At 21 weeks of age, all ZDF rats developed hyperglycemia compared to the normal ZL rats. As shown in Table 1, the untreated ZDF rats had more than a four-fold increase of fasting blood glucose levels. We monitored the progression of opaque areas by slit-lamp microscopy and observed that lens opacity appeared at 15 weeks of age and progressed linearly up to 21 weeks of age in ZDF rats. In contrast, ZL rats had normal, clear lenses at 21 weeks of age. The mean grade of cataract formation is illustrated in Figure 1A. The grade of the normal ZL rats remained 0 for the duration of the study. However, the value of the ZDF rats was more than 3, which indicated a moderate to severe lens opacity.

By western blotting, we detected multiple and robust immunoreactive bands for argpyrimidine in cataractous lenses from ZDF rats (Figure 1B). Moreover, we observed that numerous TUNEL-positive cells localized within the vicinity of argpyrimidine accumulation. ZL rats had weaker immunoreactivity for argpyrimidine and fewer TUNEL-positive cells in the lens epithelium (Figure 1C).

The NF-κB signaling pathway is affected by AGEs (^{Yamagishi et al., 2005}) and plays an important role in apoptosis (^{Romeo et al., 2002}; ^{Kowluru et al., 2003}). Thus, we investigated NF-κB activity in cataractous lenses. By immunohistochemical staining, we found the activated NF-κB mainly in the nuclei of LECs in cataractous lenses. In ZL rats, the activated NF-κB was rarely detected (Figure 2A). To evaluate NF-κB activation in a quantitative way, we also performed an ELISA-based NF-κB assay. ZDF rats presented a significantly higher activity of NF-κB than normal ZL rats (Figure 2B, P < 0.01).

High glucose has enhanced Bax expression and apoptosis in human LECs (^{Wu et al., 2008}). In retinal pericytes, NF-κB activation by high glucose has increased Bax expression (^{Podesta et al., 2000}; ^{Romeo et al., 2002}). Moreover, the Bax promoter contains an imperfect NF-κB consensus sequence (^{Dixon et al., 1997}). Therefore, to further investigate the potent pro-apoptotic role of NF-κB activation in LECs, we focused on the expression of the proapoptotic Bax protein and the anti-apoptotic Bcl-2 protein. We detected strong immunoreactivity of the Bax protein in the cytoplasm of LECs in the ZDF rats by immunofluorescence staining. However, Bcl-2 immunoreactivity did not differ between normal and diabetic rats (Figure 2A). By western blot analysis, the expression of the Bax protein was greatly increased in the ZDF rats compared to the normal ZL rats. However, the expression level of Bcl-2 did not differ between ZDF and ZL rats. The ratio of Bax to Bcl-2 protein levels in the ZDF rats was significantly higher than the ratio in the normal ZL rats (Figure 2C).

We next investigated whether argpyrimidine accumulation induced a nuclear translocation of NF-κB p65 and apoptotic cell death in HLE-B3 cells by performing immunofluorescence staining. As shown in Figures 3 and 4, MGO treatment increased argpyrimidine formation in a dose-dependent manner (Figure 3A) and elicited robust apoptotic cell death in HLE-B3 cells (Figure 4A). In untreated cells, positive staining for NF-κB p65 was hardly detectable in the nucleus. Not surprisingly, there was more positive staining for NF-κB p65 in the nucleus of the cells after treatment with 400 µM MGO. The MGO-induced increase in nuclear staining of NF-κB p65 was significantly attenuated by the treatment with 25 µM PDTC, as shown Figure pannels 3C and 3D. Interestingly, pretreatment with 0.5 µM PM not only blocked argpyrimidine formation but also inhibited the nuclear translocalization of the NF-κB p65 subunit. Moreover, apoptotic cell death by MGO was significantly suppressed with the addition of 25 µM PDTC, as shown in Figure 4A. Unlike PM treatment, however, PDTC did not inhibit argpyrimidine formation. In addition, the ratio of Bax to Bcl-2 protein levels in the MGO-treated HLE-B3 cells was significantly higher than the ratio in the non-treated cells. However, in the PDTC-treated group, this ratio was similar to normal controls (Figure 4C). These results suggested that NF-κB might play an important role in regulation of AGEs-mediated apoptosis in LECs.

Eight male 6-week-old ZDF rats (ZDF/Gmi-fa/fa) and eight Zucker lean (ZL) counterparts (ZDF/Gmi-lean) were purchased from Charles River Laboratory (Waltham, MA). Rats were individually housed in plastic cages and maintained at 24 ± 2℃ with a 12-h light:dark cycle and received a diet of Purina 5008 (Ralston Purina, St. Louis, MO) and tap water ad libitum. Fasting plasma glucose levels were analyzed using an enzymatic assay based on glucose oxidase and peroxidase (Glucose B-Test Wako, Wako Pure Chemical, Osaka, Japan). At 21 weeks of age, the right eye from each rat was enucleated under deep anesthesia, following an intraperitoneal injection of pentobarbital sodium (30 mg/kg body weight), and fixed in 10% neutralized formalin for 24 h and embedded in paraffin. The left eyes were enucleated and stored until assayed. All procedures involving rats were approved by the Institutional Animal Care and Use Committee at the Korea Institute of Oriental Medicine.

Lenses were examined weekly using a slit lamp (Kowa Genesis, Tokyo, Japan) after dilating the pupils using a tropicamide solution. Lens opacity was scored according to the classification by Ao et al. (¹⁹⁹¹): grade 0, clear normal lens; 1, peripheral vesicles; 2, peripheral vesicles and cortical opacities; 3, diffuse central opacities; and 4, matured nuclear cataract. The mean grade of cataract formation was calculated as the average of grades in all eyes.

The following antibodies were used for immunofluorescence staining of the lens sections: rabbit anti-Bax (1:250, Santa Cruz, CA), rabbit anti-Bcl-2 (1:200, Santa Cruz) and mouse anti-NF-κB antibody MAB3026 (1:200, Chemicon International, CA). The MAB3026 antibody recognizes an epitope overlapping the nuclear localization signal of the p65 subunit of the NF-κB heterodimer (^{Kaltschmidt et al., 1995}). Thus, this antibody selectively binds the IκB-free, activated form of NF-κB. NF-κB was visualized with fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgG (Santa Cruz). For detection of Bax and Bcl-2, the sections were incubated with FITC-conjugated goat anti-rabbit IgG (Santa Cruz).

In order to confirm the apoptotic cell death in argpyrimidine-enriched LECs, a sequential immunostaining was performed. The sections were first stained using a TdT-mediated dUTP nick-end labeling (TUNEL) kit according to the manufacturer's instructions (DeadEnd apoptosis detection system, Promega, WI). Apoptotic cells were detected with FITC-conjugated streptavidin (Santa Cruz). The second sequence of immunostaining using anti-argpyrimidine (1:100, Cosmo bio, Tokyo, Japan) was performed on the same sections with tetramethyl rhodamine isothiocyanate (TRITC)-conjugated goat anti-mouse IgG antibody (Santa Cruz). The slides were mounted in glycerol-based Vectashield media (Vector, CA) containing 4,6-diamidino-2-phenylindole (DAPI).

Nuclear proteins were isolated from lenses by using commercially available nuclear and cytoplasmic extraction kits (NE-PER™ Nuclear and Cytoplasmic Extraction Reagents, Pierce, IL). NF-κB DNA-binding activity was evaluated using the ELISA-based EZ-detected™ Transcription Factor Kit for NF-κB p65 (Pierce, IL) according to the manufacturer's instructions. The amount of active NF-κB p65 in each sample was obtained from a standard curve and normalized to the protein content.

HLE-B3 cells, a human LEC line immortalized by SV-40 viral transformation, were obtained from the ATCC (Manassas, VA). Cells were cultured in minimal essential media with 10% FBS at 37℃ in an incubator with 5% CO₂. Cells were plated and used for experiments upon reaching 80% confluence. Standard culture medium was replaced with fresh serum-free medium 16 h before experiments. To induce the formation of argpyrimidine, cells were then incubated with the indicated concentration of MGO for 24 h. Various concentrations of the NF-κB inhibitor pyrrolidine dithiocarbamate (PDTC) (Sigma, MO) and the AGEs inhibitor pyridoxamine (PM) (Sigma) were added to the culture medium 1 h before MGO treatment.

Intracellular formation of argpyrimidine was assessed by immunohistochemistry as previously described (^{Padayatti et al., 2001b}; ^{Padival et al., 2003}; ^{Kim et al., 2010}). To evaluate NF-κB nuclear translocalization, immunofluorescence staining for the NF-κB p65 subunit was performed on the cultured LECs with a mouse anti-NF-κB p65 antibody (Santa Cruz, CA). NF-κB was visualized with FITC-conjugated goat anti-mouse IgG (Santa Cruz). The nuclei of cultured LECs were then counterstained with DAPI.

Proteins were extracted from both the lenses and cultured LECs and then separated by SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes (Bio-Rad, CA). Membranes were probed with the anti-argpyrimidine, anti-Bax or anti-Bcl-2 antibodies, and then the immune complexes were visualized with an enhanced chemiluminescence detection system (ECL; Amersham Bioscience, NJ). Protein expression levels were determined by analyzing the signals captured on the nitrocellulose membranes using an image analyzer (Las-3000, Fuji photo, Tokyo, Japan).

Cytotoxicity assays were performed using the Cell Counting Kit-8 as described by the manufacturer (Dojindo Laboratory, Kumamoto, Japan). Apoptosis was assessed by using the TUNEL assay (DeadEnd apoptosis detection system, Promega).

The data were analyzed with the paired-t test and a one-way analysis of variance followed by Tukey's multiple comparison test. Statistical analysis was performed by using GraphPad Prism 4.0 (GraphPad, CA).