Surface saturation of drug-loaded hollow manganese dioxide nanoparticles with human serum albumin for treating rheumatoid arthritis

Abstract

Rheumatoid arthritis (RA) is a chronic inflammatory joint disease accompanied by energy depletion and accumulation of reactive oxygen species (ROS). Inorganic nanoparticles (NPs) offer great promise for the treatment of RA because they mostly have functions beyond being drug carriers. However, conventional nanomaterials become coated with a protein corona (PC) or lose their cargo prematurely in vivo, reducing their therapeutic efficacy. To avoid these problems, we loaded methotrexate (MTX) into hollow structured manganese dioxide nanoparticles (H-MnO₂ NPs), then coated them with a ‘pseudo-corona’ of human serum albumin (HSA) at physiological concentrations to obtain HSA-MnO₂@MTX NPs. Efficacy of MTX, MnO₂@MTX, and HSA-MnO₂@MTX NPs was compared in vitro and in vivo. Compared to MnO₂@MTX, HSA-coated NPs were taken up better by lipopolysaccharide-activated RAW264.7 and were more effective at lowering levels of pro-inflammatory cytokines and preventing ROS accumulation. HSA-MnO₂@MTX NPs were also more efficient at blocking the proliferation and migration of fibroblast-like synoviocytes from rats with collagen-induced arthritis. In this rat model, HSA-MnO₂@MTX NPs showed better biodistribution than other treatments, specifically targeting the ankle joint. Furthermore, HSA-MnO₂@MTX NPs reduced swelling in the paw, regulated pro-inflammatory cytokine production, and limited cartilage degradation and signs of inflammation. These results establish the therapeutic potential of HSA-MnO₂@MTX NPs against RA.

1. Introduction

Rheumatoid arthritis (RA) is a multi-factorial systemic, inflammatory and progressive autoimmune disease that involves immune cell infiltration, synovial tissue hyperplasia and bone destruction (Mueller et al., 2021; Wu et al., 2021a; Li et al., 2022a). Current treatments against RA include disease-modifying anti-rheumatic drugs (DMARDs), non-steroidal anti-inflammatory drugs, and glucocorticoids, as well as biologics (Zhao et al., 2021a). Among these, DMARDs may be the most effective (Radu and Bungau, 2021). In particular, low-dose methotrexate (MTX), a second-line DMARD, reduces symptoms and improves bone quality by inhibiting the synthesis of purines and pyrimidines, promoting the release of adenosine, and controlling inflammatory cytokines (Katturajan et al., 2021; Wang et al., 2022a). As a folic acid antagonist, MTX exerts anti-proliferative and pro-apoptotic effects (Wood and Wu, 2015). However, poor patient compliance, nonspecific distribution, and high risk of gastrointestinal side effects limit the efficacy of MTX as well as other anti-RA drugs (Wang et al., 2018; García-González and Baker, 2022).

Nanobiotechnology has begun to offer ways to overcome these challenges (Wang et al., 2022b). Many novel delivery systems for MTX have been developed to improve its efficacy and safety, such as liposomes, dendrimers, endogenous carrier materials (albumin, red blood cells), and multifunctional nanoparticles (NPs; Abolmaali et al., 2013; Li et al., 2022b). Such NPs are particularly attractive because they can be engineered to address the disease’s multiple pathophysiological pathways (Zheng et al., 2021; Zhao et al., 2021b).

One disease pathway is oxidative stress (da Fonseca et al., 2019): RA is associated with higher levels of reactive oxygen species (ROS), which damage bone and cartilage and promote inflammatory cytokine production, which exacerbates the damage (Phull et al., 2018). Therefore, elevated ROS levels may serve as a trigger for releasing drugs specifically in tissues affected by RA.

Hollow-structured manganese dioxide nanoparticles (H-MnO₂ NPs) can not only carry drugs such as MTX but also scavenge ROS. In addition, they can release their drug cargo selectively in ROS-rich, acidic environments such as RA tissue, after which the MnO₂ can be converted to nontoxic Mn²⁺ and rapidly cleared by the kidneys (Jia et al., 2021; Wang et al., 2021; Li et al., 2021a; Wu et al., 2021b; Qiu et al., 2022a). Despite this potential, H-MnO₂ NPs face efficacy and safety hurdles before they can be used in the clinic (Ajdary et al., 2018; Ravindran et al., 2018).

One obstacle is that nanomaterials have a high surface free energy due to their large relative surface area. Thus, proteins and other macromolecules in the blood easily adsorb onto their surface to form a protein corona (PC) through electrostatic adsorption, non-covalent interactions between sulfhydryl groups, and hydrophobic interactions (Docter et al., 2015; Cai and Chen, 2019). Immunoglobulins, complement proteins, and plasma fibrinogen in the PC send an ‘eat me’ signal to the mononuclear phagocytic system (MPS), which clears the NPs (Sobczynski and Eniola-Adefeso, 2017; de Castro et al., 2021). Therefore, saturating potential surface binding sites on NPs might reduce such clearance and thereby ensure efficient drug delivery.

One strategy may be to coat the NPs with an endogenous protein before injecting them into the body. One of the most promising candidates is albumin, which accounts for 40%–60% of total protein in plasma, where its concentration is 35–46 mg/mL (Amano et al., 2020; Sheinenzon et al., 2021). Albumin has been widely used in nanomaterials as a carrier and scaffold (An and Zhang, 2017), as well as a surface coating to reduce NP clearance from the blood (Spada et al., 2021). Albumin may offer an additional advantage to drug-delivering NPs in the context of RA: albumin, as a major source of energy and amino acids, tends to accumulate in tissues with high energy demand (Lamichhane and Lee, 2020; Zheng et al., 2022), and inflamed joints in RA show high energy demand because of inflammatory responses mediated by fibroblast-like synoviocytes (FLS) and classically activated macrophages (Loftus and Finlay, 2016; Asif Amin et al., 2017; Yang et al., 2020; Chang and Kan, 2021; Lu et al., 2022). In parallel, FLS invade joints and proliferate excessively there (Wu et al., 2021c). Thus, coating NPs with albumin may help target the particles to inflamed joints in RA patients. Furthermore, the extracellular matrix of inflamed joints contains abnormally high levels of ‘secreted protein acidic and rich in cysteine’ (SPARC), which binds strongly to albumin (Zheng et al., 2022).

Here we loaded MTX into H-MnO₂ NPs and then saturated surface binding sites with human serum albumin (HSA) in an effort to maximize the accumulation of NPs in inflamed tissue, where they could release MTX and reduce ROS (Figure 1). We compared the therapeutic efficacy of these precoated NPs with naked NPs and free MTX in vitro and in vivo.

Figure 1.

Schematic diagram of preparation and therapeutic mechanism of HSA-MnO₂@MTX NPs against rheumatoid arthritis. CIA: collagen-induced arthritis; FLS: fibroblast-like synoviocytes; HIF-1α: hypoxia-inducible factor-1α; H₂O₂: hydrogen peroxide; IL: interleukin; HSA: human serum albumin; H-MnO₂: hollow structured manganese dioxide; MTX: methotrexate; NP: nanoparticle; ROS: reactive oxygen species; TNF-α: tumor necrosis factor-α.

2. Methods and materials

2.1. Materials

2′,7′-Dichlorofluorescin diacetate (DCFH-DA) and 4′,6-diamidino-2-phenylindole (DAPI) were obtained from Solarbio Science & Technology (Beijing, China). Tetraethyl orthosilicate (TEOS, ≥99.99%) was purchased from Macklin (Shanghai, China). 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was purchased from BioFroxx (Guangzhou, China). The 4-chlorobenzenesulfonate salt of 1,1′-dioctadecyl-3,3,3′,3′-tetramethyl indodicarbocyanine (DID) was purchased from Beyotime (Shanghai, China). MTX (98%) disodium salt was purchased from Yuanye Bio-Technology (Shanghai, China). Immunization-grade bovine type II collagen as well as complete and incomplete Freund’s adjuvant were obtained from Chondrex (Woodinville, WA, USA).

2.2. Cell culture

The mouse macrophage cell line RAW264.7 (Number: CTCC-001-0048) was obtained from the Meisen Chinese Tissue Culture Collections (Zhejiang, China) and cultured in Dulbecco’s modified Eagle’s medium (Gibco, USA) supplemented with 10% fetal bovine serum (Sijiqing, Huzhou, China) and 1% (v/v) penicillin/streptomycin (Solarbio) at 37 °C in a humidified atmosphere containing 5% CO₂.

FLS from a rat model of collagen-induced arthritis (CIA-FLS, Number: BNCC359259) were purchased from the BeNa Culture Collection (Henan, China) and cultured in Roswell Park Memorial Institute 1640 (RPMI 1640; Gibco) supplemented with 10% fetal bovine serum (Sijiqing) and 1% (v/v) penicillin/streptomycin (Solarbio) at 37 °C in an atmosphere containing 5% CO₂.

2.3. Animal research information

2.3.1. Ethical approval statement

The authors have adhered to the ARRIVE guidelines. All experiments were performed in accordance with the Regulations of the Experimental Animal Administration, issued by the State Committee of Science and Technology of the People’s Republic of China (November 14, 1988). Ethical approval to conduct the study was obtained from the Animal Care and Ethics Committee of Southwest Medical University (ethical approval No. 20211123-002).

2.3.2. Animal model

Collagen-induced arthritis (CIA) animal model was necessary for pharmacodynamic evaluation in vivo in this experiment, which has the pathological manifestations closest to RA in humans and could not be replaced with any other sample. Male Sprague–Dawley rats were chosen as the model animals based on the previous research results of our group, which method was perfect and had a high success rate and stronger survival ability during the experiment.

2.3.3. Other statement

Male Sprague–Dawley rats (n = 38) and BALB/c (n = 10) mice were purchased from the Laboratory Animal Center of Southwest Medical University (Luzhou, China), which is approved by the Animal Care and Ethics Committee of Southwest Medical University. All animals were kept in a temperature-controlled environment, with free access to food and water, and a light/dark cycle of 12/12 h.

Ming Jia, Minrui Wang, Yan Liu, and Chenglong Wang were involved in conducting the entire animal experiment in vivo. The induction experiment of CIA model began on November 30, 2021, and the starting time of other in vivo experiments was calculated from this starting point.

2.3.4. Animal anesthesia

Rats and mice were anesthetized in the same way. Each rat or mouse before the procedure was restrained for a two-person injection technique and anesthetized by a one-time intraperitoneal injection (refer to Section 2.3.6) at 40 mg/kg sodium pentobarbital.

2.3.5. Animal blood collection

After anesthesia, rats or mice were fixed in a supine position and the fur of the chest area was cut off. After skin disinfection, needles were inserted into the heart with a syringe at the place where the heartbeat was strongest at the third to the fourth intercostal area on the left side. The syringe is slowly drawn out to encourage blood to enter the syringe and transfer to the collection vessel after collection. Whole blood samples were centrifuged for 10 min at 2500 rpm to obtain upper plasma samples.

2.3.6. Animal euthanasia

Humane endpoint or experimental terminative indicator: (1) the animal is unable to stand without anesthesia or sedation; (2) animal weight reduction of 15%–20%; (3) the animal completely loses appetite for 5 days or desire to drink for 3 days.

Rats or mice at the humane endpoint were euthanized in the same way. Details are as follows: each rat or mouse was restrained for a two-person injection technique and euthanized by intraperitoneal injection at 200 mg/kg sodium pentobarbital. Rats were held in dorsal recumbency at an approximately 30° angle (head lowermost) by one person who supported each rat and restrained the left pelvic limb. The other person restrained the rat’s right pelvic limb and injected it. Each injection was performed in the right caudal quadrant of the abdomen at the level of the coxofemoral joint and approximately 5 mm to the right of the midline, while the needle was directed cranially at a 45° angle to the body wall. Death was confirmed by identifying cessation of heartbeat.

2.4. Synthesis of HSA-MnO₂@MTX NPs

Briefly, 3–7 mL of TEOS were added to a mixture of anhydrous ethanol, ultrapure water, and ammonia, and the mixtures were incubated at 45, 50, 55 or 60 °C to obtain monodisperse silica (SiO₂) NPs. SiO₂ NPs were added as ‘templates’ to aqueous solutions of KMnO₄ at mass ratios of 1: 7.5, stirred at room temperature for 6 h, and left standing overnight. The precipitates were collected by centrifugation and transferred to 100 mL 2 M Na₂CO₃ at 60 °C, and the temperature was maintained for 12 h, leading to H-MnO₂ NPs (Pan et al., 2019).

HSA-MnO₂@MTX NPs were prepared by stirring MTX and H-MnO₂ at mass ratios of 1: 3, 1: 1, or 3: 1 for 12 h at room temperature, then the precipitate was collected by centrifugation and immediately dispersed in HSA (40 mg/mL) and incubated with oscillation on a shaker for 6 h at 37 °C. To prepare NPs loaded with fluorescent dye, MTX was replaced with DID in this procedure, and all steps were performed in the dark.

After the above preparation, mice plasma (PM) extracted from BALB/c mice (method referred to Section 2.3.5) was further added and incubated at 37 °C to prepare PM + MnO₂@MTX, PM + MnO₂@DID, PM + HSA-MnO₂@MTX or PM + HSA-MnO₂@DID NPs.

2.5. Characterization of HSA-MnO₂@MTX NPs

The hydrodynamic size and ζ potential of HSA-MnO₂@MTX NPs were measured by dynamic light scattering (Malvern Zetasizer Nano ZS, UK), while their morphology was observed using transmission electron microscopy (TEM, JEM1200EX, Japan). Ultraviolet-visible absorption spectra were obtained using a UV-2600 spectrophotometer (Shimadzu, Japan). The encapsulation efficiency (EE) and loading capacity (LC) for MTX were determined by high-performance liquid chromatography (LC-2030, Shimadzu). EE was calculated as EE (%) = (W₂ – W₁)/W₂ × 100% and LC was calculated as LC (%) = (W₂ – W₁)/(W₂ – W₁ + W₃) × 100%, where W₁ was the weight of free MTX; W₂, the weight of total MTX; and W₃, the weight of H-MnO₂ NPs.

To analyze the profile of MTX release from NPs, 1 mL HSA-MnO₂@MTX NPs (1.5 mg/mL) was added to dialysis bags with a molecular weight cutoff of 1000 Da, which were submerged in 50 mL of phosphate-buffered saline (PBS) at pH 5.5 (with 10 mM H₂O₂) or pH 7.4 at 37 °C. At predetermined time points (0.5, 1, 2, 5, 6, 7, 8, 10, 24, and 48 h), a sample of dialysate (500 µL) was collected and replaced with an equal amount of fresh medium. The concentration of MTX was quantified by high-performance liquid chromatography.

2.6. NP uptake by lipopolysaccharide (LPS)-activated RAW264.7

LPS-activated RAW264.7 as the classically activated macrophage model. RAW264.7 cells were seeded into 24-well plates and cultured for 24 h in the absence or presence of 5 µg/mL LPS (Solarbio). Next, cells were incubated for 1–2 h with NPs in which MTX was replaced by DID (the cultures from LPS-RAW264.7 were then incubated for 25 min in the dark at room temperature with 200 µL of fluorescein isothiocyanate-labeled phalloidin (50 nM; Solarbio) per well. Next, the cells were fixed with paraformaldehyde. Nuclei were stained with DAPI, then NP internalization was assessed by confocal laser scanning microscopy (SP8, Leica, Germany).

LPS-activated RAW264.7 were treated with HSA-MnO₂@MTX NPs (20 µg/mL) for 4 h, washed with PBS, and then pre-fixed for 0.5 h at 4 °C in a small amount of 2.5% glutaraldehyde. The cells were scraped off and centrifuged into a pellet, which was fixed overnight at 4 °C in glutaraldehyde. Next, the cells were fixed with 1% osmium tetroxide, dehydrated in increasing concentrations of acetone, embedded in epoxy resin, and thin-sectioned. Sections were examined using TEM (JEM-1400FLASH, Japan) at the LiLai Biomedical Experiment Center (Chengdu, China).

2.7. Cytotoxicity of NPs

RAW264.7 were seeded into 96-well plates and incubated for 24 h with NPs containing different MTX concentrations. Subsequently, the medium was replaced by 200 µL of MTT solution (0.5 mg/mL) for 4 h at 37 °C. MTT solution was discarded and 150 µL of Dimethyl sulfoxide (DMSO) was added to each well, followed by shaking for 15 min at 37 °C. Absorbance at 490 nm was measured using a microplate reader (BioTek Cytation 5, USA). Cytotoxicity of MTX or HSA-MnO₂@MTX NPs against CIA-FLS was assessed using a similar method.

2.8. Intracellular ROS levels

RAW264.7 cells were seeded into 12-well plates and incubated with 5 µg/mL LPS for 24 h. The culture medium was replaced by different formulations, all of which contained the equivalent of 20 µg/mL MTX. After incubation for 2 h, 10 µL of 10 mM DCFH-DA was added to each well, followed by incubation for 30 min. Fluorescence images were captured using a fluorescence microscope (Olympus, Japan).

In assays of intracellular H₂O₂, LPS-activated RAW264.7 were incubated with HSA-MnO₂@MTX NPs (20 µg/mL MTX) for 4 h at 37 °C, cells were harvested and sonicated, and H₂O₂ was assayed using a commercial kit (Solarbio, kit number: BC3595). The cells were collected and 1 mL of acetone was added. Next, ultrasound was performed at 200 W for 3 s to break up the cells, which process was repeated 30 times. The supernatant was collected by centrifugation at 8000×g at 4 °C for 10 min. Then, each reagent was added in sequence, while standard and blank tubes were set as required. After standing at room temperature for 5 min, 200 µL of each sample were inoculated into 48-well plates and the absorbance was determined at 415 nm by using a microplate reader (BioTek Cytation 5, USA). Calculate the H₂O₂ content according to the following formula:

$$\begin{array}{l}{H}_2{O}_2\text{concentration\hspace{0.17em}}(\text{mmol}/\text{cell})=\\ 0.002\times \left(A_{\text{sample}}-A_{\text{blank}}\right)/\left(A_{\text{standard}}-A_{\text{blank}}\right)\times {10}^{-4}\end{array}$$

2.9. Intracellular levels of inflammatory cytokines

RAW264.7 were seeded into 24-well plates and incubated with 5 µg/mL LPS for 24 h. The culture medium was replaced for 24 h by different formulations, all of which contained the equivalent of 20 µg/mL MTX. Medium was collected and assayed for the inflammatory cytokines TNF-α (kit number: AD2726Mo), IL-1β (kit number: AD3364Mo), and IL-10 (kit number: AD2837Mo) as well as hypoxia-inducible factor 1α (HIF-1α; kit number: AD3185Mo) using commercial enzyme-linked immunosorbent assays (ELISAs; Andy Gene, Beijing, China). Each sample to be tested was inoculated in a 48-well plate, followed by warming, liquid mixing, washing, enzyme addition, color developing, termination, and other processes, while the standard sample concentration gradient and blank control group were set in the well plate. Finally, the absorbance of each hole was determined at 450 nm by using a microplate reader (BioTek Cytation 5, USA). Draw the standard curve with the concentration and absorbance value of the standard substance and calculate the corresponding concentration of each sample, and then multiply by the corresponding dilution times to get the actual concentration of the sample.

2.10. Migration assay

For the wound scratch assay, CIA-FLS were cultured at a density of 2 × 10⁵ cells per well in 6-well plates and grown to confluence. The monolayer was scratched using a 200-µL pipette tip and washed with PBS to remove detached cells. Then, the cells were cultured for 24 h in serum-free medium with MTX or HSA-MnO₂@MTX NPs (20 µg/mL MTX). Untreated CIA-FLS served as control. CIA-FLS were imaged 0 and 24 h later under a CKX3-SLP microscope (Olympus).

2.11. Cell apoptosis

CIA-FLS were seeded at 2 × 10⁵ cells per well in 6-well plates and grown to confluence. The culture medium was replaced by MTX or HSA-MnO₂@MTX NPs (20 µg/mL MTX) for 8 h. Viable and dead cells were detected using a Calcein-AM/PI kit (Solarbio, Beijing, China, kit number: CA1630). The cells were collected, rinsed twice with 1× assay buffer, resuspended in 1 mL, incubated for 25 min at 37 °C with Calcein-AM (1–2 µL), then for 5 min at 37 °C with propidium Iodide (PI; 3–5 µL). Mitochondrial membrane potential was analyzed using a JC-1 kit (Solarbio, Beijing, China, kit number: M8650). The cells were mixed with 1 mL of JC-1 staining solution and incubated at 37 °C for 20 min. The supernatant was removed by aspiration, and cells were washed twice with cold 1× JC-1 staining buffer. Then 2 mL of cell culture medium was added. Fluorescence images of all stained samples were taken with an inverted fluorescence microscope (Olympus).

2.12. Rat model of collagen-induced arthritis

The rat model of CIA (n = 30) was established. Details are as follows: bovine type II collagen was emulsified with an equal volume of complete Freund’s adjuvant, and 100 µL was injected subcutaneously in the tail of rats. Seven days later, 100 µL of an emulsion containing equal amounts of incomplete Freund’s adjuvant and bovine type II collagen was injected subcutaneously into the tail of rats (Wang et al., 2022a). Swelling and erythema were observed in the ankle and paw on day 14. Healthy rats (n = 8) as the negative control did not undergo any manipulations.

2.13. Biodistribution

Healthy (n = 3) and CIA (n = 3) rats were randomly selected from Section 2.12. The expression of SPARC in the ankle synovium of healthy and CIA rats was detected by immunohistochemical staining at the LiLai Biomedical Experiment Center (Chengdu, China).

On day 15 after arthritis induction, CIA rats (n = 6) were randomly selected from Section 2.12 and divided into two groups (n = 3 per group)). All the rats were deprived of food and water, and stripped of their fur from the waist down. The next day, the rats in the two groups were injected with free DiD and HSA-MnO₂@DID which contain equal amounts of 5 µg DiD through the tail vein, respectively. After the rats were anesthesia (method refers to Section 2.3.4) at 1, 4, 7, and 10 h, they were placed in supine position on the imaging platform to observe the distribution and intensity of fluorescence in the rats by the Carestream MI system (Gel Logic 6000 PRO, USA).

The rats were euthanized (method refers to Section 2.3.6) after completion of above process, and the heart, liver, spleen, lung, kidney, and ankle joint were separated. Remove excess blood from tissues and organs by rinsing them with saline, then use filter paper to remove water stains and place them in a utensil. Finally, the distribution of fluorescence in tissues and organs of rats was observed and recorded by the Carestream MI system.

2.14. Measurement of hemolysis

Whole blood samples were collected from CIA rats (n = 3, randomly selected from Section 2.12) on day 17 after arthritis induction (method refers to Section 2.3.5). Three times the volume of normal saline was added to the collected whole blood, mixed, and centrifuged at 2500 rpm for 10 min. Discard the supernatant and repeat 3 times. Finally, a total of 4 drops of overstocked red blood cells were absorbed with a pasteurized dropper and transferred to 8 mL saline to obtain 2% red blood cell suspension.

A 2% red blood cell suspension was incubated with water, saline, or different concentrations of NP preparations at 37 °C for 4 h. Absorbance at 540 nm was quantified using a microplate reader. The hemolysis rate was calculated using the formula (Lyu et al., 2021).

$$\text{Hemolysis\hspace{0.17em}ratio\hspace{0.17em}}\left(\text{\%}\right)=\frac{\left[A\left(\text{sample}\right)‐A\left(\text{saline}\right)\right]}{\left[A\left(\text{water}\right)‐A\left(\text{saline}\right)\right]}\times 100\text{\%}$$

where A indicates absorbance.

2.15. Therapeutic index monitoring

CIA rats (n = 12) were randomly selected from Section 2.12 and divided into four groups (n = 3 per group): saline, MTX, MnO₂@MTX, and HSA-MnO₂@MTX. On days 16, 19, 22, 25, and 28 after arthritis induction, 200 µL of saline, free MTX, MnO₂@MTX, or HSA-MnO₂@MTX were injected once through the tail vein at the same MTX concentration of 1.0 mg/kg (power calculations refer to Section 2.5). Three healthy rats without any treatment served (n = 3) as negative control group and the group of saline was used as positive control group. During the treatment period, ankle diameter, paw thickness, and paw volume were measured every 3 days. Ankle joints were photographed after the treatment was completed.

2.16. Measurement of inflammatory cytokines in vivo

Plasma samples were collected from rats (same batch as Section 2.15) on day 30 after CIA induction (method refers to Section 2.3.5), and serum levels of TNF-α (Kit number: AD3238Ra), IL-6 (Kit number: AD3249Ra), HIF-1α (Kit number: E-30270) and IL-10 (Kit number: AD3254Ra) were measured using commercial ELISAs (Andy Gene). Each sample to be tested was inoculated in a 48-well plate, followed by warming, liquid mixing, washing, enzyme addition, color developing, termination, and other processes, while the standard sample concentration gradient and blank control group were set in the well plate. Finally, the absorbance of each hole was determined at 450 nm by using a microplate reader (BioTek Cytation 5, USA). Draw the standard curve with the concentration and absorbance value of the standard substance and calculate the corresponding concentration of each sample, and then multiply by the corresponding dilution times to get the actual concentration of the sample.

2.17. Micro-CT

On day 30 after CIA induction, rats (same batch as Section 2.15) were euthanized (method refers to Section 2.3.6) and joint tissues were collected. Then, the muscle and fascia were removed as clean as possible and fixed in 4% paraformaldehyde for 48 h. Finally, the samples were scanned using Micro-CT to observe the degree of bone erosion. Scanning parameters are as follows: voltage: 80 kV; current: 500 µA; exposure time: 1800 ms. SIEMENS Inveon Research Workplace V4.2 was used for 3D reconstruction to obtain high-resolution images.

2.18. Histological examination

The above joint tissues (same batch as Section 2.15) were further stained with hematoxylin and eosin (H&E), safranin O, or toluidine blue at the LiLai Biomedical Experiment Center. The fixed tissue was decalcified in a pre-prepared 15% Ethylenediaminetetraacetic acid (EDTA) solution, followed by automated dehydration, embedding, sectioning, dewaxing, and hydration. (1) H&E stain: the tissue was stained with hematoxylin for 10–20 min and then differentiated in acid alcohol for 5–10 s. It was then subjected to a blue return process in a weak alkaline solution at 50 °C after water washing. Subsequently, it was stained with eosin for 3–5 min after being washed again in 85% alcohol for 3–5 min. (2) Safranin O stain: the tissue underwent staining in freshly prepared Weigert’s stain for 3–5 min and differentiation in acetic acid solution for 15 s. After water washing, it was immersed in fast green stain for 1.5–3 min before being placed into Safranin O stain for an additional 2 min. The tissue slices were then rinsed with acetic acid solution followed by water. (3) Toluidine blue stain: the tissue was placed into a preheated (50 °C) aqueous solution of toluidine blue and incubated at 56 °C for 20 min. After washing with water, soak in 70% alcohol for 1 min, then differentiate with 95% alcohol.

After the above staining process was completed, the sections were dehydrated with gradient alcohol, transparent with xylene, sealed with neutral gum, and finally moved to microscopy for image acquisition.

2.19. Statistical analysis

Statistical analysis was performed using GraphPad Prism 7 (GraphPad Software, La Jolla, CA, USA). All data were shown as mean ± standard deviation (SD). Pairwise differences were assessed for significance using Student’s t-test. Differences among three or more groups were assessed by one-way analysis of variance and Dunnett’s multiple comparison test. Differences associated with p < .05 were considered statistically significant.

3. Results

3.1. Synthesis and characterization of HSA-MnO₂@MTX NPs

Size affects NP migration and accumulation in inflammatory tissues: larger size prolongs retention in the blood and slows penetration into inflammatory tissue, while smaller size has the opposite effect (Yu et al., 2020). According to the manufacturing process, a uniform layer of mesoporous MnO₂ was grown on the surface of SiO₂ NP by adding KMnO₄. To this end, we screened the prescription of SiO₂ NP size in terms of reaction temperature and the volume of TEOS. Synthesis of SiO₂ NPs at a reaction temperature of 45 °C and in a volume of 3 mL of TEOS led to the ideal size of ∼100 nm, as measured by dynamic light scattering (Figure 2(A)). TEM revealed that H-MnO₂ NPs had a spherical hollow structure with actual particle size of about ∼100 nm (Figure 2(B)). The optimal drug loading formulation ratio of MTX: H-MnO₂ NPs was 1: 1 where LC was 39.01 ± 0.01% and EE was 63.97 ± 0.02% (Figure 2(C)). The profile of UV-Vis proved that MTX and HSA could be successfully attached to NPs (Figure 2(D)). Meanwhile, filling the NPs with MTX and coating them with HSA substantially changed their morphology with actual particle size of about ∼120 nm (Figure 2(E)). HSA-MnO₂@MTX NPs released about 30% of MTX during 48 h in solution at pH 7.4, while about 80% of MTX was released within 8 h at pH 5.5 with 10 mM H₂O₂ which is attributed to the fact that H-MnO₂ could undergo redox reaction with H₂O₂ to release O₂, generate Mn²⁺ and promote the release of MTX. These results suggest that the NPs should selectively release their cargo in the acidic, inflammatory environment of inflamed joints and not in the blood (Liu et al., 2020; Figure 2(F)). Subsequently, we evaluated the formation of PC by observing the changes in size and ζ potential from different NPs with rat plasma treatment in vitro. As shown in Figure 2(G), the absolute values of ζ potential for H-MnO₂, MnO₂@MTX, and HSA-MnO₂@MTX NPs were beyond 20 mV with good stability. Interestingly, the electronegative ζ potential of NPs increases gradually with the synthesis process, which may be attributed to the electronegativity of both HSA and MTX. However, the absolute value of ζ potential of MnO₂@MTX NPs was close to 0 mV after PM treatment, which indicated that the stability of MnO₂@MTX NPs became worse. These results imply that PM has an effect on the ζ potential of NPs, and has less effect on NPs saturated with HSA than on bare NPs. Although the absolute value of ζ potential decreased after PM treatment, it could still maintain about 18 mV. In addition, the hydrodynamic particle size of MnO₂@MTX NPs was also significantly increased after PM treatment, while that of HSA-MnO₂@MTX NPs was not found (Figure 2(H)). These results confirmed from another perspective that the preferential saturation of HSA on the surface of MnO₂@MTX NPs resisted the formation of PC and improved the stability of NPs.

Figure 2.

Characterization of HSA-MnO₂@MTX NPs. (A) The size of SiO₂ NPs was measured by DLS under different reaction temperatures and TEOS volumes. (B) TEM images of H-MnO₂ NPs. Scale bar = 100 nm. (C) The encapsulation efficiency and loading capacity of H-MnO₂ NPs at different ratios (MTX: H-MnO₂ NPs, w/w) were measured by HPLC. (D) The image was obtained from GraphPad Prism 9.0.0 by processing the raw data from ultraviolet-visible absorption spectra of free MTX, H-MnO₂, MnO₂@MTX, HSA, and HSA-MnO₂@MTX. (E) TEM images of HSA-MnO₂@MTX NPs. Scale bar = 100 nm. (F) Cumulative release of HSA-MnO₂@MTX NPs at pH 7.4 or pH 5.5 (with 10 mM H₂O₂). (G–H) ζ potential (G) and size (H) of H-MnO₂, MnO₂@MTX, HSA-MnO₂@MTX, PM + MnO₂@MTX, and PM + HSA-MnO₂@MTX NPs. Data are mean ± SD (n = 3). DLS: dynamic light scattering; H₂O₂: hydrogen peroxide; IL: interleukin; HSA: human serum albumin; H-MnO₂: hollow structured manganese dioxide; MTX: methotrexate; NP: nanoparticle; PM: plasma; TEM: transmission electron microscopy.

3.2. Cellular uptake of HSA-MnO₂@MTX NPs in vitro

The formation of PC on the surface of NPs promotes its recognition and clearance by the MPS where macrophages are the ‘main force’ (Cai and Chen, 2019). Thus, we used RAW264.7 as the macrophage model to evaluate whether pretreatment with HSA can reduce the phagocytosis and clearance of MnO₂@MTX NPs. As shown in Figure 3(A), compared with MnO₂@DID NPs, the fluorescence intensity and distribution of MnO₂@DID NPs in macrophages were significantly increased after PM treatment. In contrast, no significant changes were found in the fluorescence intensity of HSA-MnO₂@DID NPs within macrophages before or after PM treatment (Figure 3(B)). In addition, the fluorescence intensity of HSA-MnO₂@DID NPs in macrophages is also weaker than that of MnO₂@DID NPs. These imply that HSA-MnO₂@DID NPs can be uptake less than MnO₂@DID NPs by macrophages, and also indicate that HSA pretreatment is conducive to preempting the saturation of PC-binding sites on the surface of MnO₂@DID NPs to avoid the formation of PC and decrease clearance.

Figure 3.

Cellular uptake of HSA-MnO₂@MTX NPs. (A, B) Representative confocal microscopy images showing uptake of MnO₂@DID, PM + MnO₂@DID, HSA-MnO₂@DID, and PM + HSA-MnO₂@DID NPs by RAW264.7. Scale bar, 50 µm. (C) Representative confocal microscopy images showing uptake of HSA-MnO₂@DID and MnO₂@DID NPs by LPS-activated RAW264.7. Scale bar = 25 µm. (D) Bio-TEM images of LPS-activated macrophages following exposure to HSA-MnO₂@MTX NPs (20 µg/mL) for 4 h. DID, the 4-chlorobenzenesulfonate salt of 1,1′-dioctadecyl-3,3,3′,3′-tetramethyl indodicarbocyanine; HSA: human serum albumin; H-MnO₂: hollow structured manganese dioxide; MTX: methotrexate; TEM: transmission electron microscopy. PM: plasma.

Next, using confocal microscopy, we investigated the internalization of NPs by RAW264.7 cells that had been activated by LPS to evaluate their specific uptake. In these experiment results, it was evident that LPS-activated macrophages internalized more HSA-MnO₂@DID than MnO₂@DID NPs (Figure 3(C)), which may be attributed to the higher energy demand of LPS-activated macrophages driving the internalization of HSA-MnO₂@DID NPs. To further demonstrate whether NPs can successfully escape from the endosome/lysosome system, we characterized the intracellular location of HSA-MnO₂@MTX NPs using biology TEM (Bio-TEM) and found that it localized mainly in the cytoplasm of LPS-activated macrophages (Figure 3(D)).

3.3. Intracellular levels of inflammatory cytokines and ROS

The determination of nontoxic dose concentration is the prerequisite for subsequent cell experiments. Therefore, we evaluated the cytotoxicity of NPs with concentration gradients on RAW264.7 by the MTT method. As shown in Figure 4(A), the viability of RAW264.7 decreased after treatment with NPs containing increasing MTX concentrations, possibly reflecting the drug’s anti-proliferative activity. Although the toxicity of MTX encapsulated in H-MnO₂ NPs can be mitigated to a certain extent, the safety of H-MnO₂ NPs as an exogenous inorganic nanomaterial is still problematic. Interestingly, pretreatment with HSA could decrease the toxicity of both and significantly raise the upper limit for safe administration. Based on these results, we chose the concentration of 20 µg/mL for subsequent culture experiments. Next, we confirmed that HSA-MnO₂@MTX NPs alleviated the intracellular inflammatory response (Figure 4(B–D)). Compared to other groups, HSA-MnO₂@MTX NPs significantly downregulated TNF-α and IL-1β while up-regulating the anti-inflammatory mediator IL-10. Notably, MnO₂@MTX NPs downregulated TNF-α and IL-1β both better than H-MnO₂ NPs or MTX, implying the positive synergistic effect by H-MnO₂ NPs and MTX. In addition, NPs that contain H-MnO₂ were superior to free MTX in the down-regulation of HIF-α (Figure 4(E)), and HSA-MnO₂@MTX NPs also effectively scavenged ROS and suppressed levels of H₂O₂ in LPS-activated macrophages (Figure 4(F,G)), indicating that it can effectively utilize intracellular H₂O₂ to promote the production of oxygen and the consumption of ROS.

Figure 4.

Intracellular levels of inflammatory cytokines and ROS. (A) Cytotoxicity of formulations with different MTX concentrations against RAW264.7 was assessed by MTT assay. Data are shown as mean ± SD (n = 5). (B–E) Levels of (B) TNF-α, (C) IL-1β, (D) IL-10, and (E) HIF-α in medium from treated LPS-activated RAW264.7 cell cultures. Data are shown as mean ± SD (n = 3). (F) Intracellular levels of H₂O₂ in LPS-activated RAW264.7 after various treatments. Data are shown as mean ± SD (n = 3). (G) Fluorescence images of LPS-activated RAW264.7 incubated with DCFH-DA after different treatments for detecting ROS. Scale bar = 100 µm. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. PBS. DCFH-DA, 2′,7′-dichlorofluorescin diacetate; HSA: human serum albumin; H-MnO₂: hollow structured manganese dioxide; MTT: 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; MTX: methotrexate.

3.4. HSA-MnO₂@MTX NPs inhibit proliferation of CIA-FLS

Based on the hyperproliferative properties of FLS under inflammation and the original inhibitory function of MTX, we evaluated whether HSA-MnO₂@MTX NPs also could inhibit the proliferation of CIA-FLS. Compared with free MTX, HSA-MnO₂@MTX NPs better inhibited their proliferation in proportion to MTX concentration (Figure 5(A)). For CIA-FLS, compared to MTX, HSA-MnO₂@MTX NPs also could better inhibit their excessive migration (Figure 5(B)). Meanwhile, calcein AM/PI staining after HSA-MnO₂@MTX NP treatment showed the higher dead/live cell ratio, which implied its ability to induce cell death more strongly (Figure 5(C)). The cell death was further determined to be apoptosis, based on the increase in the ratio of JC-1 polymer (red fluorescence) to JC-1 monomer (green fluorescence; Figure 5(D)), signaling a decrease in mitochondrial membrane potential. Thus, we conclude that HSA-MnO₂@MTX NPs are the most effective treatment to block inflammatory responses initiated by CIA-FLS. These findings likely reflect that CIA-FLS takes up NPs better than free MTX.

Figure 5.

HSA-MnO₂@MTX NPs inhibit proliferation and induce apoptosis of CIA-FLS. (A) Cell viability of formulations containing different MTX concentrations against CIA-FLS was detected by MTT assay. Data are shown as mean ± SD (n = 4). ****p < 0.0001 vs. MTX. (B) Representative images showing the migration ability of CIA-FLS in a wound scratch assay after 24-h treatment with MTX or HSA-MnO₂@MTX NPs. The yellow lines show the edges of the cell migration. (C) Representative images showing cell apoptosis indicated by calcein AM/PI staining of CIA-FLS after 24h treatment with MTX or HSA-MnO₂@MTX NPs. Scale bar, 100 µm. (D) Representative images showing loss of mitochondrial membrane potential, as indicated by JC-1 staining, in CIA-FLS after 24-h treatment with MTX or HSA-MnO₂@MTX NPs. Scale bar = 100 µm. CIA-FLS that were not exposed to any formulation served as controls. HSA: human serum albumin; H-MnO₂: hollow structured manganese dioxide; MTX: methotrexate.

3.5. Biodistribution of HSA-MnO₂@MTX NPs in vivo

Studies have shown that SPARC overexpression in inflammatory joints facilitates the targeted accumulation effect of albumin-based nanomaterials (Zheng et al., 2022). Therefore, before evaluating the biodistribution of HSA-MnO₂@MTX NPs, we first compared the expression of SPARC in ankle joints from healthy and CIA rats. Immunohistochemical results showed that SPARC was noticeably more prevalent in the arthritic ankle joints (Figure 6(A)). Then, we replaced MTX with DID and, using near-infrared fluorescence imaging, visualized the distribution of HSA-MnO₂@DID NPs or free DID in CIA rats at 1, 4, 7, and 10 h after CIA induction. Compared with free DID, HSA-MnO₂@DID NPs exhibited stronger and longer-lasting fluorescence in joints (Figure 6(B,C)). These results in vivo, which were confirmed by fluorescence analysis of the organs ex vivo (Figure 6(D)), suggest that the carrier structure of HSA-MnO₂ significantly enhances the accumulation of drug at the disease site, which may lead to higher therapeutic efficacy for NPs in vivo.

Figure 6.

Biodistribution of HSA-MnO₂@MTX NPs in CIA rat models. (A) Immunofluorescence staining against SPARC in ankle tissue from healthy and CIA rats. Scale bar = 40 µm. (B) Fluorescence distribution of free DID and HSA-MnO₂@DID NPs in joints of CIA rats at different times after CIA induction. (C) Semi-quantitative mean fluorescence intensity in joints of CIA rats. Data are shown as mean ± SD (n = 3). (D) Ex vivo imaging of free DID and HSA-MnO₂@DID NPs in ankle joints and organs. Heart (1), liver (2), spleen (3), lung (4) and kidney (5). *p < 0.05, **p < 0.001, ****p < 0.0001 vs. free DID. HSA: human serum albumin; H-MnO₂: hollow structured manganese dioxide; DID: 4-chlorobenzenesulfonate salt of 1,1′-dioctadecyl-3,3,3′,3′-tetramethyl indodicarbocyanine.

3.6. Therapeutic efficacy of HSA-MnO₂@MTX NPs in vivo

According to the animal model establishment scheme and treatment cycle shown in Figure 7(A), at 16 days after arthritis induction, rats were randomly divided into four groups, which were treated in different ways (once for 3 days) until the 28th day. Dynamic measures to monitor which included ankle diameter, paw thickness, and paw volume during treatment found that all groups showed varying degrees of downtrend, but the saline group showed the slowest decrease and remained the most severely affected by the disease after treatment. However, CIA rats treated with HSA-MnO₂@MTX NPs showed the greatest significant reduction in ankle diameter and paw volume compared to the saline group, whereas MTX and MnO₂@MTX NPs were not found to be superior (Figure 7(B–D)). Photographing the treated inflamed joints revealed that compared to the untreated healthy group, swelling, and deformation remained severe after saline treatment, while MTX, MnO₂@MTX, and HSA-MnO₂@MTX groups showed an increasing degree of improvement (Figure 7(E)). Micro-CT was used to further observe the bone structure after different treatments and found that the saline group still had serious bone erosion, while HSA-MnO₂@MTX group showed the most significant bone amelioration and repair (Figure 7(F)).

Figure 7.

Therapeutic efficacy of HSA-MnO₂@MTX in vivo. (A) Schematic diagram of the CIA rat establishment and treatment protocol. (B–D) Changes in (B) ankle diameter, (C) paw volume, and (D) paw thickness of CIA rats over time. (E) Photographs of hind limbs from CIA rats after different treatments. (F) Micro-computed tomography (micro-CT) images of joints after different treatments. Data are shown as mean ± SD (n = 3). *p < 0.05, **p < 0.01 vs. saline. HSA: human serum albumin; H-MnO₂: hollow structured manganese dioxide; MTX: methotrexate.

The expression levels of TNF-α, IL-1β, HIF-α, and IL-10 in the serum of CIA and healthy rats were measured as indicators of inflammation. All treated groups showed different degrees of improvement, with animals treated by HSA-MnO₂@MTX NPs showing significant down-regulation of TNF-α, IL-1β, and HIF-α, as well as up-regulation of IL-10 (Figure 8(A)). These findings are consistent with the experiments in vitro.

Figure 8.

Inflammatory indicators after treatment in vivo. (A) Levels of TNF-α, IL-6, HIF-1α, and IL-10 in serum of CIA rats. *p < 0.05, **p < 0.01 vs. saline. (B–D) Staining of ankle joints of rats using (B) hematoxylin–eosin, (C) safranin O, or (D) toluidine blue after different treatments. Magnification: 100×. (E) Quantification of hemolysis rate from CIA rats after different treatments. (F) Representative images of hemolysis of blood from CIA rats after different treatments at equivalent MTX concentration of 500 µg/mL. Data are shown as mean ± SD (n = 3). HSA: human serum albumin; H-MnO₂: hollow structured manganese dioxide; MTX: methotrexate.

Next, H&E staining of the ankle sections showed that the most severe synovial hyperplasia and inflammatory infiltration were observed in the saline group, whereas HSA-MnO₂@MTX NPs prevented such pathology (Figure 8(B)). In addition, safranin O and toluidine blue staining also showed serious cartilage degeneration and damage in the saline group, and the reversal of cartilage damage at HSA-MnO₂@MTX was better than that at MTX or MnO₂@MTX (Figure 8(C,D)).

The biosafety of HSA-MnO₂@MTX NPs was evaluated by measuring the hemolysis in CIA rats. 2% erythrocyte suspensions were incubated with NPs at 37 °C for 4 h, resulting in the hemolysis rates in all groups being less than 5% which was within the normal range, regardless of MTX concentration (Figure 8(E,F)).

4. Discussion

Currently, the research and application of inorganic nanomaterials in RA disease models are becoming more and more widespread and common, especially the combination of multiple therapies triggered by composite inorganic materials has successively shown great advantages in recent years (Huang et al., 2021; Qiu et al., 2022b). However, the stability and retention of inorganic nanomaterials in body systems often need to be considered. In this study, we developed MTX-loaded NPs that were saturated on their surface with HSA before injection, leading to better localization to inflamed joints, better internalization by LPS-activated macrophage, and stronger therapeutic effects in CIA rats than free MTX or NPs that were not precoated with HSA. We attribute the efficacy of HSA-MnO₂@MTX NPs to several factors: (1) precoating of the NPs with albumin at physiological concentrations prevents PC formation in vivo; (2) inflamed tissues express high levels of SPARC, which bind strongly with HSA and thereby help recruit and retain NPs; and (3) the acidic, ROS-rich microenvironment in inflamed joints induces MTX release.

Our study confirmed the effectiveness of this approach. Coating MnO₂@MTX NPs with HSA reduced their uptake by macrophages in the blood and improved their ability to target sites of inflammation. The superiority of HSA-MnO₂@MTX NPs over naked MnO₂@MTX NPs highlights the usefulness of precoating NPs with HSA. HSA-MnO₂@MTX NPs also inhibited the hyperproliferation and migration of CIA-FLS and reduced the expression of inflammatory cytokines and ROS in inflammatory activating macrophages, indicating that it helped MTX to further ‘personalization’ strategies within multiple cells to obtain synergistic therapies.

However, it must be acknowledged that the serum microenvironment in vivo is a dynamic process, which is still different from the simulated environment in vitro. These methods are unlikely to prevent PC formation completely, given the heterogeneity of proteins in the blood (Li et al., 2021b; Lei et al., 2023; Wang et al., 2023). Although some methods such as modifying specific albumin binding domains have been used to modulate the composition of PC in vivo by increasing the surface albumin content on NPs (Li et al., 2023), how to ensure that NPs are as saturated or occupied by albumin as possible and are not easily altered still needs further exploration and demonstration. In addition, different clinical states of single diseases or comorbidities can also change the composition of PC to have different effects on the biological behavior of NPs (Xu et al., 2022).

Moreover, MTX has a well-defined toxicity profile and its clinical use is usually limited by severe hepatorenal toxicity and intestinal damage (Cao et al., 2019; Mei et al., 2020). Hepatotoxicity is considered to be the most predominant, caused by MTX and its metabolite, which is mainly manifested by histological alterations in the liver and leads to hepatic fibrosis and cirrhosis (Cao et al., 2019; Ezhilarasan, 2021). The potential mechanisms by which MTX induces hepatotoxicity have not yet been fully identified, but one of the mechanisms that has been proposed is the dysregulation of cellular antioxidant defenses, which contributes to the accumulation of ROS leading to hepatocyte injury (Moghadam et al., 2015). Therefore, MTX combined with H-MnO₂ NPs may be beneficial in reducing MTX hepatotoxicity by neutralizing the harmful effects of free radicals against hepatocyte oxidative stress injury, but this still requires more experimental results to confirm this, which is a potential future direction for this study. Therefore, future research needs to refine into specific diseases or develop universal albumin-based NPs. Further research should clarify long-term toxicity and metabolism in the body, as well as explore methods to ensure reliable, clinical-grade manufacturing quality to bring nanomedicines into the clinic (Lu et al., 2021). In parallel, animal models that more accurately resemble RA and other inflammatory diseases should be developed and applied to the development of nanomedicines.

5. Conclusions

To sum up, we constructed a novel multifunctional MTX-based nano-delivery by mimicking the formation program of PC in vitro. It could reduce recognition and clearance by the body system and deliver cargo selectively to the disease site while mitigating oxidative stress and exhibiting clear anti-inflammatory effects in vitro and in vivo. Our work may provide a new design concept and inspiration for more effective nanoplatforms in the future.

Funding Statement

This work was supported by the Sichuan Province and Technology Program [Grant Nos. 2023NSFSC0620, 2022YFS0614, 2022YFS0622] and the Key Project of Application and Basic Research of Southwest Medical University [Grant No. 2021ZKZD016].

Author contributions

Ming Jia: writing – original draft. Ming Jia, Wei Ren, and Minrui Wang: investigation, methodology, formal analysis, visualization, and validation. Yan Liu and Chenglong Wang: investigation, data curation, and validation. Zongquan Zhang, Maochang Xu, and Nianhui Ding: methodology and resources. Chunhong Li and Hong Yang: conceptualization, resources, writing – review and editing, resources, supervision, project administration, and funding acquisition. All authors have given approval to the final version of the manuscript.

Ethical approval

The authors have adhered to the ARRIVE guidelines. All experiments were performed in accordance with the Regulations of the Experimental Animal Administration, issued by the State Committee of Science and Technology of the People’s Republic of China (November 14, 1988). Ethical approval to conduct the study was obtained from the Animal Care and Ethics Committee of Southwest Medical University (ethical approval No. 20211123-002).

Disclosure statement

The authors report there are no competing interests to declare.

Data availability statement

The data that support the findings of this study are available from the corresponding author, upon reasonable request.