Effects of the SEMA4B gene on hexavalent chromium [Cr(VI)]-induced malignant transformation of human bronchial epithelial cells

Abstract

Our previous study identified the potential of SEMA4B methylation level as a biomarker for hexavalent chromium [Cr(VI)] exposure. This study aimed to investigate the role of the SEMA4B gene in Cr(VI)-mediated malignant transformation of human bronchial epithelial (BEAS-2B) cells. In our population survey of workers, the geometric mean [95% confidence intervals (CIs)] of Cr in blood was 3.80 (0.42, 26.56) μg/L. Following treatment with various doses of Cr(VI), it was found that 0.5 μM had negligible effects on the cell viability of BEAS-2B cells. The expression of SEMA4B was observed to decrease in BEAS-2B cells after 7 days of treatment with 0.5 μM Cr(VI), and this downregulation continued with increasing passages of Cr(VI) treatment. Chronic exposure to 0.5 μM Cr(VI) enhanced the anchorage-independent growth ability of BEAS-2B cells. Furthermore, the use of a methylation inhibitor suppressed the Cr(VI)-mediated anchorage-independent growth in BEAS-2B cells. Considering that Cr levels exceeding 0.5 μM can be found in human blood due to occupational exposure, the results suggested a potential carcinogenic risk associated with occupational Cr(VI) exposure through the promotion of malignant transformation. The in vitro study further demonstrated that Cr(VI) exposure might inhibit the expression of the SEMA4B gene to promote the malignant transformation of BEAS-2B cells.

Graphical Abstract

Graphical Abstract.

Introduction

Hexavalent chromium [Cr(VI)] is extensively utilized in diverse industries and poses potential health risks through routes of exposure, such as ingestion, dermal contact, or inhalation of particulate matter containing Cr(VI). Occupational exposure predominantly occurs in industries like electroplating, leather processing, printing, dyeing, welding, and the production and application of Cr-containing products. Prolonged exposure to Cr(VI) has been linked to the development of tumors in various organs, along with health concerns such as skin irritation and asthma.^1,2 Elevated tumor development risk has been identified in association with Cr(VI) exposure through epidemiological studies and animal experiments, resulting in its classification as a human carcinogen.³ The inhalation of Cr fumes and dust stands as the primary occupational exposure pathway, heightening the susceptibility of workers to develop lung cancer. Previous epidemiological research has indicated a substantial 2- to 80-fold increased risk of lung cancer in individuals exposed to Cr(VI) within their respective workplace environments.⁴ Therefore, there is an urgent need to further elucidate the molecular mechanisms underlying Cr(VI)-induced cancers and identify suitable targets for early detection.

Multiple mechanisms have been implicated in the carcinogenicity of Cr(VI), encompassing signal transduction dysregulation, genomic instability, aberrant epigenetic alterations, and perturbed gene expression.^5,6 Under normal physiological conditions, a non-specific phosphate/sulfate anion transport protein facilitates the passage of Cr(VI) across the cell membrane. Once inside the cell, Cr(VI) undergoes the transformation into insoluble Cr(III), which then interacts with DNA and forms DNA-Cr adducts, leading to DNA damage and genomic instability.⁷ Cr(VI) exposure has also been linked to abnormal alterations in epigenetic patterns, encompassing histone modifications, non-coding RNA (such as microRNA and lncRNA), and DNA methylation. It has been demonstrated that Cr(VI) induces histone modifications that impede the recruitment of RNA polymerase II, leading to suppressed expression of specific genes.⁸ Furthermore, individuals occupationally exposed to Cr(VI) have shown a positive correlation between blood Cr concentrations and the methylation levels of targeted DNA repair genes. Notably, in both 16HBE cells and occupationally exposed individuals, there was a negative association between methylation levels at CpG sites and the corresponding mRNA levels of DNA repair genes.⁹

In our previous studies, our research group has focused on examining the effects of Cr(VI) on DNA methylation.¹⁰ Specifically, in a previous study conducted by our group, we utilized high-throughput methylation microarrays to analyze DNA methylation patterns in an occupational population exposed to Cr(VI). Furthermore, we evaluated the aberrant methylation in cell lines treated with Cr(VI).¹¹ The findings from this study indicated the presence of hypermethylation in the SEMA4B gene and a subsequent reduction in SEMA4B gene mRNA expression levels among the occupationally exposed population to Cr(VI). As a result, the aim of this current study is to further explore the impact of the SEMA4B gene in relation to Cr(VI) exposure.

A significant body of evidence from various studies consistently supports the notion that the majority of lung cancers originate from epithelial cells, particularly those of the bronchial variety.¹² Consequently, the development and progression of lung cancer are tightly intertwined with the malignant transformation of bronchial epithelial cells. Specifically, human bronchial epithelial cells that have been transfected with the SV40 virus and exhibit a sub-diploid karyotype are commonly denoted as BEAS-2B cells. These cells exhibit the ability to develop indefinitely and can be easily cultivated and passaged in vitro. In accordance with the selection criteria for cell transformation studies, BEAS-2B cells do not possess the capacity for spontaneous transformation and have not yet displayed features of malignant transformation.¹³ Interestingly, previous investigations have demonstrated that BEAS-2B cells can undergo malignant transformation when exposed to certain toxic and hazardous substances.^14,15 In the present study, we treated BEAS-2B cells with 0.5 μM of Cr(VI) to establish a cell model of Cr(VI)-induced malignant transformation. Subsequently, we analyzed the impact of altering the SEMA4B gene on the process of Cr(VI)-induced cell malignant transformation.

Materials and methods

Occupational epidemiological study

A cross-sectional study was conducted on a group of workers employed in an electroplating factory situated in the Zhejiang province, China. The survey encompassed 71 workers who had been consistently exposed to Cr(VI) for a median duration of 2 years. The research protocol received approval from the Ethics Committee of Hangzhou Medical College (LL2022-08), and written informed consent was obtained from each participant. Blood samples were collected, and the concentration of Cr in the blood was quantified using Inductively Coupled Plasma Mass Spectrometry (ICP-MS, Nexion300D, PerkinElmer, USA), employing a detection method previously described by our papers.^16,17 Briefly, whole blood samples were equilibrated at room temperature for 20 min and gently mixed. Then, 200 μL blood was diluted 20-fold with a 5% tetramethylammonium hydroxide solution (Sigma, USA), followed by a 10-min incubation at standard room temperature. Throughout the testing, whole blood reference standards (Seronorm, Norway) were concurrently processed with test samples. The limit of detection (LOD) for Cr in blood samples was 0.02 μg/L. During the course, repeated measurements were performed for each specimen to ensure precision. The ICP-MS in our lab is not only nationally accredited but also subject to stringent testing and regular maintenance.

Cell culture and conditions

The BEAS-2B cell line was acquired from the China Center for Type Culture Collection (CCTCC, Beijing, China). These cells were cultured in DMEM medium (Gibco, USA), which was supplemented with 10% fetal bovine serum (FBS) obtained from Bovogen Biological (Australia). Maintained at a temperature of 37 °C with a humidified atmosphere containing 5% CO₂, the cell culture incubator (SANYO, Japan) fostered optimal growth conditions. All cell experiments were performed in triplicate to ensure the reliability of the results.

Cytotoxicity assay

Potassium dichromate (Sigma, USA) was utilized as the source of Cr(VI) exposure. BEAS-2B cells in the logarithmic growth phase were detached using 0.25% trypsin (Gibco, USA) and subsequently seeded in 96-well plates with a density of 5,000 cells per well. To observe the effects of different concentrations of Cr(VI) on cell viability, BEAS-2B cells were treated with various concentrations of Cr(VI) (0, 0.1, 0.3, 0.5, 0.7, and 0.9 μM). The cytotoxic impact of Cr(VI) on BEAS-2B cells was evaluated using the Cell Counting Kit-8 test (CCK-8, Beyotime) following the manufacturer’s instructions.

Short-term and long-term treatments

Based on the results of the cytotoxicity assay, a concentration of 0.5 μM Cr(VI) was selected for both short-term and long-term experiments. In the Cr(VI) treatment group and the corresponding control group, a consistent interval of 3 days was maintained between consecutive cell passages. The passaging process was executed when the cell confluence reached approximately 80%, ensuring that the cells were at an optimal density for proliferation. Subsequent to each passage, once the cells had fully adhered to the new culture surface and formed a stable monolayer, Cr(VI) was introduced into the treatment group cultures as per the experimental protocol. After a 7-day exposure to Cr(VI), the expression levels of the relevant genes were examined. The BEAS-2B cells were continuously exposed to Cr(VI) for long-term culture, and the degree of malignant transformation was determined using the anchorage-independent growth ability assessed through a colony formation assay. At passages 4, 14, 24, 34, and 44 of the BEAS-2B cells exposed to Cr(VI), samples were collected, and the expression levels of the relevant genes were assessed.

Real-time quantitative PCR

In both short-term and long-term Cr(VI) treatments, total RNA was isolated from BEAS-2B cells. The cDNA was reverse transcribed from the extracted RNA using the Takara PrimeScriptRT kit (Takara, Japan) following the manufacturer’s protocol. The cDNA stock solution obtained from reverse transcription was used for real-time quantitative PCR amplification. The gene expression levels were normalized by GAPDH. The primers of each target gene are shown in Table 1.

Table 1.

Primers for PCR amplification of target genes.

Gene	Forward primer(5′-3′)	Reverse primer(5′-3′)
GAPDH	ACGGATTTGGTCGATTGGG	CTCGCTCCTGGAAGATGGTG
SEMA4B	GTAGTTGTTGCTGCCGTCGTC	CGCGCTGTCCTCTCGTGT
P21	ACTGTGATGCGCTAATGGC	ATGGTCTTCCTCTGCTGTCC

Colony formation assay

The assessment of anchorage-independent growth capability was conducted by employing a colony formation assay, following an established experimental protocol.¹⁸ BEAS-2B cells were seeded at a density of 3,000 cells per well in 1% methylcellulose-supplemented DMEM containing 10% FBS, in the absence of Cr(VI) supplementation. Subsequently, the cells were incubated for a period of two weeks, and all colonies (≥50 μm) were counted. The results were counted relative to the control. Representative photographs were taken under an inverted fluorescence microscope. Upon observing significantly higher colony formation in 44th-generation cells compared to controls, a colony formation assay was performed on these cells using the methylation inhibitor 5-Aza-dC (Sigma, USA), which is both a SEMA4B inhibitor and has broad-spectrum methylation inhibitory properties. Given the absence of any specific methyltransferase inhibitors tailored for SEMA4B, a treatment protocol was adopted following a previous study,¹⁹ administering 0.5 μM of 5-Aza-dC to the cells under investigation.

Wound healing assay

Cell migration ability was measured by wound healing assay. 4 × 10⁵ cells per well were seeded into 6-well plates and allowed to grow until confluent. Subsequently, a sterile micropipette tip was employed to create a uniform scratch across the monolayer, following which, the detached cells were gently washed away using phosphate-buffered saline (PBS). The culture medium was then replaced with serum-free medium for incubation. Photographic documentation of the wound healing process was captured at 0, 24 h, and 48 h post-scratch. The scratch migration distance was calculated as: migration distance = (D0 - Dn)/D0, where D0 represents the initial width of the scratch and Dn denotes the residual scratch width at the designated time point for measurement.

Statistical analysis

SPSS 23.0 software (IBM Corp., Armonk) was applied for statistical analysis. The comparison of means between two groups was conducted using Student’s t-test. For multiple group comparisons, one-way ANOVA with the Bonferroni test was employed. Statistical significance was defined as a p-value below 0.05.

Results

Levels of Cr in the occupationally exposed population

Table 2 presents the measured blood chromium (Cr) levels in occupationally exposed populations from various countries and regions. It can be observed that the blood Cr concentrations in our study participants, who were exposed to Cr(VI), were found to be comparable to the average Cr levels observed in other countries or regions.

Table 2.

Blood Cr levels in the populations occupationally exposed to Cr(VI) in different countries and regions.

Industry	Region	Years	N	Blood Cr, (μg/L)	Range	references
Electroplating	Eastern China	2023	71	Geometric mean (95%CI): 3.80 (0.42, 26.56)	0.42–29.07	This study
Chromate production and application	China	2022	515	Geometric mean: 2.38	0.48(P5)–14.40 (P95)	Zhang et al. 202242
Chromate production and application	Northern China	2021	455	Geometric mean: 6.42 (6.08, 6.79)	-	Hu et al. 202143
Electroplating	Eastern China	2019	162	Median: 6.37	0.04–58.92	Our previous study20
Chromate production	Shandong, China	2018	106	Mean (SD): 18.91(10.35)	5.95–48.1	Feng et al. 201844
Welding	Poland	2020	67	Median (IQR): 1.25 (0.12–2.68)	-	Stanislawska et al. 202045
Mixed	Brazil	2020	50	Mean(SD): 2.02 (0.20)	-	Muller et al. 202246
Electroplating	Egypt	2018	41	Median: 3.30	0.09–7.20	El Safty et al. 201847
Printing	Thailand	2015	75	Mean (SD): 1.24 (1.13)	0.1–4.21	Decharat 201548

Cytotoxicity of Cr(VI)

Figure 1 presents the results of the viability assessment conducted on BEAS-2B cells following exposure to Cr(VI), revealing a clear time-dependent and dose-dependent correlation. According to the findings, concentrations of Cr(VI) ranging from 0.1 μM to 0.5 μM did not exhibit a significant impact on the cell survival in comparison to the control group (P > 0.05). Consequently, a concentration of 0.5 μM was selected as the treatment dosage for subsequent experimental procedures.

Fig. 1.

Relative cell viability of BEAS-2B cells treated with Cr(VI). ^*P < 0.05; ^***P < 0.001.

Expression levels of the SEMA4B gene in the short-term exposure to Cr(VI)

The relative expression of SEMA4B mRNA in the treated cells was lower than that in the control cells after 7 days of Cr(VI) exposure (P < 0.05), as shown in Fig. 2. These results indicated that the expression level of SEMA4B in BEAS-2B cells could be reduced after short-term exposure to Cr(VI), which was in agreement with our previous results obtained in Cr(VI)-treated HMy2.CIR cells.¹¹

Fig. 2.

Effects of short-term Cr(VI) treatment on the SEMA4B expression levels in BEAS-2B cells. ^*P < 0.05.

Expression levels of SEMA4B and P21 genes in the long-term exposure to Cr(VI)

The impact of continuous exposure to low-dose Cr(VI) on gene expression levels in different passages of BEAS-2B cells was investigated using real-time quantitative PCR. Figure 3 presents the evaluation of SEMA4B and p21 gene expression levels in both the Cr(VI)-treated group and the control group at 10-generation intervals starting from the 4th generation. The results revealed a progressive decline in the expression levels of the SEMA4B gene as the number of treatment generations increased. Conversely, the relative mRNA expression level of the p21 gene in the Cr(VI)-treated group initially displayed an increase, followed by a subsequent decrease. In addition, the Cr(VI)-treated group had enhanced proliferation ability and grew faster than the control cells at passage 44 based on the CCK-8 assay, as shown in Fig. 4a. This enhanced proliferation potential may potentially contribute to the malignant transformation of the cells.

Fig. 3.

Effect of long-term Cr(VI) treatment on the expression levels of SEMA4B and P21 genes in BEAS-2B cells. ^*P < 0.05; ^**P < 0.01; ^***P < 0.001.

Fig. 4.

Long-term treatment with Cr(VI) induced malignant transformation of BEAS-2B cells. Cells at were trypsinized and seed for cell proliferation assay (44th-passage cells) (a), for colony formation assay (24th-, 34th-, and 44th-passage cells) (b and c), and for wound healing assay (44th-passage cells) (c and d). Quantitative results were shown using bar graph from four independent experiments. ^*P < 0.05; ^**P < 0.01; ^***P < 0.001.

Long-term Cr(VI) treatment induced malignant transformation of BEAS-2B cells

The colony formation rate and cell migration ability of BEAS-2B cells subjected to chronic Cr(VI) treatment exhibited a significant increase compared to the control group, as depicted in Fig. 4. There was no difference in the ability of colony formation between the 24 and 34 passages of continuous exposure to Cr(VI) compared to the control group (Fig. 4b). The results indicated that the 44th-passage cells in the 0.5 μM Cr(VI) long-term treatment increased the number of colonies formed (Fig. 4b and c) and enhanced cell migration by wound healing assays (Fig. 4d and e), suggesting that Cr(VI) treatment promoted the malignant transformation of BEAS-2B cells.

Effects of methylation inhibitor of Cr(VI)-induced anchorage-independent growth

The impact of methylation inhibitor treatment on the colonies of the Cr(VI) long-term treated group was examined and depicted in Fig. 5. Interestingly, a significant decrease in the colonies was observed following the application of the methylation inhibitor (P < 0.001). These results imply that the treatment with the methylation inhibitor may effectively reduce the methylation of cells in the Cr(VI) long-term treated group, consequently influencing the process of Cr(VI)-induced malignant transformation.

Fig. 5.

Effect of the methylation inhibitor on Cr(VI)-induced malignant transformation of BEAS-2B cells. Standardized colony formation rates (a) and representative pictures (b) of colonies are presented. ^**P < 0.01; ^***P < 0.001.

Discussion

In this study, BEAS-2B cells were chronically exposed to 0.5 μM Cr(VI), and this prolonged exposure led to an increased capacity for colony formation and cell migration, indicating successful establishment of a cell malignant transformation model. The concentration of 0.5 μM (approximately 26 μg/L) used in this study was similar to the reported blood Cr concentration in populations exposed to Cr(VI).²⁰ These results suggested that the Cr level in blood could reach 0.5 μM in humans, implying that our in vitro experimental findings may have relevance to actual Cr(VI) exposure situations in humans, especially in occupational settings. It was also observed that long-term exposure to Cr(VI) decreased the expression of the SEMA4B gene, indicating its potential role in the malignant transformation of BEAS-2B cells caused by Cr(VI).

SEMAs are a family of proteins involved in various intercellular communications, including cell migration, immune response, and other physiological functions. Several studies have demonstrated the involvement of signaling proteins, including SEMAs, in regulating various cancer-related features such as cell proliferation, metastasis, and invasion.^21,22 Many SEMA4 family members have been reported to be associated with tumorigenesis and progression. SEMA4C has been shown to attract tumor-associated macrophages and promote tumor growth,²³ SEMA4D has been linked to immune cell dysfunction in non-small cell lung cancer (NSCLC),²⁴ and SEMA4B has been upregulated in prometastatic B cells in renal cell carcinoma.²⁵ Previous studies have suggested the involvement of SEMA4B in the growth and metastasis of NSCLC,^26,27 as well as its potential as a biomarker for lymph node metastasis and poor prognosis in lung adenocarcinoma.^28,29 However, the role of SEMA4B in lung cancer remains unclear.³⁰ A recent study established that elevated SEMA4B expression in lung adenocarcinoma tumor tissues was associated with an unfavorable prognosis, and notably, its suppression diminished the proliferative potential of cancer cells.³¹ Previous studies reported that overexpression of SEMA4B inhibited proliferation and metastasis of A549 cells, suggesting a potential dual role of SEMA4B in tumor inhibition and promotion.³²

The PI3K signaling cascade is a downstream effector of the SEMAs. In NSCLC, SEMA4B has been shown to suppress matrix metalloproteinase 9 (MMP9) expression by inhibiting the PI3K pathway, which in turn effectively restrains NSCLC invasion and retards tumor progression.²⁶ The same research group further substantiated that SEMA4B exerted an antiproliferative effect on NSCLC cells, again through interference with the PI3K pathway.²⁷ The PI3K pathway serves as a principal upstream regulator of FoxO, whereupon phosphorylated FoxO translocates from the nucleus to the cytoplasm, thereby impairing its ability to express nuclear-bound genes, including the cell cycle inhibitor p21. In the current study, chronic exposure to Cr(VI) led to a significant decline in SEMA4B expression within a BEAS-2B cell malignant transformation model, coinciding with accelerated cellular proliferation. P21 is known to mediate the p53-induced G1 phase cell cycle arrest.³³ A study scrutinized the activity and expression of p21 in BEAS-2B cells acutely exposed to arsenic and Cr(VI), which indicated that acute exposure of BEAS-2B cells to either arsenic or Cr(VI) led to p21 activation and increased expression. However, p21 expression was suppressed in arsenic-transformed BEAS-2B cells.³⁴ This aligns with our results where short-term Cr(VI) exposure caused a decline in p21 levels, whereas long-term Cr(VI) treatment resulted in elevated p21 expression during malignant transformation of cells. Therefore, it is plausible that the SEMA4B gene may modulate changes in cell proliferation by influencing the normal expression levels of the p21.

DNA methylation alterations due to Cr(VI) exposure have been extensively studied.¹⁴ Reduced expression of p16INK4a and abnormal upregulation of p16INK4a promoter methylation in lung cancer tissues of workers with long-term (>15 years) Cr(VI) exposure suggested that p16INK4a promoter hypermethylation was involved in the carcinogenic effects of Cr(VI).³⁵ Previous studies indicated that DNA hypermethylation caused by Cr(VI) exposure might hinder DNA repair systems, leading to the accumulation of genetic damage and ultimately carcinogenesis.³⁶ Another study found hypermethylation in the hedgehog-interacting protein (HHIP) promoter region in Cr(VI)-transformed human bronchial epithelial cells, and elevated CXCL5 expression levels were associated with DNA hypomethylation and histone modifications induced by Cr(VI) exposure.³⁷ Thus, Cr(VI) exposure may participate in the malignant transformation process by altering gene methylation statuses.

A previous study by our group found that elevated methylation and altered expression levels of the SEMA4B gene in the peripheral blood of Cr(VI) exposed workers.¹¹ In this previous study, short-term treatment of HMy2.CIR cells with 5 μM Cr(VI) revealed a decrease in the expression of SEMA4B, which is consistent with the results of short-term treatment of BEAS-2B with Cr(VI) in our study. We further explored the results of long-term exposure from the actual exposure of the population, conducted in vitro experiments of long-term treatment with Cr(VI). In addition, few people have previously studied the role of SEMA4B gene in the malignant transformation of cells, and no studies have explored the changes of SEMA4B and p21 genes in each generation during long-term treatment with Cr(VI). Therefore, our study provided a new target for the study of the malignant transformation of Cr(VI). Given that workers could be exposed to these doses of Cr(VI) in blood according to our previous occupational survey,²⁰ the development of preventive systems should be considered in the future.

Further attempts were made to carry out methylation intervention experiments. 5-Aza-dC, a potent DNA methyltransferase (DNMT) inhibitor, leading to DNA demethylation and ultimately gene reactivation processes.³⁸ It has consistently been demonstrated that 5-Aza-dC efficiently promotes genomic DNA demethylation within cells, with numerous research studies employing it to reverse methylation patterns of specific genes. The application of 5-Aza-dC resulted in activation of the p53-p21 and p16-Rb pathways through DNA demethylation in immortalized fish cell lines,³⁹ significantly mitigated Mn-induced p53 methylation in BV2 microglia,⁴⁰ and inhibited CXCL14 methylation while augmenting its expression in KYSE150 and KYSE510 cell lines.⁴¹ This study followed suit by treating BEAS-2B cells with 5-Aza-dC. It was found that methylation inhibitor treatment suppressed the Cr(VI)-mediated increase in the anchorage-independent growth capacity of BEAS-2B cells.

There are some advantages to our research. This study grounded our investigation in real-world population exposure scenarios, and the role of SEMA4B in malignant transformation has not been previously explored. Scarce research exists on how gene expression changes over multiple generations during chronic Cr(VI) treatment. Furthermore, this study has certain limitations due to the absence of direct evidence concerning protein expression change. This study centered on elucidating critical molecular events through mRNA expression profiling. Nonetheless, mRNA levels alone may not provide a comprehensive understanding of gene function regulation, especially when considering post-translational modifications and actual protein activity. Indeed, knockout and overexpression experiments constitute essential components for in-depth research.

Conclusions

In summary, our findings demonstrated that treatment with 0.5 μM of Cr(VI), a concentration commonly found in human blood, promoted the malignant transformation of BEAS-2B cells. We further observed a decrease in the expression of the SEMA4B gene during the process of Cr(VI)-induced malignant transformation in vitro.

Author contributions

Conceptualization, Yao Qin, Huadong Xu, Yongyong Xi, Lingfang Feng; methodology, Yao Qin, Huadong Xu; validation, Yao Qin; formal analysis, Yao Qin, Huadong Xu, Zhaoqiang Jiang; investigation, Lingfang Feng; resources, Junfei Chen, Biao Xu, Xiaowen Dong, Yongxin Li; data curation, Huadong Xu, Yongyong Xi, Lingfang Feng, Zhaoqiang Jiang; writing—original draft preparation, Yao Qin; writing—review and editing, Huadong Xu, Yao Qin; supervision, Jianlin Lou; project administration, Jianlin Lou; funding acquisition, Jianlin Lou.

Funding

This study was funded by the National Natural Science Foundation of China (82273609, 82003426, and 81872602), Zhejiang Provincial Natural Science Foundation of China (LY21H260002), and Zhejiang Medical Health Science and Technology Foundation (2019RC145).

Conflict of interest statement: The authors declare no conflict of interest.

Data availability

The data are available from the corresponding author upon request.