Increased caveolin 1 by human antigen R exacerbates Porphyromonas gingivali-induced atherosclerosis by modulating oxidative stress and inflammatory responses

Establishment of P. gingivalis-induced atherosclerotic mouse model

This study was approved by the Animal Experiment Ethics Committee of the Second People’s Hospital of Lanzhou (LZSKQYY-LL-ZD-2024-01). A total of 15 apolipoprotein E-deficient (ApoE^−/−) mice (males, 6 weeks old, 20–22 g) were purchased from the medical experimental center of Lanzhou University. The mice were kept in a standard environment with controlled temperature (21–23°C) and humidity (60% ± 5%) under specific pathogen-free conditions in individually ventilated cages. They had access to food and water ad libitum. To study and analyze the relationship between AS formation and P. gingivalis infection, we randomly divided ApoE^−/− mice into three groups: ApoE^−/− group, ApoE^−/− + P. gingivalis + si-NC group, and ApoE^−/− + P. gingivalis + si-Cav-1 group (n = 5). The ApoE^−/− mice were injected with P. gingivalis (10⁷ colony-forming unit in 0.1 mL of phosphate buffered saline [PBS]) through the tail vein once a week.^[19] The small interfering RNA (siRNA) sequences transfected into the mice were as follows: CCACCTTCACTGTGACGAAAT (siCav-1) and TTCTCCGAACGTGTCACGT (si-NC). siRNA (100 μg) was injected into the ApoE^−/− mice through the tail vein every 3 days 5 times (20 μg/injection). After 16 weeks, the mice were exposed to isoflurane (1.5–2%; 26675-46-7, Yaji Biological, Shanghai, China) and then euthanized through cervical dislocation,^[20] and blood and aorta samples were collected. Blood collected from the orbital vascular plexus was centrifuged at 3000×g for 8 min to obtain the serum. Finally, the blood and aorta were stored at −80°C for experiments.

Cell culture and transfection

P. gingivalis strain American type culture collection 33277 and HUVECs were obtained from the American Tissue Culture Collection (33277, Manassas, VA, USA) and ScienCell Research Laboratories (8000, Carlsbad, CA, USA), respectively. HUVECs were authenticated by examining short tandem repeat profiling, cell morphology and growth characteristics, and detected mycoplasma contamination using the Mycoplasma polymerase chain reaction (PCR) Detection Kit Mycoplasma Detection Kit (iCell-MD01-001, iCell Bioscience Inc, Shanghai, China). An endothelial cell (ScienCell, Carlsbad, CA, USA) medium containing 1% endothelial cell growth supplement (1052, ScienCell, Carlsbad, CA, USA), 1% penicillin/streptomycin (ST488S, Beyotime, Shanghai, China), and 5% fetal bovine serum (C0235, Beyotime, Shanghai, China) was prepared, and the HUVECs were seeded into the medium. The medium was then placed in a carbon dioxide (CO₂) incubator. The cells were observed, and when the density was at least 80%, the cells were digested with trypsin (P4201, Beyotime, Shanghai, China). Passage manipulation was then performed on the cells. The HUVECs were stimulated with P. gingivalis (multiplicity of infection [MOI]: 25, 50, and 100) for 24 h or with P. gingivalis (MOI: 100) for 6, 12, and 24 h. Non-targeted control siRNA (si-NC: 5'-UUC UCC GAA CGU GUC ACG UTT-3'), specific siRNA against Cav-1 (siCav-1: 5'-CCC ACU CUU UGA AGC UGU UTT-3'), HuR (si-HuR: 5'-UUC AUC GUC CUG UGU CGA ACC-3'), and pcDNA3.1-HuR were synthesized by Shanghai Sangon Biotech Co., Ltd. Lipofectamine 3000 (10 μL; lipo3000, Thermo Fisher, Waltham, MA, USA) or si-Cav-1 or si-NC (100 nM, 10 μL) was, respectively, added in OPTI-MEM (250 μL) and stood alone at room temperature for 5 min, followed by mixed incubation for 20 min to obtain liposome transfection complex. When the confluence of the HUVECs reached 70–80%, the complexes were transferred to a six-well culture plate and incubated at 37°C in 5% CO₂ for 48 h. The knockdown efficiency was then assessed through Western blot.

RNA stability and reverse transcription quantitative PCR (RT-qPCR)

The HUVECs were treated with actinomycin D (5 μg/mL; 50–76–0, Sigma-Aldrich, St. Louis, Missouri, USA) for 0, 2, and 4 h and then incubated with TRIzol reagent (15596–026; Thermo Fisher, Waltham, MA, USA). RNA was extracted, and the remaining mRNA was analyzed through RT-qPCR.

Total RNA was extracted from the cells with TRIzol reagent and then reverse transcribed to produce complementary DNA, which was used as a template for subsequent quantitative PCR (qPCR). A qPCR machine (PIKOREAL96; Thermo Fisher, Waltham, MA, USA) and SYBR Premix EX TaqTM kits (DRR820A, TakaRa, Japan) were used. The cycle threshold value was determined, and then, the 2^−ΔΔCt method was used in calculating the Cav-1 level. The following primers pairs were used: Cav-1: 5'-GCG ACC CTA AAC ACC TCA AC-3' (forward) and 5'-ATG CCG TGT CAA ACT GTG TGT C (reverse); GAPDH: 5'-GCA CCG TCA AGG CTG AGA AC-3' (forward); and 5'-TGG TGA AGA CGC CAG TGG A-3' (reverse).

Western blotting

Radioimmunoprecipitation assay lysis buffer (P0013B, Beyotime, Shanghai, China) was used to extract the total protein, followed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) electrophoresis and transfer of the protein to polyvinylidene fluoride (PVDF; IPVH00010, Millipore; Billerica, MA, USA) membranes. SDS-PAGE gels (10%) were prepared, loaded at 30 μg of protein per well, electrophoresed onto PVDF membranes, and blocked in 5% bovine serum albumin for 120 min. Rabbit primary antibodies (Cav-1, 1:1000, 3267, Cell Signaling Technology, Danvers, MA, USA; HuR, 1:1000, ab200342, Abcam, Cambridge, UK; intercellular cell adhesion molecule-1 (ICAM-1), 1:1000, ab109361, Abcam, Cambridge, UK; β-actin, 1:1000, ab8226, Abcam, Cambridge, UK) were added, and the membranes were placed on ice. On the following day, the membranes were washed after the incubation cassette was removed and then re-suspended at room temperature for 30 min. Horseradish peroxidase -conjugated goat-antirabbit immunoglobulin G (IgG) (1:5000, ab6721, Abcam, Cambridge, UK) was added, and the membranes were incubated at room temperature for 120 min. The membrane was washed 3 times with Tris-buffered saline containing Tween-20 (ST673, Beyotime, Shanghai, China) and developed through chemiluminescent detection (ChemiDoc XRS+, Bio-Rad ChemiDoc, California, USA). The relative protein gray values were calculated by normalizing to β-actin protein as a loading reference using Image J (National Institutes of Health, Bethesda, MA, USA).

Enzyme-linked immunosorbent assay (ELISA) analysis and reactive oxygen species (ROS) detection

The levels of IL-6, tumor necrosis factor-α (TNF-α), and IL-1β in the cells and sera were detected using relevant product professional detection ELISA kits (IL-6, PI330, PI326, Beyotime, Shanghai, China; TNF-α, PT518, PT512, Beyotime, Shanghai, China; IL-1β, PI305, PI301, Beyotime, Shanghai, China). The detection was conducted in strict accordance with the instructions. A ROS assay kit (ab113851, Abcam, Cambridge, UK) was used in quantitatively assessing ROS levels in the cell samples. The cells were attached to 96-well culture plates, washed with buffer, and stained with 2',7'-dichlorodihydrofluorescein diacetate for 45 min at 37°C. The images and fluorescence intensity were captured and measured using a fluorescence microscope (excitation/emission at 485 nm/535 nm; IX73, Olympus, Tokyo, Japan) and ImageJ software (National Institutes of Health, Bethesda, MA, USA). In addition, the ROS levels in the aortic cryostat sections were assessed using the fluorescent probe dihydroethidium (DHE, D23107, Invitrogen, Waltham, MA, United States). The freshly prepared 5 μm-thick sections of mouse aortas were incubated with 5 μM DHE solution in the dark at 37°C for 10 min. The sections were rinsed with PBS 3 times at 4°C for the removal of excess probes. Subsequently, fluorescence imaging was performed using an excitation wavelength of 518 nm and emission wavelength of 606 nm. The resulting fluorescence, which is indicative of ROS levels, was measured and quantified using ImageJ software (National Institutes of Health, Bethesda, MA, USA).

Cell counting kit-8 (CCK-8)

The HUVECs were digested and prepared as cell suspensions, which were then seeded into a 96-well plate at a density of 2 × 10³cells/well for 24 h under standard conditions. At 0, 6, 12, or 24 h, each well was treated with CCK-8 reagent (10 μL; CA1210, Solaibio, Beijing, China) and mixed for 2 h. Using a microplate reader (Multiskan FC, Thermo Fisher, Waltham, MA, USA), the absorbance at 450 nm was measured to detect cell viability.

Cell apoptosis

An annexin V- Fluorescein Isothiocyanate (FITC)/propidium iodide (PI) apoptosis detection kit (556547, BD Bioscience, San Jose, CA, USA) was used in cell apoptosis analysis. HUVECs were resuspended with 500 μL of binding buffer, 5 μL of annexin V-FITC, and 5 μL of PI in the dark at 4°C for 15 min. Finally, apoptosis levels were analyzed with a flow cytometry kit (FACSVERSE, BD Bioscience, San Jose, CA, USA) according to the manufacturer’s instruction.

RNA immunoprecipitation (RIP) and biotin pull-down

The binding of Cav-1 and HuR was analyzed using a MEGshortscript T7 kit (AM1354, Thermo Fisher, USA) and an EZ-Magna RIP RNA-binding protein immunoprecipitation kit (17-701, Millipore, Billerica, MA, USA). For the RNA immunoprecipitation (RIP) experiment, a buffer containing control IgG-conjugated magnetic beads or HuR antibodies and cell lysates were prepared. Each sample was incubated at 4°C for 12 h. For the biotin pull-down experiment, streptavidin-conjugated beads containing Cav-1 mRNA 3’ and 5’ UTRs or coding regions were prepared, and cell lysates were added. Each sample was left to stand at 4°C for 3 h. The eluents obtained with the two procedures were respectively analyzed through Western blotting and qPCR.

Vascular oil red O staining

The aorta fixed with 4% paraformaldehyde was cut along the long axis of the blood vessel under a dissection microscope and washed twice with PBS, and the blood vessel was expanded and fixed on a white bottom plate. The aorta was immersed in an Oil Red O staining solution (O8010, Solarbio, Beijing, China) and incubated at 37°C for 1 h. The differentiation of the cells was induced in 75% ethanol until the lipid droplets were orange or bright red and then photographed. The area of lesion positively stained by Oil Red O in the entire aorta or aortic root was evaluated using Image J software (National Institutes of Health, Bethesda, MA, USA).

Statistical analysis

After data collection, statistical analysis was performed using the Statistical Package for the Social Sciences 18.0 software (IBM Corp., USA) and GraphPad Prism 7 (GraphPad Software Inc., San Diego, CA, USA). Whether the data were normally distributed and were determined with the Shapiro– Wilk test, and the homogeneity of variance was verified with Levene’s test. Data between groups were analyzed using t-tests and one-way analysis of variance. Then, Tukey’s post hoc test was performed for pairwise comparisons. Data were expressed as mean ± standard deviation, and P < 0.05 was considered significant.

P. gingivalis suppressed cell viability and increased Cav-1 expression level

To study the relationship of P. gingivalis to the development of AS, we employed the different multiplicity of P. gingivalis to infect the HUVECs and then detected cell viability. The CCK-8 assay results showed that P. gingivalis infection remarkably decreased the cell viability of the HUVECs, and the inhibitory effect on cell viability was evident at 100 MOI (P < 0.05); [Figure 1a]. Moreover, the cell viability of the HUVECs markedly weakened as infection time increased (P < 0.05); [Figure 1b]. Given that P. gingivalis may contribute to Cav-1 expression,^[15] we explored and analyzed HUVECs infected with P. gingivalis and examined their Cav-1 levels. The results showed that after infection with P. gingivalis, Cav-1 levels showed a significant upward trend with time and concentration (P < 0.05); [Figure 1c and d].

Close

Figure 1:

Porphyromonas gingivalis suppressed cell viability and increased caveolin 1 (Cav-1) expression level. (a) The viability of human umbilical vein endothelial cells (HUVECs) infected with P. gingivalis at different concentrations (multiplicity of infection [MOI]: 0, 25, 50, 100; n = 5) with cell counting kit-8 (CCK-8). (b) The viability of HUVECs infected with P. gingivalis at a specified MOI (100) for a specified period of time (0, 6, 12, and 24 h) was analyzed with the CCK-8 assay. (c and d) The concentration- and time-dependent expression of Cav-1 in the HUVECs exposed to P. gingivalis (n = 3) was analyzed through Western blot analysis. a and c: *P < 0.05 versus MOI (0) group; b and d: *P < 0.05 versus time (0 h) group.

Export to PPT

Cav-1 triggered P. gingivalis-induced oxidative stress, inflammatory responses, and cell apoptosis of HUVECs

To explore whether Cav-1 is involved in P. gingivalis-accelerated AS, we exposed the HUVECs to P. gingivalis to mimic P. gingivalis-accelerated AS in vitro and then decreased Cav-1 level with siRNA. Western blotting revealed that HUVECs subjected to P. gingivalis showed substantial increase in Cav-1 level, whereas the transfection of si-Cav-1 reduced the P. gingivalis-accelerated upregulation of Cav-1 (P < 0.05); [Figure 2a]. P. gingivalis substantially promoted the production of IL-1β, IL-6, and TNF-α, while silencing of Cav-1 dramatically restored these changes in HUVECs (P < 0.05); [Figure 2b-d]. Consistently, the production of ROS induced by P. gingivalis was substantially abolished by si-Cav-1 (P < 0.05); [Figure 2e and f]. Flow cytometry showed that silencing of Cav-1 significantly reduced cell apoptosis in the HUVECs infected with P. gingivalis (P < 0.05); [Figure 2g and h]. The Western blot results further revealed that P. gingivalis exposure increased the expression of ICAM-1, whereas Cav-1 silencing in HUVECs reversed this change (P < 0.05); [Figure 2i].

Close

Figure 2:

Caveolin 1 (Cav-1) triggered Porphyromonas gingivalis (P. gingivalis)-induced oxidative stress, inflammatory responses, and cell apoptosis of human umbilical vein endothelial cells. (a) Cav-1 protein levels detected using Western blotting. (b-f) Interleukin (IL)-1β, IL-6, tumor necrosis factor-α, and reactive oxygen species levels detected using enzyme-linked immunosorbent assay kits and 2',7'-dichlorodihydrofluorescein diacetate fluorescence probe, respectively, scale bar = 100 μm (n = 5). (g and h) Detected cell apoptosis levels using flow cytometry (n = 3). (i) Detected intercellular cell adhesion molecule-1 (ICAM-1) protein levels using Western blotting. (n = 3). si-NC: small interfering RNA negative control, si-Cav-1: small interfering RNA Caveolin 1, V-FITC: V-Fluorescein Isothiocyanate.*P < 0.05.

Export to PPT

HuR directly bound and stabilized Cav-1 mRNA

The 3’ UTR of Cav-1 mRNA contains ARE structure.^[18] HuR, as an RNA-binding protein, can bind to a variety of 3’ UTR mRNAs containing the ARE structure and improve their stability. Therefore, we hypothesized that HuR exerts a regulatory effect on Cav-1 mRNAs. In HUVECs, when HuR expression decreased or increased, Cav-1 protein expression was decreased and increased, respectively (P < 0.05), but its mRNA level did not change (P > 0.05). This result indicated that HuR had a regulatory effect on Cav-1 mRNA stability [Figure 3a-c]. Subsequently, we collected cells at different time points after the addition of actinomycin D and detected the remaining amount of Cav-1 mRNA through RT-qPCR. The results showed that the half-life of a Cav-1 mRNA was significantly shortened after si-HuR transfection (P < 0.05); [Figure 3d]. Biotin pull-down assay demonstrated that HuR bound with the 3’ UTR of the Cav-1 mRNA but not with the 5’ UTR or coding region [Figure 3e]. By performing an RIP assay, we found that the HuR antibody pulled down more Cav-1 mRNA than the control IgG antibody (P < 0.05); [Figure 3f].

Close

Figure 3:

Human antigen R (HuR) directly bound to the caveolin 1 (Cav-1) messenger ribonucleic acid (mRNA) and stabilized it. (a-c) The levels of HuR and Cav-1 protein and mRNA levels after transfection small interfering RNA HuR (si-HuR), HuR overexpression plasmid (OEHuR), or corresponding negative controls in human umbilical vein endothelial cells (HUVECs) (n = 3) were measured through Western blotting and reverse transcription quantitative polymerase chain reaction (RT-qPCR). (d) The levels of Cav-1 in si-HuR- or negative control small interfering RNA -transfected HUVECs were measured through RT-qPCR after actinomycin D treatment (n = 5). (e) Biotin pull-down assay detection of the binding between HuR and the 3' untranslated region (UTR), coding region, and 5’ UTR of the Cav-1 mRNA (n = 3). (f) RNA immunoprecipitation experiments detection of the binding of HuR to Cav-1 in HUVECs treated with anti-HuR and anti-immunoglobulin G (negative control; n = 5). UTR: Untranslated region, si-NC: small interfering RNA negative control, OE-NC: Overexpression plasmid negiative control, ns: no significant. ^nsP > 0.05, *P < 0.05.

Export to PPT

Cav-1 triggered P. gingivalis-induced inflammatory responses, oxidative stress, and cell apoptosis of HUVECs by binding to HuR

We further confirmed that Cav-1 activated by HuR affected HUVEC function. The co-overexpression of HuR suppressed the inhibitory effects of si-Cav-1 on inflammatory responses and oxidative stress, increasing IL-1β, IL-6, and TNF-α levels and promoting ROS production (P < 0.05); [Figure 4a-e]. Simultaneously, Cav-1 inhibition significantly reduced apoptosis; however, the co-overexpression of HuR and siCav-1 increased apoptosis (P < 0.05); [Figure 4f and g]. In addition, Cav-1 inhibition substantially diminished the protein levels of Cav-1 and ICAM-1, while HuR overexpression enhanced the decrease of Cav-1 and ICAM-1 expression induced by Cav-1 inhibition in HUVEC subjected to P. gingivalis (P < 0.05); [Figure 4h].

Close

Figure 4:

Caveolin 1 (Cav-1) triggered Porphyromonas gingivalis (P. gingivalis)-induced inflammatory responses, oxidative stress, and cell apoptosis of human umbilical vein endothelial cells (HUVECs) by binding to human antigen R (HuR). HUVECs were transfected with small interfering ribonucleic acid HuR (si-HuR) alone or together with overexpression HuR (OE-HuR) plasmid and then subjected to P. gingivalis. (a-e) Enzyme-linked immunosorbent assay kits and 2',7'-dichlorodihydrofluorescein diacetate fluorescence probe were used in measuring the levels of interleukin (IL)-1β, IL-6, tumor necrosis factor-α, and reactive oxygen species in the supernatant of the HUVECs; Scale bar = 100 μm; (n = 5). (f and g) Flow cytometry was used in detecting changes in apoptosis level in the HUVECs (n = 5). (h) Changes in Cav-1, HuR, and intercellular cell adhesion molecule-1 (ICAM-1) levels were detected through Western blotting (n = 3). si-NC: small interfering RNA negative control, OE-NC: Overexpression plasmid negiative control, V-FITC: V-Fluorescein Isothiocyanate. *P < 0.05.

Export to PPT

Cav-1 was involved in P. gingivalis-induced atherosclerotic plaque formation

To investigate the effect of P. gingivalis on AS and the role of Cav-1, we infected the ApoE^−/− mice with P. gingivalis and injected them with si-Cav-1 through adenovirus infection. We then analyzed plaque formation in the outflow tract of the mouse aorta after P. gingivalis infection through Oil Red O staining. The results showed that after P. gingivalis infection, plaque formed in the inner wall of the outflow tract of the aorta, the intimal thickening was evident, and a large amount of lipid was deposited in the plaque. By contrast, P. gingivalis-accelerated plaque formation diminished after Cav-1 inhibition (P < 0.05); [Figure 5a and b]. Western blotting showed that ApoE^−/− mice infected with P. gingivalis had increased Cav-1 levels, whereas the ApoE^−/− mice treated with si-Cav-1 had significantly decreased Cav-1 levels (P < 0.05); [Figure 5c] Moreover, P. gingivalis infection with ApoE^−/− mice substantially aggravated the IL-6, IL-1β, and TNF-α as well as ROS production, which could be restore after si-Cav-1 treatment (P < 0.05); [Figure 5d-h].

Close

Figure 5:

Caveolin 1 (Cav-1) was involved in Porphyromonas gingivalis-induced atherosclerotic plaque formation. ApoE^−/− mice were injected with si-Cav-1 and then infected with P. gingivalis. (a) Oil Red O-stained thoracoabdominal aortas of the ApoE^−/− mice (n = 5). The white arrow showed areas of atherosclerotic lesions on the surface of the aorta. (b) Oil Red O-stained mouse aortic roots of the ApoE^−/− mice (n = 5). Scale bar = 500 μm. Magnification is 20x. The black arrow shows areas of atherosclerotic lesions in the aortic root. (c) Western blotting detection of Cav-1 in mouse aortic tissues (n = 5). (d-h) Enzyme-linked immunosorbent assay and dihydroethidium fluorescence probe detection of the levels of proinflammatory cytokines and reactive oxygen species in the supernatant of the mouse aortic tissues (n = 5). Scale bar = 200 μm. Magnification is 50x. *P < 0.05.

Export to PPT