This article is mentioned in:

Introduction

According to data from 2024, colorectal cancer (CRC) is the second leading cause of cancer-related mortality worldwide and is the third most frequently diagnosed cancer (1). The development of CRC is a complex, multistep process influenced by a combination of environmental and genetic factors (2). Key risk factors include advanced age, family history, inflammatory bowel disease, dietary habits and lifestyle factors such as smoking and physical inactivity (3). The incidence and mortality rates of CRC vary globally, with higher rates typically observed in developed countries (4); these differences may be due to lifestyle variables, such as diets heavy in fat and poor in fiber, a lack of physical activity and greater screening rates that result in more frequent diagnoses. Despite advances in screening programs and the introduction of various therapeutic interventions, including radiation, chemotherapy, surgery and targeted therapy, CRC remains a significant public health issue. Patient prognosis largely depends on the tumor stage and molecular profile of the cancer (5). Although progress has been made, many questions remain unanswered regarding the understanding of the molecular mechanisms of CRC, particularly in terms of early diagnosis and therapeutic strategies. For example, aberrant activation of the Wnt/β-catenin signaling pathway is a common molecular event in CRC, but its specific role in disease progression has not been fully elucidated (6). The heterogeneity of CRC presents challenges in achieving optimal clinical outcomes, emphasizing the need for innovative diagnostic biomarkers, therapeutic strategies and prognostic indicators to enhance patient management and outcomes in CRC.

Lectin galactoside-binding soluble 4 (LGALS4), a member of the galactose lectin family, serves a role in a variety of biological processes (7). In CRC, the expression level of LGALS4 is closely associated with tumor progression, invasiveness and drug resistance. It has been reported that LGALS4 expression is high in normal colonic epithelial cells and significantly lower in CRC tissues, suggesting that it may function as a tumor suppressor (7). Loss of function of LGALS4 has been associated with enhanced proliferation, migration and invasion of tumor cells (7). At the molecular level, LGALS4 is able to induce cell cycle arrest by regulating the levels of cell cycle-related proteins, such as Cyclin D1, p21 and p15, thereby inhibiting the proliferation of tumor cells (8). In addition, LGALS4 can inhibit tumor development by affecting the Wnt/β-catenin signaling pathway, an aberrant activation pathway commonly found in CRC (9). Increased expression of LGALS4 levels reduces the expression level of β-catenin and upregulates inhibitory factors of the Wnt signaling pathway, such as Ephrin B1, which suppresses the proliferation and invasion of tumor cells. At the clinical level, the expression level of LGALS4 correlates with the prognosis of patients with CRC and the decrease in its expression may predict the severity of the disease and the response to treatment (10). In addition, the expression level of LGALS4 correlates with tumor resistance to the chemotherapeutic agent oxaliplatin, suggesting its potential for therapeutic application (10).

Glycolysis, a crucial metabolic pathway, produces pyruvate from glucose conversion while producing ATP and NADH (11). In CRC, glycolysis is significantly upregulated, an occurrence known as the Warburg effect, where cancer cells predominantly rely on aerobic glycolysis even when oxygen is abundant (12). This metabolic reprogramming supports rapid cell proliferation and tumor growth, making glycolysis a focal point in CRC research. Zuo et al (13) investigated the impact of the long non-coding RNA maternally expressed gene 3 (MEG3) on glycolysis in CRC, reporting that MEG3 suppresses glycolysis by promoting the ubiquitin-dependent degradation of c-Myc, a key regulator of glycolytic genes. Overexpression of MEG3 significantly reduces glycolysis, glycolytic capacity and lactate production in CRC cells; conversely, MEG3 knockdown produces the opposite result. Additionally, MEG3 activation by vitamin D suggests a potential therapeutic value in CRC treatment through glycolysis modulation. Similarly, Zhu et al (14) reported that microRNA (miR)-146b-5p enhances cell proliferation, glycolysis and invasiveness in CRC by specifically targeting the pyruvate dehydrogenase E1 subunit b. Overexpression of miR-146b-5p enhanced these processes, while knockdown inhibited them, underscoring its oncogenic role. Zhu et al (15) identified five glycolysis-related genes (enolase 3, glypican 1, prolyl 4-hydrxylase subunit a 1, sperm associated antigen 4 and stanniocalcin 2) that form a prognostic model, highlighting the significant influence of aerobic glycolysis on CRC development. These studies underscore the critical role of glycolysis and its potential for CRC therapy, offering valuable insights into tumor metabolism and potential therapeutic strategies.

Despite the established role of glycolysis in CRC and the known tumor-suppressive function of LGALS4, the precise mechanisms by which LGALS4 regulates glycolysis and its potential as a therapeutic target in CRC remain understudied. The present study aimed to address these gaps by investigating the role of LGALS4 in modulating glycolysis and its interplay with the β-catenin signaling pathway in CRC cells. These findings introduced a novel perspective for CRC treatment, highlighting the potential of LGALS4 as a therapeutic target and prognostic indicator, which has not been extensively reported in previous research.

Materials and methods

Data acquisition and analysis of differentially expressed genes (DEGs)

COAD samples from the TCGA were obtained via the Clinical Bioinformatics Assistant website (https://www.aclbi.com/static/index.html#/). The GSE26571CRC microarray dataset was obtained from the GEO database (https://www.ncbi.nlm.nih.gov/gds/). The TCGA-COAD dataset comprised 455 tumor samples and 41 normal control samples, while the GSE26571 dataset included 12 CRC samples and 5 control samples. Probe IDs were transformed into gene symbols and differential expression analysis was employed using the ‘Limma’ package (version 3.46.0) in R (version 4.1.2; Posit Software). According to previous literature, genes with a fold change (FC) >2 were identified as upregulated DEGs and those with an FC <0.5 as downregulated DEGs (16,17).

Identification and enrichment analysis of overlapping DEGs

The overlapping DEGs were subjected to Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses using the Database for Annotation, Visualization and Integrated Discovery. The GO analysis included the biological process (BP), molecular function (MF) and cellular component (CC) categories to comprehensively explore the functional roles of the identified overlapping DEGs.

Construction of prognostic risk model for overlapping DEGs

The Least Absolute Shrinkage and Selection Operator (LASSO) Cox regression model was used to identify genes associated with patient prognosis. The λ parameter in the LASSO model was used to adjust the strength of selection of the variables in the model to prevent overfitting. Smaller values of λ, which allowed more variables to enter the model but may lead to excessive model complexity and larger values of λ, which increased the penalization strength and retained only the most significant variables, simplified the model structure. The optimal λ value, λ.min=0.0382 was determined, through 10-fold cross-validation to ensure that the model achieved the best balance between predictive accuracy and simplicity. This process of optimizing λ enabled the model to select only those genes with the strongest associations with patient survival outcomes, which included ribosomal protein S17 (RPS17), G2/mitotic-specific Cyclin B1 (CCNB1), karyopherin subunit a2 (KPNA2), TNF receptor associated protein 1 (TRAP1), complement C8 chain (C8G), LGALS4, lectin galactoside-binding soluble-2 (LGALS2), carbonic anhydrase 4 (CA4), and solute carrier family 6 member 8 (SLC6A8). The risk score was calculated using the following formula: Risk score=∑(coefficient × gene expression). Where the coefficient represented the coefficient of each gene in the LASSO model and gene expression was the expression level of the corresponding gene in the sample. The risk score showed the effect of each gene on patient prognosis, with positive coefficients indicating positive correlation and negative coefficients indicating negative correlation. A model capable of assessing the prognostic risk of patients was constructed, with high risk scores predicting poorer prognostic outcomes. The score of risk for each sample was calculated according to the following formula: Risk score=(0.1219) × RPS17 + (−0.0582) × CCNB1 + (−0.0442) × KPNA2 + (−0.0137) × TRAP1 + (0.0885) × C8G + (−0.1236) × LGALS4 + (−0.009) × LGALS2 + (−0.0063) × CA4 + (0.0012) × SLC6A8. The COAD cohort from the TCGA database were stratified into low- and high-risk groups according to the expression patterns of the aforementioned genes. Kaplan-Meier (KM) analysis was employed to ascertain the overall survival (OS) probability for the two risk groups. The median survival time was calculated and the statistical significance of survival differences between the groups was evaluated using the log-rank test. To clarify relative risk, hazard ratios (HRs) were calculated for the high-risk group. Additionally, Receiver Operating Characteristic (ROC) curves were generated using the timeROC package (https://CRAN.R-project.org/package=timeROC) and the Area Under the Curve (AUC) values were calculated to evaluate the predictive ability of the prognostic models for patient survival at 1, 3 and 5 years. Higher AUC values indicated stronger prognostic prediction capabilities.

Construction of prognostic nomogram and expression analysis of prognostic significant genes

Univariate and multivariate Cox regression analyses were performed on signature genes and specific clinical predictive variables, such as patient age and sex, using the ‘forestplot’ package (version 2.0.1; http://cran.r-project.org/web/packages/forestplot/vignettes/forestplot.html). For each variable, the HRs, 95% CIs and P-values were calculated. Key prognostic factors were identified based on variables with P<0.05. Nomograms were created to predict the 1-, 3- and 5-year survival probabilities using the rms software (version 3.6.1; https://cran.r-project.org/web/packages/rms/index.html; provided by R Foundation for Statistical Computing) (18). The consistency index (C-index) was determined to assess the prediction accuracy of the model. The performance of the model was then assessed by predicting survival curves for 1, 3 and 5 years and by generating a calibration curve to evaluate the accuracy of these predictions. The expression levels of C8G, LGALS4 and RPS17 in different samples were examined in the TCGA-COAD and GSE26571 datasets using the SangerBox platform (version 3.0; http://vip.sangerbox.com/home.html).

Cell lines and culture

Based on previous literature, the LoVo, HCT-116 and SW480 cell lines were selected and NCM460 cells were used as the control cell line (19–21). LoVo, HCT-116 and SW480 cells were obtained from the Cell Bank of the Chinese Academy of Sciences. NCM460 cells were purchased from Jennio Biotech Co., Ltd. All cells were cultured in DMEM, 1% penicillin-streptomycin and 10% FBS (Gibco; Thermo Fisher Scientific, Inc.). Cells were maintained at 37°C with 5% CO2 and 20% O2.

Vector construction and transfection

Total RNA was isolated from NCM460 cells using the TRIzol® kit (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. cDNA was synthesized using the PrimeScript RT kit (Takara Bio Inc.). cDNA was amplified using the CloneAmp™ HiFi PCR Premix (Takara Bio, Inc.) to amplify the corresponding LGALS4 fragment, which was cloned into the pcDNA™3.1(+) vector (Thermo Fisher Scientific, Inc.). The correct insertion of the gene was verified by DNA sequencing and the recombinant plasmid was transformed and amplified in Escherichia coli to prepare plasmid DNA for cell transfection. For transfection, 1 µg plasmid DNA was used per well, and transfection was performed at 37°C for 24 h. The LGALS4 overexpression plasmid was introduced into LoVo and HCT-116 cells using Lipofectamine® 3000 (Invitrogen; Thermo Fisher Scientific, Inc.) as the transfection reagent, and cells were incubated for 48 h post-transfection before further experimentation. Cells stably expressing LGALS4 were screened by resistance. The selection method involved puromycin at a concentration of 1 µg/ml for selection and 0.5 µg/ml puromycin for maintenance. The amplification primer sequences used were as follows: Forward (F), 5′-CTCGAGATGGCCTATGTCCCC-3′ and reverse (R), 5′-TCTAGATTAGATCTGGACATAGGACAAGG-3′.

Cell treatment

CRC cells were treated with various agents to investigate their effects on glucose metabolism and cell viability. Glucose was administered at concentrations of 10.0, 1.0 and 0.5 mM. To inhibit glucose transport and hexokinase activity, cells were treated with cytochalasin B (Cyto-B) at 20 µM and 3-bromopyruvate (3-BrPA) at 10 µg/ml. Additionally, the 5-fluorouracil (5-FU) anticancer drug was used at 50 µg/ml as an and XAV-939, a β-catenin inhibitor, was applied at 10 µM to examine its impact on CRC cells. All treatments were administered for 48 h.

Reverse transcription-quantitative PCR (RT-qPCR)

TRNzol reagent (cat. no. DP424; Tiangen Biotech Co., Ltd.) was used to extract the total RNA from CRC cells following the manufacturer's instructions. cDNA synthesis was performed using a PrimeScript RT kit (Takara Bio Inc.). RT-qPCR was conducted on the StepOnePlus Real-Time PCR System (Applied Biosystems; Thermo Fisher Scientific, Inc.) using SYBR Green PCR Master Mix (Takara Biotechnology Co., Ltd.). The thermocycling conditions were as follows: Initial denaturation at 95°C for 10 min, followed by 40 cycles of denaturation at 95°C for 15 sec, annealing at 60°C for 30 sec and extension at 72°C for 30 sec. Gene expression levels were quantified, normalized to GAPDH and calculated using the 2−ΔΔCq method (22). Table I lists the primer sequences for the genes investigated in the present study.

Table I. Primer sequences used for reverse transcription-quantitative PCR.

Table I.

Primer sequences used for reverse transcription-quantitative PCR.

Gene	Sequence (5′-3′)
LGALS4	F: CCGGACATTGCCATCAACAG
	R: CAAAGCTCTTGCCTGTGGGA
BAX	F: GAGCTAGGGTCAGAGGGTCA
	R: CCCCGATTCATCTACCCTGC
BCL2	F: ACCTACCCAGCCTCCGTTAT
	R: GAACTGGGGGAGGATTGTGG
CASP3	F: GCTGGATGCCGTCTAGAGTC
	R: ATGTGTGGATGATGCTGCCA
CASP9	F: ATTGCACAGCACGTTCACAC
	R: TATCCCATCCCAGGAAGGCA
CTNNB1	F: AAAGCGGCTGTTAGTCACTGG
	R: CGAGTCATTGCATACTGTCCAT
MYC	F: CACCACCAGCAGCGACTCT
	R: GGCACCTCTTGAGGACCAGT
PKM	F: GACTGCCTTCATTCAGACCCA
	R: GGGTGGTGAATCAATGTCCAG
SLC2A1	F: ATGCGGGAGAAGAAGGTCAC
	R: GTTGACGATACCGGAGCCAA
LDHA	F: GGCCTGTGCCATCAGTATCT
	R: GGAGATCCATCATCTCTCCC
HK2	F: CCTGTGAATCGGAGAGGTCC
	R: ATTTTGGCGTCACAACTGCT
GAPDH	F: ACCACAGTCCATGCCATCAC
	R: TCCACCACCCTGTTGCTGTA

[i] F, forward; R, reverse; LGALS4, lectin galactoside-binding soluble 4; CTNNB1, catenin beta 1; MYC, Myc proto-oncogene; PKM, pyruvate kinase M1/M2; SLC2A1, solute carrier family 2 member 1; LDHA, lactate dehydrogenase A; HK2, hexokinase 2.

Western blot (WB) assay

Protease and phosphatase inhibitors (CoWin Biosciences) were added to RIPA lysis buffer (Beijing Solarbio Science & Technology Co., Ltd) to facilitate the preparation of protein lysates from CRC cells. The BCA Protein Assay Kit (Beyotime Institute of Biotechnology) was used to measure the protein concentration. Proteins in equal quantities (20 µg/lane) were separated using 10% SDS-PAGE and then transferred onto PVDF membranes (Beyotime Institute of Biotechnology). The membranes were blocked with 5% skim milk at room temperature for 1 h. After blocking, the membranes were washed three times with Tris-buffered saline containing 0.1% Tween-20. They were then incubated with primary antibodies against LGALS4 (1:1,000; cat. no. ab175185), Cyclin dependent kinase 1 (CDK1) (1:1,000; cat. no. ab265590), Cyclin A2 (1:1,000; cat. no. ab227277), Cyclin B1 (1:2,000; cat. no. ab181593), Bax (1:1,000; cat. no. ab32503), Bcl-2 (1:1,000; cat. no. ab194583), caspase-9 (1:1,000; cat. no. ab32539), caspase-3 (1:1,000; cat. no. ab90437), β-catenin (1:1,000; cat. no. ab227499), c-myc (1:1,000; cat. no. ab32072), solute carrier family 2 member 1 (GLUT1) (1:5,000; cat. no. ab195021), pyruvate kinase M2 (PKM2) (1:1,000; cat. no. ab154816), lactate dehydrogenase A (LDHA) (1:1,000; cat. no. ab300637) and hexokinase 2 (HK2) (1:1,000; cat. no. ab209847) (all from Abcam) overnight at 4°C. After rinsing, the membranes were incubated with secondary antibodies conjugated with HRP (1:5,000; cat. no. D110011-0100; Sangon Biotech Co., Ltd) for 1 h at room temperature. GAPDH (1:5,000; cat. no. KC-5G4; Aksomics Inc.) was used as an internal reference. Protein bands were visualized using an ECL kit (Tiangen Biotech Co., Ltd.) and imaged using a ChemiDoc system (Bio-Rad Laboratories, Inc.). Densitometric analysis was performed using ImageJ software (version 2.0.0; National Institutes of Health).

Cell counting kit-8 (CCK-8) assay

The CCK-8 assay kit (cat. no. KGA317s; Nanjing KeyGen Biotech Co., Ltd) was used to evaluate cell viability. CRC cells were seeded in 96-well plates at a density of 5×103 cells/well. Following transfection with either the vector control or LGALS4 overexpression plasmid, CCK-8 reagent was added to each well and incubated for 2 h at 37°C. A microplate reader (Shanghai Kehua Bio-Engineering Co., Ltd.) was used to measure the absorbance of each sample at 450 nm after 0, 1, 2, 3, 4 and 5 days.

Cell invasion and migration assays

Transwell assays were used to assess cell invasion and migration. Transfected CRC cells (5×104 cells) were suspended in serum-free DMEM in the upper chamber of the plate. Subsequently, DMEM supplemented with 10% FBS was added to the lower chamber of the plate. Cells were incubated at 37°C in a 5% CO2 incubator for 24 h to allow migration. After the incubation period, cells that had migrated to the underside of the Transwell membrane were fixed using 4% paraformaldehyde at room temperature for 15 min. Cells were then stained using DAPI for 10 min at room temperature to visualize their nuclei. Finally, the number of migrating cells in the field of view were visualized using an inverted fluorescence microscope. For the cell invasion experiments, the top chamber of the Transwell was coated with Matrigel (BD Biosciences) at 37°C for 20 min, as previously reported (23).

Flow cytometry

For the assessment of cellular phenotypes via flow cytometry, CRC cells were enzymatically dissociated using 0.25% trypsin-EDTA (Thermo Fisher Scientific, Inc.) at 37°C for 5 min to obtain a single-cell suspension. The cells were then lysed with 100 µl ice-cold RIPA buffer (containing protease and phosphatase inhibitors) on ice for 30 min at 4°C to ensure complete cell lysis. The cells were then subjected to dual staining with propidium iodide (PI) for DNA content analysis and annexin V conjugated to FITC (BD Biosciences) for the detection of apoptotic cells, following the manufacturer's recommended protocols. Staining was performed at room temperature for 15 min in the dark. This procedure allowed for the discrimination of viable, early apoptotic and necrotic cells based on their distinct fluorescence characteristics. The PI staining also facilitated the evaluation of the cell cycle distribution by reflecting the DNA content at various stages (24–26). For cell cycle analysis, cells were also treated with RNase A (10 µg/ml) for 15 min at 37°C to ensure accurate DNA content analysis. The stained cells were analyzed using a CytoFLEX flow cytometer (Beckman Coulter, Inc.) and the resulting data were processed with FlowJo software (version 10.6.0; FlowJo LLC)to quantify the apoptotic rate and to determine the cell cycle phase distribution. The percentage of cells in the G1, S and G2 phases were calculated and statistically compared across different experimental conditions [transfection with either LGALS4 overexpression plasmid or vector control, followed by treatment with 5-FU (5.0 µM) or 0.5 mM glucose] to assess the modulation of cell cycle progression by the treatments administered.

Glucose uptake assay

2-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino]-2-deoxyglucose (2-NBDG; cat. no. 186689-07-6; Anjiekai Biological Medicine Technology) was used as a glucose tracer. Cells were seeded at 1×105 cells/well in 6-well plates in four replicates and incubated overnight at 37°C and 5% CO2. The following day, cells were subjected to glucose starvation for 4 h. Then, one well was treated with DMEM containing glucose as a negative control, while the other wells were incubated with 2-NBDG for 2 h at 37°C. After the incubation, cells were collected by trypsinization, followed by centrifugation at 4°C for 5 min at 300 × g, and rinsed twice with PBS. Subsequently, flow cytometry was performed using a CytoFLEX flow cytometer). The geometric mean fluorescence intensity of the cells was measured, with emission at 530 nm and excitation at 488 nm. Data analysis was conducted using FlowJo software, version 10.6.0.

Colony formation assay

A colony formation assay was used to assess the colony formation capacity of cells. Cells were seeded at a concentration of 2×103 in 60 mm plates and cultured for 2 weeks at 37°C with 5% CO2, under varying glucose concentrations (10.0, 1.0 and 0.5 mM). After fixation in methanol at room temperature for 30 min, cells were stained using alkali nitro tetrazolium blue chloride at room temperature for 20 min. A chemiluminescence imager (Clinx) was used to image the cells, and the colonies were counted manually. Colonies were defined as cell clusters containing ≥50 cells. Each experiment was performed thrice to ensure accuracy.

Lactate and ATP assays

To measure lactate production, 1×105 cells/well were seeded in 24-well plates and cultured at 37°C with 5% CO2 for 24 h. This process was repeated three times. For ATP production measurement, 1×105 cells/well were seeded in 96-well plates and incubated at 37°C with 5% CO2 for 24 h in five replicates. Subsequently, the medium in both assays was replaced with DMEM containing 1 mM glucose and the cells were incubated overnight at 37°C with 5% CO2. The following day, lactate in the culture media was measured using a commercial lactate assay kit (cat. no. A109-2; Nanjing Jiancheng Bioengineering Institute). The relative ATP concentration was also measured using a commercial ATP assay kit (cat. no. S0026; Beyotime Institute of Biotechnology). Both kits were used according to the manufacturers' instructions.

Statistical analysis

Statistical analysis was conducted using R software. Each experiment was carried out in triplicate and the results were presented as the mean ± SD. For survival curve analysis, Kaplan-Meier analysis was used to evaluate the overall survival, and the statistical significance between survival curves was assessed using the log-rank test. Differences between groups were evaluated using a one-way ANOVA when there were more than two experimental groups, followed by post-hoc analysis with Tukey's test. When comparing only two groups, an unpaired Student's t-test was applied P<0.05 was used to indicate a statistically significant difference.

Results

Identification and enrichment analysis of overlapping DEGs in TCGA-COAD and GSE26571 datasets

A total of 1,171 downregulated DEGs and 1,521 upregulated DEGs were identified in COAD samples compared with normal samples in the TCGA database (Fig. 1A). From the GSE26571 dataset, 356 upregulated DEGs and 338 downregulated DEGs were identified in colon cancer samples compared with normal samples (Fig. 1B). Subsequently, using bioinformatics platforms, a cross-analysis of the upregulated and downregulated DEGs from the TCGA-COAD dataset and the GSE26571 dataset identified a total of 175 overlapping DEGs (Fig. 1C and D). The enrichment analysis demonstrated significant enrichment in terms related to ‘DNA replication’ (GO: 0006260, BP), ‘response to zinc ion’ (GO: 0010043, BP), ‘CMG complex’ (GO: 0071162, CC), ‘small ribosomal subunit’ (GO: 0015935, CC), ‘RNA binding’ (GO: 0003723, MF) and ‘cadherin binding’ (GO: 0045296, MF) (Fig. 1E), as well as pathways involved in the ‘cell cycle’, ‘DNA replication’ and ‘ribosome’ (Fig. 1F).

Figure 1. Identification and enrichment analysis of overlapping DEGs in TCGA and GSE26571 datasets. Volcano plots for screening DEGs in the (A) TCGA-COAD and (B) GSE26571 datasets. Venn diagrams showing the cross-analysis of upregulated and downregulated DEGs in the (C) TCGA-COAD and (D) GSE26571 datasets. (E) GO and (F) KEGG enrichment analysis of the overlapping DEGs. DEGs, differentially expressed genes; TCGA, The Cancer Genome Atlas; COAD, colon adenocarcinoma; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; sig, significant; up, upregulated; down, downregulated.

Prognostic analysis of nine signature genes in the risk model

A total of 175 overlapping DEGs (Table SI) were analyzed using the LASSO Cox regression method. By employing cross-validation to identify the optimal lambda value of 0.0382, nine genes (Table SII) were identified that served as significant predictors of patient outcomes (Fig. 2A and B). These genes, identified through their ability to discriminate between high-risk and low-risk groups, were characterized by their expression patterns that strongly associated with survival rates. The high-risk group, as determined by the risk model, exhibited increased mortality and diminished overall survival compared with the low-risk group, a finding further supported by the Kaplan-Meier survival analysis (Fig. 2C). A median survival time of 5 years was identified for both groups, with a HR of 2.537, indicating a pronounced impact on survival probability (Fig. 2D). Moreover, the model's predictive accuracy was validated by the ROC curve, which demonstrated an area under the curve of 0.696 at the 5 year mark, signifying a reliable prediction of survival outcomes. This selection process and subsequent analysis underscored the robustness of the risk model and suggested a potential role for the identified genes in CRC prognosis.

Figure 2. Prognostic analysis of nine signature genes in overlapping DEGs. (A) Least Absolute Shrinkage and Selection Operator-Cox regression model analysis on 175 overlapping differentially expressed genes. The absolute values of the coefficients for the different genes were plotted against their corresponding L1 regularization norm values. (B) The relationship between 10-fold cross-validation partial likelihood deviation and log(λ). (C) Risk model analysis of the selected sample data. (D) Kaplan-Meier survival curve analysis of the high-risk group and the low-risk group in the risk model (E) Receiver operating characteristic curve analysis of the risk model at 1, 3 and 5 years post-treatment. AUC, area under the curve; RPS17, ribosomal protein S17; CCNB1, G2/mitotic-specific Cyclin-B1; KPNA2, karyopherin subunit 2; TRAP1, TNF receptor associated protein 1; C8G, complement C8 chain; LGALS, lectin galactoside-binding soluble 4; CA4, carbonic anhydrase 4; SLC6A8, solute carrier family 6 member 8.

Nomogram analysis of key prognostic variables and screening of hub genes

A total of three factors were identified as statistically significant after analyzing 9 genes and 2 clinical variables in the risk model: C8G, LGALS4 and RPS17 (Fig. 3A and B). Using these insights, a predictive model with a C-index of 0.634 was constructed (Fig. 3C). The calibration curve showed the highest consistency with model predictions at 1 year, followed by 3 and 5 years, indicating that these variables have predictive power for patient survival (Fig. 3D). The expression of the three prognostic significant genes was assessed in both the TCGA-COAD and GSE26571 datasets. These findings indicated that C8G was significantly under-expressed in tumor samples in the TCGA-COAD dataset, but was not significantly expressed in the GSE26571 dataset. LGALS4 was notably under-expressed in the tumor samples of both datasets, while RPS17 was significantly over-expressed (Fig. 3E and F). A previous study reported that LGALS4 is a potential prognostic factor in CRC patients, but its role in CRC glycolysis remains unclear (27). Therefore, LGALS4 was selected as a hub gene for further investigation.

Figure 3. Nomogram analysis of key prognostic variables and screening of hub genes. (A) Univariate and (B) multifactorial Cox regression analysis of signature genes and clinical prognostic variables. (C) Nomograms of key prognostic variables, along with the C-index, were used to evaluate the model's predictive power. A C-index close to 1 indicated high predictive accuracy, while a C-index of 0.5 indicated no predictive power. (D) The dashed lines represented the ideal calibration curves of the nomogram Expression levels of C8G, LGALS4 and RPS17 in the (E) TCGA-colon adenocarcinoma and (F) GSE26571 datasets. Purple represented normal samples and red represented tumor samples. *P<0.05, ***P<0.001 vs. normal group. TCGA, The Cancer Genome Atlas; pro, probability.

Overexpression of LGALS4 inhibited the proliferation, migration and invasion of CRC cells

The expression of LGALS4 was assessed in normal cells (NCM460) and CRC cells (LoVo, HCT-116 and SW480) using RT-qPCR and WB. These results demonstrated that LGALS4 was significantly under-expressed in LoVo and HCT-116 cells compared with normal cells (Fig. 4A-C). Therefore, these two cell lines were selected for further experiments. The RT-qPCR and WB analyses showed efficient overexpression of LGALS4 in LoVo and HCT-116 cells (Fig. 4D-F). Based on the results of the CCK-8 assay, on day 5 of cell culture, a significant decrease in cell viability of cells overexpressing LGALS4 was observed in the LoVo cell line, which was ~25% compared with the control cells. For the HCT-116 cell line, the cell viability of cells overexpressing LGALS4 also showed a significant decrease to ~50% compared with the control (Fig. 4G-H). These results suggested that LGALS4 may inhibit cell proliferation by modulating the expression of cell cycle-related proteins or their activities, which in turn inhibits cell proliferation. The Transwell assays demonstrated that the invasive and migratory abilities of LoVo cell lines were significantly decreased upon LGALS4 overexpression compared with control cells (Fig. 4I-L). Specifically, the invasion capacity was decreased 4-fold and the migration capacity was decreased 2.5-fold compared with the control. A similar effect was demonstrated in the HCT-116 cell line, in which the invasive capacity was reduced by 3-fold and the migratory capacity by 2-fold compared with control cells. This effect may stem from the effects of LGALS4 on cell adhesion, reorganization of the cytoskeleton or degradation of the extracellular matrix, which are all critical aspects of the cell migration and invasion process (28).

Figure 4. Overexpression of LGALS4 inhibits the proliferation, migration and invasion of CRC cells. (A) RT-qPCR was used to detect the expression levels of LGALS4 in normal cells (NCM460) and CRC cells (LoVo, HCT-116 and SW480). (B) WB analysis was performed to detect the protein expression levels of LGALS4 in normal cells (NCM460) and CRC cells (LoVo, HCT-116, SW480). (C) Semi-quantification of the WB analysis showing the protein expression levels of LGALS4 in normal cells (NCM460) and CRC cells (LoVo, HCT-116, SW480). *P<0.05 vs. NCM460. (D) RT-qPCR was used to assess the overexpression efficiency of LGALS4 in LoVo and HCT-116 cells. (E) WB analysis further confirmed the overexpression efficiency of LGALS4 in LoVo and HCT-116 cells. (F) Semi-quantification of the WB analysis further confirming the overexpression efficiency of LGALS4 in LoVo and HCT-116 cells. The Cell Counting Kit-8 assay was used to detect the effect of LGALS4 overexpression on the viability of (G) LoVo and (H) HCT-116 cells. The Transwell assay was used to detect the effect of LGALS4 overexpression on the (I and J) invasion and (K and L) migration capacities of CRC cells. *P<0.05, **P<0.01 vs. Vector. RT-qPCR, reverse transcription-quantitative PCR; WB, western blot; LGALS4, lectin galactoside-binding soluble 4; over, overexpression; CRC, colorectal cancer.

LGALS4 overexpression induces cell cycle arrest in CRC cells

To explore the effect of LGALS4 overexpression on the cell cycle progression of CRC cells, flow cytometric analysis was performed on LoVo and HCT-116 cells that were transfected with either an LGALS4 overexpression vector or a control vector. These results demonstrated a significant effect of overexpression of LGALS4 on the cell cycle distribution of LoVo and HCT-116 cells. Compared with control cells, the proportion of cells overexpressing LGALS4 was significantly higher in the G1 phase and significantly lower in the S phase. In LoVo cells, overexpression of LGALS4 resulted in a ~1.5-fold increase in the proportion of cells in G1 phase and a ~10-fold decrease in the proportion of cells in S phase compared with the control cells. For HCT-116 cells, overexpression of LGALS4 resulted in a ~1.8-fold increase in the proportion of G1-phase cells and a ~15-fold decrease in the proportion of S-phase cells compared with control cells (Fig. 5A-D). The expression levels of cell cycle-related proteins Cyclin B1, CDK1 and Cyclin A2 were further analyzed as these proteins serve a key role in driving the cell cycle transition from G1 to S phase (29). These results demonstrated that LGALS4 overexpression significantly decreased the protein expression levels of CDK1, Cyclin B1 and Cyclin A2 in CRC cells compared with the control group (Fig. 5E-G). These results suggested that LGALS4 overexpression may cause a G1 phase cell cycle arrest by downregulating the levels of cell cycle regulatory proteins.

Figure 5. LGALS4 overexpression induces cell cycle arrest in CRC cells. (A) Flow cytometric analysis showing the effect of LGALS4 overexpression on the cell cycle in LoVo cells. (B) Quantification of flow cytometry data showing the effect of LGALS4 overexpression on the cell cycle in LoVo cells. (C) Flow cytometric analysis showing the effect of LGALS4 overexpression on the cell cycle in HCT-116 cells. (D) Quantification of flow cytometry data showing the effect of LGALS4 overexpression on the cell cycle in HCT-116 cells. (E) Western blot analysis showing the protein expression levels of cell cycle-related proteins CDK1, Cyclin B1 and Cyclin A2 in LoVo and HCT-116 cells following LGALS4 overexpression. Semi-quantification of protein expression levels of CDK1, Cyclin B1 and Cyclin A2 in (F) LoVo and (G) HCT-116 cells following LGALS4 overexpression, presented as bar graphs. *P<0.05, ***P<0.001 vs. vector. CRC, colorectal cancer; LGALS4, lectin galactoside-binding soluble 4; over, overexpression; CDK1, Cyclin dependent kinase 1.

Overexpression of LGALS4 promoted CRC cell apoptosis

Flow cytometry was used to analyze the effect of LGALS4 overexpression on the apoptosis of LoVo and HCT-116 cells. These results demonstrated that the apoptosis rate of CRC overexpressing LGALS4 was significantly increased by ~2.5-fold compared with control cells (Fig. 6A-C). RT-qPCR results showed that LGALS4 overexpression significantly increased the expression levels of CASP3, BAX and CASP9 and decreased the expression level of BCL2 (Fig. 6D and E). These results were also confirmed by WB assay results (Fig. 6F-H).

Figure 6. Overexpression of LGALS4 promotes CRC cell apoptosis. Flow cytometric analysis showing the effect of LGALS4 overexpression on apoptosis in (A) LoVo and (B) HCT-116 cells. (C) Quantification of apoptosis in LoVo and HCT-116 cells following LGALS4 overexpression. RT-qPCR analysis showing the relative mRNA expression levels of apoptosis-related genes (BAX, BCL2, CASP3, and CASP9) in (D) LoVo and (E) HCT-116 cells following LGALS4 overexpression. (F) Western blot analysis showing the protein expression of Bax, Bcl-2, caspase-3, and caspase-9 in LoVo and HCT-116 cells following LGALS4 overexpression. Semi-quantification of protein expression levels of Bax, Bcl-2, caspase-3 and caspase-9 in (G) LoVo and (H) HCT-116 cells following LGALS4 overexpression. *P<0.05 vs. vector. CRC, colorectal cancer; RT-qPCR, reverse transcription-quantitative PCR; LGALS4, lectin galactoside-binding soluble 4; CASP3, caspase-3; CASP9, caspase-9.

Overexpression of LGALS4 inhibited aerobic glycolysis in CRC

Cancer cells typically rely on higher glucose concentrations to maintain their rapid glycolytic processes. By depriving glucose, the nutrient-limited conditions in the tumor microenvironment can be mimicked and the adaptation and survival of cancer cells to stressful conditions can be studied (30). Colony formation assays delineated the glucose dependency of LGALS4-overexpressing cells, demonstrating an inherent reliance on exogenous glucose for optimal proliferation. Compared with control cells, LGALS4 overexpressing cells showed enhanced survival under glucose-scarce conditions, particularly at a glucose concentration of 0.5 mM, suggesting that they may have acquired metabolic adaptations to the glucose-deficient microscopic environment (Fig. 7A-D). Subsequent flow cytometric analysis, utilizing Annexin V-FITC staining, corroborated the diminished apoptotic propensity of LGALS4-overexpressing CRC cells subjected to glucose deprivation at 0.5 mM, underscoring their enhanced survival kinetics relative to control cell populations (Fig. 7E-G). Cyto-B and 3-BrPA target the initiation steps of the glycolytic pathway such as glucose uptake and the first phosphorylation reaction, respectively (31). The use of these inhibitors could aid in the understanding of the metabolic adaptations of LGALS4 overexpressing cells when the glycolytic pathway is inhibited. Targeted inhibition of glucose transport and hexokinase activity was performed using Cyto-B at a concentration of 20 µM and 3-BrPA at a concentration of 10 µg/ml, respectively. The cytotoxic effects of these inhibitors on LGALS4-overexpressing cells were ascertained through CCK-8 assays following a 48 h incubation. These results demonstrated a preservation of cell viability among LGALS4-overexpressing cells, compared with control cells, suggesting the potential presence of a recalibrated metabolic phenotype conferring resistance to glycolytic inhibition (Fig. 7H and I). The increased survival of cyto-B-treated control cells may be due to the fact that the inhibitory effect of cyto-B on glucose transport did not completely block the energy supply of the cells, and the cells may be sustained by other metabolic pathways such as fat oxidation or amino acid metabolism. In addition, the concentration of cyto-B may not be sufficient to completely inhibit glucose uptake, or the cells may be somewhat adapted to cyto-B treatment.

Figure 7. Overexpression of LGALS4 inhibits aerobic glycolysis in CRC cells. (A) Colony formation assays were performed on LoVo cells after overexpression of LGALS4 and treatment with 10.0, 1.0 and 0.5 mM of Glu. (B) Semi-quantification of colony formation in LoVo cells overexpressing LGALS4. (C) Colony formation assay showing the effect of LGALS4 overexpression on HCT-116 cells following treatment with 10.0, 1.0 and 0.5 mM Glu. (D) Semi-quantification of colony formation in HCT-116 cells overexpressing LGALS4. Flow cytometric analysis showing the apoptosis of (E) LoVo and (F) HCT-116 cells treated with 0.5 mM Glu after LGALS4 overexpression. (G) Quantification of apoptosis in LoVo and HCT-116 cells treated with 0.5 mM Glu after LGALS4 overexpression. Quantification of cell viability in (H) LoVo and (I) HCT-116 cells overexpressing LGALS4 or a control vector under treatment with 0.5 mM Glu, 20 µM Cyto-B, and 10 µg/ml 3-BrPA. *P<0.05, **P<0.01 vs. vector. CRC, colorectal cancer; Glu, glucose; Cyto-B, cytochalasin B; 3-BrPA, 3-Bromopyruvate; LGALS4, lectin galactoside-binding soluble 4; over, overexpression; ns, not significant.

LGALS4 overexpression enhanced 5-FU-induced apoptosis and inhibited glucose metabolism in CRC cells

5-FU is an anticancer drug used to treat various types of cancer, including CRC. It typically inhibits DNA synthesis by interfering with the biosynthesis of pyrimidine nucleotides (32). Flow cytometry analysis showed that overexpression of LGALS4 significantly increased apoptosis of CRC cells compared with control cells, and the pro-apoptotic effect was significantly increased when combined with 5-FU treatment compared with the control (Fig. 8A-C). WB analysis demonstrated significantly increased protein expression levels of apoptosis markers caspase-9, caspase-3 and Bax, and significantly decreased protein expression levels of Bcl-2 in CRC cells overexpressing LGALS4 compared with controls. The effects of LGALS4 overexpression on these markers were significantly enhanced by 5-FU treatment (Fig. 8D-F). Additionally, 2-NBDG uptake, ATP production and lactate levels were measured in LoVo and HCT-116 cells. These results showed that overexpression of LGALS4 significantly decreased ATP production, lactate levels and glucose uptake in CRC cells compared with the control (Fig. 8G-8K). These results suggest that LGALS4 overexpression may enhance 5-FU-induced apoptosis in CRC cells and disrupt glucose metabolism, further inhibiting cell viability.

Figure 8. LGALS4 overexpression enhances 5-FU-induced apoptosis and inhibits glycolysis in colorectal cancer cells. Flow cytometric analysis was performed to assess apoptosis in (A) LoVo and (B) HCT-116 cells following LGALS4 overexpression, with or without treatment with 50 µg/ml 5-FU. (C) Quantification of apoptosis in LoVo and HCT-116 cells following LGALS4 overexpression, with or without 50 µg/ml 5-FU treatment. (D) Western blot analysis was performed to detect the expression of apoptotic proteins (caspase-3, caspase-9, Bax and Bcl-2) in LoVo and HCT-116 cells following LGALS4 overexpression, with or without 50 µg/ml 5-FU treatment. Semi-quantification of the western blot analysis results for apoptotic proteins (caspase-3, caspase-9, Bax and Bcl-2) in (E) LoVo and (F) HCT-116 cells following LGALS4 overexpression, with or without 50 µg/ml 5-FU treatment. *P<0.05 vs. vector; #P<0.05 vs. vector + 5-FU. (G and H) Flow cytometry was used to measure the uptake of 2-NBDG in (G) LoVo and (H) HCT-116 cells following LGALS4 overexpression. (I) Quantification of 2-NBDG uptake in LoVo and HCT-116 cells transfected with Vector or over-LGALS4. (J) A lactate detection kit was used to measure lactate release from LoVo and HCT-116 cells following LGALS4 overexpression. (K) An ATP detection kit was used to measure ATP production in LoVo and HCT-116 cells following LGALS4 overexpression. *P<0.05 vs. vector. 5-FU, 5-Fluorouracil; LGALS4, lectin galactoside-binding soluble 4; over, overexpression; 2-NBDG, 2-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino]-2-deoxyglucose.

LGALS4 modulated β-catenin signaling to inhibit glycolysis in CRC cells

To ascertain the impact of LGALS4 overexpression on glycolysis-related proteins in CRC cells, conducted RT-qPCR analysis was performed on CRC cells that were transfected with either an LGALS4 overexpression vector or a control vector. These findings showed that overexpression of LGALS4 significantly downregulated the expression levels of key glycolysis-related factors, such as CTNNB1, MYC, solute carrier family 2 member 1 (SLC2A1), pyruvate kinase M1/2 (PKM), HK2 and LDHA compared with control cells (Fig. 9A and B). WB analysis showed similar results, with a significant decrease in the expression levels of these proteins following LGALS4 overexpression compared with the controls (Fig. 9C-E). HCT-116 and LoVo cells were treated with the β-catenin inhibitor XAV-939, in addition to inducing LGALS4 overexpression. These results demonstrated that inhibition of β-catenin significantly enhanced the downregulation of the glycolysis-related factors that were induced by LGALS4 overexpression (Fig. 9F-J). These findings suggest that LGALS4 overexpression may inhibit the expression of key glycolysis-related proteins in CRC cells and this effect is further potentiated by β-catenin inhibition. The Wnt/β-catenin signaling pathway serves a central role in cell proliferation, migration and invasion and LGALS4 potentially inhibits the malignant behavior of tumor cells by inhibiting this signaling pathway and reducing the expression of glycolysis-related genes.

Figure 9. LGALS4 affects β-catenin signaling to inhibit glycolysis. RT-qPCR analysis of glycolysis-related factors, including CTNNB1, MYC, PKM, SLC2A1, LDHA, and HK2, in (A) LoVo and (B) HCT-116 cells following LGALS4 overexpression. (C) WB analysis showing the protein expression of glycolysis-related factors in LoVo and HCT-116 cells following LGALS4 overexpression. Semi-quantification of the WB analysis results for glycolysis-related factors (β-catenin, c-myc, PKM2, GLUT1, LDHA and HK2) in (D) LoVo and (E) HCT-116 cells following LGALS4 overexpression. *P<0.05 vs. vector. RT-qPCR analysis of glycolysis-related factors in (F) LoVo and (G) HCT-116 cells treated with or without the β-catenin inhibitor XAV-939 following LGALS4 overexpression. (H) WB analysis showing the protein expression of glycolysis-related factors in LoVo and HCT-116 cells treated with or without the β-catenin inhibitor XAV-939 following LGALS4 overexpression. Semi-quantification of the WB analysis results for glycolysis-related factors in (I) LoVo and (J) HCT-116 cells treated with or without the β-catenin inhibitor XAV-939 following LGALS4 overexpression. *P<0.05 vs. vector; #P<0.05 vs. vector + XAV-939. RT-qPCR, reverse transcription-quantitative PCR; WB, western blot; LGALS4, lectin galactoside-binding soluble 4; CTNNB1, catenin beta 1; MYC, Myc proto-oncogene; PKM, pyruvate kinase M1/M2; SLC2A1, solute carrier family 2 member 1; CRC, colorectal cancer; PKM2, pyruvate kinase type M2; GLUT1, solute carrier family 2 member 1; LDHA, lactate dehydrogenase A; HK2, hexokinase 2; over, overexpression.

Discussion

CRC is a multifaceted disease characterized by genetic and molecular alterations that drive its progression and impact patient outcomes (33). To better understand the molecular underpinnings of CRC, the present study performed a bioinformatics analysis of DEGs using the TCGA-COAD and GSE26571 datasets. This analysis showed overlapping DEGs, primarily involved in processes such as DNA replication, cell cycle and ribosome function. Genes involved in DNA replication and cell cycle pathways may be associated with rapid proliferation and tumor development in CRC. CRC is closely associated with aberrant DNA replication, as evidenced by the prevalence of DNA polymerase e, catalytic subunit mutations such as p.S297F in colorectal and endometrial carcinomas (34). Furthermore, Zurlo et al (35) demonstrated that Cladosporol A induces G1-phase cell cycle arrest in CRC cells, particularly HT-29 cells, by upregulating the expression of p21(waf1/cip1), which reduces Cyclin levels and inhibits CDK activity. Zinc is a cofactor for many enzymes and transcription factors and is essential for cell proliferation and differentiation (36). Genes responsive to zinc ion levels may be involved in regulating cellular adaptation to changes in the microenvironment (37). Ribosomes are sites of protein synthesis and their biosynthesis is closely linked to the metabolic requirements of cell growth and tumor cells (38). Calcineurin is a cell adhesion molecule involved in cell-cell interactions and maintenance of tissue structure. Changes in the expression of its related genes may affect the invasiveness and metastatic ability of tumor cells (39). Another study by Zurlo et al (40) reported that Cladosporol A inhibits CRC proliferation by enhancing the expression of p21(waf1/cip1) through Sp1-peroxisome proliferator activated receptor γ interaction. This compound also induces β-catenin degradation, thereby suppressing the β-catenin/T-cell factor (TCF) pathway and promoting E-cadherin expression, which impedes cell cycle progression. Prognostic analysis of overlapping DEGs identified significant prognostic genes, including C8G, LGALS4 and RPS17. Expression analysis demonstrated significantly decreased expression levels of C8G and LGALS4 in tumor samples from the TCGA-COAD and GSE26571 datasets. Yu et al (41) identified RPS17 as a hub gene in the co-expression network of differentially expressed genes in CRC with microsatellite instability, suggesting its significant role in the ribosome pathway's involvement in CRC.

LGALS4 is a protein encoded by the LGALS4 gene in humans. It belongs to the galectin family and has the ability to bind and recognize β-galactoside sugars (42). LGALS4 is primarily expressed in the gastrointestinal tract, where it serves essential roles in cell-cell adhesion, epithelial differentiation and mucosal immunity (43). Its involvement in various physiological processes, including intestinal homeostasis, inflammation and cancer progression has previously been reported. Watanabe et al (44) reported that elevated levels of circulating Galectin-4 in patients with CRC correlate with disease progression, suggesting its potential as a follow-up marker post-surgery. Galectin-1 may be useful for patient screening and Galectin-4 can complement CEA/CA19-9 in enhancing CRC monitoring. Additionally, Zhou et al (45) demonstrated that surface profiles of CRC cells and tumor-infiltrating lymphocytes from surgical samples align with prognostic categories and minimal antigenic panels, including Galectin-4, providing potential predictors for disease relapse and patient survival. Furthermore, Satelli et al (7) reported that Galectin-4 acts as a potential tumor suppressor in CRC, with its downregulation observed in adenomas and invasive carcinomas. Overexpression induces cell cycle arrest, reduces migration and sensitizes cells to apoptosis, suggesting its significance in CRC biology through interaction with Wnt signaling proteins and downregulation of Wnt target genes. In the present study, overexpression of LGALS4 limited the capacity of CRC cells to migrate, proliferate and invade and disrupted CRC cell cycle distribution. Meanwhile, overexpression of LGALS4 promoted apoptosis in CRC cells. This suggests that LGALS4 could potentially serve as both a therapeutic target and a potential prognostic marker, underscoring its importance in cancer research and clinical applications.

Glucose is a primary energy source for cells and a simple sugar. It is vital for cellular metabolism and is transported into cells through glucose transporters. Once inside the cell, glucose undergoes glycolysis, a metabolic process known as the energy investment phase and the energy payback phase, which produces ATP and NADH by converting glucose to pyruvate (46). Glycolysis, which takes place in the cytoplasm, consists of 10 enzyme-catalyzed stages. ATP is consumed during the energy investment phase to phosphorylate glucose and its intermediates. The energy payback phase involves the production of NADH and ATP. Glycolysis is an anaerobic process, meaning it does not require oxygen and is essential for energy production under both aerobic and anaerobic conditions (47). This pathway is crucial for cells with high energy demands, such as muscle and cancer cells. Zhou et al (48) reported that Dioscin inhibits aerobic glycolysis in CRC cells by degrading S-phase kinase-associated protein 2 via Cadherin 1, consequently reducing CRC proliferation. Additionally, Zhou et al (49) showed that increased expression of PKM2 in CRC promotes aerobic glycolysis, cell proliferation and migration, suggesting PKM2 may act as a promising therapeutic target for CRC. Moreover, Zhao et al (50) reported that Sam68, an RNA-binding protein, promotes aerobic glycolysis in CRC by regulating the alternative splicing of PKM2, enhancing glycolysis and cell proliferation. This underscores the potential of Sam68 as a target for therapy in CRC. In the present study, overexpression of LGALS4 significantly enhanced the survival of CRC cells under low glucose conditions. Although this phenomenon appears contradictory to the effect of LGALS4 downregulating key factors of glycolysis, this may potentially be due to the reprogramming of metabolic pathways in CRC cells facilitated by LGALS4. LGALS4 may have maintained the metabolic demand and energy supply of cells under low-glucose conditions either by activating non-glycolytic energy-generating pathways or by enhancing the efficiency of cellular utilization of nutrients. In addition, LGALS4-overexpressing CRC cells exhibited relative resistance to glucose deprivation and glycolytic inhibition, which may be related to its effects on cell cycle and apoptotic pathways, making the cells more tolerant to metabolic stress. LGALS4 may enhance cellular adaptation during glycolytic inhibition by regulating signaling pathways related to cell survival, such as the β-catenin signaling pathway. The decreased expression level of LGALS4 observed in CRC tissues may be related to its role as a tumor suppressor. Downregulation of LGALS4 expression in a variety of cancers correlates with a better prognosis, whereas in the present study, the downregulation of LGALS4 may be related to metabolic reprogramming of the tumor cells to adapt to rapid proliferation and to evade immune surveillance. The present study showed that LGALS4-overexpressing CRC cells exhibited increased survival under glucose deprivation conditions and showed tolerance to glycolytic inhibition compared with negative controls. Flow cytometry results confirmed that LGALS4 significantly reduced apoptosis induced by glucose deprivation. These findings suggest that therapeutic pathways targeting aerobic glycolysis may provide new strategies for CRC treatment in the future.

5-FU is an anticancer drug used to treat various cancers, such as breast cancer, hepatocellular carcinoma (HCC) and CRC (51). A previous study by Zou et al (52) reported that 5-FU induces cytotoxicity and apoptosis in cancer cells through a ROS-mediated mitochondrial pathway and allicin enhances the antitumor activity of 5-FU by increasing apoptosis and reducing mitochondrial membrane potential in HCC cells. Similarly, Zou et al (53) showed that insulin pretreatment enhances the anticancer effects of 5-FU in esophageal and colonic cancer cells by increasing 5-FU uptake, promoting apoptosis and upregulating the expression of cleaved caspase-3, thereby inhibiting cell proliferation more effectively. Additionally, Zuo et al (54) showed that 5-FU inhibits cell proliferation and induces apoptosis in HepG2 liver cancer cells, similar to the impacts of chikusetsusaponin IV and V, which also promote cell cycle arrest and enhance apoptotic protein activation, underscoring their potential in cancer treatment. The present study demonstrated that overexpression of LGALS4 promoted apoptosis and inhibited aerobic glycolysis in CRC cells. Based on this finding, it was hypothesized that 5-FU may have a synergistic effect with LGALS4 overexpression by further enhancing the inhibitory effect on tumor cells. Specifically, LGALS4 overexpression in CRC cells enhances the apoptotic response to 5-FU treatment and induces metabolic changes, reducing glycolytic activity and energy production. This may reduce the metabolism and elimination of 5-FU, thereby accumulating 5-FU in tumor cells and enhancing its anticancer effect. 2-NBDG, a glucose analog, is used to measure glucose uptake (55). Glycolysis, a metabolic process, transforms glucose into energy and lactate (56). Lactate release is a product of glycolysis and ATP, the main energy source within the cell, is produced through glycolysis (57). Therefore, inhibition of glycolysis by 5-FU will affect ATP production, leading to energy deficiency, while reducing lactate release and 2-NBDG uptake (58). Zhou et al (59) reported that Gefitinib causes A549 and H1975 non-small cell lung cancer cells to undergo programmed cell death and inhibit glycolysis, as evidenced by reduced glucose uptake, lowered ATP levels and increased apoptosis rates. Additionally, Zu et al (60) showed that by linking glucose metabolism with lipid synthesis, ATP citrate lyase (ACL) catalyzes the conversion of citrate to acetyl-CoA and oxaloacetate by utilizing ATP and CoA. ACL, overexpressed in various types of cancer, including CRC, serves as a possible target for therapeutic cancer by disrupting glycolysis-driven lipogenesis. Moreover, Zuo et al (61) reported that miR-4443 downregulates TRIM14, suppressing energy metabolism and metastasis in papillary thyroid carcinoma (PTC). miR-4443 inhibits ATP production and aerobic glycolysis by targeting TRIM14, indicating its role in PTC progression and energy regulation. In the present study, this was evidenced by increased apoptotic markers, decreased 2-NBDG uptake and decreased lactate and ATP levels. These results suggested that LGALS4 may exert anti-CRC effects by inhibiting glycolysis.

The β-catenin signaling pathway is essential for a number of cellular functions, including cell proliferation, differentiation and migration. β-catenin is phosphorylated by a destruction complex when Wnt ligands are not present, leading to its degradation (62). Wnt activation inhibits the destruction complex, enabling β-catenin to accumulate and move into the nucleus (63). There, it interacts with TCF/lymphoid enhancer-binding factor (LEF) transcription factors to activate target gene expression, such as Cyclin D1 and c-Myc, promoting cell growth and survival. Dysregulation of this pathway is associated with various diseases, including cancer, making it an important target for therapeutic intervention (64). Zou et al (65) reported that elevated circular (circ) RNA circ_0068464 levels in CRC contribute to cell migration, proliferation and activation of the Wnt/β-catenin signaling pathway. Its interaction with miR-383 further exacerbates CRC progression, suggesting therapeutic potential in targeting this pathway. Similarly, a study by Zou et al (66) showed that increased circCASK in CRC promotes tumor growth and invasion by upregulating six homeobox 1 expression. Forkhead box c2 transcriptionally induces circCASK expression, thereby activating the Wnt/β-catenin signaling pathway and accelerating the development of CRC. In addition, there is a close connection between β-catenin and glycolysis. Zhou et al (67) reported that Dihydrolipoamide S-acetyltransferase (DLAT), a glycolysis-related gene, is overexpressed in HCC, contributing to poor prognosis. Its downregulation inhibits PI3K/Akt and Wnt/β-catenin signaling pathways, highlighting the role of DLAT as a potential therapeutic target in HCC. Additionally, Zhou et al (68) reported that cryptotanshinone suppresses breast cancer cell migration, invasion and proliferation by targeting glycolysis-related proteins, such as PKM2. This suggests a potential connection between β-catenin and glycolysis in breast cancer, suggesting PKM2 may be a promising therapeutic avenue. The interaction between LGALS4 and the β-catenin signaling pathway may impact the metabolic properties of CRC cells. Overexpression of LGALS4 may interfere with the intranuclear accumulation of β-catenin or its interaction with TCF/LEF and inhibit the transcription of genes downstream of the Wnt signaling pathway such as c-Myc, PKM2 and GLUT1. In addition, LGALS4 may regulate the expression of the metabolic enzymes LDHA and HK and activate the AMPK signaling pathway by altering intracellular ATP levels or AMP/ATP ratios, thereby inhibiting metabolic enzyme expression. The effect of LGALS4 on the expression or function of the glucose transporter protein GLUT1 may reduce glucose uptake, thereby affecting glucose-dependent metabolic pathways. As a tumor suppressor protein, overexpression of LGALS4 inhibits tumor cell growth and metabolism through a variety of mechanisms, including downregulation of genes closely related to tumor metabolism. LGALS4-induced cell cycle arrest and promotion of apoptosis may also indirectly affect the expression of metabolism-related genes as these cellular processes are closely related to the metabolic status of cells. The RT-qPCR and WB analyses in the present study showed that LGALS4 overexpression significantly reduced the levels of glycolysis-related proteins in CRC cells. When CRC cells were subjected to 10 µM β-catenin inhibitor XAV-939 for 72 h, the expression levels of glycolysis-related proteins further decreased. This indicated that LGALS4 may affect β-catenin signaling to inhibit glycolysis, thereby potentially inhibiting CRC development.

The present study investigated the role of LGALS4 in CRC glycolysis. Although LGALS4 has been identified as a potential prognostic factor for patients with CRC, its specific impact in tumor glycolysis has not been fully elucidated. Through analysis of the TCGA-COAD and GSE26571 databases, it was demonstrated that LGALS4 expression was significantly downregulated in CRC tissues and strongly correlated with patient survival, suggesting a potentially important role in CRC development. Glycolysis is a major energy source for cancer cells and its aberrant activation is tightly linked to rapid proliferation and invasiveness of tumors. It was hypothesized that LGALS4 may affect the metabolic properties of CRC cells by regulating the glycolytic pathway. The present preliminary data suggested that LGALS4 overexpression inhibited glycolytic activity in CRC cells and that this effect may be associated with changes in the β-catenin signaling pathway. Given the role of β-catenin in the regulation of glycolytic gene expression, it could be suggested that LGALS4 may regulate glycolysis through this signaling pathway. Therefore, the function of LGALS4 in CRC glycolysis and its associated molecular mechanism were analyzed.

In summary, overexpression of LGALS4 exerted a multifaceted inhibitory effect in CRC cells, significantly affecting key biological properties of tumor cells. Firstly, it inhibited the glycolytic process, reducing the cell's dependence on glucose and decreasing lactate production and ATP generation, thereby directly limiting the cell's energy supply. Second, LGALS4 caused cell cycle arrest in the G1 phase, which prevented normal cell cycle progression by decreasing the expression levels of cell cycle-related proteins such as CDK1, Cyclin B1 and Cyclin A2. In addition, LGALS4 overexpression promoted apoptosis, which was closely related to the changes in the expression of apoptosis-related proteins such as Bax, Bcl-2, caspase-3 and caspase-9, increasing the rate of apoptosis. Meanwhile, LGALS4 affected the β-catenin protein signaling pathway, decreasing the expression of glycolysis-related factors such as β-catenin protein, c-Myc, GLUT1, PKM2, HK2 and LDHA. A previous study has shown that the β-catenin protein signaling pathway is a signaling pathway that serves a central role in cell proliferation, migration and invasion (69). LGALS4 may also cause metabolic reprogramming, which further reduced aerobic glycolysis in CRC cells and inhibited the rapid proliferation and invasiveness of tumor cells. Notably, LGALS4 overexpressing cells showed increased sensitivity to the chemotherapeutic drug 5-FU, which may be achieved by enhancing 5-FU-induced apoptosis. LGALS4 reduced 2-NBDG uptake and decreased ATP production and lactate release, suggesting a potential direct effect on cellular energy metabolism.

The present study suggested that LGALS4 may provide a new potential target for CRC therapy. As a protein whose expression is downregulated in CRC and is associated with patient survival, LGALS4 has the ability to regulate glycolysis and promote apoptosis in tumor cells, which provides a scientific basis for the development of new therapeutic approaches. By inhibiting glycolysis, LGALS4 is able to reduce the energy supply of tumor cells, directly targeting their metabolic needs for rapid proliferation. In addition, LGALS4 overexpression promoted apoptosis, providing a new therapeutic avenue for inducing tumor cell death. More importantly, LGALS4 increased the sensitivity of tumor cells to chemotherapeutic agents such as 5-FU, which may help to improve the efficacy of existing therapeutic regimens, particularly in drug-resistant tumors. Meanwhile, the regulatory effect of LGALS4 on the β-linker protein signaling pathway provided a new perspective on the control of tumor cell proliferation and invasion. These properties not only demonstrate the potential of LGALS4 in individualized medicine, but also highlight its advantages in overcoming existing therapeutic limitations. With further research, LGALS4 may be a key factor in improving treatment outcomes for patients with CRC.

Although the present study demonstrated the potential role of LGALS4 in CRC in in vitro experiments and bioinformatics analyses, there were a number of limitations. First, the findings need to be further validated by in vivo models to ensure the accuracy and reliability of the biological effects. Second, the sample size and population representation may limit the generalizability of the findings. In addition, the long-term effects and specific molecular mechanisms of LGALS4 need to be explored in depth. To address these limitations, future studies should conduct in vivo experiments in animal models to assess the actual therapeutic potential and safety of LGALS4. The mechanism of LGALS4 downregulation of the expression of factors such as β-catenin, c-Myc, PKM2, GLUT1, LDHA and HK is currently unknown. Techniques such as chromatin immunoprecipitation sequencing and RNA sequencing should be used to investigate how LGALS4 affects the transcriptional activity of the β-catenin signaling pathway and its associated target genes. Through methods such as immunoprecipitation and mass spectrometry, the direct or indirect interactions between LGALS4 and proteins such as β-catenin and c-Myc can be explored in addition to how these interactions affect their functions. In addition, metabolomics approaches should be used to analyze the metabolic changes in LGALS4 overexpressing cells to understand how LGALS4 regulates cellular metabolic pathways, particularly glycolytic processes. By observing the effects of LGALS4 overexpression on the cell cycle and apoptosis, how these cellular events are linked to the regulation of the expression of metabolism-related genes could be determined. In the present study, apoptosis and cell cycle distribution of CRC cells were quantified using PI staining combined with flow cytometry. Although this method provided valuable information on cell cycle status and apoptosis rates, it also has its inherent limitations. For example, PI staining may not be able to distinguish between the various stages of the cell cycle and there may be some bias in the assessment of cell survival status. In addition, due to the limitations of the present study conditions, 5-bromo-2-deoxyuridine (BRDU) staining was not performed to further validate cell proliferation. However, BRDU staining should be performed in future studies to complement the results of PI staining and provide additional validation of the findings of the present study. By combining these a more comprehensive understanding of the biology of CRC cells could be expected and these results could potentially provide a more solid experimental basis for future studies.

The present study highlighted the role of LGALS4 in CRC and its potential as a therapeutic target. Bioinformatics analysis demonstrated that LGALS4 was significantly downregulated in CRC and was associated with patient survival. Overexpression of LGALS4 resulted in a significant upregulation of caspase-3 and caspase-9, a phenomenon that may be achieved through multiple mechanisms. First, LGALS4 overexpression promoted apoptosis in CRC cells. Caspase-3 and caspase-9 act as key executors in the apoptotic pathway and caspase-9 acts as an initiating caspase to activate effector caspase-3, thereby triggering apoptosis. Second, LGALS4 overexpression inhibited aerobic glycolysis in CRC cells and affected the β-catenin signaling pathway, which serves a crucial role in cell survival and apoptosis. By decreasing the activity of the β-catenin signaling pathway, LGALS4 may promote the upregulation of caspase-3 and caspase-9, which in turn drives the apoptotic process. In addition, G1-phase cell cycle arrest caused by LGALS4 overexpression may have triggered a cellular stress response that activated apoptotic pathways including caspase family proteins. LGALS4 may also directly or indirectly regulate the expression of caspase-3 and caspase-9, which acted as post-transcriptional modifiers affecting the stability or translational efficiency of specific genes. Furthermore, possible intracellular feedback mechanisms may upregulate the expression of apoptosis-related genes upon detection of survival stress or abnormal signals to remove damaged cells. Taken together, the upregulation of caspase-3 and caspase-9 levels in CRC cells by LGALS4 overexpression may be due to its direct effect on apoptotic pathways and its potential inhibitory effect on the β-catenin signaling pathway, which exerts an antitumor effect in CRC.

The present study demonstrated the important role of LGALS4 in CRC and its value as a potential future therapeutic target. These findings showed that LGALS4 expression was downregulated in CRC tissues and correlated with poor patient prognosis, suggesting its role as a tumor suppressor. Functionally, overexpression of LGALS4 inhibited glycolysis and inhibited cell cycle progression in CRC cells, leading to G1 phase arrest while promoting apoptosis. In addition, the regulatory effect of LGALS4 on the β-catenin signaling pathway may have an inhibitory effect on the proliferation, migration and invasion of tumor cells. Notably, LGALS4 overexpressed cells increased higher sensitivity to the commonly used chemotherapeutic drug 5-FU, which potentially provides a novel research avenue to improve the efficacy of chemotherapy. The present study also highlighted the impact of LGALS4 in the metabolic reprogramming of tumors, indicating potential new perspectives for metabolically targeted therapies. These results further the current understanding of the molecular mechanisms of CRC and provide directions for future individualized treatment strategies and the development of new drugs, which may improve the treatment outcome of patients with CRC.

Supplementary Material

Supporting Data

Supporting Data

Acknowledgments

Not applicable.

Funding

Funding: No funding was received.

Availability of data and materials

The data generated in the present study may be requested from the corresponding author.

Authors' contributions

SL, KY and JY were responsible for conception and design of the study. CX, KY, ZQ and YC were responsible for data acquisition. SL, LY and BS were responsible for data analysis and interpretation. SL, TZ, JX and YC were responsible for statistical analysis. SL and BS drafted the manuscript. SL and KY confirm the authenticity of all the raw data. All authors read and approved the final version of the manuscript.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References