Ameliorative Effect of Curcumin Nanoparticles against Monosodium Iodoacetate-Induced Knee Osteoarthritis in Rats

Aim This study is aimed at evaluating the use of curcumin-loaded polylactic-co-glycolic acid nanoparticles (CUR-loaded PLGA NPs) as a treatment against monosodium iodoacetate- (MIA-) induced knee OA. Materials and Methods Eighteen rats were assigned to three groups (n = 6), namely, normal control group that received intra-articular injections (IAIs) of saline, an OA control group that received an IAIs of MIA (2 mg/50 μL), and a treatment group (MIA+CUR-loaded PLGA NPs) that received IAIs of CUR-loaded PLGA NPs (200 mg/kg b.wt). Results The CUR NP treatment against knee OA alleviated radiographic alternations and histopathological changes and inhibited the upregulation in the serum levels of interleukin-1β, tumor necrosis factor-α, interleukin-6, and transforming growth factor-beta and the downregulation in interleukin-10. CUR NP-treated joints also decreased the mRNA expression of nuclear factor-kappa B and inducible nitric oxide synthase and the protein expression of matrix metalloproteinase-13 and caspase-3. Finally, CUR-loaded PLGA NP treatment mitigated the loss of type II collagen, which resulted in a significant reduction in malondialdehyde level and increased the glutathione content and superoxide dismutase activity compared with that of the OA group. Conclusion This study demonstrated that the administration of CUR NPs could provide effective protection against MIA-induced OA and knee joint histological deteriorated changes due to its anti-inflammatory, antioxidant, and antiapoptotic properties.


Introduction
Osteoarthritis (OA) is a progressive and degenerative illness that happens in the whole joint and can result in articular cartilage degeneration, subchondral bone thickening, synovium inflammation, osteophyte formation, and meniscal degeneration [1]. Treatments for OA are mainly divided into three categories: nonpharmacological treatments, pharmacological treatments limited to analgesics and/or nonsteroidal anti-inflammatory drugs (NSAIDs), and surgical treatments [2]. Current available medications for OA, except joint replacement surgery, are essentially palliative and cannot hinder articular cartilage degradation [3]. Furthermore, long-term use of NSAIDs can cause adverse gastrointestinal, renal, and cardiovascular effects [4]. This calls for the development of a structural OA disease drug that is safe, provides symptomatic relief, and hinders the progression of cartilage degeneration.
Curcumin (CUR) (diferuloylmethane), a polyphenol compound obtained from turmeric, is a naturally available molecule that has robust anticatabolic, antioxidant, antiinflammatory, and antirheumatic properties [5][6][7]. Accordingly, CUR seems to be a promising approach in OA therapy. Nevertheless, the therapeutic efficiency of CUR is extremely restricted because of its low water solubility and limited oral bioavailability [8]. Nevertheless, studies have reported that CUR's biological activity could be efficiently enhanced using nanotechnology-based drug delivery [9,10]. Various studies have explored CUR nanoparticles (NPs) and CUR encapsulation with various substances such as liposomes and polymers to overcome its inherent drawbacks [11,12]. Polylactic-co-glycolic acid (PLGA) is considered as an effective biodegradable polymeric NPs that was authorized by US Food and Drug Administration for use in drug delivery systems due to its low toxicity, controlled and sustained-release properties, and biocompatibility with tissue and cells [13,14].
On the other hand, the intra-articular route for drug delivery has significant potency and fewer systemic side effects compared to that of oral delivery [15]. Accordingly, this study was designed to explore the possible feasibility of the intra-articular injections (IAIs) of CUR-loaded PLGA NPs to treat monosodium iodoacetate-(MIA-) induced knee OA in a rat model.

Materials and Methods
2.1. Animals. This study used mature male Wistar rats weighing 130-150 g, which were kept in a standard 12 : 12 light/dark cycle in well-ventilated rooms. One week before initiating the experiments, the rats were housed in sterilized cages to be adapted to the laboratory with free access to water and pellets. All methods for handling, use, and euthanasia of the animals in this study were certified by the Experimental Animal Ethics Committee of Faculty of Science, Beni-Suef University, Egypt, and the ethical authorization number is BSU/FS/2018/15. CUR-loaded PLGA NPs were synthesized by solvent solid-in-oil-in-water emulsion (s/o/w) evaporation based on the method published by Niazvand et al. [17]. The PLGA/chloroform solution (oil phase) was prepared by dissolving 60 mg of PLGA in 1 mL of chloroform. Then, 6 mg of CUR was added to the solution and sonicated, resulting in a solid/oil emulsion. Thereafter, ethanol and 2% PVA (1 : 1) was added to the emulsion and sonicated for 10 min to produce a solid/oil/weight (s/o/w) emulsion. The s/o/w emulsion was further sonicated and agitated by a magnetic stirrer for 5-6 h to evaporate the solvent (chloroform). The sample was subsequently centrifuged for 10 min at 15,000 g before being rinsed three times using distilled water. The sample was allowed to freeze-dry for 24 h to get a dry powder. The obtained NPs were kept at 4°C.

2.4.
Characterization of CUR-Loaded PLGA NPs. The surface morphology of CUR-loaded PLGA NPs was investigated with a scanning electron microscope (Zeiss Sigma 500 VP Analytical FE-SEM, Carl Zeiss Germany) and Xray diffraction (XRD) analysis (model no: 202964, Pananlytical Empyrean company). In addition, their zeta potential and size were detected by Malvern (Malvern Instruments Ltd) following the method by Moaty et al. [18].
2.5. Experimental Design. As described in Figure 1, the rats were randomly allocated into three groups (n = 6). At 0, 14, 18, 22, and 26 days, the normal control group took IAIs of 50 μL of saline into the left side knee joint, whereas the other two groups received IAIs of MIA (2 mg) into the left side knee joints on day 0. Then, rats in the OA control group were given saline IA injections on days 14, 18, 22, and 26, while the treatment group (MIA+CUR-loaded PLGA NPs) received intra-articular injections of CUR-loaded PLGA NPs at a dose of 200 mg/kg. MIA and CUR-loaded PLGA NPs injection doses were based on the previous studies by Ragab et al. [16] and Niazvand et al. [17], respectively.
2.6. Knee Diameter Measurement (Swelling). A manual caliper was used to assess the variations in the anteriorposterior diameter values of the knee joints among all groups [19]. The measurements were obtained on days 0, 7, 14, 21, and 28 after MIA injection.

X-Ray Examination.
On day 30 post-MIA injection, animals from all groups were anesthetized using diethyl ether, and both hind limbs were extended and fixed on the table with tape. Radiographs of the left knees (anterior-posterior position) were captured using an X-ray device with a 60 cm focal film distance at 55 kV and 3 mA.

Quantitative Reverse Transcription Polymerase Chain
Reaction (qRT-PCR) Assay. The mRNA levels of nuclear factor-kappa B (NF-κB), type II collagen, and inducible nitric oxide synthase (iNOS) were determined by real-time qRT-PCR, for which the Qiagen tissue extraction kit (Qiagen, USA) was applied for the total RNA isolation. Then, 0.5-2 μg total RNA was applied for cDNA synthesis utilizing a Fermentas kit (USA). Applied Biosystem software version 3.1 (StepOne, USA) was utilized for real-time qPCR amplification and analysis. The qRT-PCR assay was done with primer sets optimized for the annealing temperature. The sequences of the primers are listed in Table 1.
2.11. Western Blot Assay. The impact of CUR-loaded PLGA NPs on the protein expression levels of NF-κB50 and NF-κB65 and cleaved caspase-3 was investigated using knee samples maintained at 80°C. Briefly, all samples were homogenized in a radio-immunoprecipitation (RIPA) buffer supplemented with proteinase inhibitors and centrifuged, and the protein concentration was assayed with Bradford assay in the obtained clear supernatant. Proteins (30 mg) were separated on SDS-PAGE, moved to nitrocellulose membranes, and blocked with TRIS using 3% bovine serum albumin and a Tween 20 (TBST) buffer at ambient temperature for 1 hour. Subsequently, membranes were left in the incubator with primary antibodies against NFκBp50 (Cat# 14-6732-81; eBioscience), NF-κBp65 (Cat# 14-6731-81, eBioscience), and cleaved caspase-3 (cat # PA5-114687, Thermo Fisher Scientific). Following the washing with TBST, the prepared membranes were left in the incubator with the corresponding secondary antibodies, after which a chemiluminescence kit (BIORAD, USA) was applied. The developed blots were scanned, and image analysis software was utilized to measure the obtained band intensity of the targeted proteins against the control sample after being normalized by beta-actin on a Chemi Doc MP imager.
2.12. Histopathological Analysis. The left knee joints were assembled from all groups, followed by fixing in 10% neutral buffered formalin for 48 hours, and then decalcified using 20% EDTA for two weeks. The joints were excised sagittal and processed to get 4-6 μm paraffin-embedded sections. Obtained sections were then stained using hematoxylin and eosin (H&E) before being examined under a light microscope.

Mediators of Inflammation
Endogenous peroxidase and nonspecific binding sites for antibodies were suppressed by treating the sections for 10 min with hydrogen peroxide (0.3%) and for 20 min with 5% normal bovine serum (1 : 5 diluted tris buffer saline [TRIS]) at ambient temperature, respectively. Obtained sections were rinsed with PBS, and then, 10% normal goat serum was applied for 30 min to minimize nonspecific binding. Obtained sections were then incubated in anti MMP13 primary antibody (cat no: GB11247; Servicebio, China) for 1 h, followed by incubation in streptavidin horseradish peroxidase (Dako-K0690) and biotinylated secondary antibody (Dako Universal LSAB Kit) for 15 min and then incubated in 3-diaminobenzidine tetrahydrochloride (Sigma-D5905; Sigma-Aldrich Company Ltd., Gillingham, UK) substrate kit for 10 min to achieve immunolabelling. Afterward, the nuclei were stained using Harry's hematoxylin stain, dehydrated ethanol, cleared in xylene, and then mounted in DPX. Antibody binding was observed with high-power light microscopy.
2.14. Statistical Analysis. Statistical analyses were performed using SPSS software version 25.0 (SPSS Inc., Chicago, IL, USA). All results are presented as means ðMÞ ± standard errors of means (SEMs), with P < 0:05 meaning a statistically significant difference.

Results
3.1. Characterization of CUR-Loaded PLGA NPs. In this study, the prepared CUR-loaded PLGA NPs were shown using a scanning electron microscope micrograph (Figure 2), which were spherically shaped. Meanwhile, the XRD pattern of CUR-loaded PLGA NPs (Figure 3) showed an absence of There were six samples in each group, and data are described as means ± SEM. For each parameter, means (which have different superscript symbols) are statistically significant at P < 0:05.   Mediators of Inflammation marked crystalline domains in CUR-loaded PLGA NPs, implying that these NPs were in the disordered-crystalline phase or the amorphous or the solid-state solubilized form in the polymer matrix. Further, the average size (Figure 4(a)) and zeta potential (Figure 4(b)) of the CUR-loaded PLGA NPs were 265.2 nm and −6.86 mV, respectively.

3.2.
Effect of CUR-Loaded PLGA NPs on Swelling (Knee Diameter Measurements). As shown in Figure 5, MIA induced an increase in the knee diameters of the OA control and treatment groups when compared to the measurements before the injection. Nevertheless, two weeks of CUR-loaded PLGA NP IAIs markedly reduced swelling in the left knee joints.

Effect of CUR-Loaded PLGA NPs Using X-Ray Imaging.
Radiographic detection was carried out to observe changes in the knee joints after CUR-loaded PLGA NP treatment. Compared to that of the normal control group (Figure 6 Table 2) showed a significant increase in serum TNF-ɑ, IL-6, IL-1β, and TGF-β levels and a decrease in IL-10 levels compared to those of the normal control group (P < 0:05). Meanwhile, the OA knee joints from the treatment group showed a marked reduction in TNF-ɑ, IL-6, IL-1β, and TGF-β levels, along with an elevation in IL-10 levels as compared to those of the OA group.

Effect of CUR-Loaded PLGA NPs on Lipid Peroxidation
and Antioxidant Status. The enhanced level of serum lipid peroxidation product (MDA) in MIA-treated knee joints was accompanied by a substantial reduction in GSH level and SOD activity. Meanwhile, CUR-loaded PLGA NP IAIs significantly lowered the level of MDA and boosted the GSH content and SOD activity (Table 3).
3.6. Effect of CUR-Loaded PLGA NPs on NF-κB, iNOS, and Type II Collagen mRNA Expression. OA knee joints showed a marked (P < 0:05) elevation in the mRNA expression levels of NF-κB and iNOS, along with a significant decline in type II collagen compared with those of the normal group (Table 4). Moreover, OA knee joints treated with CUR exhibited a significant (P < 0:05) downregulation in NF-κB and iNOS levels and inhibited the loss of type II collagen mRNA expression level in comparison with that in OA control rats.   Mediators of Inflammation cartilage showed that chondrocytes in the normal control group were placed neatly, and the dyeing was uniform (Figure 9(a)). In the OA control group, obtained sections had a myriad of pathological alternations such as cracks, fibrillation, disorderly arranged cells, empty lacunae, hyperchromatic nuclei, and a magnificent reduction in the number of chondrocytes (Figures 9(b)-9(d)). In contrast, the MIA +CUR-loaded PLGA NP group (Figures 9(e)) showed a notable decrease in the severity of cartilage degradation, as the CUR treatment offered effective protection against OA progression. MIA+CUR-loaded PLGA NP-treated cartilage appeared to have a smooth surface, orderly arranged chondrocytes, less loss of cells, and intact subchondral bone compared to those of MIA-treated cartilages without treatment.  There were six samples in each group, and data are described as means ± SEM. For each parameter, means, which have different superscript symbols, are statistically significant at P < 0:05. in the chondrocytes of the articular cartilage to assess its protein expression. When compared with the normal control group, which almost had no positive staining for There were six samples in each group, and data are described as means ± SEM. For each parameter, means, which have different superscript symbols, are statistically significant at P < 0:05. There were six samples in each group, and data are described as means ± SEM. For each parameter, means, which have different superscript symbols, are statistically significant at P < 0:05.   (Figure 10(a)), MIA induced an elevation in MMP-13 content in the articular cartilage of the OA control group (Figure 10(b)). However, sections (Figure 10(c)) showed that the MMP-13 content was notably diminished in the articular cartilage of the MIA+CUR-loaded PLGA NP group as compared to that of the OA control group.

Discussion
Currently, evidence shows that there is a significant correlation between the incidence and progression of OA and inflammation, oxidative stress, and excessive catabolic activity [24].
In the current work, we utilized MIA to generate histological and biochemical changes in the articular cartilage that resemble OA conditions in humans [25]. Moreover, in search of inexpensive and beneficial treatments against OA, we prepared CUR-loaded PLGA NPs and investigated their potency against inflammatory mediators, oxidative stress, and chondrocyte apoptosis in MIA-induced OA in a rat model (Figure 11).
The sizes of the CUR-loaded PLGA NP detected in our study were smaller than 300 nm and were consistent with those reported by Gonzales et al. [26]. While the XRD pattern in our study displayed no typical CUR peaks when entrapped in NPs, Khan et al. [27] elucidated that the absence of any noticeable crystalline domains of CUR implies that CUR loaded on PLGA NPs is in the disordered-crystalline phase or the amorphous or the solidstate solubilized form in the polymer matrix. This disordered-crystalline phase, or CUR, inside the polymeric matrix, allows for a controlled release of the encapsulated drug from the NPs.
Following Yabas et al. [28], our radiographic results revealed that CUR NP intra-articular administration markedly lessened the MIA-induced radiographic abnormalities in the knee joints of the treated rats represented by normal joint space and smooth surface of articular cartilage.
The NF-κB signaling pathway is claimed to be one of the main signaling pathways that promote the progression of OA [29]. Although NF-κB stays inactive in the cytoplasm under normal conditions, upon adequate stimulation, e.g., by inflammatory cytokines or the inflammatory microenvironment, IκB kinase (IKK) activity phosphorylates IκB proteins and causes their degeneration, which allows free NF-κB complexes to translocate from the cytoplasm into the nucleus and stimulate a variety of inflammation-related genes. Furthermore, it triggers extracellular matrix degradation, chondrocyte apoptosis, pannus formation, and, eventually, pathological cartilage destruction ( Figure 11) [5,30].
Alternatively, various studies assumed that CUR hinders OA-related inflammation by blocking the NF-κB signaling  Mediators of Inflammation pathway and preventing chondrocyte apoptosis [31,32]. As a result, OA-related inflammation is suppressed and progression is slowed [11,33,34]. Subsequently, our study discussed the effect of CURloaded PLGA NP IAIs on NF-κB gene and NF-κB-regulated genes participating in inflammation, such as TNF-α, IL-6, IL-1β, and TGF-β. Our data in harmony with Alhusaini et al. [30] propose that CUR administration blocked NF-κB activation by inhibiting IκB degradation and phosphorylation and suppressing the translocation of NF-κB into the nucleus, thereby impeding the inflammatory response of the cells.
Additionally, in agreement with [35], our presented data revealed that CUR considerably boosted the serum levels of IL-10, a robust anti-inflammatory immunosuppressive and chondroprotective cytokine, indicating its potent antiinflammatory capacity.
Furthermore, CUR-loaded PLGA NP treatment during OA inhibited the expression of iNOS, which is an inflammation-induced enzyme that catalyzes the production of the proinflammatory mediator nitric oxide (NO), further demonstrating its anti-inflammatory effects. Several in vivo and in vitro studies [36,37] have shown that CUR treatment reduces iNOS production in various inflammatory diseases.

Mediators of Inflammation
Most recently, Cheragh-Birjandi et al. [37] postulated that CUR administration can regulate NO levels by suppressing the activation of the N-terminal kinase (JNK), p38, and NF-κB pathways.
NF-κB pathway activation not only upregulates proinflammatory mediators but additionally mediates the chondrocyte activation triggered by the released extracellular matrix (ECM) products, e.g., fibronectin fragments, which   Figure 11: Therapeutic effect of CUR-loaded PLGA NPs against cartilage damage due to inflammation, extracellular matrix (ECM) degradation, and apoptosis in MIA-induced knee OA. 10 Mediators of Inflammation in turn promote the expression of a variety of matrixdegrading enzymes, including metalloproteinases (MMPs) [38][39][40]. In the early stage of OA, high upregulation of MMP-13 expression, a prominent catabolic enzyme, leads to severe deterioration of cartilage as it induces type II collagen, which is a reason for nearly 90% of the ECM. The loss of type II collagen is a critical stage that determines the irreversible progression of OA ( Figure 11) [41]. Therefore, this study explored MMP-13 and collagen type II expression levels as indicators of the anticatabolic activity of CUR NPs in the OA knee joint. Herein, IAIs of CUR NPs prevented OA exacerbation by diminishing MMP-13 expression [42] and the degradation of type II collagen [6]. Kumar et al. [43] posited that CUR possesses a suppressing effect on MMPs and variable cellular signaling pathways, e.g., Janus kinase STAT and NF-κB/ mitogen-activated protein kinase/phosphoinositide 3-kinase, which eventually ameliorate OA and prevent the further damage of the cartilage.
Further, the effect of CUR treatment on cell death in knee OA has also been elucidated. Although apoptosis is a crucial process in keeping the homeostasis of several tissues, the high rate of chondrocyte death is a well-known pathological feature of OA [44]. MIA-induced TNF-α activates the tumor necrosis factor receptor (TNFR) or death receptors, eventually triggering the extrinsic pathway of apoptosis ( Figure 11). While Ding et al. [45] postulated that in MIAinduced OA, the intrinsic pathway of apoptosis can be triggered by reactive oxygen species (ROS). ROS accumulation could induce oxidative stress, thus altering mitochondrial function and promoting the excretion of cytochrome c and activation of proapoptotic factors, e.g., caspase-3 [44,46]. Moreover, mitochondrial impairment was shown to enhance the response to cytokine-induced chondrocyte inflammation by producing ROS and activating NF-κB [47].
Conversely, CUR has been shown to restore mitochondrial functions, scavenge free radicals and ROS [48,49], and suppress lipid peroxidation [50]. CUR also enhances the activity of other antioxidants such as SOD, catalase, GSH, and glutathione peroxidase in various diseases [30,51]. Based on these studies, we scrutinized the role of CUR-loaded PLGA NPs in suppressing the activation of the apoptosis mediators related to OA. Our data confirmed the antioxidant efficacy of CUR NPS as they considerably averted the increase in MDA lipid peroxidation product and enhanced the antioxidant status (GSH concentration and SOD activity). Moreover, consistent with the results of Buhrman et al. [52], our results revealed the antiapoptotic efficacy of CUR NPS, as they notably hindered upregulation in caspase-3 levels. The antioxidant, anticatabolic, and antiapoptotic potency can be mostly attributed to the hydroxyl, methoxy, α, β-unsaturated carbonyl, or diketone groups present in CUR [53].
MIA-induced knee OA inflammation led to severe histopathological alterations in the architecture of the articular cartilage, such as clefts, disorderly arranged cells, hyperchromatic, and a significant reduction in chondrocyte number [31]. Whereas CUR-loaded PLGA NP treatment did not affect the regeneration of the cells, it halted OA progression by reducing inflammation and chondrocytes apoptosis which is depicted by milder tissue defects and pronounced articular cartilage and subchondral bone integrity. Our findings are consistent with Yabas et al. [28], which revealed that CUR lessened OA severity by restoring the architecture of the knee joint. Wang et al. [54] postulated that CUR moderately enhanced the integrity of the articular cartilage by blocking the NF-κB/hypoxia-inducible factor 2α signaling pathway.
Therefore, current findings suggest that OA-related deteriorations and progression are halted due to the ability of CUR-loaded PLGA NPs to modulate various mediators such as NF-κB, MMP-13, and oxidative stress.

Conclusion
Overall, this study demonstrates that intra-articular treatment with CUR-loaded PLGA nanoparticles is a compelling candidate for improving joint health in knee OA due to their antiinflammatory, anticatabolic, and antioxidant characteristics. However, for clinical applications, further research studies with a longer treatment period and bigger sample size should be carried out to explore its underlying mechanisms.

Data Availability
The authors confirmed that all data generated or analyzed during this study are included in this published article.