AFF4 promotes tumor progression and cisplatin resistance by modulating the PTEN/PI3K/AKT/mTOR axis to accelerate glycolysis in lung adenocarcinoma

Abstract

Background

Although aerobic glycolysis contributes to malignancy and drug resistance in human cancers, the vital regulators of glycolysis in lung adenocarcinoma (LUAD) remain largely unknown. Transcription factor AF4/FMR2 family member 4 (AFF4) is the scaffolding protein of the super elongation complex (SEC) and regulates the transcription of cancer-related genes. However, the role of AFF4 in glycolysis and LUAD development remains unidentified.

Methods

AFF4 expression was assessed in LUAD cells and tissues using bioinformatics analysis, western blotting, and immunohistochemical staining. Changes in cell proliferation, migration, and invasion were determined using in vitro and in vivo loss- and gain-of-function assays. Additionally, glycolysis levels were assessed using metabolite determination assays of glucose and lactate. The underlying mechanisms were elucidated via transcriptome sequencing, cleavage under targets (CUT) &Tag, dual-luciferase reporting assay, and a series of rescue experiments.

Results

AFF4 was overexpressed in wild-type and cisplatin-resistant LUAD cells and acted as a prognostic indicator in patients with LUAD. AFF4 enhanced the tumorigenic characteristics and cisplatin resistance of LUAD cells by accelerating glycolysis. Meanwhile, glycolysis inhibition restored the AFF4 overexpression-induced increase in cell proliferation and migration and rendered AFF4-overexpressing LUAD cells sensitive to cisplatin. Mechanistically, AFF4 promoted glycolysis by modulating the phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT)/mammalian target of rapamycin (mTOR)/ signaling pathway. AFF4 downregulated phosphatase and tensin homolog (PTEN) expression by directly targeting its promoter, activating the PI3K/AKT/mTOR pathway. Additionally, transcription factor Yin Yang 1 (YY1) upregulated AFF4 by binding to its promoter, further influencing glycolysis and oncogenesis.

Conclusion

AFF4 drives metabolic reprogramming, tumor progression, and cisplatin resistance through PTEN-mediated activation of the PI3K/AKT/mTOR signaling pathway, highlighting AFF4 inhibition as a potential therapeutic strategy in LUAD.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13578-025-01455-1.

Background

Lung and bronchus cancer was responsible for the highest estimated cancer-related deaths in 2023, according to the cancer statistics of the USA [1]. Lung adenocarcinoma (LUAD) is among the most complex cancers, predominantly affecting females and smokers [2–4]. Despite advancements in gene-targeted therapies, such as epidermal growth factor receptor tyrosine kinase inhibitors and immunotherapies using programmed cell death protein 1 (PD1)/programmed death-ligand 1 (PDL1) monoclonal antibodies, poor prognoses and high recurrence rates persist in patients with LUAD. Insufficient targeted treatment and drug resistance are the primary limitations in LUAD management [5–7]. Therefore, identifying novel therapeutic targets is a critical research focus.

Aerobic glycolysis, a deviation from oxidative phosphorylation observed in healthy cells, signifies a distinct aberration in the energy metabolism of tumor cells and is recognized as a hallmark of cancer malignancy [8]. To supply sufficient adenosine triphosphate (ATP) energy to support malignant behaviors and improve glycolysis efficiency, cancer cells typically upregulate the expression of vital glycolytic enzymes and glucose transporters, such as glucose transporter 1 (GLUT1) [9], hexokinase 2 (HK2) [10], 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 3 (PFKFB3) [11], and pyruvate kinase M2 (PKM2) [12]. Increased glycolysis causes lactic acid accumulation, creating an acidic tumor microenvironment in tumor cells, which promotes cell proliferation, metastasis, and host immune escape [13–15]. Additionally, there is growing recognition that glycolysis contributes to chemotherapy resistance in cancer cells by impeding apoptosis and sustaining redox homeostasis [16, 17]. He et al. reported that reduced glycolysis caused by HK2 downregulation promotes cisplatin sensitivity of lung cancer cells [18]. Thus, prompting glycolysis inhibition is a potential treatment strategy for patients with cisplatin-resistant LUAD. However, the intricate molecular mechanisms regulating aerobic glycolysis in LUAD require further elucidation.

The super elongation complex (SEC) is a dynamic regulatory complex comprising AF4/FMR2 family members 1/4 (AFF1/AFF4), eleven-nineteen Lys-rich leukemia (ELL) family members 1/2/3 (ELL1/2/3), eleven-nineteen leukemia (ENL), ALL1-fused gene from chromosome 9 (AF9), and positive transcription elongation factor b (p-TEFb). The SEC regulates the release of paused RNA polymerase II (RNPII) from its docking site to promote efficient transcriptional elongation. Specifically, AFF1 and AFF4 are scaffold proteins belonging to the ALF family that provide binding sites for other components. Notably, every SEC component can integrate with histone methyltransferase mixed lineage leukemia (MLL) to form fusion proteins, causing the occurrence and development of mixed lineage leukemia; this highlights the relationship between SEC constituents and disease pathology [19–22]. Moreover, we previously found that AFF1 suppresses cell proliferation and migration through the neurotensin (NTS)–carbamoyl-phosphate synthase 1 (CPS1)–glutathione peroxidase 2 (GPX2) axis in LUAD cells, suggesting the therapeutic potential of ALF family genes in LUAD [23]. Furthermore, AFF4 functions as a pivotal facilitator of melanoma cell progression by modulating c-Jun activity and augmenting the oncogenic potential of head and neck squamous carcinoma cells by regulating SRY-box 2 (SOX2) expression. Conversely, AFF4 reportedly impedes the metastatic spread of colorectal cancer by enhancing CDH1 expression. Hence, AFF4 has a complex and multifaceted influence on cancer [24–26]. Despite such insights, the precise roles and underlying molecular mechanisms of AFF4 in LUAD remain unclear.

In the present study, the oncogenic role of AFF4 in LUAD was uncovered and its pivotal role in the promotion of cell proliferation, migration, and cisplatin resistance was elucidated. Our results revealed that AFF4 augmented the PI3K/AKT/mTOR signaling pathway by directly suppressing PTEN transcription, subsequently stimulating aerobic glycolysis and tumor development. Additionally, YY1 was identified as a vital upstream regulator of AFF4 in regulating glycolysis and tumor development. Therefore, AFF4 presents a promising therapeutic target for LUAD treatment in both primary and cisplatin-resistant tumor cells.

Methods

Cell lines and plasmids

The human embryonic kidney cell line 293T, human normal lung epithelial cell line Beas-2B, human LUAD cell lines A549, H1299, PC9, H358 and H1975, and the cisplatin-resistant cell line A549 DDP were obtained from Procell Life Science & Technology Co., Ltd. (Wuhan, China). 293T, A549, H1299, PC9 and H358 cells were cultivated in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 100 U/mL penicillin/streptomycin. H1975 cells were maintained in RPMI-1640 medium supplied with 10% FBS and 1% penicillin–streptomycin. A549 DDP cells were cultured in Ham’s F12K complete medium with 2 µg/mL cisplatin. All cells were cultured in an incubator with 5% CO₂ at 37 °C.

For genetic manipulation, interfering shRNAs were engineered into the PLKO.1 lentiviral vector using AgeI and EcoRI restriction endonucleases. The coding domain sequences of AFF4 and YY1 were inserted into the pCDH plasmid or the pcDNA5 vector via homologous recombination methods. Wild type (WT) and mutant sequences of the PTEN and AFF4 promoters were incorporated into the pGL3-basic plasmid using homologous recombination techniques. The detailed sequences of all PCR primers used in the procedures are presented in Table S1.

Clinical sample collection and immunohistochemistry (IHC)

Totally 8 pair of LUAD tissues and their adjacent tissues were collected from patients who underwent tumor resection, with none having received treatment prior to surgery. All patients were diagnosed with lung adenocarcinoma based on pathological examination. Informed consent was obtained from all study participants.

For IHC staining, tissue samples were fixed in formalin and embedded in paraffin. Following dehydration with 70% ethanol, the tissues were incubated with 0.3% hydrogen peroxide for 30 min at 25 °C to block endogenous peroxidase activity. The sections were then incubated with diluted blocking serum, followed by overnight incubation with the primary antibodies at 4 °C, and subsequently with fluorescently conjugated secondary antibodies for 2 h at 25 °C. The intensity and distribution of the positively stained areas in the IHC analysis were quantified using ImageJ software. Briefly, the IHC Toolbox plugin was employed to select the color blocks corresponding to the positive areas. Subsequently, the entire region of positive staining was outlined to ensure that all positive areas were accurately included. Optical densities were then analyzed and recorded. This process was repeated for each immunohistochemical image to compile average optical density data for all positive areas. Finally, differential analysis and Pearson correlation analysis were conducted using GraphPad Prism 9.0 software. The experimental protocols for this study were approved by local research ethics committee (approval number: KY202416).

Lentivirus-mediated RNA inference and gene overexpression

PLKO.1 or pCDH, together with psPAX2 and pMD2.G plasmids, were co-transfected into 293T cells using Lipofectamine 2000 (CN2541156, Invitrogen). After 96 h of transfection, the virus was harvested and passed through a 0.45-µm filter. Target cells were seeded in six-well plates (5 × 10⁴ cells/well) a day before infection. The virus, supplemented with polybrene (BL628A, Biosharp) at a final concentration of 8 µg/mL, was added to the cells. The medium was replaced after 24 h of infection, and a final concentration of 2 µg/mL puromycin (BS111-25 mg, Biosharp) was added. After 72 h of infection, cells were passaged or collected for further analyses.

Real-time quantitative PCR (qPCR) and transcriptome sequencing

Total RNA was extracted from LUAD cells using an RNA kit (SparkJade, China). For transcriptome sequencing, the extracted RNA was treated with DNase I for 15 min at 25 ℃ to remove contaminating DNAs. Subsequently, cDNA synthesis was performed using a reverse transcription kit (Takara, Japan), following the manufacturer’s instructions. Subsequent qPCR analyses were conducted using the SYBR Green Master Mix Kit (Yeasen, China). The sequences of all qPCR primers used in the experiments are presented in Table S1.

Western blotting

Total protein was isolated from normal lung epithelial and LUAD cells using RIPA lysis buffer (Beyotime, China). The protein concentrations of the lysates were quantified using a bicinchoninic acid (BCA) Protein Assay Kit (Vazyme, China). Proteins were separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto a polyvinylidene fluoride membrane (PVDF). The membrane was blocked in 5% skim milk diluted in tris buffered saline with Tween 20 (TBST), followed by incubation with primary antibodies at 4 ℃ overnight and appropriate secondary antibodies directed against rabbit or mouse IgG at 4 ℃ for 2 h. Immunoreactive bands were visualized using chemiluminescence detection. The details of the antibodies are provided in Table S2. The original bands and full-length blots are available in the supplementary materials.

Cell proliferation assay

Cell counting kit-8 (CCK-8; K1018, Apexbio) was used to assess cell proliferation. In brief, after infection for 72 h, 2 × 10³ LUAD cells in 100 µL of DMEM were seeded into each well of a 96-well plate, with 15 wells allocated for each sample. To each well, 10 µL of CCK-8 reagent was added, and the plate was incubated at 37 °C with 5% CO₂ for 2 h. The optical density (OD) was measured at 450 nm using a microplate photometer (Thermo Fisher Scientific, Waltham, MA, USA). The CCK-8 assay was repeated daily for 5 d; the results were used to plot the cell viability curve, with the time points designated along the abscissa and the corresponding OD values on the ordinate.

Colony formation assay

Colony formation assays were conducted in six-well plates (1 × 10³ LUAD cells/well) and incubated at 37 °C with 5% CO₂ for 10–14 d. The culture medium was replaced every 3 d to ensure optimal growth conditions. Following incubation, the colonies were stained with 1% crystal violet and enumerated for statistical analysis.

Flow cytometric analysis of Ki67 staining, cell cycling, and apoptosis

AFF4-deficient or AFF4-overexpressing LUAD cells were cultured in a six-well plate (2 × 10⁵ cells/well). For Ki67 staining, the cells were fixed with cold 70% ethanol and incubated with Ki67 antibody at 4 °C for 2 h. For the cell cycle assay, cells were stained with propidium iodide (PI), following the manufacturer’s protocol. Cell apoptosis was assessed using the Annexin V-fluorescein isothiocyanate Apoptosis Detection Kit (KTA0002, Abbkine), with cells collected in a binding buffer and subsequently analyzed on a flow cytometer (Beckman, California, USA) with software FlowJo.

Wound-healing assay

Cells were seeded into six-well plates at a density of 2 × 10⁵ cells/well and incubated at 37 °C with 5% CO₂ for 48 h, followed by starvation in DMEM containing 0.1% FBS for 24 h. The cell monolayer was scratched using a 200-µL plastic pipette tip and washed once with phosphate-buffered saline (PBS). Cells were further incubated in FBS-free DMEM and monitored every 12 h post-wounding using an inverted microscope. Scratch closure was quantitatively assessed by measuring the relative distance between the edges of the scratch using the ImageJ software.

Transwell cell migration and invasion assay

Cell migration and invasion assays were conducted using 24-well plates with transwell chambers containing 8-µm pores. Briefly, for the migration assay, 5 × 10⁴ cells were suspended in 200-µL of serum-free medium and seeded in the upper chamber. For the invasion assay, 5 × 10⁴ cells were placed in the upper chamber pre-coated with matrigel (BD Biosciences). The lower chamber was filled with 500 µL of medium containing 10% FBS as a chemoattractant. After 14–16 h of incubation, the chambers were removed, and the cells were stained with 0.1% crystal violet. Subsequent imaging was performed using an inverted microscope.

Glucose uptake assay

A commercial glucose assay kit (#F006-1-1, Nanjing Jiancheng Bioengineering) was used to quantify glucose consumption, and the results were normalized to the cell count. For the 2-(7-nitro-2,1,3-benzoxadiazol-4-yl)-d-glucosamine (2-NBDG) uptake test, lentivirus-infected cells were seeded in six-well plates. Subsequently, 72 h post-infection, the cells were incubated with 100 µM 2-NBDG for 30 min in a glucose-free medium. The cells were imaged with a fluorescence microscope, and the fluorescence intensities were quantified using ImageJ software. Briefly, the fluorescence intensity for each group is calculated using the following formula: Mean fluorescence intensity (Mean) = Total fluorescence intensity of the region (IntDen) / Area of the region (Area). The mean fluorescence intensity of the experimental group is then compared to that of the control group, normalized, and presented in bar graphs for statistical analysis.

Lactate production detection assay

Lactate production in LUAD cells was assessed using a lactate detection kit (#A019-2-1, Nanjing Jiancheng Bioengineering), following the manufacturer’s protocol. The absorbance was determined using a Bio-Rad microplate reader at the specified wavelength according to the manufacturer’s instructions.

Xenograft models

Five-week-old female nude mice (BALB/c, Beijing Laboratory Animal Research Center, China) were used to establish the xenograft tumor models. For proliferation studies, 2 × 10⁶ cells were subcutaneously inoculated into the right flank of the mice (n = 4/group). Tumor dimensions were measured with a vernier caliper every 3 d, and the volume was calculated using the formula: V = a × b²/2, where ‘a’ is the long diameter, and ‘b’ is the short diameter of the tumor.

In metastasis studies, 5 × 10⁶ cells were administered intravenously via the tail vein (n = 4/group). Lung tissues were excised from mice exhibiting abnormal behaviors with the humane euthanasia by cervical dislocation. Pulmonary nodules were counted, and lung tissues were prepared for subsequent histological staining.

IHC staining was performed on the sections using the primary antibodies listed in Table S2, according to the previously described procedures. The lung tissue sections from metastasis models were used for hematoxylin/eosin (H&E) staining. All experimental protocols were reviewed and approved by Research Ethics Committee of Anhui Medical University (approval number: 20240822).

Dual-luciferase reporter assay

The pRL-SV40 vector, along with the ectopic expression vector pcDNA5 and pGL3-basic plasmid, were transfected into 293T cells using Lipofectamine 2000. After 48 h of transfection, cells were lysed using a luciferase assay kit (E1500, Promega). The luciferase signals for Firefly and Renilla were measured separately. The ratio of Firefly to Renilla luciferase activity was calculated to determine the luciferase activity.

Cleavage under targets and tagmentation sequencing (CUT&Tag-seq)

The Hyperactive Universal CUT&Tag Assay Kit (TD904, Vazyme) was used to prepare a library, as per the manufacturer’s instructions. Briefly, 1 × 10⁵ A549 or PC9 cells were collected and attached to ConA beads for 10 min at 25 °C before incubation with 1 µg of the primary antibody at 4 °C overnight. The secondary antibody was subsequently added, and the cells were incubated for 1 h at 25 °C. Next, 0.04 µM pA/G–Tnp was added for 1 h at 25 °C to cleave the genome and attach a unique adaptor sequence for library construction. Proteinase K, buffer LB, and DNA extract beads were used to inhibit the reaction. Next, the cells were separated with a magnet, washed twice with 80% ethanol, and the DNA was eluted in double-distilled water for sequencing.

Cleavage under target and release using nuclease and quantitative PCR (CUT&RUN-qPCR)

The NovoNGS^® CUT&Tag 4.0 High-Sensitivity Kit (Novoprotein, N259-YH01) was used to prepare the qPCR samples, as per the manufacturer’s instructions. Briefly, the same steps conducted for CUT&Tag were performed to prepare a library. Subsequently, a 5 µL DNA sample was used to construct the library and cleaned using magnetic beads. The supernatant was adsorbed and collected with a magnetic stand for subsequent qPCR assays.

Bioinformatic analysis

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) signaling pathway analyses were conducted using the Sangerbox 3.0 (http://sangerbox.com/). The analyses were configured with minimum and maximum gene sets of 5 and 5000, respectively. Gene Set Enrichment Analysis (GSEA) was conducted using the GSEA 4.1.0 (https://www.gsea-msigdb.org/gsea/index.jsp) and msigdb_v7.4 gene sets. Statistical significance was determined with a P-value < 0.05 and a false discovery rate (FDR) < 0.25.

CUT&Tag data analysis was executed on a Linux system. First, quality control was conducted using fastp with default parameters following the raw data download. Alignment was performed against the UCSC hg19 reference genome using bwa, and peak calling was performed using macs3. Finally, the bam files were converted to bigwig format using bedGraphToBigWig, and the resulting bigwig files were visualized using IGV software.

JASPAR (http://jaspar.genereg.net/), GTRD (http://gtrd.biouml.org/), TFDB (https://guolab.wchscu.cn/AnimalTFDB), and PROMO (https://alggen.lsi.upc.es/cgi-bin/promo_v3/promo/promoinit.cgi?dirDB=TF_8.3) were used to identify the potential transcription factors binding to AFF4 promoter.

Data analysis

Statistical analyses were performed using GraphPad Prism 9.0 (GraphPad Software Inc., San Diego, CA, USA). The t-tests (and nonparametric tests) were used for comparisons between two groups. In addition, a two-way analysis of variance (ANOVA) was performed for analyses involving multiple groups. The levels of statistical significance were denoted as follows: *p < 0.05, ** p < 0.01, *** p < 0.001, and **** p < 0.0001, with ‘ns’ indicating lack of significance. All experiments were conducted at least twice to ensure the reliability and reproducibility of the results.

Results

AFF4 expression was upregulated in LUAD and aided in predicting poor patient prognosis

To evaluate the potential role of AFF4 in LUAD pathogenesis, survival analysis was conducted using the Kaplan–Meier Plotter online database. Elevated AFF4 expression was correlated with reduced overall survival time in patients with LUAD (Fig. 1A). Subsequently, two distinct Gene Expression Omnibus (GEO) datasets, GSE115002 and GSE10072, were analyzed to quantify the mRNA expression levels of AFF4. The analysis revealed a marked upregulation of AFF4 in LUAD specimens compared with their non-malignant tissue counterparts (Fig. 1B). The endogenous mRNA and protein expression levels of AFF4 in three LUAD (A549, PC9, and H1299) and normal pulmonary bronchial epithelial (Beas-2B) cell lines were then examined. AFF4 was expressed at relatively higher levels in LUAD cell lines than that in Beas-2B cells (Fig. 1C-E). Hence, these cell lines were selected for subsequent analyses. Parallel observations of AFF4 upregulation in LUAD were substantiated in the clinical tissue specimens (Fig. 1F-G). These findings indicate that AFF4 is significantly upregulated in LUAD and represents a prognostic biomarker in patients with LUAD.

Fig. 1.

AFF4 is a prognostic marker in lung adenocarcinoma (LUAD). A Kaplan–Meier analysis (http://kmplot.com/) showing overall survival in LUAD patients with low and high expression of AFF4. B The differential expression of AFF4 in paired tumor tissues (tumor) and adjacent non-tumor lung tissues (normal) as shown in GSE10072 (including 58 LUAD tissues and 49 adjacent normal tissues) and GSE115002 (including 52 LUAD tissues and 52 adjacent normal tissues) datasets. C, D qPCR (C) and western blotting (D) analysis of the relative mRNA and protein levels of AFF4 in LUAD cells (A549, PC9, and H1299) and normal lung epithelial cells (Beas-2B). β-actin served as an internal control. E Quantification of AFF4 protein in Beas-2B, A549, PC9, and H1299 cells. β-actin served as an internal control. F Representative immunohistochemistry (IHC) images of AFF4 protein in LUAD tumor tissues and non-tumor tissues (n = 8; scale bars, 100 μm). G Quantification of AFF4-positive cells. Statistical analyses were performed using the t-tests (B, F) and two-way ANOVA tests (C, E). All data are presented as the mean ± standard error of the mean; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001

AFF4 deficiency inhibited the malignant behaviors of LUAD cells

To investigate the biological role of AFF4 in LUAD cells, AFF4 expression was silenced using short hairpin RNA (shRNA) lentiviral transfection in A549, PC9, and H1299 cells, and loss-of-function assays were performed. The knockdown efficiency was validated based on the mRNA and protein levels (Fig. S1A-C). The CCK8 assay revealed that AFF4 deficiency impaired LUAD cell proliferation (Fig. 2A). Moreover, colony formation assays demonstrated a significant reduction in the survival and proliferative capacity of single cancer cells following AFF4 silencing (Fig. 2B). The proliferation potential of these cells was further assessed through Ki67 staining, revealing a decreased proportion of positively stained cells in the AFF4-knockdown groups compared with the control group (Fig. 2C).

Fig. 2.

AFF4 knockdown inhibits cell proliferation, migration, and invasion. A, B Cell counting kit-8 (CCK-8) (A) and clone formation assays (B) were used to determine cell proliferation of LUAD cells infected with control virus (NnoT sh) and two different AFF4 viruses (AFF4 sh1 and AFF4 sh2); (right) statistical analysis of violet stained cells. C Flow cytometry analysis showing the relative proportion of Ki67-positive cells in control and AFF4-deficient LUAD cells. D Representative scratch images of control and AFF4-deficient cells in 0 h compared with 36 h (PC9, H1299) or 48 h (A549); (right) calculations of mean distances. E Transwell assay detection of the cell migration and invasion abilities in NnoT sh and AFF4 sh LUAD cells; (right) record of migrated and invaded cell counts. Statistical analyses were performed using two-way ANOVA tests (A, B, D, E). All data are presented as the mean ± standard error of the mean; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001

Subsequently, the effects of AFF4 knockdown on the migratory and invasive capabilities of LUAD cells were investigated. The wound-healing assay revealed that AFF4 deficiency prolonged the scratch-healing time of the LUAD cells (Fig. 2D). Additionally, transwell assays highlighted a significant reduction in the ability of cells to migrate from the upper to lower chamber following AFF4 knockdown (Fig. 2E). These findings indicate that AFF4 is oncogenic in LUAD cells by enhancing cell proliferation and migration.

Knockdown of AFF4 resulted in cell cycle arrest at the G2/M phase in LUAD cells

To further elucidate the role of AFF4 in cell proliferation, we examined the impact of AFF4 on cell cycle. Flow cytometry results indicated that AFF4 disruption decreased the proportion of cells in the S phase and increased the number of cells in the G2/M phase, suggesting that AFF4 knockdown induced cell cycle arrest in the G2/M phase (Fig. S2A-B). Furthermore, the expression levels of several cell cycle-associated genes, including CDC20, CDCA8, CDK1, and CCNB1, were significantly decreased in cells with impaired AFF4 function (Fig. S2C). These results highlight the significance of AFF4 in promoting LUAD cell proliferation by modulating cell cycle progression.

AFF4 inhibited apoptosis in LUAD cells

Resistance to apoptosis is a significant malignant characteristic of tumor cells [8]. Our results showed that the knockdown of AFF4 accelerated apoptosis in LUAD cells (Fig. S3A). In line with this observation, the expression of the pro-apoptotic protein Caspase-3 was upregulated, while that of the anti-apoptotic protein Bcl2 was downregulated following AFF4 deficiency (Fig. S3B-C). To investigate whether the effects of AFF4 on cell proliferation, migration, and invasion are due to its role in inhibiting apoptosis, we treated AFF4 knockdown LUAD cells with the apoptosis inhibitor Z-VAD-FMK. The results showed that the administration of this apoptosis inhibitor was unable to reverse the suppression of cell proliferation, migration, and invasion cause by AFF4 knockdown (Fig. S3D-E). This finding indicates that cell apoptosis is not the primary phenotype resulting from the loss of AFF4 function.

Ectopic AFF4 expression promoted LUAD cell proliferation and metastasis

To further verify the oncogenic role of AFF4 in LUAD tumor progression, we engineered A549, PC9 and H1299 cells that overexpressed AFF4 (Fig. S1D-F) and conducted gain-of-function assays. Exogenous AFF4 overexpression significantly enhanced the proliferation, migration, and invasion capabilities of LUAD cells (Fig. S4A–E). Therefore, our findings support the notion that AFF4 contributes to the aggressive behaviors of LUAD cells. Considering that using cancer cell lines with low endogenous expression of the target gene for overexpression studies can better determine whether the overexpression of the target gene has oncogenic properties. To identify LUAD cell lines with low expression of AFF4, we further evaluated the expression levels of AFF4 in normal lung cells (Beas-2B) as well as in another two LUAD cell lines (H358 and H1975) using western blotting. The results revealed that AFF4 expression was significantly higher in H358 cells compared to Beas-2B. In contrast, H1975 exhibited lower AFF4 expression, which was even lower than that observed in Beas-2B cells (Fig. S5A–B). Consequently, we selected H1975 cells for AFF4 overexpression experiments (Fig. S5C–E). Phenotypic assays demonstrated that the overexpression of AFF4 in H1975 cells significantly enhanced cell proliferation, migration, and invasion capabilities, further supporting the oncogenic role of AFF4 (Fig. S5F–G).

To confirm the specificity of AFF4’s function, we overexpressed AFF4 in AFF4-knockdown A549 cells. CCK8 and transwell assays indicated that the overexpression of AFF4 could restore the reduced cell proliferation and migration abilities caused by AFF4 knockdown (Fig. S6A-B). Additionally, we knocked down AFF4 in H1299 cells that had previously undergone AFF4 overexpression. The phenotypic experimental results demonstrated that the increased cell proliferation and migration abilities induced by AFF4 overexpression reverted to control levels after AFF4 knockdown (Fig. S6C-D). Overall, these findings confirm the specificity of the AFF4 knockdown and underscore the crucial roles of AFF4 in regulating cell proliferation and migration in lung adenocarcinoma cells.

AFF4 improved LUAD cell resistance to cisplatin

To investigate the role of AFF4 in LUAD cell cisplatin resistance, the correlation between AFF4 expression and the half-maximal inhibitory concentration (IC₅₀) of cisplatin was analyzed using The Cancer Genome Atlas (TCGA) database. The result highlighted a positive correlation between AFF4 expression and cisplatin IC₅₀ (Fig. 3A). Furthermore, AFF4 was significantly upregulated in cisplatin-resistant A549 cells (A549 DDP) as observed in the GEO dataset (Fig. 3B), suggesting a potential relationship between AFF4 and cisplatin resistance in LUAD.

Fig. 3.

AFF4 improves LUAD cell resistance to cisplatin. A Correlation between AFF4 expression and half-maximal inhibitory concentration (IC₅₀) of cisplatin based on the TCGA database (n = 516) (https://cancergenome.nih.gov). B Relative mRNA expression of AFF4 in wild-type (WT) A549 cells and cisplatin-resistant A549 cells (A549 DDP) from GEO dataset GSE157692 (n = 4). C Determination of IC₅₀ value of cisplatin in control and AFF4 overexpressed A549 and PC9 cells. D Apoptosis in A549 and PC9 cells with or without AFF4 overexpression after treatment with 10 µg/mL cisplatin for 48 h. Right panel: calculation of apoptosis cells. E Cisplatin IC₅₀ values in WT and A549 DDP cells. F, G AFF4 expression was detected in the WT and A549 DDP cells at mRNA (F) and protein levels (G). H Quantitative analysis of AFF4 protein. I Cisplatin IC₅₀ values in A549 DDP cells with or without AFF4 impairment. J Apoptosis in control and AFF4-deficient A549 DDP cells following 10 µg/mL of cisplatin treatment. Right panel: calculation of apoptosis cells. K-L Cell proliferation in AFF4-deficient A549 DDP cells as assessed using the CCK-8 (K) and colony formation assays (L); (right) violet-stained cell count. M Transwell analysis of AFF4-deficient A549 DDP cell migration and invasion. Statistical analyses were performed using t-tests (B, C, E, F, H) and two-way ANOVA tests (D, I-M). All data are presented as the mean ± standard error of the mean; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001

Next, we treated AFF4-overexpressing cells with cisplatin and found that AFF4 overexpression increased the resistance of LUAD cells to cisplatin (Fig. 3C). Exogenous overexpression of AFF4 also diminished apoptosis in LUAD cells following cisplatin treatment (Fig. 3D). To further verify the impact of AFF4 on cisplatin sensitivity, we selected A549 DDP cells, which exhibited a nearly five-fold increase in cisplatin tolerance compared with the WT A549 cells (Fig. 3E). As expected, AFF4 was highly expressed in A549 DDP cells (Fig. 3F-H). Silencing AFF4 endowed the A549 DDP cells with cisplatin sensitivity (Fig. 3I-J). Further investigations of the effects of AFF4 on A549 DDP cells showed that impaired AFF4 expression inhibited cell proliferation and migration (Fig. 3K–M). Collectively, these findings indicate that AFF4 is upregulated in cisplatin-resistant cells and contributes to cisplatin resistance in LUAD cells.

AFF4 enhanced glycolysis to promote cell proliferation, migration, and cisplatin resistance

To explore the underlying mechanism by which AFF4 promotes LUAD progression and cisplatin resistance, transcriptome sequencing was performed using AFF4-knockdown and control A549 cells. A total of 797 differentially expressed genes (DEGs) were screened based on a fold change (FC) ≥ 2 and FDR < 0.05, with 263 upregulated and 534 downregulated (Fig. S7A). GO functional analysis revealed that these DEGs were primarily involved in cellular protein metabolic processes and the cell cycle (Fig. S7B). Furthermore, the KEGG pathway analysis linked the DEGs to apoptosis and the mTOR signaling pathway (Fig. S7C). Subsequent GSEA revealed the participation of AFF4 in the glycolytic pathway (Fig. 4A and S7D). Consistent with these findings, the expression levels of vital glycolytic proteins, including HK2, PDK1, GLUT3, PGK1, and LDHA were notably decreased in AFF4-deficient cells (Fig. S8A-B). Meanwhile, AFF4 overexpression significantly upregulated these glycolysis related proteins (Fig. S8C-D). Additionally, metabolite determination assays demonstrated that AFF4 silencing significantly reduced glucose uptake and lactate production in LUAD and cisplatin-resistant cells (Fig. 4B, C). This was further supported by 2-NBDG staining, a fluorescent glucose analog, which revealed a decrease in glucose consumption in AFF4-deficient cells (Fig. 4D-E). The promotional effects of AFF4 on glycolysis were also confirmed in AFF4-overexpressing cells (Fig. S9A–D). Additionally, the expression level of AFF4 presented a positive correlation with glycolytic proteins PGK1 and LDHA in LUAD tissues (Fig. S9E-F).

Fig. 4.

AFF4 enhances glycolysis to promote cell proliferation, migration, and cisplatin resistance. A Gene set enrichment analysis (GSEA) of transcriptome sequencing data. B, C Glucose uptake (B) and lactate production (C) in A549, PC9, H1299 and A549 DDP cells with or without AFF4 silencing. D Representative images of reduced glucose consumption in A549, PC9, H1299 and A549 DDP cells after AFF4 silencing; 2-NBDG was added to cells at a final concentration of 100 µM for 30 min (scale bars, 50 μm). E Relative fluorescence level of 2-NBDG. F, G Glucose uptake (F) and lactate production (G) in control and AFF4-overexpressing LUAD cells (A549 and PC9) treated with vehicle or 2-deoxy-D-glucose (2-DG); cells were treated with 10 mM 2-DG for 48 h. H, I CCK-8 (H) and transwell assay (I) analyses of the proliferation, migration, and invasion of control and AFF4-overexpressing LUAD cells (A549 and PC9) with or without 2-DG treatment; (right) calculated migrated and invaded cell counts. Statistical analyses were performed using two-way ANOVA tests (B, C, E, F–I). All data are presented as the mean ± standard error of the mean; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001

To assess the impact of glycolytic inhibition on the oncogenic capabilities, AFF4-overexpressing LUAD cells were treated with the glycolysis inhibitor, 2-deoxy-D-glucose (2-DG). The 2-DG treatment substantially decreased glucose uptake and lactate production (Fig. 4F, G and S9G, H). The addition of 2-DG effectively suppressed the AFF4 overexpression-induced proliferation and migration enhancement in LUAD cells (Fig. 4H, I), and increased the cisplatin sensitivity of AFF4 overexpressed LUAD cells (Fig. S9I, J). Overall, these results suggest that AFF4 promotes tumor progression and cisplatin resistance by modulating glycolytic metabolism.

AFF4 promoted aerobic glycolysis by activating the PI3K/AKT/mTOR signaling pathway

GSEA further revealed that AFF4 was associated with PI3K/AKT/mTOR and mTOR1 pathways (Fig. 5A and S7D). Consistently, AFF4 knockdown reduced the levels of phosphorylated PI3K, AKT, and mTOR compared with controls (Fig. 5B-C). Previous studies have highlighted the essential role of mTORC1 in regulating hypoxia inducible factor 1 subunit alpha (HIF1α) translation and glycolytic activity [27, 28]. Specifically, the AKT/mTOR/HIF1α signaling pathway has been recognized for its significant impact on promoting glycolysis and lactate production, thereby contributing to the “metabolic reprogramming” of cancer cells [29, 30]. Therefore, we examined the protein level of HIF1α in AFF4 silenced LUAD cells. The Western blot results indicated that HIF1α expression was downregulated in AFF4-deficient cells (Fig. 5B-C), which indicates that the dysregulation of mTOR1 caused by AFF4 modulates the glycolytic pathway through the regulation of HIF1α.

Fig. 5.

AFF4 promotes aerobic glycolysis by activating the PI3K/AKT/mTOR signaling pathway. A The PI3K/AKT/mTOR and mTOR1 signaling pathways were enriched in the GSEA of the AFF4 transcriptome sequencing data. B Western blotting analysis of PI3K, p-PI3K AKT, p-AKT, mTOR, p-mTOR and HIF1α abundance in control and AFF4-deficient LUAD cells (A549 and PC9). C Quantification of PI3K, p-PI3K AKT, p-AKT, mTOR, p-mTOR and HIF1α proteins. D, E Glycolysis metabolites glucose (D) and lactate (E) in control and AFF4-overexpressing A549 and PC9 cells treated with vehicle or artemisinin (ART; p-AKT inhibitor); A549 and PC9 were treated with 50µM or 100 µM artemisinin, respectively, for 48 h. F, G Effects of artemisinin on the proliferation (CCK-8 assay, F), migration, and invasion (transwell assay, G) of AFF4-overexpressing A549 and PC9 cells treated with vehicle or artemisinin; (right) migrated and invaded cell counts. Statistical comparisons were performed using two-way ANOVA tests (C–G). All data are presented as the mean ± standard error of the mean; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001

Based on these findings, we hypothesized that AFF4 modulated glycolytic reprogramming by activating the PI3K/AKT/mTOR pathway. To test this hypothesis, AFF4-overexpressing LUAD cells were treated with the AKT inhibitor artemisinin, which inhibits AKT phosphorylation, to assess its effects on cell proliferation, metastasis, and glycolytic metabolism. Artemisinin treatment reversed the AFF4 overexpression-induced elevated glucose uptake and lactate production in LUAD cells (Fig. 5D, E). Similarly, AFF4-driven cell proliferation and metastasis enhancements were significantly mitigated by the inhibition of AKT signaling (Fig. 5F, G). Additionally, we treated AFF4-overexpressing LUAD cells with the mTORC1 inhibitor rapamycin. Following this treatment, we conducted metabolite determination assays and cellular function experiments. Our findings revealed that rapamycin inhibited the increase in glucose uptake and lactate production caused by AFF4 overexpression (Fig. S10A-B), while also weakening the enhanced cell proliferation and migration capabilities induced by AFF4 overexpression (Fig. S10C-D). This suggests that the modulation of glycolysis and oncogenic progression by AFF4 is mediated through the PI3K/AKT/mTOR signaling pathway.

AFF4 activated the PI3K/AKT/mTOR signaling pathway by transcriptionally inhibiting PTEN

To better understand how AFF4 regulates the PI3K/AKT/mTOR pathway, we probed our transcriptome sequencing data. Among the top 30 DEGs involved in the PI3K/AKT/mTOR pathway, PTEN was markedly upregulated in AFF4-deficient cells (Figs. 6A and S11A). Meanwhile, the mRNA and protein levels of PTEN were lower in LUAD cells than in normal cells (Fig. S11B-D). Moreover, the analysis of mRNA expression profiles from the GSE115002 and GSE10072 datasets revealed a significant downregulation of PTEN in LUAD samples (Fig. 6B). Subsequently, we investigated PTEN expression in LUAD cells following AFF4 knockdown or overexpression. Consistent with the initial findings, PTEN was upregulated in AFF4-deficient cells and downregulated in AFF4-overexpressing cells (Fig. 6C-E and S11E-G).

Fig. 6.

AFF4 activates the PI3K/AKT/mTOR signaling pathway via transcriptional inhibition of PTEN. A Top 20 upregulated genes related to the PI3K/AKT/mTOR signaling pathway from GSEA. B PTEN expression in LUAD and normal tissues in the GSE115002 and GSE10072 datasets. C, D The mRNA (C) and protein levels (D) of PTEN in A549 and PC9 cells expressing NonT sh and AFF4 sh. E Quantification of AFF4 and PTEN proteins. F Gene tracks of CUT&Tag sequencing data in A549 and PC9 cells showing enrichment of AFF4 in the PTEN promoter site. G CUT&RUN-qPCR analysis of AFF4 binding to the PTEN promoter region, as shown by relatively higher enrichment of the anti-AFF4 than anti-IgG antibodies. H Schematic images of three predicted AFF4-binding motifs and mutant sites (Mut1, Mut2, and Mut3) at the PTEN promoter region. I Luciferase reporter plasmids with WT and three mutant PTEN promoters transfected into 293T cells expressing control or AFF4 to assess PTEN promoter activity. J, K Level of glucose uptake (J) and lactate production (K) in AFF4-deficient A549 and PC9 cells following PTEN knockdown. L CCK-8 assays were used to evaluate the effects of PTEN silencing on cell viability regulated by AFF4 overexpression. M Migration and invasion (transwell assay) of AFF4-deficient LUAD cells after PTEN impairment; (right) statistical analysis of migrated and invaded cells. Statistical analysis were performed using t-tests (B, G) and two-way ANOVA tests (C, E, I–M). All data are presented as the mean ± standard error of the mean; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001

Next, we performed the CUT&Tag sequencing assay to determine whether AFF4 regulated PTEN expression by direct binding and observed that AFF4 had a strong binding at PTEN promoter region (Fig. 6F). Moreover, the CUT&RUN-qPCR results also confirmed the enrichment of AFF4 at PTEN promoter (Fig. 6G). Exploration using a dual-luciferase reporter assay with three PTEN promoter mutants demonstrated that AFF4 overexpression significantly reduced the transcriptional activities of the WT, Mut1, and Mut2, but not Mut3 variants. Hence, the core sequence corresponding to Mut3 was crucial for AFF4-mediated transcriptional repression of PTEN (Fig. 6H, I). Given that both AFF4 and AFF1 belong to the ALF family and that AFF1 also plays a role in the malignant progression of LUAD cells [23, 31], it is essential to explore their potential relationships in LUAD. Our qPCR results indicated that AFF4 does not influence the expression of AFF1 (Fig. S11H). Furthermore, the knockdown of AFF1 did not impact the expression of PTEN (Fig. S11I). Consequently, our data suggest that the transcriptional regulation of PTEN by AFF4 occurs independently of AFF1.

To verify whether AFF4 facilitates LUAD glycolysis and development via PTEN, we transfected AFF4-knockdown A549 and PC9 cells with a shPTEN plasmid and observed re-activation of the PI3K/AKT/mTOR signaling pathway (Fig. S11J, K). Moreover, AFF4 deficiency-related glycolysis suppression—as evidenced by glucose uptake and lactate production—and reduced proliferation, migration, and invasion, were reversed following PTEN knockdown (Fig. 6J–M). Taken together, these results indicate that AFF4 promotes glycolysis and oncogenic progression of LUAD cells by activating the PI3K/AKT/mTOR signaling pathway through the direct transcriptional suppression of PTEN.

YY1 activated AFF4 expression and glycolysis

While the role of AFF4 in transcriptionally regulating gene expression is well documented, the mechanisms underlying the modulation of AFF4 expression within LUAD remain elusive. Accordingly, we performed an integrative analysis using the JASPAR, GTRD, TFDB, and PROMO databases to identify potential transcription factors that regulate AFF4 expression. YY1 was identified as a potential regulator (Fig. 7A). An examination of the mRNA expression profiles from datasets GSE115002 and GSE10072 further indicated YY1 upregulation in LUAD tumor samples compared with their normal counterparts (Fig. 7B). To elucidate the relationship between YY1 and AFF4 expression, we conducted IHC assays on LUAD tissues from the same patient. The results indicated that AFF4 is highly expressed in areas with high YY1 expression, and conversely, AFF4 is also lowly expressed in regions with low YY1 expression. Furthermore, we observed a significant positive correlation between the expression levels of YY1 and AFF4. These findings suggest that YY1 plays a regulatory role in the transcription and expression of AFF4 in LUAD (Fig. S12A, B).

Fig. 7.

YY1 binds to the AFF4 promoter to activate its expression. A Venn diagram of JASPAR, GTRD, TFDB and PROMO, identifying YY1 as a potential transcription factor of AFF4. B YY1 expression in LUAD tumor and paired non-tumor tissues in the GSE115002 and GSE10072 datasets. C, D Changes in AFF4 expression at the mRNA (C) and protein levels (D) after YY1 knockdown. E Quantification of YY1 and AFF4 proteins. F ChIP-seq tracks showing the binding of YY1 to the AFF4 promoter region. G Enrichment of YY1 in the AFF4 promoter determined by CUT&RUN-qPCR. H Three potential YY1 binding sites and mutants in the AFF4 promoter. I Luciferase reporter assay showing the activity of the AFF4 promoter in 293T cells with control or YY1 overexpression. J Western blot analysis of YY1 deficiency and AFF4 overexpression efficacy in A549 and PC9 cells. K Quantification of YY1 and AFF4 proteins. L, M The glycolytic metabolite glucose (L) and lactate detection (M) in YY1-deficient A549 and PC9 cells with or without AFF4 overexpression. N, O CCK-8 (N) and transwell (O) assay analysis of the influence of AFF4 overexpression on cell proliferation, migration, and invasion in YY1-deficient A549 and PC9 cells; (right) migrated and invaded cell counts. Statistical analyses were performed with t-tests (B, C, E, G) and two-way ANOVA tests (I, K-O). All data are presented as the mean ± standard error of the mean; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001

Subsequently, we assessed the effect of YY1 on AFF4 expression and observed decreased AFF4 levels in YY1-depleted cells (Fig. 7C-E). To gain further insights into the role of YY1 in AFF4 transcriptional regulation, we obtained YY1 ChIP-seq data of A549 and PC9 cells from the ENCODE database. Strong YY1 binding was observed in the promoter region of AFF4 (Fig. 7F). Furthermore, CUT&RUN-qPCR results showed the enrichment of YY1 at the AFF4 promoter site (Fig. 7G). Experiments with AFF4 luciferase reporter constructs harboring WT and various mutant promoters in YY1-overexpressing 293T cells revealed that YY1 overexpression enhanced the luciferase activity of the AFF4 promoter with WT, Mut1, and Mut3 constructs, but not with the Mut2 construct (Fig. 7H, I). Hence, we concluded that YY1 regulates AFF4 transcription by specifically targeting the AFF4 promoter, particularly at sites coinciding with the Mut2 mutation. Consistent with these findings, we noted that YY1-deficient cells exhibited low glucose uptake and lactate production, which were nearly restored by exogenous AFF4 expression (Fig. 7J–M). Similarly, the reduced proliferation and metastatic capabilities of cells with YY1 depletion were reversed by AFF4 overexpression (Fig. 7N, O). These findings provide novel insights into the transcriptional regulation of AFF4 in LUAD, highlighting YY1 as a pivotal upstream activator.

Role of AFF4 in vivo

To assess the effect of AFF4 on tumor progression and metastasis in a live model, lung cancer xenografts were established in nude mice using AFF4-deficient or control A549 cells. AFF4 suppression caused a marked reduction in tumor growth, size, and weight (Fig. 8A–C). The modulatory effect of AFF4 on PTEN expression was substantiated in these animal models, as revealed by western blotting and IHC staining (Fig. 8D-G). Moreover, IHC analysis revealed a notable decrease in Ki67 and vimentin-positive cells, accompanied by an increase in E-cadherin-positive cells in the AFF4-knockdown group. The decreased levels of PGK1 and LDHA proteins also suggest a strong association between AFF4 and glycolysis in vivo (Fig. 8F, G). In the metastasis murine model, a decreased number of lymph nodes and reduced clustering of aberrant cells was noted in the groups lacking AFF4 (Fig. 8H, I). Moreover, significant alterations in E-cadherin and vimentin levels were observed in the groups with impaired AFF4 function (Fig. 8J, K).

Fig. 8.

Role of AFF4 in vivo. A Images of xenografts from mice subcutaneously injected with control or AFF4-deficient A549 cells (n = 4). B, C Tumor volume (B) and tumor weights (C) of the control and AFF4-deficient xenograft mice. D Western blot analysis of AFF4 and PTEN abundance in xenograft mice. E Quantification of AFF4 and PTEN proteins. F IHC analysis of AFF4, PTEN, Ki67, E-cadherin, vimentin, LDHA, and PGK1 expression in AFF4-deficient xenograft mice (scale bars, 100 μm). G Statistical analysis of positive-stained cells. H Representative images of lung metastatic nodes in xenograft mice administered control or AFF4-deficient A549 cells (n = 4). I Representative hematoxylin/eosin staining of lung metastatic nodes in xenograft mice injected with control or AFF4-deficient A549 cells (scale bars, 100 μm). J Representative IHC images of E-cadherin- and vimentin-positive cells in control and AFF4-deficient xenograft mice (scale bars, 100 μm). K Quantification of positive-stained cells. L Schematic model of the potential mechanism by which AFF4 regulates glycolysis, tumor progression, and chemoresistance. Created with BioRender (https://www.biorender.com/). Statistical analyses were performed using two-way ANOVA tests (B, C, E, G, K). All data are presented as the mean ± standard error of the mean; *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001

Overall, our in vitro and in vivo studies revealed the oncogenic role of AFF4, supporting its influence on tumor metabolism and progression (Fig. 8L).

Discussion

AFF4 has recently been recognized as a promising therapeutic target across various cancer types due to its ability to modulate cell proliferation, migration, and stemness in melanoma, ovarian, colorectal, and bladder cancer [24, 26, 32, 33]. Aerobic glycolysis is a prevalent form of metabolic reprogramming that promotes tumor growth by providing enormous energy and biosynthetic precursors [29, 34]. Recent studies have highlighted the role of AFF4 in nucleotide metabolism in pancreatic cancer, underscoring its potential to alter cellular metabolism [35]. In this study, we concluded that AFF4 stimulates aerobic glycolysis, enhancing cell proliferation, migration, and invasion in LUAD. Our findings highlight the regulatory capacity of AFF4 in LUAD cell metabolism, providing an innovative and targeted approach for LUAD treatment by modulating glycolytic pathways.

Cisplatin is a first-generation platinum-based chemotherapeutic agent and is used at various stages of NSCLC treatment [36]. The molecular mechanisms underlying cisplatin resistance in lung cancer are intricate and involve several factors, including drug metabolism, DNA repair processes, evasion of apoptosis, epigenetic regulation, and the influence of the tumor microenvironment [37]. The increased aerobic glycolysis promotes cell survival through metabolic reprogramming, which leads to cisplatin resistance in lung cancer cells [38–40]. Here, we observed that AFF4 expression was elevated in cisplatin-resistant LUAD cells, whereas disrupting AFF4 expression reduced glycolysis, proliferation, and migration in these cells. Moreover, glycolysis inhibition prevented AFF4-induced cisplatin resistance in LUAD cells, suggesting that AFF4 facilitates cisplatin resistance by enhancing glycolysis. Previous studies revealed that the lactate production caused by glycolysis could promote the repair of DNA damage caused by cisplatin [41]. Moreover, the glycolytic enzyme HK2, which is induced by AFF4 as demonstrated by our data, can interact with chromatin-associated proteins such as 53BP1 and Rad51, thereby facilitating DNA damage repair [42, 43]. On the other hand, the upregulation of glycolysis results in the accumulation of lactate, which inhibits the proliferation and function of CD8 + T cells and natural killer (NK) cells [44–46]. Furthermore, this lactate accumulation contributes to the immunosuppressive environment within the tumor by promoting the polarization of tumor-associated macrophages (TAMs) towards the M2 phenotype and facilitating the differentiation and expansion of myeloid-derived suppressor cells (MDSCs) [47–50]. Collectively, the elevation of glycolytic levels induced by AFF4 enhances the resistance of lung cancer cells to cisplatin by improving DNA damage repair and altering the tumor microenvironment.

The PI3K/AKT/mTOR signaling pathway is a classic intracellular pathway that is abnormally regulated in human cancers and contributes significantly to aerobic glycolysis. Hence, the regulators of this signaling pathway represent underlying therapeutic targets to suppress tumor growth and chemotherapy resistance [51, 52]. For example, PTEN primarily acts as a negative regulator of the PI3K/AKT/mTOR signaling pathway by dephosphorylating phosphorylated PI3K and inhibiting glycolysis [53]. In fact, diminished or absent PTEN expression commonly occurs in NSCLC and is associated with malignant cell behavior [54]. Our findings support the role of AFF4 in promoting glycolysis and tumor progression by activating the PI3K/AKT/mTOR signaling pathway through PTEN inhibition. Numerous microRNAs have been identified as enhancers of the PI3K/AKT/mTOR pathway, propelling the malignant attributes of NSCLC by directly targeting PTEN for suppression [55–57]. However, the transcriptional regulatory mechanisms governing PTEN expression remain largely unexplored. Our study revealed that AFF4 inhibits the promoter activity of PTEN by binding to specific regions of its promoter, thereby reducing the transcriptional levels of PTEN. Decreased PTEN expression facilitated PI3K/AKT/mTOR pathway activation and glycolysis in AFF4-deficient LUAD cells, which indicated that AFF4 acts as an upstream transcriptional repressor of PTEN in promoting tumor progression. Previous studies have indicated that the promoter activity of PTEN is influenced by epigenetic modifications. Specifically, an increase in the levels of H3K9Me2 and H3K27Me3 in the PTEN promoter region, along with a decrease in H3K9Ac enrichment, can lead to the repression of PTEN transcription. Additionally, DNA methylation within the PTEN promoter region also plays a role in regulating its activity [58–61]. Given that AFF4 acts as a transcriptional suppressor, it is plausible that it may regulate these histone modifications and DNA methylation in the PTEN promoter region by recruiting epigenetic regulators, thereby further suppressing PTEN promoter activity.

YY1 is a transcription factor containing a C2H2-type zinc finger that is crucial in regulating various cellular processes [62–64]. Recent discoveries have revealed the significant influence of YY1 on glycolysis, notably through its regulatory impact on LDHA and GLUT3 expressions [65, 66]. Despite these findings, the specific mechanism by which YY1 influences glycolysis in lung cancer remains only partially elucidated. Herein, we found that YY1 deficiency adversely affected AFF4 expression. Meanwhile, the reintroduction of AFF4 into YY1-deficient LUAD cells restored glycolysis and cell growth, suggesting that AFF4 is a critical mediator of YY1 in the regulation of cellular metabolism in lung cancer. This underscores the potential of targeting AFF4 to influence YY1-regulated metabolic pathways in lung cancer therapy.

Collectively, AFF4 promotes the malignant characteristics and cisplatin resistance of lung adenocarcinoma cells by enhancing glycolysis, positioning it as an excellent direct therapeutic target for both primary and cisplatin-resistant LUAD patients. Additionally, small molecule inhibitors that target AFF4 can be used in conjunction with cisplatin to increase its capacity for oxidative damage. Furthermore, these AFF4 inhibitors may be combined with immunotherapeutic agents, such as PD-1/PD-L1 inhibitors, to enhance the effectiveness of immunotherapy by reversing the immunosuppressive environment mediated by lactate.

Conclusions

Our study reveals the carcinogenic role of AFF4 in LUAD and elucidates the mechanism by which AFF4 modulates cellular glycolysis reprogramming. Additionally, the upstream and downstream factors, as well as the signaling pathways implicated in the regulatory axis of AFF4 in LUAD cells, have been defined. Furthermore, we establish the association between AFF4 and chemoresistance in human cancers for the first time. Collectively, our findings underscore the potential of AFF4 as a novel and promising therapeutic target for managing LUAD.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

LUAD

Lung adenocarcinoma

AFF4

AF4/FMR2 family member 4

SEC

Super elongation complex

PI3K

Phosphoinositide 3-kinase

AKT

Protein kinase B

mTOR

Mammalian target of rapamycin

HIF1α

Hypoxia inducible factor 1 subunit alpha

PTEN

Phosphatase and tensin homolog

YY1

Yin Yang 1

PD1

Programmed cell death protein 1

PDL1

Programmed death-ligand 1

ATP

Adenosine triphosphate

GLUT1

Glucose transporter 1

HK2

Hexokinase 2

PFKFB3

6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 3

PKM2

Pyruvate kinase M2

AFF1/AFF4

AF4/FMR2 family members 1/4

ELL1/2/3

Eleven-nineteen Lys-rich leukemia (ELL) family members 1/2/3

ENL

Eleven-nineteen leukemia (ENL)

AF9

ALL1-fused gene from chromosome 9 (AF9)

p-TEFb

Positive transcription elongation factor b (P-TEFb)

RNPII

RNA polymerase II

MLL

Methyltransferase mixed lineage leukemia

NTS

Neurotensin

CPS1

Carbamoyl-phosphate synthase 1

GPX2

Glutathione peroxidase 2

SOX2

SRY-box 2

DMEM

Dulbecco’s Modified Eagle Medium

FBS

Fetal bovine serum

WT

Wild type

IHC

Immunohistochemistry

qPCR

Quantitative PCR

BCA

Bicinchoninic acid

SDS-PAGE

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis

PVDF

Polyvinylidene fluoride membrane

CCK-8

Cell counting kit-8

OD

Optical density

PI

Propidium iodide

PBS

Phosphate-buffered saline

2-NBDG

2-(7-nitro-2,1,3-benzoxadiazol-4-yl)-d-glucosamine

H&E

Hematoxylin/eosin

CUT&RUN-qPCR

Cleavage under target and release using nuclease and quantitative PCR

GO

Gene Ontology

KEGG

Kyoto Encyclopedia of Genes and Genomes

GSEA

Gene Set Enrichment Analysis

shRNA

Short hairpin RNA

TCGA

The Cancer Genome Atlas

DEGs

Differentially expressed genes

Author contributions

Xufeng Yao, Qian Chai and Yuhao Ma performed most of experiments. Guomeng Li, Tiantian Jia, Xiaohang Zhang, Tao Xia, Xiaozheng Wei, Xueyi Feng, Yanke Zhang, Yaqiang Zhang, Xueqin Wang, and Danye Han participated in cell cycle experiments, Western blotting assay, CUT&Tag assay, luciferase report assay, animal experiments and data analysis. Qian Dai, Lei Zhao, Zongwei Li designed and supervised the study, prepared and modified the manuscript. All authors contributed to and approved the manuscript.

Funding

This work was supported by grants from National Natural Science Foundation of China (Grant number: 82103299), Natural Science Foundation of Anhui Provincial Universities (Grant number: 2024AH050653, 2022AH050790), Fuyang City’s “14th Five-Year Plan” Key Clinical Specialty Construction Project, the Major Project of Anhui Provincial Health Commission (Grant number: AHWJ2024Aa40028), and The Postgraduate Innovation Research and Practice Program of Anhui Medical University (Grant number: YJS20230145 & YJS20240076).

Data availability

The transcriptome sequencing data and CUT&Tag data are deposited in the NCBI’s GEO dataset under the accession numbers GSE262833 and GSE262461. The accessible link of transcriptome sequencing data is https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE262833 (code: ihepmscqzrkjxgf). The accessible link of CUT&Tag data is https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE262461 (code: yryrawyutxodzob). The online GEO datasets GSE115002, GSE10072 and GSE157692 were downloaded from GEO database (https://www.ncbi.nlm.nih.gov/geo/). The YY1 ChIP-seq data were downloaded from the ENCODE datasets ENCFF968ZQN (A549 IgG), ENCFF274QYN (A549 YY1), ENCFF249TDO (PC9 YY1) of ENCODE database (https://www.encodeproject.org/). All data generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The immunohistochemistry of clinical samples and all animal procedures were approved by Research Ethics Committee of Anhui Medical University (approval number: KY202416, 20240822).

Consent for publication

All authors have given their consent for the publication of the article.

Competing interests

The authors declare no competing interests.