*2.2. HPLC Analysis*

HPLC analyses were performed using an Agilent 1200 HPLC instrument (Agilent Technologies, Santa Clara, CA, USA) equipped with a binary pump (G1312A), vacuum degasser (G1322A), auto-sampler (G1329A), column compartment (G1316A), and diode array detector (DAD, 1365B). Data were collected and analyzed using the Agilent ChemStation software. Chromatographic separation was conducted using a Luna C18(2) (250 × 4.6 mm, 5.0 μm; Phenomenex, Torrance, CA, USA), and the column temperature was maintained at 40 ◦C. The mobile phase consisted of 0.1% formic acid in water (A) and acetonitrile (B) with gradient elution for better separation. The gradient solvent system was optimized as follows: 95–55% A (0–40 min), 55–0% A (40–41 min), 100% B (41–45 min), and 95% A (45–55 min). The flow rate was 1 mL/min. The detection was conducted at 240 nm, and the injection volume of each sample was 10 μL. To test for linearity, standard solutions at five levels were prepared by serially diluting the stock solution. Each analysis was repeated three times, and the calibration curves were fitted by linear regression. The limit of detection (LOD) and limit of quantification (LOQ) data obtained under the optimal chromatographic conditions were determined using signal-to-noise (S/N) ratios of 3 and 10, respectively.

#### *2.3. Determination of Preventive E*ff*ect of CPA4-1 on AGE Formation*

Bovine serum albumin (BSA) (Sigma-Aldrich, St. Louis, MO, USA) was mixed with 0.2 M glucose or fructose in 50 μM phosphate bu ffer at 10 mg/mL. Various concentrations of CPA4-1 extract or aminoguanidine (AG) (Sigma-Aldrich) were added, and the mixture was incubated at 37 ◦C for 14 days. After incubation, the fluorescence products of glycated BSA was determined using a spectrofluorometer (Synergy HT; BIO-TEK, Winooski, VT, USA) at excitation/emission wavelength of 350/450 nm. The 50% inhibition concentration (IC50) for AGE formation was calculated via interpolation from the concentration-inhibition curve.

#### *2.4. Breaking E*ff*ect of CPA4-1 on Preformed AGE-Collagen Complexes*

The ability of CPA4-1 to break preformed AGEs was evaluated using a previously described method [22]. Briefly, 1 μg of glycated BSA (AGEs-BSA) (MBL International, Woburn, MA, USA) was pre-incubated in collagen-coated 96-well plates for 24 h, and the collagen-AGEs-BSA complexes were then incubated with CPA4-1. Collagen-AGE-BSA crosslinking was detected using mouse anti-AGEs primary antibody (Clone No. 6D12; Trans Genic Inc., Kobe, Japan), horseradish peroxidase-linked goa<sup>t</sup> anti-mouse IgG secondary antibody, and H2O2 substrate containing 2,2--azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) chromogen. Breakdown levels were measured as the percentage of decrease in optical density (OD = 410 nm). We calculated the IC50 (μg/mL) as 50% inhibition of collagen-AGE-BSA crosslinking, in which the crosslinking inhibition percent was calculated as follows:

Inhibition of collagen-AGE-BSA crosslinking (%) = 100 − abserbance of sample abserbanceofcontrol× 100.

Breaking of AGE-induced crosslinking was expressed as the percentage decrease in optical density. The breaking percentage was calculated according to the above equation.

#### *2.5. Cell Culture and Determination of Preventive E*ff*ect of CPA4-1 on AGE Formation in ARPE-19 Cells*

ARPE-19 cells were purchased from the American Type Culture Collection (ATCC CRL-2302; Manassas, VA, USA) and maintained at 37 ◦C in a humidified 5% CO2 incubator [23]. To examine the inhibitory e ffect of CPA4-1 on AGE formation, the cells were treated with CPA4-1 (10, 20, or 50 μg/mL in DMSO) for 1 h before the addition of 25 mM glucose and 500 μg/mL BSA (Roche Diagnostics, Basel, Swiss). After that, the cells were incubated for 48 h and then subjected to Western blotting analysis.

#### *2.6. Western Blot Analysis*

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed, as described previously [4]. Each sample (25 μg/mL) was fractionated on a 10% SDS-PAGE, after which the proteins were transferred to a polyvinylamide gel membrane (Millipore, Billerica, MA, USA) using traditional tank transfer system (Mini Trans-Blot cell, Bio-rad, Hercules, CA, USA). Membranes were probed with antibodies against AGEs (Clone No. 6D12; Trans Genic Inc.), occludin (Invitrogen, Carlsbad, CA, USA), and β-actin (Sigma-Aldrich), each at 1:1000 dilution. Signals were detected using a WEST-one ECL solution (Intron, Korea) and captured on a Fuji Film LAS-3000 (Tokyo, Japan).

#### *2.7. Animals and Experimental Design*

Animal experiments were conducted by Qu-Best Bio Co., Ltd. (Yongin City, Korea), according to the National Institutes of Health Guide for the Care and Use of Laboratory Animals (Approval number: QJE14015). Male C57BL/KsJ db/db mice and their age-matched lean littermates (db/+) were purchased from Japan SLC (Shizuoka, Japan). At 8 weeks of age, the db/db mice were randomly assigned into four groups (n = 10): NOR, normal control mice; DM, db/db mice; FENO, db/db mice treated with fenofibrate (100 mg/kg); CPA4-1-100, db/db mice treated with CPA4-1 (100 mg/kg). CPA4-1 was dissolved in the vehicle (0.5% *w*/*v* carboxymethyl cellulose solution) at a concentration of 5 mg/mL. The mice received daily gastric gavage of fenofibrate (100 mg/kg) or CPA4-1-100 (100 mg/kg), and db/+ mice received the same vehicle treatment for 12 weeks. Blood glucose level was measured with an automated biochemistry analyzer (HITACHI917; Hitachi, Japan), and the glycated hemoglobin (Hb1Ac) level was determined by a commercial kit (Roche Diagnostic, Mannheim, Germany).

#### *2.8. Measurement of BRB Permeability*

At autopsy, mice were anesthetized by intraperitoneal injection of 10 mg/kg zolazepam (Zoletil, Virbac, Carros, France) and 10 mg/kg xylazine hydrochloride (Rumpun, Bayer, Frankfurt, Germany). The peritoneal and thoracic cavities were opened to secure the heart, and 50 mg/mL fluorescein-dextran (10 kDa Mw, Sigma-Aldrich) and 10 mg/mL Hoechst 33342 (Sigma-Aldrich) dissolved in 1 mL sterile phosphate-bu ffered saline (PBS) were injected into the left ventricle. After 5 min, the eyeballs were removed, fixed in 4% paraformaldehyde for 2 h, and the retina was separated from the eyecup. The separated retina was placed on a slide, mounted with an aqueous mounting medium, and observed under a fluorescence microscope with digital capture (BX41 microscope; Olympus, Tokyo, Japan).

#### *2.9. Preparation of Trypsin-Digested Retinal Vessel*

The isolated retinas were placed in 10% formalin for 2 days. After fixation, the retina was incubated in trypsin (3% in sodium phosphate bu ffer containing 0.1 M sodium fluoride) for 60 min. The vessel structures were separated from retinal cells by gentle rinsing in distilled water. The vascular specimens were mounted on a slide and subjected to periodic acid-Schi ff staining. The specimens were then analyzed under a microscope with digital capture (BX41 microscope; Olympus). The number of acellular capillaries per mm<sup>2</sup> of the capillary area was determined by counting 10 selected microscopic fields.

## *2.10. Immunohistochemical Staining*

Each eye was enucleated and fixed with 4% paraformaldehyde for 24 h. The retina was isolated under a dissecting microscope and entirely washed with water and incubated in 3% trypsin in sodium phosphate bu ffer for 1 h. The trypsin digests were subjected to immunofluorescence staining, as previously described [13]. The slides were incubated with a mouse anti-occludin antibody (Invitrogen) for 1 h at room temperature. Signals were detected using a rhodamine-conjugated goa<sup>t</sup> anti-mouse antibody (Santa Cruz Biotechnology, Paso Robles, CA, USA) and detected by fluorescence microscopy (BX51; Olympus).
