RORα inhibits proliferation and chemoresistance through AKR1A1-induced glucose and lipid reprogramming in gastric cancer

doi:10.21203/rs.3.rs-4864679/v1

Download PDF

Research Article

RORα inhibits proliferation and chemoresistance through AKR1A1-induced glucose and lipid reprogramming in gastric cancer

https://doi.org/10.21203/rs.3.rs-4864679/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background Abnormal glycolysis and lipid synthesis play important roles in the occurrence and development of gastric cancer (GC).Moreover, dysregulation of circadian genes was associated with metabolic reprogramming in the tumor microenvironment. This study aimed to decipher the role of retinoic acid-related orphan receptor alpha (RORα) in glucose and lipid reprogramming of GC.

Methods The effects on GC growth in vitro and in vivo were studied using gain- and loss-of-function experiments. Glycolysis and lipid metabolism of GC cells were detected using seahorse assay and test kit. Moreover, regulatory mechanisms underlying RORα deletion were explored using half-life, Co-Immunoprecipitation (Co-IP), Chromatin immunoprecipitation (Chip), luciferase reporter and immunofluorescence co-localization assays in GC cells. In addition, the relationship of RORα with E47 and AKR1A1 was analyzed using clinicopathological retrospective analysis in two cohorts.

Results RORα deletion promoted proliferation and 5-FU chemoresistance by enhancing glycolytic activity and lipid synthesis. In contrast, SR1078, a RORα activator, reversed these phenomena and had a synergistic inhibitory effect on cell proliferation through combination with 2-deoxygulose glucose or atorvastatin. Mechanistically, RORα deletion promoted aldo-keto reductase family 1 member A1 (AKR1A1)-induced glycolysis and lipid metabolism to enhance cell proliferation and 5-FU chemoresistance. In detail, the protein stability of the AKR1A1 transcription factor E47 is modulated by β-catenin through binding to form a transcription heterodimer. RORα deletion enhanced wnt3a and β-catenin expression and indirectly inhibited E47 protein stability, leading to attenuated AKR1A1 transcriptional activity. Moreover, AKR1A1 expression was associated with clinicopathological parameters, including age, tumor-node-metastasis stage and tumor grade; AKR1A1 was found to be an independent predictor of GC prognosis and was negatively correlated with RORα expression. Moreover, the disease-free survival time of GC patients with RORα-high/E47-low/AKR1A1-low expression patterns was optimal. Furthermore, RORα, E47 or AKR1A1 expression alone or in combination with each other was associated with response status, the standard uptake value and lipid synthesis.

Conclusions These findings revealed a novel mechanism by which RORα regulates glucose and lipid reprogramming and might be a promising target for GC treatment.

RORα

β-catenin

E47

AKR1A1

glucose and lipid reprogramming

gastric cancer

Gastric cancer (GC) is one of the most common cancers worldwide (1,2). Despite the development of early diagnosis and therapeutic treatments, the incidence and mortality of advantaged GC remain unsatisfactory for patients due to aging populations, poor dieting customs and living circumstances and susceptibility gene heredity over the past several decades (3,4). As the course of treatment progresses, most patients develop chemoresistance, leading to chemotherapy failure and even rapid tumor recurrence. Moreover, patients are subjected to excessive drug administration to overcome chemoresistance. Eventually, the side effects are aggravated, resulting in a negative cycle(1,3). Thus, it is essential to explore novel targets to improve GC therapeutics.

Glucose and lipid reprogramming are regarded as hallmarks of cancer development through abundant energy uptake in vascular disorders, extreme nutrient deficiencies and anoxic microenvironments(5). Moreover, the circadian rhythm plays a vital role in immune regulation to prevent invasion of tumor cells(5). Thus, there is much crosstalk between metabolic reprogramming and the circadian rhythm under physiological and pathological conditions. On the one hand, maintaining normal metabolic processes requires a circadian rhythm. On the other hand, aberrant regulation inevitably causes abnormal progression, apoptosis, and even metastasis (5). In previous studies, the dysregulation of MYC disturbed the molecular clock by directly inducing REV-ERBα to dampen the expression and oscillation of BMAL1 to affect glucose metabolism (6). The circadian gene HKDC1 promoted glycolysis, tumorigenesis and EMT through activating NF-kB signaling pathways in GC (7). Moreover, in mice in which the REV-ERBα and REV-ERBβ genes were double deleted, wheel running behavior and lipid homeostasis were disrupted through the inhibition of circadian expression of core circadian genes(8). In addition, GPAM is a key enzyme involved in lipid biosynthesis and is associated with HCC progression. The complex of BMAL1 and EZH2 is recruited to the promoter of GPAM to suppress its transcriptional activity(9). Retinoid-related orphan receptor alpha (RORα), a circadian gene, participates in metabolic regulation, and its dysregulation is associated with a variety of chronic diseases, immunity, inflammation and cancers (10–13). However, the role of RORα in glucose and lipid reprogramming of GC is poorly understood. Moreover, SR1078, a RORα activator, was verified to suppress oncogenesis(14). Thus, the application of SR1078 in the treatment of GC is valuable.

In this study, we revealed that RORα deletion promotes proliferation and 5-FU chemoresistance by attenuating AKR1A1-induced glycolysis and lipid synthesis in GC. Moreover, SR1078 reversed these effects and synergistically inhibited GC proliferation in the presence of 2-deoxygulose glucose (2-DG) or atorvastatin. Overall, RORα might be a novel therapeutic target for GC patients to suppress tumor progression and overcome chemoresistance.

Patients and specimens

Samples (paraffin-embedded sections and frozen liver tissues) and clinicopathological data of GC patients were collected from The First Affiliated Hospital of Anhui Medical University (Hefei, Anhui) from August 2016 to September 2023. The inclusion criteria for eligible patients were confirmed gastric adenocarcinoma and Eastern Cooperative Oncology Group Performance Status scores between 0 and 2 (15). The clinical stage was assessed by chest, abdomen and pelvis enhanced CT or MRI according to the 7th Edition of the International Union against Cancer Tumor–Node–Metastasis (TNM) classification(16). The standard uptake value (SUV) was determined by 18F-FDG PEC/CT. Imaging diagnosis was determined by at least two radiologists from the Department of Nuclear Medicine. The collection diagram and demographic characteristics are illustrated in Figure S1 and Table 1. Disease-free survival (DFS) time was defined as the time of local recurrence, distant metastasis or death from any complication. The response to treatment was defined as partial response (PR), complete response (CR), stable disease (SD) or progressive disease (PD) in locally advanced GC patients (cT4a/bN+) who received neoadjuvant chemotherapy according to the RECIST 1.1 guidelines (17). All patients signed informed consent before the study, which was approved by the human Ethics Committee of Anhui Medical University (20180344, Hefei, Anhui).

Neoadjuvant chemotherapy regimens

Patients received 3 cycles of treatment based on the XELOX and FOLFOX-6 regimens according to the Chinese Society of Clinical Oncology (CSCO) and National Comprehensive Cancer Network (NCCN) guidelines as follows (18,19). 1): The XELOX regimen included oxaliplatin (120 mg/m² on day 1 in each cycle, intravenously) plus capecitabine (1000 mg/m²twice daily on days 1-14, orally); 2): The FOLFOX-6 regimen included fluorouracil (400 mg/m² on day 1 followed by 2400 mg/m²on46 hoursin each cycle, intravenously), leucovorin (400 mg/m² on day 1 in each cycle, intravenously) plus oxaliplatin (400 mg/m² on day 1 in each cycle, intravenously).

Single-cell sequencing data downloading, preprocessing and analysis

Single-cell RNA transcription data were downloaded from the NCBI Gene Expression Omnibus (GEO) database (www.ncbi.nlm.nih.gov/geo/) with 10X Genomics in 10 fresh tissues (GSE163558), including 3 primary tumor tissues, 1 adjacent gastric tissue and 6 tissues with GC cell metastasis (2 lymph nodes, 1 ovary, 1 peritoneum and 2 livers). These tissues were invaded by GC cells. The functions of Read10x, CreatSeuratObject and Merge in the R package ‘Seurat’ were utilized to create data for R software to obtain high-quality cells.

DoubletFinder was used to remove double cells. The inclusion criteria were as follows: fewer than 80000 total genes, fewer than 5% mitochondrial genes and fewer than 50 housekeeping genes.

The top 2000 highly expressed genes were identified through the function FindALLMarkers according to dispersion and average expression levels. The functions Normalizedata and Harmony were utilized to standardize scaling and eliminate batch effects, respectively. The functions FindNeighbors and FindClusters were utilized to gather different clusters through differentially expressed genes at a resolution of 0.6.

The copy number analysis was performed by the R package inferCNV (v1.0.6) using the default parameters. Endothelial and fibroblasts were used as controls to identify highly CNV epithelial cells, which were regarded as cancer cells.

The 123 5-FU-related genes were collected from published references, including 83 resistance genes and 40 sensitive genes (20). The AUCell algorithm was utilized to score each tumor cell line. Based on the scores, we nomenclatured each cluster to compare the expression levels of RORα among these cell types. The ridge and box plots were generated using the ggolot package and ggstatsplot package, respectively.

RNA transcriptome analysis

The RNA-sequencing transcriptome was generated by OE Biotech Co., Ltd. (Shanghai, China) in the primary and adjacent tissues of 5 GC patients. Total RNA was extracted with TRIzol reagent (Invitrogen, CA, USA) according to the manufacturer’s instructions. RNA purity and quantification were performed using a NanoDrop 2000 Spectrophotometer (Thermo Scientific, USA). RNA integrity was assessed using an Agilent 2100 Bioanalyzer (Agilent, Technologies, Santa Clara, CA, USA). Libraries were generated with a VAHTS Universal V6 RNA-seq Library Prep Kit according to the manufacturer’s instructions.

Transcription factor prediction via bioinformatic analysis

The online bioinformatics tools AnimalTFDBI, TRANSFAC PATCH and MATCH were used to identify the potential transcription factors of AKR1A1 (Table S1). JASPAR (http://jaspar.genereg.net/) was used to evaluate the highest binding score of the two candidates.

Cell culture and treatment

Human GC cell lines (AGS and MKN-74) and a mouse forestomach carcinoma (MFC) cell line were obtained from Procell Life Science and Technology (Wuhan, China). These cells were cultured according to the protocol. Cells were treated with SR1078 (5 µM, GlpBio, Montclair, CA, USA), 2-DG (8 mM, Sigma‒Aldrich, MO, USA) and atorvastatin (10 µM, GlpBio, Montclair, CA, USA) for 24 h. SR1078 was dissolved in phosphate-buffered saline (PBS). Other reagents, including fluorouracil (GC14466, GlpBio, Montclair, CA, USA), were dissolved in dimethyl sulfoxide (DMSO).

CRISPR-Cas9 gene deletion and small interfering RNA (siRNA) transfection

Single guide RNA (sgRNA) oligonucleotides were cloned and inserted into LV-U6-spsgRNA(RORα)-CMV-SV40-NLS-spCas9-NLS-Flag-P2A-Puro-T2A-EGFP-WPRE (human) and packaged as lentiviruses (BrainVTA, Wuhan, China). An alignment of the DNA sequences of the human and mouse RORα genes revealed 91% similarity according to the NCBI database. The cells were infected with a virus expressing Cas9 and gRNA for 12 hours. After culturing with 2 μg/ml puromycin for 72 h, we found that the fluorescence (30-50%) of the AGS (MOI=0.5), MKN-74 (MOI=0.3) and MFC (MOI=0.5) strains was satisfactory. Western blotting was used to detect protein expression levels after transfection. The RORα-KO sgRNA sequence (human) was GTAATCGACAGTGTTGGCAG. siRNA-E47 (RiboBio, Guangzhou, China) and siRNA-β-catenin (RiboBio, Guangzhou, China) were dissolved in 1,2-dierucoyl-sn-glycero-3-phosphocholine (DEPC). The cells were cultured in 1640 medium without fetal bovine serum (FBS). The siRNAs and Lipofectamine 3000 (Thermo Scientific, USA) were diluted and incubated at room temperature for 30 minutes. The mixtures were cultured for 6 h in an incubator, and then the medium was changed to complete medium for 24 h until the next series of experiments.

Lentivirus infection for knockdown and overexpression

The lentiviral particles pLVX-shRNA-AKR1A1-Puro (LVP0916) and pLV4ltr-PGK-ZsGreen(2A)PURO-CMV (LVP0917, E47 overexpression) were purchased from TSINGKE (Beijing, China). Cells were seeded into 6-well plates and cultured with 8 μg/ml polybrene complete medium for 24 h. Next, AGS (MOI=10, 10) and MKN-74 (MOI=5, 5) cells were transfected with virus for knockdown or overexpression, respectively. The cells were cultured in an incubator for 24 h, after which the medium was changed to complete medium for western blot analysis to determine the transfection efficiency.

Colony formation assay

Cells (1200 cells/well) were seeded into 6-well plates and cultured in an incubator through a counting plate. The wells were fixed and stained with 0.1% crystal violet for 10 min after 10 days. The number of colonies in each well was calculated three times.

Cell Counting Kit-8 (CCK-8) assay

Cells were seeded into 96-well microplates and cultured with fluorouracil according to concentration gradients. The mixtures were incubated in 10 µl of enhanced Cell Counting Kit-8 (CCK-8, C0042, Beyotime, Haimen, China) for 1 h according to the manufacturer's protocol. The optical density (OD) values were measured at 450 nm using a microplate reader (Thermo Scientific, USA). The relative absorbance values were analyzed through OD values.

Flow cytometry analysis

Pretreated cells were cultured with fluorouracil (50 µg/ml) for 24 h. The cell suspension was collected and then incubated with 5 µl of Annexin V-FITC and 5 µl of PI through an Apoptosis Kit (E-CK-A211, Elabscience, Wuhan, China) at room temperature for 15 min. The percentage of cell apoptosis was determined through flow cytometry (Beckman Coulter, USA).

Mouse model construction

Six- to eight-week-old mice (C57, Vital River Laboratory Animal Technology, Beijing, China) were randomly divided into different groups. A total of 1 × 10⁶MFC cells were injected into the subcutaneous flank of mice to construct a subcutaneous tumor model for 4 weeks. A total of 5 × 10⁵MFC cells were injected into the portal or caudal vein of mice to establish a model of liver metastasis or necrosis for 2 weeks. The weight and volume (1/2 × long diameter × short diameter) of the mice were measured weekly, and the mice were sacrificed, and samples were collected for immunohistochemistry and hematoxylin-eosin (H-E) staining after an adequate breeding time. The liver metastatic nodules and extent of necrosis were observed and calculated by microscopy (OLYMPUS, Japan). These studies complied with international protective guidelines for laboratory animals and ethical standards. This project was approved by the Ethics Committee of Anhui Medical University (20180365, Hefei, Anhui).

Immunohistochemistry, H&E and Oil Red O staining

The sections were deparaffinized and immersed in boiling citrate buffer (pH=6.0). Primary antibodies against RORα, Ki-67, proliferating cell nuclear antigen (PCNA), multidrug resistance 1 (MDR1), E47 and AKR1A1 were incubated with the cells for 2 h (Table S2). 3,3'-diaminobenzidine (ZSGB‑BIO, OriGene Technologies, Beijing, China) and 20% hematoxylin were used for staining at room temperature. The mean optical density (MOD) method was used to assess relative expression levels in previous studies (21). High and low expression was judged as the proportion grade multiplied by the staining intensity score. The proportion grades were as follows: 3 (> 76%), 2 (26–75%), 1 (26–50%), and 0 (< 25%). The staining intensity scores were as follows: 2 (brownish and yellow brown), 1 (light staining) and 0 (no staining). A sample score of 0–2 was defined as low expression, and 3–6 was defined as high expression. Frozen liver tissues were collected, and Oil Red O staining was performed by Servicebio Company (Wuhan, China). A TissueGnostics viewer (TG, Austria) was used to obtain images, which were quantified with Image J software.

Immunofluorescence colocalization

Cells were seeded into 6-well plates containing slides for 24 h. The slides were fixed with 4% paraformaldehyde for 10 min, permeabilized with 0.5% Triton X-100 for 10 min and then blocked with bovine serum albumin (BSA) for 2 h at room temperature. After washing with PBS three times. The blocked cells were incubated with primary antibodies against β-catenin and E47 at 4°C overnight and then incubated with fluorescent secondary antibodies for 1 h at room temperature (Table S2). 4′,6-Diamidino-2-phenylindole (DAPI) was used to stain the nuclei for 10 min. TG was used for image acquisition.

Western blot analysis

Cell lysis was performed with RIPA buffer. The protein concentration of the samples was determined by a BCA assay (KeyGen Biotech, Nanjing, China). The protein extract was separated by 12% SDS‒PAGE and transferred to a polyvinylidene fluoride (PVDF) membrane. Primary antibodies against RORα, AKR1A1, E47, β-catenin, wnt3a and β-actin were incubated overnight at 4°C (Table S2). The membranes were incubated with secondary antibodies for 2 h at room temperature. The bands were imaged with a chemiluminescence system (Tanon, Shanghai, China).

Quantitative real-time PCR (qPCR)

Total RNA was isolated using TRIzol (Invitrogen, USA) according to the manufacturer’s instructions. cDNA was generated using a cDNA kit (TaKaRa, Shiga, Japan), and PCR was performed using GoTaq® Green Master Mix (Promega, USA) on a 7900 Thermal Cycler (Thermo Fisher Scientific, USA) through denaturation, renaturation and extension of primers (Table S3).Relative mRNA expression levels were calculated through the 2⁻^ΔΔ^Cq method(22). The β-actin gene was used as an internal control for normalization.

Seahorse assay

The extracellular acidification rate (ECAR) was determined with a Seahorse XFe24 analyzer (Seahorse Bioscience, USA) according to the manufacturer’s guidelines. Briefly, the pretreated cells were seeded into 24-well plates after sensor calibration. Next, XF test medium was added to each well, and the plates were incubated for 1 h at room temperature. Subsequently, the ECAR was detected following sequential treatment with glucose (10 mM), oligomycin (1 μM), and 2-DG (100 mM).

ATP production measurement

AnATP assay kit (S0026, Biyotime, Haimen, China) was used to detect ATP consumption according to the manufacturer’s instructions. The cell lysate was centrifuged for 20 min at 40°C to collect the supernatant, which was subsequently incubated with diluted ATP detection reagent for 5 min. The relative unit light (RUL) values were measured by a microplate reader and were converted into ATP concentrations according to a standard curve.

Triglyceride and phospholipid measurements

The cellular contents of triglycerides and phospholipids were measured with a triglyceride assay kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) and an EnzyChrom™ phospholipid assay kit (BioAssay Systems, CA, USA) according to the manufacturer’s instructions. Briefly, the OD values of the lysed cells were measured with a microplate reader at 500 or 570 nm. The concentration was calculated by a standard curve.

Coimmunoprecipitation (Co-IP) assay

The cell supernatant was collected and incubated with antibodies against E47 or β-catenin (Table S2) overnight at 4°C. Protein A/G PLUS-Agarose beads (10μl, sc-2003, Santa Cruz Biotechnology, USA) were added and incubated at room temperature for 1 h. After transient centrifugation. The immunoprecipitated agarose beads were collected and resuspended in loading buffer. Next, these samples were subjected to immunoblotting after boiling.

Chromatin immunoprecipitation (Chip)-qPCR assay

A chip assay kit (P2078, Beyotime, Haimen, China) was used to perform the chip assay. The detailed process of the experiment was described in a previous study(23). The primary antibodies against E47 and β-catenin are listed in Table S2. Immune complexes were obtained after crosslinking, ultrasonication and washing. The purified DNA fragments were obtained through decrosslinking and were subjected to qPCR.

Luciferase reporter assay

The region of the AKR1A1 promoter from +291 to -1664 bp was inserted into the pGL3-basic vector (Addgene, #212936) for construction of the luciferase reporter plasmid. Next, the cells were cotransfected with the E47 overexpression or control vector and luciferase reporter plasmids with Lipofectamine 3000 for 24 h. A luciferase assay was performed with a Pierce^TMluciferase kit (Promega, USA) according to the manufacturer’s instructions. The sequences used were as follows: GCCAACAGTTACTTTGTTCCCT (forward) and ATGTCCATTTTGATCTGCAAAGT (reverse).

Half-life of E47 mRNA and protein expression

Cells were treated with actinomycin D (5 µg/mL, Aladdin, Shanghai, China) or cycloheximide (20 µg/ml, MCE, USA) for the indicated durations. The mRNA abundance at each time point was normalized to the initial abundance.

Analysis of drug combinations

The effect of drug combination was determined by the SynerFinder 3.0 web application (https://synergyfinder.org/) according to the Zero Interaction Potency (ZIP) model(24). The percentage of relative cell viability was calculated from the OD values. The white and red regions represent superposition and synergy, respectively. The summary synergy scores were as follows: 1) less than -10 indicated antagonistic effects; 2) from -10 to 10 indicated additive effects; and 3) greater than 10 indicated synergistic effects.

Statistical analysis

Statistical analysis was performed with SPSS 19.0 and GraphPad Prism 10.0.0 software. Comparisons between different groups were performed using ANOVA or the Mann‒Whitney test. Survival analysis was performed by the Kaplan‒Meier method and log-rank test. The correlation analysis was performed by the chi-squared test. Univariate and multivariate survival analyses were performed with a Cox proportional hazards model. P value < 0.05 was considered a statistically significant difference.

RORα is associated with cell proliferation and 5-FU chemoresistance in GC

We investigated the role of RORα in cell proliferation and chemoresistance through functional experiments both in vivo and in vitro. First, we utilized CRISPR-Cas9 gene deletion technology to eliminate the endogenous expression of RORα, and the results were verified by western blot analysis. Moreover, SR1078 enhanced RORα expression in GC and MFC cells (Figure 1A). Next, colony formation assay showed that RORα deletion promoted colony formation. However, these effects were reversed in GC cells treated with SR1078 (Figure 1B). In previous studies, the role of RORα in chemoresistance was unclear. We utilized single-cell sequencing analysis and found that 35532 high-quality cells were selected for analysis after removing double cells and debris. As shown in Figure S2A, B and C, these cells were grouped into 17 clusters and were annotated based on known cell lineage-specific marker genes, including epithelial cells (“EPCAM”, “KRT19” and “CLDN4”), endothelial cells (“RAMP2”, “PECAM1”, “KDR” and “VWF”), fibroblasts (“ACTA2”, “COL1A2”), T lymphocytes (“CD2”, “CD3D”, “CD3E” and “CD3G”), B lymphocytes (“CD79A” and “CD79B”), and myeloid cells (“LYZ” and “MNDA”) (20,25). As shown in Figure S2D, nearly all of the epithelial cells exhibited high copy number variations (CNVs), which were assigned to cancer cells. Furthermore, all of the cancer cells (1163 cells) were exacted and clustered into 15 clusters and termed resistant (445 cells), sensitive (664 cells) or intermediate (64 cells) according to the AUCell scores (Figure S2E, F and G). As shown in Figure S2H, among these cell lines, we further analyzed the expression pattern of RORα and showed that the sensitive subtypes harbored significantly greater expression levels of RORα than did the resistant and intermediate subtypes (P=0.01). These results indicated a detrimental effect of RORα reduction on 5-FU-based chemotherapy in patients with GC. We further verified that RORα deletion promoted resistance to fluorouracil. In contrast, SR1078 attenuated resistance in GC cells, as determined by CCK-8 and flow cytometry assays (Figure 1C and D). In vivo, compared with those in the CON group, the tumor volume, weight, ki-67, PCNA and MDR1 expression in the RORα-KO-MFC group were greater. These effects were reversed in the SR1078 group (Figure 1E and F). Moreover, the RORα-KO-MFC group displayed more necrotic regions and more metastatic nodules than did the CON group. These effects were also reversed in the SR1078 group (Figure 1G and H).

RORα is involved in glucose and lipid reprogramming in GC

We explored the effect of RORα on glucose and lipid reprogramming. RORα deletion enhanced glycolysis, glycolytic capacity and ATP consumption. These effects were reversed by SR1078 in GC cells (Figure 2A and B). Moreover, the levels of triglycerides and phospholipids were increased in RORα-KO GC cells but decreased in GC cells treated with SR1078 (Figure 2C). 2-DG has been verified to disturb glucose metabolism to suppress the development of malignant tumors(26). Moreover, atorvastatin, an HMG-CoA reductase inhibitor, has the ability to reduce blood lipids, which significantly suppresses tumors(27). We then explored whether SR1078 can synergistically treat GC with 2-DG or atorvastatin. We found that the combination of SR1078 with 2-DG or atorvastatin significantly inhibited GC proliferation through colony formation (Figure 2D). Moreover, in vivo, we found 2-DG or atorvastatin group inhibit tumor growth and proliferation compared with Vehicle group. SR1078 could significantly promote the effect of 2-DG or atorvastatin (Figure S3A, B and C). Next, we utilized OD values to calculate cell viability under different concentration combinations. The synergistic effects of SR1078 combined with 2-DG or SR1078 combined with atorvastatin were revealed (Figure 2E).

RORα deletion promotes proliferation and chemoresistance through AKR1A1-induced glucose and lipid reprogramming in GC

We collected 5 paired tissues from GC patients for RNA transcriptome analysis. The RORα mRNA expression in these tumor tissues was lower than that in the corresponding control (adjacent) tissues (Figure 3A). As shown in Figure 3B and C, Gene set enrichment analysis (GSEA) indicated the involvement of the TCA cycle and fatty acid beta-oxidation. The profile of glycolysis- and lipid metabolism-associated genes revealed 6 potential candidates, and AKR1A1 was the most enriched according to DESeq2 software based on the criteria of a q-value <0.05 and a fold change > 2 or <0.5 (28,29). To verify the regulatory effect of RORα on the AKR1A1 gene. Western blot analysis revealed that AKR1A1 expression was increased in RORα-KO GC cells and decreased in GC cells treated with SR1078 (Figure 3D). To test whether RORα deletion affects glucose and lipid reprogramming through the mediation of AKR1A1 and thus promotes proliferation and 5-FU chemoresistance, we introduced shRNA-AKR1A1 into RORα-KO GC cells (Figure S4A and B). Next, colony formation assay showed that shRNA-AKR1A1 obviously offset the promotive effect of RORα-KO on the progression of GC cells (Figure 3E). Moreover, CCK-8 and flow cytometry assays showed that shRNA-AKR1A1 obviously offset the resistance of RORα-KO GC cells to fluorouracil (Figure 3F and G). Subsequently, we explored the effect of shRNA-AKR1A1 on glucose and lipid reprogramming in RORα-KO GC cells. shRNA-AKR1A1 offset the promotive effect of RORα-KO on glycolysis, glycolytic capacity and ATP consumption in GC cells (Figure 3H and I). Moreover, shRNA-AKR1A1 also offset the promotive effect of RORα-KO on the levels of triglycerides and phospholipids in GC cells (Figure 3J).

AKR1A1-induced proliferation, chemoresistance, and glucose and lipid reprogramming are regulated by E47 in GC

RORα tends to be recruited to response elements and specific regions of promoters to participate in transcriptional processes. On the other hand, RORα is involved in multiple epigenetic modifications through binding to activators and suppressors (30,31). Thus, the bioinformatic tools AnimalTFDB, TRANSFAC PATCH and TRANSFAC MATCH were utilized to predict potential transcription factors of AKR1A1, and RORα was not found to be a transcription factor (Table S2). A Venn diagram revealed that there are 2 candidates (Figure 4A). JASPAR was used to assess the scores of the candidates, indicating that E47 is a potential AKR1A1 transcription factor (Figure 4B). To verify our prediction. We generated E47-overexpressing GC and MFC cells (Figure S4C and D). Western blot and q-PCR assays showed that E47 overexpression significantly enhanced AKR1A1 mRNA and protein expression in GC cells (Figure 4C). Moreover, a luciferase reporter assay showed that E47 overexpression enhanced AKR1A1-luciferase activity in GC cells (Figure 4D). In addition, a chip assay confirmed that E47 is recruited to the AKR1A1 promoter in GC cells (Figure 4E). These results indicated that E47 is a positive transcription factor for AKR1A1 in GC cells. Next, we found that E47 overexpression reversed the effect of shRNA-AKR1A1 on the proliferation of GC cells through colony formation assay (Figure 4F). Moreover, E47 overexpression also reversed the effect of shRNA-AKR1A1 on fluorouracil resistance in GC cells (Figure 4G). Subsequently, we explored the effect of E47 overexpression on glucose and lipid reprogramming in shRNA-expressing AKR1A1 GC cells. E47 overexpression reversed the suppressive effect of shRNA-AKR1A1 on glycolysis, glycolytic capacity and ATP consumption in GC cells (Figure 4H and I). Moreover, E47 overexpression reversed the suppressive effect of shRNA-AKR1A1 on the levels of triglycerides and phospholipids in GC cells (Figure 4J).

RORα attenuates E47 stability by inhibiting the wnt3a/β-catenin signaling pathway in GC cells

To explore whether E47 is involved in the regulatory relationship between RORα and AKR1A1. Western blot analysis revealed that the RORα-KO-mediated increase in AKR1A1 protein expression was abolished by the silencing of E47 (Figure 5A). However, SR1078 attenuated AKR1A1 protein expression in GC cells overexpressing E47 (Figure 5B). These results indicated that the regulation of AKR1A1 expression by RORα does not depend entirely on E47. Next, a half-life assay showed that RORα deletion significantly enhanced E47 protein stability but did not affect E47 mRNA stability (Figure 5C and D). In previous studies, TCF-4 exhibited strong binding to β-catenin but failed to bind RORα directly to form a heterodimer(32). Moreover, RORα can attenuate the Wnt/β-catenin signaling pathway in colon cancer(33). Thus, we hypothesized that RORα indirectly affects E47 stability through the regulation of β-catenin. Subsequently, an immunofluorescence assay showed that E47 and β-catenin can colocalize in the nucleus of GC cells (Figure 5E). Co-IP assay further verified the direct binding of β-catenin to E47 to form a heterodimer in GC cells (Figure 5F). A chip assay revealed that β-catenin is recruited to the AKR1A1 promoter in GC cells (Figure 5G). A half-life assay showed that siRNA-β-catenin significantly reduced E47 protein stability (Figure 5H). These results indicated that β-catenin can directly affect E47 stability through binding to each other in GC cells. Subsequently, we found that the effect of RORα on E47 protein stability was abolished by silencing β-catenin in GC cells (Figure 5I). Western blot analysis revealed that wnt3a and β-catenin expression levels were increased in RORα-KO GC cells and decreased in GC cells treated with SR1078 (Figure 5J). These results demonstrated the detailed mechanisms by which RORα attenuates E47 protein stability by inhibiting wnt3a/β-catenin signaling, leading to a reduction in AKR1A1 transcriptional activity (Figure 5K).

The relationships of RORα with E47 and AKR1A1 in GC patients

Finally, we analyzed the clinical expression of RORα, E47 and AKR1A1 in GC patients. We found that E47 expression was increased in stomach adenocarcinoma-tumor (STAD) tissues but that AKR1A1 expression did not significantly differ in timer 2.0 database (Figure 6A). Moreover, E47 and AKR1A1 were associated with poor overall survival (OS) according to the KM-plotter database (Figure 6B). However, the GEPIA database showed no significant difference (Figure 6C). In cohort 1, the number of GC patients with high AKR1A1 and E47 expression was greater than that with low AKR1A1 and E47 expression (Figure. 6D). Moreover, AKR1A1 expression was associated with several clinicopathological parameters, including age, tumor grade and TNM stage. However, no correlation was detected for E47 (Table 2). Furthermore, E47 and AKR1A1 were associated with poor DFS (Figure 6E). Multivariate and univariate analyses indicated that AKR1A1 can be a predictor of GC prognosis, but this effect was not detected for E47 (Figure 6F and Table 3). Next, we investigated the possible associations among RORα, E47 and AKR1A1. Analysis of the GEPIA database revealed that AKR1A1 is positively correlated with E47 and negatively correlated with RORα, which is similar to the results obtained for cohort 1. However, RORα was not correlated with E47 (Figure 6G and Table 4). Cohort 1 patients were assigned to three groups to determine the combined effect on prognosis. We found that group I (RORα-high/E47-low/AKR1A1-low) had the optimal DFS, and group II (RORα-low/E47-high/AKR1A1-high) had the poorest DFS, indicating the existence of the RORα/E47/AKR1A1 axis and the independence of RORα on E47 (Figure 6H). Subsequently, we compared the percentage of patients who achieved a response status in cohort 2 patients with RORα, E47 and AKR1A1 expression alone or in combination (groups I to IV). We found that RORα-high or E47-low expression alone and in combination group I (RORα-high/E47-low), group III (RORα-high/AKR1A1-low) or group IV (E47-low/AKR1A1-low) obtained a greater ratio in the PR or CR. In contrast, RORα-low or E47-high expression alone and in combination group V (E47-high/AKR1A1-high) resulted in a greater ratio of SD or PD (Figure 6I). Moreover, 18F-FDG PEC/CT was utilized to detect glycolytic activity. We found that the SUVmax was lower in patients with RORα-high expression alone and in combination group I (RORα-high/E47-low) or group III (RORα-high/AKR1A1-low) than in patients with RORα-high/AKR1A1-low expression alone or in combination groups (Figure 6J). Furthermore, we detected lipid metabolism through an Oil Red O staining assay in the livers of cohort 1 patients and found that lipid droplet formation was greater in group I (RORα-low/E47-high/AKR1A1-high) than in all other groups (Figure 6K). These results further verified the regulation of the RORα/E47/AKR1A1 axis in glucose and lipid reprogramming.

RORα, a circadian gene, is abundantly expressed in a variety of mammalian organs, including the testis, kidney, brain and liver(33). At present, it is widely recognized that RORα expression is reduced and plays an important role in multiple cancers through the regulation of immunity, circadian rhythm, energy metabolism and inflammation(30,34–36). Moreover, chemoresistance is a long-term challenge in the treatment of malignant tumors because of numerous complicated mechanisms, including transporter pumps, oncogenes, tumor suppressor genes, autophagy, epithelial–mesenchymal transition (EMT), mitochondrial changes, cancer stemness, energy metabolism, DNA damage and exosomes (37). Thus, we hypothesized that RORα can inhibit progression and chemoresistance through glucose and lipid reprogramming in GC. In the present study, we found that RORα deletion promoted GC progression by increasing glycolytic activity and lipid synthesis. In contrast, SR1078 reversed these effects. Moreover, single-cell sequencing analysis indicated that RORα may be involved in 5-FU chemoresistance, which was verified by a series of gain- and loss-of-function experiments both in vivo and in vitro. In addition, the RORα activator SR1078 synergistically inhibited GC progression in combination with 2-DG or atorvastatin. These results revealed a novel role for RORα in GC.

According to RNA transcriptome analysis, GSEA revealed that differentially expressed genes (DEGs) involved in the TCA cycle and fatty acid beta-oxidation are involved in the regulatory mechanism of RORα in GC. Next, Venn diagram analysis revealed 6 glycolysis- and lipid metabolism-associated DEGs. Subsequently, AKR1A1 exhibited the greatest change, indicating that AKR1A1 is a potential downstream gene. AKR1A1 belongs to the aldo-keto reductase (AKR) protein superfamily(38). This enzyme is dependent on NADPH and possesses broad substrate specificity for the transformation of sugars, lipid aldehydes and ketosteroids(39,40). Previous studies revealed that AKR1B1, AKR1B10, AKR1C1, AKR1C2 and AKR1C3 play important roles in diabetic complications, hepatocarcinogenesis, endometrial cancer, premenstrual syndrome, leukemia, prostate cancer and breast cancer(41). However, AKR1A1 has rarely been studied in tumors. Previous studies have shown that AKR1A1 can promote radioresistance in laryngeal cancer cells by suppressing p53 activation(42). Moreover, AKR1A1, a glycolysis-associated gene, is closely related to the prognosis of patients with lung adenocarcinoma and has proven valuable for the clinical management of patients (43). In the present study, we found that AKR1A1 expression is enhanced in RORα-KO GC cells. In contrast, SR1078 inhibited AKR1A1 expression. Moreover, a series of functional experiments both in vivo and in vitro demonstrated that RORα deletion promoted proliferation and chemoresistance through AKR1A1-induced glucose and lipid reprogramming in GC.

RORα, a transcription factor, tends to regulate the transcriptional activity of target genes(31). Thus, we preliminarily believe that this is the transcription factor of AKR1A1. However, bioinformatic tools for AnimalTFDB, TRANSFAC PATCH and TRANSFAC MATCH revealed that RORα is not a transcription factor of AKR1A1, and 2 candidates were predicted through Venn diagrams. Finally, JASPAR revealed that E47 may be a potential transcription factor for AKR1A1. In previous studies, E47, a classical transcription factor, was shown to participate in the transcriptional regulation of target genes in tumors. For instance, E47 can be recruited to the promoter to transactivate ΔNp63 gene expression in squamous cell carcinoma cells(44). Moreover, E47 can bind to the E-cadherin promoter directly to inhibit E-cadherin transcription, leading to epithelial-mesenchymal transition (EMT) and colon cancer metastasis(45). In the present study, we found that E47 overexpression significantly enhanced AKR1A1 mRNA and protein expression in GC cells. Moreover, E47 could be recruited to the AKR1A1 promoter and enhance its transcriptional activity. These results demonstrated that E47 is a transcription factor of AKR1A1. Next, a series of functional experiments both in vivo and in vitro demonstrated that AKR1A1-induced proliferation, chemoresistance, and glucose and lipid reprogramming are enhanced by E47 overexpression in GC.

The activation of eukaryotic transcription is a complicated process that is performed with the assistance of various elements and factors(46). Thus, we hypothesized that the regulation of AKR1A1 expression by RORα is mediated by E47. We found that the increase in AKR1A1 expression induced by RORα KO in GC cells is dependent on E47. Moreover, SR1078 attenuated AKR1A1 expression in GC cells overexpressing E47. This broken dependence indicates that its regulation is not completely dependent. Moreover, RORα deletion significantly enhanced E47 protein stability but did not affect E47 mRNA stability. These results indicated two relationships: 1) RORα may directly bind to E47 to form a heterodimer to affect its protein stability, and 2) RORα indirectly affects E47 protein stability. We ruled out the possibility that RORα directly binds to E47, suggesting that RORα indirectly regulates E47 protein stability by attenuating the wnt3a/β-catenin signaling pathway according to previous studies(32,33). Therefore, we found that E47 and β-catenin can form a heterodimer and colocalize in the nucleus in GC cells. Moreover. Chip assays revealed that β-catenin is recruited to the AKR1A1 promoter in GC cells. In addition, E47 protein stability could be reduced by siRNA-mediated inhibition of β-catenin in GC cells. Moreover, siRNA-β-catenin offset the ability of RORα-KO to promote E47 protein stability in GC cells. These results verified our hypothesis that RORα inhibits E47 protein stability by attenuating β-catenin expression. Subsequently, we further verified that Wnt3a and β-catenin expression is increased in RORα-KO GC cells and decreased in GC cells treated with SR1078, indicating that RORα attenuates the Wnt3a/β-catenin signaling pathway in GC cells.

Importantly, analysis of the timer 2.0 database demonstrated that the E47 and AKR1A1 expression levels in STAD tissues were greater than those in normal tissues. However, the difference in AKR1A1 expression levels was not significant. Moreover, the KM-plotter database showed that E47 and AKR1A1 are associated with poor prognosis, but there was no significant difference in the GEPIA database. Furthermore, the GEPIA database also showed AKR1A1 is positively correlated with E47 and negatively correlated with RORα. In our present study, AKR1A1 and E47 tended to be highly expressed and to have prognostic value, and AKR1A1 expression was associated with age, grade and TNM stage and could be an independent predictor in patients with cohort 1 patients. Regarding the relationships among RORα, E47 and AKR1A1, we found that RORα expression was negatively correlated with AKR1A1. GC patients with RORα-high/E47-low/AKR1A1-low expression pattern exhibited optimal DFS. In contrast, RORα-low/E47-high/AKR1A1-high exhibited the poorest DFS. These results further verified the abovementioned regulatory axis. Interestingly, in cohort 2, RORα, E47 and AKR1A1 expression alone or in combination was associated with the SUVmax and response to neoadjuvant therapy in GC patients. Moreover, the expression pattern of RORα-low/E47-high/AKR1A1-high revealed significant lipid metabolism in the liver of cohort 1 patients compared to all other patterns. Notably, this axis is involved in glucose and lipid reprogramming.

RORα deletion promoted proliferation and 5-FU chemoresistance by attenuating AKR1A1-induced glucose and lipid programming in GC. Moreover, SR1078, an RORα activator, significantly inhibited not only GC proliferation but also the synergistic effects of 2-DG and atorvastatin. Mechanistically, E47, a transcription factor of AKR1A1, is modulated by β-catenin through affecting its protein stability by binding to form a transcription heterodimer. Thus, RORα activated the wnt3a/β-signaling pathway and indirectly inhibited E47 protein stability, leading to the attenuation of AKR1A1 transcription. In summary, these findings provide promising insight into the regulatory mechanism of glycolysis and lipid metabolism, suggesting that RORα plays a therapeutic role in GC.

Acknowledgments

Not applicable.

Conflict of interest statement

Ethics approval and consent to participate This study was carried out following the World Medical Association’s Declaration of Helsinki and was approved by the Research Ethics Committee of The First Affiliated Hospital of Anhui Medical University. Written informed consent was obtained from all participants.

Consent for publication This paper has not been published or accepted for publication. It is not under consideration at another journal. The publications were approved by all the authors.

Competing interest The authors declare that there are no conflicts of interest.

Funding This study was funded by the Natural Science Foundation of Anhui Province (grant no. 2008085MH1294) and the University Collaboration Innovation Project of Anhui Province (grant no. GXXT-2022-065).

Author contributions XSW designed and drafted the manuscript. YWW collected the clinicopathological data and performed the analysis. MDC performed parts of the functional experiments. FXC assisted in handling the mice and performed the immunohistochemical staining. MDC performed the bioinformatic analysis. TS performed the supervision.

Data Availability Statement The data that support the findings of our study were stored and shared in Melendy Data (doi: 10.17632/pyfxvwm5cd.2) upon reasonable request.

Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA and Jemal A: Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 68, 2018.
Herrero R, Park JY and Forman D: The fight against gastric cancer - The IARC Working Group report. Best Pract Res Clin Gastroenterol 28, 2014.
Torre LA, Siegel RL, Ward EM and Jemal A: Global cancer incidence and mortality rates and trends - An update. Cancer Epidemiology Biomarkers and Prevention 25, 2016.
Kim H, Hwang Y, Sung H, et al.: Effectiveness of gastric cancer screening on gastric cancer incidence and mortality in a community-based prospective cohort. Cancer Res Treat 50, 2018.
Wang Z, Ma L, Meng Y, Fang J, Xu D and Lu Z: The interplay of the circadian clock and metabolic tumorigenesis. Trends Cell Biol, 2023.
Altman BJ, Hsieh AL, Sengupta A, et al.: MYC Disrupts the Circadian Clock and Metabolism in Cancer Cells. Cell Metab 22, 2015.
Wang MQ, Chen YR, Xu HW, et al.: HKDC1 upregulation promotes glycolysis and disease progression, and confers chemoresistance onto gastric cancer. Cancer Sci 114, 2023.
Cho H, Zhao X, Hatori M, et al.: Regulation of circadian behavior and metabolism by REV-ERB-α and REV-ERB-β. Nature, 2012.
Yang Y, Yang T, Zhao Z, et al.: Downregulation of BMAL1 by MiR-494-3p Promotes Hepatocellular Carcinoma Growth and Metastasis by Increasing GPAM-mediated Lipid Biosynthesis. Int J Biol Sci 18, 2022.
Chai C, Cox B, Yaish D, et al.: Agonist of RORA Attenuates Nonalcoholic Fatty Liver Progression in Mice via Upregulation of MicroRNA 122. Gastroenterology 159, 2020.
Ma X, Chen H, Li L, Yang F, Wu C and Tao K: CircGSK3B promotes RORA expression and suppresses gastric cancer progression through the prevention of EZH2 trans-inhibition. Journal of Experimental and Clinical Cancer Research 40, 2021.
Mao W, Xiong G, Wu Y, Wang C, Clair DS, Li J Da and Xu R: RORα suppresses cancer-associated inflammation by repressing respiratory complex I-dependent ROS generation. Int J Mol Sci 22, 2021.
Choi WS, Lee G, Song WH, et al.: The CH25H–CYP7B1–RORα axis of cholesterol metabolism regulates osteoarthritis. Nature 566, 2019.
Byun JK, Choi YK, Kang YN, et al.: Retinoic acid-related orphan receptor alpha reprograms glucose metabolism in glutamine-deficient hepatoma cells. Hepatology 61, 2015.
Mischel AM and Rosielle DA: Eastern Cooperative Oncology Group Performance Status #434. J Palliat Med 25, 2022.
Lee MH, Choi D, Park MJ and Lee MW: Gastric cancer: Imaging and staging with MDCT based on the 7th AJCC guidelines. Abdom Imaging 37, 2012.
Schwartz LH, Litière S, De Vries E, et al.: RECIST 1.1 - Update and clarification: From the RECIST committee. Eur J Cancer 62, 2016.
Wang FH, Zhang XT, Li YF, et al.: The Chinese Society of Clinical Oncology (CSCO): Clinical guidelines for the diagnosis and treatment of gastric cancer, 2021. Cancer Commun 41, 2021.
Glasgow RE, Hayman JA, Hochwald S, et al.: NCCN Gastric Cancer, Version 3.2016 Clinical Practice Guidelines in Oncology. JNCCN-Journal of the National Comprehensive Cancer Network 14, 2015.
Li S, Yu W, Xie F, et al.: Neoadjuvant therapy with immune checkpoint blockade, antiangiogenesis, and chemotherapy for locally advanced gastric cancer. Nat Commun 14, 2023.
Wang X, Jia R, Chen K, Wang J, Jiang K and Wang Z: RORα and REV-ERBα are Associated With Clinicopathological Parameters and are Independent Biomarkers of Prognosis in Gastric Cancer. Technol Cancer Res Treat 20, 2021.
Livak KJ and Schmittgen TD: Analysis of relative gene expression data using real-time quantitative PCR and the 2-ΔΔCT method. Methods, 2001.
Wang X, Zhang J, Wu Y, et al.: RORα inhibits gastric cancer proliferation through attenuating G6PD and PFKFB3 induced glycolytic activity. Cancer Cell Int 24, 2024.
Ianevski A, Giri AK and Aittokallio T: SynergyFinder 3.0: an interactive analysis and consensus interpretation of multidrug synergies across multiple samples. Nucleic Acids Res 50, 2022.
Hirz T, Mei S, Sarkar H, et al.: Dissecting the immune suppressive human prostate tumor microenvironment via integrated single-cell and spatial transcriptomic analyses. Nat Commun 14, 2023.
Pajak B, Siwiak E, Sołtyka M, et al.: 2-Deoxy-D-Glucose and its analogs: From diagnostic to therapeutic agents. Int J Mol Sci 21, 2020.
Zhang X, Coker OO, Chu ESH, et al.: Dietary cholesterol drives fatty liver-associated liver cancer by modulating gut microbiota and metabolites. Gut 70, 2021.
Love MI, Huber W and Anders S: Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15, 2014.
Anders S and Huber W: Differential expression of RNA-Seq data at the gene level–the DESeq package. DESq Manual, 2012.
Jetten AM: Retinoid-related orphan receptors (RORs): critical roles in development, immunity, circadian rhythm, and cellular metabolism. Nucl Recept Signal 7, 2009.
Popov VM, Wang C, Shirley LA, et al.: The functional significance of nuclear receptor acetylation. Steroids, 2007.
Miravet S, Piedra J, Miró F, Itarte E, De Herreros AG and Duñach M: The transcription factor Tcf-4 contains different binding sites for β-catenin and plakoglobin. Journal of Biological Chemistry 277, 2002.
Lee JM, Kim IS, Kim H, et al.: RORα Attenuates Wnt/β-Catenin Signaling by PKCα-Dependent Phosphorylation in Colon Cancer. Mol Cell 37, 2010.
Xiong G and Xu R: RORα Binds to E2F1 To Inhibit Cell Proliferation and Regulate Mammary Gland Branching Morphogenesis. Mol Cell Biol 34, 2014.
Sun X, Dongol S, Qiu C, et al.: MiR-652 promotes tumor proliferation and metastasis by targeting RORA in endometrial cancer. Molecular Cancer Research 16, 2018.
Taheri M, Omrani MD, Noroozi R, Ghafouri-Fard S and Sayad A: Retinoic acid-related orphan receptor alpha (RORA) variants and risk of breast cancer. Breast Dis 37, 2017.
Ou A, Alfred Yung WK and Majd N: Molecular mechanisms of treatment resistance in glioblastoma. Int J Mol Sci 22, 2021.
Jez JM, Bennett MJ, Schlegel BP, Lewis M and Penning TM: Comparative anatomy of the aldo-keto reductase superfamily. Biochemical Journal 326, 1997.
Bohren KM, Bullock B, Wermuth B and Gabbay KH: The aldo-keto reductase superfamily. cDNAs and deduced amino acid sequences of human aldehyde and aldose reductases. Journal of Biological Chemistry 264, 1989.
Hyndman D, Bauman DR, Heredia V V. and Penning TM: The aldo-keto reductase superfamily homepage. In: Chemico-Biological Interactions. vol. 143–144, 2003.
Penning TM: The aldo-keto reductases (AKRs): Overview. Chem Biol Interact 234, 2015.
Kim J-S, Chang JW, Park JK and Hwang S-G: Increased aldehyde reductase expression mediates acquired radioresistance of laryngeal cancer cells by modulating p53. Cancer Biol Ther 13, 2012.
Liu C, Li Y, Wei M, Zhao L, Yu Y and Li G: Identification of a novel glycolysis-related gene signature that can predict the survival of patients with lung adenocarcinoma. Cell Cycle 18, 2019.
Xu J, Li F, Gao Y, et al.: E47 upregulates ΔNp63α to promote growth of squamous cell carcinoma. Cell Death Dis 12, 2021.
Zhu G, Li X, Guo B, Ke Q, Dong M and Li F: PAK5-mediated E47 phosphorylation promotes epithelial–mesenchymal transition and metastasis of colon cancer. Oncogene 35, 2016.
Xie J, Libri D and Porrua O: Mechanisms of eukaryotic transcription termination at a glance. J Cell Sci 136, 2023.

No competing interests reported.

Download PDF

Reviewers invited by journal
19 Aug, 2024
Editor assigned by journal
07 Aug, 2024
Submission checks completed at journal
07 Aug, 2024
First submitted to journal
05 Aug, 2024

You are reading this latest preprint version

RORα inhibits proliferation and chemoresistance through AKR1A1-induced glucose and lipid reprogramming in gastric cancer

Status:

Version 1

Abstract

Figures

Introduction

Methods

Results

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1