**1. Introduction**

Hyperglycemia induces the formation and accumulation of advanced glycation end products (AGEs), and these products are present at high levels in the blood and tissue of diabetic patients [1,2]. AGEs are accumulated at high levels in the tissues of patients with age-related diseases, such as chronic obstructive pulmonary disease, cardiovascular diseases, osteoporosis, and neurodegenerative diseases [3]. AGEs are formed by oxidative and non-oxidative reactions, and they a ffect the biochemical and physical properties of proteins in tissues. AGE formation is triggered by high glucose-induced oxidative stress and fluorescent protein cross-linking [4].

Diabetic retinopathy (DR) is a complication of diabetes that causes damage to retinal blood vessels [5]. Elevated AGE levels increase the breakdown of the blood-retinal barrier (BRB), adhesion of leukocytes, and retinal vascular injury, leading to serious impairment of vision. The BRB consists of

inner and outer nuclear layers. The inner nuclear layer of the BRB consists of tight junctions between endothelial cells and pericytes, whereas the outer nuclear layer of the BRB is formed by tight junctions between retinal pigment epithelial cells [6]. The advent of anti-vascular endothelial growth factor (VEGF) has shown a remarkable effect in DR patients; however, most of DR patients have failed to achieve significant clinical visual improvement. The treatment of DR remains challenging. Inhibition of AGE formation has been suggested as a therapeutic target for improving insulin resistance in diabetes with obesity [7]. For example, fenofibrate, a peroxisome proliferator-activated receptor alpha (PPARα) agonist, was approved for slowing down the progression of DR in patients with type 2 diabetes mellitus in October 2013 in Australia [8]. Moreover, pyridoxamine, an inhibitor of AGE formation, has been shown to ameliorate insulin resistance in obese, type 2 diabetic mice [9]. The identification of inhibitors of AGE formation from natural sources has gained much attention.

During the last 15 years, we have screened inhibitors of AGE formation from natural products [10–12]. *Aster koraiensis* extract prevents retinal pericyte apoptosis in streptozotocin (STZ)-induced diabetic rats [13]. *Osteomeles schweinae* extract inhibits methylglyoxal (an active precursor in the formation of AGEs)-induced apoptosis in human retinal pigment epithelial cells [14]. Cinnamomi Ramulus (the twig of *Cinnamomum cassia* Blume; Lauraceae) and Paeoniae Radix (the root of *Paeonia lactiflora* Pallas; Paeoniaceae) have been shown to exert efficacy in inhibiting the formation of AGEs in our previous study. Cinnamomi Ramulus has traditionally been used for its anti-inflammatory, antioxidant, and neuroinflammatory effects [15]. Its marker compounds include coumarin, cinnamyl alcohol, and cinnamic acid. In humans, the effect of cinnamon is controversial; it significantly decreases plasma glucose to the baseline levels, without causing adverse effects nor significant glycemic and inflammatory indicators in patients with type 2 diabetes [16,17]. Paeoniae Radix has been used in traditional medicine for treating inflammatory diseases owing to its anti-allergic, immunoregulatory, and analgesic effects [18]. The marker compounds of Paeoniae Radix include gallic acid, albiflorin, paeoniflorin, and benzoic acid [19]. In a preliminary study, we evaluated the efficacy of inhibition of AGE formation with different combinations of the two herbs to obtain the best formulation. It showed a different inhibitory effect according to the ratio, and it was the best at CPA 4-1 (Cinnamomi Ramulus:Paeoniae Radix = 1:8). Here, we tested a mixture of the CPA4-1 to investigate the optimum ratio for inhibiting AGE formation in the human retinal pigment epithelial cells (ARPE-19). In addition, we examined the therapeutic efficacy of CPA4-1 in preventing DR in db/db mice, a well-established model of obesity-induced type 2 diabetes with retinal neurodegeneration [20,21].

#### **2. Materials and Methods**

#### *2.1. Preparation of the CPA4-1*

Cinnamomi Ramulus and Paeoniae Radix were purchased from a traditional herbal medicine store in Daejeon, Republic of Korea, in April 2016 and identified by Prof. Ki Hwan Bae (College of Pharmacy, Chungnam National University, Republic of Korea). Voucher specimens of Cinnamomi Ramulus (KIOM-CIRA-2016) and Paeoniae Radix (KIOM-PARA-2016) have been deposited in the Herbarium of Korea Institute of Oriental Medicine (KIOM), Republic of Korea. The herbal combination was prepared at a Cinnamomi Ramulus to Paeoniae Radix ratio of 1:8 (*m*/*m*). For preparing CPA4-1 extract, 20 g of Cinnamomi Ramulus and 160 g of Paeoniae Radix were weighed accurately and mixed. Distilled water (1080 mL) was added to the mixed herbs (180 g) and extracted at 100 ◦C for 3 h using a reflux extractor (MS-DM607, M-TOPS, Seoul, Korea). The extract solutions were filtered and evaporated under reduced pressure using a rotary evaporator (N-1200A; Eyela, Tokyo, Japan) at 50 ◦C and then freeze-dried using a freeze dryer (FDU-2100; Eyela) at −80 ◦C for 72 h to obtain an extract powder of CPA4-1 (18.5 g; yield, 10.3%). This sample extract (100 mg) was dissolved in 50% methanol (10 mL), and the solution was filtered through a 0.45-μm syringe filter (Whatman, Clifton, NJ) prior to injection. Standard stock solutions of five reference standards (all at 1 mg/mL) were prepared in HPLC-grade MeOH, stored at <4 ◦C, and used for HPLC analyses after serial dilution in MeOH.
