The cells from human OSCC cell lines, UPCI: SCC084 and UPCI: SCC131, were grown in DMEM supplemented with 10% fetal bovine serum and 1x antibiotic/antimycotic solution (Sigma-Aldrich, St. Louis, MO) in a humidified chamber with 5% CO₂ at 37°C. These cell lines were a kind gift from Prof. Susanne Gollin, University of Pittsburgh, Pittsburgh, PA (28). These cell lines were generated following Institutional Review Board guidelines from consenting patients undergoing surgery for squamous cell carcinoma of the oral cavity at the University of Pittsburgh Medical Centre (28). SCC131 cells were derived from a T2N2bM0 lesion on the floor of the mouth of a 73-year-old Caucasian male, and SCC084 cells were derived from a T2N2M0 lesion of the retromolar trigone of a 52-year-old Caucasian male (28). A total of 39 matched OSCC tumor and their adjacent normal oral tissues from the patients were obtained at the Kidwai Memorial Institute of Oncology (KMIO), Bangalore from July 13, 2018 to November 14, 2018. The samples were obtained as surgically resected tissues from oral cancerous lesions and adjacent normal tissues (taken from the farthest margin of surgical resection) in the RNALater (Sigma-Aldrich, St. Louis, MO) and transferred to -80°C until further use (13, 14). The clinicopathological information for the 39 patients is given in Supplementary Tables S1 , S2 . Tumors were classified according to tumor, node, and metastasis criteria (29).

For the overexpression studies, SCC131 or SCC084 cells were seeded at 2×10⁶ cells/well in a 6-well plate and transfected with an appropriate construct or a combination of different constructs using Lipofectamine 2000 (Thermo Fisher Scientific, Waltham, MA) according to the manufacturer’s instructions. For the dual-luciferase reporter assay, 5 × 10⁴ cells/well were transfected in 24-well plates with different constructs. The dual-luciferase reporter assays were performed post 48 h of transfection using the Dual-Luciferase Reporter Assay System (Promega, Madison, WI) according to the manufacturer’s protocol. The pRL-TK control vector (Promega, Madison, WI) was co-transfected in the cells for normalizing the transfection efficiency (10).

Total RNA from tissues and cells was isolated using TRI-Reagent (Sigma-Aldrich, St. Louis, MO) and quantitated using a NanoDrop 1000 spectrophotometer (Thermo Fischer Scientific, Waltham, MA). The first-strand cDNA was synthesized from 2 µg of total RNA using a Verso cDNA Synthesis Kit (Thermo Fischer Scientific, Waltham, MA).

The expression level of miR-198 was determined by qRT-PCR using the miR-Q technique by Sharbati-Tehrani et al. (30). The details of the primers are given in Supplementary Table S3 . The RT-qPCR analysis was performed using the DyNAmo ColorFlash SYBR Green qPCR Kit in StepOne Plus Real-Time PCR and QuantStudio 3 PCR Systems (Thermo Fischer Scientific, Waltham, MA). GAPDH and 5S rRNA were used as normalizing controls (13). The following equation ΔCt_gene = Ct_gene-Ct_{normalizing control}, was used to calculate the fold change. Ct represents cycle threshold value, and ΔCt represents the gene expression normalized to GAPDH or 5S rRNA.

The miR-198 (pmiR-198) construct was generated in pcDNA3-EGFP, using human genomic DNA as a template and gene specific primers ( Supplementary Table S4 ) by a standard laboratory method. The TOPORS-ORF construct (pTOPORS) in the pcDNA 3.1(+) vector (Invitrogen, Waltham, MA, USA) was obtained from the GenScript (cat # OHu18284C, ORF size- 3138 bp, Piscataway, NJ). For the validation of a direct interaction between miR-198 and the 3’UTR of TOPORS by the dual-luciferase reporter assay, we cloned the 3’UTR of TOPORS in a sense orientation in the pMIR-REPORT miRNA Expression Reporter Vector System (Thermo Fischer Scientific, Waltham, MA) downstream to the luciferase (Luc) gene using human genomic DNA as a template. Briefly, the TOPORS sequences were obtained from the UCSC Genome browser (https://genome.ucsc.edu/) and the 769 bp long fragment was amplified from human genomic DNA using specific primers ( Supplementary Table S4 ). The fragment was first cloned in a cloning vector pBluescript II KS (+) [pBSKS (+)] and the construct was named as pBSKS-TOPORS-3’UTR-S. From this vector, we excised the 3’UTR fragment with Pme I and Mlu I and ligated it to the pMIR-REPORT vector, and this construct was named as pMIR-REPORT-TOPORS-3’UTR-S. To validate the sequence specificity of the interaction between miR-198 and the TOPORS 3’UTR, the target site for miR-198 present in the 3’UTR of TOPORS was abrogated by site-directed mutagenesis (31). For this, the construct pBSKS-TOPORS-3’UTR-M was generated using specific primers ( Supplementary Table S4 ) and pBSKS-TOPORS-3’UTR-S (carrying the TOPORS 3’UTR in a sense orientation) as the template. The mutated TOPORS 3’UTR was excised from this construct with Pme I and Mlu I and ligated in the pMIR-REPORT vector (also digested with the same enzymes) to generate the pMIR-REPORT-TOPORS-3’UTR-M construct. To determine if the 3’UTR of TOPORS is responsible for miR-198-mediated regulation of TOPORS, the pTOPORS-3′UTR-S, and pTOPORS-3′UTR-M constructs were generated by cloning the relevant 3′UTR sequences downstream to the TOPORS-ORF at EcoR V and Not I restriction sites in the pTOPORS construct ( Supplementary Table S4 ). All constructs used in the study were validated by restriction enzyme digestion and Sanger sequencing on a 3730xl DNA analyser (Thermo Fisher Scientific, Waltham, MA, USA).

Protein lysates from cells were prepared using the CelLytic M Cell Lysis Reagent (Sigma-Aldrich, St. Louis, MO). The proteins were then resolved on an SDS-PAGE and then transferred on a PVDF membrane (Pall Corp., Port Washington, NY). The membrane was blocked using 5% fat-free milk powder in 1× PBST (1xX Phosphate-Buffered Saline, 0.1% Tween 20 Detergent), and the signal was visualized using an appropriate antibody and the Immobilon Western Chemiluminescent HRP substrate (Milipore, Billerica, MA). The following antibodies were used: anti-rabbit-TOPORS (1:1,000 dilution; cat# ab86383, Abcam, Cambridge, CAMBS, UK), anti-mouse β-actin (1:10,000 dilution; cat# A5441, Sigma-Aldrich, St. Louis, MO), p53 wild-type (1: 1,000 dilution; cat# AH00152, Thermo Fisher Scientific, Waltham, MA), p21 Waf1/Cip1 (1:2,500 dilution; cat#2946, CST, Danvers, MA) and secondary HRP-conjugated goat anti-rabbit/HRP-conjugated goat anti-mouse (1.5:5,000 dilution; Bangalore Genei, Bangalore, India) antibodies.

To analyze the rate of cell proliferation, a trypan blue dye exclusion assay was employed (13, 32). Briefly, 2x10⁶ cells were seeded in 6-well plates. After 16 hr, cells were transfected with the desired construct(s) as described above. Post 24 h of transfection, they were trypsinized, counted and reseeded at the density of 30,000 cells/well in triplicates in four 24-well plates (one plate for each day; total 4 days). This is day 0. Starting from day 1 (24 h after reseeding), cells from each well were trypsinized. The cells were then diluted 1:10 with 0.4% trypan blue dye (Sigma-Aldrich, St. Louis, MO) prepared in 1XPBS, allowed to stain for 10 min at room temperature and counted using a hemacytometer. This was done for all 4 days. The number of viable cells/mL was calculated using the following formula: number of viable cells/mL = the average count per square (unstained/live cells) x dilution factor x 10⁴ (13).

The CaspGLOW Fluorescein Active Caspase-3 Staining Kit (Biovision, Mountain View, CA) was used to analyze the rate of apoptosis in cells transfected with the appropriate constructs (10). After transfection, FITC-DEVD label was added to cells, and the rest of the steps were followed as per the manufacturer’s protocol. The fluorescence percentage was measured, using FACSCalibur Flow cytometer (BD Biosciences, New Jersey, NJ), using the FL-1 channel and the BD Accuri C6 software.

The anchorage-independent growth is one of the hallmarks of cancer, where the cancer cells have the ability to grow independently on a solid surface and was assessed by the number of colonies formed in soft agar (10, 13, 14). After transfection of cells with different constructs, 6,000 cells were plated in 1 ml of 0.35% Difco Noble Agar (Difco, Mumbai, India) diluted with DMEM culture media in a 35 mm dish. After 20-25 days, colonies were stained with 0.01% crystal violet (Sigma-Aldrich, St. Louis, MO), counted and imaged using a Leica Inverted Microscope DMi1 (Leica Microsystems, Wetzlar, Germany).

Although Duan et al. (33) have earlier identified an independent promoter for miR-198, a thorough analysis of the primers used by them for the promoter sequence revealed the presence of the promoter in another gene, homogentisate 1,2-dioxygenase (HGD; NC_000003.12-120628172-120682239, complement) and not in the FSTL1 gene (NC_000003.12-120392293-120450992, complement), which harbours the MIR198 gene. Therefore, we used the following databases to predict the putative MIR198 gene promoter: DBTSS (http://dbtss.hgc.jp/), MatInspector (https://www.genomatix.de) and Switchgear (https://switchgeargenomics.com). The predicted DNA sequence for the MIR198 promoter was then cloned upstream to the luciferase ORF in the pGL3-Basic vector (Promega, Madison, OH), using a standard laboratory method and specific primers ( Supplementary Table S5 ). The resulting promoter construct, pMIR198-prom was then used for the promoter reporter assay using the Dual-Luciferase Reporter Assay system (Promega, Madison, WI, USA), according to manufacturer’s instructions.

For demethylation experiments, SCC131 cells were grown in T-75 flasks till 80% confluency and treated with a freshly prepared 5-Azacytidine at 5 μM final concentration (Sigma-Aldrich, St. Lous, MO) for 5 days with the drug and the medium being replaced every 24 h. DMSO (Sigma-Aldrich, St. Lous, MO) was used as a vehicle control. Following treatment, cells were harvested, and the genomic DNA was extracted using a Wizard Genomic DNA Purification Kit (Promega, Madison, WI, USA), according to the manufacturer’s instructions. An aliquot of 2 µg of total genomic DNA from each treatment was treated with sodium bisulphite (10, 14). Using sodium bisulphite treated genomic DNA from control (DMSO) and 5-Azacytidine treated SCC131 cells as templates, the MIR198 promoter-specific fragment was amplified ( Supplementary Table S6 ) using methylation specific primers and cloned in the pTZ57R TA cloning vector (Thermo Fisher Scientific, Waltham, MA). Ten random TA clones were selected for each treatment and Sanger sequenced as described above. Methylation-specific primers were designed using the MethPrimer database (http://www.urogene.org/methprimer/).

A consensus approach was employed to predict the gene targets for miR-198 using following mRNA target prediction algorithms like miRDB (http://mirdb.org), TargetScanHuman (https://www.targetscan.org/vert_80/), PicTar (https://pictar.mdc-berlin.de/cgi-bin/PicTar_vertebrate.cgi) and CoMeTa (https://cometa.tigem.it/) ( Supplementary Table S7 ).

To analyze the effect of miR-198-mediated targeting of TOPORS on tumor growth, 2×10⁶ SCC131 cells were transfected with 900 nM of a synthetic miR-198 mimic or 900 nM of a mock control (scrambled oligos) separately. After 24 h of transfection, cells from both groups were suspended separately in 150 μl of incomplete DMEM and then injected into the right flank of a female BALB/c athymic 6-week-old nude mouse subcutaneously (10, 13, 14). Tumors were allowed to grow in animals of the two experimental sets, and tumor volumes were measured using a Vernier’s calliper every alternate day till the termination of the experiment. At the end of the study, animals were euthanized in CO₂ atmosphere under sterile conditions by cervical dislocation. Tumor volumes were calculated using the formula: V = L×W²×0.5, where L and W represent the length and width of the tumor, respectively (10, 13, 14). The animals were photographed, and the tumor xenografts were harvested, photographed and weighed at the end of the study by trained personnel. Mice were maintained on a 12:12 h light/dark cycle in proper cages with sufficient food and water. Efforts were taken to alleviate suffering of the animals, and they were handled and treated only by trained personnel. Animals were consistently monitored for general health and behaviour. miRIDIAN microRNA hsa-miR-198 Hairpin mimic (cat# C-300532-05-0020) and miRIDIAN microRNA Negative Control #1/Mock (cat# CN-001000-01-XX) were purchased from Dharmacon (Lafayette, CO).

The statistical significance of the comparison between any two experimental data sets was calculated using the student’s t test in the GraphPad Prism 8 software (Boston, MA, USA). The statistical significance of comparisons between multiple data sets was calculated by ANOVA (one-way analysis of variance). Comparisons in data sets were considered significant when p-values were ≤0.05 (*), <0.01 (**), <0.001 (***), <0.0001 (****) or non-significant (ns) when the p-value was >0.05.

We first validated the upregulation of miR-198 in 5-Azacytidine treated SCC131 cells, using RT6-198 and short-miR-198 primers specific to miR-198 ( Supplementary Table S3 ). A significant upregulation in the miR-198 level was observed in 5-Azacytidine treated cells compared with the vehicle (DMSO) treated cells, thus validating the human miRNA microarray data ( Figure 1A ).

DNA hypermethylation is one of the key players for the reduced expression of tumor suppressor miRNAs in cancer. As 5-Azacytidine is a global hypomethylating agent which can reactivate the transcription of tumor suppressor miRNAs, and miR-198 was observed to be upregulated post 5-Azacytidine treatment of SCC131 cells, we hypothesized that miR-198 could be a tumor suppressor miRNA. We, therefore, wanted to explore if overexpression of miR-198 in OSCC cells has an anti-proliferative role in OSCC cells. We then transfected SCC131 and SCC084 cells with pcDNA3-EGFP (vector control) and pmiR-198 separately and performed the trypan blue dye exclusion cell counting assay. The results showed that miR-198 significantly reduced the rate of proliferation of OSCC cells compared with those transfected the vector control, suggesting that it negatively regulates cell proliferation and functions as a tumor suppressor in OSCC cells ( Figure 1B ).

To understand the underlying mechanism for the upregulation of miR-198 in 5-Azacytidine treated SCC131 cells, we hypothesized that the demethylation of the MIR198 promoter can cause upregulation of miR-198 levels. Since the promoter of miR-198 is unknown, we first retrieved the putative MIR198 promoter sequence from the DBTSS, Switchgear and Matinspector databases ( Supplementary Figure S1A ). The 839 bp sequence ( Supplementary Figure S1A ) was cloned in the pGL3-Basic vector. The MIR198 promoter construct (pMIR198-prom; Supplementary Figure S1B ) along with the positive control pGL3-Control and the negative control pGL3-Basic were then transfected separately in SCC131 and SCC084 cells and the dual-luciferase reporter assay was performed. The results showed a significant promoter activity for the MIR198 promoter construct in both SCC131 and SCC084 cells compared with the pGL3-Basic vector ( Supplementary Figure S1C ), suggesting that the putative MIR198 promoter sequence indeed represents an independent promoter for the MIR198 gene.

After characterizing the MIR198 promoter, we analysed the methylation status of the MIR198 promoter, using bisulphite sequencing PCR (BSP). We have designed BSP specific primers ( Supplementary Table S6 ) from nucleotide positions -213 to +190, which harbours 3 CpG sites ( Figure 2A ). Using sodium bisulphite treated genomic DNA from control and 5-Azacytidine treated SCC131 cells as templates, we amplified the MIR198 promoter-specific fragment with BSP-specific primers ( Supplementary Table S6 ), cloned them in a TA cloning vector, and Sanger sequenced 10 random clones for each of the control and 5-Azacytidine treated SCC131 cells. The results showed that the percentage of methylation at the MIR198 promoter region reduced from 76.66% in the control-treated SCC131 cells to 50% in 5-Azacytidine treated cells ( Figure 2B ), suggesting that demethylation of the MIR198 promoter is responsible for upregulation of miR-198 following 5-Azacytidine treatment of SCC131 cells.

MiRNAs are known to regulate gene expression by binding to their cognate mRNA targets. As we have observed miR-198 to have an anti-proliferative role in OSCC cells, we wanted to elucidate its target gene(s) and the mechanism through which it exerts its effect. To this end, we have used a consensus approach by employing four different mRNA target prediction algorithms (e.g., TargetScanHuman, miRDB, PicTar and CoMeTa) to identify gene target(s) of miR-198. We found TOPORS (topoisomerase I binding, arginine/serine-rich, E3 ubiquitin protein ligase), NRIP1 (nuclear receptor interacting protein 1), PBX1 (pre-B-cell leukemia homeobox 1), PDCD1LG2 (programmed cell death 1 ligand 2), SLC2A1 [solute carrier family 2 (facilitated glucose transporter), member 1], PUM2 (pumilio RNA-binding family member 2), MET (met proto-oncogene), VCP (Valosin Containing Protein) and FUT8 [(fucosyltransferase 8 (alpha (1, 6) fucosyltransferase)] as potential gene targets of miR-198 predicted by all the four algorithms ( Supplementary Table S7 ). Among these FUT8, VCP, PUM2, PBX1 and MET are already explored targets of miR-198. A thorough literature search was performed for the remaining targets considering their impact in cancer, known role, implication in pathways etc. TOPORS is known to directly interact and regulate the expression of p53 (34). The role of TP53 in cancer progression is well established. Also, other pathways (MAPK, PI3K/AKT, HGF/MET, JAK/STAT and FGFR1) are already explored for miR-198 (27). However, none of the studies investigated the regulation of p53 after ectopic expression of miR-198. Thus, we decided to pursue TOPORS as a gene target for miR-198.

To determine if miR-198 targets and regulates TOPORS, we transfected pmiR-198 and pcDNA3-EGFP separately in SCC131 cells and assessed the levels of TOPORS transcript and protein ( Figure 3A ). The results showed that miR-198 reduces the levels of TOPORS transcript and protein ( Figure 3A ), suggesting miR-198-mediated regulation of TOPORS. To determine if there is a dose-dependent regulation of TOPORS by miR-198, we transfected different quantities of the pmiR-198 construct in SCC131 cells. The results showed that miR-198 reduces the levels of TOPORS transcript and protein in a dose-dependent manner ( Figure 3B ). We next looked for the effect of 5-Azacytidine treatment on the expression of both TOPORS transcript and protein in SCC131 cells. The results showed a concomitant reduction in the levels of TOPORS transcript and protein with an increased level of miR-198 in 5-Azacytine treated cells compared with the control-treated cells ( Figure 3C ). These observations further strengthen the significance of miR-198-mediated regulation of TOPORS.

The target prediction programs showed one putative target site (TS) in the 3’UTR of TOPORS from nucleotide positions 543-549 for binding of the miR-198 seed region (SD) ( Figure 4A ). The CLUSTALW alignment (https://www.genome.jp/tools-bin/clustalw) showed that this TS is conserved across species ( Figure 4A ). To validate the direct interaction between miR-198 and the 3’UTR of TOPORS, we employed the use of a gold standard assay, the dual-luciferase reporter (DLR) assay. The DLR assay was performed using various constructs illustrated in Figure 4B . Using the site-directed mutagenesis, we generated a negative control construct, pMIR-REPORT-TOPORS-3’UTR-M, by abrogating the TS for miR-198 in the 3’UTR of TOPORS. If miRNA binds to the 3’UTR of the target gene TOPORS, the luciferase activity significantly reduces compared with control transfected cells. Therefore, we co-transfected SCC131 cells with pMIR-REPORT-TOPORS-3’UTR-S containing the intact 3’UTR of TOPORS in a sense orientation and pmiR-198 or pMIR-REPORT-TOPORS-3’UTR-S and pcDNA3-EGFP (vector control) and quantified the luciferase reporter activity. We observed a significant reduction in the luciferase activity in cells co-transfected with pMIR-REPORT-TOPORS-3’UTR-S and pmiR-198 compared with those co-transfected with pMIR-REPORT-TOPORS-3’UTR-S and pcDNA3-EGFP ( Figure 4C ). As expected, cells co-transfected with pmiR-198 and pMIR-REPORT-TOPORS-3’UTR-M showed the luciferase activity comparable to those co-transfected with pMIR-REPORT-TOPORS-3’UTR-S and pcDNA3-EGFP. This is attributed to the lack of binding of miR-198 to the TOPORS 3’UTR, as the TS is abrogated in the pMIR-REPORT-TOPORS-3’UTR-M construct ( Figures 4B, C ). We also analysed the binding of miR-198 to its known gene target FUT8 as a positive control. We observed that the cells co-transfected with pmiR-198 and pMIR-REPORT-FUT8-3’UTR-S showed a significantly reduced luciferase activity compared with those co-transfected with pMIR-REPORT-FUT8-3’UTR-S and pcDNA3-EGFP, confirming the binding of miR-198 to the 3’UTR of FUT8. The above observations confirm that miR-198 binds directly to the TS in the 3’UTR of TOPORS in a sequence-specific manner.

To understand the biological relevance of the interaction between miR-198 and TOPORS, we determined the levels of miR-198 and TOPORS transcripts in 39 matched normal oral tissue and OSCC patient samples, using qRT-PCR ( Figure 5 ). We found miR-198 to be significantly downregulated in 29/39 (74.35%) OSCC samples compared with their matched normal oral tissue samples (viz., patient no. 63, 3, 8, 33, 47, 49, 52, 56, 62, 64, 70, 76, 80, 10, 14, 15, 17, 24, 32, 43, 44, 45, 48, 51, 59, 61, 65, 66 and 67) and significantly upregulated in tumor samples compared with their matched normal oral tissue samples in 5/39 (12.8%) (viz., patient no. 68, 55, 50, 57 and 60) ( Figure 5A ). Moreover, no significant change in its levels was observed in 5/39 (12.8%) OSCC samples compared with their normal tissue counterparts (viz., patient no. 54, 25, 46, 53 and 6). Further, we found TOPORS to be significantly upregulated in 19/39 (48.7%) OSCC samples compared with their matched normal oral tissue samples (viz., patient no. 68, 8, 33, 47, 49, 64, 70, 76, 10, 14, 15, 24, 43, 44, 48, 50, 60, 65 and 66) ( Figure 5B ) and significantly downregulated in 17/39 (43.58%) OSCC samples compared with their matched normal oral tissues (viz., patient no 54, 63, 3, 46, 52, 53, 55, 56, 62, 6, 32, 45, 51, 57, 59, 61 and 67) ( Figure 5 ). Moreover, no significant change was reported in its levels in 3/39 (7.6%) (viz., patient no. 25, 80 and 17) OSCC samples compared with their matched normal tissue samples ( Figure 5B ). In patient numbers 8, 33, 47, 49, 64, 70, 76, 10, 14, 15, 24, 43, 44, 48, 65, and 66 the miR-198 expression is significantly downregulated compared with their matched normal oral tissue samples and concomitantly the TOPORS expression is significantly upregulated compared with their matched normal oral tissue samples. In patient numbers 55 and 57 miR-198 is significantly upregulated compared with their matched normal oral tissue samples and concomitantly TOPORS is significantly downregulated compared with their matched normal oral tissue samples. Overall, an inverse relation of such kind was observed between the levels of miR-198 and TOPORS in 18/39 (46.15%) (viz., patient no. 8, 33, 47, 49, 55, 64, 70, 76, 10, 14, 15, 24, 43, 44, 48, 57, 65 and 66) OSCC patient samples ( Figure 5 ), suggesting the biological relevance of their interaction.

MiRNAs are known to negatively regulate their target genes by binding to their 3’UTRs. As we have observed miR-198 to be a tumor suppressor miRNA, which targets TOPORS, it is interesting to study the role of TOPORS in OSCC cells. To this end, we transfected SCC131 and SCC084 cells with vector control and pTOPORS separately and checked for cell proliferation by trypan blue dye exclusion assay. We observed a significant increase in proliferation of cells transfected with pTOPORS compared with those transfected with the vector control, suggesting that TOPORS exerts a positive regulation on proliferation of OSCC cells from both cell lines ( Supplementary Figure S2 ).

Since miR-198 binds directly to the 3’UTR of TOPORS, we wanted to investigate if the function of TOPORS depends on the presence of its 3’UTR. To this end, we utilized the following different TOPORS overexpression constructs: pTOPORS harbouring a complete TOPORS ORF, pTOPORS-3’UTR-S with the 3’UTR of TOPORS cloned downstream to the TOPORS ORF in the pTOPORS construct in a sense orientation and pTOPORS-3’UTR-M with the abrogated TS in the 3’UTR of TOPORS cloned downstream to the TOPORS ORF in the pTOPORS construct. We then transfected the vector control [pcDNA3.1(+)] only or co-transfected vector control and pmiR-198 or pmiR-198 with pTOPORS, pTOPORS-3’UTR-S or pTOPORS-3’UTR-M in SCC084 and SCC131 cells and performed Western blot analysis to assess TOPORS levels ( Supplementary Figure S3 ). The results showed that the TOPORS levels is reduced in cells co-transfected with the vector and pmiR-198 compared with those transfected with the vector control only, because of miR-198 targeting the endogenous TOPORS ( Supplementary Figure S3 ). Also, an increased level of TOPORS was observed in cells co-transfected with pmiR-198 and pTOPORS as compared to those co-transfected with the vector and pmiR-198 ( Supplementary Figure S3 ). Further, we observed a reduced level of TOPORS in cells co-transfected with pTOPORS-3’UTR-S and pmiR-198 as compared with those co-transfected with pTOPORS and pmiR-198, due to the presence of a functional TS in the 3’UTR of pTOPORS-3’UTR-S ( Supplementary Figure S3 ). Further, an increased level of TOPORS was observed in cells co-transfected with pTOPORS-3’UTR-M and pmiR-198 as compared with those co-transfected with pTOPORS-3’UTR-S and pmiR-198, because of the absence of TS in the 3’UTR of pTOPORS-3’UTR-M ( Supplementary Figure S3 ). These observations suggest that TOPORS expression depends on the presence or absence of its 3’UTR and is, in part, regulated by miR-198.

We then used the above constructs to assess the effect of miR-198 mediated regulation of TOPORS on a few hallmarks of cancer such as cell proliferation, apoptosis, and anchorage-independent growth. To this end, we co-transfected pTOPORS, pTOPORS-3’UTR-S and pTOPORS-3’UTR-M constructs separately with pmiR-198 in SCC131 and SCC084 cells and assessed cell proliferation using the trypan blue dye exclusion assay. As expected, we observed a decreased rate of cell proliferation in cells co-transfected with pmiR-198 and vector compared with those transfected with the vector only ( Figure 6A ), due to targeting of endogenous TOPORS by miR-198. Further, we observed a decreased rate of cell proliferation in cells co-transfected with pmiR-198 and pTOPORS-3’UTR-S compared with those co-transfected with pmiR-198 and pTOPORS, due to the presence of a functional TS in the 3’UTR of pTOPORS-3’UTR-S ( Figure 6A ). As expected, we observed no difference in the rate of cell proliferation in cells co-transfected with pmiR-198 and pTOPORS compared with those co-transfected with pmiR-198 and pTOPORS-3’UTR-M, due to the absence of a functional TS in 3’UTR of pTOPORS-3’UTR-M ( Figure 6A ). A similar observation was made in cells from both cell lines SCC131 and SCC084 ( Figure 6A ). These observations highlighted the negative regulation of cell proliferation by miR-198, in part, via targeting the 3’UTR of TOPORS.

To determine if miR-198 regulates apoptosis by targeting the 3’UTR of TOPORS, we again used the above constructs and performed caspase-3 assay and analysed the percentage of caspase-3 positive cells in SCC131 and SCC084 cells. We observed a significant increase in the rate of apoptosis in cells co-transfected with pmiR-198 and vector control compared with those transfected with vector control only ( Figure 6B ). Also, the percentage of apoptotic cells was significantly higher in cells co-transfected with pTOPORS-3’UTR-S and pmiR-198 compared with those co-transfected with pTOPORS-3’UTR-M or pTOPORS and pmiR-198, due to binding of the miR-198 to the TS in pTOPORS-3’UTR-S ( Figure 6B ). The percentage of caspase-3 positive cells was significantly higher in cells transfected pmiR-198 and pcDNA3.1(+) compared with those co-transfected with pTOPORS and pmiR-198, due to ectopic expression of oncogenic TOPORS-ORF ( Figure 6B ). These observations suggested that miR-198 positively regulates cellular apoptosis in SCC131 and SCC084 cells, in part, by directly targeting the 3’UTR of TOPORS.

To investigate the effect of miR-198-mediated regulation of TOPORS on anchorage-independent growth of OSCC cells, we co-transfected pTOPORS, pTOPORS-3’UTR-S and pTOPORS-3’UTR-M separately with pmiR-198 in SCC084 and SCC131 cells by soft agar colony forming assay ( Figure 7 ). As expected, we observed a reduced number of colonies in cells co-transfected with pmiR-198 and the vector compared with those transfected with the vector only ( Figure 7 ). Further, a reduced number of colonies was observed in cells co-transfected with pmiR-198 and pTOPORS-3’UTR-S compared with those co-transfected with pTOPORS and pmiR-198 ( Figure 7 ), due to binding of miR-198 to the TS in the pTOPORS-3’UTR-S. Further, no difference in the number of colonies was observed in cells co-transfected with pmiR-198 and pTOPORS compared with those co-transfected with pmiR-198 and pTOPORS-3’UTR-M, due to the absence of a functional TS in pTOPORS-3’UTR-M ( Figure 7 ). These results clearly suggested a negative regulation of miR-198 on the anchorage-independent growth, in part, by targeting the 3’UTR of TOPORS.

Since we had observed an anti-proliferative role of miR-198, we hypothesized that the restoration of the expression level of the miR-198 by a synthetic miR-198 mimic could reduce the level of oncogenic TOPORS in OSCC cells, which might have an anti-tumor effect in vivo. To this end, we used in vivo pre-treated OSCC xenograft nude mouse model. We first optimized the dosage of a synthetic miR-198 mimic in SCC131 cells and found 900 nM of the mimic to be an optimum dose to reduce TOPORS at both the transcript and protein levels ( Supplementary Figure S4 ). We then injected equal numbers of pre-transfected SCC131 cells with 900 nM of miR-198 mimic (M900) or 900 nM of mimic control (MC900) separately into the right flanks of female nude mice (6 weeks of age). The mice were monitored for OSCC xenograft (tumor) growth for 40 days. As expected, there was a reduction in both tumor volume and weight in nude mice injected with miR-198 mimic treated cells compared to that of the mock control ( Figure 8 ), suggesting that miR-198 functions as tumor suppressor in OSCC.

TOPORS, an E3 ubiquitin and SUMO ligase, is known to affect the expression of several proteins, including the guardian of the genome, p53 (34–38). The p53 pathway is well known for induction of apoptosis and cell cycle arrest (39). The effect of miR-198 on the p53 signaling has not been explored. We, therefore, wanted to understand if miR-198-mediated regulation of TOPORS has any impact on the p53 pathway in OSCC ( Figure 9A ). To this end, we determined the levels of p53 and its downstream target p21 (WAF1/CIP1/CDKN1A) in OSCC cells transfected with pmiR-198 or pTOPORS, using the Western blot analysis ( Figures 9B, C ). The results showed that the cells transfected with pmiR-198 had increased levels of p53 and p21 compared with those transfected with the vector control in both SCC131 and SCC084 cells ( Figure 9B ). As expected, overexpression of miR-198 led to a reduced level of TOPORS in both SCC131 and SCC084 cells ( Figure 9B ). In contrast, cells transfected with pTOPORS showed reduced levels of p53 and p21 compared with those transfected with the vector control in both SCC131 and SCC084 cells ( Figure 9C ). These results highlighted the significant roles miR-198 and TOPORS in the p53 pathway in OSCC. These observations also indicated that miR-198 enhances signaling of p53/p21, in part, by regulating TOPORS.

Since the p53/p21 pathway is observed to be directly correlated with miR-198-mediated regulation of TOPORS ( Figure 9 ), we wanted to explore the levels of TP53 transcript in the OSCC patient samples which showed an inverse relation of miR-198 and TOPORS levels (viz., patient no. 8, 33, 47, 49, 55, 64, 70, 76, 10, 14, 15, 24, 43, 44, 48, 57, 65 and 66) by qRT-PCR using TP53 specific primers ( Supplementary Table S3 ). The results showed that TP53 was significantly downregulated in 7/18 OSCC samples (viz., patient no. 47, 49, 55, 76, 14, 43 and 66) and upregulated in 5/18 OSCC samples (viz., patient no. 8, 64, 10, 48 and 57) compared with their matched normal tissue samples ( Figure 10 ). Since TP53 is the most mutated gene in cancer, we checked the CDKN1A transcript level using CDK1NA (P21) specific primer ( Supplementary Table S3 ) as mutations in CDKN1A are less frequent and CDKN1A is a direct target and a downstream effector for p53. The results showed a significant downregulation of the CDKN1A level in 9/18 OSCC samples (viz., patient no. 8, 33, 47, 49, 76, 43, 44, 48 and 66) and its upregulation in 4/18 OSCC samples (viz., patient no. 55, 10, 57 and 65) compared to matched normal oral tissue samples ( Figure 10 ). No significant change between their levels was observed in 6/18 OSCC samples for TP53 (viz., patient no. 33, 70, 15, 24, 44 and 65) and 5/18 OSCC samples for CDKN1A (viz., patient no. 64, 70, 14, 15 and 24) compared with matched normal oral samples ( Figure 10 ). We then analyzed the expression pattern of TP53 and CDKN1A with respect to miR-198 expression in these inversely related patient samples. Interestingly, we observed miR-198 levels to be positively correlated with CDKN1A levels in 11/18 OSCC samples compared with their matched normal oral tissue samples (viz., patient no. 8, 33, 47, 49, 55, 76, 43, 44, 48, 57 and 66) ( Figure 10 ). For example, if miR-198 expression is observed to be downregulated in the tumor sample from patient no. 47, the expression of CDKN1A (downstream effector of p53) and TP53 was also downregulated and the expression of TOPORS was upregulated in this tumor sample compared with its matched normal tissue sample ( Figure 10 ). The Pearson correlation analysis indicate an inverse correlation between miR-198 and TOPORS (r = -0.08; p value=0.614) whereas the analysis of miR-198 and CDKN1A (r = 0.422; p value=0.080) indicate a positive correlation ( Supplementary Figure S5 ) but the correlation was not significant. The small sample size, tumor heterogeneity and the presence of other underlying alternate mechanisms are possible reasons. Nonetheless, these results highlighted the biological relevance of this interaction between miR-198 and TOPORS and its role in the p53/p21 pathway.