Nanoparticles preparation and characterization
The amorphous SiNPs used in the experiments were prepared by the Stöber method as previously described [36]. The shape and size of the nanoparticles were observed by scanning electron microscopy (SEM; Hitachi S-4800, Japan) and transmission electron microscopy (TEM; JEM2100, Japan). Based on the TEM results, the particle size distribution was analyzed through Image J software. The hydrodynamic size and Zeta potential of SiNPs in deionized water were measured by Zetasizer (Malvern Nano-ZS90, UK). Moreover, an inductively coupled plasma atomic emission spectrometry (ICP-AES; Agilent 720, USA) was used for the purity detection of the synthesized SiNPs, and a gel-clot limulus amebocyte lysate (LAL) assay kit (Bokang, Zhanjiang, China) for endotoxin measurement. In addition, the stock suspension of SiNPs were firstly dispersed by a sonicator (160 W, 20 kHz, 5 min; Bioruptor UCD-200, Belgium), and then diluted by the corresponding exposure media, 0.9% saline (in vivo test) or DMEM (in vitro test).
Animal studies
ApoE-/- mice at the age of 1 - 2 months is commonly used as the animal model for spontaneous atherosclerosis [86]. Male ApoE-/- mice (age, four-week; weight, 18-22 g) were obtained from the Experimental Animal Center of Capital Medical University (Beijing, China) to assess the long-term effect of SiNPs in the development of atherosclerosis. All mice were housed in sterilized filter-topped cages with free access to food and water, and maintained in a specific pathogen-free facility with a constant humidity (50 ± 5%), temperature (24 ± 1 °C) at 12/12-h light/dark cycle. After one week of acclimation, all the mice were supplied with a Western diet (21% fat, 0.15% cholesterol, 34% sucrose) for a rapid establishment of the murine atherosclerosis model. According to the development stage of atherosclerosis in ApoE-/- mice fed with high-fat diet [87], the SiNPs exposure began when mice was 9-week old. Four-week later, the mice were randomly divided into four groups, which were control group and three SiNPs groups at a dose of 1.5, 3.0, or 6.0 mg/kg·bw, respectively.
The applied dose of SiNPs was in reference to an inhalation study in mice [88], in which the doses of SiNPs were evaluated based on the real workplace exposure scenarios. It was reported the occupational exposure level of SiNPs ranged from 1.0 - 27.6 mg/m3 [89]. In considering the lack of a recommended exposure limit for amorphous SiNPs, the permissible concentration-time weighted average (PC-TWA) of amorphous silica dioxide (SiO2), 6 mg/m3 was used. Therefore, a worker (60 kg) exposed to a concentration of 6 mg/m3 SiNPs for 8 h (1 workday) without proper protection would result in an approximate pulmonary dose of 0.44 mg/kg·bw (assuming human under a light exercise condition in workplace with breathing frequency 20 breaths/min, 1024 mL/breath, and pulmonary deposition fraction for 60 nm particles of 0.45 in human) [90,91]. According to the equivalent conversion coefficient of the dose per kilogram of body weight in experiment animals and human [92], the dosage is equivalent to 5.45 mg/kg·bw in mice. Thus, we set the highest dose as 6.0 mg/kg·bw. In addition, Inoue et al. mentioned the number of NPs in ambient air ranged from 2 × 104 to 2 × 105/cm3, with mass concentrations of > 50 μg/m3 near major highways [93]. Here, the actual lung exposure dose of SiNPs (1.5, 3.0 or 6.0 mg/kg·bw) was about 40, 80, 160 μg/mouse/week (based on mice weighting 26 - 29 μg during SiNPs treatment), respectively. According to a previous study [21], the inhalation dose of a mouse is about 5 μg after a one-week exposure at the daily concentration of 50 μg/m3 near major highways (consuming the inhalation rate for mice is 0.052 m3/day, and the mice pulmonary deposition fraction for 60 nm particles of 0.25) [91]. Thereby, the applied dose (1.5, 3.0 or 6.0 mg/kg·bw) in this study was correspondingly 8, 16, or 32 times to the airborne exposure level of NPs.
Mice in SiNPs groups were administered SiNPs suspension through intratracheally instillation, once in every 7 days and 12 times in total, whereas the control mice were instilled with 0.9% saline instead. The volume of intratracheal instillation was controlled to be 50 ± 5 μl. Furthermore, the UBM of three mice per group was performed during the experiment. In addition, the body weight and food intake of mice were monitored and weighed weekly (see details in the supplementary Fig. S1). The experiment was terminated at 1 month after the last SiNPs exposure in order to observe an irreversible effect caused by SiNPs exposure. At the termination of experiment, mice were fasted overnight, blood and aortas were harvested. Serum was extracted from blood and stored at -80 °C until analyzed. All the animal experimentation was performed following the National Guidelines for Animal Care and Use, and approved by the Committee of Laboratory Animal Care and Use in Capital Medical University (Ethical number, AEEI-2018-002).
Ultrasound biomicroscopy
An ultra-high resolution color doppler ultrasound system (Vevo 2100, FUJIFILM Visualsonics, USA) equipped with MS 400/550D mechanical transducers were used, and the ultrasound imaging parameters of LCCA were measured. During the experiment, mice were anaesthetized with isoflurane gas resulting in a heart rate of approximately 500 beats/min, and the hair from the anterior chest wall was carefully shaved, and warm ultrasound transmission gel was liberally applied to ensure optimal image quality. The IMT and PWV were obtained using VEVO LAB software. The EKV two-dimensional dynamic image was analyzed by VEVO VASC software, and indicators representing vascular compliance (diameter/area percentage spread and global radial strain) were obtained. On the basis of a previous description [94], IMT is measured with the vascular lumen-intimal interface selected as the internal measurement site and the media adventitial interface as the external limit. PWV was calculated by the following formula: Length of LCCA/Time of blood flowing through the LCCA. All measurements were repeated three times. All the images were analyzed by another operator blinded to the identities of the animals.
Lipid profiles analysis
Blood samples were collected and centrifuged at 3000 rpm, 4℃ for 10 minutes. The contents of TC, TG, HDL-C, and LDL-C in mice serum were measured by an automatic biochemical analyzer (HITACHI 7180) combining commercial kits (Jiancheng, Nanjing, China). The ratio of HDL-C/LDL-C was calculated, as well as AI according to the formula: AI = (TC - HDL-C)/HDL-C [50].
Histopathological examination
For lesions throughout the aorta, the whole aortas of three mice per group were separated, cut longitudinally after removing excess adipose tissue, and stained with Oil-Red O staining (Solarbio, Beijing, China) for 10 minutes. Afterwards, the stained aortas were placed in 75% alcohol until the artery wall without lesions was cleaned. Images were captured and analyzed by Image J software. Furthermore, for the lesion at the aortic root, the entire aortic root was immersed in 4% paraformaldehyde for 24 hours, embedded into paraffin or optimal cutting temperature (OCT) for histological examination. The cross-sections of the aortic root were stained with H&E, Oil-Red O, Masson and Alizarin Red for the quantification of plaque area, lipid and collagen content, and aortic calcification. It is worth noting that the regional error in lesion size was avoided by acquiring of the sequential cross-sections throughout the entire aortic root as previously described [95]. Ultimately, the largest lesion area was selected for the comparative analysis of plaque (Supplementary material Fig. S2). All slides were scanned with Pannoramic SCAN system (3DHISTECH, Hungary), and measured with CaseViewer software (3DHISTECH, Hungary) or by Image J software. The quantification of each morphological parameter was performed by one investigator blinded for the treatment. The quantification of each morphological parameter was performed by one investigator blinded for the treatment, and reviewed by certified veterinary pathologists.
Immunohistochemical staining
Immunohistochemistry was performed in the paraffin-embedded artery root to determine the expressions of CD68 (a macrophage marker), CD36 (a principal contributor to cholesterol uptake), Bip and CHOP (biomarkers for ER stress) in situ. Briefly, the dehydrated paraffin sections were immersed in 1 mM EDTA (pH = 9) for antigen retrieval, and incubated with 3% hydrogen peroxide to abolish endogenous peroxidase. The sections were incubated with the primary antibody for CD68 (ab125212, Abcam, UK), CD36 (18836, Proteintech, USA), Bip (#3177, CST, USA), or CHOP (15204, Proteintech, USA) overnight at 4 °C, and then incubated with the corresponding secondary antibody and stained with 3,3’-diaminobenzidine (DAB). These primary antibodies were diluted with 5% BSA solution at a ratio of 1: 200. Moreover, the nucleus was stained with hematoxylin. Finally, the percentage of positive-staining area in the whole plaque of aortic root was analyzed using the Image J software. All analyses were performed by one investigator blinded for the treatment.
TEM observation of lesions
The ultrastructure of lesion was observed by using TEM (JEM2100; JEOL, Japan). In brief, the first branch of aorta was fixed by 2.5% glutaraldehyde overnight, rinsed with 0.1 M phosphate buffer, and postfixed with osmic acid for 2 hours. After being dehydrated in ethanol with concentration gradients and acetone, the sample was embedded in epoxy resin. Ultimately, the ultrathin sections (50 nm) were obtained and imaged under TEM.
Cell culture and treatment
Mouse macrophage cell line, RAW264.7 cells were cultured in DMEM (ThermoFisher, USA) with 10% fetal bovine serum (FBS; ThermoFisher, USA) at 37°C in a 5% CO2 incubator. SiNPs were diluted by DMEM to appropriate concentrations, and an ER stress inhibitor, 4-phenylbutyric acid (4-PBA; Selleck, USA) was applied (3 mM, 6 hours). The dosage of SiNPs was set according to the cell viability analysis by using MTT assay. Since a significant acute toxicity (24 hours) was seen in SiNPs-treated group at a concentration of 50 μg/ml, while simultaneously with cell viability > 70%, the exposure mode of SiNPs (50 μg/ml, 24 hours) was used in the subsequent in vitro experiments. Similarly, the application of 4-PBA was set up through MTT assay and also verified as evidenced by an efficient inhibition on the up-regulated expressions of Bip and Chop induced by SiNPs. See details in the supplementary Fig. S3. After SiNPs with or without 4-PBA treatment, cells were harvested for the following measurement.
Cellular morphology observation and particle internalization analysis
After 50 μg/ml SiNPs treatment for 24 hours, the cellular morphology and alterations of cellular ultrastructure were observed by SEM (S-4800, Hitachi, Japan), and TEM (JEM2100, JEOL, Japan). Based on the TEM image, the particle uptake and internalization were verified by energy dispersive spectrometry (EDS; Bruker-XFlash6/60, Germany).
Intracellular lipid measurement
Intracellular lipid droplets were determined by Oil-Red O staining as previously described [43]. In brief, cells were fixed with 4% paraformaldehyde, and stained with Oil-Red O working solution for 30 minutes after assimilation of 60% isopropanol. The excess dye was washed away with 60% isopropanol and the cells were observed under an Olympus IX81 microscope (Tokyo, Japan). Also, the intracellular content of total cholesterol (TC) was measured by a total cholesterol assay kit (Applygen, Beijing, China) according to the manufacturer’s protocol. Ultimately, the intracellular TC content was calibrated using protein mass.
Quantitative real-time RT-PCR
The total cellular RNA was extracted by using a RNAsimple Total RNA kit (Tiangen, Beijing, China), and reversely transcribed to cDNA using a PrimeScript™ RT reagent kit (TaKaRa, Japan). The quantitative PCR was performed by using the SYBR Premix Ex TaqTM II (Takara, Japan) in a real-time PCR machine (Bio-Rad, USA). The relative expression in mRNA levels of lipid transport (CD36, SRA1, ABCA1, ABCG1 and SRBI) and esterification (ACAT1), and also ER stress indicators (Bip and CHOP) were quantified. Each experiment was conducted in triplicate with β-actin as the internal standard. Primers used for quantitative PCR analysis were listed in the supplementary file (Table S1).
Western Blot assay
The whole cellular protein was extracted by a Protein Rapid Extraction kit (KeyGEN, China), and quantified by BCA protein assay (Dingguo, China). After denaturation, protein lysate was separated with SDS-PAGE, and transferred to a nitrocellulose membrane (Pall, Germany). The membrane was blocked with Tris-buffered saline (TBS) solution containing 5% skim milk powder for 1 hour at room temperature. After wash three times with TBST (TBS with 0.05% Tween-20), membrane was incubated with the primary antibody diluted by TBST solution (1 : 1000), including Bip (#3177, CST, USA), CHOP (#2895, CST, USA), CD36 (ab64014, Abcam, UK), GADPH (#5174, CST, USA) and β-actin (66009, Proteintech, USA) overnight. After three-time wash with TBST, membrane was incubated with the corresponding fluorescent secondary antibody (LI-COR, Gene Company Limited, Hong Kong) for 1 hour at room temperature, and ultimately detected using Odyssey® CLx imaging system (Gene Company Limited, Hong Kong). At least three independent experiments were performed. The relative expression level of protein was analyzed by Image Studio™ quantification software (Gene Company Limited, Hong Kong) with β-actin or GAPDH as internal control, and normalized to the control group.
Statistical analysis
Data were expressed as mean ± standard deviation. Significant differences in ultrasound data were analyzed by ANOVA of repeated measurement data. The t-test of independent samples was used to analyze the significant difference of the intracellular cholesterol content and lipid influx/efflux factor expression in Raw264.7 cells, as well as CD36 expression in mouse plaque lesions. Significant differences in the remaining data were analyzed by one-way ANOVA. The LSD test was selected for post hoc test of homogeneous data, whereas Dunnett's T3 test for the post hoc test of heterogeneous data. A two-tailed Pearson correlation test was applied to determine the correlation between the lesion areas in aortic root and serum lipid levels. All data were analyzed by SPSS 20.0 software, and p value < 0.05 indicates statistical significance.