Glucose-Dependent Insulinotropic Polypeptide Suppresses Foam Cell Formation of Macrophages through Inhibition of the Cyclin-Dependent Kinase 5-CD36 Pathway

Glucose-dependent insulinotropic polypeptide (GIP) has been reported to have an atheroprotective property in animal models. However, the effect of GIP on macrophage foam cell formation, a crucial step of atherosclerosis, remains largely unknown. We investigated the effects of GIP on foam cell formation of, and CD36 expression in, macrophages extracted from GIP receptor-deficient (Gipr−/−) and Gipr+/+ mice and cultured human U937 macrophages by using an agonist for GIP receptor, [D-Ala2]GIP(1–42). Foam cell formation evaluated by esterification of free cholesterol to cholesteryl ester and CD36 gene expression in macrophages isolated from Gipr+/+ mice infused subcutaneously with [D-Ala2]GIP(1–42) were significantly suppressed compared with vehicle-treated mice, while these beneficial effects were not observed in macrophages isolated from Gipr−/− mice infused with [D-Ala2]GIP(1–42). When macrophages were isolated from Gipr+/+ and Gipr−/− mice, and then exposed to [D-Ala2]GIP(1–42), similar results were obtained. [D-Ala2]GIP(1–42) attenuated ox-LDL uptake of, and CD36 gene expression in, human U937 macrophages as well. Gene expression level of cyclin-dependent kinase 5 (Cdk5) was also suppressed by [D-Ala2]GIP(1–42) in U937 cells, which was corelated with that of CD36. A selective inhibitor of Cdk5, (R)-DRF053 mimicked the effects of [D-Ala2]GIP(1–42) in U937 cells. The present study suggests that GIP could inhibit foam cell formation of macrophages by suppressing the Cdk5-CD36 pathway via GIP receptor.


Introduction
Cardiovascular disease (CVD) is one of the devasting complications in diabetes and accounts for the increased risk of mortality in these patients all over the world [1,2]. Indeed, the Emerging Risk Factors Collaboration study revealed that after adjustment for traditional coronary risk factors, the risk of death from cardiovascular causes increased by 2.3-fold in patients with diabetes compared with non-diabetic subjects [1].
In the subendothelial space, low-density lipoprotein (LDL) is changed to oxidized LDL (ox-LDL) by oxidative modifications of apolipoprotein B100, which could stimulate adhesion molecule and chemokine expression within the atherosclerotic plaques, thereby promoting esterification of free cholesterol to cholesteryl ester and foam cell formation of macrophages, one of the initial characteristic features of atherosclerotic CVD [3][4][5].
Biomedicines 2021, 9, 832 2 of 10 Scavenger receptor CD36 has been shown to contribute to ox-LDL uptake by macrophages and subsequent foam cell formation within the atherosclerotic lesions [3,5].
Glucose-dependent insulinotropic polypeptide (GIP), which is one of the incretins produced by K-cells in the small intestine in response to lipids and/or sugars, has been known to promote secretion of insulin in a glucose-dependent manner [6][7][8][9]. Besides its blood glucose-lowering action, GIP has been reported to have atheroprotective actions in animal models [10][11][12][13][14][15]. Indeed, we have previously reported that chronic infusion of active GIP  attenuates aortic plaque formation in apolipoprotein E-null (Apoe −/− ) mice, whose actions were totally independent of blood pressure, body weight, food intake, and plasma lipid or glucose levels [14,15]. Furthermore, ex vivo-treatment with active GIP  has also been found to suppress the foam cell formation of, and CD36 gene expression in, macrophages isolated from Apoe −/− mice [14,15]. However, the underlying molecular mechanism for the inhibitory effects of GIP on foam cell formation of macrophages remains largely unknown. In other words, how GIP could inhibit foam cell formation of, and CD36 gene expression in, macrophages are not sufficiently understood.
Cyclin-dependent kinases (Cdks) have principal roles in regulation of cell cycle, transcription, and differentiation [16,17]. Cyclin-dependent kinase 5 (Cdk5) is considered to be unique because in contrast to other Cdk members, Cdk5 is not a modulator of cell cycle procession [18][19][20], but a regulator of gene modulation and cell survive [21]. Cdk5 could phosphorylate lysine-serine-proline motif of neurofilaments, which plays a crucial role in neuronal cell development, differentiation and migration in supernumerary spinal and cranial motor neurons, while its functional disorder is involved in neurodegenerative disorders, such as Alzheimer's disease [21][22][23]. Recently, Cdk5 has been reported to contribute to endothelial cell senescence [24], and truncated regulatory subunit of Cdk5 has been shown to be accumulated within the atherosclerotic lesions, and long-term suppression of Cdk5 attenuates the progression of atherosclerosis in Apoe −/− mice by reducing the inflammatory reactions [18]. Moreover, Cdk5 is abundantly expressed in macrophages and could mediate the lipopolysaccharide-induced inflammatory reactions [19]. Furthermore, we have recently found that advanced glycation end products (AGEs), aging molecules formed at an accelerated rate under diabetes, stimulate macrophage foam cell formation via the activation of Cdk5-CD36 pathway [25]. Since AGEs are localized in monocyte/macrophagederived foam cells within the atherosclerotic plaques, macrophage foam cell formation evoked by AGEs could cause the atherosclerotic plaque instability and resultantly increase the risk of CVD in diabetes [26][27][28][29][30]. These observations suggest that the Cdk5-CD36 pathway in macrophages could be a therapeutic target for CVD. However as far as we know, there is no paper to show the effects of GIP on foam cell formation and Cdk5-CD36 pathway in macrophages. Therefore, we investigated here whether [D-Ala 2 ]GIP , an agonist for GIP receptor, could inhibit the macrophage foam cell formation by suppressing the Cdk5-CD36 pathway via GIP receptor interaction by using macrophages extracted from [D-Ala 2 ]GIP(1-42)-administrated GIP receptor-deficient (Gipr −/− ) and Gipr +/+ mice, [D-Ala 2 ]GIP(1-42)-treated macrophages isolated from Gipr −/− and Gipr +/+ mice, and human U937 macrophages.

Animal Experiments
The protocol and design of this experiments were approved by the Animal Care Committee of Showa University (permission number: 04141) and Akita University Graduate School of Medicine (approval number: a-1-2520). All experiments, surgeries or sacrifices, were conducted with efforts to minimize the suffering using general anesthesia of isoflurane. Gipr −/− and Gipr +/+ mice were bred (backcrossed to C57BL/6J strain for >18 generations to minimize variability of gene) as described previously [31]. A total of 10 male Gipr −/− and Gipr +/+ mice, respectively, at 7 weeks old were transferred from Akita University Graduate School of Medicine to Animal Institute of Showa University School. The mice were kept on a standard food with free water in the room controlled at 21 • C temperature, under a 12-h light and dark cycle and 40-60% humidity. At 9 weeks old, the mice were subcutaneously infused with [D-Ala 2 ]GIP(1-42) at 25 nmol/kg/day or saline by osmotic mini-pumps. At 13 weeks old, we collected blood samples and peritoneal macrophages from the mice after intraperitoneal injection of thioglycolate broth as described previously [14,15,[32][33][34][35][36]. Also, macrophages were first isolated from Gipr −/− and Gipr +/+ mice at 21 weeks old, respectively, and then exposed to [D-Ala 2 ]GIP(1-42) at 1 nmol/L for 18 h as described previously [14,15].

Dil-ox-LDL Uptake into Macrophages
The U937 cells were treated with 10 µg/mL Dil-ox-LDL in RPMI 1640 medium including 10% FCS at 37 • C in 5% CO 2 for 18 h [25,32,36]. After twice washing with PBS gently, immunofluorescence was observed using Keyence BZ-X710 microscope and analyzed with the Keyence BZ-X710 software (Osaka, Japan). The quantification of fluorescent intensity of red color per cells was calculated as described previously [25,32,36].

Statistical Analysis
Data were presented as mean ± SD. The statistical analyses above two groups were performed by appropriate ANOVA; Unpaired t-test was used to compare two groups. The correlation between two groups was analysed by Peason's correlation test. All analyses were performed using PRISM (version 7.05, GraphPad Inc., San Diego, CA, USA). Differences were defined statistically significant at p < 0.05.  Table 1. There were no significant differences of food intake, body weight, heart rate, SBP or DBP, Total-C, HDL-C, triglycerides, insulin, Total-GIP, FBG and HbA1c among 4 groups. Table 1. Laboratory characteristics of Gipr −/− mice and Gipr +/+ mice at 13 weeks old.

Discussion
We have previously reported that chronic infusion of active GIP(1-42) at 25 nmol/kg/day, the same concentration used in the present experiments, significantly suppresses the foam cell formation of macrophages and subsequent progression of atherosclerosis in Apoe −/− mice [14,15]. However, the underlying molecular mechanisms for this remain largely unclear. To address the issue, we first examined the effects of [D-

Discussion
We have previously reported that chronic infusion of active GIP(1-42) at 25 nmol/kg/day, the same concentration used in the present experiments, significantly suppresses the foam cell formation of macrophages and subsequent progression of atherosclerosis in Apoe −/− mice [14,15]. However, the underlying molecular mechanisms for this remain largely unclear. To address the issue, we first examined the effects of [D-Ala 2 ]GIP(1-42), a dipeptidyl peptidase-4-resistant GIP receptor agonist on foam cell formation of, and CD36 gene expression in, macrophages isolated from Gipr −/− and Gipr +/+ mice. We found here that subcutaneously long-term infusion of [D-Ala 2 ]GIP(1-42) to mice significantly inhibited the foam cell formation evaluated by the radioactivity of cholesterol [ 3 H]oleate and CD36 gene expression in macrophages isolated from Gipr +/+ mice compared with vehicletreated mice, while these beneficial effects of [D-Ala 2 ]GIP(1-42) were not observed in macrophages isolated from Gipr −/− mice. These observations suggest that the inhibitory effects of [D-Ala 2 ]GIP(1-42) on foam cell formation of, and CD36 gene expression in, mouse macrophages could be mediated through the interaction with GIP receptor. CD36 is one of the main scavenger receptors, which could regulate foam cell formation of macrophages, the early characteristic features of atherosclerosis [3,5]. We have recently reported that (1) neutralizing anti-CD36 antibody inhibits ox-LDL uptake into AGE-exposed U937 macrophages [25] and (2) [D-Ala 2 ]GIP(1-42) attenuates cholesterol accumulation of macrophages in Apoe −/− mice in association with the reduction of CD36 expression [14,15]. These observations suggest that the suppressive effect of [D-Ala 2 ]GIP(1-42) on foam cell formation could be mediated by reduction of CD36 gene expression in macrophages.
Cdk5 is constitutively expressed in macrophages, which could contribute to inflammatory reactions in these cell types [19]. In this study, gene expression levels of Cdk5 and Recently, we have found that (R)-DRF053, a selective inhibitor of Cdk5 significantly inhibits foam cell formation of AGE-exposed macrophages by reducing CD36 gene expression [25]. These findings suggest that reduction of Cdk5 gene expression by [D-Ala 2 ]GIP(1-42) may contribute to its suppressive effects on macrophage foam cell formation.
Interaction of GIP with the GIP receptor induces activation of adenosine monophosphateactivated protein kinase (AMPK) through the phospholipase C and calcium/calmodulindependent protein kinase pathway [11,[41][42][43]. Recently, crocin, a carotenoid compound, has been found to activate AMPK and subsequently improve metabolic dysfunction in diabetic mice via the suppression of Cdk5 expression [44,45]. These observations suggest that GIP and its receptor interaction may inhibit macrophage foam cell formation by suppressing the Cdk5-CD36 pathway via the activation of AMPK (Figure 3). mation of, and CD36 gene expression in, macrophages isolated from Gipr −/− and Gipr +/+ mice. We found here that subcutaneously long-term infusion of [D-Ala 2 ]GIP  to mice significantly inhibited the foam cell formation evaluated by the radioactivity of cholesterol [ 3 H]oleate and CD36 gene expression in macrophages isolated from Gipr +/+ mice compared with vehicle-treated mice, while these beneficial effects of [D-Ala 2 ]GIP  were not observed in macrophages isolated from Gipr −/− mice. These observations suggest that the inhibitory effects of [D-Ala 2 ]GIP(1-42) on foam cell formation of, and CD36 gene expression in, mouse macrophages could be mediated through the interaction with GIP receptor. CD36 is one of the main scavenger receptors, which could regulate foam cell formation of macrophages, the early characteristic features of atherosclerosis [3,5]. We have recently reported that (1) neutralizing anti-CD36 antibody inhibits ox-LDL uptake into AGE-exposed U937 macrophages [25] and (2) [D-Ala 2 ]GIP(1-42) attenuates cholesterol accumulation of macrophages in Apoe −/− mice in association with the reduction of CD36 expression [14,15]. These observations suggest that the suppressive effect of [D-Ala 2 ]GIP(1-42) on foam cell formation could be mediated by reduction of CD36 gene expression in macrophages.
Cdk5 is constitutively expressed in macrophages, which could contribute to inflammatory reactions in these cell types [19]. In this study, gene expression levels of Cdk5 and Recently, we have found that (R)-DRF053, a selective inhibitor of Cdk5 significantly inhibits foam cell formation of AGE-exposed macrophages by reducing CD36 gene expression [25]. These findings suggest that reduction of Cdk5 gene expression by [D-Ala 2 ]GIP(1-42) may contribute to its suppressive effects on macrophage foam cell formation.
Interaction of GIP with the GIP receptor induces activation of adenosine monophosphate-activated protein kinase (AMPK) through the phospholipase C and calcium/calmodulin-dependent protein kinase pathway [11,[41][42][43]. Recently, crocin, a carotenoid compound, has been found to activate AMPK and subsequently improve metabolic dysfunction in diabetic mice via the suppression of Cdk5 expression [44,45]. These observations suggest that GIP and its receptor interaction may inhibit macrophage foam cell formation by suppressing the Cdk5-CD36 pathway via the activation of AMPK (Figure 3).  Our study has some potential limitations. First, [D-Ala 2 ]GIP(1-42) is a human type of GIP agonist. Amino acid sequences of human GIP  is almost identical to that of mouse [46]. This could support the compatibility of [D-Ala 2 ]GIP  in mouse experiments. Second, (R)-DRF053 is not a specific selective inhibitor of Cdk5, but it inhibits Cdk1 and other Cdk members among other kinases. However, we cannot obtain an additional structure unrelated inhibitor of Cdk5. It would be interesting to examine the effects of [D-Ala 2 ]GIP(1-42) on CD36 gene expression and form cell formation of macrophages in Cdk5-knockout mice. Third, Cdk5 gene expression in the mice was not examined in the present experiments because of the lack of samples. Therefore, although an inhibitor of Cdk5, (R)-DRF053 inhibited CD36 mRNA levels in U937 cells and that there was a positive correlation between Cdk5 and CD36 mRNA levels ( Figure 2F-H), the conclusion that inhibition of the Cdk5-CD36 pathway via the GIP receptor suppresses macrophage foam cell formation cannot be definitely proved without confirmation in mouse experiments. Fourth, protein expression levels of CD36 were not evaluated, however we found that CD36 gene expression and foam cell formation in mouse and human macrophages were correlated with each other. Therefore, CD36 protein expression could be functionally correlated with foam cell formation of macrophages [5], and CD36 gene expression leads to reflect surface cellular expression of the protein. Additionally, some reports showed that Cdk5 activity is highly correlated with level of Cdk5 gene expression [17,47]. Finally, it would be interesting to examine the effects of AMPK inhibitor on foam cell formation of [D-Ala 2 ]GIP(1-42)-exposed U937 cells.

Conclusions
We found here that [D-Ala 2 ]GIP(1-42) could inhibit foam cell formation of macrophages through Cdk5-CD36 pathway via GIP receptor. Inhibition of Cdk5-CD36 pathway by GIP in macrophage may be a novel therapeutic target for atherosclerotic cardiovascular disease.

Informed Consent Statement: Not applicable.
Data Availability Statement: All data used in this study are available from the corresponding author on reasonable request.