AMPKα2 promotes tumor immune escape by inducing CD8+ T-cell exhaustion and CD4+ Treg cell formation in liver hepatocellular carcinoma

RNA-seq data for 31 normal tissues and LIHC samples were obtained from the Genotype-Tissue Expression (GTEx, https://​gtexportal.​org/​home/​) and the Cancer Genome Atlas (TCGA, https://​www.​cancer. gov/tcga) databases, respectively. The normal tissue types included adipose, adrenal gland, bladder, blood, blood vessel, bone marrow, brain, breast, cervix uteri, colon, esophagus, fallopian tube, heart, kidney, liver, lung, muscle, nerve, ovary, pancreas, pituitary, prostate, salivary gland, skin, small intestine, spleen, stomach, testis, thyroid, uterus, and vagina. Single-cell RNA-seq (scRNA-seq) data for LIHC were collected from the China National Genebank Database (CNGBdb, https://​db.​cngb.​org/​search/​project/​CNP0000650); accession code: CSE0000008).

Based on prior reports in the literature, 106 AMPK substrates were extracted [11]. The expression correlation between AMPK subunits and AMPK substrates in 31 normal tissues was determined by Pearson correlation analysis (Correlation coefficients > 0.2).

To explore differences in the expression of AMPK subunits between tumor samples and their matched normal tissue controls in pan-cancer, we used the gene set cancer analysis (GSCA) online tool (http://​bioinfo.​life.​hust.​edu.​cn/​GSCA/​). A false discovery rate-adjusted p-value less than 0.05 indicated a significant difference.

Based on the median PRKAA2 level, patients were divided into high- and low-expression subgroups. The R packages “survival” and “survminer” were employed for Kaplan–Meier survival analyses. The abundance of each infiltrating immune cell type was analyzed using a single-sample gene set enrichment analysis. Wilcoxon signed-rank test was performed and p less than 0.05 implied a significant difference.

We performed cell communication analysis using CellphoneDB [20]. Average expression levels were calculated based on the annotated ligand-receptor pairs attained from the STRING database. The ligand-receptor pairs with p < 0.05 values were identified, and the interactions between two cell types using these identified pairs were analyzed. Cytokines play a critical role in cell communication; thus, cytokine signaling, based on the transcriptomic profiles, was determined using CytoSig.

To detect the association between PRKAA2 expression and T-cell evolution during tumor progression in single cells, a Pseudotime analysis was performed using R packages “monocle2” [21]. The monocle subject was built by applying the function “newCellDataSet”. Trajectory analysis was performed based on the differentially expressed genes determined by the R package “Seurat”. The “reduceDimension” function was used for dimensionality reduction, and cells were placed on Pseudotime trajectories using the “orderCells” functions.

Single-cell metabolic activity was evaluated using the R package “scMetabolism” based on a previously reported method [22]. Differences in the metabolic pathway scores between subgroups with high or low PRKAA2 expression were analyzed using Wilcoxon signed-rank test, and p less than 0.05 implied a significant difference.

The human LIHC cell line, HepG2, was acquired from the American Type Cell Culture Collection. Cells were grown in Dulbecco Modified Eagle Medium (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) containing 10% fetal bovine serum (FBS, BI) at 37 °C and 5% CO₂. PRKAA2 knockout cells were generated by transfecting cells with the lentiviral-based short hairpin RNA (shRNA) vector pGPU6/GFP/Neo (Genechem, Shanghai, China). The shRNA sequences were as follows: shRNA-1: GTGGCTTATCATCTTATCATT and shRNA-2: GTCATCCTCATATTATCAAAC. PRKAA2 mRNA levels were measured using quantitative real-time PCR (qPCR). Total RNA was extracted using an RNA extraction reagent (Takara, 9108, Japan), and 2X Super SYBR Green qPCR Master Mix (ES Science, Guangzhou, China) was used to detect PRKAA2 levels. The qPCR primer sequences for PRKAA2 and GAPDH were as follows: PRKAA2 forward, 5′-CGGGTGAAGATCGGACACTA-3′;PRKAA2 reverse, 5′-TCCAACAACATCTAAACTGCGA′; GAPDH forward, 5′-GACCTGACCTGCCGTCTA-3′;and GAPDH reverse, 5′-AGGAGTGGGTGTCGCTGT-3′.

Cells (8000/well) were seeded into 96-well plates (Servicebio, WuHan, China), and proliferation was evaluated using Cell Counting Kit-8 (Biosharp Biotechnology, Beijing, China) reagent. The optical density was read at 450 nm. For clone formation assays, cells were cultured for 14 days, fixed with 4% paraformaldehyde for 15 min, and stained with 0.1% crystal violet for 15 min (Meilunbio, Dalian, China). For scratch assays a linear wound was created in a monolayer of serum-starved cells using a 10-μL pipette tip, and cell coverage across this line was determined. For Transwell migration assays, cells were seeded into the upper chamber (8 μm; BIOFIL, Guangzhou, China) and incubated with serum-free medium. The lower chamber contained medium with 10% FBS. After 24 hours, cells were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet.

The expression levels of AMPK subunits, including AMPKα (α1 and α2), AMPKβ (β1 and β2), and AMPKγ (γ1, γ2, and γ3), in different human tissues were assessed using GTEx bulk RNA-seq data. PRKAA1, PRKAB1, and PRKAG1 were the main subunits expressed in almost all tissue types. PRKAA2, PRKAB2, and PRKAG3 were the predominant subunits expressed in muscle tissues and PRKAG2 was the major subunit expressed in heart tissues (Fig. 1). Thus, the expression of AMPK subunits was tissue-specific.

AMPK exerts biological functions by activating distinct substrates. Thus, gene expression correlations between different subunits of AMPK and AMPK substrates were assessed. As shown in Fig. 2A, different AMPK subunits significantly correlated with distinct substrates. PRKAA1 expression correlated with BRAF, C18orf25, EEF2K, and EP300 expression. PRKAA2 expression correlated with CRTC2, TNNI3, KCNA5, and TFEB expression. These results indicate that AMPK subunits may be characterized by substrate specificity. Differences in the enriched pathways for different AMPK subunits were determined using KEGG analysis. Diverse AMPK isoforms participated in different regulatory programs (Fig. 2B). These results indicate that the substrate specificity of AMPK isoforms may lead to functional differences between AMPK isoforms.

The expression levels of AMPK subunits in tumor tissue and matched adjacent normal tissue were compared using the GSCA database. Details of datasets were shown in Table S1. PRKAA2 was differentially expressed in most tumor types (Fig. 3A). Of note, PRKAA2 was significantly upregulated in LIHC, and this upregulation was in the TCGA-LIHC cohort (Fig. 3B). Subsequent Kaplan-Meier survival analyses indicated that high PRKAA2 expression was linked to unfavorable prognosis in patients with LIHC (Fig. 3C). The TME status in response to different PRKAA2 expression patterns was determined by calculating the infiltration abundances of 23 immune cell types. Patients with low PRKAA2 expression displayed a significantly higher degree of immune cell infiltration, indicative of an immune hot phenotype, compared with patients with high PRKAA2 expression, which exhibited an immune cold phenotype lacking immune cell infiltration (Fig. 3D).

The scRNA-seq dataset for LIHC was obtained from the CNGBdb (CSE0000008). The dataset included 12 primary tumors and 6 relapsed tumors (Fig. 4A). Ten cell subgroups were identified using the analysis of t-SNE clustering of single-cell samples. Based on the gene expression of cell-type specific markers, cell subgroups were annotated as known cell lineages, including immune cells (B cells, myeloid cells, NK cells, pDC, plasma cells, and T cells), malignant cells, hematopoietic stem cells, endothelial cells, and epithelial cells (Fig. 4B). PRKAA2 was predominantly expressed in malignant cells. Malignant cells were extracted and clustered. As shown in Fig. 4C, a total of 10 subgroups were obtained, and cells with high PRKAA2 expression clustered mainly in subgroups 1, 6, and 7, indicating heterogeneity in PRKAA2 expression. Malignant cells in the high PRKAA2 expression group exhibited a more malignant phenotype than cells in the low PRKAA2 expression group (Fig. 4D). Gene set variation analysis (GSVA) revealed that signaling pathways associated with tumor progression, including PI3K/Akt/mTOR signaling, Notch signaling, angiogenesis signaling, and epithelial–mesenchymal transition (EMT), were significantly enriched in malignant cells with high PRKAA2 expression (Fig. 4E). PRKAA2 expression was higher in malignant cells from relapsed tumors compared with PRKAA2 expression in malignant cells from primary tumor samples. Genes involved in TGF-β signaling and EMT were also higher in malignant cells from relapsed tumors compared with malignant cells from primary tumor samples (Fig. 4F). Conversely, DNA repair and p53 signaling pathways were more pronounced in primary tumor samples. Finally, cyclins were upregulated in malignant cells with high PRKAA2 expression, suggesting that these cells are in the activated cell cycle state (Fig. 4G). Altogether, these findings reveal that PRKAA2 may contribute to LIHC development.

To evaluate the relationship between PRKAA2 and the LIHC immune microenvironment, patients (n = 18) were divided into high and low PRKAA2 expression groups. The proportion of immune cells was significantly lower in tumors with high PRKAA2 expression compared with the proportion of immune cells in the low PRKAA2 expression group (43.8% vs. 75.84%; Fig. 5A). Single-sample GSVA to detect differentially expressed genes between the two subgroups revealed that the IFN-γ response pathways were enriched in the malignant cells of tumors with low PRKAA2 expression (Fig. 5B). This finding was confirmed using GO enrichment analysis (Fig. 5C). Notably, GO analysis also revealed that the MHC protein complex assembly signaling pathway was significantly upregulated in the low PRKAA2 expression subgroup. The interferon-gamma (IFN-γ) signaling pathway plays a key role in regulating MHC-I expression. Thus, the levels of MHC-I genes were assessed in the two subgroups. Malignant cells with high PRKAA2 expression exhibited low levels of MHC-I gene expression (Fig. 5D), indicative of weak immunogenicity. In agreement with this finding, the high-expression group also had a significantly low immunogenic cell death pathway score. The functional consequence of high PRKAA2 expression was increased immune escape (Fig. 5E).

To explore the relationship between PRKAA2 and the evolution of T cells, we first identified 6179 T cells. All T cells were clustered into eight subgroups and annotated as CD4⁺ T cells (cytotoxic CD4⁺ T cells, exhausted CD4⁺ T cells, native CD4⁺ T cells, CD4⁺ T helper (Th) cells, and CD4⁺ T regulatory (Treg) cells) and CD8⁺ T cells (cytotoxic CD8⁺ T cells, exhausted CD8⁺ T cells, and native CD8⁺ T cells) (Fig. 6A). T cells with high PRKAA2 expression were primarily enriched in exhausted T cells and CD4⁺ Treg cells. These results were confirmed by analyzing the composition and proportion of T-cell types in high and low PRKAA2 expression subgroups (Fig. 6B). Compared with the other CD8+ T-cell types, exhausted CD8+ T cells expressed higher levels of PRKAA2 (Fig. 6C-D). In addition, the expression levels of multiple immune checkpoint proteins, including CTLA4, HAVCR2, PDCD1, TIGIT, CD27, and LAG3, were higher in the exhausted CD8+ T-cell subgroup with high PRKAA2 expression (Fig. 6E-G), implying that the functional exhaustion of CD8+ T cells may be a potential mechanism for PRKAA2-mediated cancer cell immune evasion. Immune cells in the tumor microenvironment evolve with the progression of the tumor, and this dynamic process can be described by the monocle algorithm. Pseudotime and trajectory analyses showed that CD8+ T cells tended to be exhausted during tumor progression, which was associated with high expression of PRKAA2 (Fig. 6H-J). Thus, PRKAA2 may promote tumor progression by contributing to CD8+ T-cell exhaustion.

CD4+ Treg cells, which are an important component of CD4+ T-cell types, are mainly responsible for maintaining immunological tolerance and homeostasis. High PRKAA2 expression in CD4+ Treg cells significantly enhanced CD27, TIGIT, TNFRSF9, ICOS, TNFRSF4, CTLA4, TNFRSF18, and CD28 expression levels (Fig. 7A-D). This suggests that CD4+ Treg cells mediate the promotion of tumor immune escape by PRKAA2. Differentiation trajectory analysis of CD4+ T cells revealed that PRKAA2 may be involved in the formation of CD4+ Treg cells (Fig. 7E-G). Increased metabolism of Treg cells is a key factor in maintaining their immunosuppressive effects. Thus, the enrichment of metabolic-related signaling pathways in Treg cells and other T-cell types was examined. The metabolic pathways enriched in Treg cells included beta-alanine metabolism, citrate cycle (TCA cycle), glycolysis/gluconeogenesis, and oxidative phosphorylation (Fig. 7H). These signaling pathways were also enriched in T cells with high PRKAA2 expression (Fig. 7I). Taken together, our findings suggest that PRKAA2 may contribute to CD8+ T-cell exhaustion and the formation of CD4+ Treg cells to facilitate immune escape.

The interaction between malignant cells and T cells within the TME is crucial for tumor progression. Therefore, a cell-cell communication network was built to explore the role of PRKAA2 in cell communication. The expression levels of chemokines from malignant cells, including CXCL12, CXCL10, and CXCL1, were significantly changed by PRKAA2 (Fig. 8A). In addition, high PRKAA2 expression suppressed costimulatory molecules, including TNFSF4, TNFSF10, ICAM3, and APP, generated by malignant cells (Fig. 8B). However, coinhibitory signals were enhanced in malignant cells with high PRKAA2 expression (Fig. 8C). Of note, synergistic interaction between T cells and malignant cells with high PRKAA2 expression activated Treg cells and promoted T-cell exhaustion (Fig. 8D). Conversely, induction of multiple ligand-receptors pairs, including CXCL10/DPP4, CXCL1/CXCR2, TNFSF4/TNFRSF4, and FAM3C/CLEC2D, reinforced the accumulation of T cells in response to low PRKAA2 expression (Fig. 8E).

Cytokines regulate cell-cell communications within the immune system. Thus, the relationship between PRKAA2 and cytokine signaling at the single-cell level was investigated. Seventeen cytokine-related pathways were activated in malignant cells with high PKRAA2 expression; 10 of these pathways were significantly and positively associated with marker genes from T cells (Fig. 8F). Differences in the activation of cytokine signaling pathways, such as IL6 and IL3, between high and low PRKAA2 expression subgroups contributed to the distinct communication behaviors in T cells and malignant cells (Fig. 8G). Consequently, the functional role of PRKAA2 in reshaping cell-cell interactions and cytokine signaling activity in the TME suggests that PRKAA2 is a main regulator of T-cell exhaustion in LIHC.

The role of PRKAA2 in the metabolic reprogramming of malignant cells was investigated. Compared to other cell types, metabolic pathways, including glycolysis/gluconeogenesis and TCA cycle, were upregulated in malignant cells (Fig. 9A). Overall metabolic pathway abundance was significantly higher in malignant cells with high PRKAA2 expression compared with malignant cells with low PRKAA2 expression (Fig. 9B). Differences in each metabolic pathway activity between high- and low-expression subgroups are shown in Fig. 9C. Differentially expressed genes in the two subgroups were extracted (Fig. 9D) and a KEGG analysis was conducted. The analysis shows that genes were mainly enriched in metabolism-related signaling pathways such as oxidative phosphorylation, glycolysis/gluconeogenesis, and TCA; oxidative phosphorylation was the most enriched pathway (Fig. 9E). Oxidative phosphorylation was closely related to glycolysis and hypoxia, and glycolysis was substantially linked to the response to hypoxia (Fig. 9F). We also observed that PRKAA2 was significantly associated with hypoxia. Thus, associations between oxidative phosphorylation and glycolysis and between oxidative phosphorylation and hypoxia indicate that coupling between aerobic respiration and hypoxia-associated pathways may be a feature of malignant cells in the TME in response to high PRKAA2 expression.

To explore the functional role of PRKAA2 in LIHC progression, we knocked PRKAA2 down in LIHC HepG2 cells using shRNAs (Fig. 10A). Proliferation was significantly inhibited in PRKAA2-deficient HepG2 cells compared to control cells (sh-Control) (Fig. 10B). Clonogenic ability was also significantly decreased in response to PRKAA2 silencing (Fig. 10C). PRKAA2 depletion impaired the healing ability of HepG2 cells compared to sh-Control cells (Fig. 10D). Migratory and invasive behaviors of HepG2 cells were remarkably limited by PRKAA2 depletion in Transwell assays (Fig. 10E). Thus, PRKAA2 may be critical for LIHC progression and metastasis.