Kefir peptides alleviate high-fat diet-induced atherosclerosis by attenuating macrophage accumulation and oxidative stress in ApoE knockout mice

In the past decade, the high morbidity and mortality of atherosclerotic disease have been prevalent worldwide. High-fat food consumption has been suggested to be an overarching factor for atherosclerosis incidence. This study aims to investigate the effects of kefir peptides on high-fat diet (HFD)-induced atherosclerosis in apolipoprotein E knockout (ApoE−/−) mice. 7-week old male ApoE−/− and normal C57BL/6 mice were randomly divided into five groups (n = 8). Atherosclerotic lesion development in ApoE−/− mice was established after fed the HFD for 12 weeks compared to standard chow diet (SCD)-fed C57BL/6 and ApoE−/− control groups. Kefir peptides oral administration significantly improved atherosclerotic lesion development by protecting against endothelial dysfunction, decreasing oxidative stress, reducing aortic lipid deposition, attenuating macrophage accumulation, and suppressing the inflammatory immune response compared with the HFD/ApoE−/− mock group. Moreover, the high dose of kefir peptides substantially inhibited aortic fibrosis and restored the fibrosis in the aorta root close to that observed in the C57BL/6 normal control group. Our findings show, for the first time, anti-atherosclerotic progression via kefir peptides consumption in HFD-fed ApoE−/− mice. The profitable effects of kefir peptides provide new perspectives for its use as an anti-atherosclerotic agent in the preventive medicine.


Kefir peptides improve body weight change and systemic lipid profiles in HFD-induced atherosclerosis in ApoE
). Food consumption was not significantly different between HFD/Mock and HFD/KPs groups (Supplementary Table 2). After the 12-week treatment, the HFD/ApoE −/− mock group displayed a 25% increment in body weight when compared with the SCD/ApoE −/− control group (P < 0.05). Interestingly, kefir peptides (KPs) intake groups exhibited a dose-dependent reduction of body weight, 9.8% lower in low-dose (100 mg/kg) kefir peptides group (KPs-L) and 14.6% lower in high-dose (400 mg/kg) kefir peptides group (KPs-H) group, when compared with the HFD/ApoE −/− mock group (Fig. 1A). Serum total cholesterol (TC) showed a 3-fold increment in SCD/ApoE −/− control group when compared with SCD/B6 control groups, while HFD/ApoE −/− mock group showed 32% higher TC level than SCD/ApoE −/− control groups (Fig. 1B). Although serum TC showed no significant change in both the HFD/KPs-L and HFD/KPs-H groups when compared with the HFD/ApoE −/− mock group (Fig. 1B), the concentration of serum high-density lipoprotein (HDL) and low-density lipoprotein (LDL) showed a significantly improve in both dosages of KPs treatment groups (Fig. 1C,D). Serum HDL showed a 40% reduction in SCD/ApoE −/− control group when compared with SCD/B6 control groups, and further lower serum HDL level was detected in HFD/ApoE −/− mock group (P < 0.05). Administration of KPs exhibited a dose-dependent upregulation of serum HDL, 40% higher in KPs-L group and 92% higher in KPs-H group, when compared with the HFD/ApoE −/− mock group (Fig. 1C). In addition, serum LDL showed no significant difference between SCD/ApoE −/− and SCD/B6 control groups, while HFD/ApoE −/− mock group showed a 2.2-fold higher serum LDL level than SCD/ApoE −/− control groups. Administration of KPs showed a 60% reduction of serum LDL level in both of the KPs-L and KPs-H groups when compared with the HFD/ApoE −/− mock group (P < 0.05; Fig. 1D). Administration of high-dose KPs had a better Kefir peptides inhibit atherosclerotic formation in HFD-induced atherosclerotic ApoE −/− mice. Atherosclerotic plaques are composed of a lipid-rich core covered with a thin fibrous cap, containing sparse smooth muscle cells and extensive macrophages accumulation 35 . To visualize lipid deposition in the atherosclerotic plaques, aortas were stained and observed with Oil red O staining. As shown in Fig. 2A,B, both HFD/ApoE −/− KPs groups exhibited significantly less lipid deposition in aortic roots but no effect on thoracic portion of aortas compared with HFD/ApoE −/− mock group. The atherosclerotic lesion size was examined by the percentage of area of atherosclerotic plaque compared to the whole cross-sectional aortic sinus area stained with H&E (Fig. 2C). The atherosclerotic lesions showed a 2.6-fold increment in HFD/ApoE −/− mock group compared with the SCD/ApoE −/− control group. Administration of KPs showed a significant less lesion area in a dosage manner, 56% lower in KPs-L group and 75% lower in KPs-H group, when compared with the HFD/ApoE −/− mock group (Fig. 2C).
Moreover, Oil red-O staining showed that small amounts of red-stained lipid deposition could be detected in the aortic root in the SCD/ApoE −/− control group, while a 5.2-fold increment could be detected in HFD/ApoE −/− mock group compared with SCD/ApoE −/− control group. Administration of KPs inhibited lipid deposition in the aortic roots in a dose-dependent effect, 35% lower in KPs-L group and 52% lower in KPs-H group, when compared with the HFD/ApoE −/− mock group (P < 0.05; Fig. 2D). Aortic walls in the HFD/ApoE −/− mock group exhibited a 1.3-fold increment of collagen and smooth muscle fibers production and deposition (Fig. 2E) and the a 1.2-fold increment on plaque fibrotic caps thicknesses (Fig. 2F) compared with SCD/ApoE −/− control group. As anticipated, the deposition of collagen content and the fibrous caps of plaques in the KPs administration groups were thinner and contained less collagen in a dose-dependent manner, 40% thinner/31% lower in KPs-L group and 61% thinner/59% lower in KPs-H group, respectively, when compared with the HFD/ApoE −/− mock group (Fig. 2E,F). Histopathological results indicated that administration of high-dose KPs had a better effect on reduction of lesion area, lipid deposition, and plaque fibrotic caps thicknesses in the atherosclerotic plaques in Weight of mice at 20 weeks of age. Concentrations of (B) blood total cholesterol (TC), (C) high-density lipoprotein (HDL), and (D) low-density lipoprotein (LDL) in different treated mice groups were detected. Data are displayed as the mean ± SEM (n = 8). The statistical analysis was performed according to Duncan's multiple-range method. The labels at the top of columns without the same letters indicate significant differences between groups (P < 0.05).
Kefir peptides decrease oxidative stress in HFD-induced atherosclerotic ApoE −/− mice. The HFD/ApoE −/− mock group showed a 27% reduction of reduction of NO production resulted in a 6.6-fold increment of ROS activity, as assessed by the increase of DCF fluorescence; whilst, their downstream ox-LDL level also showed a 3.8-fold increment when compared with the SCD/ApoE −/− control group ( Fig. 4A-C). Administration of KPs showed a dose-dependent improvement in elevating NO production, decreasing ROS activity and ox-LDL level when compared with HFD/ApoE −/− mock group (NO level: 28% higher in KPs-L and 75% higher in KPs-H; ROS level: 31% lower in KPs-L and 66% lower in KPs-H; ox-LDL level: 13% lower in KPs-L and 39% lower in KPs-H) ( Fig. 4A-C). Administration of high-dose KPs had better inhibitory effects on ROS activities and serum oxLDL level when compared with the HFD/ApoE −/− mock group (Fig. 4).
Kefir peptides reduce plaque macrophage accumulation and modulate inflammatory response in HFD-induced atherosclerotic ApoE −/− mice. We further examined whether KPs inhibited plaque formation through modulating inflammatory response and attenuating macrophage infiltration and accumulation. In the present study, an intracellular macrophage marker, MOMA-2, was used to verify macrophage accumulation in atherosclerotic plaques. The HFD/ApoE −/− mock group showed a 12-fold increment in the content of macrophage accumulation in the lipid-rich site of atherosclerotic plaque, thereby contributing to increased Several inflammatory cytokines, such as interleukin-1β (IL-1β) and tumor necrosis factor-α (TNF-α), involved in the pro-inflammatory signaling in the atherosclerosis progression 39 . In this study, both IL-1β and TNF-α levels in serum and aortic tissue were detected. In serum, the IL-1β and TNF-α levels were increased by 3.6-fold and 2.6-fold, respectively, in the HFD/ApoE −/− mock group compared to the SCD/ApoE −/− control group (Fig. 5C,D). Furthermore, the IL-1β and TNF-α levels in aortic tissue were analyzed via western blot analysis (Fig. 5E). The aortic protein expression levels of IL-1β and TNF-α were increased by 4-fold and 1.5-fold, respectively, in the HFD/ApoE −/− mock group compared to the SCD/ApoE −/− control group (Fig. 5F). Administration of KPs showed a dose-dependent reduction of IL-1β and TNF-α levels in serum and aortic tissue compared to HFD/ApoE −/− mock group ( Fig. 5C-F). In addition, monocyte chemoattractant protein-1 (MCP-1), which can recruit monocyte/macrophage to the site of vascular inflammation, was significantly decreased in the both dosages of KPs administration groups when compared with the HFD/ApoE −/− mock group (Fig. 5E,F). The amazing results indicated that the high-dose of kefir peptides (HFD/ApoE −/− KPs-H) had a greater effect on inhibiting macrophage accumulation and modulating inflammatory response in the aorta, which showed no discernible difference in the levels of inflammation to the SCD/ApoE −/− control group (Fig. 5).

Kefir peptides suppress endothelial cell activation and THP-1 monocytes adhesion and migration under ox-LDL-conditioned cell cultures.
Overexpression and accumulation of ox-LDL in the arterial wall and subsequent monocyte trafficking across the vessel wall to differentiate into macrophages are critical steps during the atherosclerotic plaque development 40,41 . To mimic these steps, phorbol 12-myristate 13-acetate (PMA)-induced THP-1 macrophages were used and then treated with ox-LDL for 48 h and conditioned medium (CM) was collected (Fig. 6A). These ox-LDL CM were used to study the potential inhibitory effects of KPs on the early atherosclerotic processes. As anticipated, ox-LDL CM evoke endothelial activation as shown by upregulation of adhesion molecules mRNA and protein levels in HUVECs (Fig. 6). Accordingly, addition of KPs to the ox-LDL CM showed a significant inhibitory effect on upregulation of endothelial adhesion molecules VCAM-1 and ICAM-1 mRNA expressions (29% and 40% lower, respectively) as well as protein levels (flow cytometry: 40% and 25% lower; western blot: 29% and 27% lower, respectively) in HUVECs after 6 h incubation (Fig. 6B-F). Results showed that KPs had a potential inhibitory effect on oxLDL-stimulated endothelial activation.
We next investigated the inhibitory effects of KPs on the adhesion and following migration ability of THP-1 monocytes (Fig. 7A). Interestingly, THP-1 monocyte strongly adheres to a confluent monolayer of ox-LDL CM pre-incubated HUVECs under static conditions. However, addition of KPs to the ox-LDL CM pre-incubated HUVECs showed an 83% inhibitory efficacy on THP-1 monocytes adhesion (Fig. 7A,B). Furthermore, we investigated THP-1 monocytes migration ability using the transwell system (Fig. 8). Results showed that ox-LDL CM attract lots of THP-1 monocytes to the lower sites of insert and lower chamber, while a significant inhibitory effect was detected (71% and 27% reduction of monocyte in lower site of insert and lower chamber, respectively) under KPs treatment (Fig. 8). Taken together, KPs addition strongly inhibit the monocyte adhesion to endothelial and subsequent migration into the subendothelial region (Figs. 7 and 8).

Discussion
Despite the benefits of aspirin and statin, which are well-established in CVD prevention and therapy, the possibility of aftereffects must be considered 14,16 . In this study, we first demonstrated anti-atherosclerotic progression by kefir peptides consumption in ApoE knockout mice. A substantial increase in aortic lipid deposition, oxidative stress, plaque macrophage accumulation, systemic inflammatory response and aortic fibrosis were induced by HFD-induced atherosclerosis. The beneficial of HFD-induced atherosclerotic mouse model than spontaneously developing atherosclerotic lesions by SCD-fed ApoE −/− mouse model is that HFD can accelerate the progression of atherosclerosis and also elevate AST and ALT levels ( Supplementary Fig. 1A,B) to increase the risk of developing CVD. Treatment with kefir peptides completely indicated an anti-atherogenic effect in a dose-dependent www.nature.com/scientificreports www.nature.com/scientificreports/ manner. Our investigation suggested fruitful effects of kefir peptides to support novel therapy and prevention approaches for anti-atherosclerotic effects. The proposed mechanism of kefir peptides in atherosclerosis treatment are shown in Fig. 9.
Considering human atherosclerosis development takes from months to years or even decades with individual variations, thus two ideal strains of mice, ApoE −/− and LDL receptor deficient (Ldlr −/− ) mice, to study the atherosclerosis are established, which are susceptible to develop atherosclerotic lesion formation during high-fat www.nature.com/scientificreports www.nature.com/scientificreports/ or high-cholesterol diet and several features of the disease mimic to humans 42 . Although both strains develop features of type 2 diabetes and promote atherosclerosis development, Ldlr −/− mice are more prone to develop diabetic phenotype including increased body weight, subcutaneous fat accumulation, high blood glucose, and developed an insulin resistance compared to the ApoE −/− mice and the control wild-type mice during HFD challenge 43 . On a standard chow diet, ApoE −/− mice showed higher total cholesterol level in plasma compared with Ldlr −/− and the control wild-type mice, similar result detected in Fig. 1B, and eventually develop atherosclerotic lesions a few months after birth 44 . Due to the imbalance of the cholesterol deposition in the macrophage and tissue which trigger side effects related to inflammation and extracellular matrix degradation, such as Alzheimer's, steatohepatitis, and respiratory diseases 45 . Therefore, the choice of the diet is very important. A high-cholesterol www.nature.com/scientificreports www.nature.com/scientificreports/ diet but not high-fat diet in given to ApoE −/− mice with high risk factors for neurodegeneration 46 , Alzheimer disease 47,48 , retinal abnormalities 49 , and chronic obstructive pulmonary disease (COPD) 45 . Taken together, we chose the high-fat diet to induce atherosclerosis in ApoE −/− mouse model in this study.
Our previous in vivo animal study demonstrated that kefir peptides prevent hyperlipidemia in HFD-induced obese rats through the inhibition of the lipogenesis pathway through reduced fatty acid synthase (FAS) enzyme, increased p-ACC protein, and stimulation of the lipid oxidation pathway via augmented expression of p-AMPK, PPAR-α, and CPT1 50 . Moreover, we found that kefir peptides improved non-alcoholic fatty liver diseases through manipulation of the JAK2/STAT3 and JAK2/AMPK signaling pathways in a high fructose-induced fatty liver animal model 32 . These current studies indicated that kefir peptides play an important role in lipid metabolism modulation. The precise detection of lipid deposition is critical for monitoring atherosclerotic progression. Although circulating cholesterol accumulation in HFD-induced atherosclerosis was not suppressed by kefir peptide treatment, the aortic lipid deposition was dramatically abolished by kefir peptide administration, which suggested that the anti-atherogenic effect by kefir peptides did not occur through the regulation of cholesterol metabolism. Another view of this result suggested that HFD-induced atherosclerosis in ApoE −/− mice may develop serious atherosclerosis, and the hyperlipidemia was not extremely altered by kefir peptides (Fig. 1B).
ET-1, a potent vascular function indicator, is involved in vasoconstriction, free radical formation and proinflammatory response and results in the development of vascular dysfunction and cardiovascular disease 51 . Furthermore, ox-LDL, induced by ET-1 in human endothelial cells, stimulates ROS generation through NADPH oxidase as previously reported 52,53 . Our results demonstrated that ET-1, ox-LDL, and ROS were significantly decreased by kefir peptide treatment in HFD-induced atherosclerosis mice, which suggested that kefir peptides may have an effect on endothelial function protection (Figs. 3 and 4). Not surprisingly, in a previous study,  www.nature.com/scientificreports www.nature.com/scientificreports/ Friques and his colleagues 54 also found that kefir ameliorates the endothelial function in spontaneously hypertensive rats (SHR) by restoring the ROS/NO imbalance. Our results showed that no significantly change on the blood pressure (BP), including systolic, diastolic, and mean blood pressure after 12 weeks HFD and KPs administration between each group (Supplementary Fig. 1C-E).
Atherosclerosis is a chronic inflammatory disorder of aortic disease. Expressions of both ICAM-1 and VCAM-1 are produced on endothelial cells of atherosclerotic plaque development by several mediators, including ROS and ox-LDL 55 . More importantly, ICAM-1 and VCAM-1, expressed by abnormal endothelium in developing atherosclerotic plaques, are required for the circulating monocyte recruitment to atherosclerotic lesions 56 . Moreover, VCAM-1 has been shown to play a dominant role in the initiation of atherosclerosis 57 . However, evidence has demonstrated that ICAM-1 deficiency substantially protects against atherosclerosis lesion formation in ApoE −/− mice 58,59 , indicating a controversial issue, which is the dominate mediator between ICAM-1 and VCAM-1 in modulating atherosclerosis progression. Our results indicated there was no significant difference in VCAM-1 was observed among the HFD/ApoE −/− mock, HFD/ApoE −/− KPs-L, and HFD/ApoE −/− KPs-H groups; however, the expression of ICAM-1 protein was substantially suppressed in both the HFD/ApoE −/− KPs-L and HFD/ApoE −/− KPs-H groups compared to the HFD/ApoE −/− mock group (Fig. 3). Consistently, ICAM-1 inhibition markedly attenuates macrophage homing to atherosclerotic plaques in ApoE-deficient mice 60 , which suggests ICAM-1 may play a leading role in atherosclerosis procession.
Atherosclerotic lesion development is an inflammatory process accompanied by the recruitment and activation of macrophages, which trigger downstream cascade activation and enhance inflammatory cytokine secretion. Abundant evidence indicates that macrophage-mediated inflammation comprises a central role in atherosclerotic development and may trigger acute thrombotic vascular disease, stroke, myocardial infarction, and sudden cardiac death 61,62 . Our study identified macrophages in the blood stream and homing to atherosclerotic lesions through MCP-1 chemoattractant, which is secreted by aortic endothelial cells. Suppressing the accumulation of lesion macrophages by kefir peptide consumption effectively decreased the inflammatory cytokine IL-1β and TNF-α production (Fig. 5). The anti-inflammatory properties of kefir products have been demonstrated in a mouse model and humans 63 . Our results further suggest that kefir peptides may be absorbed into the blood and influence atherosclerotic development through its immune modulation ability.

Conclusion
In summary, our results indicated that atherosclerotic lesion development in HFD-induced atherosclerotic ApoE −/− mice was improved by oral administration of kefir peptides. We identified reduced aortic lipid deposition, oxidative stress, macrophage accumulation in plaques, systemic IL-1β and TNF-α levels, and aortic root fibrosis and enhanced endothelial function following kefir peptide intake compared with the SCD/ApoE −/− control group. Furthermore, the in vitro cell studies also demonstrated that kefir peptides suppress endothelial cell activation and THP-1 monocytes adhesion and migration under ox-LDL-conditioned cell cultures. These results suggested that kefir peptides play a role in anti-atherosclerosis potentially by modulating the immune cell responses, reducing ROS and ox-LDL productions, and regulating cytokine related pathways. The profitable impacts of kefir peptides provide new perspectives for its use as an anti-atherosclerotic agent in the preventive medicine. Kefir peptides were dissolved in phosphate-buffered saline (PBS; pH 7.4) and orally administered daily for 12 weeks. The mice were sacrificed by intra-peritoneal injection of pentobarbital (60 mg/kg) at 19 weeks of age, after 12 weeks of kefir peptides administration. The heart, aorta, blood, and tissues were collected for further examination.
Western blot analysis. The thoracic aortas and cell lines were homogenized in 300 μl of an RIPA buffer (Sigma-Aldrich, St. Louis, MO, USA) for protein extraction. The protein (50 μg) was then separated via 10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and electrotransferred onto polyvinylidene difluoride (PVDF) membranes. The membranes were incubated in a blocking solution (5% bovine serum albumin) at room temperature for 2 h, followed by incubation with a primary antibody (MCP-1, ICAM-1, VCAM-1, ET-1, TNF-α, IL-1β, α-tubulin or β-actin; Abcam Inc.) overnight at 4 °C. After washing, the membranes were incubated with anti-rabbit or anti-mouse horseradish peroxidase (HRP) conjugated secondary antibody. The membranes were developed using an enhanced chemiluminescence (ECL) western blot detection system (GE Healthcare Biosciences, Pittsburgh, PA, USA) as previously described 50 . Histological and immunofluorescent (IF) staining. The C57BL/6 and ApoE −/− mice were scarified, the thoracic cavity was opened and perfused with PBS via the left ventricle, and then aortic artery was collected. Aortic sinus tissues were fixed in 4% paraformaldehyde overnight, embedded in paraffin, and cut into sections for hematoxylin and eosin (H&E) staining. Atherosclerotic plaques sizes were examined and quantified from four independent sets of H&E-stained section. Oil red O staining was performed to identify the lipid deposition as previously described 32 . Briefly, frozen aortic sinus tissue sections (10-12 µm) and en face aortic samples were stained with Oil red O (Sigma-Aldrich) for 10 min at 37°C, washed and counterstained with hematoxylin for 1 min to determine lipid accumulation. Representative photomicrographs were captured using Olympus Masson's trichrome staining was performed to identify the collagen fibers contents in the aortic sinus tissue sections as previously described 65 A ll analyses were followed the protocol and performed by a pathologist who was blinded to the experimental procedure.

Cell lines. Human monocytic cell line (THP-1) was purchase from Bioresource Collection and Research
Center (Hsinchu, Taiwan). Cell lines were maintained in RPMI-1640 media supplemented with 10% heat-inactivated FBS (Life Technologies Co., Camarillo, CA, USA), 1% penicillin/streptomycin and 50 μM β-mercaptoethanol (Sigma-Aldrich) and were incubated at 37oC in a 5% CO 2 incubator. Human umbilical vein endothelial cell line (HUVECs) was purchased from Lonza Walkersville, Inc. (Walkersville, MD, USA). Cell lines were maintained in the endothelial cell basal medium (EBM-2, Lonza) supplemented with an endothelial cell growth SingleQuot kit (EGM-2, Lonza) and were incubated at 37 °C in a 5% CO 2 incubator. Preparation of conditioned medium. As shown in Fig. 6A, THP-1 cells were differentiated to adherent macrophages by overnight culture in culture medium supplemented with 100 ng/ml phorbol 12-myristate 13-acetate (PMA, Sigma-Aldrich), and then with 35 μg/mL ox-LDL (Life Technologies Co.) for 48 h. After stimulation, supernatant was collected and removed cell debris by centrifugation and passed through 0.2 μm filters. The conditioned medium is referred as an ox-LDL CM and the collected medium from unstimulated THP-1 macrophages is referred as a control CM. Fig. 7A, 3×10 4 HUVECs were seeded on 48 well plate in EBM-2/ EGM-2 medium, to complete confluence. Cells were incubated with ox-LDL CM with or without KPs (100 μg/ml) for 4 h. After stimulation, HUVECs were washed twice with DPSB and 5×10 5 CM-DiI-labeled THP-1 monocytes were added to each well of a 48 well plate and incubated at 37 °C in a 5% CO 2 incubator for 30 min. Then, each well was washed three times with DPBS and high-power field (HPF) digital images were captured using Olympus IX71 microscope with an AxioCam MRc camera. Adhered cells per HPF was counted and calculated using ImageJ software (National Institute of Health, USA). Fig. 8A, 1×10 5 CM-DiI labeled THP-1 monocytes were added to transwell inserts (Millipore, Danvers, MA, USA) with 8 μm pores and incubated in complete medium with or without KPs (100 μg/ml). Assemble transwell inserts in the chamber of 24-wells culture plate with ox-LDL CM or control CM and incubated at 37 °C in a 5% CO 2 incubator for 2 h. Cells on the top of the transwell insert were removed using a cotton swab and only migrated cells on the lower site membrane of transwell insert or suspend in the lower chamber were analyzed. The transwell insert membrane were carefully cut and mounted with FluoreGuard Mounting Medium (Biosystems, Barcelona, Spain) on a glass slide. The high-power field (HPF) digital images were taken using Zeiss AxioScope A1 microscope with an AxioCam MRc camera. Cells per HPF was counted and calculated using ImageJ software. To determine the migrated cell numbers in lower chamber using the haemacytometer.

Migration assay. As shown in
Flow cytometry analysis. Flow cytometry was used to examine the ICAM-1 and VCAM-1 expressions on the endothelial cells surface according to manufacturer's instructions (Abcam Inc.). Briefly, detached cells were fixed with 10% formalin for 20 min, and then incubated 1 h at room temperature with the following antibodies