Overexpression of NK4 gene in TU212 affects migratory activity in laryngeal squamous cell carcinoma

Yixuan Huo¹Wei Zhang²Fan Yang²Wenhua Shao³Guozheng Cong³Shoukai Zhang^4*

• ¹Ningxia Medical University, Yinchuan, China
• ²Lanzhou University, Lanzhou, China
• ³Lanzhou Veterinary Research Institute Chinese Academy of Agricultural Sciences, Lanzhou, China
• ⁴Gansu Provincial Hospital, Lanzhou, China

According to global epidemiological data, laryngeal cancer, a major type of malignant head and neck tumor, accounts for approximately 1% of all malignant tumors (1). Approximately 90% of the pathological types are squamous cell carcinomas (2), and malignant transformation originates from the abnormal proliferation of laryngeal mucosal epithelial cells (3). Despite the multidisciplinary integration of innovative therapies in recent years, such as surgical resection, radiotherapy, sensitization chemotherapy, and PD-1/PD-L1 immune checkpoint inhibitors, the incidence and mortality rates are increasing (4). Laser minimally invasive surgery (TLM) for early-stage patients can effectively remove tumors while preserving laryngeal function as much as possible (5). In surgery combined with radiotherapy (6), chemotherapy, and targeted therapy, the 5-year survival rate for earlystage laryngeal cancer (T1-T2) is approximately 80%, while the 5-year survival rate for locally advanced laryngeal cancer (T3-T4) is approximately 50% (7,8).This clinical dilemma is primarily attributable to the unique molecular and pathological characteristics of laryngeal cancer. LSCC is prone to local infiltration, cervical lymph node metastasis, and chemotherapy resistance (9). To date, no accurate biomarkers have been determined for treatment response in LSCC. Therefore, exploring the molecular regulatory network of invasion and metastasis in LSCC may provide a theoretical basis for the development of new strategies for targeted therapy.Clinical studies have shown that the aberrant activation of the hepatocyte growth factor (HGF)/c-Met signaling pathway is associated with tumorigenesis and progression. Met, a receptor for HGF, is a proto-oncogene product of a heterodimeric tyrosine kinase (10). After binding to HGF, it initiates downstream signaling pathways, such as MAPK, STAT, PI3K, and NF-kB, to regulate cell proliferation, invasion, and migration (11). Serum HGF concentrations in patients with advanced and poorly differentiated laryngeal squamous cell carcinoma are significantly higher than those in patients with early stage and highly differentiated disease (12). Studies have shown that NK4, an HGF antagonist, inhibits HGF/c-Met-induced tumor growth, metastasis, and invasion antagonism ultimately leading to apoptosis. NK4 consists of a hairpin structural domain at the N-terminal end of the a-chain of HGF and four Kringle regions, which is obtained from the cleavage of HGF by proteolytic digestion. NK4, a splice variant of HGF, competes with c-Met for binding and is a specific antagonist of HGF (13). An adenoviral vector carrying a stably expressed NK4 gene was found to inhibit the growth of a bone metastatic tumor model and prolonged the survival of mice (14). Human placental-derived mesenchymal stem cells expressing NK4 (PDMSC-NK4) have been reported to exert antitumor effects on glioblastoma cells (15). Moreover, NK4 inhibits cancer growth in vivo by suppressing IDO expression in tumors and promoting NK cell accumulation (16).Lentiviruses can efficiently deliver target genes into tumor cells and stably integrate them into the host cell genome to achieve long-term and stable gene expression, thereby aiding in the genetic modification of tumor cells and the study of regulatory gene functions. Using a lentiviral packaging system to construct cell lines that stably express genes is a powerful tool for gene therapy and mechanistic research on tumors. For instance, localized STAT3 knockdown in the brain, achieved via lentiviral vectors encoding STAT3 shRNA, was established, and the NLRP3-STAT3 interaction was validated. This model system was then utilized to determine if JAK2 inhibition confers neuroprotection following ischemic stroke (17). Furthemore, previous research has also reported a lentiviral overexpression vector co-expressing the CopGFP and PuroR reporter genes was successfully engineered. Validation in hepatocellular carcinoma (HCC) cells demonstrated its efficacy in promoting the expression of a target gene with an extended coding sequence (18). Lentivirus-mediated gene transfer was employed to establish a stable cell line expressing the avb3 integrin, which exhibited enhanced susceptibility to foot-and-mouth disease virus (FMDV). This model provides a valuable tool for investigating avb3 integrin functions (19). Based on this literature, we used lentiviral packaging technology to construct a TU212 cell line of laryngeal squamous cell carcinoma overexpressing NK4 and verified its biological functions. The aim of this study was to investigate the function and mechanism of action of NK4, providing a theoretical basis for assessing its role in the progression of laryngeal cancer, as well as its potential therapeutic value and its clinical application. The TU212 cell line was purchased from Beina and cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum and 1% antibiotics at 37°C in a 5% CO2 incubator. PLV-NK4-Flag plasmid was constructed by the research team in the earlier stage. Human embryonic kidney cells (HEK-293T), psPAX2 (Addgene, 12260), and pMD2.G (Addgene, 12259) were generously provided by the Lanzhou Veterinary Research Institute of the Chinese Academy of Agricultural Sciences. RPMI 1640 cell culture medium (Gibco) and 0.25% EDTA (Gibco) trypsin were purchased from Gansu Pengcheng Biotechnology Development Co. Ltd. Fetal bovine serum (FBS) was purchased from Guangzhou Lucheng Biotechnology Co. Ltd. Polyplus jet PRIME transfection reagent was purchased from Gene Biotechnology Co. Ltd. Western antibody dilution buffer was purchased from Biotime Biotechnology Co. Ltd. Mouse anti-Flag antibody was purchased from Sigma-Aldrich (Shanghai) Trading Co. Ltd., and the protein prestained marker was purchased from Invitrogen. After reviving HEK-293T cells to the second generation, the cells were found to be in good condition and seeded at a density of 5 × 10 6 cells in a 100-mm culture dish. When the cell density reached approximately 70%, 3 mg of pLV-puro-3×-Flag-JMJD6 plasmid, 2 mg of psPAX2 packaging plasmid (Addgene, 12260), and 1 mg of pMD2.G envelope plasmid (Addgene, 12259) were transfected using jetPRIME transfection reagent. The supernatants containing the lentivirus were collected 24 and 48 h post-transfection, filtered for purification, and the packaged lentivirus was used to infect the TU212 cells. After 48 h, the cells were digested with trypsin, and the medium containing 10 mg/ml of puromycin was added for selection. The medium was changed after 24 h, and the medium containing 5 mg/ml of puromycin was added to continue selection. Finally, the surviving cells were subcloned using the limiting dilution method, and the selected monoclonal cells were expanded and cryopreserved. Cells cultured for five generations were mixed with loading buffer, and 20 ml of the protein sample was used for SDS-PAGE. After electrophoresis, the proteins were transferred onto a nitrocellulose membrane at a constant voltage of 100 V. The membrane was blocked at room temperature for 2 h with 5% non-fat milk solution prepared in TBST followed by overnight incubation at 4 C with the corresponding primary antibody. After washing with TBST, the secondary antibody was added and incubated at room temperature for 2 h. Finally, an ECL chemiluminescent solution was used for fully automated chemiluminescent imaging (Figures 1A,B). Cells that were continuously passaged for 10 generations were subjected to indirect immunofluorescence analysis. The results were observed using a laser confocal microscope (×60) (Figure 1C). Western blot analysis was performed to detect EMT markers (E-cadherin, Snail, MMP9, and Slug) for PLV-NK4-TU212, with bactin as the internal control (Figure 2A), and to show expression of P53 and Bcl-2 genes related to apoptosis (Figure 2B). The cell concentration was adjusted to 2 × 10 4 ml followed by inoculation into a 96-well plate and culturing for 24 h. The original culture medium was removed, the MTT solution was added, and the cells were incubated at a constant temperature for 4 h. The resulting supernatant was discarded before adding 100 ml of DMSO solution to each well. The cultures were shaken thoroughly for 10 min, and then the absorbance was measured at 570 nm. All cultures were continuously monitored for 7 days (Figure 2C). Cells were seeded at a density of 2.5 × 10 6 /mL in a six-well plate and cultured in a complete medium until 100% confluence was reached (Figures 4A,B). A sterile pipette tip was used to create vertical scratches at the bottom of the well followed by washing with PBS. Next, 2% serum-containing RPMI 1640 medium was added to each well, and the plate was placed in a 37°C incubator with 5% CO 2 . Images were captured at 0 and 48 h using an inverted light microscope. The scratch healing rate was calculated as follows: Scratch healing rate = (0 h -48 h)/0 h × 100% (Figure 3). Results are expressed as mean ± standard deviation. All data were analyzed using GraphPad Prism 9 (GraphPad Software, Inc., CA) (Figure 4C). Intergroup differences were assessed using Student's t-test or unpaired two-tailed t-test. Two-way analysis of variance (ANOVA) was used for comparisons among multiple groups. A p-value <0.05 was considered statistically significant. The cells, WT-TU212 and PLV-NK4-TU212, were resuscitated and cultured in a medium containing 10% FBS and 1% dual antibiotics until the third generation. Total RNA was extracted using the TRIzol method, and eukaryotic mRNA was enriched with a poly (A) tail using magnetic beads with Oligo(dT) and then fragmented with a buffer. Using fragmented mRNA as a template and random oligonucleotides as primers, the first strand of cDNA was synthesized using the M-MuLV reverse transcriptase system. Subsequently, the RNA strand was degraded with RNase H, and the filtered to obtain a high-quality database (Figures 4A,B). The shortread alignment tool Bowtie2 (20) was used to align the clean reads to the ribosomal database of the species removing reads aligned to the ribosome without allowing mismatches. The remaining unmapped reads were used for subsequent transcriptome analysis. HISAT2 software was used to perform alignment analysis based on the reference genome. The distribution of reads in the exonic, intronic, and intergenic regions of the reference genome was determined based on the total mapped read alignment results (Figure 4C). Based on the TPM values of each gene, the expression distribution of different sample genes or transcripts are displayed (Figure 4D). Principal component analysis was conducted using R (http://www.r-project.org/) (Figure 4E). The Pearson correlation coefficient between the expression levels of any two samples was calculated, and the correlation between any two samples was displayed in the form of a heatmap to examine the reproducibility among repeated samples within the test group (Figure 4F). The correlation coefficient between pairs of samples was calculated using the FPKM/TPM values of genes in the samples to assess the reproducibility of samples within the group (Figure 4G). Genes were selected based on gene abundance, and a Venn diagram was used to identify shared and unique genes between groups (Figure 4H). DESeq2 software was used to normalize the read count, calculate the probability of hypothesis testing (p-value) according to the model, and finally correct for multiple hypothesis testing to obtain the FDR value (false discovery rate), with the following conditions for difference screening: p-value < 0.05, |log 2 FC| > log 2 (1.5) (Figures 4A,B). Differential genes were mapped to each term of the GO database (http://www.geneontology.org/), and the number of differential genes per term was calculated to obtain a list of differential genes with a certain GO function. The number of differential genes was used to generate statistics of GO entries that were significantly enriched in differential genes. KEGG is the main public database on pathways, which identifies the top biochemical metabolic pathways and signaling pathways in which differential genes are involved through significant pathway enrichment. DO (Disease Ontology) is a database that describes the relationship between gene functions and diseases. We map the differentially expressed genes to the terms in the DO database (http://disease-ontology.org/) and calculate the number of differentially expressed genes for each term resulting in a list of differentially expressed genes associated with a specific DO function and a count of these genes. We then apply hypergeometric testing to identify DO entries that are significantly enriched among the differentially expressed genes compared to the entire background.The Reactome database gathers various reactions and biological pathways from several species. We mapped the differentially expressed genes to the various terms in the Reactome database (https://reactome.org/) and calculated the number of differentially expressed genes for each term resulting in a list of differentially expressed genes with a specific Reactome function and a count of these genes. We then applied hypergeometric testing to identify Reactome entries that are significantly enriched among the differentially expressed genes compared to the entire background. The transcription factor database (http://bioinfo.life.hust.edu.cn/ AnimalTFDB4/#/) was used to annotate genes for transcription factors (TF), to determine whether a gene is a member of a specific transcription factor family, and summarize how many genes are in each TF family.Alternative splicing (AS) is an important gene regulation mechanism in eukaryotes. We used rMATS to detect five types of alternative splicing events as follows: SE, RI, MXE, A5SS, and A3SS. We can perform differential AS analysis on samples with biological replicates. The lentiviral NK4 overexpression cell lines selected under pressure with puromycin were passaged (Figure 1A), and fifthgeneration cell lines were identified by western blotting (Figure 1B), as shown in Figures 1A,B. NK4 was stably expressed in passaged cell lines 3, 4, and 6. The remaining cells were further cultured to the 10th generation, and the expression of the NK4 gene in the cell lines was detected using an indirect immunofluorescence assay employing a mouse-derived anti-Flag tag antibody and an anti-488 fluorescent antibody as the primary and secondary antibodies, respectively, to verify the expression of the NK4 gene in the cell lines. The results showed that NK4 was stably expressed and localized in the nucleus. We performed Western blot analysis of several EMT (epithelial-mesenchymal transition) markers (E-cadherin, Snail, MMP9, and Slug) for cell line. The expression of the proteins Snail, MMP9, and Slug was decreased in PLV-NK4-TU212, which proved that overexpression of the NK4 gene inhibited the process of EMT (Figure 2A). Immunoblotting showed that, compared to the control group, the expression of the apoptosis-related genes P53 and Bcl-2 was upregulated in PLV-NK4-TU212. These results indicate that NK4 gene overexpression induces apoptosis (Figure 2B). The results of the MTT assay showed that the OD value of the PLV-NK4-TU212 group was significantly reduced compared with that of WT-TU212, and the OD value of the PLV-NK4-TU212 group was significantly reduced compared with that of the PLV-NC-TU212 group (Figure 5). It indicates that overexpression of NK4 in laryngeal squamous cell carcinoma TU212 cell line inhibited cell proliferation. The scratch healing experiment tested the migration ability of stable cell lines, and the results showed that, compared to the PLV-NC-TU212 and WT-TU212 groups, the cell migration rate in the PLV-NK4-TU212 group decreased at 48 h suggesting that NK4 overexpression can inhibit the migration ability of TU212 cells. We used RNA-seq to compare the gene expression profiles between PLV-NK4-TU212 and WT-TU212 cell lines, and the results showed that a total of 320 genes had significant differences in expression, of which 189 genes were upregulated and 131 genes were downregulated (Figures 4A,B). GO annotations were enriched in terms of molecular function, cellular component, and biological process. In the biological process, we mainly focused on cellular process, biological regulation, regulation of biological process, metabolic process, and response to stimulus; in the cellular component, we only focused on cellular anatomical entity and protein. anatomical entity and protein-containing complex. In terms of molecular function, binding and catalytic activity were differentially enriched (Figure 4C). KEGG enrichment analysis showed that among the top 20 significantly enriched pathways, they were mainly related to HIF-1, MAPK, and PI3K-AKt signaling pathways (Figure 4D). The significant enrichment of organ system cancer and cell type cancer in the Disease Ontology (DO) enrichment analysis of differentially expressed genes suggests a dual regulatory mechanism of differentially expressed genes in cancer occurrence (Figure 4E). Differentially expressed genes are enriched in the hemostasis pathway, and their abnormalities are closely related to the process of tumor metastasis (Figure 4F). Transcription factor statistics are mainly focused on zf-C2H2 (Figure 6A), and alternative splicing (AS) is an important gene regulatory mechanism in eukaryotes. rMATS is used to detect five types of alternative splicing events (SE, RI, MXE, A5SS, and A3SS), and it can perform differential AS analysis on samples with biological replicates. The analysis results show that exon skipping accounts for the highest proportion (Figure 6B). Laryngeal cancer is a malignant tumor originating in the laryngeal tissues of the larynx (21). Clinical manifestations include progressive hoarseness, a sensation of a foreign body in the throat, difficulty breathing, difficulty swallowing, and enlargement of cervical lymph nodes. Risk factors include long-term smoking, excessive drinking, and human papillomavirus (HPV) infection (22). Treatment for throat cancer emphasizes a balance between organ Transcription factor statistics and variable shear analysis (A-D).function preservation and tumor control (23). However, the hidden nature of the disease often leads to a delayed diagnosis (24). Longterm throat edema caused by radiotherapy affects airway function (25). Although research on gene expression markers (such as TP53 and PIK3CA mutations) and immunotherapy targets (such as PD-L1) provides a direction for precise treatment, unified testing standards for clinical applications are lacking (25). Therefore, research on the utilization of gene expression profiles combined with molecular typing and artificial intelligence data analysis to promote personalized and advanced treatment optimization is of great significance (25,26).HGF/c-MET signaling is involved in the metabolic reprogramming of tumor cells in head and neck squamous cell carcinoma regulating cell proliferation and invasion capabilities and affecting immune surveillance and activation in the tumor microenvironment (27,28). NK4 is an intramolecular fragment of HGF that binds to MET but does not activate receptor signaling (29). Previous studies have shown that NK4 adenovirus (Ad-NK4) effectively inhibits the viability, invasiveness, and tumorigenicity of human mesothelioma cells. Additionally, Ad-NK4 suppressed the characteristics of cancer stem cells (30). Human placenta-derived mesenchymal stem cells (PDMSCs) combined with NK4 inhibit the proliferation ability of glioblastoma cells (15). Heightened HGF expression in head and neck squamous cell carcinomas correlates negatively with patient survival(4). Identification of prognostic genes in the oral squamous cell carcinoma (OSCC) microenvironment revealed that HGF expression significantly correlates with the infiltration levels of B cells, CD4⁺T cells, CD8⁺T cells, macrophages, and neutrophils (31). Blood tests conducted on grouped clinical patients showed that serum HGF concentrations in patients with poorly differentiated tumors, high tumor grades, and advanced clinical stages were significantly higher than those in patients with well-differentiated tumors, low tumor grades, and early clinical stages (p < 0.05) (32). In the present study, we successfully constructed lentiviral vectors carrying FLAG-tagged NK4 overexpression and negative control plasmids. These were successfully packaged into lentiviruses and used to infect the laryngeal squamous cell carcinoma TU212 cell line resulting in the establishment of stable NK4 gene-expressing TU212 cell lines, PLV-NK4-TU212 and PLV-NC-TU212. The stable transfection of the NK4 gene in these cells overcomes the drawbacks of low transfection efficiency and unstable gene expression, thus providing an effective and convenient tool for further research. This study validated the effects of the stable transfected cell lines on migration, proliferation, and apoptosis induction capabilities through MTT assays and immunoblotting to detect MET and apoptotic molecular markers. Further validation of the experimental conclusions by Zhang et al. showed that NK4 gene overexpression can inhibit the proliferation, migration, and invasion abilities of the laryngeal squamous cell carcinoma cell line AMC-HN-8 inducing its apoptosis. NK4 not only negatively regulates the HGF/c-Met signaling pathway but also inhibits tumor angiogenesis independently of the HGF/c-Met pathway. Studies show that NK4 inhibits VEGF-induced angiogenesis by suppressing the phosphorylation of ERK and ETS-1 in cultured endothelial cells and in an in vivo rabbit model (33), and the anti-angiogenic effect of NK4 is independent of c-Met. When NK4 is combined with globin, it inhibits the extracellular assembly of fibronectin, thereby suppressing the spread of fibronectin-dependent endothelial cells (34). NK4 inhibits the paracrine loop of HGF, indirectly suppressing the expression of VEGF in tumor cells, thereby exerting an antiangiogenic effect (35). RNA sequencing is an important technique for analyzing differential gene expression in tumors. By analyzing the expression of differential genes, key genes related to specific biological processes can be identified. Total RNA was extracted from PLV-NK4-TU212 and PLV-NC-TU212 cells followed by library construction and sequencing using an Illumina platform. The sequencing results revealed that NK4 caused significant differences in the expression levels of 320 genes, with 189 upregulated and 131 downregulated genes (p < 0.05). Gene Ontology enrichment analysis showed that the distribution of differential genes was primarily focused on biological processes, while Kyoto Encyclopedia of Genes and Genomes pathway enrichment indicated that the pathways involved were mainly related to the HIF-1, MAPK, and PI3K-Akt signaling pathways. This suggests that the expression of the NK4 protein triggered a strong cellular response.The results obtained from experiments on laryngeal cancer cell lines in this study are consistent with those of previous reports on the anticancer effects of NK4 cells. This suggests that NK4 could serve as a potential anticancer gene and play an important role in counteracting the growth, invasion, and metastasis of laryngeal cancers.Due to time constraints, this study did not conduct in vivo experimental validation or more in-depth pathway screening and mechanism verification.