Establishment of cell model for evaluating cell migration-related functions

A head and neck squamous carcinoma cell model was established to expose cells to nicotine using the CCK-8 assay to evaluate the proliferation capabilities of Cal27, Fadu, HN6, and Tu686 cells after nicotine exposure. Cell scratch assays and Transwell experiments were conducted to quantitatively assess the migration and invasion capabilities of cells following nicotine exposure and CHRNA5 knockdown. Western blot was used to quantify the expression levels of epithelial-mesenchymal transition (EMT)-related markers in cells exposed to nicotine following CHRNA5 knockdown. The expression of nAChR mRNA and protein was detected using qRT-PCR and western blot. Immunofluorescence staining was used to examine the expression levels of CHRNA5 and Ki-67 post-CHRNA5 knockdown, as well as the subcellular localization of CHRNA5 in tumor cells. Clonogenic assays were used to evaluate the clonogenic dependency and proliferation capabilities of cells post-CHRNA5 knockdown. Using sh-CHRNA5 Cal27 and Tu686 cell lines exposed to 10μmol/L nicotine, samples were collected for RNA sequencing. qRT-PCR and western blot validated the transcription of CES1 mRNA and its protein translation post-CHRNA5 knockdown. The interaction between CHRNA5 and CES1 was investigated using molecular docking and molecular dynamics simulations, along with co-immunoprecipitation experiments and immunofluorescence colocalization. Differentially expressed genes were subjected to pathway enrichment analysis, and western blot experiments validated the expression of MEK/ERK proteins and their phosphorylation levels. The antibodies used were MEK/p-MEK (CST, 8727T/9154T) and ERK/p-ERK (CST, 4695T/4370T) for MEK and ERK proteins, respectively. For CES1, antibodies from Abcam (ab53008), Proteintech (67079-1-Ig), and ABclonal (A11478) were used. CHRNA5 antibodies were obtained from Abcam (ab259859) and Proteintech (66363-1-Ig).

Molecular docking and molecular dynamics simulation

Protein Docking: The amino acid sequences of CHRNA5 and CES1 were retrieved from the UniProt protein database (https://www.uniprot.org/). Protein structure files for CHRNA5 (P30532) and CES1 (P23141) were obtained from the Alphafold database (https://alphafold.ebi.ac.uk/). Homology modeling based on amino acid sequences was performed using the I-TASSER database (https://seq2fun.dcmb.med.umich.edu//I-TASSER/). The GRAMM-X online tool (https://gramm.x3dna.org/) was used to dock CHRNA5 (3742 atoms) as the receptor and CES1 (4408 atoms) as the ligand. Protein-protein interactions were explored and visually analyzed using Pymol (version 2.4) and PDBePISA (https://www.ebi.ac.uk/pdbe/pisa/). Small-molecule docking: The SDF file of the MEK/ERK pathway inhibitor PD98059 was downloaded from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/). AutoDock software was used for the dehydration and hydrogenation of small molecules as well as for identifying rotatable bonds. Similarly, it processed the CHRNA5 protein structure to prepare it for docking with the small molecules, PD98059 and CHRNA5.

Molecular dynamics simulations were performed using Gromacs version 2022.3. Preprocessing of the small molecule included the addition of the GAFF force field using AmberTools22 and the hydrogenation and RESP potential calculation of the small molecule using Gaussian 16W. After the simulation, the trajectory data were thoroughly analyzed using the built-in analysis tools of the simulation software. This analysis included the calculation of root mean square deviation (RMSD), root mean square fluctuation (RMSF), protein gyration radius (Rg), and solvent-accessible surface area (SASA). These calculated results provide important foundational data for subsequent detailed analysis.

Nude mouse tumorigenesis experiment

Initially, 5–6-week-old nude mice were selected for this study. The experiment proceeded with the preparation of Cal27 cells, which were stably transfected with sh-CHRNA5 lentivirus and a control group; these cells were revived, cultured, and resuspended at a density of 5×10⁷ cells/mL. Subsequently, the mice were randomly divided into two experiments: Experiment 1 comprising the NC (negative control) and SH (sh-CHRNA5) groups, and Experiment 2, which included the SH group and the SH+NIC (SH + nicotine) group, with eight mice in each experiment, distributed into subgroups of 4. For the injection phase, a 100μL cell suspension was subcutaneously injected into the right axilla of each mouse.

In the post-injection care phase, the SH+NIC group mice were exposed to smoke twice daily, with no less than 4h between sessions, following a regimen mimicking the acute model of Chronic Obstructive Pulmonary Disease (COPD). This involved at least 5 days a week, with each session lasting 1h under smoke conditions that adhered to ISO/CIR standards, employing no less than four cigarettes per animal per session, ensuring a smoke concentration greater than 350mg/m³. To potentially shorten the smoking modeling period, lipopolysaccharide (LPS) was injected on days 15th and 45th days. The total smoking duration was extended to 35 days.

Tumor development was closely monitored throughout the experiment. Starting on day 7 post-injection, the length and width of each tumor were measured using a caliper, and measurements were recorded every subsequent 7 days. On the 35th day, the mice were euthanized by cervical dislocation, after which the tumor length and width were measured to calculate the volume (mm³=1/2LW²). Finally, the tumors were excised, weighed, and photographed for further analysis.

Paraffin embedding and immunohistochemistry

Paraffin blocks, including those from nude mice tumorigenesis experiments (NC, SH, and SH+NIC groups) and HNSC tissues (laryngocarcinoma and hypopharyngeal carcinoma and their controls), were sectioned into 4–5μm thick slices and mounted on slides. This setup also encompasses laryngocarcinoma and normal tissue combination microarray chips, featuring 29 squamous carcinoma cases and 10 normal tissues, with two-point sampling for one case. In the following section, the process includes drying, deparaffinization, rehydration, and then the immunohistochemistry phase. This phase starts with heat-induced antigen retrieval using sodium citrate buffer, followed by inactivation of endogenous enzymes, serum blocking to prevent non-specific binding, primary and secondary antibody incubations, and DAB staining for color development. After staining, sections were dehydrated, cleared, and sealed with neutral gum for preservation. Slide mounting, photography, and data quantification: The slides were sealed with neutral gum, observed, and photographed under a microscope. Staining quantification was performed using ImageJ software with the help of the IHC Profiler plugin to implement the Histochemistry Score (H-score). The H-score was calculated as (percentage of weak staining×1)+(percentage of moderate staining×2)+(percentage of strong staining×3), with higher scores indicating stronger overall positive intensity.

Nicotine exposure promotes the transcription and translation of the nicotinic acetylcholine receptor (nAChR) gene CHRNA5

Using qRT-PCR, we systematically investigated the specific impact of nicotine exposure on the transcription levels of nicotinic acetylcholine receptor (nAChR) family genes. We examined the mRNA expression of 15 nAChR genes (including CHRNA1, CHRNA2, CHRNA3, CHRNA4, CHRNA5, CHRNA6, CHRNA7, CHRNA9, CHRNA10, CHRNB1, CHRNB2, CHRNB3, CHRND, CHRNE, CHRNG) in Cal27, Fadu, HN6, and Tu686 cells after exposure to 10μmol/L nicotine. We found that in Cal27, HN6, and Tu686 cells, the expression of the CHRNA5 gene was significantly upregulated (p<0.05), whereas in FaDu cells, although there was an increase in expression, the difference was not statistically significant (Fig. ​(Fig.2A2A).

Open in a separate window

Fig. 2

Nicotine exposure promotes the transcription and translation of CHRNA5.

A 10μmol/L nicotine significantly increases the mRNA expression levels of CHRNA5 in Cal27, Fadu, HN6, and Tu686 cells; B 10μmol/L nicotine significantly increases the protein expression levels of CHRNA5 in the cells; C Expression of CHRNA5 protein in the cells exposed to 10μmol/L nicotine observed under laser confocal microscopy. CHRNA5 protein is primarily expressed in the cytoplasm and cell membrane, with nicotine significantly increasing the protein expression levels of CHRNA5 in the cells; D Quantitative analysis of CHRNA5 protein expression in the cells exposed to 10μmol/L nicotine under laser confocal microscopy. Nicotine significantly increases the protein expression levels of CHRNA5 in the cells. The quantitative representation of average fluorescence intensity in the 3D graph is depicted using a rainbow color spectrum, where red indicates the lowest expression and purple indicates the highest expression. (*p<0.05, ** p<0.01, *** p<0.001).

Additionally, through western blot experiments, we investigated the translational level of nAChR genes. The results indicated a significant increase in CHRNA5 protein expression in Cal27, Fadu, HN6, and Tu686 cells after nicotine exposure (Fig. ​(Fig.2B).2B). Moreover, using an inverted fluorescence microscope and a laser scanning confocal microscope to observe immunofluorescence staining, we found that nicotine exposure significantly elevated the expression of CHRNA5 in head and neck squamous carcinoma cells. We also noted that CHRNA5 was primarily distributed in the cell membrane and cytoplasm (p<0.05; Fig. 2C/D).

Nicotine exposure reversal through knockdown of CHRNA5 reduces migration and invasion of head and neck squamous carcinoma cells, EMT, and single-cell proliferative capacity

To investigate the role of CHRNA5 in the migration and invasion of head and neck squamous carcinoma cells, we successfully constructed sh-CHRNA5 stable transfectants in Cal27 and Tu686 cells. Initially, we performed in vitro wound-healing assays to analyze the relative migration distance of the cells under serum-free conditions for 48h. The results indicated that in both Cal27 and Tu686 cells, the migration capability towards the scratched area was reduced in the sh-CHRNA5 group compared to that in the sh-NC group (p<0.01). However, after nicotine exposure, the migration capability was restored and slightly enhanced, significantly improving cell “healing” ability (p<0.0001) (Fig. ​(Fig.3A).3A). Subsequently, through Transwell migration and invasion assays, we systematically explored the impact of CHRNA5 on the migration and invasion abilities of head and neck squamous carcinoma cells. In the Transwell assays, compared to the sh-NC group, the sh-CHRNA5 group showed significantly reduced migration and invasion abilities of Cal27 and Tu686 cells (p<0.01). Notably, after nicotine exposure, the cell migration and invasion abilities were significantly restored (p<0.001) (Fig. ​(Fig.3B).3B). Western blot analyses were conducted to examine the changes in the expression of the EMT-related proteins E-cadherin and N-cadherin after CHRNA5 knockdown. The results showed that, compared to the sh-NC group, the expression of E-cadherin was upregulated in the sh-CHRNA5 group (p<0.05); after nicotine exposure in the sh-CHRNA5 group, the expression of E-cadherin showed a downward trend, although it was not statistically significant. In Tu686 cells, the expression of N-cadherin was significantly decreased in the sh-CHRNA5 group, while in Cal27 cells, it showed a downward trend without statistical significance. In both cell types, after nicotine exposure in the sh-CHRNA5 group, the expression of N-cadherin showed an upward trend, but it was not statistically significant (Fig. ​(Fig.3C3C).

Open in a separate window

Fig. 3

Nicotine exposure reversal through knockdown of CHRNA5 reduces migration and invasion of HNSC cells, EMT, and single-cell proliferative capacity.

A Knockdown of CHRNA5 reduces scratch migration in head and neck squamous cell carcinoma cells, while nicotine exposure re-promotes migration in Cal27 and Tu686 cells; B Knockdown of CHRNA5 reduces migration and invasion capabilities in Cal27 and Tu686 cells, while nicotine exposure re-promotes migration and invasion; C Knockdown of CHRNA5 reduces EMT ability in Cal27 and Tu686 cells, while nicotine exposure re-promotes EMT ability; D Knockdown of CHRNA5 reduces single-cell proliferation in Cal27 and Tu686 cells, while nicotine exposure re-promotes clonogenicity; E Knockdown of CHRNA5 reduces Ki-67 protein expression in Cal27 and Tu686 cells, while nicotine exposure re-promotes Ki-67 expression. The quantitative representation of average fluorescence intensity in the 3D graph is depicted using a rainbow color spectrum, where red indicates the lowest expression and purple indicates the highest expression. (* p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001).

To investigate the impact of CHRNA5 on the clonogenic dependency and proliferative capability of individual head and neck squamous carcinoma cell populations (cell clusters formed through continuous in vitro proliferation over six generations or more), clonogenic assays were conducted. The results demonstrated a significant reduction in the clonogenic dependency and proliferative ability of individual cell populations in the sh-CHRNA5 group compared with the sh-NC group in both Cal27 and Tu686 cells, as evidenced by a decrease in the clone formation rate (p<0.01). Notably, nicotine exposure restored the clonogenic dependency and proliferative capacity of these cells, as indicated by an increase in the clone formation rate (p<0.05) (Fig. ​(Fig.3D).3D). Ki67, a nuclear protein associated with cell proliferation, is primarily located in the nucleolar periphery during the interphase of cell division and recruited to condensed chromosomes during mitosis. Typically, Ki67 is only expressed in proliferating cells. The expression level of this protein effectively reflects the proliferative status of cells. Therefore, immunofluorescence assays were performed on Cal27 and Tu686 cells to assess the abundance of the Ki-67 protein. The results showed that the sh-CHRNA5 group exhibited a significant reduction in Ki-67 protein expression in both Cal27 and Tu686 cells (p<0.05). Following nicotine exposure, a significant increase in Ki-67 protein expression was observed (p<0.01) (Fig. ​(Fig.3E3E).

Correspondingly, we successfully established an OE-CHRNA5 stable transfection cell line in Fadu cells. First, an in vitro wound healing assay was performed, and the results showed that, compared to the OE-NC group, the migration ability of Fadu cells toward the scratch area was significantly enhanced in the OE-CHRNA5 group (p<0.01) (Fig. ​(Fig.4A).4A). Subsequently, the Transwell assay indicated that the migration and invasion abilities of Fadu cells were increased in the OE-CHRNA5 group (p<0.05) (Fig. ​(Fig.4B).4B). We also analyzed the changes in the expression of EMT-related proteins E-cadherin and N-cadherin after CHRNA5 overexpression by Western blot. The results showed that in the OE-CHRNA5 group, E-cadherin expression was downregulated (p<0.05), while N-cadherin expression was upregulated (p<0.05) (Fig. ​(Fig.4C).4C). Next, we conducted a colony formation assay, and the results showed that in the OE-CHRNA5 group, the clonogenicity and proliferation ability of individual cell populations in Fadu cells increased, as evidenced by the elevated colony formation rate (p<0.05) (Fig. ​(Fig.4D).4D). Finally, an immunofluorescence assay was performed to assess the abundance of the Ki-67 protein. The results showed that Ki-67 protein expression was significantly elevated in the OE-CHRNA5 group (p<0.01) (Fig. ​(Fig.4E4E).

Open in a separate window

Fig. 4

Overexpression of CHRNA5 increases the migration and invasion, EMT and single-cell proliferative capacity of Fadu cells.

A Overexpression of CHRNA5 increases scratch migration of Fadu cells; B Overexpression of CHRNA5 increases the migration and invasion of Fadu cells; C Overexpression of CHRNA5 increases the EMT ability of Fadu cells; D Overexpression of CHRNA5 increases the single-cell proliferative capacity of Fadu cells; E Overexpression of CHRNA5 increases the Ki-67 protein expression of Fadu cells. The quantitative representation of average fluorescence intensity in the 3D graph is depicted using a rainbow color spectrum, where red indicates the lowest expression and purple indicates the highest expression. (* p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001).

Transcriptomics reveals differential gene expression after CHRNA5 knockdown and Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis of differential genes

RNA integrity number (RIN) represents a numerical assessment of RNA integrity, ranging from 1 to 10, as provided by the Agilent Bioanalyzer for total RNA. A lower value indicates more significant degradation, with 10 representing the highest integrity. In this study, all nine samples (TU-NC-01, TU-SH-01, TU-NIC-01, TU-NC-02, TU-SH-02, TU-NIC-02, CA-NC-03, CA-SH-03, CA-NIC-03) had RIN values of 10, indicating high integrity and classifying them as Category A samples suitable for subsequent experiments. Reference-based transcriptome sequencing was performed for all nine samples, with each sample revealing gene expression data for more than 17,806 genes. After analyzing the differential gene expression data between the NC (control) and SH (sh-CHRNA5) groups, a total of 300 upregulated genes and 832 downregulated genes were identified. Among these genes, the expression level of the CES1 gene was significantly lower in the SH group, with statistically significant differences (p<0.01) (Fig. ​(Fig.5A).5A). Further cluster analysis of differential gene expression levels between the NC and SH groups showed that the expression level of CES1 was lower in the SH group (Fig. ​(Fig.5B5B).

Open in a separate window

Fig. 5

Transcriptomic sequencing with participant inclusion.

A Volcano plot illustrating differential gene expression. Gray indicates genes with non-significant differential expression, while red and blue denote genes with significant differential expression. B Heatmap of differentially expressed genes. Red signifies relatively high expression of protein-coding genes, whereas blue represents relatively low expression of protein-coding genes. Enrichment analysis. C Differential gene enrichment analysis using Gene Ontology (GO) (NC-SH). D Enrichment analysis of differential genes using Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways (NC-SH). The horizontal axis represents the Enrichment Score, while the vertical axis represents the categories. Bubble size reflects the number of differentially expressed protein-coding genes, and color changes from blue to red indicate the significance of enrichment p-values, with larger bubbles being more significant. E Enrichment analysis of differential genes using Reactome pathways (NC-SH). F Enrichment analysis of differential genes using WikiPathways (SH-NIC).

In transcriptomics, Gene Ontology (GO) enrichment analysis for the NC and SH groups revealed the regulation of Biological Processes such as cell migration, cell adhesion, and positive regulation of cell migration (Fig. ​(Fig.5C).5C). Concurrently, Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis indicated that these regulations are associated with pathways, such as the MAPK signaling pathway (Fig. ​(Fig.5D).5D). Additionally, Reactome enrichment analysis for the NC and SH groups showed associations with post-translational protein phosphorylation (Fig. ​(Fig.5E),5E), while differential gene analysis between the SH and NIC groups in WikiPathways enrichment revealed relevance to FBXL10 enhancement of MAP/ERK signaling in diffuse large B-cell lymphoma (Fig. ​(Fig.5F5F).

Molecular docking and co-immunoprecipitation analysis between CHRNA5 and CES1

A rigid protein-protein docking study was conducted between CHRNA5 and CES1 using the GRAMM-X web server to perform free molecular docking with the receptor CHRNA5 (P30532) and ligand CES1 (P23141) data, which were homology-modeled using the I-TASSER database. After 3000 rounds of free matching, the best structure was obtained with an energy of −586 units. Pymol software was utilized to assemble the docking structure into a visualized complex, and specific hydrogen bonding sites compatible with docking in the PDBePISA database are shown in Table ​Table1.1. The docking surface area between the two structures is 2049.3Ų, with a binding energy of −26.3kcal/mol. As shown in Fig. ​Fig.6A,6A, CHRNA5 and CES1 formed hydrogen bonds through amino acid residue sites (ILE 259-ASN 211, CYS 283-LYS 498, TYR 273-ASP 126, and THR 404-PHE 6).

Open in a separate window

Fig. 6

Gromacs molecular dynamics simulations of the CHRNA5-CES1 protein docking model and its mutant model.

A Illustration of the original protein docking model, where CHRNA5 is shown in blue and CES1 in yellow, with hydrogen bond interactions represented by dashed lines. Free energy 2D and 3D landscapes of the original (B, D) and mutant (C, E) models, where the x-axis represents root-mean-square deviation (RMSD) and the y-axis represents the protein radius of gyration (Gyrate), both reflecting the protein energy states. The color spectrum indicates energy levels, with blue representing lower energy and red representing higher energy. A lower energy state indicates more stable protein binding, while the increased red energy area in the mutant model signifies higher instability. F Changes in hydrogen bonds between the original docking model and the mutant model. It demonstrates that the number of hydrogen bonds between proteins decreases in the mutant model (red), confirming the successful construction of the mutant model. G RMSD of the original docking model, with CHRNA5 in black, CES1 in red, and the overall docking model in blue, indicating that post-docking, the energy fluctuation range is small and remains in a low-energy state, suggesting stable binding. H RMSD changes of the original (black) and mutant (red) models, showing that the mutant model has greater energy fluctuation and remains in a high-energy state, indicating increased instability. I Gyrate changes in the original (black) and mutant (red) models, demonstrating that the mutant model exhibits greater energy fluctuation and remains in a high-energy state, signifying increased instability. J Solvent Accessible Surface Area (SASA) of the mutant model (red) fluctuates more widely, indicating increased instability. K Compared to the root mean square fluctuation (RMSF) of CHRNA5 and CES1 in the original docking model, the RMSF fluctuation range of both proteins increases in the mutant model, further indicating increased instability.

To further validate the original protein docking model, we mutated the predicted four amino acid sites to Alanine, which lacks a side chain, to form a mutant docking model [24, 25]. As shown in Fig. ​Fig.6B/C/D/E,6B/C/D/E, the 2D and 3D free energy landscape plots display the root-mean-square deviation (RMSD) on the x-axis and the radius of gyration (Gyrate) on the y-axis, both reflecting the protein energy states. The rainbow color scale represents energy levels, with blue indicating lower energy and red indicating higher energy. A lower energy state reflects more stable protein binding, while an increase in red areas in the mutant model indicates higher instability. As illustrated in Fig. ​Fig.6F,6F, compared to the original docking model, the number of hydrogen bonds between proteins in the mutant model is significantly reduced, confirming the successful construction of the mutant model. During the docking process between CHRNA5 and CES1, the RMSD values remained stable, and the combined RMSD was lower, indicating a smaller energy fluctuation range and a more stable binding in the original docking model (Fig. ​(Fig.6G).6G). However, in the mutant model, the RMSD fluctuation range was larger and remained in a high-energy state, indicating increased instability (Fig. ​(Fig.6H).6H). Similarly, the Gyrate fluctuation range in the mutant model was larger and remained in a high-energy state, indicating greater instability (Fig. ​(Fig.6I).6I). The solvent-accessible surface area (SASA) also exhibited greater fluctuation in the mutant model, further indicating increased instability (Fig. ​(Fig.6J).6J). Compared to the root-mean-square fluctuation (RMSF) of CHRNA5 and CES1 in the original docking model, the RMSF fluctuation range of both proteins was larger in the mutant model, suggesting increased instability (Fig. 6K, L). In conclusion, CHRNA5 and CES1 form a stable protein-protein docking model, but mutating their binding sites significantly reduces stability.

To further confirm the interaction between CHRNA5 and CES1 proteins, a cell model overexpressing Flag-CHRNA5 was first constructed in Cal27 and Tu686 cell lines, and the specific results are shown in Fig. 7A–C. Results from qPCR and western blot experiments indicated that, compared to the control group, both mRNA and protein expression levels of CHRNA5 were significantly increased in these two cell lines (p<0.05), demonstrating the successful construction of the cell model overexpressing Flag-CHRNA5. Subsequently, co-immunoprecipitation experiments were conducted in which CHRNA5 was tagged with FLAG. This allowed for the detection of CHRNA5 by identifying the FLAG tag and directly detecting CHRNA5. From the bands in the input group, it is clear that protein samples from both Cal27 and Tu686 cells contained CHRNA5 and CES1. In the IP IgG group, the bands were essentially not displayed, indicating that the results of the negative control group were normal. After overexpressing Flag-CHRNA5 in Cal27 and Tu686 cells, IP was performed using magnetic beads conjugated with anti-Flag antibody. Western blot analysis using antibodies against FLAG and CES1 showed bands at the positions of CHRNA5 and CES1, explicitly indicating an interaction between CHRNA5 and CES1 (Fig. 7D, E).

Open in a separate window

Fig. 7

Immunoprecipitation of CHRNA5-CES1.

A mRNA level validation after overexpression of CHRNA5. B Protein level validation after overexpression of CHRNA5. C Quantitative analysis of protein levels after overexpression of CHRNA5. D Immunoprecipitation validation of CHRNA5 and CES1 in Cal27 cells. E Immunoprecipitation validation of CHRNA5 and CES1 in Tu686 cells. F Confocal microscopy of triple-stained CHRNA5 (green), CES1 (red), and DAPI (blue) in Cal27 cells; G Scatter plot depicting the colocalization of CHRNA5 and CES1; H Line graph illustrating the colocalization of CHRNA5 and CES1.

Immunofluorescence colocalization analysis of CHRNA5 and CES1 was also conducted, as shown in Fig. 7F–H, with triple staining in Cal27 cells: CHRNA5 (green), CES1 (red), and DAPI (blue). Quantitative analysis of the data revealed a Pearson’s correlation coefficient of 0.8242 and a Manders’ overlap coefficient of 0.9068 for the co-precipitation of CHRNA5 and CES1, indicating a high degree of colocalization and a high level of authenticity in the relationship between the two channels. These analytical results support a deeper understanding of the multifaceted relationship between CHRNA5 and CES1, revealing their complex and diverse associations from multiple perspectives, including physical, co-expression, and genetic interactions.

Verification at the cellular level of the mechanism by which CHRNA5 regulates CES1 expression through the MEK/ERK pathway

The enrichment analysis shown in Fig. ​Fig.55 revealed that the knockdown of CHRNA5 might be associated with the MEK/ERK pathway. To further verify this finding, western blot experiments were conducted in the Cal27 and Tu686 cell lines. The MEK/ERK pathway inhibitor PD98059 was chosen as a tool for the experiments. PD98059 is an effective and selective inhibitor of MEK with an IC50 of 5μM. It functions by binding to the inactive form of MEK, preventing the activation of MEK1 and MEK2 by upstream kinases, and acts as an inhibitor of the ERK1/2 signaling pathway. As shown in Supplementary Fig. S1, the effect of the small molecule PD98059 on the protein structures of CHRNA5 and CES1 was analyzed. It was found that out of 100 docking attempts between CHRNA5 and CES1, optimal docking resulted in only one hydrogen bond formation (THR 130 and SER 305), indicating weak binding and minimal effect.

The experimental design was divided into five groups: NC (control group), SH (knockdown group), SH+IN (knockdown + inhibitor group), SH+NIC (knockdown + nicotine group), and SH+NIC+IN (knockdown + nicotine + inhibitor group). The results showed that activated MEK (p-MEK/MEK) significantly decreased with the knockdown of CHRNA5 (all p<0.0001). PD98059 was able to further reduce p-MEK/MEK in the SH group (Cal27, p<0.001), whereas nicotine was able to reverse the reduction of p-MEK/MEK in the SH group (Tu686, p<0.05). PD98059 also inhibited the nicotine-induced increase in p-MEK/MEK (Cal27, p<0.01; Tu686, p<0.05). Similarly, activated ERK (p-ERK/ERK) significantly decreased with the knockdown of CHRNA5 (all p<0.0001). PD98059 was able to further reduce p-ERK/ERK compared to the SH group (Cal27, p<0.01), but nicotine did not significantly reverse the reduction of p-ERK/ERK in the SH group. PD98059 inhibited nicotine-induced increase in p-ERK/ERK (all p<0.05). Additionally, CES1 decreased with the knockdown of CHRNA5 (Cal27, p<0.05), and PD98059 was able to further reduce CES1 on the basis of the SH group (Tu686, p<0.05), whereas nicotine did not significantly reverse the reduction of CES1 in the SH group. PD98059 inhibited nicotine-induced increase in CES1 (Cal27, p<0.05) (Fig. 8A–C).

Open in a separate window

Fig. 8

Cellular knockdown of CHRNA5 influences CES1 expression through the MEK/ERK pathway.

A Cal27 cells. B Tu686 cells. C Quantitative analysis of p-MEK/MEK, p-ERK/ERK, and CES1/GAPDH ratio. Cellular overexpression of CHRNA5 affects CES1 expression through the MEK/MEK pathway. D Cal27 cells. E Tu686 cells. F Quantitative analysis of p-MEK/MEK, p-ERK/ERK, and CES1/GAPDH ratio.

The following experiments utilized a cell model constructed to overexpress CHRNA5 for further validation, which was divided into five groups: NC group (control group), NC+IN group (control + inhibitor group), OE group (overexpression group), OE+IN group (overexpression + inhibitor group), and OE+IN+NIC group (overexpression + inhibitor + nicotine group). The experiments demonstrated that the activation of MEK (p-MEK/MEK) increased with the overexpression of CHRNA5 (Cal27, p<0.001; Tu686, p<0.05). PD98059 reduced p-MEK/MEK in the NC group (Tu686, p<0.05) and reduced p-MEK/MEK in the OE group (Tu686, p<0.05). Nicotine reversed the reduction in p-MEK/MEK levels in the OE+IN group, showing a trend but the difference was not statistically significant. Similarly, the activation of ERK (p-ERK/ERK) increased with the overexpression of CHRNA5 (all p<0.05). PD98059 reduced p-ERK/ERK levels in the NC group (Cal27, p<0.01) and OE group (Cal27, p<0.05). Nicotine reversed the reduction in p-ERK/ERK in the OE+IN group; however, the difference was not statistically significant. CES1 increased with the overexpression of CHRNA5 (Cal27, p<0.001), and PD98059 reduced CES1 in the NC group (Tu686, p<0.01) and OE group (Tu686, p<0.05). Nicotine reversed the reduction in CES1 in the OE+IN group, showing a trend but the difference was not statistically significant (Fig. 8D–F).

Morphological comparison of tumor formation in nude mice and validation of CES1 regulation via the MEK/ERK pathway in tumorigenic tissues

Tumorigenesis was induced in nude mice using sh-CHRNA5 Cal27 stable transfectants, which were then divided into the NC group (control group), SH group (knockdown group), and SH+NIC group (knockdown + nicotine group). Tumor formation in each group of nude mice was measured every seven days. A comparison between the NC and SH groups revealed that from day 14 onwards, the tumor volume in the NC group was significantly larger than that in the SH group (p<0.01). Comparing the SH group with the SH+NIC group, from day 28 onwards, the tumor volume in the SH+NIC group was significantly larger than that in the SH group (p<0.05) (Fig. 9C, F). After tumor removal on day 35, both visual inspection and measurements showed that the tumor volume in the NC group was significantly larger than that in the SH group (weight, p<0.05; volume, p<0.0001), and the tumor volume in the SH+NIC group was significantly larger than that in the SH group (weight, p<0.05; volume, p<0.001) (Fig. 9A, B, D, E).

Open in a separate window

Fig. 9

Nude mouse tumor morphology, PCR, and Western blot experiments.

A Comparison of tumor tissues between the NC group and SH group. B Quantification of tumor weight in the NC group and SH group. C Changes in tumor volume in the NC group and SH group. D Comparison of tumor tissues between the SH group and SH+NIC group. E Quantification of tumor weight in the NC group and SH+NIC group. F Changes in tumor volume in the NC group and SH+NIC group. G Validation of MEK/ERK pathway and CES1 protein expression changes in nude mouse tumors. H Quantitative analysis of MEK/ERK pathway and CES1 protein expression in nude mouse tumors. I Comparison of CHRNA5 and CES1 transcription levels in tumor tissues between the NC group and SH group.

Subsequent western blot experiments on tumor tissue proteins revealed that activation of MEK (p-MEK/MEK) significantly decreased under CHRNA5 knockdown (p<0.001), whereas nicotine reversed the reduction of p-MEK/MEK in the SH group. Similarly, activated ERK (p-ERK/ERK) was also significantly decreased after CHRNA5 knockdown (p<0.0001); however, nicotine did not significantly reverse the trend of reduced p-ERK/ERK in the SH group. Additionally, the expression level of CES1 significantly decreased after CHRNA5 knockdown (p<0.05), and nicotine showed a trend towards reversing the reduction in CES1 in the SH group, although the effect was not significant (Fig. 9G, H). qRT-PCR validation of CES1 transcription levels in the NC and SH groups showed that CES1 mRNA expression levels significantly decreased with the knockdown of CHRNA5 (p<0.001), consistent with the protein expression trend (Fig. ​(Fig.9I9I).

Pathological staining of nude mouse tumor tissue and human head and neck squamous carcinoma tissue

Pathological section staining was performed on the tumor tissues of the three groups of tumor-bearing nude mice (NC, SH, and SH+NIC), as shown in Fig. 10A. HE staining of all three groups exhibited characteristics of undifferentiated carcinoma, with significant pleomorphism, densely arranged nucleoli observable within the nuclei, and multiple mitotic Fig. 10A. No obvious lung metastatic tissue was found in the HE staining of lung tissues.

Open in a separate window

Fig. 10

HE staining and immunohistochemistry of HNSC tissue.

A Nude mouse tumor histopathology experiment (NC group, SH group, and SH+NIC group). HE staining of tumor tissues and lung tissues. B Immunohistochemical staining of CHRNA5. C Quantitative analysis of H-score. D Immunohistochemical staining of Pan-ck and PCNA. E Immunohistochemical staining of Ki-67 and E-cadherin. F Quantitative analysis of H-score. G Histopathology experiment on laryngocarcinoma and hypopharyngeal carcinoma tissues. Immunohistochemical staining of CHRNA5. H Immunohistochemical staining of Pan-ck. I Quantitative analysis of H-score. J Immunohistochemical staining of E-cadherin. K Immunohistochemical staining of Ki-67. L Immunohistochemical staining of PCNA. M Quantitative analysis of H-score. N Immunohistochemical experiments on laryngeal cancer tissue microarray. Immunohistochemical staining of CHRNA5. O Quantitative analysis of H-score.

As shown in Fig. 10B, C, the protein expression level of CHRNA5 in the SH group was lower than that in the NC group, with a significantly lower immunohistochemical score (H-score) than that in the NC group (p<0.05), further confirming the successful construction of the Cal27 cell nude mouse tumor model. At the same time, the protein expression levels of Pan-ck (p<0.01), PCNA (p<0.05), and Ki-67 (p<0.001) in the SH group were also lower than those in the NC group, with significantly lower H-score, while the protein expression level of E-cadherin in the SH group was higher than that in the NC group, with a significantly increased H-score (p<0.001) (Fig. 10D–F). In the SH+NIC group, the protein expression level of CHRNA5 was higher than that in the SH group, with a significantly higher H-score than the SH group (p<0.01). Nicotine reversed the expression level of CHRNA5, leading to increased protein expression levels of Pan-ck (p<0.01), and Ki-67 (p<0.01), with a significantly increased H-score. Simultaneously, the protein expression level of E-cadherin decreased, with a significantly reduced H-score (p<0.01) (Fig. 10D–F).

Tissue samples were collected from patients with laryngocarcinoma treated surgically at our hospital and controls (3:3), as well as from patients with hypopharyngeal carcinoma and controls (3:3). Immunohistochemical staining revealed that the protein expression levels of CHRNA5 and Pan-ck in laryngocarcinoma and hypopharyngeal carcinoma were higher than those in the control group, with a significantly increased H-score (p<0.05). Simultaneously, the protein expression levels of E-cadherin in laryngocarcinoma and hypopharyngeal carcinoma were lower than those in the control group, with a significantly reduced H-score (p<0.05). In hypopharyngeal carcinoma, the protein expression levels of PCNA and Ki-67 were higher than those in the control group, with a significantly increased H-score (p<0.05) (Fig. 10G–M). Subsequently, 29 patients with laryngocarcinoma and 10 normal control tissues were collected, with two tissue sites per case stained on an immunotissue microarray (58 cases of laryngeal cancer and 20 cases of normal control tissue). The results showed that CHRNA5 (p<0.01) was highly expressed in patients with laryngocarcinoma, with a significantly increased H-score (Fig. 10N, O).