AL355338 acts as an oncogenic lncRNA by interacting with protein ENO1 to regulate EGFR/AKT pathway in NSCLC

Eighty paired NSCLC tissues and paratumor tissues were collected from the surgery specimen archives of the Second Affiliated Hospital of Soochow University. Patient follow-up continued to December 2020. No recruited patients received preoperative chemotherapy or radiotherapy. This research was approved by the institutional clinical research ethics committee of the Second Affiliated Hospital of Soochow University. Patient stratification was based on the new TNM staging criteria in lung cancer (8th edition) [14].

The human normal lung epithelial cell line HBE and four human NSCLC cell lines (A549, PC9, H1650, H1299) were obtained from ATCC (Manassas, VA, UA) and cultured in Dulbecco's modified Eagle’s medium (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (Gibco) at 37 °C, in a humid 5% CO₂ incubator. Two small interfering RNA (siRNA) constructs (GenePharma, Shanghai, China) were designed to knock down AL355338 expression, and non-specific siRNA served as the negative control; all siRNAs were transfected with Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). For ectopic AL355338, ENO1, and EGFR expression, the appropriate cDNA was amplified with reverse transcription polymerase chain reaction (RT-PCR) and subcloned into a pcDNA-4/TO vector. Neofect transfection reagent (Neofect biotech, Beijing, China) was used for plasmid transfection. The sh-AL355338, sh-ENO1 and sh-NC lentiviruses were purchased from GenePharma. The siRNA and sh-RNA sequences are listed in Additional file 1: Table S1.

Total RNA was extracted from cells or tissues with TRIzol reagent (Omega, Norcross, GA, USA) according to the manufacturer’s instructions. RNA quality and quantity were assessed with the NanoDrop2000 platform (Thermo Fisher Scientific, Waltham, MA, USA). cDNA Synthesis kits (TaKaRa, Otsu, Japan) were used for mRNA reverse transcription. Real-time PCR analysis was performed with SYBR Green PCR Kits (TaKaRa) on the 7500 Sequence Detection System (Applied Biosystems, Foster City, CA, USA). The amplified transcript levels of genes were normalized to β-actin with the optimized comparative 2^−ΔΔCt value method. The primer sequences are in Additional file 1: Table S1.

The nuclear and cytoplasmic fractions were generated with a PARIS™ Kit (Ambion, Austin, TX, USA) according to the manufacturer protocol. Nuclear and cytoplasmic RNA levels were normalized to U6 and β-actin, respectively using real-time PCR.

FISH analyses were performed with RNA FISH kits (RiboBio, Guangzhou, China) using a previously described method [15]. The AL35538 probe sequence was 5′-DIG-ACTTCGAGACCAGCATGGCCAACATGGTGAAGCCC-DIG-3′. Nuclei were stained with DAPI, and high-resolution images were taken with a confocal microscope (ZEISS, Oberkochen, Germany). Percentages of positive stained cells were quantified as: 1 = 0–25%, 2 = 26–50%, 3 = 51–75%, and 4 = 76–100%. Staining intensity scores were: 0 = none, 1 = weak, 2 = moderate, and 3 = strong. The final scores were calculated by multiplying the percentage and intensity scores of positive cells. Samples were divided into low (score 0–6) and high (score 7–12) expression groups.

Transfected cells were harvested and fixed in pre-cooled 70% ethanol for 1 h on ice. Then cells were washed with cold phosphate-buffered saline (PBS) three times and incubated with RNase for 30 min at 37 °C. After staining the cells with propidium Iodide for 15 min at room temperature, they were analyzed with a flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA). The percentages of cells in the G0/G1, S, and G2/M phases were calculated using ModFit software (Verity Software, Topsham, ME, USA).

Cell proliferation rates were measured with Cell Counting Kit-8 (bimake, Shanghai, China) according to the manufacturer’s protocol. Briefly, 5 × 10³ cells per well were seeded into 96-well plates. At 0, 24, 48, and 72 h after cell transfection, 10 μL CCK-8 solution was added to each well. After incubating he plates for 2 h at 37 °C, the optical density at 450 nm was measured for each well. For colony formation assays, transfected or control A549 and PC9 cells were harvested 24 h after transfection, and 500 cells per well were seeded into 6-well plates. After 2 weeks of incubation, the colonies were fixed in 4% paraformaldehyde and stained with crystal violet solution. Cell proliferation was also assessed using ethynyldeoxyuridine (EdU) analyses (RiboBio) according to the manufacturer’s instructions as previously reported [16].

Wound healing assays were carried out as previously described [17]. Transwell assays were performed using Boyden chambers containing transwell membrane filters (Corning, Corning, NY, USA). The detailed protocol was published elsewhere [17]. Images were collected with an inverted microscope (Olympus, Tokyo, Japan) and analyzed with ImageJ software (National Institutes of Health, Bethesda, MD, USA). At least five random fields of view were analyzed in each chamber.

As a glucose analogue, ¹⁸F-FDG uptake reflects the intracellular glucose uptake level. In vitro experiments were performed as previously reported [16]. Lactate production was measured in cells that were seeded into 12-well plates in triplicate for 24 h and then refreshed with 1 mM glucose culture medium overnight. Culture supernatants were collected to measure lactate concentrations (Nanjing Jiancheng Bioengineering Institute, Nanjing, China), and lysed cell pellets were used to measure ATP levels (Nanjing Jiancheng) using the manufacturer’s instructions. ¹⁸F-FDG uptake and lactate production were normalized to number of cells in each sample; ATP levels were normalized to cell protein mass.

Western blotting was performed following a standard protocol [18]. The antibodies used are listed in Additional file 1: Table S2. Immunoreactive bands were detected using ECL western blot kits (Amersham Biosciences, Little Chalfont, UK).

IF was performed as described elsewhere [18]. Protein expression and localization were observed under a confocal microscope (ZEISS). Antibodies used for IF are listed in Additional file 1: Table S2.

Biotin-RNA pull-down assays were performed as described elsewhere [15]. Briefly, full-length sense, antisense, and serial deletion sequences of AL355338 were amplified with a T7-containing primer and reverse transcribed using MAXIscript™ T7 Transcription Kits (Thermo Fisher Scientific). Pierce™ RNA 3′ End Desthiobiotinylation Kits (Thermo Fisher Scientific) were used to label targeted RNA. The biotin-labeled AL355338 probe was incubated with total cell lysates of A549 cells using Pierce™ Magnetic RNA–Protein Pull-Down Kits (Thermo Fisher Scientific). Eluted proteins were purified by silver staining and subjected to mass spectrometry (MS) or western blotting.

Magna RNA-binding protein immunoprecipitation kits (Millipore, Billerica, MA, USA) were used to determine specific protein binding with targeted RNA according to the manufacturer’s protocol and as described previously [15]. Briefly, cells were lysed with RIP lysis buffer, and lysates were incubated overnight at 4 °C with magnetic beads conjugated with human anti-ENO1 antibody or anti-mouse immunoglobulin G (IgG). The next day, co-immunoprecipitated RNA was extracted for qRT-PCR analyses. Total RNA (input controls) and IgG controls were simultaneously assayed to confirm that the RNA specifically bound to ENO1.

Reciprocal exogenous and endogenous co-IP assays were performed as described elsewhere reported [18]. IP samples and input controls were analyzed by western blot. The co-IP antibodies are listed in Additional file 1: Table S2.

PLA (Duolink®, Sigma-Aldrich, St. Louis, MO, USA) was performed to evaluate interactions between ENO1 and EGFR as previously reported [19]. Briefly, A549 and PC9 cells were grown on glass coverslips in 24-well plates at 37 °C overnight. The next day, cells were fixed in 4% paraformaldehyde for 20 min and permeabilized with 0.1% Triton X-100 for 5 min. After blocking for 30 min, the cells were incubated with anti-ENO1 and anti-EGFR antibodies diluted in blocking buffer at 4 °C overnight. The next day, cells were incubated with species-specific PLA probes at 37 °C for 1 h, then 1× ligase for 30 min, and finally 1× polymerase diluted with amplification stock solution at 37 °C for 100 min to amplify the fluorescence signal. Finally, coverslips were mounted on slides with anti-fluorescence quenching sealing liquid containing DAPI. Images were acquired under confocal microscopy (ZEISS).

Four-week-old male BALB/c nude mice were purchased from the Animal Center of Soochow University. Animal experiments were carried out according to the protocol approved by the Animal Care and Ethics Committee of the Second Affiliated Hospital of Soochow University. A total of 20 mice were subcutaneously injected with 5 × 10⁶ A549 cells in the right flank with sh-AL355339, and sh-NC was given in the left flank to establish an NSCLC xenograft model. Tumor volume was measured every 5th day and calculated according to the equation V = length × width² × 1/2. Micro-PET/CT (Super Nova® PET/CT, Pingseng, China) was performed 3 weeks after cell injection to measure ¹⁸F-FDG uptake by xenograft tumors. All tumor-bearing mice were fasted for 8 h prior to PET/CT. ¹⁸F-FDG (0.1 mL, 7.4 MBq) was injected via tail vein. After 30 min, they were anesthetized with 2% isoflurane in oxygen and immobilized for 15-min scans. ¹⁸F-FDG uptake was measured as maximum standard uptake (SUV_max) in the regions of interest covering the tumors. After 3 weeks, mice were humanely sacrificed by euthanasia. The mice were inhaled with CO₂ for 5 min. The CO₂ was gradually filled into an airtight chamber at 30% of the chamber volume/min rate. The xenograft tumors were harvested and measured.

Stably transfected A549 cells (1 × 10⁶/0.2 ml PBS) were injected into the tail vein of nude mice. Micro-CT scans were performed 5 weeks later to observe lung metastasis. After 5 weeks, mice were humanely sacrificed by euthanasia. Following sacrifice, the lungs were surgically dissected and embedded in paraffin for hematoxylin and eosin (HE) staining. The numbers of pulmonary metastatic nodules were compared between groups.

IHC was performed and quantified as previously described [20].

All data are expressed as mean ± standard deviation. Statistical analysis was performed with SPSS (version 22.0, IBM Corp., Armonk, NY, USA) and graphs were generated with GraphPad Prism software (version 8.0, GraphPad Inc., San Diego, CA, USA). Student’s t-tests or one-way analyses of variance were used to assess significant differences between groups. Categorical data were compared with Chi-square tests. Survival analyses were performed using Kaplan–Meier analyses.

To identify glycolysis-related lncRNAs in NSCLC, we performed RNA-seq assay with ¹⁸F-FDG high-uptake NSCLC tissues and pair-matched normal tissues from our previous study [16]. We detected 222 upregulated genes (fold change > 1.5, false discovery rate < 0.005). The top 20 significantly upregulated lncRNAs (Additional file 1: Figure S1A) were submitted for Gene Ontology analysis, and the results showed that they were most enriched in metabolic pathways (Additional file 1: Figure S1B). To screen more functional lncRNAs, our next investigation used more stringent filtering criteria (> 400 bp, not fully overlapped with other coding transcripts, efficient PCR amplification). In the current study, we focused on the uncharacterized AL355338, which was the top-scoring upregulated lncRNA in the RNA-seq data, and explored its function in NSCLC development. AL355338 is a 913-bp transcript on human chromosome 13 with 2 exons (Additional file 1: Figure S1C). The Coding-Potential Assessment Tool (CPAT), PhyloCSF tool, and ORFfinder all predicted that AL355338 had no coding potential (Additional file 1: Figure S1D, E). To explore the clinical significance of AL355338, we analyzed RNA-seq data from the publicly available TCGA dataset (StarBase V3.0 platform). AL355338 expression was significantly upregulated in lung adenocarcinoma and lung squamous carcinoma tissues (Additional file 1: Figure S2A, B). Kaplan–Meier survival analysis revealed that patients with high AL355338 expression has shorter overall survival than patients with low AL355338 expression (Additional file 1: Figure S2C, D).

QRT-PCR analysis confirmed that AL355338 was increased in tumor tissues compared with the corresponding paratumor tissues in a panel of 80 specimens from NSCLC patients (Fig. 1A). Furthermore, AL355338 expression was remarkably higher in metastatic samples compared with non-metastatic samples, and elevated AL355338 expression correlated with advanced TNM stage (Fig. 1B, C). Kaplan–Meier survival analysis of overall survival showed that high AL355338 expression was associated with shorter survival (Fig. 1D; P = 0.032). We next detected AL355338 with FISH and evaluated the relationship between its expression and clinicopathological characteristics. Consistent with the qRT-PCR results, more positive AL355338 expression signals were observed in tumor tissues compared to adjacent normal tissues (Fig. 1E). Expression score analyses demonstrated that expression percentages of AL355338 were higher in stage III + IV than in stage I + II NSCLC samples (Fig. 1F). High AL355338 expression was positively associated with T stage, N stage, distant metastasis, and TNM stage (Table 1). However, it was not related to patient age, sex, or tumor location. Uni- and multivariate regression analyses revealed that AL355338 expression was an independent predictor of NSCLC aggressiveness and could predict clinical outcome (Fig. 1G, H). Collectively, these results suggested that AL355338 was upregulated in NSCLC and might be a useful prognostic biomarker.

To explore the potential function of AL355338 on cellular behavior, we first detected the expression levels of AL355338 in NSCLC cell lines. Its expression in NSCLC cell lines was significantly higher than that in the normal lung epithelial cell line HBE (Additional file 1: Figure S3A). We then examined its subcellular localization in NSCLC cells. Cellular FISH experiments and analyses of RNA isolated from the cytoplasmic and nuclear compartments of NSCLC cells showed that AL355338 was mainly present in the cytoplasm (Fig. 2A, B). Then, we downregulated AL355338 by transfecting two different siRNAs in A549 and PC9 cells. qRT-PCR was performed to verify the knockdown efficiency of AL355338; both si-AL355338#1 and si-AL355338#2 successfully downregulated AL355338 expression to 20–30% in A549 and PC9 cells (Fig. 2C).

Flow cytometry analyses of cell cycle distribution indicated that AL355338 downregulation significantly decreased the number of cells in S phase, with an increase of cells in G0/G1 phase for both A549 and PC9 cells (Fig. 2D–F). CCK-8 and cell colony formation assays demonstrated that the proliferation and colony formation ability of A549 and PC9 cells were impaired after AL355338 knockdown (Fig. 2E, G, H). Subsequent EdU assays indicated that DNA synthesis by A549 and PC9 cells was markedly reduced after silencing AL355338 (Fig. 2I). These results indicated that AL355338 was effective in facilitating NSCLC cell proliferation.

To further assess the migratory and invasive ability conferred by AL355338, we performed transwell and wound healing assays. The transwell results in Fig. 3A–C show that downregulation of AL355338 suppressed the migratory and invasive activities of A549 and PC9 cells compared to the negative control. Wound healing assays showed the same pattern (Fig. 3D–F). The initiation of epithelial-mesenchymal transition is a driving force of tumor invasion and metastasis. Analyses of the protein expression of epithelial and mesenchymal markers revealed that AL355338 silencing reduced N-cadherin and vimentin expression but enhanced E-cadherin expression in A549 and PC9 cells (Fig. 3G). Consistently, IF staining also showed that the fluorescence intensity of Vimentin was attenuated in AL355338 downregulated A549 cells (Additional file 1: Figure S3B).

Considering AL355338 expression was upregulated in association with high ¹⁸F-FDG uptake, we investigated the effect of AL355338 in glycolysis regulation. Glucose enzyme expression levels were examined, and the results showed that alteration of AL355338 expression could positively affect hexokinase II, pyruvate kinase isozymes M2, and lactate dehydrogenase A protein levels (Fig. 3H). To confirm the effect of AL355338 on glycolysis, we measured ¹⁸F-FDG uptake, lactate release, and ATP production to reflect the flux distribution of glucose. AL355338 knockdown led to a decreases in all three parameters (Fig. 3I–K). These results demonstrated that AL355338 could regulate glycolytic metabolism in NSCLC.

To determine whether AL355338 expression could affect NSCLC cell growth in vivo, A549 cells were stably transfected with sh-AL355338 or negative control sh-NC lentivirus. In vivo xenograft experiments showed that tumor volumes in the sh-AL355338 group were obviously smaller than in the sh-NC group (Fig. 4A). Notably, tumor weights were markedly reduced in the sh-AL355338 group compared with controls (Fig. 4B). In addition, xenograft glycolysis uptake ability was observed by ¹⁸F-FDG micro-PET/CT imaging. Tumor glycolysis was significantly blocked in sh-AL355338 xenografts, with lower SUV_max (mean SUV_max: 1.15) compared with sh-NC xenografts (mean SUV_max: 2.08; Fig. 4C, D). IHC revealed that the tumor tissue grown from sh-AL355338 cells displayed lower Ki-67, ENO1, and EGFR staining than those formed from sh-NC cells. FISH analyses confirmed the inhibitory effect of knocking down AL355338 expression in A549 cells (Fig. 4E).

We then investigated the function of AL355338 on tumor metastasis in vivo. Micro-CT images showed more metastatic lesions in the lungs of mice in the sh-NC group, while fewer and smaller metastatic lung nodules were seen in the AL355338 knockdown group (Fig. 4F). Summarized data on HE staining analysis demonstrated that more metastatic lung foci were formed in the sh-NC group compared to the AL355338 knockdown group (Fig. 4G). These results indicated that inhibiting AL355338 expression in NSCLC cells could efficiently suppress tumor metastasis in vivo.

To investigate the underlying mechanism of AL355338-mediated glycolysis and carcinogenesis in NSCLC, we set out to identify intracellular AL355338-binding factors in an unbiased manner. Biotinylated AL355338 and negative control were incubated with total protein extracts from A549 cells and pulled down with streptavidin (Additional file 1: Figure S4A). MS was conducted following RNA pull-down to screen endogenous AL355338 binding proteins (Additional file 1: Table S3). From the MS data files, we selected ENO1 for further binding validation as it is a key enzyme responsible for the glycolytic pathway. Subsequent western blotting of proteins retrieved from AL355338 pull-down assays confirmed that the ENO1 protein directly bound to the sense AL355338 sequence but did not bind the antisense control (Fig. 5A, upper). In addition, RIP assays with an ENO1 antibody and IgG antibody control were performed using A549 and PC9 cellular extracts. Correspondingly, we observed significantly higher AL355338 enrichment with the ENO1 antibody compared to IgG (Fig. 5A, lower). Both assays demonstrated that AL355338 directly interacted with ENO1. To better identify which region of ENO1 interacted with AL355338, we performed RIP assays using FLAG-tagged truncated forms of ENO1. The results demonstrated that the ENO1-C domain (237–434 aa) was essential for AL355338 binding (Fig. 5B). To map the ENO1-binding regions of AL355338, we constructed four AL355338 deletion fragments. Subsequent RNA pull-down assays followed by western blotting revealed that the ENO1 specific binding sequence was located between 528 and 913 nucleotides (Fig. 5C). RNA FISH of AL355338 combined with IF of ENO1 confirmed cytoplasmic co-localization of AL355338 and ENO1 in A549 cells (Fig. 5D).

A recent study demonstrated that lncRNAs interact with proteins through several modes of action, such as regulating protein expression and function, modulating protein–protein interactions, or guiding protein subcellular localization [21]. To investigate the molecular effect of AL355338 on ENO1, western blotting was performed to measure ENO1 expression after A549 and PC9 cells were subjected to different treatments with AL355338. The results showed that AL355338 positively regulated ENO1 protein expression (Fig. 5E). Similar trends were observed in IF assays (Additional file 1: Figure S4B). Interestingly, loss or gain of function of AL355338 had no influence on ENO1 mRNA expression (Additional file 1: Figure S4C, D), suggesting that AL355338 may regulate ENO1 protein stability. Next, cycloheximide (CHX) chase assays were performed to observe the effect of AL355338 on ENO1 protein degradation. Western blotting showed that ENO1 protein was more stable in AL355338-transfected cells, while it had a shorter half-life in AL355338-silenced cells (Fig. 5F, G). Moreover, the proteasome inhibitor MG132 rescued the reduction of ENO1 caused by AL355338 repression (Fig. 5H), suggesting that AL355338 elevated ENO1 levels by reducing its degradation. As reported, cytoplasmic protein degradation involves the ubiquitin–proteasome pathway [22]. Thus, whether AL355338 was involved in the ubiquitin-mediated ENO1 degradation was tested with in vitro ubiquitination assays. The results illustrated that upregulation of AL355338 decreased ENO1 ubiquitination, whereas silencing of AL355338 enhanced ENO1 ubiquitination (Fig. 5I). Collectively, our data indicated that AL355338 directly bound the ENO1 protein and enhanced its stability in NSCLC cells.

To clarify whether the RNA–protein interaction between AL355338 and ENO1 mediated the key oncogenic function of AL355338 in NSCLC, complementary rescue experiments were performed. Stable ENO1-knockdown A549 and PC9 cells were constructed and then transfected with AL355338 overexpression plasmid. Transfection efficiency was confirmed by qRT-PCR and western blot (Fig. 6A, B). As was expected, CCK-8 and cell colony formation assays showed that ectopic AL355338 expression increased the proliferation activity and colony formation ability of A549 and PC9 cells, while ENO1-knockdown partially abrogated the effects induced by AL355338 upregulation (Fig. 6C, D). EdU assays revealed that increased DNA synthesis capability of A549 and PC9 cells induced with AL355338 was reversed by ENO1 silencing (Fig. 6E). Cell migration and invasion abilities were also assessed, and results demonstrated that AL355338 exerted minimal effects on cell migration and invasion in the absence of ENO1 (Fig. 6F). Importantly, the stimulatory effect of AL355338 on epithelial and mesenchymal marker expression was recovered by repressing ENO1 (Fig. 6G). In glycolysis metabolism assays, ENO1 downregulation dramatically blocked the AL355338-induced increase in ¹⁸F-FDG uptake, as well as lactate production by A549 and PC9 cells (Fig. 6H, I). The above results suggested that AL355338 regulation of glycolytic metabolism and cell proliferation were dependent on ENO1 stability in NSCLC cells.

The EGFR/AKT signaling pathway influences tumor cell survival in many cancers. As previous reports have demonstrated that ENO1 could regulate the AMPK-mediated or FAK-mediated AKT pathway [23, 24], we sought to determine whether ENO1 could interact with EGFR and subsequently activate its downstream AKT kinase signaling pathway. To verify the relationship between ENO1 and EGFR in NSCLC cells, we first examined their TCGA database expression levels. We extracted ENO1 and EGFR data from 34 tumors and found that both were notably upregulated in diverse cancer types, especially lung adenocarcinoma and lung squamous carcinoma tissues (Additional file 1: Figure S5A, B). As shown in Additional file 1: Figure S5C, the correlation coefficient of ENO1 and EGFR reached 0.32 in the NSCLC cohort (P < 0.001, Spearman correlation analysis). Besides, ENO1 shared a similar mutation status with EGFR in NSCLC samples (P < 0.001, Additional file 1: Figure S5D).

Next, we aimed to investigate whether ENO1 directly interacted with EGFR. Reciprocal co-IP assays were performed using both exogenous labeled antibodies and endogenous antibodies to precipitate ENO1 and EGFR, respectively. The co-IP results confirmed physical binding between ENO1 and EGFR proteins in NSCLC cells (Fig. 7A, B). IF staining confirmed ENO1 and EGFR co-localization in A549 and PC9 cells (Fig. 7C). Furthermore, in situ PLA was used to visualize native protein complexes. High densities of PLA-positive protein complexes that included ENO1 and EGFR were observed in A549 and PC9 cells, which was in accordance with the co-IP results (Fig. 7D). Intriguingly, AL355338 knockdown noticeably impaired the ENO1 and EGFR interaction, while AL355338 overexpression had the opposite effect (Fig. 7E, F).

To understand the potential relationship between AL355338 and the EGFR/AKT pathway, we investigated EGFR and AKT expression. Western blot analyses showed that AL355338 suppression decreased the levels of phosphorylated EGFR and AKT, but total protein levels were not affected. Overexpression of AL355338 enhanced EGFR and AKT phosphorylation, indicating that the EGFR/AKT pathway was a downstream effector of AL355338. We then assessed the ability of ENO1 to overcome AL355338-mediated activation of the EGFR/AKT pathway in NSCLC cells. Notably, ENO1 knockdown abrogated AL355338-mediated increases of EGFR and AKT phosphorylation, whereas ENO1 overexpression reversed sh-AL355338-mediated inactivation of the EGFR/AKT pathway (Fig. 7G, H). Based on these results, we inferred that AL355338 could regulate EGFR/AKT pathway activity by targeting ENO1/EGFR interaction.

To further establish the relationship between AL355338 and ENO1 in NSCLC tissues, we first evaluated ENO1 protein expression with IHC in 80 NSCLC and pair-matched normal tissue samples (Fig. 8A). The results revealed that 76.3% (61/80) NSCLC samples showed increased ENO1 expression compared with their pair-matched normal tissues (Fig. 8B). ENO1 protein levels positively correlated with AL355338 levels (r = 0.382, P < 0.001, Fig. 8C), suggesting potential regulation of ENO1 by AL355338 in clinical NSCLC tissues. Meanwhile, TCGA datasets revealed that ENO1 expression was significantly upregulated in lung adenocarcinoma and lung squamous carcinoma tissues compared to normal tissues (Fig. 8D, E). Kaplan–Meier analysis of the 80-patient NSCLC cohort showed that high ENO1 expression was associated with poor prognosis (P = 0.009, Fig. 8F). This was validated in the TCGA cohort; survival analyses showed that patients with low ENO1 expression had longer overall survival (P = 0.025) and disease-free survival (P = 0.044) compared to those with high ENO1 expression (Fig. 8G, H).