AGE-RAGE Axis Stimulates Oxidized LDL Uptake into Macrophages through Cyclin-Dependent Kinase 5-CD36 Pathway via Oxidative Stress Generation

Advanced glycation end products (AGEs) are localized in macrophage-derived foam cells within atherosclerotic lesions, which could be associated with the increased risk of atherosclerotic cardiovascular disease under diabetic conditions. Although foam cell formation of macrophages has been shown to be enhanced by AGEs, the underlying molecular mechanism remains unclear. Since cyclin-dependent kinase 5 (Cdk5) is reported to modulate inflammatory responses in macrophages, we investigated whether Cdk5 could be involved in AGE-induced CD36 gene expression and foam cell formation of macrophages. AGEs significantly increased Dil-oxidized low-density lipoprotein (ox-LDL) uptake, and Cdk5 and CD36 gene expression in U937 human macrophages, all of which were inhibited by DNA aptamer raised against RAGE (RAGE-aptamer). Cdk5 and CD36 gene expression levels were correlated with each other. An antioxidant, N-acetyl-l-cysteine, mimicked the effects of RAGE-aptamer on AGE-exposed U937 cells. A selective inhibitor of Cdk5, (R)-DRF053, attenuated the AGE-induced Dil-ox-LDL uptake and CD36 gene expression, whereas anti-CD36 antibody inhibited the Dil-ox-LDL uptake but not Cdk5 gene expression. The present study suggests that AGEs may stimulate ox-LDL uptake into macrophages through the Cdk5–CD36 pathway via RAGE-mediated oxidative stress.


Introduction
Diabetes increases the risk of atherosclerotic cardiovascular disease, and half of diabetic patients die from cardiovascular disorders [1]. According to the recent report of the International Diabetes Federation (IDF) Diabetes Atlas in 2019, the number of diabetes patients aged 20-79 is estimated to be 463 million worldwide. Macrovascular complications, such as coronary artery disease, peripheral artery disease, and stroke, could account for the comorbidities and increased risk of death in patients the inhibition of Cdk5 not only attenuates the expression of inflammatory genes in endothelial cells, but also suppresses the development of atherosclerosis in apolipoprotein E-deficient mice [29]. In addition, Cdk5 is constitutively expressed in macrophages and could contribute to inflammatory reactions in lipopolysaccharide-stimulated macrophages [30]. These observations led us to speculate that Cdk5 could be a modulator of foam cell formation in AGE-exposed macrophages and that the inhibition of Cdk5 in macrophages may have favorable effects on foam cell formation of macrophages. Thus, we examined here whether and how Cdk5 is involved in AGE-induced CD36 gene expression and foam cell formation of macrophages using various types of inhibitors for the AGE-signaling pathway.

RAGE-Aptamer Inhibited the AGE-Induced Dil-ox-LDL Uptake, and Cdk5 and CD36 Gene Expression in U937 Cells
We first examined the effects of AGEs on ox-LDL uptake, and Cdk5 and CD36 gene expression in U937 macrophages. As shown in Figure 1A-D, compared with non-glycated bovine serum albumin (BSA), AGEs remarkably increased ox-LDL uptake into macrophages evaluated by 1,1 -dioctadecyl-3,3,3 ,3 -tetramethylindocarbocyanine perchlorate (Dil)-ox-LDL immunofluorescent staining, which was significantly inhibited by the treatment with RAGE-aptamer. AGE-BSA significantly up-regulated Cdk5 and CD36 mRNA levels in U937 cells, both of which were attenuated by RAGE-aptamer ( Figure 1E,F). Cdk5 and CD36 gene expression levels were correlated with each other ( Figure 1G).
Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 3 of 11 in macrophages and could contribute to inflammatory reactions in lipopolysaccharide-stimulated macrophages [30]. These observations led us to speculate that Cdk5 could be a modulator of foam cell formation in AGE-exposed macrophages and that the inhibition of Cdk5 in macrophages may have favorable effects on foam cell formation of macrophages. Thus, we examined here whether and how Cdk5 is involved in AGE-induced CD36 gene expression and foam cell formation of macrophages using various types of inhibitors for the AGE-signaling pathway.

RAGE-Aptamer Inhibited the AGE-Induced Dil-ox-LDL Uptake, and Cdk5 and CD36 Gene Expression in U937 Cells
We first examined the effects of AGEs on ox-LDL uptake, and Cdk5 and CD36 gene expression in U937 macrophages. As shown in Figure 1A-D, compared with non-glycated bovine serum albumin (BSA), AGEs remarkably increased ox-LDL uptake into macrophages evaluated by 1,1′dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate (Dil)-ox-LDL immunofluorescent staining, which was significantly inhibited by the treatment with RAGE-aptamer. AGE-BSA significantly up-regulated Cdk5 and CD36 mRNA levels in U937 cells, both of which were attenuated by RAGE-aptamer ( Figure 1E,F). Cdk5 and CD36 gene expression levels were correlated with each other ( Figure 1G).

Figure 1.
Effects of RAGE-aptamer on Dil-ox-LDL uptake, Cdk5 and CD36 gene expression in advanced glycation end product (AGE)-exposed U937 cells. U937 macrophages were treated with 100 µg/mL AGE-bovine serum albumin (AGE-BSA), 100 µg/mL non-glycated BSA in the presence or absence of 100 nmol/L RAGE-aptamer, or 100 nmol/L control-aptamer in Roswell Park Memorial Institute (RPMI) 1640 medium supplemented with 10% fetal bovine serum including 100 µg/mL streptomycin and 100 U/mL penicillin at 37 °C in 5% CO2 for 24 h. The cells then were incubated with 10 µg/mL Dil-ox-LDL in the same RPMI 1640 medium for 18 h to evaluate the fluorescence intensity.  Effects of RAGE-aptamer on Dil-ox-LDL uptake, Cdk5 and CD36 gene expression in advanced glycation end product (AGE)-exposed U937 cells. U937 macrophages were treated with 100 µg/mL AGE-bovine serum albumin (AGE-BSA), 100 µg/mL non-glycated BSA in the presence or absence of 100 nmol/L RAGE-aptamer, or 100 nmol/L control-aptamer in Roswell Park Memorial Institute (RPMI) 1640 medium supplemented with 10% fetal bovine serum including 100 µg/mL streptomycin and 100 U/mL penicillin at 37 • C in 5% CO 2 for 24 h. The cells then were incubated with 10 µg/mL Dil-ox-LDL in the same RPMI 1640 medium for 18 h to evaluate the fluorescence intensity.

Effects of NAC, (R)-DRF053 and Anti-CD36 Antibody on AGE-Exposed U937 Cells
We then examined how AGEs increased the Dil-ox-LDL uptake into U937 cells. As shown in Figure 2, an antioxidant, N-acetyl-l-cysteine (NAC), mimicked the effects of RAGE-aptamer on AGE-exposed U937 cells; it significantly inhibited the AGE-induced Dil-ox-LDL uptake, as well as Cdk5 and CD36 gene expression in macrophages. A selective inhibitor of Cdk5, (R)-DRF053, dihydrochloride attenuated the Dil-ox-LDL uptake and CD36 gene expression in AGE-exposed U937 cells, whereas anti-CD36 antibody inhibited the Dil-ox-LDL uptake but not Cdk5 gene expression ( Figure 2). normalized by the intensity of GAPDH mRNA-derived signals and then related to the control values with BSA. The correlation between Cdk5 and CD36 gene expression levels was determined by Pearson's correlation test. Number = 8 for each group. Results are presented as mean ± standard deviation. ★ ★ ★ p < 0.005 vs. treatment of AGE-BSA.

Effects of NAC, (R)-DRF053 and Anti-CD36 Antibody on AGE-Exposed U937 Cells
We then examined how AGEs increased the Dil-ox-LDL uptake into U937 cells. As shown in Figure 2, an antioxidant, N-acetyl-L-cysteine (NAC), mimicked the effects of RAGE-aptamer on AGEexposed U937 cells; it significantly inhibited the AGE-induced Dil-ox-LDL uptake, as well as Cdk5 and CD36 gene expression in macrophages. A selective inhibitor of Cdk5, (R)-DRF053, dihydrochloride attenuated the Dil-ox-LDL uptake and CD36 gene expression in AGE-exposed U937 cells, whereas anti-CD36 antibody inhibited the Dil-ox-LDL uptake but not Cdk5 gene expression ( Figure 2).

Discussion
We recently found that AGEs increase the foam cell formation of, and gene expression of CD36 in, human macrophages [26]. However, the underlying molecular mechanisms remain largely unknown. To address the issue, we first examined the effects of RAGE-aptamer on Dil-ox-LDL uptake, and Cdk5 and CD36 gene expression in AGE-exposed human macrophage U937 cells. We found here that RAGE-aptamer significantly suppressed the AGE-induced foam cell formation evaluated by ox-LDL uptake, and Cdk5 and CD36 gene expression in U937 cells. We previously showed that RAGE-aptamer at 100 nmol/L, the same concentration used in the present experiments, significantly inhibited the binding of AGEs to RAGE and attenuated oxidative stress and inflammatory reactions in human cultured mesangial cells and melanoma cells, whereas 100 nmol/L control-aptamer did not affect the binding of AGEs to RAGE at all [16,18]. These observations suggest that the AGE-induced foam cell formation and the gene expression of Cdk5 and CD36 in U937 cells could be mediated by the interaction with RAGE. In other words, RAGE-aptamer inhibited these harmful effects of AGEs on macrophages by blocking the binding of AGEs to RAGE. Compared with neutralizing antibodies or peptides, aptamers have less immunogenicity and are thermally stable [16,18,36]. Moreover, aptamers can more efficiently penetrate into various tissues over antibodies due to their small size, their pharmacokinetics can be easily improved by chemical modification, and their production costs are low [18,36]. As macrophage foam cell formation is one of the crucial elements of atherosclerosis [20] and the concentration of AGE-BSA used in the present experiments (100 µg/mL) is comparable with that of an in vivo-diabetic situation [2,26], our present study suggests that inhibition of the AGE-RAGE axis by RAGE-aptamer in macrophages may attenuate the progression of atherosclerosis in vivo.
AGEs exert deleterious effects on numerous types of cells through the interaction with RAGE via intracellular oxidative stress generation [37][38][39][40]. In this study, we found that an antioxidant, NAC, mimicked the beneficial effects of RAGE-aptamer on U937 cells; NAC inhibited Dil-ox-LDL uptake into, and Cdk5 and CD36 mRNA up-regulation in, AGE-exposed U937 cells, whereas NAC alone did not affect these parameters in nonglycated BSA-treated cells. These observations suggest that the AGE-RAGE-induced oxidative stress could up-regulate Cdk5 and CD36 mRNA levels in, and induce foam cell formation of, U937 macrophages.
In the present study, gene expression levels of Cdk5 and CD36 in AGE-exposed U937 cells were correlated with each other. CD36 is one of the main scavenger receptors that can mediate ox-LDL uptake into macrophages [20], whereas Cdk5 has been shown to contribute to inflammatory reactions in macrophages [30]. Therefore, we next investigated the involvement of Cdk5 and CD36 in AGE-induced foam cell formation using various types of inhibitors. To the best of our knowledge, we are the first to find that an inhibitor of Cdk5 significantly blocked the AGE-induced CD36 gene expression and Dil-ox-LDL uptake, while an anti-CD36 antibody inhibited the uptake of Dil-ox-LDL into macrophages. In addition, the Cdk5 gene expression levels were not significantly suppressed by the treatment with anti-CD36 antibody. These findings suggest that CD36 could be mainly involved in AGE-induced foam ell formation and that AGE-RAGE-mediated oxidative stress generation may up-regulate CD36 mRNA levels in U937 cells via Cdk5 gene expression. Oxidative stress is reported not only to induce the translocation of Cdk5 from cytoplasm to the nucleus, which could alter its substrate specificity [41], but also to increase the Cdk5 activity in experimental motor neuron disease [42]. Cdk5 has been shown to induce the phosphorylation of peroxisome proliferator-activated receptor γ (PPARγ) and, subsequently, to stimulate its transcriptional activity, being involved in adipogenesis and insulin resistance [43,44]. In addition, PPARγ is known to stimulate macrophage foam cell formation via the transcriptional induction of CD36 [45,46]. Since CD36 expression is enhanced by oxidative stress via PPARγ [47,48], our present study suggests that AGE-RAGE-induced oxidative stress generation may stimulate macrophage foam cell formation through the transcriptional induction of CD36 via the Cdk5-PPARγ pathway (Figure 3). Further studies are needed to clarify whether RAGE-aptamer could attenuate the development and progression of experimental atherosclerosis in animal models by inhibiting macrophage foam cell formation via the suppression of Cdk5-CD36 pathway. needed to clarify whether RAGE-aptamer could attenuate the development and progression of experimental atherosclerosis in animal models by inhibiting macrophage foam cell formation via the suppression of Cdk5-CD36 pathway. The present study has some potential limitations. First, we previously showed that RAGEaptamer alone does not have nonspecific effects on mesangial cells, endothelial cells, and melanoma cells [16,18]. In addition, we found here that, when compared with control-aptamer, RAGE-aptamer significantly inhibited the AGE-induced Dil-ox-LDL uptake, and Cdk5 and CD36 gene expression in macrophages. Therefore, it is unlikely that RAGE-aptamer has off-target effects on macrophages. Second, to justify the 24-h exposure experiments with AGEs, it would be helpful to examine the timedependent effects of AGEs on macrophages. Third, although an antioxidant, NAC, inhibited the harmful effects of AGEs on U937 cells, it would also be interesting to measure intracellular oxidative stress in AGE-exposed macrophages. Fourth, we here examined the effects of AGEs on macrophage foam cell formation focusing only on ox-LDL uptake, and Cdk5 and CD36 gene expression in vitro. It remains unclear whether continuous administration of RAGE-aptamer or NAC could inhibit foam cell formation within the atherosclerotic plaques of animal models via the inhibition of the Cdk5-Cd36 pathway. Fifth, in this study, we evaluated CD36 gene expression with quantitative real-time RT-PCR using a standard curve. Although we did not examine protein expression levels of CD36 in the present experiments, we found here that CD36 gene expression and Dil-ox-LDL uptake in U937 cells were correlated with each other, and that both were significantly inhibited by the treatment with RAGE-aptamer or NAC. As protein expression levels of CD36 are functionally correlated with ox-LDL uptake of macrophages [20], gene expression of CD36 could reflect the cell surface expression of this protein. In addition, the activity of Cdk5 was reported to be highly correlated with that of Cdk5 mRNA [28,49]. In the present study, we found that (R)-DRF053 dihydrochloride, a selective inhibitor of Cdk5, blocked the AGE-induced foam cell formation and CD36 gene expression levels of macrophages. These observations suggest the involvement of the CD36-Cdk5 pathway in ox-LDL The present study has some potential limitations. First, we previously showed that RAGE-aptamer alone does not have nonspecific effects on mesangial cells, endothelial cells, and melanoma cells [16,18]. In addition, we found here that, when compared with control-aptamer, RAGE-aptamer significantly inhibited the AGE-induced Dil-ox-LDL uptake, and Cdk5 and CD36 gene expression in macrophages. Therefore, it is unlikely that RAGE-aptamer has off-target effects on macrophages. Second, to justify the 24-h exposure experiments with AGEs, it would be helpful to examine the time-dependent effects of AGEs on macrophages. Third, although an antioxidant, NAC, inhibited the harmful effects of AGEs on U937 cells, it would also be interesting to measure intracellular oxidative stress in AGE-exposed macrophages. Fourth, we here examined the effects of AGEs on macrophage foam cell formation focusing only on ox-LDL uptake, and Cdk5 and CD36 gene expression in vitro. It remains unclear whether continuous administration of RAGE-aptamer or NAC could inhibit foam cell formation within the atherosclerotic plaques of animal models via the inhibition of the Cdk5-Cd36 pathway. Fifth, in this study, we evaluated CD36 gene expression with quantitative real-time RT-PCR using a standard curve. Although we did not examine protein expression levels of CD36 in the present experiments, we found here that CD36 gene expression and Dil-ox-LDL uptake in U937 cells were correlated with each other, and that both were significantly inhibited by the treatment with RAGE-aptamer or NAC. As protein expression levels of CD36 are functionally correlated with ox-LDL uptake of macrophages [20], gene expression of CD36 could reflect the cell surface expression of this protein. In addition, the activity of Cdk5 was reported to be highly correlated with that of Cdk5 mRNA [28,49]. In the present study, we found that (R)-DRF053 dihydrochloride, a selective inhibitor of Cdk5, blocked the AGE-induced foam cell formation and CD36 gene expression levels of macrophages. These observations suggest the involvement of the CD36-Cdk5 pathway in ox-LDL uptake into macrophages. Sixth, the magnitude of decrease in CD36 gene expression in U937 cells by various inhibitors was similar to that in Dil-ox-LDL uptake. This observation suggests that CD36 may be the main receptor that could mediate the ox-LDL uptake into macrophages. However, it would be interesting to assess the effects of RAGE-aptamer on other scavenger receptors and cholesterol efflux. Furthermore, experiments with lentiviral or retroviral knockdown of CD36 could prove our speculation. Seventh, we did not examine the effects of AGEs on the immunophenotype change of macrophages. Such information would improve the present works. Eighth, it would also be helpful to investigate whether AGEs could stimulate macrophage foam cell formation through the transcriptional activation of CD36 gene via the Cdk5-induced activation of PPARγ and if RAGE-aptamer could prevent the progression of atherosclerosis in diabetic animals by suppressing the Cdk5-PPARγ-CD36 pathway.

Preparation of AGE-BSA
BSA (25 mg/mL) was incubated under sterile conditions with 0.1 mol/L d-glyceraldehyde in 0.2 M NaPO 4 buffer (pH 7.4) at 37 • C for 7 days. Then, unincorporated sugars were removed by dialysis against phosphate-buffered saline, as described previously [26,50]. Control nonglycated BSA was incubated in the same conditions except for the absence of D-glyceraldehyde. The extent of lysine modification of AGE-BSA was 65% as described previously [51].

Uptake of Dil-ox-LDL into Macrophages
The U937 macrophages were incubated in the presence of 10 µg/mL Dil-ox-LDL in RPMI 1640 medium supplemented with 10% fetal bovine serum including 100 µg/mL streptomycin and 100 U/mL penicillin at 37 • C in 5% CO 2 for 18 h [21,26]. After gentle washing with PBS three times, the adherent cells were mounted in Vectashield mounting medium (H-1500, Vector Laboratories, Burlingame, CA, USA) and were imaged with BZ-X710 microscope/software (Keyence, Osaka, Japan). The fluorescent intensity of red color area per cells was quantified as described previously [21,26].

Gene Expression Levels
The total RNA was isolated with QIAGEN reagents (Hilden, Germany), and the gene expression was analyzed by real-time RT-PCR using the gene expression assay of SYBR Green or TaqMan, and a sequence detection system (ABI PRISM 7900, Life Technologies, Palo Alto, CA, USA) as described previously [21][22][23]26,58]. Values were normalized by the intensity of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mRNA-derived signals and then related to the control values with BSA. Primers and probes were as follows: Cdk5, NM_001164410.3, NM_004935.4; CD36, Hs00169627_ml; Gapdh, Hs99999905_ml.

Statistical Analysis
Values are shown as mean ± standard deviation. Analysis of variance (ANOVA) was used for statistical comparison among multiple groups. Pearson's correlation test was performed to examine the correlation between two groups. All statistical analyses were performed by PRISM 7.0 software (GraphPad, San Diego, CA, USA), and p < 0.05 was defined as statistically significant.

Conclusions
The present study is the first to show that the AGE-RAGE axis could stimulate ox-LDL uptake into macrophages via the oxidative stress-Cdk5-CD36 pathway. Inhibition of the AGE-RAGE axis in macrophages may be a novel therapeutic target for the management of accelerated atherosclerosis in diabetes mellitus.