Association of single nucleotide polymorphism miRNA-146a (rs2910164) with increased predisposition to oral squamous cell carcinoma in central India population

Abstract

BACKGROUND:

miRNAs play a crucial role in the genesis of cancer, either as tumor suppressor genes or as oncogenes. Single Nucleotide Polymorphisms (SNPs) in the seed region of microRNAs (miRNAs) can dysregulate their levels in the tissues and thereby affect carcinogenesis. The association of SNP in miR-146a (rs2910164) with the risk of oral squamous cell carcinoma (OSCC) has not been understood.

OBJECTIVE:

In the present study, we have determined the association and functional significance of miR-146a (rs2910164) SNP with susceptibility to OSCC predisposition.

METHODS:

In the present case-control study, we enrolled 430 subjects from central India (215 OSCC cases and 215 healthy controls). We performed genotyping by Kompetitive Allele Specific PCR (KASP), and their correlation with OSCC susceptibility was analyzed. miRNA expression profiling in tumor tissues and adjacent normal tissues from six OSCC patients was done by a NanoString n-Counter-based assay. Subsequently, gene ontology and pathway analysis were performed with FunRich version 3.13.

RESULTS:

The CC genotype of rs2910164 miR-146a was significantly associated with the increased risk for OSCC (CC vs GC, OR $=$ 2.62; 95% CI: 1.48–4.66; $p$ value $=$ 0.001). However, the GC genotype was protective with GC vs CC (OR $=$ 0.38, 95%CI $=$ 0.21–0.67, $p$ -value $=$ 0.001), and GC vs GG (OR $=$ 0.58, 95%CI $=$ 0.37–0.89, $p$ -value $=$ 0.01).

CONCLUSION:

Our finding suggests that SNP rs2910164 of miR-146a may be a genetic risk factor for OSCC susceptibility in the Central India population. However, more extensive multicenter studies are required to validate these findings.

1. Introduction

Oral Squamous Cell Carcinoma (OSCC) is a locally aggressive tumor that arises in the oral cavity and lips and represents the most common type of Head and Neck Cancer (HNC) [1]. According to the Global Cancer Observatory (GCO) data, in 2020, the global incidence of OSCC was 377,713. Asia contributed the most, with the highest number, 248,360, followed by Europe and North America [2]. In India, the estimated incidence of OSCC in 2020 was around 135,929, which accounted for 10.3% of the total cancer burden of the country (GLOBOCAN 2020) [1]. Major risk factors of OSCC are smokeless tobacco, betel-quid chewing, alcohol, smoking, and unhygienic oral condition. These factors work synergistically towards the initiation, development, and progression of OSCC and collectively increase the risk by 35% [3]. Human Papillomavirus (HPV) and Epstein-Barr Virus (EBV) are also accepted as risk factors and are associated with cancer of the oropharynx and nasopharynx [4]. Other than these risk factors, genetic variations like polymorphisms in DNA repair genes and cellular regulatory networks can modify the susceptibility to OSCC. Single nucleotide polymorphisms (SNPs) in microRNA (miRNA) genes plays an active role in establishing susceptibility towards OSCC through their involvement in various cellular proliferative, developmental, metabolic, and immunological mechanisms [5].

MicroRNAs are small non-coding RNA molecules, approximately 18–24 nucleotides long, bind to 3 ${}^{\prime }$ -UTR/CDS/5 ${}^{\prime }$ -UTR of their target mRNAs. This interaction leads to either translational repression or degradation of the target mRNA, depending on the degree of complementarity between the miRNA seed sequence and the target mRNA [6]. Unlike mRNA, miRNAs undergo nuclear and cytoplasmic processing for maturation. miRNA biogenesis starts with the transcription of a primary-miRNA (pri-miRNA) transcript by RNA Polymerase II. This transcript is several kilobases long and undergoes several modifications, such as 5 ${}^{\prime }$ capping and the addition of a poly-A tail [6]. The stem-loop of pri-miRNA is recognized by DGCR8 and sent to DROSHA, an RNAase III enzyme that cleaves the stem to form precursor-miRNA (pre-miRNA) of 60–70 nucleotide length with a hairpin structure [7]. Pre-miRNA comes out in the cytoplasm with the help of exportins and is further cleaved to mature miRNA by DICER, another RNAase III enzyme. One of the strands of mature miRNA, i.e., antisense or guide strand, is loaded with Argonaute proteins to form an RNA-induced silencing complex (RISC), and the other strand (sense or passenger strand) degrades out [6, 7].

SNPs are the most common types of genetic variations that can alter target binding affinity, expression pattern, and abnormal maturation. A strong association exists between SNPs in miRNA gene, promoter, and target binding sites with genetic susceptibility to various cancers [5]. Furthermore, SNPs are specific to ethnicity, and exploration in diverse populations will further help in the development of diagnostic, prognostic as well as therapeutic modalities for cancer [5]. miR-146a is known to have an antagonistic effect on inflammation and dampening of adaptive as well as innate immune responses [8]. It has recently gained attention due to its major role in M2 macrophage polarization and NF-kB regulation [9]. rs2910164 (G $>$ C) SNP in miR-146a is located in the seed region and exhibits a transcript variant that dysregulates many cellular functions and has been implicated in the pathogenesis of various diseases, including cancer [10].

Previous studies have shown that rs2910164 in miR-146a is significantly associated with non-small cell lung cancer (NSCLC), gastric cancer, sporadic breast cancer, colorectal cancer, and digestive system cancer [11, 12, 13, 14]. Although previous studies have explored the role of rs2910164 in miR-146a and its significance with a predisposition to multiple cancers, its role in OSCC is not well studied. Therefore, the present study has analyzed the association of rs2910164 of miR-146a with OSCC susceptibility, miR-146a expression levels in tumor tissue of OSCC patients, and in silico functional characterization.

2. Methodology

2.1 Study design

The present study is a hospital-based case-control study. A total of 430 subjects, patients with OSCC ( $N=$ 215), and 215 disease-free subjects (controls) with no evidence of OSCC or any oral lesions, were enrolled in the study. Written informed consent was taken from all the participants. Adults (18–70 years of age) with histologically confirmed cases of OSCC were enrolled at All India Institute of Medical Sciences, Bhopal. Disease-free adult healthy subjects with no previous history of malignancy were recruited as control subjects. In order to match the exposure, more than 80% of control subjects with a history of tobacco consumption (smokeless tobacco/gutkha/supari/surti) were recruited in the study. None of the control subjects had first-degree relatives with a history of oral cancer. The frequency of controls was matched to cases on age ( $\pm$ 5 years), and all the participants were from Central India. Venous blood (2 ml) was collected from each participant. The protocol was approved by the Institutional Human Ethics Committee (IHEC), AIIMS Bhopal. In the present study, the prominent lesion site in OSCC patients was left buccal mucosa (33%) and tongue (35%).

2.2 SNP genotyping

Blood samples were collected in EDTA vials, and genomic DNA from all cases and controls was isolated by the conventional phenol-chloroform method [15]. Genotyping of rs2910164 in miR-146a was performed using Kompetitive Allele Specific PCR (KASP) genotyping assay, and the post-run data were analyzed using the SNPviewer software, LGC Biosearch Technologies [16]. All assays were performed on 96-well plates. Primers for each tag SNP were designed using Kraken ${}^{\text{TM}}$ software (LIMS platform) with sequence 5 ${}^{\prime }$ CCTGGACTGCAAGGAGGGGTCTTTGCACCATCTCTGAAAAGCCGATGTGTATCCTCAGCTTTGAGAACTGAATTCCATGGGTTGTGTCAGTGTCAGACCT[G/C]TGAAATTCAGTTCTTCAGCTGGGATATCTCTGTCATCGTGGGCTTGAGGACCTGGAGAGAGTAGATCCTGAAGAACTTTTTCAGTCTGCTGAAGAGCTTGGAAGACTGGAGACAGAAGGC-3 ${}^{\prime }$ . KASP assays were performed as described previously [16].

2.3 RNA Extraction and expression profiling

A total of six biopsy-proven newly diagnosed adult patients (cases) of OSCC were included for mRNA expression study, and independent diagnosis, and histological examinations of each patient was made by physicians and pathologists following standard criterion (Table 3). Three patients had tumor in the buccal mucosa and three in the tongue. None of the patients received curative treatment (surgery, chemo- or radiotherapy) before their enrollment. Tumor tissue (approx. 2 mm $\times$ 2 mm $\times$ 2 mm) and adjacent normal tissue (approx. 2 mm $\times$ 2 mm $\times$ 2 mm), preferably 2 cm away from the tumor margin, were obtained from these patients with their written consent. None of the patients was COVID positive or had recent history of COVID. As per case record proforma there was no history of other chronic disease in patients under investigation. Total RNA was isolated from tumor and adjacent normal tissues using TRIzol reagent (Invitrogen) and Aurum ${}^{\text{TM}}$ RNA extraction kit (Bio-Rad Laboratories), as described previously [17]. miRNA expression profiling was done with nCounter ${}^{\mathrm{\circledR }}$ gene expression panels with 770 miRNA targets (retrieved from mirBase) [18]. Nanostring nCounter is a multiplex RNA hybridization assay that utilizes target specific reporter (color coded) and capture probe and observation under nCounter instrument [18].

2.4 Target prediction and functional annotations

Target genes for miR-146a-5p were predicted using the multiMiR package of R [19]. The multiMiR data-base represents data from fourteen external databases consisting of miRecords, TarBase, and miRTarBase (validated miRNA-Target database), DIANA-microT, MicroCosm, miRanda, ElMMo, miRDB, PITA, PicTar, TargetScan (Predicted miRNA-Target databases), miR2Disease, PhenomiR, and Pharmaco-miR (disease-/drug-related miRNA databases. Subsequently, the impact of SNP in miR-146a was predicted for any change in the miR-146a target site, like disruption or creation of a new target site with miRSNPV3, MicroSNiPer, multiMiR (R) [10, 20].

The second level of validation was applied to further select significant miRNA targets at the expression level in oral cancer. A list of downregulated genes in OSCC was downloaded from the TCGA databases ( $|$ Log2FC $|$ $<$ 1.5, $q$ -value $<$ 0.05, limma methods). The intersection of the Venn diagram was taken as target genes downregulated with upregulation of miR-146a-5p in oral cancer.

2.5 Pathway analysis

The molecular mechanism and significant pathway associated with miR-146a-5p dysregulation were found using FunRich version 3.1.3 with Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) based on p-value. Further, the protein-protein interaction (PPI network) was made with the STRING database (http://string-db.org/), and ENCORI Pan-Cancer analysis was performed to gain insight into the global scenario of the target genes with head and neck squamous cell carcinoma (HNSCC).

Table 1

Distribution of selected demographic variables of OSCC Patients and Healthy subjects

Factors	Healthy, $N=$ 215 ${}^1$	OSCC patients, $N=$ 215 ${}^1$	$P$ ${}^2$
Age (years)	46.5 (15.9)	48.2 (13.1)	0.5
Gender			0.4
Female	29	24 (%)
Male	71	76 (%)
Lesion site			$>$ 0.9
Alveolus	NA	13 (%)
Buccal Mucosa	NA	37 (%)
Gingivobuccal sulcus	NA	6.7 (%)
Hard palate	NA	3.3 (%)
Tongue	NA	40 (%)
Histopathology
Poorly differentiated	NA	31 (%)
Moderately differentiated	NA	18 (%)
Well differentiated	NA	51 (%)
Smokeless Tobacco chewing
No	NA	36 (%)
Yes	NA	64 (%)
Smoking			0.4
No	67	74 (%)
Yes	33	26 (%)

${}^1$ Mean (Standard deviation) or Frequency (%); ${}^2$ Wilcoxon rank sum test; Pearson’s Chi-squared test; Fisher’s exact test.

2.6 Statistical analysis

The distribution of selected demographic factors between cases and controls was examined using the Wilcoxon rank-sum test, Pearson’s Chi-square test, and Fisher’s exact test. The goodness-of-fit test was performed to assess any deviation of genotypic frequency in the controls from the Hardy-Weinberg equilibrium (HWE). The odds ratio (ORs), 95% confidence interval (CI), and p-value were used for the analysis and performed using the Chi-square test with five different models; Allele, Co-dominant, Dominant, Recessive, and Heterozygous. All statistical tests were carried out using SPSS-25 software (SPSS, Inc.) and the gtsummary package of R 4.2.1 tool. All p-values were two-sided, and p-value less than 0.05 was considered statistically significant.

3. Results

3.1 Study population and genotyping

The present study had 215 OSCC cases and 215 healthy controls. The distribution of selected demographic variables is summarized in Table 1. The mean age of subjects was 46.2 $\pm$ 15.8 and 48.7 $\pm$ 13.1 years in healthy and OSCC patients, respectively, and there was no statistical difference between the two groups ( $p$ -value $=$ 0.12). Similarly, there were no measurable differences in gender and weight distribution between the cases and controls ( $p=$ 0.2). As per the inclusion criterion (mentioned in methods), all controls were tobacco consumers, while 63% of cases were tobacco consumers (Table 1). History of smoking and alcohol consumption were not significantly correlated with OSCC. As reported previously [17], left buccal mucosa and tongue were the most prominent sites of the oral cavity, with 34% and 35% of the total OSCC cases. Histopathological evidences represent 51% of patients with well differentiated squamous cell carcinoma, followed by poorly differentiated (31%) and moderately differentiated (18%) (Table 1).

Table 2

Genotypic distribution of miR-146a-5p rs2910164 in OSCC patients and healthy control subjects with various genotyping models

SNP (rs2910164)	OSCC patients ( $n=$ 215 subjects)%	Healthy control ( $n=$ 215 subjects)%	OR (95% CI)	Corrected $p$ -value
Allele
G	57.8	59.1	$1$
C	42.2	40.9	1.05 (0.8–1.38)	0.72
Co-dominant
GG	36.2	29.1	$1$
GC	43.4	60.1	0.58 (0.37–0.89)	0.01
CC	20.4	10.8	1.52 (0.83–2.79)	0.22
Dominant
GG	36.2	29.1	$1$
GC+CC	63.8	70.9	0.72 (0.98–1.08)	0.14
Recessive
GG $+$ GC	79.6	89.2	$1$
CC	20.4	10.8	2.12 (1.23–3.67)	0.007
Heterozygous
GC vs CC	$-$	$-$	0.38 (0.21–0.67)	0.001
GC vs GG	$-$	$-$	0.58 (0.37–0.89)	0.01
Homozygous
GG vs GC	$-$	$-$	1.72 (1.12–2.6)	0.01
CC vs GC	$-$	$-$	2.62 (1.48–4.66)	0.001

Table 3

Demographic and clinicopathological details of OSCC patients included in miR-146a expression profiling

Factors	Buccal mucosa, $N=$ 3 ${}^1$	Tongue, $N=$ 3 ${}^1$	$P$ ${}^2$
Gender			$>$ 0.9
Female	0	33
Male	100	67
Staging			$>$ 0.9
Stage-I	33	0
Stage-II	33	67
Stage-III	0	33
Stage-IVA	33	0
Tobacco_History			$>$ 0.9
No	0	33
Yes	100	67

${}^1$ Mean (Standard deviation) or Frequency(%); ${}^2$ Fisher’s exact test.

The observed genotype frequencies for the rs2910164 were in agreement with that expected under the HWE in the controls ( $X^2=$ 2.985, $P=$ 0.225), which confirms the randomness in the control sample. The different genotype models (Co-dominant, Dominant, Heterozygous, Recessive, Homozygous), and respective allele frequency of the rs2910164 are given in Table 2. As shown in Table 2, in the allelic frequency distribution, ’G’ was the major allele in healthy controls and OSCC cases with 57.8% and 59.1%, over ‘C’ allele with 42.2% and 40.9% respectively. Heterozygous genotype GC was common in the central Indian population (Frequency; 60.1% control; and 43.4% cases), and CC was rare for miR-146a (Frequency; 10.8% control; and 20.4% cases for rs2910164 of miR-146a. Importantly, the CC genotype was significantly associated with the increased risk for OSCC (Homozygous model, CC vs GC, OR $=$ 2.62, 95%CI: 1.48–4.66, $p$ value $=$ 0.001). Similarly, in recessive models the genotype CC was associated with OSCC risk in comparison to GG $+$ GC (Recessive model, CC vs GG $\mathrm{+}$ GC, OR $=$ 2.12, 95%CI was 1.23–3.67, $p$ value $=$ 0.007). We also found that the GC genotype was significantly associated with OSCC predisposition in heterozygous models GC vs CC (OR $=$ 0.38, 95%CI was 0.21–0.67, $p$ value $=$ 0.001), and GC vs GG (OR $=$ 0.58, 95%CI was 0.37–0.89, $p$ value $=$ 0.01). Moreover, GC genotype showed negative correlation with OSCC risk. However, in allelic models, we could not find any significant association with OSCC susceptibility (OR $=$ 1.05 with 95%, CI 0.80–1.38). This showed the recessive nature of C allele, which is further validated by the fact that the homozygous genotype CC shows significance in recessive model (OR $=$ 2.12, 95% CI $=$ 1.23–3.67, $p$ value $=$ 0.007), but not in dominant models (OR $=$ 0.72, 95% CI $=$ 0.44–1.27, $p$ value $=$ 0.35). Further, rs2910164 of miR-146a was not significantly correlated with OSCC staging, Histopathology, Tumor site, Age (years), and Gender (Table 4).

Figure 1.

A) miR-146a expression profile in OSCC tumor and adjacent normal tissue. Figure shows the upregulation of miR-146a in tumor tissue by 1.42-fold in comparison to normal tissue. B) The ENCORI Pan-Cancer analysis of miR-146A expression in Head and Neck Squamous Cell Carcinoma and Overall survival with LogRank $p$ -value $=$ 0.1.

Figure 2.

Twenty potential target genes were gained by rs2910164 of miR-146a. The genes were selected as potential targets based on their expression levels in oral cancer.

3.2 Expression profiling and bioinformatics analysis

Table 4

rs2910164 of miR-146A correlation with clinical features of OSCC patients

Clinical features	CC, $N=$ 44 ${}^1$	GC, $N=$ 98 ${}^1$	GG, $N=$ 73 ${}^1$	$P$ ${}^2$
OSCC staging				0.9
Stage I and II	13	20	20
Stage III and IV	87	80	80
Histopathology				0.8
Poorly differentiated	12	24	17
Moderately differentiated	19	15	24
Well differentiated	69	61	59
Tumor site				0.8
Lower lip	0	3	3
Tongue	22	34	47
Buccal mucosa	67	46	38
Other sites of oral cavity	11	17	12
Age (Years)				$>$ 0.9
$<$ 45 Years	44	40	40
$>$ 45 Years	56	60	60
Gender				0.3
Female	37	22	29
Male	63	78	71

${}^1$ Frequency(%); ${}^2$ Fisher’s exact test; Pearson’s Chi-squared test.

Figure 3.

The FunRich version 3.1.3 based Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis for target genes gained by miR-146A with rs2910164. A) Activated pathway: Pyrimidine metabolism, Transcriptional activation of p53 and p21, Activation of PUMA and NOXA; B) Transcription factors, C) Biological processes, D) COSMIC.

We found that the expression of miR-146a-5p was upregulated in tumor tissue by 1.42-fold compared to normal tissue (Fig. 1A). The survival analysis with univariate Cox proportional hazard regression model showed no significance with up-regulation of miR-146a expression ( $p=$ 0.1). Further, miR-146a expression correlation analysis with various clinical features reveals a trend of positive correlation between miR-146a expression and OSCC staging ( $p=$ 0.056, $r=$ 0.8). A total of 381 target genes gained by rs2910164 of miR-146a were identified with multiMiR R package and miRSNP-v3; among them, 20 common genes were differentially expressed in OSCC tumors (TCGA database) by Venn diagram (Fig. 2). The FunRich version 3.1.3 based GO and KEGG pathway enrichment analysis divulges additional pathway activation and regulation of biological processes (Fig. 3). These processes might play an essential role in the development and progression of the tumor through transcriptional activation of p53-responsive genes, cell cycle inhibitor p21, pyrimidine metabolism, PUMA, and NOXA activation (Fig. 3). Selected target genes were also entailed in the regulation of some key transcription factors (TF) involved in the maintenance of cellular homeostasis and thus indirectly affects homeostasis (Fig. 3). Based on GO and KEGG analysis, spleen-associated tyrosine kinase (SYK), tumor protein (TP53), F-box/WD repeat-containing protein 7 (FBXW7), uridine cytidine kinase 1 (UCK1), argininosuccinate lyase (ASL), establishment of sister chromatid cohesion N-acetyltransferase 2 (ESCO2), and dihydropyrimidinase (DPYS) were the most presumable target gained by rs2910164 of miR-146a. A protein network was built using STRING, and TP53, SYK, and FBXW7 were identified as hub genes, and are the most likely targets of miR-146a. ENCORI, a Pan-Cancer Analysis Platform showed the upregulation of TP53 and SYK, and downregulation of FBXW7 in HNSCC (Fig. 4). FBXW7 downregulation was significantly associated with overall survival in HNSCC with $p=$ 0.0017 and a hazard ratio of 0.65 (Fig. 4).

Figure 4.

The ENCORI Pan-Cancer analysis for final target genes gained by miR-146a with rs2910164. Box plot of gene expression profile in Head and Neck Squamous Cell Carcinoma FBXW7 (A), TP53 (B), SYC (C); Overall survival for FBXW7 in Head and Neck Squamous Cell Carcinoma (D).

Figure 5.

Correlation between miR-146a expression and clinical features of OSCC patients. A) Correlation between miR-146a expression and OSCC staging, B) Correlation between miR-146a expression and histology grade of OSCC, C) Correlation between miR-146a expression and OSCC tumor volume, D) Correlation between miR-146a expression and TNM staging.

In selected six OSCC patients, there was a strong positive correlation ( $r=$ 0.8) between miR-146a expression in tumor tissue and overall OSCC staging. However, it did not reach statistical significance ( $p$ -value $=$ 0.056), probably due to small sample size (Fig. 5). Histology grading ( $r=$ 0.62; $p$ -value $=$ 0.19), tumor volume ( $r=$ 0.77; $p$ value $=$ 0.073), and TNM staging ( $r=$ 0.48; $p$ -value $=$ 0.34) also showed a trend of positive correlation with miR-146a expression in OSCC tumors (Fig. 5).

4. Discussion

In the present study, the prominent lesion site was the Tongue (40%) followed by the Buccal mucosa (37%) in OSCC patients (Table 1). We found that rs2910164 of miR-146a was significantly associated with OSCC risk in the central India population. Specifically, the frequency of the CC genotype was significantly increased in patients with OSCC population, and it was positively correlated with OSCC risk. Conversely, the GC genotype was protective and negatively associated with OSCC risk. The ‘G’ allele was the major allele in both OSCC and healthy populations, and ‘C’ was the alternate allele. In addition, we observed 1.42-fold miR-146a upregulation in tumor tissue, possibly due to rs2910164 SNP in miR-146a. Concurrently, MIRNA SNP DISEASE Database (MSDD) data further supports our findings and suggest that rs2910164 may affect the expression of miR-146a. The elevated miR-146a expression has been reported previously in tumor tissue from OSCC and other cancer types, including breast cancer, hepatocellular carcinoma, ovarian cancer, and lung cancer [21, 22, 23]. Twenty target genes gained by SNP rs2910164 of miR-146a have an essential role in the cell cycle, nucleotide metabolism, and NOTCH pathway, and might directly correlate with tumor development and progression.

miR-146a has shown a crucial role in the expression of epidermal growth factor receptor (EGFR), and TNF receptor-associated factor 6 (TRAF6). It has also been shown to regulate NF- $\kappa$ B by directly binding to EGFR and plays a prime role in various clinical phenotypes, including cancer [24]. The key reason for studying the miR-146a rs2910164 in the central India population was the notable role of this SNP in multiple cancer predisposition in various populations and being unexplored in the Indian population [25, 26, 27, 28, 29]. miR-146a can bind with 3 ${}^{\prime }$ -UTRs of the IL-1 receptor-associated kinase 1 (IRAC1) and TRAF6 that are associated with enhanced inflammation [27]. The additional pathways activated due to rs2910164 polymorphism might assist in OSCC aggressiveness. Hung et al. 2012, revealed that rs2910164 (G $>$ C polymorphism) in pre-miR-146a was associated with the expression pattern of miR-146a and more likely to upregulated in advanced nodal stage OSCC patients [22]. A similar association rs2910164 with predisposition has been reported in other cancer types. For example, a meta-analysis 2,794 cases and 2,840 controls showed that the SNP rs2910164 was significantly associated with increased risk to lung cancer (CC vs. GG: OR $=$ 1.30, 95% CI: 1.13–1.50, $p<$ 0.001; CC vs. GC $+$ GG: OR $=$ 1.27, 95% CI: 1.13–1.42, $p<$ 0.001; C vs. G: OR $=$ 1.15, 95% CI: 1.08–1.24, $p<$ 0.001) [30]. These observations corroborate with our findings, except for the allelic model. Similarly, Dezfuli et al., 2020 found that the CC genotype of rs2910164 of miR-146a has been shown to significantly increase (OR $=$ 2.93, 95% CI $=$ 1.07–7.9, $p=$ 0.034) the risk of NSCLC in the Iranian population, but could not find any significant association with type or staging [27]. Furthermore, the CC vs CG genotype of rs2910164 of miR-146a was significantly associated with more synchronous liver metastasis, higher NUMB (Notch signaling) expression, and chemotactic activity [13]. Another study examined the association of rs2910164 SNP with susceptibility to childhood acute lymphoblastic leukemia (ALL) in the Taiwan population [31]. They found that the G allele and the GG genotype were negatively associated with childhood ALL risk (OR $=$ 0.66, 95%CI $=$ 0.52-0.85, $p=$ 0.0011 and OR $=$ 0.40, 95%CI $=$ 0.23–0.67, $p=$ 0.0004, respectively) [31]. All these results support our findings and highlight the significance of the CC genotype of miR-146a (rs2910164) in predisposing individuals to several types of cancer, including OSCC.

On the contrary, rs2910164 has also been associated with decreased miR-146a expression in some cancers like papillary thyroid carcinoma, esophageal squamous cell carcinoma, colorectal cancer, and diffuse large B cell lymphoma [32, 33, 34]. An article reviewed the existing data available with the functional role of rs29010164 G/C with miR-146a expression and provided a plausible explanation for the contradictory findings in colorectal cancer [28]. This study found the involvement of unknown mRNA targets and differences in selected individual races are the key factors for contradictory results, and it enhances the need for SNP-based study in a diverse population [28].

We observed that the expression profile of target genes TP53, SYK, and FBXW7 gained by rs2910164 of miR-146a varies in HNSCC (Fig. 4). Pan-Cancer analysis showed that FBXW7 was downregulated in HNSCC cases and significantly correlated with reduced overall survival. FBXW7 is an important transcription factor for reprogramming induced pluripotent stem cells (iPSCs), and its loss of expression correlates with the prediction of recurrence in colorectal liver metastasis [35, 36]. Additionally, regulatory factor as miR-146a might impact FBXW7 expression and function.

5. Conclusion

In conclusion, the present study shows that the rs 2910164 of miR-146a polymorphism is associated with the OSCC risk. Notably, CC genotype of rs2910164of increases the risk of OSCC in the central Indian population, and the GC genotype may protect the individuals. Furthermore, this SNP can dysregulate miR-146a expression in tumors and results in additional target yield, which affects downstream metabolic pathways, biological processes, and transcription factors involved in OSCC development and progression, especially FBXW7. Therefore, miR-146a might play a crucial role in the pathogenesis of OSCC either by initiating the new lesions or by accelerating the transformation of oral potentially malignant lesions into a malignant disease. Thus, it might help in the screening of individuals at risk for OSCC. Furthermore, future studies are warranted to understand the functional role of rs2910164 in the OSCC pathogenesis at cellular and in vivo levels.

Funding

This work was supported by the Department of Health and Research, New Delhi, to ST (R. 12013/11/2021-HR/E-office: 8111586). In addition, AK would like to thank the Indian Council of Medical Research (ICMR 5/13/93/2020/NCD-III) for the Grant-In-Aid.

Author’s contributions

Conception: Shikha Tiwari, Ritu Pandey, and Ashok Kumar.

Investigation: Ritu Pandey, Supriya Vishwakarma, Tulasi Sindhuja, and Sana Hasmi.

Interpretation or analysis of data: Shikha Tiwari and Ashok Kumar.

Preparation of the manuscript: Shikha Tiwari.

Manuscript editing: Ashok Kumar.

Revision for important intellectual content: Vinay Kumar, Saikat Das, and Ashok Kumar.

Supervision: Vikas Gupta, Rajeev Nema, and Ashok Kumar.

Supplementary data

The supplementary files are available to download from http://dx.doi.org/10.3233/CBM-230064.

sj-pdf-1-cbm-10.3233_CBM-230064.pdf - Supplemental material

Supplemental material, sj-pdf-1-cbm-10.3233_CBM-230064.pdf