Abstract Background The landscape of the tumor microenvironment (TME) is intricately linked to the development of head and neck squamous cell carcinoma (HNSCC) and significantly influences immunotherapy efficacy. Recent research has underscored the pivotal role of PNCK in cancer progression, yet its relationship with immunotherapy remains elusive. Methods We leveraged sequencing data from our cohort and public databases to evaluate PNCK expression, prognostic significance, and immune efficacy prediction. In vitro and in vivo experiments explored the role of PNCK in HNSCC progression. Animal models assessed the therapeutic effects and survival benefits of PNCK knockdown combined with immune checkpoint inhibitors (ICIs). Single-cell transcriptomics analyzed the impact of PNCK on the TME, and proteomic studies elucidated the mechanisms. Results PNCK exerts multifaceted critical roles in the progression of HNSCC. Lower PNCK expression is associated with improved prognosis, enhanced immune cell infiltration, and increased responsiveness to ICIs. Conversely, PNCK promotes HNSCC cell migration, invasion, proliferation, colony formation, zebrafish angiogenesis, and tumor growth in mice. Moreover, targeting PNCK enhances sensitivity to ICIs and leads to significant alterations in the T-cell and B-cell ratios within the TME. These changes are attributed to the inhibition of nuclear transcription of PNCK-phosphorylated ZEB1, which restricts cytokine release and inflames the immune microenvironment to regulate the TME. Conclusions Inhibition of PNCK may be a potential strategy for treating HNSCC, as it may activate the immune response and improve the TME, thereby enhancing the efficacy of immunotherapy for HNSCC patients. Keywords: Immunotherapy, Tumor Microenvironment, Head and Neck Cancer __________________________________________________________________ WHAT IS ALREADY KNOWN ON THIS TOPIC * Clinical evidence demonstrates that immunotherapies are effective in managing cancer progression; however, resistance to immune checkpoint inhibitors (ICIs) is common, often resulting from genetic changes in tumor signaling. Previous reports have identified that PNCK is implicated in cancer progression and immune response modulation. Nevertheless, its role in head and neck squamous cell carcinoma (HNSCC), along with its association with the tumor microenvironment (TME) and immunotherapy, remains underexplored. WHAT THIS STUDY ADDS * Our study contributes significantly to the existing body of knowledge by providing novel insights into the multifaceted role of PNCK in HNSCC. We demonstrate that PNCK expression is inversely correlated with patient prognosis and immune cell infiltration, suggesting that high PNCK levels contribute to an immune-suppressive TME. By using both in vitro and in vivo models, we show that PNCK promotes HNSCC progression through enhanced cell migration, invasion, proliferation, and tumor growth. Importantly, we reveal that targeting PNCK can enhance the efficacy of ICIs by modulating the TME, increasing T-cell and B-cell infiltration, and reducing immune evasion. These findings suggest that PNCK inhibition could be a promising strategy to overcome ICI resistance in HNSCC. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY * The combined evidence from our study indicates that PNCK is a critical regulator of HNSCC progression and the immune landscape of the TME. Clinically, our findings suggest that PNCK could serve as both a prognostic biomarker and a therapeutic target in HNSCC. Targeting PNCK may improve clinical outcomes or HNSCC patients by enhancing their responsiveness to ICIs, thereby providing a new avenue for treatment strategies. Introduction Head and neck squamous cell carcinoma (HNSCC) constitutes a significant global health issue, ranking as the sixth most prevalent cancer globally.[46]^1 2 Recent clinical evidence has underscored the efficacy of immunotherapies, particularly immune checkpoint inhibitors (ICIs), in managing the progression of HNSCC.[47]^3 4 Nevertheless, resistance to ICIs is a common occurrence among patients, which can be attributed to a multitude of factors, including genetic alterations in tumor-intrinsic signaling pathways. Tumor cells deploy intrinsic regulators to establish a suppressive tumor microenvironment (TME) to evade immune surveillance.[48]5,[49]7 These insights have catalyzed a surge in research efforts aimed at elucidating the molecular and cellular mechanisms underlying clinical resistance to ICIs. A prevalent mechanism by which malignant cells evade the rejuvenating effect of ICIs on tumor-targeting immunity is the establishment of an “immune desert” TME.[50]^8 Consequently, there has been a substantial commitment to the development of strategies that “inflame” the TME to facilitate the infiltration of immune effector cells and enhance clinical sensitivity to ICIs.[51]^9 PNCK, a member of the calmodulin kinase family, has attracted interest due to its regulation of intracellular calcium levels and its impact on downstream signaling pathways.[52]10,[53]14 Although PNCK has been associated with breast and renal cancers, indicating its capacity as an oncogene,[54]^15 its role in HNSCC development and its interaction with the TME have not been comprehensively examined. This research seeks to address the gaps by examining the multifaceted role of PNCK in HNSCC. Specifically, we aim to elucidate (1) the function of PNCK in promoting the malignant progression of HNSCC through a combination of in vitro and in vivo approaches; (2) the complex relationship between PNCK and the TME, and its implications for tumor immunotherapy; and (3) the primary regulatory mechanisms by which PNCK influences the immune therapeutic response in HNSCC. By employing a comprehensive experimental framework, this study aims to shed light on the functional significance of PNCK in HNSCC and its potential as a therapeutic target. The findings could contribute to the development of more effective treatment strategies for patients with HNSCC, ultimately improving their clinical outcomes. Methods and materials [55]Online supplemental figure S1 illustrates the technology roadmap for this study. Clinical sample collection Tumor biopsy samples were collected from 193 head and neck tumor patients at Fujian Cancer Hospital (January 2015–January 2018) and staged per the eighth edition TNM criteria ([56]online supplemental table S1). Animal model and cell lines Pathogen-free female C3H mice (6–8 weeks old) were obtained from SPF (Beijing) Biotechnology Co. The culture standards for wild-type (WT) zebrafish are outlined in The Zebrafish Book. Mouse SCC7 (RRID: CVCL_V412) and human CAL27 (RRID: CVCL_1107) cells were obtained from our Radiobiology Research Laboratory. The experimental procedures are detailed in [57]online supplemental materials. Tumor challenge and treatment SCC7 cells were resuspended in 0.2 mL of phosphate-buffered saline (PBS) and injected subcutaneously into the flanks of C3H or NSG mice. In the first set of animal experiments without combination therapy, 1×10^6 SCC7 cells were injected per mouse. In the second set of experiments, designed to study the effects of PNCK KD combined with ICIs under a higher tumor burden, 5×10^6 SCC7 cells were injected per mouse. Treatment began on day 5 post-tumor cell injection and was administered every 3 days, with a total dosage of 600 µg. Tumor size was determined using the formula: long diameter×(short diameter)^2/2 and charted over time to create a growth curve. Mice were euthanized immediately if their tumor volume reached 2000 mm³ during the course of the experiment. In survival experiments, a tumor size over 1000 mm³ was regarded as a significant event. For ICI therapy, mice with WT or Pnck-KD SCC7 cells received intraperitoneal injections of anti-mouse CTLA4 mAbs (200 µg per mouse; BioXCell, Cat No. BE0032, clone: UC10-4F10-11, RRID: AB_1107598), anti-PD-L1 mAbs (200 µg per mouse; MCE, Cat No. HY-[58]P99145, clone: 10F.9G2), or corresponding isotype control mAbs on days 5, 8, and 11. Lentiviral stable infection constructed knockdown and overexpressed PNCK cell lines The experimental procedures are detailed in [59]online supplemental materials. RNA isolation, cDNA synthesis, and RT-qPCR We isolated total RNA using TRIzol reagent and synthesized first-strand cDNA from 1 µg of RNA using the RT PCR Kit (Promega, Cat No. LS2052). RT-qPCR was performed with SYBR Green Master Mix (Promega, Cat No. LS2062). Cycling conditions were set at 95°C for 30 s, followed by 40 cycles of 95°C for 5 s and 60°C for 30 s. Relative levels were calculated using the relative quantification 2^-ΔΔCT method. All primer sequences are shown in [60]online supplemental table S2. Western blot The detailed experimental procedures are accessible in [61]online supplemental materials. Zebrafish knockdown and overexpression model construction Zebrafish embryos at 30 min postfertilization were microinjected in agarose dishes. RNA target editing was employed to knock down the endogenous pnck gene.[62]^16 Overexpression (OE) involved dividing embryos into control (CMV-mCherry) and experimental (CMV-pnck) groups, each repeated three times. Injected embryos (6 nL volume) were cultured in E3 buffer and observed at 48 hpf under in vivo fluorescence microscopy. Changes in angiogenesis rate after pnck gene manipulation were analyzed computationally and statistically. The experimental procedures are detailed in [63]online supplemental materials. Flow cytometric analysis of tumor-infiltrating CD8^+ T cells Mice were euthanized at the later stages of the experiment, and tumor samples were digested for 30 min at 37°C with 0.1 mg/mL DNase I and 1 mg/mL Collagenase D. The resulting single-cell suspensions were filtered through a 70 µm cell strainer (Miltenyi Biotec, Cat No. 130-098-462) and incubated with Fc-block (1:200, BD Pharmingen, Cat No. 553142) to prevent non-specific binding. Cells were stained with FITC-conjugated anti-mouse CD45 (1:200, BD Pharmingen, Cat No. 553079), BV510-conjugated anti-mouse CD8 (1:200, BD Pharmingen, catalog no. 563068), and a fixable viability dye (1:1000, BD Pharmingen, Cat No. 565388) on ice in the dark for 40 min. Flow cytometry was performed using a CytoFLEX S (Beckman Coulter), and data were analyzed with CytExpert software. Single-cell dissociation, sequencing, and data analysis The experimental procedures of single-cell dissociation, sequencing, and data analysis are detailed in [64]online supplemental materials. Multiple immunofluorescence The tissue samples were first fixed in 4% paraformaldehyde, embedded in paraffin, and then sectioned into 4 µm thick slices. These sections were deparaffinized in xylene and rehydrated through a graded series of ethanol solutions. Antigen retrieval was performed using citrate buffer (pH 6.0) in a microwave for 20 min, followed by cooling at room temperature. The sections were then blocked with 5% bovine serum albumin for 1 hour at room temperature to prevent non-specific binding. Primary antibodies specific to the target antigens were applied to the sections and incubated overnight at 4°C. After washing with PBS, the sections were incubated with secondary antibodies conjugated to different fluorophores for 1 hour at room temperature in the dark. Nuclei were counterstained with DAPI for 10 min. The stained sections were then mounted with antifade mounting medium and visualized under a fluorescence microscope. Images were captured using appropriate filters to detect the different fluorophores. 4D-label free phosphorylation quantitative proteomics Samples were treated with dithiothreitol and iodoacetamide for protein denaturation and alkylation, followed by trypsin digestion (1 mg/mL, 1/50 w/w) and peptide desalting. Phosphopeptides were enriched, and chromatography was performed using a Sep-Pak C18 cartridge column using a gradient of mobile phases (A: 99.9% H2O, 0.1% FA; B: 99.9% ACN, 0.1% FA). Mass spectrometry settings included capillary voltage, gas temperature, and flow rate. Data were processed with MaxQuant against the uniprot-Mus musculus-10 090-2023.2.1.fasta database. The experimental procedures are detailed in [65]online supplemental materials. Shotgun proteomics The experimental procedures are detailed in [66]online supplemental materials. Co-immunoprecipitation The experimental procedures are detailed in [67]online supplemental materials. Molecular docking analysis PNCK and ZEB1 protein-protein docking was performed using GRAMM-X ([68]http://gramm.compbio.ku.edu/). Structural domains of PNCK and ZEB1 were obtained from the Protein Data Bank (PDB, [69]http://www.rcsb.org/). Protein–protein interactions were visualized and analyzed using Pymol (V.2.4) and PDBePISA ([70]https://www.ebi.ac.uk/pdbe/pisa/). Multicytokine assay Tumor tissue was homogenized, and cell supernatants were collected for a multicytokine assay using a Luminex 200 system. A panel of 31 mouse cytokines, including TNF-α, IFN-γ, GM-CSF, IL-10, IL-16, IL-1b, IL-2, IL-4, IL-6, CX3CL, CXCL1, CXCL10, CXCL11, CXCL12, CXCL13, CXCL16, CXCL5, CCL11, CCL12, CCL17, CCL19, CCL2, CCL20, CCL22, CCL24, CCL27, CCL3, CCL4, CCL5, CCL7, and CCL1, was analyzed according to the manufacturer recommendations. Public data downloading and processing Detailed methods of bioinformatics analysis are detailed in [71]online supplemental material. Statistical analysis Statistical analysis was performed by using GraphPad Prism V.8.4.1. For comparisons between two groups, a two-tailed unpaired Student’s t-test or Wilcoxon test was employed. For comparisons involving more than two groups, one-way analysis of variance (ANOVA) or two-way ANOVA was used, followed by Tukey’s multiple comparison test. Survival analyses used the two-sided log-rank test, and categorical data were compared using the χ^2 test. Spearman correlation was used to assess variable associations. A p<0.05 was considered statistically significant. Results PNCK is associated with poor prognosis and promotes the development of “immune-desert” HNSCC To investigate the role of PNCK in cancer and its alterations in malignancy, we initially performed an exploratory analysis to examine the relationship between the expression pattern of PNCK and patient prognosis using sequencing data from our institutional samples (n=193) as well as public datasets (TCGA-HNSC, n=503). We found that PNCK is highly expressed in malignant HNSC cells. Analysis of TCGA-HNSC transcriptomic and protein data revealed significantly higher PNCK levels in tumor tissues compared with normal tissues ([72]figure 1a, [73]online supplemental figure S2a,b). At the single-cell level, PNCK expression was upregulated in malignant cells ([74]online supplemental figure S2c). High PNCK expression correlated with higher clinical and pathological T stages ([75]online supplemental figure S2d), higher N stage, lymphovascular infiltration ([76]online supplemental figure S2e), and advanced clinical and histological stages ([77]online supplemental figure S2f). Next, we observed that PNCK exhibits significant prognostic predictive ability in HNSCC. Data from Fujian Provincial Cancer Hospital (n=193) cohorts and TCGA-HNSC (n=503) indicated that high PNCK expression is associated with lower survival times ([78]figure 1b, [79]online supplemental figure S3a). One-way and time-dependent receiver operating characteristic (ROC) curves confirmed its predictive effectiveness for overall survival, with area under the curves (AUCs) exceeding 0.7 ([80]online supplemental figure S3b,c). Furthermore, long-term follow-up revealed that patients from our center with low PNCK expression had a lower recurrence rate compared with those with high PNCK expression ([81]figure 1c). These findings suggest that PNCK may serve as a promising prognostic indicator in HNSCC. Figure 1. An initial exploration into the predictive significance of PNCK for prognosis and immune-inflammatory response. (a) Differential expression of PNCK in TCGA-HNSC cancer tissues and normal tissues. Two-tailed paired Student’s t-test. (b) Kaplan-Meier ‌progression-free survival (PFS) curve for 193 head and neck tumor patients with high or low PNCK expression in the cohort from Fujian Provincial Cancer Hospital. Two-sided log-rank test. (c) Recurrence rates in patients from Fujian Provincial Cancer Hospital with high versus low PNCK expression. The red bar represents relapse, and the blue bar indicates non-relapse. Two-tailed χ^2 test. (d, e) GSEA revealing pro-carcinogenic-related and immune-related pathways correlated with PNCK expression in the in-house dataset (n=193 biologically independent samples). High expression of PNCK was enriched in the Hedgehog signaling pathway (d), whereas low expression was enriched in the cytokine-cytokine receptor pathway (e). (f) Significant differences in immune score, stromal score, and ESTIMATE score between PNCK high and low expression groups. The upper, middle and lower horizontal lines of the box represent the upper, median and lower quartile respectively. Two-tailed unpaired Student’s t-test. Data are presented as mean values±SD. (g) Association of PNCK expression with diverse immune cells and the immune score. Two-tailed Spearman correlation is reported. (h) The heatmap illustrates the differential cytokine expression profiles between high and low PNCK expression groups. (i) Association between PNCK expression and immune checkpoint marker expression. Two-tailed Spearman correlation is reported. HNSC, head and neck squamous cell. *** indicates p < 0.001, and **** indicates p < 0.0001. [82]Figure 1 [83]Open in a new tab Given the clinical implications of PNCK expression, we delved further into its intrinsic role and mechanistic effects within the TME. KEGG enrichment analysis showed that high expression of PNCK was enriched in pro-carcinogenic-related pathways, such as the Hedgehog signaling pathway ([84]figure 1d), whereas low expression was enriched in active immune-inflammatory response pathways, such as the cytokine-cytokine receptor pathway ([85]figure 1e). These findings imply that PNCK alterations may influence the immune microenvironment, warranting further study of immune cell infiltration. Consequently, we assessed alterations in the immune microenvironment and found that high PNCK expression correlates with decreased immune cell infiltration within the TME. Analysis of stromal, immune, and ESTIMATE scores showed significant differences between high and low PNCK expression groups ([86]figure 1f, [87]online supplemental figure S4a). Subsequent evaluation of cellular infiltration demonstrated that the levels of most immune cells, including T cells and B cells, significantly inversely correlated with PNCK expression, as confirmed by TIMER ([88]figure 1g) and the ssGSEA algorithm ([89]online supplemental figure S4b). High PNCK expression was also negatively correlated with genes related to CD8+T cell activation, such as CD8A, CD8B, CTLA4, GZMA, and GZMB ([90]online supplemental figure S4c). Since cytokines are central to intercellular communication, particularly in immune responses, we observed significant differences in the activities of chemokines and interleukins, crucial cytokine family members, between patients with varying PNCK expression levels ([91]figure 1h). Furthermore, significant differences in IFNγ scores were observed between patients with high and low PNCK expression ([92]online supplemental figure S4d). In summary, low PNCK expression was associated with higher immune cell infiltration, indicating an “immuno-inflammatory” profile, while high PNCK expression indicated an “immune desert” profile. Transitioning from characterization of the immune microenvironment to predicting immune therapy efficacy is a significant step for clinical applications. In this regard, PNCK expression was negatively associated with common immune checkpoint markers ([93]figure 1i, [94]online supplemental figure S4d). Using the TIDE algorithm to analyze immunotherapy response rates in the TCGA-HNSC and melanoma ([95]GSE100797) cohorts, we found that patients with lower PNCK expression were more likely to respond to immunotherapy ([96]online supplemental figure S4e,f). This suggests that individuals with reduced PNCK expression may have an increased likelihood of benefiting from ICI therapy, potentially due to a more active immune-inflammatory response, such as the activation of multiple immune cells and the release of proinflammatory cytokines. Cellular-level validation of PNCK promoting malignant biological behaviors in HNSCC We hypothesize that PNCK may facilitate malignant behaviors such as metastasis and proliferation in cancer cells, based on its highly enriched presence in pro-oncogenic pathways. To further investigate our hypothesis, we initially confirmed PNCK OE and knockdown in CAL27 and SCC7 cells ([97]figure 2a,b). Migration and invasion of tumor cells are critical steps in the process of cancer metastasis.[98]^17 In the wound healing assay, PNCK knockdown significantly inhibited cell migration after 24 hours ([99]figure 2c). Migration and invasion assays showed fewer cells passing through the membrane and reduced Matrigel penetration in the knockdown group after 48 hours ([100]figure 2d), indicating that PNCK knockdown inhibits HNSC cell migration and invasion. In parallel, the capacity for unregulated proliferation is a key biological feature of tumor cells.[101]^18 The CCK8 assay revealed that knocking down PNCK significantly inhibited cell growth, peaking at 48 hours, with similar results in SCC7 cells ([102]figure 2e). Colony formation was also significantly reduced in the PNCK-KD group ([103]figure 1f). EdU staining showed more cells in the DNA replication phase in the knockdown group ([104]figure 2g,h). PNCK knockdown increased G1 phase distribution and slightly decreased S and G2 phase cell numbers ([105]figure 2i), suggesting G1 phase arrest and cell cycle disruption. These findings suggest that PNCK plays a facilitative role in the malignant behavior of cells and may represent a potential therapeutic target for cancer treatment. Figure 2. Cellular-level validation of PNCK promoting malignant biological behaviors in HNSCC. (a) RT-qPCR verification of PNCK knockdown efficiency. A two-tailed unpaired Student’s t-test was performed. Data are presented as mean values±SD. (b) Western blot verification of PNCK knockdown and overexpression efficiency. The numbers below each panel represent the relative protein expression, with the NC group set to 1. (c) A scratch assay was performed to assess cell migration. The wound healing was monitored and quantified using a two-tailed unpaired Student’s t-test. Data are presented as mean values±SD. (d) Transwell migration and invasion assays were conducted to evaluate the migratory and invasive capabilities of HNSCC cells with altered PNCK expression, using a two-tailed unpaired Student’s t-test. Data are presented as mean values±SD. (e) Cell proliferation was assessed using the CCK8 assay. Two-way ANOVA with Tukey’s multiple comparison test. Data are presented as mean values±SD. (f) The ability of HNSCC cells to form colonies was evaluated using a colony formation assay, using a two-tailed unpaired Student’s t-test. Data are presented as mean values±SD. (g) EdU incorporation assay was performed to detect proliferating cells. Blue represents cell nuclei stained with Hoechst, and red represents EdU-positive cells; scale bar: 100 µm. (h) Flow cytometry was used to quantify EdU-positive cells. (i) The cell cycle distribution of HNSCC cells with altered PNCK expression was analyzed by flow cytometry. ANOVA, analysis of variance; HNSCC, head and neck squamous cell carcinoma. [106]Figure 2 [107]Open in a new tab Pnck gene knockdown suppresses tumor growth by inducing an antitumor immune response Although PNCK exhibited oncogenic characteristics in vitro, these results did not account for the complex interplay between PNCK and the TME, particularly the immune system, which is crucial for tumor development. To address this limitation, we conducted in vivo studies using zebrafish and mouse models and validated our findings with tissue microarrays from patients with HNSCC. The zebrafish, recognized as a robust in vivo model, reveals a high degree of genetic homology with its human counterpart ([108]figure 3a).[109]^19 Functional studies in zebrafish demonstrate that knockdown of pnck attenuates vascular fluorescence, whereas pnck OE enhances it, establishing a potential link between pnck expression and angiogenesis ([110]figure 3b,c). Next, WT and Pnck-KD SCC7 cells were implanted in immunocompetent mice, showing that Pnck knockdown significantly inhibited HNSC growth, tumor mass ([111]figure 3d,e,f) and lung metastasis ([112]figure 3g,h). To further confirm that the tumor suppressive effect was mediated by the immune system, WT and Pnck-KD SCC7 cells were injected into severely immunodeficient NSG mice. Under conditions of immunodeficiency, the trend of tumor reduction was weakened without statistically significant difference ([113]online supplemental figure S5a). This highlights the role of the immune system in restricting tumor progression when PNCK is knocked down. Immunohistochemical staining of 55 primary tumors and 5 matched non-cancerous tissues showed significantly higher PNCK expression in tumors ([114]figure 3i). Patients with high PNCK expression had significantly larger tumors ([115]figure 3j), indicating that PNCK is elevated in HNSCC and correlates with tumor volume, suggesting an oncogenic effect. Figure 3. Knockdown of PNCK significantly inhibits in vivo growth of HNSCC. (a) Homology comparison of PNCK gene between zebrafish and humans. The squares and letters in various colors at the top represent different amino acid sequences, while the blue squares at the bottom indicate overlapping portions of these sequences. (b, c) Comparison of zebrafish blood vessel fluorescence intensity after PNCK knockdown (b) and overexpression (c).A two-tailed unpaired Student’s t-test was performed. Data are presented as mean values±SD. (d) Subcutaneous tumor growth in C3H mice (n=7 biologically independent samples). (e) Comparison of mouse tumor weights. A two-tailed unpaired Student’s t-test was performed. Data are presented as mean values±SD. (f) Mouse tumor growth curve. A two-tailed unpaired Student’s t-test was performed. Data are presented as mean values±SD. (g) HE staining of lung metastatic nodules in mice (n=8 biologically independent samples). (h) The pie chart shows the percentage of mice that developed pulmonary metastases. Two-tailed χ^2 test. (i) Difference in PNCK-positive area and HE staining between normal and tumor tissues in HNSCC tissue microarray (n=60 biologically independent samples). A two-tailed unpaired Student’s t-test was performed. Data are presented as mean values±SD. (j) Difference in tumor volume between PNCK high and low expression groups in HNSCC tissue microarray, and corresponding HE staining (n=60 biologically independent samples). A two-tailed unpaired Student’s t-test was performed. Data are presented as mean values±SD. HNSCC, head and neck squamous cell carcinoma. * indicates p < 0.05, ** indicates p < 0.01, *** indicates p < 0.001, and **** indicates p < 0.0001. [116]Figure 3 [117]Open in a new tab Drawing on the observation of tumor-suppressive role in immunocompetent C3H mice, we posited that the inhibition of Pnck may exert its antineoplastic effects predominantly through the augmentation of antitumor immunity. To delineate this mechanism, we initiated a transcriptomic analysis via high-throughput sequencing of RNA from three randomly selected Pnck KD and WT tumor specimens, each derived from the transplanted tumor models of these mice. We initially identified 421 differentially expressed genes (DEGs) between Pnck KD and WT tumor tissues, with 115 upregulated and 306 downregulated ([118]figure 4a). Upregulated DEGs included immune cell chemokines like Ccl24, Ccr10, Cxcr2, Vtcn1, and Ackr4 ([119]figure 4a), indicating an elevated secretion of immunoinflammatory cytokines. GSVA pathway enrichment analysis showed significant upregulation of immune-related pathways in Pnck-KD tumors, such as cytokine generation, and immunoprocesses by circulating immune proteins ([120]figure 4b). In the TCGA-HNSC dataset, PNCK low-expression group pathways were associated with complement and coagulation cascades and cytokine-cytokine receptor interactions, indicating antitumor immunity. In contrast, high PNCK expression correlated with cell cycle and DNA replication pathways, indicating tumor proliferation ([121]figure 4c). Immunohistochemistry of mouse tumor tissues showed higher CD4, CD8, and CD11b expression in the Pnck-KD group, negatively correlated with PNCK expression ([122]figure 4d), suggesting that Pnck knockdown activates antitumor immune processes. Figure 4. Knockdown of the Pnck gene induces antitumor immune response. (a) Volcano plot showing DEGs in the sequencing results, including upregulation of genes encoding immune cell chemotactic factors, like Ccl24, Ccr10, Vtcn1, Ackr4 and Cxcr2. Differentially expressed genes were identified with the threshold of |log[2](fold change)| >1 and false discovery rate (FDR)<0.05. (b) Heatmap showing the enrichment and upregulation of immune-related pathways in the Pnck-KD group. (c) GSEA enrichment analysis of complement and cytokine-related pathways in the PNCK low-expression group, and GSEA enrichment of cell cycle and DNA repair-related pathways in the high-expression group. (d) Immunohistochemistry demonstrated the expression levels of CD4, CD8, and CD163 in PNCK high and low groups. (e) Infiltration abundance of CD8 T cells in mouse tumors in the WT and Pnck-KD group (n=3 biologically independent samples). [123]Figure 4 [124]Open in a new tab These results indicate that Pnck knockdown induces activation of immune-related pathways, characterized by elevated secretion of immunoinflammatory mediators and strengthened antitumor immune responses. Pnck knockdown enhanced the therapeutic response to ICIs in the HNSCC treatment ICIs have transformed the treatment landscape for various cancers, yet response rates remain low. Tumor cells can evade ICIs by creating an “immune desert” microenvironment. Strategies to “inflame” the TME and enhance immune cell infiltration are being explored to improve ICI responsiveness.[125]^8 We have found that the TME demonstrated an enhanced immunoinflammatory profile in conjunction with decreased PNCK expression within our internal cohort and Pnck-KD mice. This finding prompted us to hypothesize that PNCK knockdown may sensitize tumors to ICI therapy. To test this hypothesis, we organized eight treatment groups of mice: a control group (WT+IgG, n=10), a knockdown group (KD+IgG, n=10), a group receiving anti-CTLA-4 monotherapy (WT+anti-CTLA4, n=10), a group receiving combination therapy (KD+anti-CTLA4, n=10), a group receiving anti-PD-L1 monotherapy (WT+anti-PD-L1, n=10), a group receiving combination therapy (KD+anti-PD-L1, n=10), a group receiving anti-CTLA4 and anti-PD-L1 combination therapy (WT+anti-CTLA4+anti-PD-L1, n=10) and another group receiving anti-CTLA4 and anti-PD-L1 combination therapy (KD+anti-CTLA4 + anti-PD-L1, n=10). Treatment began on day 5 post-tumor cell injection and was administered every 3 days, with a total dosage of 600 µg ([126]figure 5a). Combined anti-CTLA-4 or anti-PD-L1 therapy significantly reduced tumor growth ([127]figure 5b,c). Notably, in the PNCK-KD group, the concurrent administration of both therapies resulted in the most significant reduction in tumor growth rate and volume. In the monotherapy group, 40% of the mice exhibited complete tumor regression, whereas in the double-agent combination group, 70% of the mice demonstrated complete tumor regression ([128]figure 5c). Pnck knockdown with ICIs significantly prolonged survival ([129]online supplemental figure S5b). Cured mice did not develop new tumors on reinoculation, indicating durable antitumor immune memory ([130]figure 5d). Furthermore, the anti-CTLA-4 combination group exhibited significant upregulation of common immune cell markers and cytokines, including CD4, CD8, IFN-γ, and PD-L1, within tumor tissues, suggesting a heightened immune activity ([131]figure 5e). Figure 5. PNCK knockdown enhances the efficacy of immune checkpoint inhibition therapy. (a) The experimental schedule for the SCC7 cell challenge in the C3H mice tumor model is as follows. Intraperitoneal injections of ICIs were administered at a dose of 200 µg per mouse on the 5th, 8th, and 11th days after tumor implantation. The experiment was terminated on the 23rd day, and the tumors were removed for subsequent analysis. If the tumor volume reached 2000 mm³ during the experiment, euthanasia was performed immediately. (b) Overall tumor growth curves in different groups of mice. Mice were intraperitoneally treated with 200 µg anti-PD-L1 or 200 µg anti-CTLA4 on days 5, 8 and 11 after tumor inoculation. A rat IgG isotype antibody was applied as a control. Two-way ANOVA with Tukey’s multiple comparison test. Data are presented as mean values±SEM. The statistical comparisons among these groups were listed as follows: WT+IgG vs KD+IgG, p=0.0258; WT+CTLA4 vs KD+CTLA4, p=0.0994; WT+PD-L1 vs KD+PD-L1, p=0.0016; WT+CTLA4 + PD-L1 vs KD+CTLA4 + PD-L1, p=0.0412; KD+IgG vs KD+CTLA4, p=0.0016; KD+IgG vs KD+PD-L1, p=0.0177; KD+CTLA4 vs KD+CTLA4 + PD-L1, p=0.0066; KD+PD-L1 vs KD+CTLA4 + PD-L1, p<0.0001. (c) Individual tumor growth curves and percentages in different groups of mice. (d) Tumor growth curve in mice with cured tumors after subcutaneous reimplantation of HNSCC. Two-way ANOVA with Tukey’s multiple comparison test was performed. Data are presented as mean values±SD. (e) Expression levels of immune markers and cytokines in tumors of mice in different groups. Kruskal-Wallis with Dunn’s multiple comparison test are reported. ANOVA, analysis of variance; ICIs, immune checkpoint inhibitors. [132]Figure 5 [133]Open in a new tab Our findings suggest that Pnck knockdown may enhance the sensitivity to ICI treatment, and that the combined approach of PNCK knockdown and ICI therapy may induce an inflammatory response within the TME. Single-cell RNA-sequencing unveils distinct immune microenvironment in PNCK knockdown versus WT tumors To elucidate the impact of PNCK knockdown on TME heterogeneity and alterations in immune cell populations, we conducted single-cell RNA-sequencing analysis on tumor cells derived from Pnck-KD and WT tumors. The sorted cells underwent 10X single-cell transcriptome sequencing ([134]online supplemental figure S6a). 21,617 cells were obtained and analyzed by UMAP clustering, identifying 21 subpopulations. These included five immune cell types (T cells, B cells, monocytes/macrophages, NK cells, and dendritic cells) and three non-immune cell types (epithelial cells, fibroblasts, and endothelial cells) ([135]online supplemental figure S6b,c). The immune cell composition differed significantly between the KD and WT groups, with the Pnck-KD tumors enriched in T cells and B cells, while monocytes/macrophages were more prevalent in WT tumors ([136]online supplemental figure S6d,e). Immunohistochemical analysis further corroborated the immune cell heterogeneity between the KD and WT groups. As illustrated in [137]online supplemental figure S7, this analysis robustly validated the differential expression of T cells, dendritic cells, and macrophages at the protein level, consistent with the findings from single-cell RNA sequencing. Next, we analyzed differential gene expression within the same immune cell subpopulations between the Pnck-KD and WT groups. Numerous differential genes were identified within these subpopulations ([138]online supplemental figure S6f). To further delineate the heterogeneity within the TME, we conducted subgroup analyses of T cells and B cells. Among the T cells, we identified five T cell subpopulations: naïve T cells, proliferative T cells, IFN-responsive T cells (IFN T), exhausted T cells (Tex), progenitor exhausted CD8+T cells (Tpex) and regulatory T cells (Treg) ([139]figure 6a–d). The proportions varied significantly between the Pnck-KD and WT groups. Tex cells were 16.96% in Pnck-KD tumors but 32.7% in WT tumors, while Tpex cells were significantly expanded in Pnck-KD tumors ([140]figure 6e). Moreover, protein expression analysis demonstrated a higher prevalence of Tpex cells (Tcf7^+ Pd1^+ Cd8^+) in Pnck-KD samples compared with WT samples ([141]figure 6f, [142]online supplemental figure S8a). Within these subgroups, the differential gene expression analysis revealed numerous genes with significantly altered expression patterns, including a range of functional gene alterations ([143]online supplemental figure S8b). KEGG enrichment analysis indicated upregulation of immune-related pathways, such as cytokine-cytokine receptor interaction and antigen processing, T cell receptor signaling pathway, and natural killer cell-mediated cytotoxicity in Tpex subpopulation, cell cycle and PI3K-Akt signal pathway in proliferative T cells, reflecting an activated immune state ([144]figure 6g, [145]online supplemental figure S8c). Figure 6. Heterogeneity analysis of T cell subtypes. (a, b) UMAP analysis shows the distribution of T cell subtypes in different samples. Each point represents a single cell, colored by its subtype. (c) Bubble plots showing the expression of commonly used marker genes for each T cell subtype, used for annotating cell subtypes. The size of the bubble represents the percentage of cells expressing the marker gene, and the color intensity indicates the expression level. (d) UMAP analysis shows the distribution of different T cell subtypes. Different colors represent different T cell subtypes. (e) Proportion of each T cell subtype in the WT and knockdown groups. Bar graphs illustrate the percentage of each subtype in the two groups. (f) Multiplex immunofluorescence probed the content of Tpex cells (TCF7^+ PD1^+ CD8^+) in PNCK-KD samples. (g) KEGG functional enrichment analysis of DEGs in the Tpex cell subtype. The bar graph shows the top enriched pathways, with the length of the bar indicating the enrichment score. DEGs, differentially expressed genes. [146]Figure 6 [147]Open in a new tab Tumor-infiltrating B lymphocytes constitute a vital element within the TME, functioning to regulate the immune response and exerting a profound influence on the trajectory of tumor development. In our study, significant B cell infiltration was observed in KD tumors ([148]figure 7a). We identified four B cell subsets: naïve B cells, memory B cells, regulatory B cells, and plasma B cells ([149]figure 7b,c, [150]online supplemental table S3). Memory B cells were notably increased in KD tumors, while regulatory B cells predominated in WT tumors ([151]figure 7d). Most cells expressed the activation marker Tnfrsf8 (Cd30), indicating active B cells in the TME ([152]figure 7e). Regulatory B cells, which negatively regulate immune responses, were more abundant in WT tumors, suggesting immunosuppression. We found numerous DEGs between WT and KD groups ([153]figure 7f). Notably, Ifi30 was significantly upregulated in KD tumors, associated with immune-promoting B cells and favorable outcomes, while Ppp1r14b and C1galt1 were upregulated in WT tumors, linked to immunosuppressive B cells and poor prognosis ([154]figure 7g). The differentiation timeline illustrated that naïve B cells are progenitors of other subtypes ([155]figure 7h). Figure 7. Heterogeneity analysis of B cell subtypes. (a) UMAP analysis shows the distribution of B cells in different groups. Each point represents a single cell, colored by its subtype. (b) UMAP analysis shows the distribution of B cell subtypes in different samples. Each point represents a single cell. Different colors represent different B cell subtypes. (c) Bubble plots showing the expression of commonly used marker genes for each B cell subtype, used for annotating cell subtypes. The size of the bubble represents the percentage of cells expressing the marker gene, and the color intensity indicates the expression level.(d) Proportion of each B cell subtype in the WT and knockdown groups. (e) Expression of the B cell marker Tnfrsf8 (Cd30). (f) Differential genes in regulatory B cells between different groups. (g) Prognostic value of specific genes in regulatory B cells, based on survival analysis. A Two-tailed log-rank test was performed (h) Pseudotime analysis shows the developmental trajectory of B cell subtypes. Figure 7 [156]Open in a new tab PNCK regulates ZEB1 nuclear translocation to remodel the tumor immunosuppressive microenvironment As a kinase, PNCK phosphorylates its interacting proteins, thereby initiating a cascade of events that may be a pivotal mechanism driving the reconfiguration of the immunoinflammatory microenvironment. Guided by this hypothesis, we postulated that changes in PNCK expression would be associated with shifts in the phosphate content within the microenvironment. We, therefore, measured phosphorylated protein levels and found that PNCK knockdown corresponded to a reduction in overall phosphorylated protein concentration, while OE led to an increase ([157]figure 8a). To elucidate the mechanisms by which PNCK may contribute to immune evasion, we further performed phosphoproteomic analyses to investigate the phosphorylation landscape under WT PNCK (WT, n=3) and OE (n=3) conditions. Our analysis revealed 1267 differentially phosphorylated proteins, with 817 showing increased and 450 showing decreased phosphorylation, indicating a general trend toward enhanced protein phosphorylation on PNCK OE ([158]figure 8b,c). Additionally, intersecting upregulated phosphorylated proteins with those interacting with PNCK in WT and OE groups identified 32 PNCK OE-specific upregulated phosphorylated proteins ([159]figure 8d). Figure 8. PNCK regulates ZEB1 nuclear translocation and reshapes the tumor immune-suppressive microenvironment. (a) Quantification of overall phosphorylated protein levels in various cell samples to assess changes in the phosphorylation status. A two-tailed unpaired Student’s t-test was performed. Data are presented as mean values±SD. (b) Differential phosphorylated proteins between the PNCK overexpression group and the WT group. (c) Heatmap showing the expression pattern of differentially phosphorylated proteins in the PNCK OE and WT groups. (d) Venn diagram showing the intersection of upregulated phosphorylated proteins and proteins interacting with PNCK in WT and OE groups. (e) Co-immunoprecipitation (CO-IP) validation demonstrates the interaction between PNCK and ZEB1. (f) Surface diagram of the docking model and their interfacing residues between ZEB1 and PNCK protein (ZEB1, blue; PNCK, yellow; hydrogen bond interaction, dotted line). (g) Western blot analysis shows changes in ZEB1 expression levels in response to alterations in PNCK expression. (h) Western blot analysis shows the upregulation of ZEB1 and phosphorylated ZEB1 (p-ZEB1) in the nuclear protein fraction of the PNCK OE group. GAPDH was used as a cytoplasmic protein loading control, and Histone H3 was used as a nuclear protein loading control. (i) Cytokine array analysis of numerous immune-related cytokines across multiple experimental groups, including KD, NC(KD), OE, NC(OE), OE+siZEB1, and OE+siNC. Figure 8 [160]Open in a new tab Previous studies have reported that the transcriptional repressor ZEB1 in melanoma cells is associated with reduced CD8+T cell infiltration and is a key determinant of melanoma immune evasion.[161]20,[162]22 We also detected diminished infiltration of functional T cells into the TME following PNCK knockdown, prompting us to consider the possibility of a relationship between PNCK and ZEB1. Consequently, we probed the functional interplay between PNCK and ZEB1, assessing whether a reciprocal relationship exists between their expressions. We initially validated that PNCK interacts with ZEB1 in SCC7 cells ([163]figure 8e). Protein–protein docking simulations depicted a stable docking interface between PNCK and ZEB1, with ZEB1 (blue) and PNCK (yellow) forming a hydrogen bond through Tyr-23 and Tyr-229 (yellow dashed line), yielding a binding energy of −14.6 kcal/mol, indicating a robust interaction ([164]figure 8f). To ascertain whether this interaction elicits alterations in downstream functions, we noted that ZEB1 expression was regulated in accordance with PNCK levels, such that the PNCK knockdown group displayed diminished ZEB1 expression, whereas the PNCK OE group exhibited elevated ZEB1 expression ([165]figure 8g). Concurrently, in the PNCK OE group, we observed elevated levels of ZEB1 and phosphorylated ZEB1 (p-ZEB1) in nuclear proteins compared with the control group ([166]figure 8h). This indicates that PNCK OE promotes the accumulation of ZEB1 and its phosphorylated form in the nucleus, suggesting that PNCK facilitates ZEB1 nuclear translocation. Moreover, multicytokine profiling revealed heightened cytokine levels in the PNCK knockdown group, indicative of a more pronounced immune response ([167]figure 8i). Given the already low cytokine levels in the NC samples, the suppressive effect of PNCK OE on cytokine production may not be readily observable. In such a low baseline context, the additional inhibitory impact of PNCK OE becomes challenging to detect. To clarify the mechanistic link between ZEB1, PNCK, and cytokine production in the tumor, a rescue experiment was conducted. In this experiment, knocking down ZEB1 in the PNCK overexpressing cell line resulted in a marked increase in cytokine levels, effectively reversing the effect of PNCK OE ([168]figure 8i). This suggests that PNCK modulates cytokine expression through ZEB1, thereby influencing the overall cytokine profile in the TME. Our data indicate that PNCK promotes the nuclear translocation of ZEB1, which subsequently leads to the release of a set of inflammatory cytokines, thereby enhancing the immunoinflammatory state within the TME ([169]online supplemental figure S9). Discussion Immunotherapy has shown great potential in oncology, but its efficacy varies among patients, leading to overtreatment or undertreatment issues.[170]^23 A key factor in resistance is the deficiency of functional T cells in “immune desert” tumors.[171]^24 Building on the observation that PNCK OE is linked to the “immune desert” phenotype and resistance to ICIs, we demonstrated that PNCK deficiency inhibits tumorigenesis in immunocompetent mice. This tumor suppression was predominantly attributed to the activation of antitumor immunity. Tumor development and progression depend on interactions between tumor cells and TME, which includes endothelial cells, fibroblasts, immune cells, and extracellular components.[172]25,[173]27 The TME sustains cancer hallmarks such as proliferative signaling, resistance to cell death, angiogenesis, invasion, metastasis, inflammation, and immune evasion.[174]^28 29 It also significantly influences treatment outcomes, making it crucial for developing new therapeutic strategies.[175]^30 31 Our study identified PNCK as an oncogene, where its inhibition exhibited a degree of tumor-suppressive effect in vitro and in immune-deficient mice. Notably, this effect was significantly enhanced in immune-competent mice, with the added benefit of sensitization to ICIs. Specifically, PNCK knockdown elicits tumor-suppressive effects, augmenting the infiltration of effector T and B cells, which suggests immune-dependent antitumor effects. Understanding the TME is crucial for developing effective therapeutic strategies against cancer. Using 10X single-cell RNA sequencing, this study delved into the role of PNCK in modulating the TME. Our results showed significant heterogeneity in immune cells between Pnck-KD and WT TME. On the one hand, T cells play a pivotal role in anticancer responses, eliminate tumor cells, inhibit immune evasion, and form immune memory.[176]^32 Tpex cells play a crucial role in the response to immunotherapy, highlighting the importance of increasing their presence in the TME to enhance treatment outcomes and extend patient survival.[177]^33 Notably, Tpex cells were found in greater abundance in the PNCK-KD group, indicating improved antigen presentation and a stronger immune response, which are key factors in effective cancer immunotherapy. Moreover, exhausted T cells, especially prevalent in the PNCK-WT group, secrete more anti-inflammatory cytokines, suppressing immune responses, and express higher levels of inhibitory receptors that inhibit T cell activity and impair disease defense on interaction with other cells. Their limited persistence and proliferation further compromise the immune system. On the other hand, cancer immunotherapy research underscores the pivotal role of B cells in anti-tumor immunity, as they produce antibodies targeting tumor cells to initiate immune responses that suppress tumor growth and spread.[178]^34 35 Our study revealed that B cells were more abundant in the PNCK-KD group compared with the WT group. In the WT group, the predominant B cell subpopulation was regulatory B cells, which possess potential immunosuppressive functions. This finding further suggests that the microenvironment becomes more immunologically active following PNCK knockdown. Thus, targeting PNCK could create favorable conditions for the immune system to combat tumors effectively, offering insights for developing immunotherapeutic interventions against cancer. Although our findings reveal immune microenvironment heterogeneity induced by Pnck alteration, further research is needed to clarify specific regulatory mechanisms and clinical applications. Cytokines are pivotal in shaping the TME, regulating immune suppression. For instance, IL-12 can activate proinflammatory cytokines and enhance T cell and dendritic cell function, overcoming immune suppression.[179]^36 The precise control of cytokine expression, particularly surface-anchored IL-12, can repolarize the tumor immune microenvironment, enhancing T cell therapy efficacy.[180]^37 Reports indicate that ZEB1 can regulate immune responses in the TME by inhibiting the expression of proinflammatory cytokines or promoting anti-inflammatory cytokines.[181]^38 This regulatory function may affect the function and distribution of immunosuppressive cells such as myeloid-derived suppressor cells and Tregs, further reinforcing tumor immune evasion. ZEB1, a pivotal molecule in epithelial-mesenchymal transition, has garnered attention for its dual role in both tumorigenesis and immune regulation, particularly through its influence on extracellular matrix remodeling and immune evasion. Our study suggests PNCK, an upstream regulator, influences cytokine expression by modulating ZEB1 phosphorylation, impacting immune cell infiltration and activity. Targeting PNCK or its regulation of ZEB1 may restore immune activity in the TME, enhancing immunotherapy efficacy. In summary, this study elucidates the pivotal role of PNCK in influencing the TME and sensitizing HNSCC to ICIs, thereby establishing a significant scientific basis for the advancement of novel immunotherapeutic approaches. supplementary material online supplemental file 1 [182]jitc-12-10-s001.pdf^ (3.4MB, pdf) DOI: 10.1136/jitc-2024-009893 Acknowledgements