Nose to brain delivery of naringin-loaded chitosan nanoparticles for potential use in oxaliplatin-induced chemobrain in rats: impact on oxidative stress, cGAS/STING and HMGB1/RAGE/TLR2/MYD88 inflammatory axes

ABSTRACT Objectives Oxaliplatin induces chemobrain in cancer patients/survivors. Nutraceutical naringin has antioxidant and anti-inflammatory properties with low oral bioavailability. Our aim was to formulate naringin in chitosan nanoparticles for nose to brain delivery and assess its neuroprotective effect against oxaliplatin-induced chemobrain in rats. Methods Naringin chitosan nanoparticles were prepared by ionic gelation. Rats were administered oral naringin (80 mg/kg), intranasal naringin (0.3 mg/kg) or intranasal naringin-loaded chitosan nanoparticles (0.3 mg/kg). Naringin’s neuroprotective efficacy was assessed based on behavioral tests, histopathology, and measuring oxidative stress and inflammatory markers. Results Selected nanoparticles formulation showed drug loading of 5%, size of 150 nm and were cationic. Intranasal naringin administration enhanced memory function, inhibited hippocampal acetylcholinesterase activity, and corrected oxaliplatin-induced histological changes. Moreover, it reduced malondialdehyde and elevated reduced glutathione hippocampal levels. Furthermore, it decreased levels of inflammatory markers: NF-kB and TNF-α by 1.25-fold. Upstream to this inflammatory status, intranasal naringin downregulated the hippocampal protein levels of two pathways: cGAS/STING and HMGB1/RAGE/TLR2/MYD88. Conclusion Intranasal naringin-loaded chitosan nanoparticles showed superior amelioration of oxaliplatin-induced chemobrain in rats at a dose 267-fold lower to that administered orally. The potential involvement of cGAS/STING and HMGB1/RAGE/TLR2/MYD88 pathways in the mechanistic process of either oxaliplatin-induced chemobrain or naringin-mediated neuroprotection was evidenced.


Introduction
Chemobrain/Chemofog is a general term used to explain chemotherapy-induced cognitive impairments during or after the treatment experienced by cancer patients/survivors.These include difficulties with attention, concentration, planning, and working memory [1].It represents an obstacle to patients trying to get on with day-to-day activities.Understanding chemo-induced cognitive impairment pathophysiology is a challenging problem.Presently, it is theorized as the result of a neuronal injury caused by the alterations in the blood -brain barrier (BBB).This allows the cytotoxic concentrations of the drugs to reach the brain, causing oxidative stress, hippocampal structural and functional changes, and hence, altered brain connectivity precipitating disturbances in the executive and memory functions [2].
Oxaliplatin (OXP) is a third-generation organoplatinum chemotherapeutic agent widely used for the treatment of colorectal, pancreatic, and gastric cancers.It acts by disrupting DNA replication and transcription and exerts its cytotoxic effects by forming intra-strand DNA adducts.However, preclinical studies have documented that OXP may induce cognitive impairment via promoting disturbances in antioxidant mechanisms, dysregulation of cytokine and neurotransmitter levels, and consequent behavioral alterations [3].To date, no drug interventions are approved for preventing cognitive deficits associated with chemotherapy, keeping it a challenge facing medicinal research.
Flavonoids are natural polyphenolic compounds found in fruits and vegetables exhibiting various biological and pharmacological properties.Naringin (4′, 5, 7-trihydroxy flavanone 7-rhamnoglucoside, NRG) is a flavone glycoside, which is obtained from grapefruit and related citrus species.It falls under class II according to the biopharmaceutical classification system and is characterized by poor aqueous solubility [4] and considerable first-pass metabolism limiting its oral bioavailability [5].Following oral administration, it is metabolized by intestinal bacterial naringinase enzyme into the bioavailable naringenin, which is readily absorbed and can cross BBB [6].NRG was reported to have numerous beneficial biological actions including antioxidant, anti-inflammatory, and neuroprotective effects [7,8].This suggests that it might have a protective role against OXP-induced cognitive impairment, but this has not been investigated yet.
Drug delivery to the brain can be realized through nasal administration.This can be achieved directly through olfactory neural pathway and trigeminal nerves or indirectly through systemic absorption followed by crossing BBB if the drug molecule is small and lipophilic.However, drug present in the nasal cavity faces a lot of hurdles such as removal by mucociliary clearance and enzymatic degradation.Nanoparticles showed promising results in exploiting and enhancing nose to brain delivery by increasing drug's residence time and protecting it from degradation facilitating its transport to central nervous system (CNS) [9].This was evident, where delivery of drugs treating various neurological diseases was improved using lipid and polymeric nanoparticles [10][11][12].Of particular interest are chitosan nanoparticles, which produced outstanding results in nose to brain drug delivery owing to its cationic and mucoadhesive nature [13].Chitosan (CS) is a natural polysaccharide with repeating units of D-glucosamine and N-acetyl-D-glucosamine.It is biocompatible, biodegradable, and extensively used in drug delivery applications [14][15][16].Nasally administered chitosan nanoparticles (CS NPs) enhanced pharmacodynamic response and brain targeting efficiency of rotigotine against an animal model of Parkinson's disease [17].Moreover, they increased concentration of sitagliptin levels in brain by 5-fold compared to its nasally administered solution [18].Furthermore, thiolated CS NPs mitigated scopolamine-induced amnesia in mice [19].Therefore, nasal administration of NRG-loaded CS NPs is expected to enhance its brain delivery.
So far, the effect of NRG against OXP-induced chemobrain in rats has not yet been addressed.Therefore, the aim of the present study was to evaluate the possible protective role of oral NRG, intranasal (IN) nanoformulated NRG in CS NPs, and IN NRG solution against OXP-induced hippocampus impairment and the consequently mediated memory deficits, oxidative stress, and inflammatory signaling in rats.

Preparation and characterization of naringin-loaded chitosan nanoparticles
Chitosan NPs were prepared by the ionic gelation method as previously described [20].Briefly, chitosan was dissolved overnight in 0.1% acetic acid and its pH was adjusted to 5 using 1N sodium hydroxide.A certain volume of TPP dissolved in phosphate-buffered saline (PBS, pH = 7.4), without NRG in blank NPs and with NRG in drug loaded NPs, was added drop wise to chitosan solution under stirring (AREC.X VELP Scientifica, Usmate, Italy) as detailed in Table 1.CS:TPP ratio was kept at 2:1 in all formulations as it was found to be optimum for production of small sized CS NPs [21].The obtained NPs were centrifuged at 10,000 rpm (9500 g) (Z216 MK, Hermle, Wehingen, Germany) at 20°C for 10 minutes for separation.Sedimented NPs were redispersed in PBS and adjusted to a concentration of 0.3 mg/ml for further experiments.To determine the entrapment efficiency (EE%) and drug loading (DL%), the concentration of NRG in the supernatant was measured at a wavelength of 279 nm using a UV-Vis spectrophotometer (UV-1601 PC, Shimadzu, Kyoto, Japan).This analytical method was developed and validated in our laboratory and had a linearity range of 4-40 μg/ml.The limit of detection (LOD) and limit of quantification (LOQ) were found to be 1.59 μg/ml and 4.83 μg/ml, respectively.The regression equation (R 2 = 0.9993) was: where y is the absorbance and x is the concentration of naringin in μg/ml.EE% and DL% were calculated using the following equations: where: W t is the total weight of naringin used in the formulation, W f is the weight of free naringin remaining in the supernatant, and W s is the total weight of solid content.
The prepared CS NPs were characterized for their particle size (PS), polydispersity index (PDI) and zeta potential (ZP) using Zetasizer Nano ZS (Malvern Instruments, Malvern, United Kingdom).PS was determined in disposable polystyrene cells at 25°C applying dynamic light scattering technique.While ZP was measured in disposable plain folded capillary zeta cells and was calculated through electrophoretic mobility technique applying the Helmholtz -Smoluchowski equation.
Morphological examination of naringin-loaded chitosan nanoparticles (NRG CS NPs) was carried out using electron microscopy.Sample was sonicated for 10 minutes in a bath sonicator (Crest Ultrasonics Corp., New Jersey, U.S.A).Then a few drops were loaded on carbon coated copper grid, stained by 1% phosphotungistic acid and left to dry.Afterwards the grid, loaded with the sample, was examined by high-resolution transmission electron microscope (HR-TEM) (JOEL, JEM-2100, Tokyo, Japan) operated at 200 KV.

In-vitro release of naringin from chitosan nanoparticles
1 ml of NRG CS NPs and pure NRG dissolved in PBS (pH = 7.4) at a concentration of 0.3 mg/ml were placed in a dialysis tube (MWCO 12,000-14,000, Spectra/Por ® , Spectrum Laboratories, California, U.S.A).Afterwards, the dialysis tube was sunk in 10 ml of PBS (pH = 7.4) as dissolution medium at 37°C and kept under continuous shaking in a shaking water bath (GFL 1083, Gesellschaft für Labortechnik, Burgwedel, Germany).Samples were withdrawn at 0.5, 1, 2, 4, 6, 8 and 24 hours and replaced with fresh buffer.The concentration of NRG was assayed spectrophotometrically at 279 nm using a UV-Vis spectrophotometer (UV-1601 PC, Shimadzu, Kyoto, Japan).

Animals
Male rats (Wister, 8 weeks old, 160-180 gm) were purchased from Nile Co. for pharmaceutical and chemical industries (Cairo, Egypt).They were housed in cages (85 × 45 × 25 cm 3 ), five animals per cage, in an air-conditioned atmosphere (25°C).All efforts have been made to reduce animals' suffering and use the lowest number of animals sufficient to produce reproducible results.Rats were given food and water ad libitum.Ethical guidelines of the care and use of laboratory animals (U.K. Animals Act, 1986) and Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines [22] were adopted.The study protocol was approved by the Research Ethical Committee (REC) at Faculty of Pharmacy, Ain Shams University (Cairo, Egypt) (Approval code: ACUC-FP-ASU RHDIRB2020110301 REC#123).

Experimental design
Sixty male rats were randomly distributed among 5 groups (12 rats each) and treated for 4 weeks as follows: Group I received saline (0.9% sodium chloride) i.p. twice per week and oral 0.5% carboxymethyl cellulose (CMC) in distilled water daily.Groups II-V were administered OXP (5 mg/ml in 0.9% saline) in a dose of 4 mg/kg twice weekly, i.p. for 4 weeks.Additionally, group III received NRG in a dose of 80 mg/kg orally from a stock suspension of 8 mg/ml, once daily for 28 days.Group IV received NRG formulated in CS NPs (0.3 mg/kg, intranasal from a stock formulation of 0.3 mg/ml), once daily for 28 days.Group V received NRG solution (0.3 mg/kg, intranasal from a stock solution of 0.3 mg/ml).NRG for oral administration was prepared as a suspension in 0.5% CMC in distilled water, while that for nasal administration was prepared as a solution in PBS.This is attributed to the dose-limiting solubility of NRG.IN NRG administration was done via solution drip into the nasal cavity of the two nostrils.The solution was given as multiple small volume administrations by maximum 50 µl per nostril.OXP and NRG oral doses were selected based on previous studies [23,24], while NRG nasal dose was based on the maximum permissible dose according to its EE% and DL%.Afterwards, cognitive and memory function in rats was assessed via behavioral testing as follows: locomotor activity, step-through passive avoidance, and spatial memory (y maze, and Morris water maze) tests.On day (34) of the experiment, rats were sacrificed and brain samples from all groups (n = 3) were fixed in 10% buffered formalin and embedded in paraffin for histological examination.Besides, hippocampi were dissected out and samples from different groups (n = 3) were immediately snap frozen in liquid nitrogen and stored at − 80°C for western blot analysis.The remaining specimens (n = 6) were homogenized at 1:10 (w:v) in potassium phosphate buffer (pH 7.5) for further biochemical analysis.

Locomotor activity
Locomotor activity for different groups was assessed using activity monitor (Opto-Varimex-Mini Model B, Columbus Instruments, Columbus, OH, U.S.A) on day 29.Infrared beam interruptions by rats' motor activity were sensed and counted by the apparatus.Results were expressed as counts per 5 min [25].

Memory acquisition (step-through passive avoidance)
Step-through passive avoidance apparatus (UgoBasile, Italy) with light and dark compartments was used.Rats were subjected to training session on day 29 (0.2 mA −3 sec) and testing session (no electric shock) on day 30.When a rat enters the dark chamber with four paws, an electric shock is delivered.The training session's cut-off period was 90 sec.On the testing day, the step-through latency was detected indicating memory acquisition with a cut-off period of 300 sec [26].

Short term spatial memory (y maze)
Y-shaped black wooden maze with intersecting similar arms was used.On day 29, rats were placed at the center of the apparatus and left for acclimatization for 5 min.Afterwards, rats' behavior was monitored for 5 min.When a rat entered sequentially the three arms, it is termed as 'spontaneous alteration'.Results were expressed as total arm entries (TAE) and spontaneous alternation percentage (SAP) as follows: SAP = [(number of alternations)/(TAE −2)] × 100 [27].

Long term spatial memory (Morris water maze)
Starting from day 30, long term spatial memory was assessed using a circular white pool with a 150 cm diameter (Neuroscience, Osaka, Japan).Briefly, the pool was divided into four equal quadrants and filled with water (22-25°C) where a transparent platform was placed 1 cm below the water surface in what's called the 'target quadrant.'Rats were trained for 4 consecutive days via placing each animal for 120 sec into each of the three quadrants other than the target quadrant and recording escape latencies.On day 34 (testing session), the platform was removed, and water made opaque using nonfat dry milk.Each rat was placed in the quadrant facing the target quadrant and probe trial was detected in sec [28].

Histological examination
Hippocampi from different groups were fixed in 10% neutral buffered formalin for 24 h.Dehydration was then performed using serial dilutions of methyl, ethyl, and absolute ethyl alcohols.Afterwards, samples were made clear by xylene and paraffinized (56°C for 24 h).Slide microtome (Leica RM 2135 BioCut Rotary Microtome, Ramsey, U.S.A) was used to cut tissue blocks at 5 µm thickness.Staining was performed using hematoxylin and eosin.Then, glass slides were visualized using Full HD microscopic camera (Olympus xc30, Tokyo, Japan) [29].

Acetylcholinesterase (AchE) activity
AchE activity was determined according to Ellman et al. [30].Briefly, sample homogenates (10%) in ice cold phosphate buffer (pH 7.4, 1 M) were prepared.Afterwards, phosphate buffer (pH 8) was added to tissue homogenates and incubated (37°C −5 min).After incubation, acetylthiocholine iodide and DTNB (10 mM) were added to the reaction mixture.AchE in hippocampal homogenates hydrolyzes acetylthiocholine iodide which produces acetate and thiocholine.The latter, in the presence of DTNB, gives a yellow color of thionitrobenzoic acid that was recorded at 405 nm for 150 sec at 30 sec intervals.AchE activities were expressed as nM/min/g tissue.

Oxidative stress markers
Reduced glutathione (GSH) and malondiladehyde (MDA) levels were determined using kits purchased from Biodiagnostics (Giza, Egypt).Assessment of GSH was performed in accordance with Beutler et al. [31].GSH determination principle depends on the reduction of DTNB with GSH in sample homogenates forming a yellow-colored compound.The latter's color intensity is directly proportional to GSH concentration, and its absorbance can be detected at 405 nm.The results were expressed as nmol GSH/g tissue.Lipid peroxidation was determined as MDA according to the method of Satoh [32].Briefly, the reaction mixture (0.2 ml homogenate and 1.0 ml 0.6% thiobarbituric acid (TBA)) was heated in a boiling water bath (30 min).After cooling, the absorbance was measured at 534 nm.The results were expressed as nmol MDA/g tissue using 1,1,3,3-tetraethoxypropane as standard.

Inflammatory markers
The levels of p105 subunit of nuclear factor kappa-light-chainenhancer of activated B cells (p105 NF-κB) and tumor necrosis factor-alpha (TNF-α) were assessed using ELISA kits purchased from Elab science, US (Catalogue No. #E-EL-R0673 and E-EL-R2856).Results were expressed as pg/g tissue.

Cytotoxicity assay
To exclude a possible inhibitory interaction between NRG and OXP that could affect OXP cytotoxic potential, this assay was performed.The HCT-116 colorectal cancer cell line was purchased from Nawah Scientific Inc. (Mokatam, Cairo, Egypt).Cells were maintained in Roswell Park Memorial Institute Medium (RPMI) media supplemented with 100 mg/ml of streptomycin, 100 units/ml of penicillin and 10% of heat-inactivated fetal bovine serum in humidified, 5% (v/v) CO 2 atmosphere at 37°C.Sulforhodamine B (SRB) assay was used to test cell viability where 100 μL of cell suspensions (5×10^3cells) were incubated in 96-well plate for 24 h.OXP and NRG were dissolved in DMSO at a stock concentration of 100 mM.Afterwards, cells were treated with six different concentrations of OXP and NRG: 0.01, 0.1, 1, 10, 100, and 1000 µg/ml for 72 h.Then, SRB testing was adopted as mentioned previously [33].
Absorbance was measured at 540 nm using BMGLABTECH ® - FLUOstar Omega microplate reader (Ortenberg, Germany).Half-maximal inhibitory concentration (IC 50 ) was determined using GraphPad Prism software, version 7 (GraphPad Software, Inc. La Jolla, CA, U.S.A).Then, using the IC50 values obtained, combinations of OXP and NRG were tested using the same protocol to indicate the effect of NRG on the cytotoxic effect of OXP.Results were compared to control and experiments were performed in triplicates.

Statistical analysis
Data are presented as mean ± SD.Passive avoidance test results (non-parametric) were analyzed using Kruskal-Wallis test followed by Dunn's as post hoc test.Morris-water maze training results were analyzed by two-way ANOVA followed by Bonferroni post hoc test.For the rest of data, multiple comparisons were performed using one-way ANOVA followed by Tukey as a post-hoc test.A 0.05 level of probability was used as the criterion for significance.Graphs were sketched using GraphPad Prism software version 7 (GraphPad Software, Inc., La Jolla, CA, U.S.A).

Preparation and characterization of naringin-loaded chitosan nanoparticles
Chitosan NPs were successfully prepared in 2 different combinations: 0.1% CS (5 ml) mixed with 0.125% TPP (2 ml) and 0.2% CS (5 ml) mixed with 0.25% TPP (2 ml) keeping a constant ratio of CS:TPP at 2:1.Starting with EE% and DL%, at 0.1% CS, the EE% and DL% increased significantly as the amount of NRG increased (Figure 1a).The highest EE % of 24% was obtained at 1 mg NRG then decreased to 17% at 2 mg NRG.While for DL%, the highest value of 5% was recorded at 2 mg NRG compared to 3% for 1 mg NRG.Regarding 0.2% CS, lower EE% and DL% values were noticed compared to 0.1% CS formulations (Figure 1d).Formulation containing 2 mg NRG showed highest EE% and DL% with values of 11% and 2%, respectively.No significant difference was found for both EE% and DL% among different amounts of NRG for 0.2% CS formulations.0.1% CS produced smaller NPs than 0.2% CS, where all 0.1% CS formulae were below 200 nm (Figure 1b, Supplementary Figure S1).While PS of 0.2% CS formulations was between 200 and 500 nm (Figure 1e, Supplementary Figure S2).It is also worth mentioning that formulations containing 0.2% CS suffered from aggregation followed by sedimentation upon standing for few hours after preparation suggesting instability.Formulations containing 0.1% CS did not display such behavior.No significant difference in PS and PDI was found between formulations containing different amounts of NRG at 0.1% CS.On the contrary, adding NRG to 0.2% CS formulations increased PS and PDI significantly in comparison to blank 0.2% CS.
All prepared CS NPs had a positive zeta potential.No significant difference in ZP was found between 0.1% CS formulations (Figure 1c).Oppositely, all 0.2% CS formulations containing NRG displayed significantly lower, but practically irrelevant ZP compared to blank 0.2% CS preparation (Figure 1f).Morphological examination of CS NPs by TEM revealed that NPs were spherical in shape with a smaller size than that recorded by dynamic light scattering technique and presence of aggregates (Figure 2).

In-vitro release of naringin from chitosan nanoparticles
NRG solution reached 100% release after 2 hours, while NRGloaded CS NPs sustained NRG release releasing 75% after 24 hours (Figure 3).

Locomotor activity
One-way ANOVA revealed no significant difference between different treatment groups.This could exclude the effect of locomotor activity on other behavioral tests (Figure 4a).

Memory acquisition (step-through passive avoidance)
No significant difference in step-through latency was detected when comparing all treatment groups in the training session (Figure 4b).Concerning testing day results (Figure 4c), a 3.39-fold significant decrease in step-through latency was found when comparing OXP-treated and control groups.By contrast, IN NRG solution treatment revealed a significant elevation in step-through latency relative to the corresponding OXP-treated group by 3.23 folds.

Short-term spatial memory (Y maze)
Y maze test was determined in terms of TAE (Figure 4d) and SAP (Figure 4e).Concerning TAE, there was no significant difference detected among different treatment groups.However, OXP treatment showed significant reduction in SAP by 3.48 folds as compared to control group.Only the group of rats treated with IN NRG formulated in CS NPs illustrated a significant increase in SAP by 4.2 folds relative to OXPtreated rats.

Long-term spatial memory (Morris water maze)
For the training sessions, two-way ANOVA revealed a significant decrease in the latency time to reach the platform in the fourth training day in control and all NRGtreated groups relative to the corresponding OXP-treated group (Figure 4g).Regarding the testing session (Figure 4f), rats treated with OXP presented a significant reduction in probe trial as compared to the control group by 1.96 folds.On the other hand, both IN nanoformulated NRG CS NPs-treated and IN NRG solution-treated groups  illustrated significant amelioration in probe trial relative to the OXP-treated group by 1.79 and 1.86 folds, respectively.

Histological examination
Microscopic examination of dentate gyrus region of the hippocampus of control group (Figure 5a) showed normal histological structure.This was evidenced by the presence of compact granular cells with dark nuclei.In addition, the molecular layer revealed normal glial cells, as well as pyramidal cells.High power examination of hippocampus revealed three layers: molecular, pyramidal, and polymorphic layers.The tissue section of hippocampal dentate gyrus region of OXP-treated group (Figure 5b) illustrated marked cellular disorganization and shrinkage in size of large pyramidal cells, with darkened nuclei.Moreover, granular cell layers also showed shrinkage, darkly stained, spindle-shaped nuclei in a diffuse manner.Histological section of hippocampal dentate gyrus region of oral NRG-treated group (Figure 5c) revealed shrinkage in size of large pyramidal cells, with darkened nuclei.However, granular cell layers presented normal histological arrangement.Hippocampal dentate gyrus region of group IV treated with NRG formulated in CS NPs (Figure 5d) showed cellular organization and shrinkage in size of pyramidal cells, with darkened nuclei.Additionally, vacuolation of granular cell layers was also noticed.Group of rats treated with IN NRG solution revealed cellular organization with shrinkage in size of large pyramidal cells of hippocampal dentate gyrus region.

Acetylcholinesterase (AchE) activity
Group II rats presented a significant enhancement in hippocampal AchE activity relative to the control rats by 39.76%.On  the contrary, treatment with either IN nanoformulated NRG CS NPs or IN NRG solution illustrated a significant reduction in AchE activity in the hippocampus by 16.34% and 36.63%,respectively, compared to the OXP-treated group.Interestingly, both IN NRG treated groups; IV and V, showed significant reduction in AchE activity compared to oral NRGtreated group by 14.27% and 35.07%, respectively.In addition, rats treated with IN NRG solution demonstrated a significant lessening in AchE activity relative to IN nanoformulated NRG CS NPs-treated rats by 24.26% (Figure 6a).

Oxidative stress markers
The effect of OXP and NRG on hippocampal oxidative stress was estimated via measurement of GSH and MDA levels.As shown in Figure 6b, a 1.24-fold significant reduction in GSH level was found between control group and OXP-treated groups.On the other side, treatment with oral NRG or IN nanoformulated NRG CS NPs demonstrated a significant similar increase in GSH level as compared to OXP-treated rats by 1.22 folds.Although IN NRG solution seemed to restore GSH levels to that of control group, its effect was not statistically significant compared to OXP-treated group.Regarding MDA level in the hippocampus (Figure 6c), group II rats showed significant elevation in lipid peroxidation level relative to the control group by 27.96%.However, groups IV and V revealed a significant reduction in MDA level by 23.78% and 38.46%, respectively, as compared to group II and by 32.81% and 38.46%, respectively, relative to group III.

Inflammatory markers
Inflammatory response following OXP and NRG treatment was evaluated via p105 NF-κB and TNF-α determination in the hippocampus (Figure 7).Group of rats treated with OXP revealed a significant increase in both; p105 NF-κB and TNFα, relative to the corresponding control group by 1.5 and 1.39 folds, respectively.On the other side, treatment with either IN nanoformulated NRG CS NPs or IN NRG solution revealed a significant amelioration in p105 NF-κB relative to the OXPtreated group by 1.25, and 1.12 folds, respectively.Surprisingly, IN nanoformulated NRG CS NPs-treated rats presented a significant reduction in p105 NF-κB level relative to both oral NRG-treated and IN NRG solution-treated rats by 1.16 and 1.12 folds, respectively.Additionally, a significant reduction in TNF-α level was detected between all NRGtreated groups and OXP-treated rats.Moreover, IN nanoformulated NRG treatment showed significant reduction in TNF-α level relative to oral NRG treatment.

Western blotting
To check the effect of OXP and NRG on cGAS/STING and HMGB1/RAGE/TLR2/MYD88 pathways, western blot analysis was performed (Figure 8).Concerning cGAS/STING (Figure 8a and b), OXP treatment resulted in a significant elevation in the protein expression of both markers relative to the corresponding control group.However, treatment with either IN nanoformulated NRG CS NPs or IN NRG solution exhibited significant reduction in their protein expression compared to the corresponding chemobrain-induced group.Additionally, IN NRG solution-treated rats presented significant decrease in STING expression relative to oral NRG-treated rats.Regarding HMGB1/TLR2/MYD88 pathway (Figures 8c-f), group of rats treated with OXP alone illustrated a significant increase in the protein expression of HMGB1, RAGE, TLR2, and MYD88 as compared to the corresponding control group.On the other hand, IN nanoformulated NRG CS NPs-treated group exhibited lessening in HMGB1 protein expression relative to the corresponding OXP-treated group.Moreover, IN nanoformulated NRG CS NPs treatment showed significant reduction in HMGB1 protein expression as compared to both oral and IN NRG solution treatment.In addition, both IN nanoformulated NRG CS NPs-treated and IN NRG solution-treated rats illustrated a significant decrease in RAGE expression as compared to both OXP-treated and oral NRG-treated rats.All NRGtreated groups revealed a marked decrease in TLR2 expression relative to OXP-treated rats.Furthermore, TLR2 protein expression was found to be ameliorated with IN nanoformulated NRG CS NPs-treated group compared to oral NRG-treated rats.MYD88 protein expression was found to be reduced in IN nanoformulated NRG CS NPs-treated and IN NRG-treated rats relative to the corresponding OXP-treated group of rats.

Cytotoxicity assay
The effect of NRG treatment on the cytotoxic effects of OXP was assessed on HCT-116 colorectal cancer cell line (Supplementary Figure S3).It was found that the IC50 of OXP on HCT-116 cells = 1.11 ± 0.17 µg/ml.When NRG was tested for its cytotoxic effect on the same cancer cell line, its IC50 was found to be 740 ± 1.57 µg/ml.On combination of both OXP and NRG, the IC50 was noticed to be nonsignificantly decreased to 0.9 ± 0.09 µg/ml.

Discussion
Even though the valuable use of chemotherapeutic drugs in improving the survival of cancer patients is known, they are well known for having serious side effects which inversely affect the efficacy of anti-cancer treatment and patients' quality of life [34].Neurotoxicity is a debilitating side effect of many anticancer drugs including the platinum-based OXP.Ongoing studies are aiming to reveal the machinery behind such toxicity and to find hopefully a promising correction.As such, the current study investigated NRG neuroprotective ability both orally and nasally against OXP-induced cognitive impairment.Moreover, CS NPs were assessed as potential carriers for NRG nose to brain delivery.Additionally, the possible underlying molecular pathways were investigated.Generally, IN administration of NRG either as nanoformulation in CS NPs or solution showed a comparable neuroprotective role against OXP-induced cognitive impairment.However, both were superior to the higher dose of NRG orally administered, in ameliorating OXP-cognitive impairment in rats.Thus, emphasizing the impact of changing the route of administration and evading the first-pass effect.
Ionic gelation was the method of choice for preparation of CS NPs, as it is easy, simple and avoids the usage of organic solvents and application of heat.It depends on the electrostatic attraction between the cationic CS and anionic TPP.A delicate balance is required between the ratios of CS and TPP.At low quantities of both, a clear solution will be obtained indicating failure in formation of NPs, while at high quantities aggregates will be produced [35].This indicates the presence of an optimum CS:TPP ratio for formation of NPs.In our study, the CS:TPP ratio was fixed at 2:1 as it was proven to be ideal for production of CS NPs with a PS of about 100 nm [21].Increasing chitosan molecular weight and concentration was reported to be accompanied by enlargement of particles size and reduction of ZP [36,37].This agrees with our results, where 0.2% CS produced larger NPs with lower ZP than 0.1% CS.As CS molecular weight and concentration increase, the solution viscosity increases, which results in the formation of bigger particles [38].Additionally, higher CS concentration results in close packing of CS molecules causing entanglement and hindering dispersion of TPP molecules within CS increasing PS [21].
Increasing CS concentration from 0.1% to 0.2% was associated with reduction in EE% and DL% conforming to previous results [36].More CS molecules might have limited the space available within the particles for the accommodation of the guest molecule.In addition to increasing solution viscosity, which might have hindered polymer dispersion and its capacity to incorporate NRG during preparation [39].
Examination of CS NPs under TEM reported their tendency to form aggregates [37] in agreement with our findings.Similarly, it was common to observe smaller PS than that recorded by dynamic light scattering in TEM images for CS NPs [21,40].This can be related to NPs shrinkage after drying, required before TEM examination, and the difference in the measurement technique, where dynamic light scattering measures the hydrodynamic diameter.Pure NRG was released completely within 2 hrs as it was completely soluble in PBS at the prepared concentration (0.3 mg/ml).However, CS NPs retarded NRG release as it had to diffuse through the NPs onto the dissolution medium.
The protective efficacy of NRG administered orally and nasally either as solution, or nanoformulated in CS NPs against OXP-induced chemobrain was assessed based on behavioral tests, histological examination, and measurement of certain biochemical markers.With regards to the cognitive function, OXP impaired learning, short-term memory, long-term memory, memory acquisition, and spatial memory in rats as indicated by step-through passive avoidance, y maze, and Morris water maze tests.These findings agree with previous studies [41][42][43][44][45] which reported the negative effects of OXP on memory in rats.NRG-treated groups reversed OXP decline effects, with subsequently improved learning and memory acquisition.This follows other research groups which reported the memory ameliorating potential of NRG in different models [46][47][48][49][50][51].IN nanoformulated NRG CS NPs exhibited superior results as compared to orally treated NRG both in short-term and longterm spatial memory represented by y maze and Morris water maze tests, respectively.IN NRG solution was superior to oral only with respect to long term spatial memory.The distinction between orally and IN-treated NRG groups was not observed in step-through passive avoidance test, where all oral, IN nanoformulated NRG CS NPs, and IN NRG solution displayed almost similar behavior.The OXP-induced memory deficits were consistent with the hippocampal damage revealed by histological staining.NRG treatment abrogated such histopathological changes with similar efficacy in both oral and INtreated groups.
To explore the mechanism behind the declined cognitive function, biochemical investigation of AchE activity, a basic indicator of neural function, was carried out.OXP escalated AchE activity which might explain its involvement in the physiopathology of OXP-induced cognitive impairment.Similar observation was recently reported in an experimental model of OXP-induced peripheral neuropathy [52].NRG has previously succeeded in preventing the increase in AchE activity stimulated by exposure to cisplatin [47].In our study, both IN nanoformulated NRG CS NPs and IN NRG solution corrected for AchE activity with superior action by the IN NRG solution.However, orally treated NRG at a dose of 80 mg/kg failed to lessen the elevated AchE levels.This contrasts with previous findings [23] which displayed an inhibitory action of oral NRG at same dose on elevated AchE activity in an experimental model of manganese-induced neurotoxicity.
Both direct and indirect chemotherapy-induced cytotoxic effects contribute to chemobrain.On the central level, some chemotherapeutic agents can directly cross the BBB, mediating direct toxic effects.Peripherally, chemotherapy induces an oxidative stress status accompanied by the overproduction of reactive oxygen species (ROS) and pro-inflammatory cytokines which trigger several molecular mechanisms.This can compromise the tight junctions integrity of the BBB enabling the entrance of more cytokines and chemotherapeutic drugs causing brain damage [34,53].Whether OXP induces both direct and indirect neurotoxicity is not clear and needs further investigations.In one study, OXP has shown a limited ability to cross the BBB where only small levels could be detected upon brain examination of treated non-human primates [54].Another study reported the tendency of OXP to cause disassembly of the tight junction proteins, accompanied by increased ROS levels and subsequent dysfunction in a rat brain endothelial cell line [55].Regardless of the way, it is agreed on the presence of key players which mostly contribute to OXP-mediated neurotoxic actions, one of which is oxidative stress [24,52,56].This could be attributed to the tendency of platinum compounds to bind mitochondrial DNA, forming potentially lethal bifunctional lesions which fail to get repaired, causing DNA damage and subsequent increased reactive oxygen species.Together with activation of signaling pathways and compromised antioxidant defense mechanism, redox imbalance prevails [57,58].In the current investigation, the effects of OXP on GSH and lipid peroxidation levels in hippocampi provided evidence that OXP-induced oxidative stress co-exists with and contributes to its influence on cognitive impairment.The beneficial role of NRG in attenuating oxidative stress, an essential mediator of neurobehavioral disorders, is nicely reviewed by Viswanatha et al. [59].In line with this, the IN nanoformulated NRG CS NPs successfully abolished the OXP-induced changes in MDA and GSH, thereby restoring the antioxidant defense system.On the other side, IN NRG solution significantly abrogated the changes in MDA, while the orally treated NRG attenuated those of GSH levels.
Oxidative stress could activate nuclear factor-kappa B (NF-κB), a transcription factor implicated in the development and exacerbation of inflammatory reactions at diverse tissue locations.This is followed by formation of pNF-κB which then translocates to the nucleus and binds to the target gene's matching sequence, such as pro-inflammatory factors, promoting the expression of inflammatory cytokines as TNF-α, IL-1ß, and IL-6 [24].Our findings support the upregulation of p105-NF-κB and TNF-α levels by OXP.With a favorable antiinflammatory neuroprotective potential, NRG has abolished the doxorubicin-mediated inflammatory status [48].In consistency, IN nanoformulated NRG CS NPs treatment downregulated both cytokines' levels with more prominent effects as compared to orally treated NRG and as compared to IN NRG solution in case of p105-NF-κB.
Besides oxidative stress, another stimulator of inflammatory cytokines release is the cGAS/STING signaling pathway.The cyclic GMP-AMP synthase-stimulator of interferon genes (cGAS-STING) axis, a significant innate immune signaling system, has recently been shown to be critical for intrinsic antitumor immunity [60][61][62][63].Signal transmission begins with cGAS which catalyzes the production of cyclic dinucleotides (CDNs) upon detection of aberrant double-stranded DNA (dsDNA) fragments in the cytoplasm.Then CDNs bind to STING, activating it with subsequent triggering of interferon regulatory factor 3 (IRF3) and NF-κB and release of interferons and inflammatory cytokines.As part of their antitumor therapeutic efficacy, chemotherapeutic agents activate the cGAS/STING pathway where OXP has shown such potential in human hepatocellular carcinoma cells [63].Our data supports that study and provides evidence on the contribution of the cGAS/STING signaling pathway to OXP-induced central neurotoxicity.For the first time, our data provides evidence on the ameliorating action of NRG on this pathway where IN NRG either nanoformulated in CS NPs or in solution have displayed almost similar decreases of protein expressions of cGAS and STING.
Emerging evidence has revealed the key role played by the macrophage-derived high mobility group box 1 (HMGB1), a damage-associated molecular pattern (DAMP) protein, in the progression of OXP-induced peripheral neurotoxicity in rodents [64].Depending on the redox states, HMGB1 exists in two active forms, a fully reduced form which activates the receptor for advanced glycation end products (RAGE) and an oxidized form which activates Toll like receptors (TLRs) 2 and 4, which in turn upregulate the expression of myeloid differentiation factor 88 (MYD88), activating NF-κB signaling [65,66].A possible implication of HMGB1/RAGE/TLR2/MYD88 signaling cascade in OXP-induced cognitive deficits is provided in the present study as evidenced by their upregulation in OXP-treated rats.Supporting to previous reports on NRG action at these signaling markers [67][68][69], IN NRG either nanoformulated in CS NPs or solution decreased the protein expression levels of all of these markers with almost similar efficacy except for HMGB1, which the IN NRG solution could not downregulate.Expression levels of HMGB1 and TLR2 reaching those of the control were achieved by the IN nanoformulated NRG CS NPs, suggesting its powerful impact on that pathway.On the other side, our findings reveal a weak activity of NRG on this pathway when administered orally.This might be related to the dose, where higher oral doses of 100 mg/kg could be needed to observe the effect [69].Besides studying naringin neuroprotective potential, it was necessary to test whether naringin has a disturbing or inhibitory action on the cytotoxic activity of oxaliplatin or not, particularly for oral NRG administration.NRG was previously reported to inhibit the proliferation of colorectal cancer cells [70].In line with this, our data supports its cytotoxic potential but with much less potency as compared to OXP.Importantly, when combined, NRG seems not to interfere with OXP tumor killing ability as evidenced by the cytotoxicity assay on the human colorectal carcinoma cell line HCT116.This indicates that coadministration of NRG with OXP during cancer management should not disturb the cytotoxic action of the latter.
IN administration of NRG at a dose of 0.3 mg/kg either as solution or nanoformulation in CS NPs enhanced its protective action against the neurotoxicity induced by OXP in comparison to orally administered NRG at a dose of 80 mg/kg.This was evident for behavioral tests, histopathology, and biomarkers measured as indicators for oxidative stress and inflammation.It is worth mentioning that the enhancement magnitude of changing the route from oral to nasal exceeded that of changing the NRG formulation from nasal solution to nasal CS NPs.IN route is known for its fast systemic absorption and avoidance of first-pass effect, which was found to be substantial for NRG in rats [5].This could have enhanced NRG bioavailability allowing it to reach CNS as it can cross BBB [71].This might have been further augmented by BBB disruption caused by OXP.In addition, a direct nose to brain delivery is a possibility, where NRG might have reached the brain through paracellular or transcellular route across the olfactory epithelial cells or neurons [13].This pathway should allow NRG to reach the brain directly bypassing the BBB, improving its local concentration in the CNS.Moreover, NPs are known to enhance this by protecting the encapsulated drug from degradation and efflux from cells.Furthermore, CS is known for its mucoadhesive properties.Its cationic nature facilitates electrostatic interaction with the negatively charged nasal mucosa increasing the residence time.Additionally, it can transiently open tight junction allowing paracellular transport of small molecules [72].Whether the enhanced delivery and subsequently the protective action of NRG was achieved through systemic absorption or direct nose to brain delivery, or both will need further investigation to be fully elucidated.

Conclusion
By virtue of its numerous favorable pharmacological potentials, naringin proved to have a neuroprotective effect against oxaliplatin-induced chemobrain in rats.Its intranasal administration and nanoformulation in chitosan nanoparticles enhanced its efficacy and showed superior potency compared to its oral administration at a dose 267-fold lower.With respect to the growing demand for therapeutic candidates of natural origin and with few adverse effects, naringin is a promising nominee to be tested clinically in patients on chemotherapy.

Figure 4 .
Figure 4. Effect of oral naringin, intranasal nanoformulated naringin in chitosan nanoparticles, and intranasal naringin solution on locomotor activity (a), stepthrough passive avoidance training (b), step-through passive avoidance testing (c), total arm entries (d), spontaneous alterations (e), probe trial testing (f), and probe trial training (g) in oxaliplatin-induced chemobrain in rats.Data are presented as mean ± SD (n = 6) where: a, and b; statistically significant from control group and oxaliplatin-treated group, respectively, at p ≤ 0.05 using one way ANOVA followed by Tukey as a post-hoc test (a, c, f, g), repeated measures two-way ANOVA followed by Bonferroni test (b), and Kruskal Wallis followed by Dunn's test (d, e).

Figure 6 .
Figure 6.Effect of treatment with naringin on hippocampal acetylcholinesterase activity (a), reduced glutathione (b), and malondialdehyde (c) in an experimental model of chemobrain induced by oxaliplatin in rats.Data are presented as mean ± SD (n = 6) where: a, b,c, and d; statistically significant from control, oxaliplatintreated, oral naringin-treated, and intranasal nanoformulated naringin in chitosan nanoparticles treated groups, respectively, at p ≤ 0.05 using one way analysis of variance (ANOVA) followed by Tukey as a post-hoc test.

Figure 7 .
Figure 7. Effect of treatment with naringin on hippocampal inflammatory markers; p105 NF-κB (a) and TNF-α (b) in an experimental model of chemobrain induced by oxaliplatin in rats.Data are presented as mean ± SD (n = 6) where: a, b,c, d and e; statistically significant from control, oxaliplatin-treated, oral naringin-treated, intranasal nanoformulated naringin in chitosan nanoparticles treated and intranasal naringin solution treated groups, respectively, at p ≤ 0.05 using one way analysis of variance (ANOVA) followed by Tukey as a post-hoc test.

Table 1 .
Composition of the prepared formulations of chitosan nanoparticles.