Introduction
Gastric cancer (GC) is the fourth leading cause of cancer-associated death worldwide [
1]. However, the incidence rates in Europe and America are generally lower than those in Asia and Africa. The incidence of young adults exhibits a progressive rise in both high and low risk regions due to
Helicobacter Pylori infection, genetic risk factors and poor lifestyles [
1]. Unfortunately, most patients tend to suffer an advanced stage despite the development of diagnostic technology [
1,
2]. These points indicate that we still lack sufficient knowledge regarding the mechanism of GC. Therefore, it is critical and urgent to discover novel biomarkers and molecular pathways in GC.
Normal cells that do not need extra energy during oxidative phosphorylation exclude tumor cells. Aerobic glycolysis can obtain more energy to maintain rapid proliferation of tumor cells, which causes a reprogramming process of metabolism [
3]. Under this kind of requirement. Tumor cells activate various transport proteins and terminated limitation by key enzymes [
3]. In previous studies, abnormal glycolysis was shown to play an essential role in tumors through the modulation of clock genes [
4‐
6]. However, the exact clarification of the mechanism between clock genes and glycolysis remains poorly understood.
Retinoic acid-related orphan receptor α (RORα) is widely distributed in mammals to modulate the transcription of target genes in the nucleus. It tends to exhibit different specificities during the complicated process of physiology including lipid, cholesterol metabolism and immune system. [
7]. Thus, the dysregulation of RORα is associated with multiple cancers according to previous studies [
8‐
11]. However, few studies between RORα and aerobic glycolysis in tumor. Recent studies have demonstrated RORα reprograms glucose metabolism in glutamine-deficient hepatoma cells, but also inhibits aerobic glycolysis activity in hepatoma cells treated with the RORα activator SR1078 by reducing the expression of pyruvate dehydrogenase kinase 2 (PDK2), inhibiting the phosphorylation of pyruvate dehydrogenase and shifting pyruvate to complete oxidation [
12]. Moreover, Glucose 6 phosphate dehydrogenase (G6PD) and phosphofructokinase-2/fructose-2,6-bisphosphatase (PFKFB3) genes were inducers in downstream signal pathways to promote the progression of GC [
13‐
15]. Whether RORα is involved in glucose metabolism through the modulation of G6PD and PFKFB3 in GC is not clear.
Accordingly, we found RORα inhibits GC proliferation and glycolysis through a series of functional experiments in vitro and in vivo. More importantly, RORα was recruited to the promotors of G6PD and PFKFB3 genes to modulate its transcription, thereby, inhibiting GC proliferation and glycolysis. Moreover, the environment with high proliferation and high glucose modulated a negative feedback and inhibited RORα expression in GC. In addition, RORα deletion improved fluorouracil chemoresistance through inhibition of glucose uptake in GC. These findings provide a perspective on the role of RORα in GC.
Methods
Patients and specimens
GC patients (pathological diagnosis) were received circulating tumor cells (CTC), vascular endothelial growth factor (VEGF) examination and were collected paraffin-embedded sections in The First Affiliated Hospital of Anhui Medical University (Hefei, Anhui) from 2021 to 2023. The clinicopathological stage was assessed by chest, abdomen and pelvis enhanced CT or MRI according to 7th Edition of the International Union against Cancer tumour–node–metastasis (TNM) classification [
16]. The disease-free survival (DFS) time was defined as the time of recurrence locally, distant metastasis or up to 18 months. 18F-FDG PEC/CT examination, imaging diagnosis and standard uptake value (SUV) level were performed and were analyzed by three radiologists. The gender of male 116/71.6% and female 46/28, 4%. Age from 34 to 50 (21/13.0%), 51–60 (55, 14.0%) to 61–88 (86, 53.0%) and median was 63. The details of data collection was illustrated through the patient profile (Additional file
1: Fig. S1). All enrolled patients signed the informed consent which was approved by the human Ethics Committee of Anhui Medical University (20,180,344, Hefei, Anhui).
Cell culture and treatment
Human GC cell lines (AGS and MKN-74) and Mouse Forestomach Carcinoma (MFC) cell line were purchased from the Procell Life Science and Technology (Wuhan, China). These cells were cultured in RPMI 1640 medium (Procell Life Science and Technology, Wuhan, China) with 10% fetal bovine serum (Thermo Fisher Scientific, Waltham, MA, USA). Culture medium was changed every 2 days. Cells were pretreated with SR1078 (RORα activator, GC16392, GlpBio, Montclair, CA, USA), 3PO (3-(3-pyridinyl)-1-(4-pyridinyl)-2-propen-1-one, PFKFB3 inhibitor, MCE, USA), DHEA (Dehydroepiandrosterone, G6PD inhibitor, MCE, USA) and glucose for 24 h. Cells were pretreated with TGF-β1 (Transforming growth factor beta 1, Cayman chemical, USA) for 48 h. SR1078 was dissolved in phosphate buffer saline (PBS). 3PO, DHEA, TGF-β1 and fluorouracil (GC14466, GlpBio, Montclair, CA, USA) were dissolved in dimethyl sulfoxide (DMSO).
CRISPR-Cas9 gene deletion
Single guide RNA (sgRNA) oligonucleotides were cloned into LV-U6-spsgRNA(RORα)-CMV-SV40-NLS-spCas9-NLS-Flag-P2A-Puro-T2A-EGFP-WPRE (human) and packaged as lentivirus (BrainVTA, Wuhan, China). The alignment of the DNA sequences of human and mouse RORα genes showed 91% similarity according to NCBI. Herein, The AGS, MKN-74 and MFC cells were infected by virus expressing Cas9 and gRNA at 12 h. The media was resuspended with 2 μg/ml puromycin for 72 h and was verified by western blot to detect transfection efficiency. The RORα-KO sgRNA sequence (human) was GTAATCGACAGTGTTGGCAG.
Cells were seeded into 6-well (2000 cells/well) plates through repeating dilution and were cultured in incubator. The plates were collected to stain with crystal violet after 2 weeks. These colonies were counted to analyze the ability of proliferation.
CCK-8 assay
Cells were seeded into 96-well tissue microplates (5000 cells/well) to culture according to concentration and time gradient in incubator. The wells were incubated with Enhanced Cell Counting Kit-8 (CCK-8, 10 µl, C0042, Beyotime institute of biotechnology, Haimen, China) for 1 h. The Optical density (OD) values were measured at 450 nm using microplate reader (Thermo Scientific, Inc., USA) to represent the relative cell viability.
Quantitative real-time PCR (q-PCR)
Total RNA was isolated using TRIzol (Invitrogen, USA) according to the manufacturer’s instructions. The cDNA was generated using a Transcriptor first-strand cDNA synthesis kit (TaKaRa, Shiga, Japan) at 42 ℃ for 40 min and 85 ℃ for 5 min. The primers were showed in Table
1. The q-PCR was performed using the Thermo Biosysterm7500, 96 real-time PCR detection system with TaKaRa SYBR
® Green supermix (TaKaRa, Shiga, Japan) according to the manufacturer’s instructions. Every sample obtained a cycle threshold (Cq) value to determine the relative mRNA levels through 2
−ΔΔCt method [
17]. The β-actin was used as a control for normalization.
Table 1
Primer Sequences for q-PCR assay
E-cadherin | Forward | 5′–TGC CCA GAA AAT GAA AAA GG–3′ |
Reverse | 5′–GTG TAT GTG GCA ATG CGT TC–3′ |
N-cadherin | Forward | 5′–GGT GGA GGA GAA GAA GAC CAG–3′ |
Reverse | 5′–GGC ATC AGG CTC CAC AGT G–3′ |
Vimentin | Forward | 5′–GAG AAC TTT GCC GTT GAA GC–3′ |
Reverse | 5′–GCT TCC TGT AGG TGG CAA TC–3′ |
G6PD | Forward | 5′–AAA CGG TCG TAC ACT TCG GG–3′ |
Reverse | 5′–GGT AGT GGT CGT TGC GGT AG–3′ |
PFKFB3 | Forward | 5′–CAG TTG TGG CCT CCA ATA TC–3′ |
Reverse | 5′–GGC TTC ATA GCA ACT GAT CC–3′ |
β-actin | Forward | 5′–CAT GTA CGT TGC TAT CCA GGC–3′ |
Reverse | 5′–CTC CTT AAT GTC ACG CAC GAT–3′ |
Western blot
The protein lysate was obtained from cells or samples through lysis buffer dissolution, and was quantified by BCA Protein kit (Santa Cruz Biotechnology, USA). The target proteins were separated by 10% SDS-PAGE and transferred to polyvinylidene fluoride (PVDF) membrane. Subsequently, the membrane was blocked and was incubated with primary RORα (Rabbit, DF3196, 1:1000 dilution, Affinity Biosciences, Beijing, China), G6PD (Rabbit, GTX101218, 1:1000 dilution, GeneTex, USA), PFKFB3 (Rabbit, ab181861, 1:2000 dilution, Abcam, USA) and β-actin (Rabbit, 1:1000 dilution, Abcam, USA) antibodies for 24 h at 4 ℃, respectively. After washing three times with TBST. The membrane was incubated with conjugated second antibody for 2 h at room temperature. Finally, the bands were imaged by chemiluminescence system (Tanon, Shanghai, China).
Immunohistochemical staining
The paraffin-embedded sections were deparaffinized and hydrated with a series of xylene and different concentration of ethyl alcohol. The sections were incubated with 3% H
2O
2 for 30 min and then were antigen retrieved by microwave. The primary RORα (1:100 dilution), G6PD (1:1000 dilution), PFKFB3 (1:100 dilution), Ki-67 (Rabbit, 12202 T, 1:400 dilution, Cell Signaling Technology, USA) and Proliferating Cell Nuclear Antigen (PCNA, Rabbit, 13110 T, 1:4000 dilution, Cell Signaling Technology, USA) antibodies were employed to incubate for 2 h at room temperature, respectively. Subsequently, the sections were incubated with the conjugated second antibody for 30 min at room temperature. The DAB kit (ZSGB-BIO, OriGene Technologies, Beijing, China) and 20% hematoxylin were utilized to staining at room temperature. The calculation of relative protein expression levels, and definition of low and high expression levels were mentioned in our previous studies [
18].
Establishment of the GC proliferation mice model
MFC (1 × 107) cells were injected into the subcutaneous flank of 6–8 weeks-old mice (C57, Vital River Laboratory Animal Technology, Beijing, China). When the tumor volume reached 100 to 300 mm3. Partial mice were sacrificed to perform immunohistochemistry. Another part of mice treated with fluorouracil (100 mg/kg/w) subsequently through injecting into subcutaneous tissues around the tumor for 4 weeks. 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 (20,180,365).
Chromatin immunoprecipitation (Chip)-qPCR assay
The Chip assay was performed using Chip assay kit (P2078, Beyotime, China). The cells were crosslinked by 37% methanol and terminated by glycine solution (10X) at room temperature. After washing with PBS twice, cells were collected and were sufficiently lysed with SDS lysis buffer. The ultrasonic (VCX750, Sonics, USA) was performed to cut DNA fragments from 200 to 800 bp at 25% power, 4.5 s shock and 9 s clearance for 14 times. The samples were centrifuged and collected supernatant solution to dilute with dilution buffer. Then, protein A + G Agarose was added. After centrifuging the mixture. The new supernatant solution was collected and primary RORα (1 μg; Rabbit; ab278099; Abcam, USA) body was added to immune complex. After a series of centrifugation and washing. The precipitation was collected to perform the q-PCR assay.
Glycolysis assay
The extracellular acidification rate (ECAR) was performed by Seahorse XFe24 analyzer (Seahorse Bioscience, USA) according to the manufacturer’s guidelines. The sensors were immersed in calibrant to hydrate overnight before assay. Cells were seeded into 24 well plate, and the medium was added to a final volume of 250 µl. Next, cells were washed twice and the XF test medium was added to a final volume of 500 µl. Subsequently, The glucose (10 mM), oligomycin (1 μM), and 2-DG (100 mM) were added sequentially to measure the ECAR levels.
Statistical analysis
SPSS 19.0 and GraphPad Prism 10.0.0 software were performed to statistical analysis. Chi-square test was utilized to analyze the association of RORα, G6PD and PFKFB3 expression levels. Comparisons between different groups was performed using Pearson correlation analysis, t-test or two-way ANOVA. Survival analysis was performed by the Kaplan–Meier method and log-rank test. P value < 0.05 was considered as statistically significant.
Discussion
Aerobic glycolysis plays a significant role in the reprogramming tumor microenvironment and eventually results in abnormal proliferation, progression, invasion and metastasis [
3]. G6PD and PFKFB3 are key enzymes that participate in glucose metabolism [
13‐
15]. The present study revealed RORα inhibits GC proliferation through attenuating G6PD and PFKFB3 induced glycolytic activity. These findings indicated a novel interaction between RORα and glycolysis in GC.
Reportedly, RORα is one of circadian rhythm genes and is involved in multiple metabolic cycles in mammals, and its dysregulation results in disease and even cancer [
8,
10,
26]. In detail, RORα suppressed breast tumor invasion by inducing SEMA3F expression. RORα as a transcription regulator to mediate SEMA3F transcription [
8]. Moreover, RORα suppressed EMT of GC cells via Wnt/β-Catenin Pathway [
27]. In addition, RORα inhibited hepatocellular carcinoma proliferation, invasion and migration through downregulation of chemokine CXCL5 [
28]. Thus, RORα performed an antitumor function and was obtained verification. However, whether RORα modulates aerobic glycolysis to be involved in oncogenesis is still an unknow domain of exploration in GC. In our present study, we attempted to explore the association among RORα and glycolysis through bioinformatics analysis. The GSEA algorithm indicated RORα inhibits GC proliferation and glycolysis according to Reactome and Wikipathways database. Next, we collected abundant clinicopathological data from GC patients combined with a series of functional experiments in vitro and in vivo to verify this indication.
G6PD assists in the metabolism of glucose and is mainly involved in pentose phosphate pathway which induces the occurrence and progression of disease [
29]. Its metabolites including glyceraldehyde-3-phosphate and fructose-6-phosphate recycled glycolysis to obtain energy [
29]. Recent studies demonstrated the activation of G6PD regulates Warburg effect and affects GC proliferation [
15]. G6PD altered aerobic glycolysis and promoted tumor progression via pentose phosphate pathway [
30]. PFKFB3 as a glucose regulator and is associated with diabetes and cancer [
13,
14,
31,
32]. However, whether G6PD and PKFKB3 are downstream targets to induce glycolysis via RORα is not clear. Thus, we utilized TIMER 2.0 and found G6PD and PFKFB3 expression was increased in GC. Moreover, RORα expression levels revealed a negative correlation with G6PD and PFKFB3 expression levels in GC tissues. GC patients with low RORα and high G6PD or low RORα and high PFKFB3 expression patterns obtained a poorest DFS compared with other patterns. These results indicated a regulatory axis among RORα/G6PD/PFKFB3 in GC. To verify this speculation. We utilized their inhibitors DHEA and 3PO to offset the regulation of RORα in GC proliferation and glycolysis. These phenomena demonstrated RORα inhibits GC proliferation through attenuating G6PD and PFKFB3 induced glycolytic activity. Mechanically, RORα was recruited to the promoters of G6PD and PFKFB3, leading to the regulation of GC proliferation and glycolysis.
It has been reported that TGF-β1 plays an ambiguous role in tumors, where it can inhibit tumor development during the early stages while promoting metastatic spread as the disease progresses [
33]. To construct a high proliferation environment, we treated GC or MFC cells with TGF-β1 and verified a viable concentration, consistent with previous report that TGF-β1 induce GC progression [
34]. Moreover, we treated GC or MFC cells with glucose in concentration gradient to obtain an optimum environment for GC glycolysis according to previous study [
35]. Interestingly, these environments significantly inhibited RORα expression in vitro and in vivo. However, whether this negative feedback is also mediated by G6PD and PFKFB3 was not obtained a further exploration, even would appear other targets to mediate regulation.
Glycolysis enhanced chemoresistance is an ongoing and promising hotpot. In previous studies, the Fibrillin-1/VEGFR2/STAT2 signaling axis promoted cisplatin chemoresistance via modulating glycolysis in ovarian cancer cells [
36]. HKDC1 promoted cisplatin, oxaliplatin and fluorouracil chemoresistance through glycolysis in GC [
37]. Thus, inhibition of glucose uptake is a viable strategy for cancer therapy through weakening glycolysis-associated chemoresistance [
25,
38]. We found RORα deletion barely improve fluorouracil chemoresistance through inhibition of less glucose uptake, while achievement through inhibition of abundant glucose uptake in vitro and in vivo. Of note, when RORα-KO GC cells treated with high concentration fluorouracil, inhibition of less glucose uptake improved fluorouracil chemoresistance with the extension of time. In contrast, RORα-KO GC cells treated with low concentration fluorouracil did not shifting the situation of fluorouracil chemoresistance. These results indicated RORα deletion could improve fluorouracil chemoresistance through inhibition of glucose uptake in GC. Meanwhile, the improvement ability was associated with dependent concentration and time gradient.
However, a few limitations should be mentioned due to objective and subjective reasons. The absence of past medical and therapeutic history might disturb a logically rigorous among interactions of result in GC. To prevent excessive bias and obtain deeply stratified research, further demographic analysis was needed to expand the sample size, extend follow-up time and collect detailed clinicopathological data. Additionally, although we verified a hypothesis that RORα regulates GC proliferation through G6PD and PFKFB3 induced glycolysis, the downstream core molecules are still unclear. It is necessary to explore the integrated and distinct mechanism. Therefore, the limitations aforementioned opens new routes for future studies to further explore therapeutics of GC.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.