Introduction

The oral cavity is the sixth to ninth most frequently affected anatomical site by cancer, and it is estimated that there are ~275,000 additional cases of oral cancer globally each year, with squamous cell carcinoma (SCC) contributing to ~80–90% of all malignancies (Mirghani, Amen, Moreau, & Lacau St Guily, 2015). Oral squamous cell carcinoma (OSCC) originates from the oral mucosa (Mirghani et al. 2015) and is the ninth most common cancer globally, with 354,864 fresh cases detected and 177,384 mortalities reported in 2018 (Bray et al. 2018). The prognosis of OSCC is statistically poor, with a 50% five-year survival rate and a yearly treatment cost of US$2 billion (Botha et al. 2021). OSCC often implies an altered functional state of genes linked to various pathways, leading to a dysregulation of normal cell proliferation (Lindemann et al. 2018). Therefore, the development of novel biomarkers for OSCC could help in early cancer diagnosis, treatment and prognosis.

N6-methyladenosine (m6A) is the most prominent RNA modification in mammalian cells and another form of epigenetic regulation (Sun et al. 2019). The m6A methylation is carried out by the multicomponent RNA methyltransferase complex, RNA demethylases, and m6A readers, and this dynamic and reversible process regulates RNA translocation, localization, translation, and degradation (Zhang et al. 2021). It has been shown that m6A is engaged in all spheres of RNA metabolism, notably precursor mRNA splicing, 3'-end processing, nuclear export, translational regulation, and mRNA decay (Kumari et al. 2022; Roundtree et al. 2017). It has been documented that m6A modifications regulate the expression of target genes as proto-oncogenes or tumour suppressors widely involved in tumour progression, including breast (Wu et al. 2022), lung (J. Liu et al. 2018), ovarian (T. Liu et al. 2020) and liver (M. Chen et al. 2018) cancers. In the past few years, a few sporadic reports have consistently concluded that the m6A regulatory gene KIAA1429 was markedly enhanced in the tissues of head and neck squamous cell carcinoma (HNSCC), as well as significantly correlated with cancer stage, tumour grade, and lymph node metastasis (Arumugam et al. 2021; Paramasivam et al. 2021). However, its downstream mechanisms of action need to be further explored. Nonetheless, in OSCC, investigation is needed to understand the mechanisms by which KIAA1429 regulates the m6A methylation levels of specific transcripts.

Here, we sought to demonstrate the regulatory role of methyltransferase KIAA1429 on OSCC development and to explore the downstream action sites of KIAA1429. We speculated that KIAA1429 regulates the stable expression of carbonic anhydrase 9 (CA9) through m6A modification, which worsens OSCC progression. The outcomes of the present study broaden the horizon of OSCC development mechanisms and provide valuable targets for OSCC treatment.

Materials and methods

Tissue specimens

At our hospital, patients who had surgery were asked to donate 34 pairs of OSCC tissues and the adjoining healthy tissues. These samples were cryopreserved at – 80 °C for further use. All participants in our study signed informed consent before obtaining samples. The initiation of this research was approved by the ethic committee of our hospital (ethical approval code: HBZY2023-C66-05).

Cells culture

Human immortalized oral epithelial cell line (HIOEC) (cat. no. YS1814C) and OSCC cell lines, including HSC-2 (cat. no. YS2196C), SCC-25(cat. no. YS1748C), HSC-4 (cat. no. YS1792C) and SCC-9 (cat. no. YS336C) were all bought from YaJi Biological (China). The cell lines were cultured in an incubator at 37 °C containing 5% CO2 in DMEM (YaJi Biological) supplemented with 1% P/S (Zhong Qiao Xin Zhou Biotech., China) and 10% FBS (Zhong Qiao Xin Zhou Biotech.).

Cell transfection

Full length KIAA1428 (gene ID: 25,962) and CA9 (gene ID: 768) genes were inserted and cloned into a lentiviral vector expressing GFP and puromycin (RiboBio, China) to construct overexpression plasmids. The empty vector (vector) was used as the control. Small interfering RNA targeting KIAA129 (si-KIAA1429 (PMID: 31,118,692); F 5′-GAGGATGATCGACGAACAGTA-3′; R: R 5′-AAGGCTTATTAACCTCCTAGA-3′), and negative controls (si-NC) were also acquired from RiboBio (China). HSC-2 and SCC-9 cells (1 × 105) were plated into a 12-well plate and cultured for 12 h. Next, the transfection of vectors, or oligonucleotides was implemented by Lipo6000 (Beyotime, China). After 48 h, the transfection efficiency was evaluated by qRT-PCR.

qRT-PCR

The total RNA was extracted by utilizing the Trizol reagent (Generay Biotech., China). As directed by the procedure, cDNA was synthesized from an equal quantity of RNA by means of cDNA Synthesis Kit (Fermentas, Canada). Afterwards, qPCR was carried out using SYBR Green qPCR Mastermix (Qiangen, Germany). By means of the 2−ΔΔCt method, we determined the relative KIAA1429 and CA9 expressions via GAPDH as the internal control. The primers are tabulated in Table 1.

Table 1 Primer sequences used in this study
Full size table

CCK-8 assay

We used a 96-well plate to routinely culture HSC-2 and SCC-9 cells (5000 cells/well). At the specified time intervals (0-, 24-, 48-, 72-, and 96 h), CCK-8 reagent (Dojindo, Japan) was introduced into cell culture and subsequently incubated for 2 more hours at 37 ℃. For measuring optical density (at 450 nm) of these cells, a microplate reader (BIO-RAD Lab., Japan) was employed.

Transwell migration and invasion assays

HSC-2 and SCC-9 cells (2 × 105 cells/well) were resuspended in serum-free DMEM media (500 µL) and loaded onto the Matrigel-pre-coated (for invasion assay)/ uncoated (for migration assay) upper chambers. Meanwhile, the lower chambers were filled with DMEM media (750 µL) having 10% FBS. After culturing for 24 h, the fixation of invaded cells was done with methanol for 15 min before they were stained with 0.5% crystal violet. The light microscope (Olympus, Tokyo, Japan) was then employed to photograph three randomly chosen fields.

Xenograft assay

Ten ♂ BALB/c nude mice (age: ~6 weeks, weight range: 18 g to 20 g) were acquired from Hunan SJA Laboratory Animal (China) and kept in an environment free of any pathogens. This animal experiment was approved by the ethics committee at our hospital (ethical approval code: No.20230302). To establish animal models, HSC-2 cells along with sh-KIAA1429 or sh-NC lentiviral vector obtained from GenePharma (USA) were microinjected into the left axilla of mice (n = 5 per group). The shRNA target sequences were as follows: sense: 5′—AGUAUCUAAAAAUAACAGCUC—3′; antisense: 5′—GCUGUUAUUUUUAGAUACUUU—3′ (PMID: 36,463,391). Tumor development was tracked from day 0 to day 28, and the formula for determining tumor size was volume = (length × width2) × 0.5. Following 28 days, euthanasia of mice was done by CO2 inhalation, and tumor nodules were removed and weighed.

RNA m6A quantification

The RNA quality of the total extracted RNA was checked by NanoDrop (Thermo Fisher Scientific, USA). Then, EpiQuik m6A RNA Methylation Quantification Kit (Epigentek, USA) was utilized to examine the total RNA m6A modification levels. In short, RNA (200 ng) was coated with m6A standards on the detection wells and then treated with capture and detection antibody solutions. The m6A level was quantified by colorimetric measurement by taking the absorbance at 450 nm.

MeRIP assay

MeRIP was done by utilizing the Magna MeRIP m6A Kit (Cloud-seq Biotech., China). Total RNA (150 µg) extracted from si-KIAA1429- or si-NC-transfected HSC-2 and SCC-9 cells was fragmented into 100 or lower nucleotides. The pre-treated RNA was immunoprecipitated with magnetic beads precoated with anti-m6A (3 µg, ab286164, Abcam) or anti-IgG (3 µg, ab133470, Abcam) antibodies. After eluting, the RNA fragments were purified with an RNA purification kit (Qiagen). The enriched fragments were evaluated via qRT-PCR.

Western blotting

Primary antibodies including anti-CA9 (ab15086), and anti-GAPDH (ab8245), all diluted to 1: 1000, were all obtained from Abcam (UK). Whole cell proteins were extracted via RIPA lysis buffer (Beyotime, China), quantified via BCA assay kit (Beyotime), separated on 10% SDS-PAGE, and transferred onto PVDF membranes. The membranes were incubated with the primary antibodies overnight at 4 °C after blocking with 5% skim-milk. Next day, the membranes were incubated with the secondary fluorescent rabbit antibody (1:5000, Abcam) for an hour at room temperature. An ECL kit (Beyotime) was utilized to see protein bands, and the protein expression was quantified by Image J software.

Statistical analysis

The data from 3 independent experiments were examined by GraphPad Prism 7 (GraphPad, USA) and displayed as the mean ± standard deviation. Differences between groups were evaluated using two-tailed Student’s t-test for 2 groups or one-way ANOVA for 3 or more groups. A P value of < 0.05 was considered statistically significant.

Results

High KIAA1429 expression and increased m6A levels are shown in OSCC

According to the data from GEPIA (http://gepia.cancer-pku.cn/index.html), KIAA1429 was upregulated in HNSCC samples (Fig. 1A). To clarify the KIAA1429 level in OSCC, qRT-PCR was performed. The KIAA1429 expression was amplified in OSCC tissue samples as compared to control samples (Fig. 1B). Similarly, elevated KIAA1429 was observed in OSCC cells compared to HIOEC (Fig. 1C). In addition, the tissue m6A level assay showed approximately 6.5-fold higher m6A content in OSCC tissues as compared to control tissues (Fig. 1D). Similarly, m6A levels were markedly upregulated in OSCC cells as compared to HIOEC (Fig. 1E). In particular, HSC-2 and SCC-9 cells showed the most pronounced increment of KIAA1429 and m6A content compared to other OSCC cells, so these two were used for the following experiments. In conclusion, OSCC tissues and cells have high levels of KIAA1429 and m6A.

Fig. 1
figure 1

High KIAA1429 expression and increased m6A levels is shown in OSCC. A The expression levels of KIAA1429 were analyzed using the GEPIA database in 519 head and neck squamous cell carcinoma samples and 44 normal controls from TCGA and Genotype-Tissue Expression GTEx data (http://gepia.cancer-pku.cn/help.html). HNSCC, head and neck squamous cell carcinoma. B The levels of KIAA1429 in OSCC samples and normal tissues were revealed via qRT-PCR. **P < 0.001. C The levels of KIAA1429 in HIOEC and OSCC cells (HSC-2, HSC-4, SCC-25, and SCC-9) were revealed via qRT-PCR. *P < 0.05, **P < 0.001 vs. HIOEC. D The m6A levels of OSCC samples and normal tissues were revealed via RNA m6A quantification. **P < 0.001. E The m6A levels of HIOEC and OSCC cells (HSC-2, HSC-4, SCC-25, and SCC-9) were revealed via RNA m6A quantification. **P < 0.001 vs. HIOEC

Full size image

KIAA1429 silencing attenuates OSCC cell proliferation and metastasis in vitro and in vivo

We further explored the properties of KIAA1429 in HSC-2 and SCC-9 cells. Given the dysregulation of KIAA1429 in OSCC, we used loss-of-function experiments to reveal the effect of KIAA1429 silencing on the malignant actions of OSCC cells. The KIAA1429 expression was manipulated by introducing si-KIAA1429#1, si-KIAA1429#2 and si-NC into HSC-2 and SCC-9 cells. Results of qRT-PCR showed that KIAA1429 was effectively silenced in both OSCC cells and that si-KIAA1429#1, si-KIAA1429#2 had similar effects (Fig. 2A). We selected si-KIAA1429#1 for further study and renamed it as si-KIAA1429. The results of CCK-8 elucidated that silencing KIAA1429 reduced proliferation of HSC-2 and SCC-9 cells (Fig. 2B). Transwell showed that the loss of KIAA1429 decreased the migrative rate of HSC-2 and SCC-9 cells (Fig. 2C). Similarly, the invasive capability of OSCC cells was decreased (Fig. 2D). Taken together, silencing KIAA1429 retarded the proliferation as well as metastasis of OSCC cells in vitro.

Fig. 2
figure 2

KIAA1429 silencing attenuates OSCC cell proliferation and metastasis in vitro. A The efficiency of KIAA1429 interference (si-KIAA1429#1, si-KIAA1429#2) in HSC-2 and SCC-9 cells was checked by qRT-PCR. B The effect of KIAA1429 interference on cell proliferation was examined using CCK-8 assay. C The effect of KIAA1429 interference on cell migration was examined using transwell assay. D The effect of KIAA1429 interference on cell invasion was examined using transwell assay. **P < 0.001 vs. si-NC

Full size image

Thereafter, we evaluated the in vivo oncogenic potential of KIAA1429, and we performed subcutaneous tumor implantation in mice with sh-KIAA1429 and sh-NC transfected HSC-2 cells. The stable silencing of KIAA1429 reduced tumor volume (Fig. 3A), weight (Fig. 3B), and size (Fig. 3C). Taken together, silencing endogenous KIAA1429 suppressed OSCC growth in vivo.

Fig. 3
figure 3

KIAA1429 silencing attenuates tumor formation in OSCC cells in vivo. A The curve of tumor volume from the nude mice injected with sh-KIAA1429 and sh-NC. B The tumor weight was measured on the 28th day from the nude mice injected with sh-KIAA1429 and sh-NC. C The tumor samples from the nude mice given sh-KIAA1429 and sh-NC injections are shown here. **P < 0.001 vs. sh-NC

Full size image

KIAA1429 upregulates CA9 expression in a m6A-dependent manner

KIAA1429 mediates m6A methylation on RNA, and we plan to elucidate whether the function of KIAA1429 is directly dependent on its m6A catalytic activity. According to the data from GEPIA, the positive expression correlation between KIAA1429 and CA9 in HNSCC samples was observed (Fig. 4A). The qRT-PCR assay elucidated that CA9 was overexpressed in OSCC tissues as compared to normal tissues (Fig. 4B). Furthermore, the CA9 expression was positively correlated to KIAA1429 expression in our 34 OSCC tissues (Fig. 4C). KIAA1429 was ectopically under- or overexpressed in HSC-2 and SCC-9 cells, and then CA9 expression levels were assessed. Interestingly, KIAA1429 overexpression and silencing resulted in a notable increase and decrease in CA9 protein expression levels in OSCC cells, respectively (Fig. 4D). In addition, m6A-methylated CA9 was immunoprecipitated in si-KIAA1429 cell lysates with m6A antibody and then quantified by qRT-PCR. We observed that knockdown of KIAA1429 led to a marked reduction in the methylation level of CA9 (Fig. 4E). In addition, CA9 mRNA levels were reduced after si-KIAA1429 transfection (Fig. 4F). These results directly suggest that KIAA1429 promotes m6A methylation of CA9, leading to up-regulation of CA9 expression.

Fig. 4
figure 4

KIAA1429 upregulates CA9 expression via its m6A modification. A The expression correlation between KIAA1429 and CA9 in head and neck squamous cell carcinoma was analyzed by GEPIA and presented as scatter plot. This plot is the result of pair-wise gene expression correlation analysis for the given sets of TCGA and/or GTEx expression data, using methods including Pearson, Spearman and Kendall (http://gepia.cancer-pku.cn/help.html). B CA9 levels in OSCC samples and normal tissues were revealed via qRT-PCR. **P < 0.001. C The correlation between CA9 expression and KIAA1429 expression in 34 OSCC tissues was confirmed by Pearson’s correlation analysis. D The CA9 protein levels in HSC-2 and SCC-9 cells with KIAA1429 interference and overexpression was measured by western blotting analysis. **P < 0.001 vs. si-NC and ##P < 0.001 vs. Vector. (E) The level of m6A modification of CA9 was measured by MeRIP assay in KIAA1429 interference HSC-2 and SCC-9 cells. **P < 0.001 vs. si-NC. (F) The mRNA expression of CA9 in HSC-2 and SCC-9 cells with KIAA1429 interference was measured by qRT-PCR analysis. **P < 0.001 vs. si-NC

Full size image

KIAA1429 accelerates malignant phenotype of OSCC cells by interacting with CA9

To further assess whether KIAA1429 regulated the growth of OSCC cells by interacting with CA9, we co-transfected OSCC cells with CA9-OE vectors and/ or si-KIAA1429 vectors. We detected that CA9 mRNA and protein levels were notably enhanced by CA9 overexpression, whereas CA9 overexpression-induced elevated CA9 expression was alleviated by KIAA1429 silencing (Fig. 5A, B). As shown in Fig. 5C–E, CA9 upregulation increased proliferation, migration and invasion of OSCC cells, which were counteracted by KIAA1429 silencing in HSC-2 and SCC-9 cells. Thus, KIAA1429 exerts its pro-tumor potential by increasing CA9 expression in OSCC.

Fig. 5
figure 5

The promoting effects of CA9 overexpression on OSCC cell malignancy were suppressed by KIAA1429 inhibition. A–B The expression of CA9 in transfected SCC-4 and SCC-9 cells was measured by qRT-PCR (A) and western blotting (B). C The effect of KIAA1429 interference and CA9 overexpression on cell proliferation was examined using CCK-8 assay. D The effect of KIAA1429 interference and CA9 overexpression on cell migration was examined using transwell assay. E The effect of KIAA1429 interference and CA9 overexpression on cell invasion was examined using transwell assay. **P < 0.001 vs. Vector and #P < 0.05, ##P < 0.001 vs. CA9-OE

Full size image

Discussion

In the present investigation, the KIAA1429/CA9 axis was identified to play a crucial role in OSCC development. KIAA1429 was upregulated in OSCC, and its silencing lowered OSCC cell proliferation, migration and invasion in vitro and blocked tumorous growth in vivo. Furthermore, OSCC tissues as well as cell lines had high levels of m6A. Moreover, CA9 was found to be a downstream target of KIAA1429, which catalyzed its m6A modification to stabilize translation. The outcomes of this work elucidates a novel regulatory network between m6A modification and cancer progression in OSCC.

Being one of the most common epigenetic modifications in eukaryotic mRNAs, m6A methylation displays various key functions in mammals, such as in embryonic development, circadian rhythms, neurogenesis, sex determination, stress response, and tumorigenesis. Elevated levels of KIAA1429, a member of the methylation transferase family, have been reported in multiple tumors (Su et al. 2022). Lan et al. (Lan et al. 2019) reported that KIAA1429 targeted and modified GATA3 m6A methylation and its knockdown inhibited GATA3 expression in hepatocellular carcinoma cells by suppressing proliferation and metastasis. In breast cancer, KIAA1429 expression was positively correlated with migration, invasion and epithelial-mesenchymal transition of breast cancer cells by regulating the amount of m6A in SMC1A (Zhang et al. 2022). Furthermore, overexpression of KIAA1429 in OSCC predicted worse overall survival in cancer patients. Similar to that reported by Paramasivam et al. (Arumugam et al. 2021; Paramasivam et al. 2021), we also found that KIAA1429 was upregulated in OSCC tissues. We further found that KIAA1429 silencing inhibited OSCC cell malignancy in vitro and in vivo. A pro-carcinogenic role of KIAA1429 in OSCC was demonstrated. In addition, this study identified CA9 as a downstream gene of KIAA1429 and found that it mediated m6A modification of CA9 to upregulate CA9 expression. This offers a new idea and theoretical underpinning for the mechanism of m6A modification in OSCC.

CA9 is a transmembrane zinc metalloenzyme with cell adhesion functions and plays a pivotal function in tumor growth and survival in normoxic and hypoxic environments. Its expression is mainly restricted to precancerous and malignant cells and is rarely seen in normal tissues or benign lesions. Overexpression of CA9 has been linked with poor prognosis in numerous cancers. Chen et al. (Huang & Zhan 2022) found that CA9 controls apoptosis, cell proliferation and immune response in oral cancer cells. Qian et al. (Chien et al. 2012) observed that CA9 gene polymorphisms and environmental interactions such as smoking and betel nut chewing may alter susceptibility and metastasis of OSCC. Similarly, we found that CA9 was significantly elevated in OSCC and promoted cancer cell growth and metastasis in vitro, suggesting that CA9 acts as a pro-oncogenic factor in OSCC. Furthermore, functional assays showed that the pro-malignant phenotype-promoting role of CA9 overexpression in OSCC cells was reversed by KIAA1429 silencing. In conclusion, the KIAA1429-m6A-CA9 axis plays a pro-oncogenic role in OSCC.

There are some limitations of the current work, such as the lack of clinical studies on KIAA1429 and CA9. In addition, reports show that KIAA1429 is involved in cancer progression via the regulation of signaling pathways such as JAK2/STAT3, Hippo-YAP (X. Chen et al. 2023; Luo et al. 2023), so the downstream mechanism of KIAA1429 regulation of CA9 in OSCC needs to be examined in the future studies.

Conclusion

To summarize, we uncovered a novel regulatory mechanism for the high expression of KIAA1429 in OSCC. CA9 was modified by KIAA1429 in the form of m6A to promote OSCC proliferation and metastasis, implying that the KIAA1429-m6A-CA9 axis could be a potential therapeutic target in OSCC.