CAL 27 cells were cultured in Dulbecco’s modified Eagle medium (DMEM, Cytiva, Shanghai, China) supplemented with 10% fetal bovine serum (FBS, Cellmax, Lanzhou, China). The human MORC3 expression plasmid was obtained from VectorBuilder (Guangzhou, China).

The anti-MORC3 (human) and non-specific siRNAs were synthesized by Sangon (Shanghai, China). The anti-MORC3 siRNA sequence was 5′- GUG​AGG​UUG​AAU​UGC​UGG​AAA-3′, and the anti-LINC00880 siRNA sequence was 5′- AUC​UUU​UUA​GCA​UCA​UAG​ACA​CU-3’. The siRNAs were transfected into CAL 27 cells using Lipofectamine 3000 (Invitrogen, Waltham, MA, United States) according to the manufacturer’s instructions.

The tissue microarray purchased from Alenabio company (Xi’an, China) included 48 OSCC tissues and 10 normal oral mucosal tissues (Supplementary Table S1). The tissue microarray was deparaffinized by xylene and then rehydrated using a series of graded ethanol. The microarray was then incubated with Target Retrieval Solution (Dako, Glostrup, Denmark), and endogenous peroxidase was inactivated by incubation with Peroxidase-Blocking Reagent (Dako). The microarray was then incubated with mouse monoclonal anti-MORC3 antibody (Santa Cruz Biotechnology, Dallas, TX, United States) overnight at 4°C, followed by incubation with the FLEX/HRP secondary antibody (Dako) for 30 min. MORC3 expression was visualized after incubation with 3,3′-diaminobenzidine (DAB, Dako) and scanning with a Panoramic MIDI (3D HISTECH, Hungary).

The expression of MORC3 in TCGA cancers was analyzed using the Gene Expression Profiling Interactive Analysis 2 (GEPIA2) online database, with a |log2 (fold-change)| cutoff of 0.2 and p-value of 0.05. The clinical outcomes of head and neck cancer male patients were downloaded from the Human Protein Atlas (HPA) database.

The expression levels of MORC3 and interferon (IFN)-associated genes in single cells of head and neck cancer reported in a previous study (Puram et al., 2017) were downloaded using the CancerSEA database. Cells expressing MORC3 and/or IFN-associated genes were used for correlation analyses using GraphPad Prism 8.

Total RNA was purified from CAL 27 cells treated with anti-MORC3 siRNA or non-specific siRNA. RNA-seq and corresponding bioinformatics analyses were performed by Wuhan Ruixing Biotechnology Co. Ltd. (Wuhan, China). In brief, 1 µg of total RNA was used for RNA-seq library preparation, followed by high-throughput sequencing with the Illumina Novaseq 6,000 system. The alignment was performed with HISAT2 software (version 2.1.0). The expression level of each gene was calculated by fragments per kilobase of transcript per million mapped reads (FPKM), and the differentially expressed genes (DEGs) were identified by DESeq2 software (version 1.34.0) based on the criteria of fold change ≥3/2 or ≤2/3 and p-value <0.05. The function and pathway analyses of the DEGs were performed using Reactome and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment programs. Enriched pathways were further evaluated using gene set enrichment analysis (GSEA) to verify the corresponding pathways based on the expression of all genes.

Total cellular RNA was reverse transcribed using Maxima H Minus cDNA Synthesis Master Mix (Thermo Fisher, United States). The cDNA was amplified using Green Taq Mix (Vazyme Biotech, Nanjing, China) for PCR using the following primers: 5′- ACA​AAT​CAA​CAG​ACG​GCA​ACA​G-3′ and 5′-CAT​AGC​CAG​CAG​TTG​TCC​TAC​G-3′ for MORC3; and 5′- GAA​GGT​GAA​GGT​CGG​AGT​C-3′ and 5′-GAA​GAT​GGT​GAT​GGG​ATT​TC-3′ for GAPDH. The cDNA was used for qRT-PCR using ChamQ Universal SYBR qPCR Master Mix (Vazyme Biotech, Nanjing, China) with specific primers (Supplementary Table S2). The relative expression level of each gene was computed by the 2^−ΔΔCT method, in which GAPDH was used as an internal control for normalization.

Cells were lysed in 2 x sodium dodecyl sulfate (SDS) sampling buffer. Total cellular protein samples were denatured for 3 min at 95°C and separated using 10% SDS-polyacrylamide gel electrophoresis (PAGE), and proteins were then transferred to nitrocellulose membranes. The membranes were blocked with 5% skimmed milk and incubated overnight at 4°C with following antibodies: mouse monoclonal anti-MORC3 antibody (Santa Cruz Biotechnology, United States), rabbit monoclonal anti- interferon induced protein with tetratricopeptide repeats 2 (IFIT2) antibody (abcam, United States), rabbit monoclonal anti-interferon induced transmembrane protein 3 (IFITM3) antibody (Cell Signaling Technology, United States), rabbit monoclonal anti-interferon induced protein with tetratricopeptide repeats 1 (IFIT1) antibody (Cell Signaling Technology, United States), rabbit monoclonal anti-cyclin D1 (CCND1) (ABclonal, China), rabbit monoclonal anti-cyclinD2 (CCND2) (ABclonal, China), rabbit polyclonal anti-JUN (ABclonal, China), rabbit polyclonal anti-interferon regulatory factor 7 (IRF7) (Proteintech, China), rabbit polyclonal anti-PD-L1 antibody (Proteintech, China), rabbit polyclonal anti-DExD/H-Box Helicase 60 (DDX60) antibody (Proteintech, China), and mouse anti-β-actin (Abmart, China). The membranes were then incubated with horseradish peroxidase (HRP)-conjugated anti-mouse or anti-rabbit IgG antibody. Bound antibody was visualized using an enhanced chemiluminescence (ECL) kit (Merck Millipore) and X-ray films. β-actin was used as an internal loading control.

CAL 27 cells were seeded into a 12 well-plate (1 × 10⁶ cells per well). After 2 h, the cells were transfected with anti-MORC3 or non-specific (NS) siRNA using Lipofectamine 3000 (Invitrogen). At Days 2 and 4 after transfection, cells were counted using the trypan blue exclusion method.

CAL 27 cells were treated with anti-MORC3 (human) siRNA and non-specific siRNA for 48 h. The expression of PD-L1 on the surface of CAL 27 cells was analyzed by flow cytometry using an allophycocyanin (APC)-labeled anti-human PD-L1 antibody (eBioscience, United States) and a CytoFlex (Beckman, United States).

Two-group statistical comparisons were performed by a Student’s t-test using GraphPad Prism 8. p < 0.05 was considered statistically significant.

MORC3 mRNA expression was investigated in TCGA cancers. The expression of MORC3 in seven cancer tissues, including breast, kidney, lung, prostate, thyroid, or uterine cancers, was significantly lower compared to the expression in corresponding normal control tissues (Figure 1A). The expression of MORC3 protein was analyzed in a tissue microarray containing 48 OSCC tissues and 10 normal oral mucosal tissues by immunohistochemistry. The expression of MORC3 was significantly lower in the nucleus where MORC3 is normally expressed, but not the cytoplasm, of OSCC cancer cells compared to epithelial cells in normal mucosal tissues (Figures 1B, C). In addition, male patients in TCGA head and neck squamous cell carcinoma (HNSCC) with high MORC3 expression showed significantly longer survival than those with low MORC3 expression (Figure 1D). These results suggested that MORC3 has low expression in cancer and may be associated with favorable prognosis.

To elucidate the potential roles of MORC3 at a genome-wide level, MORC3 was knocked down in the CAL 27 OSCC cell line, and the molecular targets of MORC3 were explored by RNA-seq. In total, 270 significantly differentially expressed genes (DEGs) (171 upregulated and 99 downregulated) were identified after MORC3 knockdown (Figures 2A–C). Among these DEGs, the expression levels of most protein-coding genes increased, whereas the expression levels of most non-coding RNA genes decreased (Figure 2D). Four target genes, namely, DDX60, interferon induced protein 44 like (IFI44L), interferon induced protein with tetratricopeptide repeats 2 (IFIT2), and 2′-5′-oligoadenylate synthetase 1 (OAS1), were randomly selected for verification. In line with the RNA-seq results, qRT-PCR analysis demonstrated that these genes were significantly upregulated after MORC3 knockdown (Figure 2E). In contrast, overexpression of MORC3 significantly decreased the expression levels of these genes (Figures 2F–G; Supplementary Figure S1). Furthermore, the protein expression levels of DDX60 and IFIT2 were also upregulated after MORC3 knockdown but downregulated after MORC3 overexpression (Supplementary Figure S2). These results demonstrated that MORC3 regulates the expression of target genes, including both coding and noncoding genes.

To further understand the potential roles of MORC3, pathway analyses of significant DEGs were performed. Reactome pathway analysis showed that MORC3 may be involved in multiple immune regulatory pathways, especially many IFN-associated pathways, including IFNα/β/γ signaling and anti-viral mechanisms regulated by IFN-stimulated genes and ISG15 (Figure 3A). KEGG pathway enrichment analysis showed that MORC3 may regulate multiple virus-related pathways, including influenza A, hepatitis B, hepatitis C, COVID-19, Epstein-Barr virus, human papillomavirus, herpes simplex virus 1, and human immunodeficiency virus 1 (HIV-1). Importantly, viral carcinogenesis, PD-L1 expression, and the PD-1 checkpoint pathway in cancer were also enriched (Figure 3B). KEGG analysis also identified the involvement of MORC3 in the retinoic acid-inducible gene I (RIG-I)-like receptor signaling pathway, Toll-like receptor signaling pathway, and chemokine signaling pathway. GSEA confirmed that these immune- and IFN-associated pathways were significantly activated after MORC3 knockdown (Figure 3C; Supplementary Figure S3). MORC3 may also regulate cytosolic DNA-sensing pathway by increasing the expression of Z-form nucleic acid-binding protein 1 (ZBP1), a DNA and RNA sensor, as well as DEAD box protein 58 (DDX58, also called RIG-1). The target genes identified in the present study were compared with the target genes identified in the GSE182755 database, which contains data from MORC3-deficient monocytes. In total, 70 targets were shared by both OSCC cells and monocytes, indicating that these genes may be conserved across cell types (Figure 3D). Among these genes, interferon stimulated exonuclease gene 20 (ISG20), PD-L1 (CD274), signal transducer and activator of transcription 1 (STAT1), interferon induced with helicase C domain 1 (IFIH1), IRF7, interferon induced protein with tetratricopeptide repeats 2 (IFIT2), interferon induced protein 35 (IFI35), IFIT1, interferon induced protein 44 (IFI44), interferon induced protein with tetratricopeptide repeats 3 (IFIT3), interferon induced transmembrane protein 1 (IFITM1), interferon alpha inducible protein 27 (IFI27), interferon induced transmembrane protein 3 (IFITM3), and interferon alpha inducible protein 6 (IFI6) were upregulated in MORC3-deficient cells and were closely related to IFN-associated pathways (Figure 3E).

To further elucidate the roles of MORC3 in the expression of IFN-associated genes in HNSC cells, qRT-PCR was used to evaluate the expression of additional target genes associated with IFN in CAL 27 cells. In MORC3-deficient cells, the expression levels of MX dynamin like GTPase 2 (MX2), IFIT1, IRF7, interferon regulatory factor 9 (IRF9), IFITM3, IFITM1, IFI44, and IFIH1 were significantly upregulated compared to control cells (Figure 4A). In MORC3-overexpressing cells, the expression levels of MX2, IFIT1, IRF7, IRF9, IFITM3, IFITM1, and IFI44 were significantly downregulated compared to control cells (Figure 4B). Moreover, the protein expression levels of IRF7, IFIT1, and IFITM3 were upregulated after MORC3 knockdown, but downregulated after MORC3 overexpression (Supplementary Figure S4). The expression correlation between MORC3 and these genes were analyzed at the single-cell level in head and neck squamous cell carcinoma using the GSE103322 database. The expression level of MORC3 was significantly negatively associated with MX2, IFIT1, IRF7, IRF9, IFI44, IFIT2, IFI44L, and IFIH1 in HNSCC cells (Figure 4C). These results indicated that MORC3 represses IFN-associated pathways in OSCC, which may contribute to tumor suppression because weak and persistent IFN responses support tumorigenesis.

The PD-L1 immune checkpoint is a major negative regulator of immune response and is overexpressed in cancer, and the IFNγ-STAT1 pathway mediates the expression of PD-L1 (Lv et al., 2021). RNA-seq demonstrated that MORC3 knockdown significantly increased STAT1 and PD-L1 expression, which was confirmed by qRT-PCR analysis (Figure 5A). Flow cytometry analysis confirmed that MORC3 deficiency significantly increased the levels of PD-L1 protein on the surface of CAL 27 cells (Supplementary Figure S5). In contrast, MORC3 overexpression significantly decreased STAT1 and PD-L1 expression (Figure 5B). In addition, the expression level of MORC3 was significantly negatively associated with STAT1 and PD-L1 at the single-cell level in HNSCC (Figure 5C). These results demonstrated that MORC3 represses the expression of PD-L1 and its upstream activator.

MORC3 targets several lncRNAs, including LINC00880, which is overexpressed in multiple cancers (Feng et al., 2023), and LINC02574, which enhances the innate immune response (Zhang et al., 2023). RNA-seq analysis demonstrated that the expression levels of both lncRNAs increased after MORC3 knockdown, and qRT-PCR analysis confirmed these results (Figure 5D). In contrast, MORC3 overexpression significantly decreased LINC00880 expression (Figure 5E). Silencing LINC00880 by siRNA (Figure 5F) significantly decreased the expression of PD-L1 at both the mRNA (Figure 5G) and protein levels (Supplementary Figure S6). Further, LINC00880 was overexpressed in head and neck cancer (Figure 5H). These findings indicated that MORC3 represses PD-L1 expression via multiple pathways, including through the downregulation of LINC00880.

Although MORC3 is an immune-associated gene, RNA-seq analysis indicated the upregulation of several cellular proliferation-related genes, such as JUN, CCND1, CCND2, and cellular communication network factor 1 (CCN1). To evaluate if MORC3 plays a negative role in cancer cell proliferation, cell growth was analyzed in MORC3-deficient cells. MORC3 knockdown significantly promoted the cell growth of CAL 27 cells (Figure 6A). MORC3 deficiency downregulated the E-cadherin (CDH1) tumor suppressive gene and the keratin 14 (KRT14) epithelial differential marker (Figure 6B) but significantly upregulated the expression of oncogenic JUN, CCND1, CCND2, and CCN1 at the transcriptional level (Figure 6C). IL6 is a proinflammatory gene, but it also promotes tumor cell proliferation. MORC3 knockdown increased IL6 expression (Figure 6C). Western blot analysis confirmed that MORC3 knockdown significantly increased the protein expression levels of JUN, CCND1, and CCND2 (Supplementary Figure S7). These results suggested that MORC3 suppresses cancer cell proliferation.