Coenzyme Q₁₀ was purchased from Kaneka Nutrients (Pasadena, TX, USA). Trisodium glycyrrhizinate hydrate was purchased from Tokyo Chemical Industry Co. LTD. (Tokyo, Japan). Eicosapentaenoic acid (EPA) was purchased from Phycoil Biotech Korea, Inc. (Seoul, Korea). EPA was used to improve micelle stability because glycyrrhizin alone is not enough to maintain the micelle. In addition, EPA itself reduces the inflammatory response by inhibiting prostaglandin synthesis. To produce a uniform nano-emulsion (Qmicelle) (Fig. 1A), the following pretreatment is essential. Coenzyme Q₁₀, EPA, and trisodium glycyrrhizinate hydrate (at a concentration of 1 mg/mL each) were dispersed in hot water. The mixture was pre-homogenized at 10,000 rpm with a homogenizer (Multi-Purpose Homogenizer, YSTRAL, Ballrechten-Dottingen, Germany) until a homogeneous consistency was obtained. An APV2000 microfluidizer processor (SPX Flow, Leeds, UK) was used to produce nano-emulsions (Qmicelle). It has a reservoir capacity of 1,500 mL and can be operated at pressures of up to 2,000 bars. The prepared mixture at 60°C was poured into the fluidizer and Qmicelle was obtained after 10 cycles at a pressure of 1,200 bars. Each component under EPA, coenzyme Q₁₀ and trisodium glycyrrhizinate hydrate, showed single peaks at retention times of 6.0, 6.6, and 15.9 minutes, respectively (Fig. 1B), and their concentrations were 0.24, 1.0 and 0.6 mg/mL (Table 1). We detected homogeneity and size distribution by dynamic light scattering (DLS), and conformed that sizes of the Qmicelles were uniform and distributed in a narrow range around 100 nm (Fig. 1C).

All experimental procedures were reviewed and approved by the animal care and use committee of The Catholic University of Korea (CUMC-2018-0038-03). All the protocols in the study were conducted in strict accordance with their ethical guidelines for animal research. Sprague-Dawley rats (Charles River Technology, Seoul, Korea) with a starting weight of 220 to 230 g were kept in cages (Nalge Co., Rochester, NY, USA) with controlled temperature, humidity, and light in the animal care facility at the Catholic University of Korea. The rats were fed a low-salt diet (0.05% sodium chloride, Research Diets, New Brunswick, NJ, USA) daily. TAC (Prograft, Astellas Pharma, Ibaraki, Japan) was dissolved in olive oil (Sigma-Aldrich, St. Louis, MO, USA), and blended to a final concentration of 1.5 mg/mL. CoQ₁₀ (Chong Kun Dang Pharm, Seoul, Korea) was diluted in olive oil (CoQ₁₀-L) or drinking water (CoQ₁₀-W) to a final concentration of 20 mg/kg.

The rats were randomly divided into four groups (n = 9/ group). The rats in the vehicle (VH) group received 0.3 mL olive oil once a day subcutaneously. The rats in the TAC, TAC + CoQ₁₀-L and TAC + CoQ₁₀-W groups were treated with TAC (1.5 mg/kg) subcutaneously daily for 4 weeks. The TAC + CoQ₁₀-L and TAC + CoQ₁₀-W groups were co-treated with CoQ₁₀-L or CoQ₁₀-W (20 mg/kg), respectively, orally once a day. We decided the route of administration and concentrations of TAC and CoQ₁₀ based on previous studies [14,15].

Surgical instruments are sterilized before use. Following abdominal incision, phosphate-buffered saline (PBS) was used for in vivo perfusion via the abdominal aorta. After ensuring that blood was completely removed from the organs, part of the pancreas was removed for further analysis. Then the perfusion solution was changed to periodate-lysine-paraformaldehyde, and perfusion was continued for an additional 5 minutes. The remaining portion of the pancreas was collected and wax-embedded for histological studies.

Following administration of the drug, the diet and weight of the rats were closely monitored for 4 weeks. After 4 weeks of treatment, the rats were placed in a metabolic cage for 24 hours to measure water intake and urine volume. For further analysis, blood and tissue samples were collected from anaesthetized rats a day later. The TAC level in whole blood was measured by liquid chromatography-tandem mass spectrometry (LCMS/MS, Abbott Diagnostics, Abbott Park, IL, USA).

The concentration of CoQ₁₀ in the plasma and pancreatic tissues was measured by LC-MS/MS analysis. CoQ₁₀ was extracted from the plasma and tissue and quantified using LC-MS/MS method reported previously [16], with some modifications. Given that CoQ₁₀ is an endogenous substance, all calibration and quality control samples were prepared in a physiological solution (4% bovine serum albumin in PBS). Briefly, 100 μL aliquots of rat plasma or tissue homogenates treated with 10 μL of CoQ₁₀ (800 ng/mL, as the internal standard) were extracted by liquid-liquid extraction using 1 mL isopropyl alcohol, followed by LC-MS/MS analysis. The samples were analyzed using a Shimadzu Nexera X2 UPLC (Shimadzu Corporation, Kyoto, Japan) coupled with an LCMS-8050 triple quadruple mass spectrometer (Shimadzu Corporation) with an electrospray ionization interface in the positive ion mode. Chromatographic separation was achieved using a Kinetex C18 column (2.1 × 100 mm, 2.6 μm; Phenomenex, Seoul, Korea) with a mobile phase consisting of 0.1% formic acid in isopropyl alcohol and methanol at a flow rate of 0.2 mL/min. The total run time was 4 minutes per sample. Quantitation was performed using selected reaction monitoring of the transitions at m/z 863.35 > 197.15 (for CoQ₁₀) and m/z 794.35 > 197.15 (for the internal standard). The calibration curves were linear (r ≥ 0.995) from 20 to 10,000 ng/mL. The within- and between-batch precision and accuracy were within the acceptable limits of ± 15%.

Rats were fasted for 1 day following treatment with the drugs for 4 weeks before the intraperitoneal glucose tolerance test (IPGTT) was conducted. The rats were injected with 50% dextrose (1.5 g/kg), and blood glucose concentration was measured at 30, 60, 90, and 120 minutes using a glucose analyzer (Accu-Check, Roche Diagnostics, Basel, Switzerland). Thereafter, based on the IPGTT values, trapezoidal estimation was obtained, and the area under the curve of glucose (AUCg) was calculated.

After treating male Sprague-Dawley rats (250 to 300 g) with the drugs according to the in vivo experimental design, pancreatic islets were isolated using collagenase digestion as described previously [17]. The islets were incubated in Roswell Park Memorial Institute (RPMI) 1640 medium containing 10% FBS and 100 U/mL penicillin at 37°C for 24 hours. The next day, groups of 30 islets were washed with Krebs-Ringer Modified Buffer (KRB; 130 mmol/L sodium chloride [NaCl], 3.6 mmol/L potassium chloride [KCl], 1.5 mmol/L calcium chloride [CaCl2], 0.5 mmol/L magnesium sulfate [MgSO₄], 0.5 mmol/L potassium dihydrogen phosphate [KH²PO⁴], 2.0 mmol/L sodium bicarbonate [NaHCO₃], and 10 mmol/L 2-[4-(2-hydroxyethyl) piperazin-1-yl] ethanesulfonic acid [HEPES]) and 2.8 mM glucose (basal) was added. After washing with KRB, the islets were incubated in KRB containing 16.7 mM glucose for 1 hour. The supernatant was collected and the insulin level was measured using a sandwich enzyme-linked immunosorbent assay (ELISA) (Millipore Corp., St. Charles, MO, USA).

All the immunohistochemistry (IHC) procedures were performed as described previously [18]. The primary antibodies used for IHC were as follows: anti-insulin (I2018, Sigma-Aldrich; 18-0067, Invitrogen, Carlsbad, CA, USA), anti-8-hydroxy-2’-deoxyguanosine (8-OHdG; JaICA, Shizuoka, Japan), and anti-4-hydroxy-2-hexenal (4-HHE, JaICA, Shizuoka, Japan). 4-m samples were incubated with the antibodies for 12 hours at 4°C. A minimum of 20 fields per section was assessed using a color image analyzer (TDI Scope Eye version 3.6 for Windows, JN OpTIC Co. Ltd., Seoul, Korea). Briefly, images captured during insulin immunohistochemistry were quantified using the polygon program by measuring the pancreas area that was positively stained for insulin, except for vacuoles, when viewed under ×400 magnification. The measurement of 4-HHE expression was similar to that of insulin, and expressed as percentage of 4-HHE positive area in the islet. The 8-OHdG expression was quantified by counting the number of positive cells per field. Histopathologic analysis was performed on randomly selected fields of the pancreas sectioned by a pathologist blinded to the identity of the treatment groups.

All procedures were carried out strictly according to the ELISA kit protocol (Cell Biolabs, San Diego, CA, USA).

Terminal deoxynucleotide transferase-mediated dUTP nick-end labeling (TUNEL) staining was performed using the in situ Apoptosis Detection Kit (Millipore, Billerica, MA, USA) as per manufacturer’s instructions. In each tissue section, positive cells were analyzed in 20 randomly selected areas. The positive cells or areas were then quantified at ×400 magnification, and the data were analyzed with a color image analyzer (TDI Scope Eye version 3.6 for Windows).

Pancreatic tissues were fixed in 2.5% glutaraldehyde (diluted in 0.1 M phosphate buffer), post-fixed with 1% Osmium (VIII) oxide (OsO4) and then embedded in Epon 812. Ultrathin sections were prepared using the embedded tissues, and the sections were stained with uranyl acetate/lead citrate. A JEM-1200EX transmission election microscope (JEOL, Tokyo, Japan) was used for photography, and 20 randomly selected sites were scanned at ×5,000 magnification. We counted insulin granule number and measured mitochondrial area in the scanned areas using an image analyzer (TDI Scope Eye version 3.6).

The INS-1 rat insulinoma cell line was grown in conditioned RPMI-1640 medium supplemented with 10 nM HEPES, 10% fetal bovine serum, 2 mM L-glutamine, 1 mM sodium pyruvate, 50 mM 2-mercaptoethanol (all from Sigma-Aldrich), 100 IU/mL penicillin, and 100 mg/ mL streptomycin (both from Wisent Bio, Saint-Bruno, QC, Canada) in a humidified atmosphere containing 5% CO₂. Cells were plated in culture plates.

A cell counting kit-8 (CCK-8, CK04; Dojindo Molecular Technologies, Rockville, MD, USA) assay kit was used for analyzing cell viability. After plating the INS-1 cells for 1 day, the cells were treated with TAC (40 mg/mL), and same dose of either CoQ₁₀-L (0.1 pg/mL, 1 pg/mL, 10 pg/ mL, 1 ng/mL, 10 ng/mL, 100 ng/mL, and 1 μg/mL) diluted with dimethyl sulfoxide (DMSO) or CoQ₁₀-W (0.1 pg/mL, 1 pg/mL, 10 pg/mL, 1 ng/mL, 10 ng/mL, 100 ng/mL, and 1 μg/mL) diluted with water for 12 hours. According to the manufacture’s protocol, CCK-8 or propidium iodide (PI; 556463, BD Biosciences, San Jose, CA, USA) solution was added for 2 hours. Following CCK-8 treatment, the absorbance was measured at 450 nm using a Versa Max ELISA Reader (Molecular Devices, Sunnyvale, CA, USA).

To assess reactive oxygen species (ROS) production, we performed flow cytometry. INS-1 cells were plated and pre-incubated in an incubator at 37°C for 24 hours. After 24 hours, the culture medium was changed to serum-free medium containing TAC (40 mg/mL), TAC (40 mg/mL) + CoQ₁₀-L (10 ng/mL) or TAC + CoQ₁₀-W (10 ng/mL). The method followed has been described previously [19].

The data are expressed as mean ± standard error (SE) of at least three independent experiments. Multiple comparisons between different groups were carried out by oneway analysis of variance with Bonferroni’s post hoc test using IBM SPSS Statistics version 24 (IBM, Armonk, NY, USA). Results with p < 0.05 were considered statistically significant.

TAC treatment for 4 weeks significantly reduced the body weight compared to VH treatment (48 ± 6 g vs. 67 ± 3 g, p < 0.05). These changes were restored in the TAC + CoQ₁₀-L and TAC + CoQ₁₀-W groups (48 ± 6 g vs. 62 ± 1 g and 65 ± 2 g, p < 0.05). Water intake was increased in the TAC group compared with the VH group (49 ± 6 mL vs. 30 ± 2 mL, p < 0.05). In the TAC + CoQ₁₀-L and TAC + CoQ₁₀-W groups, water intake was lower than in the TAC group (35 ± 1 mL and 39 ± 4 mL vs. 49 ± 6 mL, p < 0.05). The concentration of TAC in whole blood was not affected by CoQ₁₀-L or CoQ₁₀-W treatment (9.4 ± 4 ng/mL vs. 9.3 ± 2 or 9.3 ± 3 ng/mL, p > 0.05).

Plasma CoQ₁₀ levels in the VH and TAC groups were 39 ± 7 and 25 ± 8 ng/mL, respectively. The plasma CoQ₁₀ level (Fig. 2A) was significantly higher in the TAC + CoQ₁₀-W group than the TAC + CoQ₁₀-L group (956 ± 97 ng/mL vs. 577 ± 88 ng/mL, p < 0.05). The pancreatic CoQ₁₀ level (Fig. 2B) was lower in the TAC group than in the VH group (59 ± 8 ng/mL vs. 103 ± 26 ng/mL, p < 0.05). Treatment with either TAC + CoQ₁₀-L or TAC + CoQ₁₀-W increased pancreatic CoQ₁₀ levels compared with the TAC only group (121 ± 4 ng/mL and 110 ± 14 ng/mL vs. 59 ± 8 ng/mL, p < 0.05). However, there was no significant difference between the two groups.

After the 4-week drug-treatment regimen, we performed a IPGTT to evaluate the changes in blood glucose level. Thirty minutes after injecting 50% dextrose, blood glucose was significantly higher in the TAC group than the VH group (434 ± 9 ng/mL vs. 322 ± 16 mg/dL, p < 0.05). Unlike TAC + CoQ₁₀-L treatment (430 ± 6 mg/dL), TAC + CoQ₁₀-W treatment (379 ± 14 mg/dL) decreased the blood glucose level relative to TAC for 120 minutes (Fig. 3A). The increase in AUCg values detected by IPGTT demonstrated that DM was successfully induced by TAC (273 ± 19 mg/dL/min vs. 162 ± 18 mg/dL/min, p < 0.05). The TAC + CoQ₁₀-L group had no significant difference in AUCg compared with the TAC group (241 ± 23 vs. 273 ± 19 mg/dL/min, p > 0.05). The decrease was markedly higher in the TAC + CoQ₁₀-W group (170 ± 12 mg/dL/min, p < 0.05) (Fig. 3B).

Insulin secretion was significantly reduced by TAC treatment compared to VH (11 ± 0.6 ng/mL vs. 25 ± 1 ng/mL, p < 0.05) (Fig. 4). CoQ₁₀-W increased insulin secretion (15 ± 0.5 ng/mL, p < 0.05) unlike CoQ₁₀-L (11 ± 0.2 ng/mL, p > 0.05).

Immunohistochemical staining for insulin showed that the islet sizes in the TAC group and TAC + CoQ₁₀-L group were lower than that in the VH group (8 ± 2 and 9 ± 1 vs. 16 ± 2 μm² × 10³, p < 0.05). However, the islet size in the TAC + CoQ₁₀-W group (14 ± 1 μm² × 10³) was increased compared with the TAC and the TAC + CoQ₁₀-L groups (p < 0.05) (Fig. 5).

Immunohistochemical staining for the oxidative stress markers 8-OHdG (Fig. 6A and 6B), and 4-HHE (Fig. 6C and 6D), and ELISA for serum 8-OHdG (Fig. 6E) were used to identify the anti-oxidative effects. The number of 8-OHdG-positive cells was significantly increased in the TAC group compared with the VH group (140 ± 10 per field vs. 3 ± 1 per field, p < 0.05). CoQ₁₀-W (83 ± 6 per field) significantly decreased the expression of 8-OHdG unlike CoQ₁₀-L (138 ± 6 per field). The expression of 4-HHE was significantly higher in the TAC group than in the VH group (15% ± 1% vs. 0.3% ± 1%, p < 0.05). CoQ₁₀-W (6% ± 1%) significantly decreased the expression of 4-HHE (p < 0.05) while CoQ₁₀-L (14% ± 1%) did not. Similar changes were observed in serum 8-OHdG level (Fig. 6E).

Four weeks of TAC treatment also induced apoptosis in the pancreas. As shown in Fig. 7, the number of TUNEL-positive cells was significantly higher in the TAC group than in the VH group (2.4 ± 0.3 cells per islet vs. 0 ± 0 cells per islet, p < 0.05). CoQ₁₀-W decreased TUNEL-positive cells compared with TAC (0.2 ± 0.2 cells per islet vs. 2.4 ± 0.3 cells per islet, p < 0.05), while CoQ₁₀-L (1.8 ± 0.2 cells per islet) did not (Fig. 7).

Electron microscopy was used to evaluate the changes in the mitochondrial area and insulin granule number. The average mitochondrial area was significantly lower in the TAC and TAC + CoQ₁₀-L groups than in the VH group (0.05 ± 0.006 and 0.08 ± 0.08 vs. 0.4 ± 0.03 per field, p < 0.05). However, the numbers were significantly (p < 0.05) higher in the TAC+CoQ₁₀-W group (0.2 ± 0.02 per field) (Fig. 8A and 8C). Changes in insulin granule numbers were similar to the changes in mitochondrial area (Fig. 8A and 8B).

We demonstrated the protective effect of CoQ₁₀-L and CoQ₁₀-W against cell injury during TAC treatment in INS-1, a pancreatic beta cell line. At the different doses tested, both TAC + CoQ₁₀-L and TAC + CoQ₁₀-W increased cell viability by approximately 2.8-fold compared to TAC (p < 0.05). However, there were no differences between the tested doses of TAC + CoQ₁₀-L and TAC + CoQ₁₀-W (p > 0.05) (Fig. 9).

To compare the protective effect of CoQ₁₀-L and CoQ₁₀-W on oxidative stress, we performed flow cytometry using INS-1 cells. TAC significantly increased ROS production compared to the VH (p < 0.05). Both CoQ₁₀-L and CoQ₁₀-W reduced TAC-induced ROS production; though there was no significant difference between the two groups (p > 0.05) (Fig. 10).