A TXNIP-driven bioluminescent reporter for high-throughput discovery of glycolytic inhibitors against renal cell carcinoma
Yajie Jing, Wanlu Liu, Hancheng Qin, Zhihong Chen
TL;DR
Researchers developed a bioluminescent tool to screen for drugs that inhibit glycolysis in kidney cancer cells.
Contribution
A novel TXNIP-driven bioluminescent reporter system for high-throughput discovery of glycolytic inhibitors in renal cell carcinoma.
Findings
TXNIP and MLXIP are upregulated by 2-DG in A498 RCC cells.
A TXNIP promoter-driven luciferase system detects 2-DG-like activity through bioluminescence.
The system offers a functional readout for identifying glycolysis-targeting anti-RCC drugs.
Abstract
The glycolytic inhibitor 2-deoxy-D-glucose (2-DG) has demonstrated consistent preclinical antitumor efficacy; however, the discovery of novel 2-DG-like agents for renal cell carcinoma (RCC) remains challenging due to the lack of specific, high-throughput screening (HTS) tools. In this study, RNA-seq analysis identified Thioredoxin-interacting protein (TXNIP) as a gene markedly upregulated by 2-DG in A498 RCC cells. We further confirmed that 2-DG transcriptionally upregulates the expression of both TXNIP and its transcription factor, MLX-interacting protein (MLXIP). Leveraging this mechanism, we engineered a bioluminescent reporter system by constructing a TXNIP promoter-driven luciferase construct (TXNIP-Pro-Luc2) and generating a stable A498-TXNIP-Pro-Luc2 cell line. In this system, 2-DG and its functional analogs activate the TXNIP promoter, resulting in a concentration-dependent…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 5- —The National Natural Science Foundation of China
- —The Doctoral Scientific Research Start-Up Foundation of Suzhou University
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsRedox biology and oxidative stress · bioluminescence and chemiluminescence research · Cancer Research and Treatments
Introduction
Renal cell carcinoma (RCC), accounting for about 3% of all adult malignancies, is among the most frequent cancers in the urinary system. Its most prevalent pathological subtype is clear cell renal cell carcinoma (ccRCC), comprising 70%**–**85% of cases [1]. While the incidence of primary RCC continues to rise, therapeutic options remain constrained. Therefore, developing novel targeted anti-RCC therapies is crucial for improving prevention and treatment outcomes [2, 3].
Energy metabolism reprogramming is a key feature of tumor biology, with glucose metabolism alterations being particularly notable. Unlike normal cells that maintain glucose homeostasis, tumor cells undergo metabolic reprogramming to adapt to their microenvironment. A hallmark of this reprogramming is their markedly increased glucose uptake, which fuels glycolysis and supports rapid proliferation [4]. Consequently, inhibiting glycolytic pathways can suppress tumor proliferation and induce cell death, establishing this approach as a promising therapeutic strategy [5, 6].
At present, glycolytic inhibitors show promising preclinical efficacy across diverse tumor types, with candidates like 2-deoxy-D-glucose (2-DG) demonstrating significant tumor growth suppression in animal models [7]. As a glucose analog, 2-DG competitively inhibits the activity of hexokinase 2 (HK2). Upon internalization, 2-DG is phosphorylated by HK2 to form 2-deoxyglucose-6-phosphate (2-DG-6P), which cannot undergo further metabolism, thereby disrupting glucose metabolic pathways. Disrupting glucose metabolism inhibits tumor cell growth and induces apoptosis [8–10]. Recent studies have highlighted the promising anti-tumor properties of various 2-DG derivatives, including 2-deoxy-2-fluoro-D-glucose (2-FG) and 2-deoxy-D-glucose-d (2-DG-d). These compounds, as potential targeted glycolytic inhibitors, exhibit significant clinical potential [11]. However, the therapeutic efficacy of these modified 2-DG derivatives requires validation through both in vitro and in vivo experiments. Therefore, developing efficient high-throughput screening tools is essential to expedite this drug development process. However, no such pathway-specific HTS tool for 2-DG derivatives in RCC has been reported to date, significantly hampering the advancement of this therapeutic class.
To sustain their rapid growth, proliferation, and metastasis, tumor cells must highly express glucose transporters (GLUTs) to meet their demand for glucose uptake and metabolism [12–14]. Thioredoxin-interacting protein (TXNIP), a member of the thioredoxin (TRX) superfamily, is a key regulator involved in glucose metabolism. It can inhibit glucose metabolic reprogramming in tumor cells by regulating GLUT1 to control cellular glucose uptake, thereby exerting its tumor-suppressive effects [15, 16]. In this study, we employed molecular cloning techniques to create a fusion between the promoter sequence of the TXNIP gene and the luciferase gene, thereby constructing a bioluminescent reporter system that quantitatively reflects the transcriptional activity of the TXNIP promoter. We assessed the potential of this reporter system to monitor the pharmacological effects of 2-DG in real-time, with the objective of establishing an effective tool for the screening of this class of anti-renal tumor therapeutics.
Materials and methods
Bioinformatics analysis methods
TXNIP mRNA expression levels in Kidney Renal Clear Cell Carcinoma (KIRC) and paired normal tissues were determined via the UALCAN database (http://ualcan.path.uab.edu/). Co-expression genes significantly associated with TXNIP (Pearson correlation coefficient |R| > 0.3 and P-value < 0.05) in KIRC were identified using the LinkedOmics database (http://linkedomics.org/). These genes were subsequently subjected to Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis using the DAVID bioinformatics resource (http://david.abcc.ncifcrf.gov/). The enrichment results were visualized as bar plots using the bioinformatics online platform (http://www.bioinformatics.com.cn).
Cell lines, chemicals, plasmids, and antibodies
The human renal clear cell carcinoma cell line A498 (Cat. No. CL-0254, RRID: CVCL_1056) was purchased from Procell Life Science & Technology Co., Ltd. (Wuhan, China). The cell line was authenticated by short tandem repeat (STR) profiling and maintained according to the supplier’s recommendations. Plasmids and primers were sourced as follows: the reporter plasmid pGL4.19-TXNIP-Pro-Luc2 (Cat. No. 20220805) was obtained from the Public Protein/Plasmid Library (Nanjing, China); the control plasmid pEZ-M39 (Cat. No. EX-NEG-M39), along with expression plasmids for MLXIP (Cat. No. EX-A1755-M39) and TXNIP (Cat. No. EX-M0226-M39), were purchased from GeneCopoeia (Guangzhou, China). Gene-specific primers for human TXNIP (HSQRP013766) and ACTB (HQP108762) were also acquired from the same supplier. The primary antibodies used included: rabbit polyclonal anti-MLXIP antibody (Cat. No. 13614-1-AP, RRID: AB_2282153, Proteintech, Wuhan, China), rabbit monoclonal anti-TXNIP antibody (Cat. No. A11682, RRID: AB_2935845, Nature Biosciences, Hangzhou, China), and rabbit monoclonal anti-Vinculin antibody (Cat. No. A2752, RRID: AB_2863020, Abclonal, Wuhan, China). The secondary antibody, HRP-conjugated Goat Anti-Rabbit IgG (Cat. No. AS014, RRID: AB_2769854, Abclonal, Wuhan, China), was used for immunoblotting detection. Glycolysis inhibitors were procured from the following sources: 2-Deoxy-D-glucose (Cat. No. SJ-MN0028, SparkJade, Jinan, China), 2-Deoxy-2-fluoro-D-glucose (Cat. No. HY-141637, MedChemExpress, NJ, USA), and 2-Deoxy-D-glucose-d (Cat. No. HY-13966 S, MedChemExpress, NJ, USA).
Cell culture, cell transfection, stable cell line establishment and selection
A498 cells were cultured in Minimum Essential Medium (Cat. No. PM150410, Procell, Wuhan, China) supplemented with 10% fetal bovine serum (FBS) (Cat. No. E600001-500, Sangon Biotech, Shanghai, China) at 37 °C in a humidified atmosphere with 5% CO₂. For transfection assays, A498 cells were seeded into 6-well plates at a density of 2 × 10⁵ cells per well and incubated overnight to allow complete adherence. Then, plasmid transfection was performed using HighGene Transfection Reagent (Cat. No. RM09014P, ABclonal, Wuhan, China) in strict accordance with the manufacturer’s instructions. To establish stable A498-TXNIP-Pro-Luc2 and A498-Luc2 cell lines, the corresponding plasmids were separately transfected into A498 cells. At 24 h post-transfection, the culture medium was discarded, and the cells were rinsed twice with pre-warmed (37 °C) phosphate-buffered saline (PBS). Subsequently, the cells were cultured in fresh MEM containing 10% FBS and 800 µg/mL G418 (Cat. No. G8160, Solarbio, Beijing, China) for positive selection. The G418-containing selection medium was refreshed every 2–3 days until resistant cell colonies were clearly formed and visible. Individual resistant colonies were digested with 0.25% trypsin (Cat. No. T1320, Solarbio, Beijing, China), and the obtained cells were transferred to new culture dishes for further propagation and expansion.
Western blotting
Total proteins were extracted from cultured cells using Western & IP Lysis Buffer (Cat. No. P0013, Beyotime, Shanghai, China). The protein concentration was quantified by the bicinchoninic acid (BCA) assay kit (Cat. No. PC0020, Solarbio, Beijing, China). Equal amounts of protein samples were loaded per lane for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) separation, followed by electrotransfer onto polyvinylidene fluoride (PVDF) membranes, which were pre-activated with methanol prior to electrotransfer. After electrotransfer, the membranes were blocked with 5% non-fat milk in Tris-buffered saline with Tween-20 (TBST) for 1 h at room temperature, and then incubated with primary antibodies against TXNIP, MLXIP, or Vinculin (at a dilution of 1:1,000) overnight at 4 °C. Subsequently, the membranes were washed three times with TBST and incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (at a dilution of 1:10,000) for 1 h at room temperature. Specific protein bands were visualized using an enhanced chemiluminescence (ECL) detection kit (Cat. No. RM02867, ABclonal, Wuhan, China) in accordance with the manufacturer’s instructions. Vinculin was used as the internal reference protein.
RNA extraction, RT-qPCR, and RNA-seq analysis
Total RNA was extracted from cells using the Zymo RNA Isolation Kit (Cat. No. TR154-50, Zymo Research, USA) according to the manufacturer‘s instructions. RNA concentration and purity were assessed by measuring the A260/A280 ratio using a spectrophotometer. For cDNA synthesis, 1 µg of total RNA was reverse transcribed using the All-in-One First-Strand cDNA Synthesis Kit (Cat. No. QP056, GeneCopoeia, China). Quantitative real-time PCR (RT-qPCR) was performed on a LightCycler 96 system (Roche, Germany) in a total reaction volume of 20 µL, consisting of 4 µL BlazeTaq™ SYBR Green qPCR Mix (Cat. No. QP031, GeneCopoeia, China), 2 µL mixed primers (2 µM), 2 µL cDNA template (1:5 diluted), and 12 µL nuclease-free water. The thermal cycling conditions were as follows: initial denaturation at 95 °C for 5 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 1 min. Gene expression levels were normalized to ACTB and calculated using the 2 − ΔΔCT method. For transcriptome analysis, cells were treated with either DMSO (control) or 5 mM 2-DG for 48 h (n = 3 per group). Total RNA from these samples was then submitted to Sangon Biotech (Shanghai, China) for library construction and subsequent RNA-seq analysis.
Bioluminescent imaging and luciferase assay
D-Luciferin (Cat. No. 122799 A, Revvity, Waltham, MA, USA) was added to the complete tissue culture medium to reach a final concentration of 150 µg/mL, and the cells were incubated at 37 °C in a humidified atmosphere with 5% CO₂ for 5 min in the dark. Bioluminescent photon emission was quantified using an IVIS Lumina LT Series III system (PerkinElmer, Waltham, USA), and the luminescent signal was analyzed with the dedicated software of the instrument. For the luciferase activity detection in cell lysates, the experiment was strictly performed in accordance with the manufacturer’s instructions of the Promega Luciferase Assay System (Cat. No. E1910, Promega, Madison, WI, USA). The luminescent intensity of the samples was measured on a lumiPro luminometer (Yuanpinghao, Beijing, China). The total protein concentration of cell lysates was determined for the normalization of luciferase activity results, and the final luciferase activity was expressed as relative light units (RLU) normalized to total protein content.
CCK-8, glucose uptake, and lactate production assay
A CCK-8 assay was performed using a kit (Cat. No. E606335, Sangon Biotech, Shanghai, China). A498 cells were seeded into 96-well culture plates at a density of approximately 1 × 10⁴ cells per well. Following gentle horizontal swirling (3–5 times) for even cell distribution, the plates were incubated overnight at 37 °C under 5% CO₂. Subsequently, cells underwent two PBS (37 °C) washes and were treated with drug-containing medium at various concentrations. After incubation, 10 µL of CCK-8 reagent was added to each well, followed by a further 2 h incubation. Absorbance was then measured at 450 nm using a microplate reader. Glucose uptake and lactate production were quantified using the Glucose Uptake Assay Kit (Cat. No. A154-1-1, Nanjing Jiancheng Bioengineering Institute, Nanjing, China) and Lactate Assay Kit (Cat. No. A019-2-1, Nanjing Jiancheng Bioengineering Institute, Nanjing, China) according to the respective manufacturers’ protocols.
Statistical analysis
GraphPad Prism 10.4.2 was used for statistical analysis. Data were expressed as mean value ± standard error (SEM). Depending on the experimental design, either one-way ANOVA with Dunnett’s multiple comparisons test or two-way ANOVA with Dunnett’s multiple comparisons test was applied, with statistical significance set at p < 0.05. All experiments were independently replicated three times to ensure reproducibility.
Results
2-DG and its derivatives inhibit glycolysis and cell proliferation of RCC cells
We first assessed the effects of 2-DG (0, 5, 10, and 15 mM), 2-FG (0, 1, 2, 5, and 10 mM), and 2-DG-d (0, 1, 2, 5, and 10 mM) on A498 renal cell carcinoma (RCC) cells over four days. Cells in the 0 mM groups, which received an equivalent volume of DMSO vehicle, served as the controls and exhibited nearly linear growth during this period. In contrast, treatment with increasing concentrations of each compound significantly and concentration-dependently inhibited A498 cell proliferation (Figs. 1A–C). To investigate the potential metabolic basis for this anti-proliferative effect, we measured glucose consumption and lactate production in drug-treated A498 cells. Both glucose uptake and lactate output decreased in a concentration-dependent manner with all three compounds (Figs. 1D–I). These results demonstrate that 2-DG and its derivatives effectively suppress glycolysis and inhibit the proliferation of A498 RCC cells, suggesting that glycolytic inhibition contributes to their anti-proliferative activity.
Fig. 12-DG and its derivatives inhibit glycolysis and cell proliferation of A498 cells. (A–C) Proliferation of A498 cells measured by CCK-8 assay after treatment with 2-DG (A), 2-FG (B), and 2-DG-d (C). For 2-DG treatment, the concentration gradient was set as 0 mM (vehicle control, containing equal volume of DMSO), 5 mM, 10 mM, and 15 mM; while for 2-FG and 2-DG-d treatments, the concentration gradients were consistent: 0 mM (vehicle control), 1 mM, 2 mM, 5 mM, and 10 mM. Cell viability was assessed by CCK-8 assay at the indicated time points over a 4-day period. (D-F) Glucose uptake capacity of A498 cells after 2-DG (D), 2-FG (E), and 2-DG-d (F) treatment, measured using a glucose uptake assay kit. The concentration gradients of each compound were identical to those described in the aforementioned section for cell proliferation detection. (G-I) Lactate production levels in A498 cells following treatment with gradient concentrations of 2-DG (G), 2-FG (H), and 2-DG-d (I), detected by a lactate detection kit. The concentration gradients for each compound were consistent with those used in cell proliferation detection. Data are mean ± SEM (n = 3); data in Figs. 1A-C were analyzed by two-way ANOVA with Dunnett’s multiple comparisons test; data in Figs. 1D-I were analyzed by one-way ANOVA with Dunnett’s multiple comparisons; *p < 0.05, *** p < 0.001, ****p < 0.0001
Identification of TXNIP as a 2-DG-sensitive effector gene and its functional validation in A498 cells
To identify effector genes sensitive to 2-DG treatment, we performed RNA-seq on A498 cells exposed to 5 mM 2-DG for 48 h. Using thresholds of p < 0.05 and |log2FC| ≥ 1, we identified 1,288 upregulated and 1,827 downregulated genes. Transcripts Per Kilobase of exon model per Million mapped reads (TPM) analysis showed the top three upregulated genes as INO80B-WBP1 (mean TPM: 0.00 to 21.44), KBTBD11-OT1 (0.09 to 27.76), and TXNIP (1.25 to 354.5). Notably, TXNIP exhibited the most dramatic upregulation (Figs. 2A, B). Subsequent RT-qPCR and Western blotting results demonstrated that 2-DG treatment upregulated TXNIP and MLXIP expression in A498 cells, confirming its role as a 2-DG-responsive effector gene (Figs. 2C, D). Previous studies indicate that MLXIP acts as a transcription factor for TXNIP [17–20]. To confirm this role in A498 cells, we assessed the effect of MLXIP overexpression on TXNIP expression. RT-qPCR results showed that MLXIP overexpression upregulated TXNIP expression in A498 cells, indicating that MLXIP is a transcription factor for TXNIP in these cells (Fig. 2E). Taken together, these findings identify TXNIP as a key effector gene responsive to 2-DG treatment in A498 cells. Furthermore, the concurrent upregulation of MLXIP and the established role of MLXIP as a TXNIP transcription factor strongly suggest that 2-DG induces TXNIP expression, at least in part, through an MLXIP-dependent transcriptional mechanism.
Fig. 22-DG upregulates MLXIP and TXNIP expression in A498 cells. (A) Volcano plot of differentially expressed genes (DEGs). (B) Top 10 upregulated genes from RNA-seq analysis. (C) TXNIP mRNA detected by RT-qPCR in A498 cells after 48 h of treatment with 2-DG (0, 5, 10, and 15 mM). (D) Protein expression of TXNIP and MLXIP was detected by Western blotting in A498 cells treated with 2-DG at 0, 5, 10, and 15 mM for 48 h. The 0 mM group served as the vehicle control and contained an equal volume of DMSO. (E) MLXIP and TXNIP mRNA expression analyzed by RT-qPCR in A498 cells 48 h after transfection with MLXIP plasmid (0, 2, and 4 µg), where 0 µg MLXIP plasmid corresponds to the empty vector plasmid used as control. Data are mean ± SEM (n = 3); Statistical significance was analyzed by one-way ANOVA with Dunnett’s multiple comparisons; **** p < 0.0001, *** p < 0.001
TXNIP inhibits Glycolysis in RCC cells
To investigate the functional role of TXNIP, we first analyzed its co-expressed genes in TCGA-KIRC data via LinkedOmics. Genes with significant correlations (Pearson |R| ≥ 0.3, p < 0.05) included 823 positively and 579 negatively correlated genes. Functional enrichment analysis using DAVID revealed these genes were enriched in glucose metabolism regulation pathways (Figs. 3A, B). We further investigated its clinical relevance and function. UALCAN analysis indicated that TXNIP expression is downregulated in KIRC and decreases with higher tumor grade (Figs. 3C, D), suggesting a potential tumor-suppressive role. To test this, we overexpressed TXNIP in A498 cells and confirmed that it inhibits key glycolytic metrics, namely glucose uptake and lactate production (Figs. 3E-G). These results collectively demonstrate that TXNIP acts as a glycolysis inhibitor in RCC, and its loss in tumors may promote metabolic adaptation.
Fig. 3TXNIP inhibits glycolysis in A498 cells. (A) Volcano plot of TXNIP expression-related genes in KIRC generated using LinkedOmics analysis. (B) GO enrichment and KEGG pathway analysis of TXNIP expression-related genes in KIRC performed using DAVID. (C) Comparison of TXNIP expression levels between KIRC tumors and adjacent normal tissues, and (D) analysis of TXNIP expression across different tumor grades, both using UALCAN. (E-G) Glucose uptake (E) and lactate production (F) were measured in A498 cells transfected with increasing amounts (0, 0.5, 1, and 2 µg) of TXNIP plasmid, where 0 µg TXNIP plasmid corresponds to the empty vector plasmid used as control. TXNIP mRNA analyzed by RT-qPCR in A498 cells 48 h after transfection with TXNIP plasmid (G). Data are mean ± SEM (n = 3); Statistical significance was analyzed by one-way ANOVA with Dunnett’s multiple comparisons; **** p < 0.0001, *** p < 0.001, ** p < 0.01, * p < 0.05
Establishment of the A498-TXNIP-Pro-Luc2 cell line
The recombinant luciferase reporter plasmid pGL4.19-TXNIP-Pro-Luc2 was constructed by cloning the TXNIP promoter sequence between the KpnI and XhoI restriction sites of the pGL4.19-Luc2 plasmid (Fig. 4A). Subsequently, A498 cells were transfected with either pGL4.19-TXNIP-Pro-Luc2 or pcDNA3.1-Luc2 plasmids. 48 h post-transfection, the medium was replaced with G418-supplemented MEM (800 µg/mL) for selection, yielding stable cell lines designated A498-TXNIP-Pro-Luc2 and A498-Luc2 (control). Luciferase activity assays confirmed significantly higher luminescence in A498-TXNIP-Pro-Luc2 cells compared to control cells (Fig. 4B). Bioluminescence imaging demonstrated significantly higher luminescent intensity in A498-TXNIP-Pro-Luc2 cells compared to control A498 cells (Fig. 4C). Further, to validate reporter specificity, we tested the effect of MLXIP (a TXNIP-upregulating transcription factor) on luciferase expression. In A498-TXNIP-Pro-Luc2 cells, MLXIP overexpression dose-dependently enhanced both bioluminescence signals and luciferase activity, whereas no significant effect was observed in A498-Luc2 controls (Figs. 4D-F). Furthermore, serial dilution imaging of both A498-TXNIP-Pro-Luc2 cells and A498-Luc2 cells showed that bioluminescence intensity was proportional to cell number (R² = 0.994 and R² = 0.997, respectively) (Figs. 4G, H). Collectively, these results demonstrated the successful generation of a stable TXNIP promoter-driven luciferase reporter cell line.
Fig. 4. Characterization of a stable RCC reporter cell line with TXNIP promoter-driven luciferase expression. (A) Schematic diagrams of pGL4.19-TXNIP-Pro-Luc2 constructs. The TXNIP promoter fragment, spanning from − 1166 bp to + 312 bp relative to the transcription start site (TSS), was cloned into the pGL4.19-Luc2 vector to drive luciferase expression. (B) A498-TXNIP-Pro-Luc2 cells or A498-Luc2 cells were lysed for luciferase activity analysis. (C) A498-TXNIP-Pro-Luc2 and A498-Luc2 cells were imaged using the IVIS Lumina LT system to obtain flux measurements (left panel, images). Quantified flux data were averaged (n = 3) and plotted (right panel, graph). (D, E) After transfection of the MLXIP plasmids into A498-TXNIP-Pro-Luc2 (D) or A498-Luc2 cells (E) for 48 h, imaging was performed (left panel, images). Quantified flux data were averaged (n = 3) and plotted (right panel, graph). (F) After transfection of the MLXIP plasmids into A498-TXNIP-Pro-Luc2 or A498-Luc2 cells for 48 h, the cells were lysed for luciferase activity analysis. (G, H) A498-TXNIP-Pro-Luc2 cells (G) and A498-Luc2 cells (H) were serially diluted, placed into wells of a 96-well plate, and immediately imaged. Quantified flux data were averaged (n = 3) and plotted. Data are mean ± SEM (n = 3); Statistical significance for B and C was analyzed by one-way ANOVA; for D, E, and F, it was analyzed by one-way ANOVA with Dunnett’s multiple comparisons test; ****p < 0.0001
Response of the TXNIP-Pro-Luc2 reporter to 2-DG-type glycolysis inhibitors
To evaluate whether the TXNIP-Pro-Luc2 reporter could accurately reflect the effects of 2-DG-type glycolysis inhibitors on RCC cells, we treated A498-TXNIP-Pro-Luc2 and control A498-Luc2 cells with increasing concentrations of 2-DG, 2-FG, and 2-DG-d for 48 h and measured luciferase activity. The concentration gradients for each compound are detailed in the Fig. 5 legend. The results showed that all three compounds induced a significant, dose-dependent increase in luciferase activity in A498-TXNIP-Pro-Luc2 cells (Figs. 5A–C), whereas no significant changes were observed in A498-Luc2 controls (Figs. 5D–F). We next performed bioluminescence imaging, which confirmed that 2-DG and its derivatives treatment for 48 h led to a dose-dependent elevation in luminescence signal intensity in A498-TXNIP-Pro-Luc2 cells (Figs. 5G–I). In contrast, control A498-Luc2 cells showed no significant changes in signal intensity after treatment with 2-DG and its derivatives. However, 2-DG exposure caused a marked reduction in luminescence intensity, likely due to its higher cytotoxicity at the tested concentrations (Figs. 5J–L). Together, these data demonstrate that the TXNIP-Pro-Luc2 reporter sensitively and specifically responds to 2-DG-type glycolysis inhibitors, providing a reliable tool for assessing their effects in A498 cells.
Fig. 52-DG and its derivatives activate TXNIP promoter-driven luciferase expression in A498 cells. (A–C) Luciferase activity driven by the TXNIP promoter (A498-TXNIP-Pro-Luc2 cells) after 48 h treatment with 2-DG (A), 2-FG (B), and 2-DG-d (C). For 2-DG treatment, the concentration gradient was set as 0 mM (vehicle control, containing equal volume of DMSO), 5 mM, 10 mM, and 15 mM; while for 2-FG and 2-DG-d treatments, the concentration gradients were consistent: 0 mM (vehicle control), 1 mM, 2 mM, 5 mM, and 10 mM. (D–F) Luciferase activity in control A498-Luc2 cells after 48 h treatment with 2-DG (D), 2-FG (E), and 2-DG-d (F). The concentration gradients for each compound were the same as those described for A498-TXNIP-Pro-Luc2 cells above. (G–I) After treating A498-TXNIP-Pro-Luc2 cells with 2-DG (G), 2-FG (H), and 2-DG-d (I), for 48 h, flux measurements were acquired using the IVIS Lumina LT system. (J–L) After treating A498-Luc2 cells with 2-DG (J), 2-FG (K), and 2-DG-d (L) for 48 h, flux measurements were acquired using the IVIS Lumina LT system. Top, cellular images; bottom, normalized fold induction of TXNIP-Pro-Luc2 or Luc2 treated with the indicated doses of drugs. Quantified flux data were averaged (n = 3) and plotted. The dosage of each compound was consistent with that used in the previous luciferase activity assay. Data are mean ± SEM (n = 3); Statistical significance was analyzed by one-way ANOVA with Dunnett’s multiple comparisons; ****p < 0.0001, ** p < 0.01, *p < 0.05
Discussion
The glycolytic pathway (Warburg effect) is a key metabolic target in cancer therapy [21–23], but current methods to identify its inhibitors rely on conventional biochemical metrics like glucose uptake, lactate production, and ATP levels. These approaches are technically limited by complex procedures, high variability, and low throughput, which hinders efficient drug development. Reporter genes coupled with bioluminescent imaging enable real-time, non-invasive drug response monitoring, accelerating anticancer discovery [24–26]. However, the development of such reporter systems for high-throughput screening (HTS) of glycolysis inhibitors has not been reported.
A critical first step in developing this bioluminescent screening platform was to identify effector genes within specific tumor cell types that exhibit a detectable response to drug treatment [27–29]. Therefore, to discover such effector genes in RCC cells that respond to 2-deoxy-D-glucose (2-DG), we performed RNA-seq analysis on A498 cells treated with 2-DG. The results showed that TXNIP was the most significantly upregulated gene following 2-DG treatment, a finding further confirmed by subsequent RT-qPCR and Western blotting analyses. Moreover, using the TCGA database online tool, we found that TXNIP is involved in regulating the glucose metabolism process in KIRC. Further experiments demonstrated that overexpression of TXNIP suppresses glycolysis in A498 renal cancer cells. These data establish TXNIP as a sensitive effector gene reporting on 2-DG response in A498 cells, likely providing an early readout of glycolytic inhibition.
Effector gene promoter-driven luciferase reporters represent one of the most widely utilized tools in HTS. A fundamental prerequisite for developing such systems is the confirmation that the drug of interest regulates the effector gene at the transcriptional level. In this study, we demonstrated that 2-DG treatment transcriptionally upregulates TXNIP expression in A498 cells. Although the complete mechanistic cascade warrants further investigation, our data, together with established literature, implicate the transcription factor MLXIP as a pivotal mediator in this process [30–32]. This is supported by the observations that 2-DG enhances MLXIP expression and that MLXIP overexpression alone is sufficient to activate TXNIP transcription.
Based on these findings, we propose a model in which 2-DG-induced TXNIP upregulation is mediated, at least in part, through MLXIP. This model not only provides a mechanistic rationale for utilizing TXNIP promoter activity as a functional readout for 2-DG-like compounds but also establishes this transcriptional activity as a specific indicator of their bioactivity. Consequently, we reasoned that linking the TXNIP promoter to a luciferase reporter gene would enable the development of a dynamic detection tool for identifying 2-DG-like compounds. Leveraging this mechanistic insight, we fused the TXNIP promoter with the luciferase gene to construct a bioluminescent reporter system for monitoring TXNIP transcriptional activity.
To evaluate the feasibility of using the TXNIP-Pro-Luc2 reporter system for screening glycolysis inhibitors such as 2-DG and its derivatives, we validated its performance in A498 cells. The results demonstrated that as the concentration of 2-DG increased, both cell proliferation and glycolytic activity were suppressed, while the luminescent signal from the reporter gene exhibited a dose-dependent enhancement. This confirms that the system can specifically and reliably report the bioactivity of 2-DG-like compounds. Furthermore, the core value of this system lies in its utility as a high-throughput primary screening tool. Its simple procedure, which requires only reagent addition and plate reading, along with full compatibility with high-throughput microplate workflows, enables the rapid screening of large compound libraries. Compared to conventional methods, including glucose uptake and lactate production assays which involve cumbersome sample preparation and multi-step incubations, as well as HPLC-based metabolite quantification which is inherently low-throughput and time-consuming, our system shows distinct advantages in throughput, sensitivity, and cost-effectiveness. By directly monitoring the activation of the MLXIP/TXNIP signaling axis, it specifically reflects the inhibitory effect of 2-DG-like compounds, eliminating the need for multiplexed traditional assays. Moreover, it allows early exclusion of ineffective molecules at the in vitro stage, thereby significantly reducing the cost and effort of subsequent in vivo validation. In summary, the reporter system established here represents a validated, efficient, and specific platform suitable for identifying novel anti-renal cell carcinoma lead compounds that act through a similar pathway. It is important to emphasize that within the complete drug discovery pipeline, traditional methods such as glucose uptake, lactate production assays, and HPLC-based metabolite quantification remain indispensable for validation and mechanistic investigation, whereas this reporter system is positioned as an efficient front-end screening tool. Their roles are distinct yet complementary, together forming an integrated methodological framework from initial screening to definitive confirmation.
The application of this reporter system, however, is selective and subject to several limitations. First, its functionality is intrinsically dependent on an intact MLXIP/TXNIP signaling axis. In cancer models where this pathway is impaired, the system may yield false-negative results by failing to detect glycolytic inhibitors that act through alternative mechanisms, thus limiting its generalizability and requiring pathway validation prior to any new application. Second, the reporter serves as a pathway-specific sensor, not a universal glycolysis detector. It is calibrated for 2-DG-like compounds, and its responsiveness to other metabolic inhibitors such as those targeting mitochondrial respiration or specific glycolytic enzymes like HK2 remains unestablished. This limited scope precludes its use as a comprehensive metabolic screening tool. Third, as a promoter-driven system, it detects transcriptional activation rather than direct functional inhibition. A positive signal indicates TXNIP promoter activity but does not constitute direct proof of glycolytic suppression, introducing a risk of false positives from off-target effects. Hits must therefore be confirmed with direct biochemical assays measuring metabolic parameters. Finally, the findings and validation presented in this study are derived exclusively from in vitro experiments. The performance, stability, and biological relevance of the A498-TXNIP-Pro-Luc2 reporter system within a complex tumor microenvironment, involving interactions with stromal cells, variable nutrient supplies, and hypoxic conditions, are entirely unknown and represent a critical area for future investigation. Successful performance in vitro does not guarantee equivalent functionality or predictive value in vivo.
In conclusion, by leveraging the MLXIP/TXNIP pathway’s specific responsiveness to glycolytic inhibition, we have developed a bioluminescent reporter system that overcomes key limitations of traditional metabolic assays. Since TXNIP acts as a key downstream effector, its transcriptional activation rapidly responds to 2-DG-like glycolytic inhibitors, often occurring before detectable changes in metabolic flux or ATP depletion. This feature provides an early and concentration-responsive readout for the activation of the MLXIP/TXNIP axis in response to glycolytic inhibition. Moreover, luminescence detection is inherently compatible with high-throughput automation, facilitating the drug screening process. Collectively, this system shows promise as an efficient high-throughput platform for discovering novel anti-renal cell carcinoma (RCC) agents that mimic 2-DG’s mechanism of action.
Conclusions
In conclusion, our findings demonstrate that the glycolytic inhibitor 2-DG transcriptionally upregulates TXNIP expression in renal cell carcinoma cells. Building on this molecular mechanism, we established a stable bioluminescent reporter cell line, A498-TXNIP-Pro-Luc2, which exhibits favorable concentration-dependent responsiveness and specificity to 2-DG and its functional analogues by producing a dose-dependent luminescent signal, thereby providing a direct readout of drug-induced TXNIP promoter activity. Taken together, the TXNIP-Pro-Luc2 reporter system represents a novel and practical tool for high-throughput screening of 2-DG-like glycolytic inhibitors, offering a promising preliminary screening platform to facilitate the identification of next-generation anti-renal cancer candidate compounds.
Supplementary Information
Below is the link to the electronic supplementary material.
Supplementary Material 1
The reference list from the paper itself. Each links out to its DOI / PubMed record.