PRKAR1B RNA and protein expression profiles in healthy human tissues were collected from the HPA database (https://www.proteinatlas.org/). Pan-Cancer RNA-Seq was obtained from the TCGA (https://portal.gdc.cancer.gov/) and GTEx (https://commonfund.nih.gov/GTEx) databases. PRKAR1B mRNA expression in tumor and adjacent normal samples from the TCGA database was examined using the Timer2.0 platform (http://timer.cistrome.org/). Matched normal tissue expression data from the GTEx database were also analyzed to offer a comprehensive overview. Radar plots were generated, and survival analyses, including proportional hazards assumption testing and Cox regression, were conducted using the survival package in R. Immune infiltration scores were calculated using the ssGSEA algorithm in the GSVA R package.

Cases lacking complete clinical data were excluded, and gene expression values were categorized into high and low groups using R. Kaplan-Meier survival analysis was conducted, and proportional hazards assumptions were evaluated using the survival package. Survival regression models were applied, and the ‘survminer’ and ‘ggplot2’ packages were employed to generate Kaplan-Meier curves. The predictive performance of PRKAR1B in HNSC patients was evaluated using ROC curve analysis implemented in the “pROC” R package.

To identify genes co-expressed with PRKAR1B, transcriptomic data from the TCGA-HNSC cohort were analyzed using the “limma” package in R. Pearson correlation coefficients were calculated, with significance defined as p < 0.001. Co-expressed genes were visualized using heatmaps.

In the TCGA cohort, HNSC patients were categorized into high- and low-PRKAR1B expression groups using the median PRKAR1B expression level as the threshold. Differentially expressed genes (DEGs) were determined using the ‘limma’ package in R, applying thresholds of logFC > 1 and p < 0.05 (12). DEGs were visualized using heatmaps, and functional enrichment analyses, such as Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG), were conducted using the ‘clusterProfiler’ R package (13).

The relative proportions of TIICs in HNSC samples were assessed using the CIBERSORT algorithm. Immune infiltration scores for 22 TIICs were derived from gene expression matrices and compared against the CIBERSORT reference matrix using 1,000 permutations. Wilcoxon tests were utilized to assess differences in immune infiltration between the high- and low-PRKAR1B expression groups. The relationship between PRKAR1B expression and immune checkpoint-related gene levels was examined using Pearson correlation analysis.

The relationship between PRKAR1B and immune phenotypes was evaluated to assess potential sensitivity to immune therapy. Pearson correlation analysis was carried out to evaluate relationships between PRKAR1B and 47 immune checkpoint-related genes. A threshold of p < 0.001 was set for statistical significance. Tumor mutational burden (TMB) for each sample was determined using the TMB function in R, which combined TMB data with gene expression data. Pearson correlation was employed to evaluate the association between TMB and gene expression across tumors. The pRRophetic R package, utilizing drug sensitivity data from the Cancer Genome Project, was used to predict chemotherapeutic drug sensitivity, expressed as IC50 values. The Wilcoxon test was applied to compare IC50 values between the high- and low-PRKAR1B expression groups. Immunophenoscores (IPS), sourced from the TCIA website (https://tcia.at/home), have demonstrated predictive capability for patient responses to ICI therapy (14).

Single-cell data from the Tumor Immune Single-Cell Hub (TISCH2) database (GSE103322) were analyzed to investigate PRKAR1B distribution at the single-cell level. UMAP was employed for cell clustering, annotation, and visualization (15).

HNSC cell lines SCC9 and SCC25 were procured from BNCC (BeNa Cell Culture Collection). SCC9 cells were maintained in DMEM supplemented with 10% fetal bovine serum, 100 μg/mL penicillin, and streptomycin, whereas SCC25 cells were maintained in DMEM-H/F-12K supplemented with 10% FBS and 400 ng/mL hydrocortisone. Cells were incubated at 37 °C in a humidified atmosphere with 5% CO₂.

Cells were plated in 6-well plates, then PRKAR1B siRNA and GP-transfect-Mate were supplied by GenePharma Co., Ltd., and transfection followed the manufacturer’s guidelines. The siRNA sequences used were as follows: NC siRNA: 5′-GCTTCGCGCCGTAGTCTTATCA-3′, PRKAR1B siRNA: 5′-CGUCCAGUUUGAAGAUGGATT-3′.

RNA was isolated from treated cells using the RNeasy Mini Kit (Qiagen, 74104). cDNA was synthesized using the RT Master Mix Kit (MedChemExpress, HY-K0511). qPCR was conducted on a 7500 fast real-time PCR system utilizing SYBR Green qPCR Master Mix (MedChemExpress, HYK0522). Each reaction was conducted in a final volume of 20 μL. The primers utilized were as follows: β-actin forward primer 5′-CACCATTGGCAATGAGCGGTTC-3′ and reverse primer 5′-AGGTCTTTGCGGATGTCACCGT-3′; PRKAR1B forward primer 5′-CAGGTCCTCAAAGACTGTATCGT-3′ and reverse primer 5′-ATGGGAGTCCGACTGTGAGT-3′; E-cadherin forward primer 5′-GCCTCCTGAAAAGAGAGTGGAAG-3′ and reverse primer 5′-TGGCAGTGTCTCTCCAAATCCG-3′; N-cadherin forward primer 5′-CCTCCAGAGTTTACTGCCATGAC-3′ and reverse primer 5′-GTAGGATCTCCGCCACTGATTC-3′.

Equal amounts of protein were separated using 10% SDS-PAGE and transferred onto PVDF membranes. Next, the membranes were blocked with 5% BSA in TBST for 1 hour at room temperature, followed by overnight incubation at 4 °C with anti-PRKAR1B (1:1000, Proteintech) and anti-β-actin (1:3000, Santa Cruz Biotechnology). Afterward, they were incubated with HRP-conjugated secondary antibodies (Cell Signaling) at room temperature for 2 hours. Protein bands were analyzed using ImageJ software.

Cells were fixed with 4% paraformaldehyde for 20 minutes and subsequently permeabilized using 0.1% Triton X-100 for 30 minutes. They were then incubated overnight at 4 °C with the primary antibody against PRKAR1B (1:200, Proteintech) in confocal Petri dishes. Thereafter, cells were incubated at room temperature with Plus 555-Goat Anti-Rabbit Recombinant Secondary Antibody (H+L) at a 1:500 dilution (Proteintech) for 2 hours. Finally, the cells were then stained with DAPI for 30 minutes.

CCK8 assays were conducted following the instructions provided by MedChemExpress. Briefly, cells were cultured for 0, 24, 48, and 72 hours. Following this, 10 μL of CCK-8 reagent was added to each well for 2 hours at 37 °C. Cell viability was evaluated by measuring absorbance at OD 450 nm.

Cell proliferation was assessed following siRNA transfection using the BeyoClick™ EdU Cell Proliferation Kit with AF488.

Transfected SCC9 and SCC25 cells (2 × 10⁴) were seeded into the upper chamber of Transwell inserts in 200 μL of serum-free DMEM, while the lower chamber contained 600 μL of DMEM supplemented with 10% FBS. After incubation, cells were fixed with 4% paraformaldehyde and stained with crystal violet. Non-migratory cells in the upper lumen were removed.

Cells were seeded in six-well plates and cultured until approximately 90% cell confluence. Afterward, a linear scratch was created in each well using a sterile 20 μL plastic pipette tip, following which cells were incubated in serum-free medium for 24 hours. Images of the wound area were captured at 0 and 24 hours and visualized under an inverted microscope.

Surgical paraffin-embedded tissue specimens were obtained from Jiangxi Cancer Hospital. Inclusion criteria: Pathologically confirmed primary head and neck squamous cell carcinoma; no prior surgery, radiotherapy or chemotherapy before admission; no history of head and neck tumors; complete data. Exclusion criteria: Radiotherapy, chemotherapy or other clinical interventions prior to surgery; presence of distant metastasis at initial diagnosis; severe organ dysfunction; autoimmune diseases, hematological diseases, or history of other malignant tumors; incomplete data. Immunohistochemistry staining was performed as previously described. The German semi-quantitative scoring system was used in this study.

Statistical analyses were conducted using R software version 4.2.3 and GraphPad Prism version 7.00. Data were analyzed using standard statistical tests. Two-tailed P values were reported, with p < 0.05 considered statistically significant.

The expression of PRKAR1B in various tissues under physiological conditions was analyzed using the HPA database. As anticipated, PRKAR1B expression was the highest in the cerebral cortex (Figure 1A). In addition, the TIMER database was employed to assess PRKAR1B expression across 33 tumor types and corresponding normal tissues from the TCGA database. The results revealed that PRKAR1B mRNA expression levels were reduced in Breast invasive carcinoma (BRCA), Cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC), Kidney chromophobe (KICH), Kidney renal clear cell carcinoma (KIRC), Lung adenocarcinoma (LUAD), Prostate adenocarcinoma (PRAD), and Uterine corpus endometrial carcinoma (UCEC) compared to normal tissues. On the other hand, elevated PRKAR1B mRNA expression was detected in several cancers, including Head and Neck Squamous Cell Carcinoma (HNSC), Cholangiocarcinoma (CHOL), Colon Adenocarcinoma (COAD), Esophageal Carcinoma (ESCA), Glioblastoma Multiforme (GBM), Pheochromocytoma and Paraganglioma (PCPG), Thyroid Carcinoma (THCA), Stomach Adenocarcinoma (STAD), and Rectum Adenocarcinoma (READ) (Figure 1B). These findings indicated that mRNA levels of PRKAR1B were higher in tumor tissues compared to adjacent normal tissues across various cancers, including COAD, HNSC, KICH, KIRP, LIHC, LUAD, PRAD, STAD, THCA, and CHOL (Figure 1C). To address the absence of normal tissue data for some tumors in the TIMER database, the TCGA and GTEx datasets were integrated, and radar charts were plotted to explore PRKAR1B expression in 33 tumor types. Interestingly, the analysis showed that PRKAR1B expression was up-regulated in 15 tumor types compared to their corresponding normal tissues (Figure 1D). To investigate the prognostic relevance of PRKAR1B expression, Cox regression survival analysis was performed using the TCGA cohort. Importantly, significant associations were observed in HNSC, ACC, LUAD, PAAD, and UVM. In HNSC, the hazard ratio (HR) exceeded 1 (95% CI: 1.026-1.758), positioning PRKAR1B as a potential risk factor (Figure 1E). Besides, the ssGSEA was applied to assess the infiltration of 24 immune cell types within the TCGA cohort, revealing significant correlations between PRKAR1B expression and T cell subsets, NK cells, and B cells, highlighting its potential role in the tumor microenvironment (Figure 1F).

In the present study, PRKAR1B expression was analyzed in 566 samples derived from the TCGA-HNSC cohort. The analysis revealed significantly elevated PRKAR1B expression levels in HNSC tumor tissues compared to normal and paired adjacent tissues (Figures 2A, B). However, there was no significantly PRKAR1B expressions between HPV-positive and HPV-negative tumor tissues (Supplementary Figure 1). To assess the clinical relevance of PRKAR1B expression, its correlation with various clinical factors was examined in HNSC patients, unveiling that PRKAR1B expression was correlated with tumor grade and N-stage (Figures 2C, D). Survival analysis revealed that higher PRKAR1B expression levels were correlated with poorer overall survival (OS) and progression-free survival (PFS) in patients with HNSC (P < 0.05). Furthermore, Kaplan-Meier survival curves illustrated that elevated PRKAR1B expression levels were correlated with poorer prognosis, implying its significant role in tumor progression (Figures 2E, F). Additionally, ROC analysis, performed to explore the diagnostic value of PRKAR1B level in HNSC patients, indicated an AUC of 0.795, thereby supporting its significant diagnostic potential in HNSC (Figure 2G).

Correlation analysis of the TCGA-HNSC dataset was performed to identify genes co-expressed with PRKAR1B. The top five co-expressed genes, namely C7orf50, EIF3B, TBRG4, DDX56, and BRAT1, were identified (Figures 3A). Of note, these genes have been extensively studied in various diseases. A chord diagram was established to visualize the relationship between genes whose expression correlated positively or negatively with PRKAR1B expression, displaying the top twelve genes with significant positive or negative correlations with PRKAR1B expression (Figure 3F). These genes may directly or indirectly affect PRKAR1B expression or may be regulated by PRKAR1B.

Differential expression analysis was performed on HNSC tumor samples stratified by PRKAR1B expression to explore its molecular role in tumorigenesis and progression. The top 50 DEGs were depicted in a heatmap (Figure 4A). GO and KEGG pathway enrichment analyses were conducted on these DEGs. GO functional analysis revealed that PRKAR1B is involved in skin development, epidermal development, keratinocyte differentiation, and epidermal cell differentiation (Figures 4B). At the same time, KEGG pathway analysis uncovered enrichment in several tumor-related KEGG pathways, including Cytoskeleton in muscle cells, Protein digestion and absorption, Pancreatic secretion, and Staphylococcus aureus infection (Figures 4E, F).

The CIBERSORT algorithm was adopted to evaluate correlations between PRKAR1B expression and the infiltration of different immune cell subsets. The results revealed significant differences in immune infiltration between the high- and low-PRKAR1B expression groups, especially for resting mast cells, resting dendritic cells, Tregs, and M1 macrophages (Figure 5A). Specifically, PRKAR1B expression was positively correlated with the infiltration of M2 macrophages and activated mast cells, as well as negatively correlated with the infiltration of resting mast cells, plasma cells, follicular helper T cells, and Tregs (Figure 5B). To explore the impact of PRKAR1B on immunotherapy, its relationship with immune checkpoint genes was examined, revealing correlations between PRKAR1B expression levels and the immune checkpoint genes CD276 and TNFRSF14 (Figures 6A, B). TMB, defined as the total count of base mutations per million tumor cells, is recognized for inducing tumor-specific, highly immunogenic antibodies and is emerging as a promising biomarker for immunotherapy response in cancer patients (16). Noteworthily, PRKAR1B expression showed a significant correlation with TMB in HNSC (Figure 6C). Additionally, IC50 values for various chemotherapeutic drugs were evaluated based on PRKAR1B expression levels using data derived from the Cancer Genome Project database. Lapatinib and Erlotinib exhibited greater efficacy in the high-PRKAR1B expression group, whereas Bleomycin, Doxorubicin, Vinorelbine, and Etoposide were more effective in the low-PRKAR1B expression group. These findings collectively suggest that PRKAR1B expression may predict chemotherapy responses in HNSC, reinforcing its potential as a biomarker for personalized treatment strategies (Figure 6D). Immune checkpoint genes (ICGs) are determinants of immunotherapy efficacy, and analyzing the clinical data and expression of several ICGs can assist in the identification of potential therapeutic targets for personalized treatment. To explore the specific impact of PRKAR1B on immunotherapy, the effect of ICIs on PRKAR1B was explored. Immunophenoscores were utilized to predict patient responses to different immune checkpoint inhibitor combinations. High PRKAR1B expression was associated with a higher IPS for anti-CTLA4 therapy, signaling that high PRKAR1B expression may correlate with improved immune response to ICI therapy (Figure 6J).

Furthermore, the distribution of PRKAR1B was analyzed using single-cell transcriptome data from GSE103322. A total of 20 cell clusters were identified and categorized into 11 distinct cell types based on marker gene expression: CD4Tconv, CD8+ T cells, CD8Tex, Endothelial cells, Fibroblasts, Malignant cells, Mast cells, Mono/Macro cells, Myocytes, Myofibroblasts, and Plasma cells (Figures 7A, B). PRKAR1B was highly expressed in CD8Tex, Endothelial, and Mast cells, whereas lower expression levels were detected in Myocytes (Figures 7C, D).

To validate the role of PRKAR1B in HNSC, siRNA was used to knockdown its expression in SCC9 and SCC25 cells. Transfection efficiency was validated by RT-qPCR (Figure 8A) and Western blotting (Figure 8B) and further validated by immunofluorescence. Notably, upon transfection of PRKAR1B siRNA, significantly reduced PRKAR1B levels were observed in both SCC9 and SCC25 cells (Figure 8C). CCK8 and EdU labeling assays demonstrated that PRKAR1B knockdown inhibited the proliferative abilities of SCC9 and SCC25 cells (Figures 8C, D). In addition, Transwell and wound healing assays showed that PRKAR1B silencing significantly suppressed the migratory abilities of SCC9 and SCC25 cells compared with controls. Taken together, these results confirmed that PRKAR1B knockdown may suppress the metastatic capabilities of SCC9 and SCC25 cells (Figures 9A, B). Moreover, PRKAR1B knockdown upregulated E-cadherin expression and downregulated N-cadherin expression in SCC9 and SCC25 cells (Supplementary Figure 2). Finally, immunohistochemical staining was performed to examine PRKAR1B protein expression levels in both HNSC and normal tissue samples, revealing that PRKAR1B protein was highly expressed in HNSC tissues (Figure 9C).