Evaluation of Conventional and Hyaluronic Acid-Coated Thymoquinone Liposomes in an In Vitro Model of Dry Eye

Dry eye disease (DED) is a common ocular disorder characterized by an inadequate lubrication of the eye by tears leading to inflammation and the alteration of the ocular surface. Current treatments are often limited due to their side effects and ineffectiveness. Thymoquinone (TQ) is a natural compound present in the essential oil of Nigella sativa L., with anti-inflammatory and antioxidant activities. In this study, conventional and hyaluronic acid-coated liposomes were developed to improve TQ activity at ocular level. In the present study, the cytoprotective effects of TQ or TQ liposomes were assessed against oxidative and inflammatory processes in human corneal epithelial cells (HCE-2). Hyperosmolarity conditions (450 mOsm) were used as a model of DED. Interleukin-1β (IL-1β), Interleukin-6 (IL-6) and tumor necrosis factor (TNFα) were quantified by quantitative real-time polymerase chain reaction (RT-qPCR); COX-2 and Phospho-NF-κB p65 (p-p65) by Western blotting (WB). Moreover, the mitochondrial reactive oxygen species (mtROS) levels were measured by MitoSOX assay. The hyperosmotic treatment induced a significant increase of the proinflammatory genes and proteins expression that were significantly decreased in the liposomes-treated cells. The coincubation with hyaluronic acid-coated liposomes significantly reverted the increase of mtROS production, evidently stimulated by the hyperosmotic stress. Our data suggest that TQ-loaded liposomes have potential as a therapeutic agent in dry eye disease, improving the TQ efficacy.


Introduction
Dry eye disease (DED) is a common ocular disorder affecting tens of millions of people, with prevalence in Asian compared to Western countries, indicating that the etiology of DED depends on cultural and racial factors [1]. The primary cause of DED is the decreased production of tears, such as the aqueous tear deficiency type of dry eye seen in Sjögren's syndrome, but other conditions may also contribute to an unstable tear film [2]. There are many clinical conditions of patients with DED, hence the need to give a clinical definition that encompasses most of these. Tsubota and colleagues proposed a new definition: "Dry eye is a multifactorial disease characterized by a persistently unstable and/or deficient tear film causing discomfort and/or visual impairment, accompanied by variable degrees of ocular surface epitheliopathy, inflammation and neurosensory abnormalities" [3]. Tear film instability and hyperosmolarity induced ocular surface inflammation [4]. Macromolecular alterations and damage to ocular epithelial cells and lacrimal glands caused by oxidative phatidylcholine and Plurol Oleique, a liquid lipid used to improve the loading capacity, and one formulation was coated with 0.1% w/v HA. The liposomes were developed to increase TQ solubility and availability at ocular level [19]. Interleukin-1β (IL-1β), Interleukin-6 (IL-6) and tumor necrosis factor (TNFα) were quantified by quantitative real-time polymerase chain reaction (RT-qPCR) and IL-1β, COX-2 and Phospho-NF-κB p65 (p-p65) by Western blotting (WB). The mitochondrial reactive oxygen species (mtROS) levels were measured by MitoSOX assay.

Preparation of Liposomal Formulations
The TQ-loaded liposomes (LP-TQ) and HA-coated TQ liposomes (LP-TQ-HA), were prepared according to the literature [19]. The TQ, egg phosphatidylcholine and Plurol Oleique were dissolved in dichloromethane. The solvent was evaporated and the dry lipid film was hydrated with deionized water. The dispersion was shaken for 30 min in a water bath at 50 • C and then sonicated with an ultrasonic probe for 2 min and 30 s. The HA coating was achieved by the drop-wise method [29], adding 2 mL of 0.1% w/v solution of HA in deionized water to 2 mL of LP-TQ dispersion. The LP-TQ-HA has a final concentration of 0.5 mg/mL of TQ [19].

Characterization of Liposomes
Hydrodynamic diameter, polydispersity index (PdI), and zeta-potential were determined by Dinamic Light Scattering (DLS) and Electrophoretic Light Scattering (ELS), using a Zsizer Nano series ZS90 (Malvern Instruments, Malvern, UK). The encapsulation efficiency of LP-TQ and LP-TQ-HA was determined by the dialysis bag method, as previously reported [19].

Mucoadhesion Study
The mucoadhesion interaction was determined by the zeta-potential measurements of the mucin solutions and the mixtures of mucin and formulations using a Zsizer Nano series ZS90 (Malvern Instruments, Malvern, UK) [30,31]. In total, 1 mL of LP-TQ or LP-TQ-HA formulations was mixed with 1 mL of mucin suspension at different percentages (0.5%, 1%, 2% and 3%) and incubated in the ultrasound bath for 15 min. The samples were subsequently diluted with deionized water and analyzed at the ELS for the measurement of the zeta-potential.

Human Corneal Epithelial Cells (HCE-2)
As previously reported, the human corneal epithelial cells (HCE-2) were incubated in serum-free medium supplemented with bovine pituitary extract (BPE), epidermal growth factor, hydrocortisone and insulin [32], then they were plated on coated flasks and incubated in a humidified incubator at 37 • C. The medium was changed twice a week. The cells were split in new flasks after reaching the confluence.

Analysis of In Vitro Cytotoxicity
The HCE-2 cells were plated in 24-well plates and, after they had reached the confluence, were exposed to the drugs treatment, suitably diluted. The medium was used 2.6.1. MTT Assay The viability of corneal epithelial cells exposed to TQ-free or formulated (5 µM) for 5 h was evaluated by MTT assay. From each well a part of the medium was withdrawn for LDH assay. Then, the MTT (1 mg/mL) was added to the cells [19]. After removing the MTT-containing solution, dimethyl sulfoxide (DMSO) was added to the wells to dissolve the formazan crystal formations, and the absorbance of MTT was read at 550 and 690 nm. The medium was used as a positive control. Cell viability was expressed as a percentage of cells incubated only in the vehicle at the corresponding exposure time.

LDH Assay
Damage in human corneal cells was quantitatively assessed by measuring the amount of LDH released by the damaged cells into the extracellular fluid, 5 h after drug exposure, by LDH kit, as previously described [19]. The LDH level corresponding to complete cell death was determined for each experiment by analyzing sister cultures exposed to 0.01% BAK. Background LDH release was determined in drug-unexposed control cultures and subtracted from all experimental values.

Evaluation of Mitochondrial Reactive Oxygen Species (mtROS)
As previously reported, the HCE-2 cells were stained with MitoSOX at the final concentration of 2.5 µM for 15 min, then washed with PBS, detached with accutase, and resuspended with PBS [34]. The stained cells were acquired using a MACSQuant Analyzer 10 Flow Cytometer (Miltenyi Biotec ® , Bergisch Gladbach, Germany), and the data were analyzed by Flowlogic (Miltenyi Biotec ® ).

Statistical Analysis
Experiments were repeated n times and results expressed as a mean ± SEM. The statistical significance of HCE-2 cell viability was analyzed by one-way ANOVA with a post-hoc Dunnett and gene expression; protein levels and mtROS were analyzed by oneway ANOVA followed by the post-hoc Tukey's w-test for multiple comparisons. All statistical analyses were performed by the GRAPH-PAD PRISM v. 8 for Windows (GraphPad Software, San Diego, CA, USA). A probability value (p) of < 0.05 was considered significant.

Characterization of Liposomes
The physical and chemical parameters of LP-TQ and LP-TQ-HA were reported in Table 1 [19]. The HA coating was obtained using the drop by drop coating method [29], which is based on the hydrophobic interactions and hydrogen bonds between PC and HA [35][36][37][38]. The increase of the particle sizes from 146 ± 2 nm to 166 ± 3 nm and the change of zeta potential from −26 ± 3 to −36 ± 1 mV confirmed the HA deposition. TEM analyses also demonstrated this deposition around the liposomes [19]. The high value of potential for both LP-TQ and LP-TQ-HA contributes to increase the stability of the dispersion, reducing the aggregation phenomena. Table 1. Physical and chemical characterization of TQ-loaded conventional liposomes (LP-TQ), and HA-coated TQ liposomes (LP-TQ-HA) (mean ± SEM, n = 3).

Sample
Size ( The study of the mucoadhesion is of great importance to understand the efficacy of the formulation as a potential candidate for DED treatment. Different in vitro methods were reported [39][40][41]. In this study, zeta-potential determination was applied to study the mucoadhesive properties of LP-TQ-HA [30]. This is a common approach used for the investigation of mucoadhesive characteristics of biopolymers [42,43]. The mucoadhesive properties of the liposomes were evaluated by monitoring changes in the potential following incubation of the liposomal formulation with mucin [44]. The mucin has negative charge due to the carboxyl and sulphate groups of the oligosaccharide chains. The obtained value was around -10 mV, similar to the value reported in the literature [30,45,46]. The surface property of the mucin might be changed by the adhesion of the polymer, if the polymer is mucoadhesive [31]. The zeta-potential showed only a moderate shift when LP-TQ was added to the solutions of mucin (0.5%, 1%, 2% and 3% w/v), while in the presence of LP-TQ-HA a more pronounced change is observed ( Table 2). The addition of the mucin produces a more negative potential in respect to mucin alone, according to the literature [30]. Furthermore, as the percentage of mucin increases, the interaction with the HA-coated formulation increases, and the potential approaches that of the mucin [30,47]. The DLS analyses also evidenced the increase of the sizes of HA-LP-TQ in the presence of mucin, confirming this interaction. Previously, authors reported that both the formulations reduced the possible toxicity of a high dosage (30 and 60 µM) of TQ exposure in cornea and conjunctiva cells [19].
Many studies suggest a crucial role of inflammatory and oxidative processes in the pathogenesis of DED and, in particular, tear film hyperosmolarity is accepted as a key pathogenic step, since it causes damage to the surface epithelium activating a cascade of inflammatory events at the ocular surface [48,49]. In this study, we analyzed the effect of the hyperosmolarity treatment alone or in presence of TQ and TQ-loaded liposomes at a lower concentration (5 µM) for a long time (5 h) on the cornea cell viability using LDH and MTT assays. As shown in Figure 1, the hyperosmolarity treatment did not induce cell toxicity in both conditions. mucin produces a more negative potential in respect to mucin alone, according to the literature [30]. Furthermore, as the percentage of mucin increases, the interaction with the HA-coated formulation increases, and the potential approaches that of the mucin [30,47]. The DLS analyses also evidenced the increase of the sizes of HA-LP-TQ in the presence of mucin, confirming this interaction. Previously, authors reported that both the formulations reduced the possible toxicity of a high dosage (30 and 60 µM) of TQ exposure in cornea and conjunctiva cells [19].
Many studies suggest a crucial role of inflammatory and oxidative processes in the pathogenesis of DED and, in particular, tear film hyperosmolarity is accepted as a key pathogenic step, since it causes damage to the surface epithelium activating a cascade of inflammatory events at the ocular surface [48,49]. In this study, we analyzed the effect of the hyperosmolarity treatment alone or in presence of TQ and TQ-loaded liposomes at a lower concentration (5 µM) for a long time (5 h) on the cornea cell viability using LDH and MTT assays. As shown in Figure 1, the hyperosmolarity treatment did not induce cell toxicity in both conditions.  The key role of proinflammatory cytokines in the tears of patients with DED, and other ocular surface diseases, was recently suggested by studies in literature [50]. A recent systematic review revealed that DED patients had higher tear levels of IL-1β, IL-6 and TNF-α, as compared to controls [51]. Based on these considerations, we analyzed the anti-inflammatory effect of TQ, LP-TQ and LP-TQ-HA under hyperosmolarity in HCE-2 cells. Figure 2 shows the mRNA levels of IL-1β, Il-6 and TNFα in corneal cells exposed to hyperosmotic medium. The treatment significantly increased the mRNA expression of the IL-1β, Il-6 and TNFα compared to normal control cells, as previously observed in in vitro and in vivo models [28,52]. However, the expression of these proinflammatory cytokines significantly decreased in the TQ and liposomes-treated cells (Figure 2). other ocular surface diseases, was recently suggested by studies in literature [50]. A recent systematic review revealed that DED patients had higher tear levels of IL-1β, IL-6 and TNF-α, as compared to controls [51]. Based on these considerations, we analyzed the antiinflammatory effect of TQ, LP-TQ and LP-TQ-HA under hyperosmolarity in HCE-2 cells. Figure 2 shows the mRNA levels of IL-1β, Il-6 and TNFα in corneal cells exposed to hyperosmotic medium. The treatment significantly increased the mRNA expression of the IL-1β, Il-6 and TNFα compared to normal control cells, as previously observed in in vitro and in vivo models [28,52]. However, the expression of these proinflammatory cytokines significantly decreased in the TQ and liposomes-treated cells (Figure 2). Given the modifications observed at the gene level, we investigated whether these translated into protein modifications by analyzing some proinflammatory proteins, such as Il-1β, COX-2 and NF-kappa-B p65. In this model of DED in vitro, the increased level of this proinflammatory protein was reverted by the treatment with TQ, LP-TQ and LP-TQ-HA ( Figure 3). Given the modifications observed at the gene level, we investigated whether these translated into protein modifications by analyzing some proinflammatory proteins, such as Il-1β, COX-2 and NF-kappa-B p65. In this model of DED in vitro, the increased level of this proinflammatory protein was reverted by the treatment with TQ, LP-TQ and LP-TQ-HA ( Figure 3).  Many studies suggested a key role of inflammatory and oxidative processes in the pathogenesis of DED [49,50,53]. In particular, the mitochondrial function was related to the progression of DED as well as the outcome of this disease to the modulation of mitochondrial homeostasis [54]. DED patients had high levels of reactive oxygen species (ROS) and increased inflammatory markers in the tear film [55]. The oxidized protein levels were Figure 3. The effects of TQ on proinflammatory signaling in HCE-2 cells exposed to hyperosmotic media. Cells were exposed to 450 mOsM for 5 h and then processed for WB. The TQ-free and formulated were present in the incubation medium during hyperosmolarity exposure. Many studies suggested a key role of inflammatory and oxidative processes in the pathogenesis of DED [49,50,53]. In particular, the mitochondrial function was related to the progression of DED as well as the outcome of this disease to the modulation of mitochondrial homeostasis [54]. DED patients had high levels of reactive oxygen species (ROS) and increased inflammatory markers in the tear film [55]. The oxidized protein levels were increased due to the high inflammatory activity in DED [5]. Key events in the pathogenesis of this disease were the tear film instability, its hyperosmolarity, the ocular surface damage and inflammation [49]. Hyperosmolarity caused damage to the epithelium surface by activating the cascade of inflammatory events in the ocular surface and, moreover, it induced a release of inflammatory mediators into the tears [56]. In the form of hyperosmolar culture medium, hyperosmolarity also induced oxidative stress in cultured primary HCE-2 cells. Additionally, oxidative stress affected corneal epithelial cells directly, causing irreversible macromolecular alterations and oxidative modifications of nuclear acids, lipids and proteins, and indirectly, via the increased expression of proinflammatory cytokines [28,57].
The prolonged local inflammation induced by oxidative stress was often the cause of the corneal injury [58]. An imbalance between antioxidant defenses and ROS, produced in large part by mitochondria, induces oxidative stress [59]. Chen and colleagues in 2022 demonstrated an improvement in the effects of DED reversing hyperosmolarity-mediated mitochondrial dysfunction in HCE-2 [60]. For this reason, we focused our attention on mtROS levels.
In this study, oxidative stress was determined by the quantification of mtROS levels in HCE-2 cells through the MitoSOX™ Red mitochondrial assay that detected the superoxide in the mitochondria of live cells. The measurement of mtROS levels revealed that hyperosmotic stress markedly stimulated ROS production, and this effect was reverted by the coincubation with TQ, LP-TQ and, significantly, with LP-TQ-HA at the concentration of 5 µM for 5 h, as shown in Figure 4. As previously reported, LP-TQ-HA reduced the toxicity that TQ showed at high doses in HCE-2 cells and improved the absorption at the nucleus level with a more pronounced effect for HA-coated liposomes [19]. All these results confirm that these liposomes could represent a valid therapeutic agent against DED.
Many studies reported that TQ has potential protective effects on cardiovascular disorder and mitochondrial protection, acting against oxidative stress and organelle damages [61,62]. Moreover, the antioxidant and anti-inflammatory effects of TQ have been clearly proven in many other diseases [63]. Liang and colleagues showed that TQ played an inhibitory role in UVA-induced oxidative stress, inflammation and the mitochondrial apoptosis of human skin keratinocytes [64]. The antioxidant properties of HA are well known, as reported by Almalik and co-workers in 2018, where it was shown that HA is involved in inflammatory response, due to its antioxidant scavenging activity [65,66]. According to these data, in our work, HA-coated liposomes significantly reduced the mtROS levels induced by 450 mOsM for 5 h, and this could be explained by a possible synergistic effect of TQ with HA. Furthermore, HA can also increase the adhesion to the cell due to its mucoadhesive properties [67]; in addition, it could also act as scavenger, since it is a negative-charged polysaccharide with a reductive end [68,69].

Conclusions
In the present investigation, liposomes were used as carriers to improve TQ solubility and availability at the ocular level. It investigated the efficacy of TQ liposomal formulations in a model of DED in vitro. The liposomes consist of phosphatidylcholine and Plurol Oleique, and one formulation has a HA coating that provides the mucoadhesive properties to formulation. Moreover, the antioxidative and anti-inflammatory functions of TQ were used to achieve the goal in DED. The hyperosmotic treatment induced a significant increase of the proinflammatory genes and proteins expression that was significantly decreased in the cells treated with TQ, and this effect was maintained by the liposomes. The hyperosmotic stress markedly stimulated mtROS production and the coincubation with LP-TQ-HA significantly reverted this increase, highlighting the mucoadhesive and antioxidant effect of HA. These findings suggest that these liposomes may potentially be applied as a therapeutic agent against DED, after further studies in model in vivo that could better demonstrate the benefit of the HA coating.

Data Availability Statement:
The data presented in this study are available on request from the corresponding authors.