Anthocyanin Encapsulated Nanoparticles as a Pulmonary Delivery System

Anthocyanins are known for their therapeutic efficacy for many human diseases, including cancer. After ingestion, anthocyanins degrade due to oxidation and enzymatic breakdown, resulting in reduced therapeutic efficacy. Direct delivery to target tissues and entrapment of anthocyanins increases their stability, bioavailability, and therapeutic efficacy. The objective of the present study was to develop a direct delivery system of anthocyanins into pulmonary tissues via encapsulated nanocarriers. A cyanidin-3-O-glucoside (C3G)-rich anthocyanin extract was prepared from well-ripened haskap (Lonicera caerulea L.) berries (HB) and encapsulated in three different polymeric nanocarrier systems: polyethylene glycol-poly(lactide-co-glycolide), maltodextrin, and carboxymethyl chitosan (CMC). The anthocyanin encapsulation efficiency was significantly higher in CMC (10%) than in the other two polymers. The cytotoxicity and cytoprotective effect of HB anthocyanin-encapsulated CMC (HB-CMC, 4 μg of C3G equivalent anthocyanin in 2 mg/mL nanoparticle) and anthocyanin-free CMC (E-CMC, 2 mg/mL) were tested for cytotoxicity using human normal lung epithelial BEAS-2B cells. The CMC nanoparticles were not cytotoxic for BEAS-2B cells. The HB-CMC nanoparticles reduced carcinogen-induced oxidative stress in BEAS-2B cells and restored the expression of superoxide dismutase and glutathione peroxidase enzymes. The HB-CMC nanoparticles also reduced carcinogen-induced DNA single-strand breaks and alkaline-labile sites but not the double-strand breaks. The E-CMC, HB-CMC (28 μg C3G equivalent/mouse/day for six days), or the same dose of free HB anthocyanin was administered to A/JCr mice through a nose-only passive inhalation device. C3G and its metabolites, cyanidin, peonidin-3-O-glucoside, and cyanidin-3-O-glucuronide, were detected by UPLC/ESI/Q-TOF-MS in the lungs of mice after one hour of exposure. Therefore, the CMC could be a promising noncytotoxic candidate to encapsulate HB anthocyanin. Direct delivery of anthocyanin to lung tissues enhances tissue retention, slows phase 2 metabolism, and improves therapeutic efficacy.


Introduction
Natural and synthetic biopolymers are being widely used in synthesizing nanoparticles for pharmaceuticals. Polyethylene glycol-poly(lactide-co-glycolide) (PEG-PLGA), maltodextrin (MDX), chitin and its derivatives, gelatin, pectin, and cellulose are some of the popular biopolymers [1,2]. Besides enhancing the bioavailability, nanocarriers help the encapsulants escape from enzymatic digestion and phagocytosis. Anthocyanin, composed of two phenyl benzo pyrylium (flavylium) salt derivatives and hydroxyl and methoxy groups, possesses anticancer, antioxidative, and anti-inflammatory properties [3][4][5]. Once ingested, anthocyanins undergo various structural modifications due to acidic pH in the stomach, enzymatic digestion, and microbial biotransformation, resulting in poor therapeutic efficacy as a dietary source [6][7][8]. For instance, only 1.7% of cyanidin 3-Oglucoside (C3G) was bioavailable after administering 500 mg/kg C3G to C57BL/6 J mice. Minor levels of C3G were detected in mouse tissues, 28.3 nmol h/g in the lung, 3,655 nmol h/mL in urine, and 35.3 nmol h/g in the liver within one hour of ingestion [9]. In healthy adults, serum C3G levels were 0.31 μmol h/L after 48 h of consuming 500 mg C3G. A higher bioavailability (about 12%) of C3G and its metabolic derivatives was reported in humans [10]. Direct delivery of anthocyanins into target tissues avoiding the oral route enables therapeutic dose provision and enhances therapeutic efficacy in clinical settings.
Haskap berry (HB) is rich in anthocyanins, especially C3G. Anthocyanin-rich HB reduced carcinogen-induced DNA damage, oxidative stress, and lung tumorigenesis [11][12][13]. PEG-PLGA, MDX, and carboxymethyl chitosan (CMC) were selected to encapsulate C3G-rich anthocyanins extracted from HB. The amphiphilic PEG-PLGA di-block polymer resists opsonization and phagocytosis, thereby sustaining drug circulation in the blood [14][15][16]. MDX is a carbohydrate-based polymer with D-glucose units linked by α(1→4) or α(1→6) glycosidic bonds. MDX is inexpensive and widely used in the food industry due to its good biocompatibility and biodegradability [17]. Chitosan-based encapsulants have various applications in the delivery of drugs for the treatment of cancer, gastrointestinal diseases, and pulmonary diseases. Deacetylation of chitosan generates water-soluble CMC, which has carboxyl groups [18]. The CMC nanoparticles generally have a positive surface charge and mucoadhesive properties, facilitating drug release by adhering to the mucous membrane [19]. Therefore, we hypothesized that direct delivery of HB anthocyanins to lung tissues will enhance its bioavailability and therapeutic efficacy against lung carcinogenesis. The present study demonstrates encapsulation of the anthocyanins extracted from HB in PEG-PLGA, MDX, and CMC by emulsification-solvent evaporation, antisolvent nanoprecipitation, and ionic gelation with CaCl 2, respectively. The primary objective of the study was to facilitate the direct delivery of HB anthocyanins into pulmonary tissues via encapsulated nanocarriers. North York, ON, Canada. Bovine collagen type I (PureCol® 5409) was purchased from Advanced BioMetrix, Carlsbad, CA, USA. 7-Aminoactinomycin D (7AAD, cat. no. 00-6993-50) was purchased from eBioscience Inc. San Diego, CA, USA. PLGA-PEG (5,000 : 7,000, cat. no. 765139), sodium deoxycholate (cat. no. D6750), MDX (dextrose equivalent, 4.0 -7.0, cat no. 419672), sodium dodecyl sulfate (cat no. L3771), Tween® 80 (cat no. P1754), and dichloromethane (DCM, cat no. 270997) were purchased from Sigma-Aldrich, Oakville, ON, Canada. CMC (cat. no. sc-358091) with a 90% deacetylation degree was purchased from Santa Cruz Biotechnology, Mississauga, ON, Canada. All the other chemicals used were of analytical grades. The anti-phospho-H2AX, anti-H2AX, anti-phospho-Nrf2, anti-Nrf2, superoxide dismutase (SOD), catalase, glutathione peroxidase (GPx), goat anti-mouse IgG, goat anti-rabbit IgG, rabbit anti-goat IgG, and anti-actin antibodies were from Cell Signaling The experiment was conducted in three stages. First, HB anthocyanin-encapsulated nanoparticles were prepared and characterized for their physical and chemical properties. Secondly, their cytoprotective properties were tested in cultured normal lung epithelial cells in vitro. Finally, the lung bioavailability of HB anthocyanin-encapsulated nanoparticles was tested in mice following pulmonary delivery through passive inhalation.

Preparation of Anthocyanin-Rich HB Extract.
Wellripened fresh HB (Brix value 14-16%) were obtained from the LaHave Natural Farms (Blockhouse, NS, Canada) and frozen at -20°C. The frozen berries (100 g) were ground in 500 mL deionized water (1 : 5, w/v) in a commercial blender (HBB909, Hamilton Beach Brands Inc., Glen Allen, VA, USA) for 20 min, and the juice was filtered under a vacuum. Then, the juice was purified by solid-phase column chromatography to obtain a sugar-free anthocyanin-rich HB extract (Brix value ≤0.1). The column consisted of brominated styrenic adsorbent with >250 μm particle size, 210 Å porosity, 630 m 2 /g surface area, 43-53% water content, and 780 g/L bulk density. The Brix value was measured by a refractometer. Once the water-soluble sugar fraction was eluted completely, the column-bound nonsugar anthocyanin fraction was eluted with 20%, 70%, and 95% ethanol gradient. The combined anthocyanin-rich mixture was rotary evaporated (Heidolph RotaChill, UVS400-115, Thermo Electron Corporation, Milford, MA, USA), followed by further drying under a gentle nitrogen flow (N-EVAP™ 111, Organomation Associates Inc., Berlin, MA, USA). The concentrated samples were lyophilized (Dura-Dry™ MP FD-14-85BMP1, DJS Enterprises, Markham, Ontario, Canada) under a 3,600 mT vacuum at -20°C for 24 h, and the dried powder was stored at -80°C in amber bottles until use.  [14]. Briefly, HB anthocyanin and PEG-PLGA were dissolved (1 : 10 and 1 : 20 w/w) in 3 mL of DCM at room temperature under continuous stirring for approximately 30 min. This mixture was added to a 12 mM sodium deoxycholate solution (20 mL) on an ice bath, and the mixture was probe-sonicated (model 505 Sonic, Thermo Fisher Scientific Inc., Waltham, MA, USA) at 40% power for 5 min. Then, the organic phase was evaporated entirely, and nanoparticles were separated by centrifugation (Sorvail ST 16, Thermo Fisher Scientific Inc., Waltham, MA, USA) at 14,000×g for 10 min at 4°C. The particle pellet was washed twice with deionized water and lyophilized. Anthocyaninfree PEG-PLGA nanoparticles were also prepared following the same method without HB.

MDX Nanoparticles.
The MDX nanoparticles were prepared as described by Qui and colleagues [17]. A solution of MDX (5% w/v) was prepared with 0.5% (v/v) Tween 80%. The HB anthocyanin was mixed with MDX solution (1 : 10 and 1 : 20 w/v). While stirring continuously (600 rpm), absolute ethanol (100 mL) was added dropwise to 10 mL of the above suspension. The suspension was centrifuged at 14,000×g at 4°C, rinsed with absolute ethanol twice, and lyophilized. The anthocyanin-free MDX nanoparticles were made following the same procedure without HB anthocyanin.
2.3.3. CMC Nanoparticles. The CMC nanoparticles were prepared based on previously reported methods with a few modifications [20,21]. A 1 mg/mL solution of CMC solution was prepared, and pH was adjusted (pH 7) and diluted into 0.5 mg/mL. The HB was mixed (1 : 10 and 1 : 20 w/w), and the nanoparticles were synthesized by dropwise addition of CaCl 2 (1 mg/mL) into 5 mL of the CMC-HB solution while stirring. Then, the mixture was probe-sonicated for 3 min on an ice bath at 40% power, centrifuged at 14,000×g for 15 min at 4°C, and the pellet was washed twice. The anthocyanin-free CMC nanoparticles were prepared following the same procedure without HB. All the samples were lyophilized (Figure 1(a)).

Determination of the Chemical
Characteristics of Nanoparticles 2.4.1. Anthocyanin Content. The total monomeric anthocyanin content was determined by the pH differential method, which measures the structural transformation of anthocyanins at different pH levels [22]. Briefly, 1 mg PEG-PLGA, MDX, and CMC nanoparticles were dissolved in 2 mL of DCM, deionized water, and 1% formic acid, respectively, sonicated for 10 min at room temperature, and then centri-fuged (5,000×g for 10 min) to remove structural debris. Each nanoparticle mixture was diluted in pH 1.0 buffer (potassium chloride, 0.025 M) and pH 4.5 buffer (sodium acetate, 0.4 M) in 1 : 10 (v/v) ratio, separately. The absorbance was measured at 520 nm and 700 nm using a plate reader (Tecan Infinite® M200 PRO, Morrisville, NC, USA). All the measurements were carried out in a quadruplicate. According to the Equation (1), the total anthocyanin was calculated.
Total anthocyanin mg C3G equivalant L where A = ðabsorbance at 520 nmabsorbance at 700 nm at pH 1Þ -ðabsorbance at 520 nmabsorbance at 700 nm at pH 4:5Þ; MW is the molecular weight of C3G, 449.2 g/mol; DF is the dilution factor; molar extinction coefficient (ε) is 26,900 mol/L/cm; L is the path length (1 cm); and 1,000 is the conversion factor from g to mg.  [23]. The anthocyanin-encapsulated nanoparticles were disintegrated, and 20 μL was mixed with 0.2 N Folin-Ciocalteu (100 μL). After a 5 min incubation, 80 μL of 7.5% sodium carbonate solution was added. The plate was incubated for another 2 h at room temperature before reading at 760 nm. TPC was expressed as mg gallic acid equivalent/g dry weight (mg GAE/g DW).

Anthocyanin Encapsulation
2.4.5. Antioxidant Capacity. The antioxidant capacity was determined by the ferric reducing antioxidant power (FRAP) and the DPPH-free radical scavenging assays.
2.4.6. FRAP Assay. The antioxidant capacity was determined by measuring the electron donation potential of samples, as described by Benzie and Strain and modified by Rupasinghe 3 Oxidative Medicine and Cellular Longevity [24,25]. Nanoparticles were disintegrated. A working solution consisting of 300 mM acetate buffer (pH 3.6), 20 mM ferric chloride, and 1 mM TPTZ solution (10 : 1 : 1, v/v/v) was made freshly and added (180 μL) to 20 μL of sample or standard in 96-well microplate. The absorbance was measured at 593 nm after a 6-min incubation at room temperature in the dark. The antioxidant capacity was calculated as μM Trolox equivalent/g dry weight (μM TE/g DW).
2.4.7. DPPH Assay. The antioxidant activity of anthocyaninloaded capsules was determined through the DPPH· radical scavenging assay. The assay was adapted from Blois to per-form a 96-well microplate [26]. Briefly, 0.2 mM DPPH· reagent was prepared. Nanoparticles were disintegrated and dissolved in water to a concentration gradient (   Lyophilized nanoparticles were dissolved in deionized water and placed on a TEM grid surface. Negative staining was performed with 10% uranyl acetate, and excess fluid was blotted off. The stained samples were air-dried at room temperature, and the TEM grid was then loaded into the transmission electron microscope, and images were taken.
2.5.2. Particle Size. The particle size was measured by dynamic laser light scattering using a Zetasizer in a 1 mg/mL nanoparticle suspension.
2.6. Determination of the Cytoprotective Properties of Nanoparticles In Vitro. BEAS-2B cells were cultured in T-75 flasks coated with a mixture of 0.03 mg/mL bovine collagen type I, 0.01 mg/mL BSA, and 0.01 mg/mL fibronectin and dissolved in sterile 1 × PBS. LHC-9 was used as the cell culture medium. Throughout the experiment, the cells were maintained at 37°C, 5% CO 2 , and in a 100% humidified environment (VWR3074, Hampton, NH, USA). All the experiments were conducted between passages 5 and 30. BEAS-2B cells were treated with anthocyanin-free CMC nanoparticles (E-CMC, 2 mg/mL), HB anthocyaninencapsulated CMC nanoparticles (HB-CMC, 2 mg/mL to provide 4 μg/mL C3G equivalent anthocyanin), and 4 μg/mL of HB for 3 h, separately. The treated cells were washed with 1 × PBS and exposed to 100 μM NNKOAc for another 3 h. The cells with and without NNKOAc exposure were tested to determine the cytoprotective effect of HBencapsulated nanoparticles in BEAS-2B normal bronchial epithelial cells.
where blank is the absorbance of the tested compounds without cells.
2.6.2. 7-AAD Assay. BEAS-2B cells were plated at 2 × 10 5 cells/well density in 6-well plates and incubated overnight at 37°C to encourage cell adhesion. Adhered cells were treated with 4 μg/mL anthocyanin-equivalent HB-CMC (2 mg/mL HB-CMC nanoparticles), anthocyanin-free nanoparticles (2 mg/mL), and HB extracts 4 μg/mL for 3 h at 37°C. Then, the treated monolayers were washed carefully with PBS and exposed to 100 μM NNKOAc for another 3 h. Cells and culture supernatants were harvested, combined, and centrifuged at 250×g for 5 min. Cells were resuspended in PBS, stained with 2.5 μL of 7-AAD viability staining solution at room temperature, and measured by flow cytometer (Attune NxT Acoustic Focusing Cytometer, Thermo Fisher Scientific Inc., Waltham, MA, USA) at 488 nm excitation laser and 695/40 nm emission filter.
2.6.3. DCFDA Assay. The generation of reactive oxygen species (ROS) was determined using DCFDA. Briefly, cells were cultured at 2 × 10 5 density/well in 6-well plates and incubated for 24 h. The attached cells were stained with 10 μM DCFDA and dissolved in serum-and phenol redfree DMEM medium for 45 min at 37°C. Stained cells were thoroughly rinsed with 1 × PBS and treated as described above. Then, the treated monolayers were rewashed and exposed to 100 μM NNKOAc for another 3 h. Cells and culture supernatants were harvested, centrifuged at 500×g for 5 min, and fixed in 300 μL 1% paraformaldehyde for 15 min at room temperature after washing in 1 × PBS. Fixed cells were washed in PBS and analyzed by flow cytometer at 488 nm excitation laser and 530/30 nm emission filter.

Determination of the Genoprotective Properties of
Nanoparticles In Vitro 2.7.1. Comet Assay. DNA damage in cells was evaluated using the alkaline and neutral comet assays [27]. Cells were grown in 6-well plates at a density of 1 × 10 5 cells/well and treated as explained above. After each treatment, cells were harvested and centrifuged at 125×g for 5 min. The cell pellet was washed in cold Ca 2+ and Mg 2+ ion-free PBS and resuspended in cold PBS. Agarose (0.7%, w/v) was melted in boiling water and cooled in a 37°C water bath for 20 min. Cells were mixed with molten agarose at room temperature at a ratio of 1: 10 (v/v) and immediately pipetted (30 μL) onto glass slides, and the mixture was gently spread to make a thin layer. The slides were immersed in prechilled lysis solution [2.5 M NaCl, 100 mM EDTA (pH 10), 10 mM Tris base, 1% sodium lauryl sarcosinate, and 1% Triton X-100] and incubated at 4°C for 45 min. For the alkaline comet assay, DNA unwinding was facilitated by dipping the slides in freshly prepared 0.3 M NaOH containing 1 μM EDTA (pH =9) for 20 min in the dark at room temperature, followed by gel electrophoresis at 16 V for 12 min. For the neutral comet assay, slides were immersed in cold neutral electrophoresis solution (100 mM Tris base and 300 mM sodium acetate in 1 L deionized water, pH =9) for 30 min at 4°C in the dark. The gel electrophoresis was carried out in the same neutral electrophoresis solution at 16 V for 25 min. Then, slides were immersed in 70% ethanol for 5 min and air-dried in the dark, followed by staining in CYGREEN® dye (diluted in 1: 1,000 in DI water, v/v) for 30 min at room temperature in the dark. The slides were briefly rinsed three times in DI water and dried before imaging.

2.7.2.
γH2A.X Assay. BEAS-2B cells were cultured in 6-well plates (2 × 10 5 cells/well) and treated with nanoparticles, HB extracts, and NNKOAc, as explained above. Treated cells were harvested, centrifuged at 125×g for 5 min, washed once in cold 1 × PBS, and resuspended in cold 0.5 mL PBS to get a single cell suspension. The cells were fixed in 70% ice-cold ethanol and kept at -20°C overnight. The fixed cells were permeabilized in 0.5 mL of 0.25% Triton X-100 dissolved in PBS for 5 min at room temperature. The primary antibody was diluted (1 : 250) in 1% BSA and incubated for 60 min under gentle agitation at room temperature. The cells were rinsed once in PBS and stained with fluoresceinconjugated secondary antibody (1 : 500) for another 45 min in the dark. After rinsing off excess antibodies, cells were resuspended in 0.3 mL PBS and analyzed by flow cytometry at 488 nm excitation laser and 530/30 nm emission filter.

Western
Blotting. BEAS-2B cells were collected after the treatments, as previously described, and lysed using ice-cold radioimmunoprecipitation assay lysis buffer containing freshly added protease inhibitors for 15 min on an ice bath. The cell lysates were collected at 14,000×g for 15 min at 4°C, and the protein concentration was measured by Bradford assay. Proteins were denatured and equalized using the sample buffer. Equal amounts of protein (20 μg) were loaded into 10% polyacrylamide gels and electrophoresis for 35 min at 200 V and 400 mA. Proteins were transferred to nitrocellulose membranes, and blots were incubated in 5% skim milk prepared in Tween-TBS [0.25 M Tris (pH 7.5), 150 mM NaCl, and 0.2% Tween-20] for 1 h at room temperature to block nonspecific binding. Blots were probed with primary antibodies (1 : 1,000) overnight at 4°C and then washed thoroughly with Tween-TBS (15 min ×3) and probed with HRP-conjugated donkey antirabbit or anti-mouse IgG antibody (1 : 5,000) for 1 h. The proteins of interest were visualized by enhanced chemiluminescence in the dark. Uniform protein loading was confirmed by probing with HRP-conjugated rabbit anti-actin antibody.

Lung Bioavailability of Anthocyanins from Nanoparticles.
After pulmonary delivery, the bioavailability of anthocyanin encapsulated nanoparticles in mice lungs was evaluated. This study was performed at Dalhousie University's animal care facility, following the University Committee on Laboratory Animals (protocol 2018-007). Female A/JCr albino mice at six weeks were purchased from Charles River Laboratories, Inc., Montreal, QC, Canada.
2.8.1. A/JCr Mice Model. There were three treatment groups: control (E-CMC), HB-CMC, and free HB anthocyanin inhalation. Each group consisted of 5 mice. A nose-only drug delivery device was designed to facilitate passive inhalation of each treatment (Figure 1(b)). An ultrasonic mesh nebulizer was connected to a bisphenol A-free polypropylene chamber (300 mL volume) at one end, and the opposite end of the chamber was connected to a perforated (50 mL) polypropylene tube to hold the mouse. Nanoparticles and the HB anthocyanins were dissolved separately in sterilized deionized water (4 μg C3G equivalent HB anthocyanin in 1 mL water) poured into the medication cup. The nebulizer was run at ≥0.35 mL/min flow rate at room temperature for 21 min to aerosolize each treatment. The total anthocyanin dose for each mouse per day was 28 μg C3G equivalent anthocyanin. Mice were carefully observed during the inhalation treatment for distress signs. They were exposed to the inhalation treatment for six days with two days intervals between each treatment (Figure 1(c)). At the end of the experiment, mice were anesthetized and sacrificed by cervical dislocation. Lungs were excised and stored at -80°C.

UPLC/ESI/Q-TOF-MS Analysis.
The whole lungs from each mouse were weighed and homogenized in PBS (1 : 5, w/v) in an ice bath. The homogenate was centrifuged at 14,000×g for 15 min at 4°C, and the supernatant was collected. An aliquot (500 μL) of the homogenate was mixed with two parts of cold-acetonitrile and acetone (80 : 20, v/v) containing the internal standard and stored at 4°C overnight to enhance protein precipitation. The mixture was centrifuged at 14,000×g for 15 min at 4°C; the supernatant was collected and concentrated under a gentle stream of nitrogen gas. The resultant was reconstituted in methanol, and the ultrahigh-performance liquid chromatography-electrospray ionization quadrupole time-of-flight mass spectrometry (UPLC/ESI/Q-TOF-MS) was used for metabolite analysis.
2.9. Statistical Analysis. All the experiments were performed in three independent experiments and analyzed by one-way analysis of variance (ANOVA) using Tukey's post hoc test and Bonferroni test using GraphPad Prism 5 software (GraphPad Software Inc., San Diego, CA, USA). The in vivo experiment had three treatment groups with five mice in each group. Data were presented as mean ± standard deviation (SD), and p ≤ 0:05 was considered significant between experimental groups.
The nanoparticles were <450 nm in size; however, anthocyanin-loaded CMC nanoparticles were <200 nm ( Figure 2(d)). The particle distribution intensity was less than one for all the nanoparticles; however, the CMC nanoparticles had a high particle distribution intensity (0.7-0.9). The zeta potential of the prepared nanoparticles varied depending on the polymer characteristics. For example, CMC nanoparticles had a positive zeta potential, while PEG-PLGA and MDX nanoparticles showed a negative zeta potential. The TPC and the antioxidant capacity were significantly higher in anthocyanin-encapsulated CMC nanoparticles than in the other polymers (Table 1). According to these findings, CMC was selected as the most suitable candidate for HB-anthocyanin encapsulation and tested in vitro and in vivo.
3.2. Characteristics of the CMC Nanoparticles. In CMC nanoparticles, the loading capacity was significantly higher (p < 0:001) at 1 : 5 and 1 : 10 ratios than at the 1 : 20 anthocyanin: polymer ratio. The anthocyanin: CMC ratio had a negative effect on encapsulation efficiency, while it showed a positive effect on the loading capacity (Figure 3(a)). The particle size and the zeta potential were reduced at alkaline pH (Figures 3(b) and 3(c)). The isoelectric point of anthocyanin-encapsulated CMC nanoparticles was pH 7.3.
In this study, all zeta potential values obtained at acidic pH were less than 12 mV, indicating the formation of aggregates in the dispersion conditions. Based on the anthocyanin encapsulation efficiency, loading capacity, and morphological features, CMC at 1 : 10 ratio was chosen as the best anthocyanin: polymer ratio to encapsulate the anthocyanins and used in vitro and in vivo experiments.  (Figure 4(a)). Signs of cell apoptosis, such as membrane blebbing and shrinkage of cells, were not observed in any treatment. All the treatments had over 90% cell viability and did not show significant (p > 0:05) cell death (Figures 4(b) and 4(c)). . NNKOAc did not change the expression of catalase enzyme in BEAS-2B cells but suppressed SOD and GPx expression by 13% and 20%, respectively, compared to the control. The E-CMC and HB-CMC pretreatment suppressed the NNKOAc-induced Nrf2 phosphorylation by 30% in BEAS-2B cells. The HB-CMC pretreatment restored the expression of SOD and GPx in NNKOAc-exposed BEAS-2B cells. HB-CMC nanoparticles and HB anthocyanin pretreatment reduced the expression of catalase by 10% in comparison to the NNKOAc treatment alone. Therefore, the observed cytoprotective effect of nanoparticles against ROS-mediated cytotoxicity could be partially by activating antioxidant enzymes, such as SOD and GPx, direct scavenging of ROS, or through another mechanism.

Carcinogen-Induced DNA Damage.
None of the nanoparticles or HB anthocyanin (4 μg/mL) induced DNA damage in BEAS-2B cells, whereas NNKOAc increased DNA double-strand breaks in the BEAS-2B cells. The nanoparticles or HB anthocyanin did not reduce the NNKOAcinduced DNA double-strand breaks in BEAS-2B cells. The γH2A.X (Figures 6(a) and 6(b)) and neutral comet assays (Figures 7(e)-7(h)) data had a similar pattern regarding the double-strand breaks. In contrast, HB-CMC nanoparticles and HB anthocyanin significantly reduced (p < 0:05) NNKOAc-induced DNA damage, as measured by alkaline comet assay (Figures 7(a)-7(d)). Tail length and tail DNA percentage were significantly reduced (p < 0:05) in cells pretreated with HB-CMC nanoparticles compared to NNKOAc-exposed cells. The alkaline comet data represent both single-and double-strand breaks. BEAS-2B cells in neutral comet assay showed shorter tails than the alkaline comet images (Figures 7(a) vs 7(e)).   (Table 2). Furthermore, anthocyanin-derived phenolic acids, such as hippuric acid and caffeic acid, were also observed in the lungs of mice exposed to HB.

Discussion
Anthocyanins are readily degraded and oxidized to various forms due to the oxidation of their hydroxyl groups. The HB comprises C3G, peonidin-3-O-glucoside, malvidin-3-O-glucoside, cyanidin-3-O-rutinoside, and many other phenolic acids, such as chlorogenic acid, ferulic acid, and caffeic acid. C3G is the most predominant, representing about 90% of the anthocyanins in HB [12]. The stability of anthocyanins can be enhanced by conjugating them with macromolecules, such as polysaccharides, polymers, or proteins [28]. We used synthetic (PEG-PLGA) and natural (MDX and CMC) polymers to coat HB anthocyanin to prolong and enhance lung bioavailability. The encapsulation efficiency of anthocyanins, either as nanoparticles or microparticles, ranges from 3 to 60% and depends on factors such as 11 Oxidative Medicine and Cellular Longevity polymer characteristics, anthocyanin to polymer ratio, conditions at processing (pH, time, and solvents), and method of encapsulation [14,29,30]. CMC showed a higher HB anthocyanin encapsulation efficiency and loading capacity among the selected polymers. The lower encapsulation efficiency in MDX and PEG-PLGA (<5%) in the present study may be due to the long polymer chain of MDX and/or the hydrophobic nature of PEG-PLGA (5,000 : 7,000). It is reported that chain length affects the formation, stability, and entrapment of molecules in MDX nanoparticles. MDX with low dextrose equivalent values (4.0-7.0) has long chains that aggregate quickly, resulting in poor anthocyanin content compared to the short-chain MDX (dextrose equivalent value >13) [31,32]. Additionally, PEG-PLGA (amphiphilic) in DCM creates a hydrophobic environment that negatively affects the solubility of hydrophilic HB anthocyanins, resulting in weak binding capacity with PEG-PLGA di-block polymer. The PEG: PLGA ratio also affects the drug loading 13 Oxidative Medicine and Cellular Longevity efficiency [33]. In the present study, the longer MDX polymeric chains and hydrophobic PEG-PLGA solution may have reduced the anthocyanin encapsulation efficiency and loading capacity.
The loading capacity and the amount of drug-loaded per unit weight of the nanoparticle are essential in formulations as it determines the number of particles required for a given dose [34]. The inverse association between encapsulation efficiency and loading capacity in CMC may be due to the variation of the binding capacity of anthocyanin on the CMC-Ca matrix. The carboxylate ion (COO − ) groups of CMC interact with divalent calcium ions (Ca 2+ ) and form    15 Oxidative Medicine and Cellular Longevity cross-links between polymers. Further, CMC polymers form stable bonds between NH 3 + and COO − groups. In anthocyanin mixed CMC-Ca matrix, anthocyanins form electrostatic interactions with NH 3 + and COO − groups of CMC (Figure 9(a) 1-4 bonds) [21,35,36]. It is further shown that anthocyanins can incorporate into CMC through its C=O group of the benzopyran aromatic ring [37]. The ability to form ionic interactions between anthocyanin and CMC may be the reason to show a higher encapsulation efficiency of HB anthocyanins in CMC-Ca matrix compared to PEG-PLGA and MDX. The deprotonation of CMC affects these interactions as it determines the number of reactive groups available on the polymer to interact with each other and with HB anthocyanins. For instance, at the 20 : 1 CMC: anthocyanin ratio, HB anthocyanin has enough ionic groups (on CMC) to bind, resulting in increased encapsulation efficiency but lower loading capacity due to the limitations of anthocyanins in the mixture. At the 5 : 1 CMC: anthocyanin ratio, ionic groups are limited for anthocyanins to bind, resulting in less encapsulation efficiency but higher loading  The slightly acidic pH in airway surface liquid and cytoplasm of lung cells slow the C3G release rate from HB-CMC in the lung tissues leading to prolonged retention of intact C3G than from the free HB. References are cited in the text. The Biorender free software was used to draw the illustrations. 16 Oxidative Medicine and Cellular Longevity capacity. Therefore, finding methods to increase the electrostatic interactions and ionic cross-links among C3G-CMC-Ca 2+ ion matrices is necessary. Modifying CMC polymer by altering the number of NH 2 groups and COOH on CMC and incorporating tripolyphosphate as the crosslinking agent may enhance the anthocyanin encapsulation efficiency and loading capacity. The stability of the CMC nanoparticles is pH-dependent [21]. pH determines the protonation or deprotonation levels of NH 2 (NH 3 + > COO − in acidic pH) and COOH groups (NH 3 + < COO − in alkaline pH) that finally determines the interactions between NH 3 + and COO − of CMC polymers during the synthesis of nanoparticles [21,38]. The zeta potential, the electric potential at the shear plane in a suspension, determines the stability of particles [20]. The zeta potential of CMC nanoparticles was negative at pH >7.3, which is beyond the isoelectric point, but was positive below the isoelectric point. This effect may result from COO − groups at higher pH and NH 3 + groups at lower pH [21]. Near the isoelectric point, the charges of NH4+ and COO − groups are equal. The low polydispersity index (<0.7) indicates the formation of almost monodispersed nanoparticles in all the polymers except for MDX, which showed a broad size distribution (Figure 2(b)).
NNKOAc is a nicotine-derived nitrosamine ketone (NNK) derivative in smoke and smokeless tobacco. NNK is a group I lung carcinogen that has been reported to induce oxidative stress and gene mutation both in vivo and in vitro. It has been reported that NNK generates superoxide anions and hydrogen peroxide [39]. Our previous studies have confirmed NNKOAc-induced oxidative stress in normal lung epithelial cells. NNKOAc also initiates DNA double-strand breaks in lung epithelial cells [12,40]. Therefore, NNKOAc treatment was chosen to test the cytoprotective ability of HB-CMC nanoparticles on lung epithelial cells in vitro.
The present study shows that CMC nanoparticles with and without HB anthocyanins were not cytotoxic to normal lung epithelial cells in vitro. Furthermore, 2 mg/mL of CMC nanoparticles reduced the carcinogen-induced oxidative stress but not the DNA double-strand breaks. It is reported that CMC, conjugated to drugs and unconjugated in free form, can regulate the Nrf2/ARE pathway and attenuate the activity of antioxidant enzymes SOD, catalase, GPx, and DNA damage against oxidative stress in vitro and in vivo [40][41][42][43][44]. Similarly, we found that E-CMC and HB-CMC can reduce ROS accumulation and restore the expressions of SOD and GPx in carcinogen-exposed BEAS-2B cells. However, no significant effect was found on the expression of the catalase enzyme. The HB-CMC and free HB anthocyanin improved the expression of SOD and GPx, in carcinogen-exposed cells. The FRAP and DPPH• radical scavenging antioxidant capacity assays demonstrated HB-CMC particles in scavenging free radicals. The ROS scavenging ability of CMC is reported. The hydroxyl and amino groups of CMC scavenge superoxide and hydroxyl radicals and reduce oxidative stress [45][46][47]. Therefore, ROS mediating characteristics of CMC nanoparticles may be due to the activation of SOD and GPx to a certain extent, direct scav-enging of NNKOAc-induced oxidative radicals, and activating nonenzymatic antioxidants such as glutathione.
DNA double-strand breaks in BEAS-2B cells were detected by γH2A.X immunofluorescence and neutral comet assays. At the selected doses, E-CMC, HB-CMC, and HB anthocyanins did not induce DNA damage. However, none of the nanocarriers protected the cells from NNKOAcinduced DNA double-strand breaks. Increasing the HB-CMC treatment period (>3 h) and treatment dose in future studies alter the effect of HB-CMC nanoparticles' protective effect on NNKOAc-induced genotoxicity in BEAS-2B lung epithelial cells. The neutral comet assay detects doublestrand breaks, while the alkaline comet assay measures both single-and double-strand breaks, alkali-labile sites, and DNA-DNA cross-linking. Alkaline comet assay (pH 13) is more sensitive to a minor degree of DNA damage [27]. The results of an alkaline comet assay suggest that NNKOAc also can induce single-strand breaks and/or alkali-labile sites in BEAS-2B cells. The single-strand breaks and alkali-labile sites are formed by ROS, particularly hydroxyl radicals and singlet oxygen species [48]. NNK produces hydroxyl radicals and superoxide radicals that later convert into hydroxyl radicals. The protective effect in the alkaline comet assay may be due to the reduction of oxidative stress-induced single-strand breaks and alkali-labile sites in BEAS-2B cells (Figure 9(b)).
The presence of anthocyanins in mouse lung tissues following high dose (100 and 500 mg/kg body weight) oral administration has been reported. However, after 15 min of ingestion, most anthocyanins are lost from plasma and excreted through urine, resulting in low bioavailability [49]. In the present study, we delivered HB-anthocyanins (encapsulated form and free HB) at a lower dose (28 μg C3G equivalent anthocyanin/mouse/days) directly into the mice lungs by facilitating passive inhalation. C3G, the major anthocyanin of HB, was detected in mouse lung tissues even after one hour of inhalation treatment. The results suggest that direct delivery to lung tissues prolongs the anthocyanin tissue retention and enhances tissue-specific bioavailability. Furthermore, cyanidin, the aglycone of C3G (only in HB-CMC treatment), peonidin-3-O-glucoside, and cyanidin-3-O-glucuronide (only in free HB treatment) were observed in lung tissues suggesting that lung cells can metabolize C3G into cyanidin and facilitate its phase 2 metabolism and/or part of HB-CMC and HB anthocyanins has reached the systemic circulation and undergone phase 2 liver metabolism within one hour of inhalation treatment. The absence of C3G-derived anthocyanins, such as peonidin-3-O-glucoside and cyanidin-3-O-glucuronide, in the lungs of HB-CMC nanoparticle-treated mice, indicates the protective effect of CMC envelope against enzymatic digestion of entrapped anthocyanins in the lungs. It is known that CMC has higher mucoadhesive characteristics, which facilitate active molecule absorption [48].
Additionally, the anthocyanin release rate from CMC nanoparticles is pH-dependent. For instance, the C3G release rate from C3G encapsulated CMC-Ca particles is higher (60% over six days period) under acidic conditions (pH 5.3) compared to the slightly acidic or neutral pH (20-30% over six days, pH 6-8-7.4) solutions [35]. CMC has low solubility at higher pH [50]. The mouse lung pH values range from 6.9 to 7.2 [51]. The slightly acidic environment in the mouse lung may have increased the stability of HB-CMC nanoparticles, resulting in a slower C3G release rate in mouse lungs. Hence, we only found C3G and aglycone cyanidin in HB-CMC-treated lungs. In humans, airway surface liquid/mucosa is slightly acidic (pH 6.85), and epithelial cell cytoplasm is slightly acidic or neutral (6.9-7.4) [51,52]. These conditions may facilitate extended retention of HB-CMC nanoparticles and sustained C3G release into the lung tissues (Figure 9(c)). Lung diseases can alter the pH of the airway mucosa. For example, chronic bronchitis and rhinitis create slightly alkaline (pH 7-8) conditions, while some bacterial infections and lung cancer result in an acidic environment in the lungs (pH 5-6.5) [51]. These conditions favor the retention stability and prolong the release of C3G from HB-CMC in lung tissues. HB-CMC could therefore be a potential therapeutic to treat lung diseases, particularly cancer. However, it is crucial to quantify the time cause of the release of anthocyanins and their possible metabolites in lung tissues and plasma to determine the systemic distribution of anthocyanins after pulmonary delivery.

Conclusions
To improve the bioavailability and direct delivery of C3Grich HB anthocyanins, three polymers, PEG-PLGA, MDX, and CMC, were investigated as the coating material. HB anthocyanin was successfully encapsulated by ionic gelation in CMC with Ca 2+ ion as the cross-linker, forming spherical nanoparticles. The anthocyanin: polymer ratio affects the encapsulation efficiency and loading capacity of anthocyanins in CMC. In vitro studies in normal lung epithelial cells showed that nanoparticles can attenuate carcinogen-induced oxidative stress by restoring antioxidant enzymes and protecting against DNA damage. Pulmonary delivery of an extremely low dose of encapsulated anthocyanins extended the retention of intact C3G and delayed its tissue metabolism. Therefore, we suggest that direct delivery of anthocyanins to lung tissues has therapeutic potential to treat various lung diseases, including cancer.