Lenvatinib (cat. no. HY-10981) and a PI3K inhibitor (LY294002; cat. no. HY-10108) were purchased from MedChemExpress.

The Huh7 parental (Huh7-P) and HepG2 parental (HepG2-P) LC cell lines were purchased from Zhejiang Meisen Cell Technology Co., Ltd., and their authenticity was verified through short tandem repeat profiling. The cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Beijing Solarbio Science & Technology Co., Ltd.) supplemented with 10% fetal bovine serum (FBS; Beijing Solarbio Science & Technology Co., Ltd.) in a 37°C incubator with a 5% CO₂ atmosphere. As previously reported, two lenvatinib-resistant (LR) LC cell lines (Huh7-LR and HepG2-LR) were established by gradually exposing parental Huh7 and HepG2 cells to increasing concentrations of lenvatinib (25). The Huh7-LR and HepG2-LR cells were continuously cultured with 5 or 15 µM lenvatinib, respectively.

To assess the clonogenicity of the Huh7-LR and HepG2-LR after different drug treatments, 0.5×10³ cells were seeded per well in a six-well plate, and incubated at 37°C. Cells were treated with vehicle control (DMSO), lenvatinib monotherapy, LY294002 monotherapy (20 µM) or combination therapy for 72 h. The lenvatinib concentration was 1 µM for Huh7-LR cells and 10 µM for HepG2-LR cells. To optimize drug efficacy based on pharmacokinetic profiles, lenvatinib was administered at 0 h for the entire 72-h period, while LY294002 was added at 24 h post-seeding and maintained for the subsequent 48 h. Following the completion of the treatment period, cells were cultured in a standard culture medium as aforementioned for the next 11 days. After fixation with 4% paraformaldehyde for 30 min at room temperature and staining with 0.1% crystal violet for 20 min at room temperature, the colonies consisting of >50 cells were photographed and analyzed using ImageJ v1.48 software (National Institutes of Health).

For cell cycle analysis of Huh7-LR and HepG2-LR cells, the cell concentration was adjusted to 1×10⁶/ml. After washing the cells with ice-cold phosphate-buffered saline (PBS), 1 ml of DNA staining solution and 10 µl of permeabilization solution were added, and the cells were incubated in the dark for 30 min at 4°C. A cell cycle staining kit (cat. no. CCS012; MultiSciences Biotech Co., Ltd.), which included the DNA staining solution and permeabilization solution, was employed for detection. The LR cells were arrested primarily at the G₀/G₁ phase of the cell cycle. Cell cycle analysis was performed using Attune^™ NxT (Thermo Fisher Scientific, Inc.) and data were analyzed with ModFit LT v4.1.7 (Verity Software House, Inc.). For apoptosis detection, cells were seeded at a density of 1×10⁶ cells per well in a 6-well plate and incubated overnight at 37°C. Thereafter, the cells were exposed to different drug-containing media in a 37°C incubator. The treatment groups, drug concentrations and treatment durations for Huh7-LR and HepG2-LR cells were consistent with those utilized in the clonogenic assay. Subsequently, the cells were harvested and stained with Annexin V-FITC and propidium iodide. The apoptotic rate was measured by flow cytometry (FC500; Beckman Coulter, Inc.) and analyzed using CXP Analysis software (version 2.0; Beckman Coulter, Inc.).

Matrigel (cat. no. 356234; BD Biosciences) was thawed and diluted 1:8 with serum-free culture medium at 4°C using pre-chilled consumables. Subsequently, 50 µl of the diluted Matrigel was evenly applied to each Transwell insert and these were incubated at 37°C for 30 min to allow gel solidification. Huh7-LR and HepG2-LR cells (2×10⁴) were trypsinized, resuspended in 200 µl serum-free DMEM, and added to the upper chamber of the Transwell device, while 600 µl of DMEM containing 10% FBS was added to the lower chamber. After a 72-h incubation, the chambers were removed, fixed by 4% paraformaldehyde at room temperature for 20 min, and stained with 1% crystal violet for 20 min at room temperature, while the cells in the upper chamber were wiped with a cotton swab. Images were acquired using an Olympus BX43 optical light microscope (Olympus Corporation), with three random fields of view captured per chamber.

Total RNA was extracted from LC cells using the TRIzol reagent (Thermo Fisher Scientific, Inc.) and cDNA synthesis was performed using a reverse transcription kit (Proteinssci Biotech Co., Ltd.) according to the manufacturer's protocol. Subsequently, RT-qPCR was conducted using the MagicSYBR Mixture with a CFX96 Touch^™ real-time PCR detection system (Bio-Rad Laboratories Inc.) following instructions provided by the manufacturer. The thermocycling conditions were as follows: Initial denaturation at 95°C for 15 min, followed by 40 cycles of denaturation at 95°C for 10 sec, annealing at 56°C for 30 sec and extension at 72°C for 30 sec.

The primer sequences utilized in this study (synthesized by Sangon Biotech Co., Ltd.) were as follows: LDHA-forward (F), 5′-GATTCAGCCCGATTCCGTTACC-3′; LDHA-reverse (R), 5′-AGAGACACCAGCAACATTCATTCC-3′; HK2-F, 5′-TGGAACTGGTGGA-AGGAGAAGAG-3′; HK2-R, 5′-TCTGTGCGGAAGTCATCTAGGC-3′; GLUT1-F, 5′-AGG-AAGAGAGTCGGCAGATGATG-3′; GLUT1-R, 5′-TGGAGTAATAGAAGACAGCGTT-GATG-3′; HIF-1α-F, 5′-CGCAAGTCCTCAAAGCACAGTTAC-3′; HIF-1α-R, 5′-ATTC-ATCAGTGGTGGCAGTGGTAG-3′; β-actin-F, 5′-TCG-TGCGTGACATTAAGGAG-AAGC-3′; β-actin-R, 5′-GGCGTACAGGTCTTTGCGGATG-3′.

The results were normalized to β-actin expression for mRNA measurement, and the relative expression of the target gene was calculated using the 2^−ΔΔCq comparative method (26).

Total protein was extracted by lysing liver cancer cells using radioimmunoprecipitation assay lysis buffer (Beijing Solarbio Science & Technology Co., Ltd.). Equal amounts of protein samples (~25 µg/lane), quantified by bicinchoninic acid assays, were separated by SDS-PAGE on 8–12% polyacrylamide gels and subsequently transferred onto Immobilon^®-P transfer membranes (cat. no. IPVH00010; Merck KGaA). After blocking with 5% non-fat milk at room temperature for 2 h, membranes were incubated at 4°C overnight with the following primary antibodies: Anti-GLUT1 (1:1,000; cat. no. A11727; ABclonal Biotech Co., Ltd.), anti-HK-2 (1:1,000; cat. no. A0994; ABclonal Biotech Co., Ltd.), anti-LDHA (1:1,000; cat. no. A21893; ABclonal Biotech Co., Ltd.), anti-PI3K (1:1,000; cat. no. A17433; ABclonal Biotech Co., Ltd.), anti-pPI3K (1:1,000; cat. no. AP0854; ABclonal Biotech Co., Ltd.), anti-AKT (1:1,000; cat. no. A22533; ABclonal Biotech Co., Ltd.), anti-pAKT (1:1,000; cat. no. AP0637; ABclonal Biotech Co., Ltd.) and anti-HIF-1α (1:1,000; cat. no. bs0737R; BIOSS). Thereafter, the membranes were incubated with HRP-conjugated secondary antibodies (1:3,000; cat. nos. ZB2301 and ZB2305; Beijing Zhongshan Jinqiao Biotechnology Co., Ltd.) at room temperature for 90 min, followed by incubation with an enhanced chemiluminescence reagent (cat. no. C05-07004; BIOSS). Finally, the bands were imaged using a Tanon 4800 automated imaging system (Tanon Science and Technology Co., Ltd.). The grayscale intensity of protein bands was quantified using Tanon Gel Imaging System software (version 4.2; Tanon Science and Technology Co., Ltd.).

The glucose uptake ability of cells was measured by 2-NBDG (AbMole Bioscience Inc.). Briefly, cells were seeded into 6-well plates at a density of 1×10⁶ per well and subjected to different treatments. The treatment groups, drug concentrations and exposure durations were identical to those employed in the clonogenic assay. Following treatment, 2-NBDG was added to the cells at a final concentration of 10 µM, and the cells were incubated for 30 min at 37°C. The cells were subsequently washed twice with PBS buffer to eliminate any unabsorbed probes. The mean fluorescence intensity of samples was measured by flow cytometry (FC500; Beckman Coulter, Inc.), and the median was used for statistical analysis.

The lactate contents of cell supernatant samples were determined using the Lactic Acid Content Assay Kit (cat. no. L256; Dojindo Laboratories, Inc.). After collecting the sample of cell culture supernatant, the extracellular lactate concentrations were determined using a microplate reader (Fluoroskan™FL; Thermo Fisher Scientific, Inc.) at a wavelength of 450 nm, according to the instructions provided by the manufacturer.

Intracellular ATP levels were determined using an ATP assay kit (cat. no. S0027; Beyotime Institute of Biotechnology). Cells were plated at a density of 2×10⁵ per well in 6-well plates and treated according to the experimental design as described for the clonogenic assay. After lysis, the supernatants were obtained by centrifugation at 12,000 × g for 5 min at 4°C and mixed with ATP detection working solution included in the ATP assay kit. Fluorescence intensity was measured using a microplate reader (Varioskan LUX; Thermo Fisher Scientific, Inc.) and ATP concentrations were calculated based on a standard curve. The results were normalized to protein content (BCA assay).

Statistical analyses were performed using SPSS Statistics version 26.0 (IBM Corp.) and the figures were generated using GraphPad Prism 8.0 (Dotmatics). All measurement data were obtained in triplicate and are shown as the mean ± SD. A two-tailed unpaired Student's t-test was used to compare variables between two groups. One-way ANOVA was performed for multi-group comparisons using Tukey's post hoc tests. P<0.05 was considered to indicate a statistically significant difference.

The present study aimed to investigate the mechanisms underlying lenvatinib resistance in LC. Thus, the previously established LR cell models, Huh7-LR and HepG2-LR, were used, which were generated through gradually exposing Huh7 and HepG2 cells to increasing concentrations of lenvatinib. Aberrant metabolism is a major hallmark of cancer. Aerobic glycolysis, an essential part of metabolic reprogramming, is closely associated with drug resistance in various tumors (27). Recent studies have reported that prolonged exposure to lenvatinib can enhance aerobic glycolysis, potentially driving lenvatinib resistance in LC (28,29). Therefore, the present study conducted a comparative analysis of glycolytic activity between parental cells and their LR derivatives. Glucose serves as the main energy source for tumor cells, and glucose uptake is a crucial indicator of glycolytic activity. The uptake of the fluorescence-labeled glucose analogue 2-NBDG was assessed, and a significantly increased 2-NBDG uptake in LR LC cells was observed (Fig. 1A and B). Lactate, a main metabolic byproduct of glycolysis, is of notable importance in assessing the glycolytic levels of tumor cells. Lactate assays revealed significantly higher lactate production in Huh7-LR and HepG2-LR cells compared with their parental counterparts (Fig. 1C and D). To further evaluate metabolic alterations, intracellular ATP levels were compared between the two groups. The results showed that, compared with the parental cells, the intracellular ATP levels were significantly elevated in LR LC cells (Fig. 1E and F). In addition, the expression of three key glycolysis-related genes was analyzed, namely GLUT1, LDHA and HK2. As shown in Fig. 1G and H, the mRNA expression levels of GLUT1, LDHA and HK2 were significantly upregulated in LR LC cells compared with their parental cells. Consistently, western blotting analysis further confirmed a significant increase in GLUT1, LDHA and HK2 protein expression in resistant cells (Fig. 1I and J). These findings collectively indicate that LR LC cells exhibit an enhanced glycolytic phenotype, suggesting an association between lenvatinib resistance and increased glycolysis.

Increased aerobic glycolysis has been implicated in the acquisition of various malignant tumor phenotypes, including drug resistance. To further elucidate the mechanisms underlying lenvatinib resistance in LC, the regulation of glycolysis was investigated with a focus on the PI3K/AKT/HIF-1α signaling pathway, which serves as a key regulator of glycolytic activity and plays a crucial role in promoting tumor resistance (30). The expression levels of key components within the PI3K/AKT/HIF-1α signaling pathway was compared between Huh7/HepG2 parental cells and their LR LC cells. Analysis revealed significantly elevated levels of phosphorylated PI3K/PI3K (pPI3K/PI3K), phosphorylated AKT/AKT (pAKT/AKT) and HIF-1α in Huh7-LR/HepG2-LR cells compared with their respective parental cell lines (Fig. 2A and C). Additionally, HIF-1α mRNA expression was assessed in Huh7/HepG2 parental and drug-resistant strains using RT-qPCR. The results demonstrated a significant upregulation of HIF-1α mRNA expression in Huh7-LR and HepG2-LR cells compared with their parental counterparts (Fig. 2B and D). These findings indicate that the PI3K/AKT/HIF-1α signaling pathway is aberrantly activated in LR LC cells. This dysregulation appears to be associated with acquired lenvatinib resistance in LC.

To further elucidate the role of the PI3K/AKT/HIF-1α signaling pathway in increased glycolysis and lenvatinib resistance, LR cells were treated with 20 µmol/l (31) LY294002, a broad-spectrum PI3K inhibitor. The expression levels of the key components within the PI3K/AKT/HIF-1α signaling pathway were subsequently assessed using western blotting. LY294002 significantly reduced the phosphorylation levels of PI3K and AKT, as well as the protein levels of HIF-1α in both Huh7-LR (Fig. 3A) and HepG2-LR cells compared with those in control cells and cells treated with lenvatinib (Fig. 3B). In addition, co-administration of LY294002 with lenvatinib further resulted in an even greater suppression of the PI3K/AKT/HIF-1α pathway in LR cells compared with LY294002 treatment alone (Fig. 3A and B).

To decipher the relationship between the PI3K/AKT/HIF-1α signaling pathway and glycolysis in LR LC cells, glycolytic activity following pharmacological inhibition of this pathway was investigated. The findings of the present study revealed that treatment with the PI3K inhibitor LY294002 significantly reduced glucose uptake (Fig. 4A and B), lactate production (Fig. 4C and D) and intracellular ATP levels (Fig. 4E and F), compared with both the control and lenvatinib monotherapy groups. In addition, the expression levels of glycolysis-related genes among different treatment groups were compared. In Huh7-LR cells, the mRNA expression levels of GLUT1, LDHA and HK2 were significantly decreased following LY294002 treatment compared with the control group (Fig. 4G). A similar trend was observed in HepG2-LR cells, although the reduction in GLUT1 expression compared with the control group did not reach statistical significance (Fig. 4H; P=0.053). Notably, the combination of LY294002 and lenvatinib produced a more pronounced suppression of glycolytic activity than LY294002 alone, with both treatment groups showing a significant reduction compared with the control group (Fig. 4A-H). These results demonstrate that the PI3K/AKT/HIF-1α signaling pathway plays a critical role in regulating glycolysis in LR cells. Furthermore, combined LY294002 and lenvatinib treatment attenuates glycolytic activity through synergistic suppression of the PI3K/AKT/HIF-1α pathway.

To investigate the effect of inhibiting the PI3K/AKT/HIF-1α pathway on lenvatinib resistance in LR cells, the clonogenic potential of Huh7-LR and HepG2-LR cells under different treatment conditions was evaluated. Lenvatinib monotherapy failed to inhibit colony formation in LR cells compared with the control. Notably, colony formation was significantly reduced following LY294002 treatment in both Huh7-LR and HepG2-LR cells compared with control or lenvatinib treatment alone. Additionally, the combination of LY294002 and lenvatinib led to a further reduction in colony formation compared with LY294002 monotherapy (Fig. 5A and B). Next, the invasive capacity of LC cells was assessed using Transwell invasion assays. LY294002 treatment significantly decreased the number of invading cells compared with control and lenvatinib treatment. Furthermore, the combined treatment with LY294002 and lenvatinib resulted in a significant reduction in tumor invasion in LR LC cells (Fig. 5C and D). These results demonstrated that inhibition of the PI3K/AKT/HIF-1α signaling pathway significantly enhanced the anti-proliferative and anti-invasive effects of lenvatinib in LR cells.

Subsequently, apoptosis rates in LR cells were analyzed under various treatment conditions. While lenvatinib monotherapy did not alter apoptosis rates in resistant cell lines compared with control, LY294002 significantly induced apoptosis in both LR LC cell lines. As anticipated, the combination of LY294002 and lenvatinib enhanced apoptosis in LR cells to a greater extent than LY294002 alone (Fig. 6A and B). Finally, the effect of LY294002 and lenvatinib on cell cycle distribution was examined by flow cytometry analysis. Results revealed that inhibition of the PI3K/AKT/HIF-1α signaling pathway significantly induced G₀/G₁ phase cell cycle arrest in LR cells, with the combination of LY294002 and lenvatinib further enhancing this arrest in Huh7-LR and HepG2-LR cells compared with LY294002 treatment alone (Fig. 6C and D). These results indicate that inhibition of the PI3K/AKT/HIF-1α signaling pathway significantly potentiates lenvatinib-induced G₀/G₁ phase cell cycle arrest and apoptosis. Collectively, these findings suggest that the PI3K/AKT/HIF-1α pathway plays a pivotal role in mediating lenvatinib resistance in LC. Inhibition of this signaling pathway attenuates glycolytic activity in LR LC cells, thereby resensitizing them to lenvatinib treatment.