activated macrophages using flow cytometry. Within 3 h of exposure time, ~ 98% of CUR-NE and DF-NE were taken up by activated macrophages (Fig. 3B). Next, we determined the effect of CUR-NE on CD86 expression (M₁-like macrophage marker), which is significantly expressed on RAW 264.7 macrophages post-LPS exposure. Treatment with CUR-NE did not activate macrophages toward an M₁-like phenotype even at a higher concentration (Fig. 3C). However, we observed a significant decrease in CD86⁺ expression when LPS-activated macrophages (M₁-like) were treated with CUR-NE. Although DF-NE, which was volume-matched to the highest concentration of CUR-NE, did not decrease CD86⁺ expression. There was a statistically significant difference between CUR-NE and CUR FD (Fig. 3C and D).

Fig. 3

Macrophage uptake and changes in CD86 expression. (A) Representative images of RAW 264.7 macrophages exposed to 40 $\:\mu\:$M of DiR-labeled CUR-NE and volume-matched DF-NE for 6 h and imaged for DAPI nuclei staining, Cy7 staining, and overlay of brightfield, DAPI, and Cy7. Scale bar: 20 $\:\mu\:$m. (B) Flowcytometry histogram plot comparison of cellular uptake of 40 $\:\mu\:$M CUR-NE and volume-matched DF-NE in LPS-activated RAW 264.7 macrophages. The percentage is shown as the mean ± SD. n = 3/group. (C) Representative density plots showing changes in CD86⁺ macrophages in response to treatments. (D) Quantitative changes in CD86⁺ expression in presence of varying doses of CUR-NE, CUR FD (40 $\:\mu\:$M) and DF-NE (volume-matched to highest concentration). The data is shown as the mean ± SD. n = 3/group independent cell culture, and 40,000 cells were counted. One-way ANOVA with Tukey’s multiple comparisons test. ns not significant, **p < 0.005, ****p < 0.00005. Data was analyzed using Attune NXT software

In vitro pharmacological responses

CUR-NE suppresses pro-inflammatory cytokine release from activated macrophages

Next, we investigated the anti-inflammatory properties of CUR-NE on LPS-activated macrophages by quantifying the decrease in the levels of proinflammatory cytokines, such as IL-$\:\beta\:$, IL-6, and TNF-α through ELISA assays. The levels of pro-inflammatory cytokines released from LPS-activated macrophages were significantly higher compared to the control group that was not stimulated by LPS. Treatment with CUR-NE resulted in a concentration-dependent decrease in the levels of all three pro-inflammatory cytokines. An approximately 74% decrease in TNF-α release was observed when treated with the highest concentration of CUR-NE. However, no significant decrease was observed with DF-NE (Fig. 4A). Furthermore, the highest concentration of CUR-NE resulted in approximately a 90% decrease in IL-6 release, while the lowest tested concentration (5 µM) and DF-NE did not result in a significant decrease in IL-6 (Fig. 4B). The highest concentration of CUR-NE resulted in approximately a 90% decrease in IL-$\:\beta\:$ release, whereas a 60% decrease in IL-$\:\beta\:$ release was observed at lower concentrations (5 µM) of CUR-NE. Interestingly, volume-matched DF-NE did decrease the release of IL-$\:\beta\:$. However, there was a significant difference between the extent of IL-$\:\beta\:$ suppression between CUR-NE and DF-NE. This observed superior effect with CUR-NE is attributed to the release of CUR from the NE (Fig. 4C). Next, we evaluated the effect of CUR-NE, volume-matched DF-NE, and CUR FD on the pro-inflammatory mediator nitric oxide by quantifying the total nitrite (NO₂⁻) and nitrate (NO₃⁻) production (Fig. 4D). We observed a dose-dependent decrease in NO production in the presence of CUR-NE, which was not observed with DF-NE.

For all three cytokines and in the NO assay, CUR FD (40 $\:\mu\:$M) was used as a positive control. During in vitro experiments, CUR FD was completely solubilized and was readily available to RAW 264.7 macrophages, which are maintained under static conditions. Additionally, it is shown that the degradation rate of CUR is slower in cell culture media with proteins compared to physiological conditions [44, 45].

Fig. 4

CUR-NE suppress the release of pro-inflammatory cytokines and a mediator from LPS-activated RAW 264.7 macrophages. Dose-dependent decrease in (A) TNF-α, (B) IL-6, (C) IL- $\:\beta\:$ released from LPS-activated macrophages exposed to varying doses of CUR-NE, DF-NE (volume-matched to the highest concentration of CUR-NE), and CUR FD (40 $\:\mu\:$M). (D) NO inhibition is quantified by measuring the decrease in the levels of nitrite and nitrate from LPS-activated macrophages treated with varying concentrations of CUR-NE, DF-NE (volume-matched to the highest concentration of CUR-NE), and CUR FD (40 $\:\mu\:$M). Each bar represents the mean ± SD. n = 3/group independent cell culture. One-way ANOVA with Tukey’s multiple comparisons test. ns not significant, *p < 0.05, **p < 0.005, ****p < 0.00005

CUR-NE increases anti-inflammatory IL-10 and protects against Ferroptosis

Treatment with CUR-NE showed a dose-dependent increase in IL-10, an anti-inflammatory cytokine, whereas DF-NE or CUR FD did not show an increase in IL-10 (Fig. 5A). Next, to investigate the effect of CUR-NE to protect anti-inflammatory macrophages from ferroptosis, IL-4-stimulated macrophages were simultaneously exposed to CUR-NE, DF-NE, CUR FD solution, or ferrostatin-1 (a positive control) along with RSL3, an inducer of ferroptosis. The cell viability data showed that, even at lower concentrations, ferrostatin-1, a synthetic radical-trapping antioxidant, prevented ferroptosis. CUR-NE prevented ferroptosis in a dose-response manner, with the highest dose achieving approximately 80% of cell viability. Treatment with volume-matched DF-NE and CUR FD did not improve cell viability compared to CUR-NE (Fig. 5B).

Fig. 5

CUR-NE increases the release of anti-inflammatory cytokine and protects anti-inflammatory macrophages from ferroptosis. (A) Dose-dependent increase in IL-10 from LPS-activated macrophages exposed to CUR-NE, DF-NE (volume-matched to the highest concentration of CUR-NE), and CUR FD (40 $\:\mu\:$M). Each bar represents mean ± SD. n = 3/group, independent cell culture, One-way ANOVA with Tukey’s multiple comparisons test. (B) Dose-response curve for CUR-NE, volume-matched DF-NE, CUR FD, and Ferrostatin-1. Macrophage viability was assessed post exposure to RSL3, an ferroptosis inducer. n = 6/group, ns not significant, **p < 0.005, ****p < 0.00005

CUR-NE reduces macrophage infiltration in Inflamed CFA-treated footpads

To assess the anti-inflammatory effects of CUR-NE in vivo, inflammation was induced in 6-week-old male and female Sprague Dawley rats by injecting 50 µL of CFA into the right hind paw. Two doses of CUR-NE (n = 3 rats) or DF-NE (n = 3 rats), each comprising 300 µL, were administered. The first dose was administered at the time of the insult, while the second dose was given 24 h after the CFA insult. Whole-body imaging was performed in vivo for 7 days (t = 168 h) and the fluorescence signal in the inflamed footpad was quantified (Fig. 6A and B).

Fig. 6

(A) Illustration depicting timeline of the in vivo experiment. (B) Representative NIRF images showing the accumulation of macrophages at the site of inflammation (hind right paw) during t = 24, 48, 96, 120, 144, and 168 h in male and in female rats, measured by detecting accumulation of theranostic CUR-NE or DF-NE

At baseline, no fluorescence intensity was observed in rat footpads, regardless of treatment group and sex. After injecting CFA in the footpad, i.v administration of NE results in an immediate increase in the fluorescence intensity in the CFA-injected footpad, irrespective of sex and treatment group (Fig. 7A and B). However, this immediate increase in fluorescence intensity as compared to the baseline is because of the enhanced permeability and retention (EPR) effect at the site of inflammation rather than macrophage uptake. This also helps to confirm the success of the tail vein injection, as a strong signal appears from the footpad rather than the rat tail [46]. At 24 h, the fluorescence signal increased in both sexes, but the signal increase was significantly higher for the DF-NE treatment group as compared to the CUR-NE group. It can be observed that at t = 48 h and throughout the duration of the study, the fluorescence signal intensity from the inflamed (CFA-injected) hind paw for the DF-NE group and CUR FD group was higher as compared to the CUR-NE treated group, regardless of sex (Fig. 7C and D).

Fig. 7

A-D) Quantification of the NIRF signal intensity from the ROI in CFA-injected footpads that received CUR-NE, DF-NE, and CUR FD followed by DF-NE at t = 0 h and t = 24 h to track infiltrating macrophages. The fluorescent intensities were quantified for 7 days (t = 0 h to t = 168 h) in both male and female rats. Data represents the average±SD, n = 3/treatment group. Two-way, repeat measures ANOVA, Holm-Šídák’s multiple comparisons test, ns not significant, *p < 0.05

Histopathological evaluation confirms alleviation of inflammation by CUR-NE but not DF-NE

Figure 8 shows cross sections from both male and female rat paws respectively (H&E, 4X) treated with either DF-NE or CUR-NE at 7 days (168 h) post CFA insult. The anti-inflammatory effects of CUR-NE were confirmed through histological analyses and immunofluorescence microscopy (Fig. 8A-F). We did not observe any signs of inflammatory responses in the contralateral rat foot pads regardless of sex (Fig. 8A and B). Minimal macrophages were observed post-staining with CD68, indicating normal levels in footpads. Following CFA injection (curved yellow arrows, 8 C, D), marked immune infiltration was observed on DF-NE treatment in both male and female rat paws. More pronounced edema was noted in male paws in the plantar zones (Fig. 8C) compared to female paws (Fig. 8D). Immunofluorescence staining for macrophages (CD68, green) and DF-NE (purple-pink), confirmed significant macrophage infiltration in the DF-NE treated male and female rats in the zones of plantar inflammation. Intracellular co-localization of DF-NE droplets was evident in macrophages. In comparison to DF-NE treatment, inflammation or edema was markedly less severe in intensity post treatment with CUR-NE irrespective of sex (Fig. 8E and F). Treatment with CUR-NE (Fig. 8E and F) showed scant macrophage infiltration in inflamed zones, and macrophage (CD68, green) and NE fluorescence intensity (purple-pink) were significantly lower with CUR-NE treatment compared to the DF-NE treatments (Fig. 8C and D). Together, these data corroborate the in vivo NIRF imaging findings (Fig. 7A-D) and showed that CUR-NE suppresses macrophage-mediated inflammation.

Fig. 8

H&E and immunostaining of inflamed paw sections collected at end-point from male and female rats treated with CXB-NE or DF-NE. Top Panel: Cross-section of normal male rat paw (A) and normal female rat paw, (B) showing plantar and dorsal aspects (Detail, H&E, 4X). The location of metatarsals (curved arrows) and the site of CFA injection into the plantar aspect (yellow curved arrow) are shown. Immunofluorescence staining shows nuclei (DAPI, blue) macrophages (CD68 staining, green) and NE (purple-pink). Middle Panel: Cross-section of male rat paw, (C) and female rat paw (D) following CFA injection and DF-NE treatment. Note the significant immune infiltration in the plantar zone (shown in white boxed insets) in both specimens, with edema most pronounced in the male paw sample. Immunofluorescence staining reveals co-localization of DFNE (purple-pink) macrophages (CD68 staining, green) following CFA injection. Note the zones of significant macrophage infiltration in both samples and the correlation of DFNE distribution with histopathologic presentation. The localization of DF-NE with macrophage infiltration is most pronounced in the deep dermal and fascial planes of foot. (4X and 20X). Lower Panel: Cross-section of male rat paw E) and female rat paw, F) following CFA injection and CUR-NE treatment. There is a marked reduction in immune infiltration in the plantar zones (shown in white boxed insets) in both specimens with edema scant or absent in paw samples. Immunofluorescence staining reveals co-localization of CUR-NE (purple-pink) with macrophages (CD68 staining, green) in inflammatory zones following CFA injection. Note that the intensity and distribution of macrophage infiltration in both samples are markedly reduced/different from DF-NE samples and correlate well with histopathologic presentation on H&E (4X and 20X). Scale bar: 100 $\:\mu\:$M.

CUR-NE localizes in CD40+CD68+ macrophages in Inflamed tissue

We performed further immunofluorescence to confirm colocalization of CUR-NE and DF-NE with CD40⁺CD68⁺ macrophages (M₁-like macrophages) in the male rat (Fig. 9A and B) and female rat (Fig. 9C and D). CD40⁺CD68⁺ macrophages were abundantly present in the CFA-injected footpads owing to the inflammatory reaction caused by CFA. However, CD40⁺CD68⁺ macrophages were not present in the contralateral footpads (Figures S3A-D). Our results are consistent with our previous publications, showing CD68⁺ macrophages preferentially take up the NE droplets < 150 nm in the hind foot and are detectable using the NIRF signal on histology [37].

Fig. 9

Co-localization of NEs with or without CUR with CD68⁺CD40⁺ (M₁-like) macrophages. A-B) Confocal micrograph images of CFA-injected footpad from male rat were taken to show colocalization of the DiI signal from the injected NEs and CD68⁺CD40⁺ macrophages. C-D) Confocal micrograph images of CFA-injected footpad from female rat were taken to show colocalization of the DiI signal from the injected NEs and CD68⁺CD40⁺ macrophages. Merged images show DAPI nuclear counterstain in blue. white boxes show the magnified area highlighted by white boxes, with DiI⁺CD68⁺ macrophages (white arrow) from each stain. Scale bar: 100$\:\mu\:$M

CUR-NE localizes to inflamed regions with minimal off-target accumulation

At the end of the study, ex vivo NIRF imaging technique was used to quantify the fluorescent signal per gram of organ weight in male and female rats administered CUR-NE or DF-NE. We observed that the fluorescent signal in the heart, liver, spleen, and kidneys was not significantly different between the treatments for both male and female rats (Fig. 10A and B). This biodistribution study aligned with our previous findings showing that liver and spleen accumulate theranostic NE droplets compared to the other non-targeted organs, due to their role in the reticuloendothelial system (RES) [8]. The fluorescence signal intensity from CFA-injected footpad (inflamed) was significantly higher than the fluorescence signal intensity from normal footpad (contralateral side). This shows that NEs did not produce non-specific signal, and the signal was localized only at the site of inflammation due to macrophages infiltration (Fig. 10C and D).

Fig. 10

Biodistribution of CUR-NE and DF-NE was evaluated through ex vivo imaging at day 7. (A) Fluorescence signal quantification per gram of organ weight for indicated organs in male rats. (B) Fluorescence signal quantification per gram of organ weight for indicated organs in female rats. (C) Comparison of fluorescence signal intensity from CFA-injected footpads (inflamed) verses normal footpad (contralateral) in male rats. (D) Comparison of fluorescence signal intensity from CFA-injected footpads (inflamed) verses normal footpad (contralateral) in female rats. unpaired t-test, n = 3/group, ns not significant, **p < 0.005, ***p < 0.0005, ****p < 0.00005

Discussion

In this study, we formulated a theranostic CUR-NE and tested its anti-inflammatory efficacy and ability to effectively decrease macrophage infiltration at the site of inflammation. The theranostic NE was developed using insights from a previously reported comprehensive study, which identified the critical material attributes and critical processing parameters that govern the colloidal and fluorescent stability of theranostic NE containing a high concentration (30% w/v) of perfluorocarbon and a NIRF dye [34]. The average droplet diameter of the formulated NEs with or without CUR was less than 200 nm. According to studies performed in various inflammatory rodent models, nanotherapeutics ranging in size from 100 nm to 200 nm are effectively phagocytosed by activated macrophages without further triggering immune activation [8, 30, 47]. Next, our findings from colloidal stability data indicated the suitability of NEs to be tested further in in vitro and in vivo studies. A major advantage of these NEs is that they are designed to have NIRF dye as a diagnostic component for non-invasive tracking of macrophage infiltration patterns in vivo. Despite the ability of CUR to quench signals from NIRF dyes [34] and other fluorescent nanomaterials [48], both the NEs with and without CUR had comparable DiR signal intensities. This implies that the changes observed during real-time in vivo NIRF imaging reflect the effect of CUR-NE rather than the mismatch in NIRF signal intensities between CUR-NE and DF-NE. Our in vitro release data suggested sustained release of CUR from the NE scaffold without any initial burst release as compared to the CUR solution. As compared to the physiological pH, CUR release was significantly higher in acidic conditions, typically observed in inflamed tissue. This could be attributed to the stability of CUR at low pH compared to the neutral pH. Mathematical modeling of drug release data showed that the CUR release from NE was primarily governed by principles of drug diffusion, which was in accordance with other studies reported in NE literature [49, 50].

For in vitro studies, LPS was used to activate RAW 264.7 macrophages into an M₁-like phenotype for simulating inflammatory environments. Our flow cytometry data showed increased expression of CD86, a surface marker of the M₁-like phenotype, in the presence of LPS. However, treatments with varying non-toxic doses of CUR-NE decreased expression of CD86⁺ macrophages, showing the anti-inflammatory potential of CUR-NE. This is in line with other reports showing decreased CD86⁺ macrophage population as the concentration of CUR solution was increased from 6.25 µmol/L to 25 µmol/L [51]. To further study the pharmacological effect of CUR-NE, we assessed the levels of pro-inflammatory and anti-inflammatory cytokines, which directly correlate with the extent of inflammatory insult in several diseases [52,53,54]. Increase in the levels of pro-inflammatory cytokines, such as IL-6, TNF- $\:\alpha\:$, and IL-$\:\beta\:$, are linked to increased mechanical allodynia and pain-related behaviors [55,56,57]. Thus, suppressing the levels of pro-inflammatory cytokines decreases inflammation and associated inflammatory pain. Consistent with this idea, we showed that CUR-NE can effectively decrease the release of pro-inflammatory cytokines, like IL-6, TNF- $\:\alpha\:$, and IL-$\:\beta\:$ from activated macrophages in a dose-dependent manner. Additionally, we observed increased levels of IL-10 in LPS-activated macrophages due to treatment with CUR-NE. IL-10 exerts anti-inflammatory effects by reprogramming macrophage metabolism and suppressing mammalian target of rapamycin complex 1 (mTORC1) activation [58]. Nitric oxide production is also increased in RAW 264.7 macrophages in response to LPS challenge due to P2Z/P2X7 receptor activation [59]. Chen, Y., et al. [60] reported increased production of nitric oxide synthase (NOS) following CFA injection in the plantar region of the rat footpad. However, the authors showed that administration of an NOS inhibitor decreased NO output and consequently alleviated the inflammatory thermal hyperalgesia. Our data showed that CUR-NE at the tested high concentrations successfully decreased the levels of NO output in LPS-activated macrophages, further supporting our observations that CUR-NE has anti-inflammatory effects. The release of pro-inflammatory cytokines by macrophages can induce ferroptosis in both sterile and non-sterile inflammatory [61] as well as autoimmune states [62]. Furthermore, ferroptotic cells exert strong immunogenicity owing to the increased release of damage-associated molecular pattern molecules and inflammatory cytokines, which promotes the cellular environment to be in a pro-inflammatory state [63]. Our in vitro findings indicated that CUR-NE inhibits ferroptosis in anti-inflammatory macrophages (M₂-like), showing promise as an anti-ferroptosis therapy to treat inflammation. We anticipate that by inhibiting ferroptosis in M₂-like macrophages, which are more susceptible to ferroptosis-induced cell death, would support resolution of tissue inflammation through mechanisms like sustained anti-inflammatory cytokine release.

We selected a CFA-injected rat model to evaluate whether treatment with CUR-NE would decrease the macrophage infiltration in vivo. Additionally, macrophage infiltration dynamics and experienced analgesia are reported to differ between male and female sex [8, 64, 65]. Therefore, this study was performed in male and female rats. Throughout the length of the study, we found that treatment with CUR-NE significantly reduced macrophage infiltration at the site of inflammation as compared to DF-NE and CUR FD. This was reflected by decreased NIRF signal intensities measured from the CFA-injected footpad. As expected, CUR FD did not decrease macrophage infiltration at the inflamed site as it undergoes rapid elimination (within hours) and is majorly excreted in bile in the form of its glucuronide metabolites [66, 67]. This highlights the importance of encapsulating CUR in a theranostic nanoemulsion platform for achieving in vivo anti-inflammatory effects. We further confirmed this observation with our qualitative immunohistology data, which suggested that the number of CD68⁺ macrophages decreased in excised tissues in the presence of CUR-NE as compared to DF-NE, irrespective of sex. This decrease in tissue macrophages corresponded with the inflammation resolution at the site of CFA footpad injection. Further, previous studies in different pain models have shown that a decrease in macrophage infiltration at the injured site coincides with improved pain behaviors, leading to sustained pain relief [8, 30, 32]. Importantly, these anti-inflammatory effects were observed with two low doses of NE (300 µL/injection), reaching the overall dose of ~ 2 mg of CUR/kg of rat. This represents a significant reduction in the required curcumin dose as compared to other rodent studies, where curcumin was given orally at 150–500 mg/kg or i.v. 40 mg/kg [68, 69].

We acknowledge limitations in our study that prevented us from drawing more substantive conclusions. Sex, as a biological variable, plays an important role in inflammation [70]. However, this study was not powered to statistically compare and comment on the sex-related differences observed in the macrophage infiltration patterns in the treatment groups. Such in-depth molecular investigations concerning these sex differences will be part of further studies.

Conclusions

In this study, we demonstrated the in vitro and in vivo efficacy of theranostic CUR-NE in targeting inflammation, specifically within LPS-activated macrophages and a CFA-induced inflammatory rodent model. CUR-NE effectively reduced the expression of the CD86 macrophage marker, a hallmark of M₁-like macrophage activation, while simultaneously promoting a dose-dependent decrease in pro-inflammatory cytokines and an increase in the anti-inflammatory cytokine IL-10. Notably, CUR-NE administration significantly reduced macrophage infiltration at the site of inflammation, as evidenced by diminished NIRF signals in the CFA-treated footpad, correlating with reduced local inflammation. Beyond addressing localized inflammation, the insights gleaned from this study pave the way for the development of scalable and clinically robust nanomedicine designed to modulate macrophage behavior without compromising the overall immune system, such as autoimmune diseases driven by M₁-like macrophage infiltration. This represents a step closer to precision medicine strategies for immune modulation.