Senescent epithelial cells remodel the microenvironment for the progression of oral submucous fibrosis through secreting TGF-β1

Patient samples

A total of 30 samples were collected from the center of stomatology, Xiangya Hospital from September 2019 to May 2020. 21 OSF tissues were obtained from OSF patients without OSF therapy. Nine healthy oral mucosa (NOM) samples were obtained from healthy volunteers during surgical removal of the lower third molars. All samples were diagnosed by pathological examination based on the 2005 World Health Organization classification system. All research subjects gave written informed consent in accordance with the Helsinki Declaration. This study was approved by the Medical Ethics Committee of Xiangya Hospital, Central South University (Ethical Application Ref: 201910837).

Immunohistochemistry assay

The immunohistochemistry was performed as described previously (Min et al., 2020). Tissues were fixed with 4% paraformaldehyde for 24 h, and then were dehydrated in graded alcohol solutions and embedded in paraffin. The paraffin-embedded tissues were sliced into 5-µm sections. The following primary antibodies were used in immunohistochemistry: p16 (1:500, ab108349; Abcam, Cambridge, UK), p21 (1:50, 2947T; Cell Signaling Technology, Danvers, MA, USA), alpha-smooth actin (α-SMA) (1:200, ab5694; Abcam), proliferating cell nuclear antigen (PCNA) (1:16000, 2586S; Cell Signaling Technology). Phosphate-buffered saline (PBS) was used instead of the primary antibody as a negative control, and NOM tissues were used as a positive control. All slides were observed by microscopy (Leica Microsystems, Cambridge, UK). The staining results were quantitatively evaluated using the Image-Pro-Plus 5.0 software to measure the mean optical density (MOD) of positive p16, p21, PCNA, and α-SMA in OSF tissues (MOD = integral optical density (IOD)/area of interest) (Hu et al., 2021). The epithelium or subepithelial areas of NOM and OSF tissues were marked as the interest areas for analysis.

Sudan black B staining

To detect lipofuscin accumulation in OSF epithelium, Sudan black B solution was used to stain the lipofuscin. Sudan black B staining (S109070; Aladdin) was carried out as described previously (Georgakopoulou et al., 2013).

Cell culture

Human Oral Keratinocytes (HOKs) is currently the only normal oral epithelial cell line available, which were cultured in complete Alpha MEM (10% FBS and 1% penicillin-streptomycin added) at 37°, 5% CO₂ incubator (Wang et al., 2018). As the major components of betel nut, we used arecoline to induce HOKs senescence. The half-maximal inhibitory concentrations (IC50) of arecoline in HOKs were 200 µg/ml within 24 h and 160 µg/ml within 48 h by CCK8 assay, and then arecoline concentrations were grouped accordingly. HOKs were randomly divided into six groups, and after starvation in serum-free medium overnight, they were incubated with complete medium containing different concentrations of arecoline (0, 20, 40, 80, 160, and 320 µg/ml) for 24 h. Then, all groups were replaced with complete medium, and the medium was changed every 3 days for 10–14 days.

Senescence-associated β galactosidase (SA-β-gal) activity assay

The SA-β-gal activity was performed using a SA-β-gal staining Kit (CS0030; Sigma-Aldrich, St. Louis, MO, USA) following the manufacturer’s instructions. Briefly, after senescence induction, HOKs were seeded into 24-well culture plates at a density of 1 ×10⁴ cells per well. The next day, the cells were washed three times with PBS and fixed by cell fixative solution in the SA-β-gal staining Kit for 7 min at room temperature. Then, the cells were washed three times with PBS to remove the solution. The staining mixture containing X-gal prepared following the instruction was incubated with the cells in a regular incubator at 37° overnight without CO₂. The cells were observed by microscopy (Leica Microsystems, Cambridge, UK). A total of 5 fields in each group were selected under the microscope and the total cells and SA-β-gal positive cells in each field were counted by ImageJ 5.0. The proportion of SA-β-gal positive cells was obtained by calculating the ratio of the average SA-β-gal positive cell number to the average total cell number in each group.

Immunofluorescence assay

Immunofluorescence was applied to examine the expression of phosphor-histone H2A.X (γH2A.X) in HOKs treated with different concentrations of arecoline. After senescence induction, the cells were seeded into 24-wells culture plates at 1 ×10⁴ per well density. The next day, the cells were fixed with 4% paraformaldehyde for 10 min. Subsequently, the cells were permeabilized with 0.1% Triton X-100 solution and blocked with 10% goat serum (ZLI-9056, ZSGB-BIO). After PBS washing, the cells were incubated with monoclonal rabbit anti- γH2A.X primary antibody (1:400, 9718, Cell Signaling Technology) at 4° overnight. The next day, the cells were incubated with cy3-conjugated secondary antibody (AP132C, Sigma) for 1 h in dark, and then counterstained with 4′,6-Diamidino-2-phenylindole (C1002, Beyotime) at room temperature for 1 min. Immunofluorescence images were acquired using a fluorescence microscope (Leica Microsystems, Cambridge, UK). A total of five fields in each group were selected under the microscope and the total cells and γH2A.X positive cells in each field were counted by ImageJ 5.0. The proportion of γH2A.X positive cells was obtained by calculating the ratio of the average γH2A.X positive cell number to the average total cell number in each group.

Cell proliferation assay

Cell proliferation ability was assessed by Cell Counting Kit 8 (CCK8) (A311-01; Vazyme). CCK8 assay was performed by the manufacturer’s protocol. After senescence induction, HOKs were seeded into 96-well plates at a density of 5000 cells per well and continually cultured for 0, 24, 48, 72, and 96 h. Then, CCK8 solution (10 µl) was added to each well for 2 h incubation at 37°. The plates were detected at the indicated time (0, 24, 48, 72, and 96 h) using a microplate reader (Synergy 2; Biotek, Winooski, VT, USA) at a wavelength of 450 nm.

Quantitative reverse transcription polymerase chain reaction (qRT-PCR) assay

Total RNA was extracted from HOKs treated with different concentrations of arecoline using TRIzol reagent (15596026; Invitrogen, Waltham, MA, USA). The total RNA was reverse transcribed to cDNA using RevertAid™ Master Mix, with DNase I Kit (M16325; Thermo Scientific, Waltham, MA, USA) according to the manufacturer’s protocol. qPCR was performed using Maxima SYBR Green qPCR Master Mix (K0252; Thermo Scientific, Waltham, MA, USA) on a Real-Time PCR Detection System (CFX96; Bio-Rad, Hercules, CA, USA) according to the manufacturer’s protocol. All qPCR primers were listed in Table 1. The relative mRNA expression was calculated using the 2^−ΔΔCt method. GAPDH served as housekeeping gene. The value of negative control group was used to calculate the relative fold change in expression value.

Western blotting assay

The total protein was extracted from HOKs treated with different concentrations of arecoline using SDS lysis buffer (P0013G; Beyotime). The total protein concentration was detected using Pierce™ BCA Protein Assay Kit (23225; Thermo Fisher Scientific). A total of 40 µg protein was added to 10% polyacrylamide gel and transferred to polyvinylidene fluoride (PVDF) membranes. The membranes were blocked with 5% skim milk for 1 h at room temperature. After PBST washing three times, the membranes were incubated with primary antibodies overnight at 4° overnight: p16 (1:2000, ab108349, Abcam), p21 (1:2000, 2946; Cell Signaling Technology), p53 (1:1000, 48818; Cell Signaling Technology), transforming growth factorβ1 (TGF-β1) (1:1000, ab215715; Abcam), and GAPDH (1:1000, 5174T, Cell Signaling Technology). Then, the membranes were incubated with the corresponding horseradish peroxidase-conjugated secondary antibodies (anti-Rabbit/mouse 1:5000∼10000, SA00001-1/2; Proteintech) for 1 h at room temperature. The membranes were developed using SuperSignal™ West Pico PLUS chemiluminescent substrate (34580; Thermo Fisher Scientific) and detected by a chemiluminescent imaging system (MiniChemi 610; Sage Creation, Beijing, China). and analyzed using the ImageJ 5.0 version software.

Enzyme-linked immunosorbent (ELISA) assay

Levels of TGF-β1 in supernatants were measured by the Human TGF-β1 Elisa Kit (PT880; Beyotime, Jiangsu, China) following the manufacturer’s protocol. The absorbance of samples was detected at 450 nm using a microplate reader (Synergy 2; Biotek, Winooski, VT, USA). The TGF-β1 concentration of each sample was determined based on a standard curve we prepared according to the manufacturer’s protocol.

Statistical analysis

All experiments were conducted in triplicate and repeated at least three times. All data were shown as mean ± standard deviation. Continuous variables were assessed by the student’s t test or Mann–Whitney U test according to the normality test. Spearman’s correlation was used to analyze the correlation. P value <0.05 was considered statistically significant. Statistical analysis was performed using SPSS 24.0 software.

The epithelium of OSF undergoes premature senescence

p16 and p21 belong to cyclin-dependent kinase inhibitors, which are major drivers of the cycle arrest in senescence and have been confirmed to be one of the defining features of senescent cells (Calcinotto et al., 2019). Therefore, we detected the expressions of p16 and p21 in the epithelium of OSF and healthy tissues. We found that p16 and p21 were overexpressed in OSF epithelium compared with the healthy epithelium (Figs. 1A–1D). In contrast, PCNA, which indicates cellular proliferation, was highly expressed in the epithelium of NOM (Figs. 1A, 1B, 1E). Moreover, the expressions of p16 and p21 were negatively correlated with PCNA in OSF epithelial tissues (Figs. 1H–1K).

To further confirm epithelial cells senescence in OSF, we detected the accumulation of lipofuscin in OSF epithelium by Sudan Black staining. Lipofuscin is abnormally accumulated in senescent cells, another well-established senescence marker (Georgakopoulou et al., 2013). After Sudan black B staining, we found that the lipofuscin was widely distributed in OSF epithelium (Fig. 1G).

Correlation between the epithelial cell senescence and myofibroblast activation in human OSF tissues

To further investigate the correlation between the epithelial cell senescence and myofibroblast activation in OSF tissue, we detected the expression of α-SMA that is a characteristic marker of myofibroblast activation. We observed that there were α-SMA positive spindle cells located in the lamina propria adjacent to the epithelium of OSF (Fig. 1B). In sharp contrast, we only detected rings of α-SMA positive cells representing blood vessels in the lamina propria of healthy tissues (Fig. 1A). We hardly found the α-SMA positive spindle cells in healthy tissues. Moreover, the expression of α-SMA in OSF tissues was extremely higher than that in healthy tissues (p <0.001, Fig. 1F). Further analysis revealed that α-SMA expression in OSF tissues was positively correlated with the expressions of senescence markers, p21 and p16 (p <0.01, Fig. 1H).

Arecoline induced senescence-like cell morphology and suppressed cell proliferation in HOKs

Since OSF is an oral mucosal disease associated with the chewing of betel nut, we choose arecoline to induce HOKs senescence due to it is a major component of betel nut. To confirm HOKs developing senescence, we examined senescence-associated phenotypes in HOKs.

After HOKs were treated with different concentrations of arecoline for 24 h, they gradually appeared an enlarged and flattened morphology, following another 5–7 days of culture (Fig. 2A). Meanwhile, the enlarged and flattened HOKs also showed SA-β-gal positive staining (Fig. 2A). The proportion of SA-β-gal positive HOKs was gradually increased with arecoline concentration. Among different concentrations of arecoline, the 320 µg/ml group had the highest proportion of SA-β-gal positive HOKs compared with other groups (p <0.05, Fig. 2B). Moreover, the surface area of SA-β-gal positive cells was about 10-fold more than that of SA-β-gal negative cells (p <0.05, Fig. 2C). CCK8 assay showed that compared with that of other groups, the proliferation ability of the 320 µg/ml group was largely suppressed after senescence induction (p <0.05, Fig. 2D).

Arecoline induced DNA damage response (DDR) in HOKs

The phosphorylated-histone H2A.X was detected in HOKs after senescence induction, a hallmark of DNA double-strand breakage in cells, because DNA damage stimulates DNA damage response (DDR) that triggers senescence (Di Micco et al., 2021; Gorgoulis et al., 2019). We found that γH2A.X was mainly detected in HOKs treated with high concentrations of arecoline (160 and 320 µg/ml) (Fig. 3A). The proportion of γH2A.X positive HOKs was elevated with arecoline concentration, and that in the 320 µg/ml group was significantly higher than in other groups (p <0.05, Fig. 3B). Moreover, we observed that γH2A.X was mainly expressed in the nucleus of the enlarged and flatted HOKs (Fig. 3C).

Arecoline-induced HOK senescence was associated with p53/p21 pathway

Cell cycle arrest of senescent cells is controlled by activation of the p53/p21 or p16/Rb pathway (Gorgoulis et al., 2019). Since most of HOKs in the 320 µg/ml group undergo senescence, we compared the levels of p16, p21, and p53 between the groups of the negative control (0 µg/ml) and the 320 µg/ml. We found that p21 expression was dramatically upregulated at the protein and mRNA level (p < 0.01, Figs. 4A, 4B), while p53 expression was on the rise only at the protein level in the 320 µg/ml group (p <0.05, Figs. 4A, 4C). No difference was observed in p16 at protein and mRNA levels in both groups (Fig. 4A).

The synthesis and secretion of TGF-β1 are increased in senescent HOKs

Growth factors are a prominent constituent of SASP. Given that TGF-β1 is a well-known profibrotic factor, we examined the protein and mRNA levels of TGF-β1 between the groups of the negative control (0 µg/ml) and the 320 µg/ml. The protein and mRNA levels of TGF-β1 in the 320 µg/ml group were all significantly upregulated, compared with the negative control group (p <0.05, Fig. 4D). In addition, although the density of HOKs in the 320 µg/ml group after senescence induction was approximately 30–40% of that in the negative control group, we found that the levels of TGF-β1 in the supernatant of the 320 µg/ml group were increased approximately 2-fold compared to the control group (p <0.01, Fig. 4E).