CircRNA LDLR promotes proliferation and aerobic glycolysis of gastric cancer cells by targeting CHD1 with miR-449b-5p

Background/aim Circular RNAs can serve as detection biomarkers and therapeutic targets for tumors. Our study aimed to elucidate the mechanisms associated with circRNA LDLR (circLDLR) in gastric cancer (GC) proliferation and aerobic glycolysis. Materials and methods Expression signatures of circLDLR, miR-449b-5p, and CHD1 were examined in GC samples using quantitative PCR. Proliferation ability of MKN-45 cells was assessed via CCK-8 and EdU assays, and cell apoptosis was measured by flow cytometry. Glucose uptake, lactate production, ATP/ADP ratios, and NAD+/NADH ratios in cell supernatants were quantified to evaluate aerobic glycolysis. Subcellular isolation assay, quantitative PCR, immunoblot analysis, RNA immunoprecipitation (RIP), and dual luciferase reporter assay were employed to investigate the relationship between genes. Results Expression of circLDLR and CHD1 was elevated, while miR-449b-5p expression decreased in GC. Functionally, overexpression of circLDLR enhanced proliferation and aerobic glycolysis and hampered apoptosis of MKN-45 cells. However, upregulation of miR-449b-5p or downregulation of CHD1 reversed these effects. CircLDLR acted as an miRNA spongeand regulated the expression of miR-449b-5p, thereby affecting CHD1 and accelerating GC malignant progression. Conclusion CircLDLR drives the proliferation and aerobic glycolysis of GC cells by targeting CHD1 with miR-449b-5p, which is an ideal potential target for early diagnosis and clinical treatment of GC.


Introduction
According to a 2020 CANCER TODAY report (https:// gco.iarc.fr/today),gastric cancer (GC) ranks fifth in terms of prevalence and fourth in terms of death rate globally.Since advanced cases of GC are primarily diagnosed, there is a critical need for diagnostic biomarkers that are highly efficient and reliable to improve prognosis and the 5-year survival rate (Liu et al., 2010;Zhang et al., 2020).Under aerobic conditions, glucose is often converted into lactic acid, a process known as the Warburg effect, which is the main metabolic mode to favor tumor cell survival (Vasudevan et al., 2007;Vander Heiden et al., 2009;Liu et al., 2018).It is generally believed that Warburg effect is related to poor prognosis and chemotherapy resistance in GC (Zeng et al., 2018).However, the precise pathogenesis of GC and the specific mechanism of action between GC and Warburg effect remain unclear.Therefore, exploring the molecular mechanisms of GC is advantageous for a niche to develop biomarkers for early diagnosis and potential therapeutic targets.
miRNAs can not only form a complex biological process regulation network with ncRNAs but also directly bind to target mRNAs to negatively regulate gene expression at the transcriptional level (Sexton et al., 2020).Based on this, the downstream miRNA of circLDLR and target genes of miRNA, specifically miR-449b-5p and CHD1, were screened.miR-449b-5p is recognized as a tumor suppressor that regulates tumor proliferation, metastasis, and invasion (Garber, 2004;Chen et al., 2020;Huang et al., 2020).CHD1, a regulator in maintaining genome stability, promotes DNA transcription by keeping DNA in an open and transcriptionally active state through ATP-dependent assembly, transfer, and removal of nucleosomes from DNA (Li et al., 2020;Qiu et al., 2022).It has been emphasized that CHD1 degradation suppresses glioma cell proliferation and glycolysis (Jeck et al., 2013).
The current study explored and analyzed the biological function of circLDLR in GC at the molecular level and defined the interlink between circLDLR, miR-449b-5p, and CHD1 to develop a theoretical reference to control GC progression.

Clinical samples
Forty-three pairs of tumor specimens and adjacent normal tissue samples were collected from GC patients, and all tissues were rapidly cryopreserved in liquid nitrogen.Written informed consent was obtained from each patient.This study was approved by the Institutional Review Board of Affiliated Hospital of Traditional Chinese Medicine of Xinjiang Medical University.

Quantitative PCR
To generate total RNA from tissues and cells, TRIzol reagent (Invitrogen) was applied.Subsequently, RNA samples were checked by Nanodrop 2000 (Thermo Fisher Scientific) to determine the purity and concentration.Following this, 1.5 μg of mRNA and circRNA were reverse transcribed into cDNA using SuperScriptTM III Reverse Transcriptase (Invitrogen) and subjected to PCR detection using SYBR Green master mix (CloudSeq).miRNA amplification was carried out using TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems) and TaqMan Universal Master Mix II (Applied Biosystems), followed by PCR detection in the Applied Biosystems AB7500 Real-Time PCR system.To calculate gene expression which is normalized to U6 (for miR-449b-5p) or GAPDH (for circLDLR and CHD1), the 2 −ΔΔCT method was employed.Three biological replicates were performed for each experiment.The primers are shown in Table 1.

RNAse R treatment
After the total RNA (10 μg) of MKN-45 cells was coincubated with RNAse R (3 U/g, Epicentre Technologies, USA) at 37°C for 30 min, the stability of circLDLR and LDLR mRNA was detected by quantitative PCR.2.6.Actinomycin D assay MKN-45 cells on the 6-well plates (5 × 10 5 cells/well) were treated with 2 μg/mL of actinomycin D (MCE, HY-17559), and detected at 0, 4, 8, 12, and 24 h, respectively.Linear and circular LDLR mRNA content was analyzed by quantitative PCR and normalized to the measured value at 0 h.

Subcellular isolation assay
A nucleoplasmic fractionation assay was performed using PARIS Kit (Invitrogen).MKN-45 cells (1 × 10 6 ) were collected in a cell separation buffer and centrifuged to separate the nuclei and cytoplasm.Gene expression was then measured using quantitative PCR, with 18S serving as the cytoplasmic reference and U6 as the nuclear reference.

Immunoblot analysis
Cells were lysed by adding lysis buffer (Beyotime, China) for 20 min on ice, and the protein concentration was measured using the Bradford assay (Bio-Rad, USA).Next, proteins subjected to 15% SDS-PAGE were loaded onto PVDF membranes, which were then blocked with 5% nonfat milk powder for 1 h and reacted with primary antibodies GAPDH (2118, CST, USA) and CHD1 (4351, CST, USA) overnight at 4 °C.After TBST washing, the cells were incubated with the corresponding horseradish peroxidase-conjugated secondary antibody (CST, USA) for 1 h at 37 °C and visualized using an enhanced ECL chemiluminescence kit (ultrassignal, China).2.9.CCK-8 MKN-45 cell suspension (100 μL) was placed in 96-well plate at 3 × 10 3 cells/well to calculate cell proliferation at 0, 24, 48, and 72 h.At each time point, 10 μL of CCK-8 solution (Dojindo, Japan) was added and incubated at 37 °C for 2 h.Absorbance at 450 nm was then measured on a microplate reader (Bio-Rad).

EdU experiment
The EdU assay was performed using BeyoClick EdU Cell Proliferation Kit and Alexa Fluor 555 (Beyotime).MKN-45 cells (1 × 10 5 ) were covered over 15 mm dishes overnight, treated with a 10-μM EdU solution for 2 h, fixed with 4% paraformaldehyde, permeated with 0.3% Triton X-100, and washed with 3% BSA.Finally, the cells were treated with Click Additive Solution and Hoechst 33342 and viewed under a confocal laser scanning microscope (Olympus FLUOVIEW FV1000).

Glucose uptake and lactate production
Glucose uptake analysis was performed using a glucose detection kit (BioVision, Milpitas, CA, USA).MKN-45 cells were incubated in 6-well plates at a density of 1 × 10 6 / well for 48 h.Glucose uptake = (0-h concentration − 48-h concentration)/protein concentration.
Lactate content in the culture medium was measured using a lactate detection kit (BioVision) and normalized to protein concentrations.2.12.ATP/ADP ratio ATP/ADP was measured using ApoSENSOR ADP/ATP ratio assay kit (BioVision), and luminescence was measured at SpectraMax (Molecular Devices, USA).MKN-45 cells (1 × 10 4 ) were mixed with the nucleotide release buffer, followed by 1 μL of ATP monitoring enzyme.Luminescence readings at 1 min (Data A) and 10 min (Data B) were recorded.Data C was determined after adding ADP converting enzyme.The ATP/ADP = Data A/(Data C − Data B).2.13 NAD + /NADH ratio NAD + /NADH was determined using an EnzyLight NAD + / NADH ratio assay kit (Bioassay Systems, USA).MKN-45 cells were resuspended with 100 μL of NAD + extraction buffer to determine NAD + .For NADH determination, 100 μL of NADH extraction buffer was added.After heating at 60°C for 5 min, assay buffer (20 μL) was added to the extract, followed by centrifugation and addition of working reagent to the resulting supernatant.Absorbance at 565 nm was read at 0 and 15 min.

Apoptosis assay
The apoptosis rate was detected using the Annexin V-FITC/ PI Apoptosis Detection Kit (Solarbio, CA1020).MKN-45 cells were washed with precooled PBS and resuspended in 1 mL of 1X binding buffer.The cell suspension (100 μL, 1 × 10 6 cells/mL) was mixed with 5 μL of Annexin V-FITC for 10 min, followed by 5 μL of propidium iodide for 5 min.Data were acquired on a FACScan flow cytometer (BD Biosciences, USA).

RNA binding protein immunoprecipitation (RIP) experiment
Cell lysates were prepared using RNA immunoprecipitation lysis buffer supplemented with protease inhibitors and RNAse inhibitors, and then mixed with magnetic beadconjugated Ago2 or anti-IgG for 6 h at 4 °C.After elution, the immunoprecipitates were purified and analyzed by quantitative PCR.The necessary materials for this procedure were obtained from the EZ-Magna RIP kit (Millipore).

Xenotransplantation and immunohistochemistry
The ).At the end of week 4, euthanasia was performed to harvest the tumors.The tumors were fixed with 10% formalin and prepared into paraffin sections (5 μm) for immunohistochemical analysis of Ki-67 (ab15580, Abcam).

Data analysis
The data are presented as mean ± standard deviation (SD).Student's t-test was used for comparisons between two groups, while one-way analysis of variance (ANOVA) was employed for comparisons among multiple groups.Each experiment was conducted with at least three biological replicates.A p-value <0.05 was considered statistically significant.

CircLDLR expression signature in GC
Quantitative PCR analysis demonstrated higher levels of circLDLR in GC tissues compared to normal tissues (Figure 1A), as well as in AGS and MKN-45 cells compared to GES-1 cells (Figure 1B).The correlation between circLDLR and clinicopathological features of GC patients was clarified.The results revealed that high expression of circLDLR was correlated with advanced tumor stage and poor differentiation of TNM (Table 2).Using the bioinformatics website circBank (http:// www.circbank.cn/),it was determined that circLDLR has a predicted length of 295nt and originates from exons 13 and 14 of the LDLR gene (Figure 1C).Testing the stability of circLDLR using RNAse R and actinomycin D assays revealed that circLDLR was resistant to RNAse R digestion and exhibited a longer half-life compared to linear RNA (Figures 1D and 1E).In addition, subcellular isolation experiments confirmed the main localization of circLDLR in the cytoplasm of MKN-45 cells (Figure 1F).

circLDLR induces proliferation and aerobic glycolysis of GC cells
circLDLR gain-and loss-of-function assays were conducted by transfecting circLDLR-oe and si-circLDLR into MKN-45 cells, respectively, and quantitative PCR validated that circLDLR in the cells was stably upregulated or downregulated (Figure 2A).CCK-8 assay and EdU assay manifested that silencing circLDLR was suppressive for the proliferative activity of .Moreover, flow cytometry revealed that depletion of circLDLR promoted apoptosis in MKN-45 cells (Figure 2D).In addition, circLDLR deficiency in MKN-45 cells lowered glucose uptake, lactate production, ATP/ADP ratio, and increased NAD + /NADH ratio (Figures 2E-2H).Consistent with the above results, when circLDLR was overexpressed, cell proliferation and aerobic glycolysis were enhanced, while apoptosis was disrupted.

Tumor growth mediated by circLDLR in vivo
After xenotransplantation with MKN-45 cells, tumors developed in nude mice.Interestingly, circLDLR knockdown impaired the tumorigenic ability of MKN-45 cells, as evidenced by reductions in tumor volume and weight (Figures 7A-7C).Additionally, circLDLR knockdown led to decreased expression of CHD1 protein (Figure 7D) and reduced levels of Ki-67 in the tumors (Figure 7E).

Discussion
Due to the complex and heterogeneous nature of GC, progress has been limited to control tumor development.
CircRNAs are widely and abnormally expressed in tumors, including GC, and indicate an association with tumor progression (Memczak et al., 2013;Zheng et al., 2016;Wang et al., 2019).Particularly, the injection of synthetic circRNA into patients may be an effective approach for future cancer treatment (Wang et al., 2022).In pursuit of potential regulators of GC, the current study evaluated circLDLR on proliferation, apoptosis, and aerobic glycolysis of GC, and eventually confirmed that circLDLR/ miR-449b-5p/CHD1 feedback aggravates GC malignancy.
CircLDLR is transcriptively processed from its parent LDLR and is involved in ovarian steroidogenesis to mediate follicular dysplasia in polycystic ovary syndrome.Furthermore, a recent report has mentioned that abundant circLDLR is closely related to the malignant progression of colorectal cancer cells.Based on this, the current study focused on circLDLR-related mechanisms in GC.It was found that circLDLR was stably and highly expressed in GC, mainly in the cytoplasm of GC cells.Furthermore, silencing circLDLR inhibited GC tumor growth; overexpressing circLDLR stimulated the proliferative and antiapoptotic abilities of MKN-45 cells while silencing circLDLR had the opposite effect.Considering that aerobic glycolysis is a hallmark in the development of GC (Luo et al., 2020;Pu et al., 2020), aerobic glycolysis-related indicators were tested, finally confirming the strengthening effect of circLDLR overexpression on aerobic glycolysis and the inhibitory effect of circLDLR knockdown.
CircRNAs can participate in biological processes through various mechanisms (Holdt et al., 2016;Huang et al., 2019).Among them, the main biological function of circRNA in tumors is to promote miRNA degradation by binding with miRNA (Meng et al., 2022).At present, studies on circLDLR have confirmed that it affects papillary thyroid cancer and colorectal cancer as ceRNA sponge miRNA.Similarly, the current study confirmed the targeting relationship between circLDLR and miR-449b-5.It has been discussed that miR-449b-5p blocks the malignant behaviors of cancer cells, thereby regulating tumor progression in various cancers including glioma, colorectal cancer, and lung adenocarcinoma (Hall and Georgel, 2007;Marfella et al., 2007;Jiang et al., 2021).In the setting of GC, the current study measured miR-449b-5p downregulation in tumor cells and identified that miR-449b-5p upregulation counteracted proliferation, apoptosis, and aerobic glycolysis mediated by circLDLR overexpression.
Subsequently, CHD1, an upregulated protein in GC, was screened out as a target of miR-449b-5p.Classified as a member of the CHD family, CHD1 is involved in transcription, replication, recombination, and DNA repair processes (Salzman et al., 2013;Kristensen et al., 2019).In tumor progression, CHD1 has been considered to accelerate glioma growth in the network of circRNA and miRNA.The current study elaborated that CHD1 was coregulated by circLDLR and miR-449b-5p, and furthermore, CHD1 deficiency attenuated the tumorpromoting effects induced by circLDLR in GC.
Despite these insights, there are limitations to be addressed in subsequent experiments.Future studies will delve into the molecular mechanisms by which circLDLR mediates the miR-449b-5p/CHD1 axis to affect the proliferation and aerobic glycolysis of GC cells, including its interactions with glycolytic genes and signaling pathways.Moreover, additional clinical samples should be collected for verification and experimental purposes.

Conclusion
CircLDLR in GC can target CHD1 through miR-449b-5p, and then participates in the proliferation and aerobic glycolysis of GC cells, which provides a new target and reference for the clinical treatment of GC patients.

Figure 1 .
Figure 1.CircLDLR expression signature in GC.A: quantitative PCR to detect circLDLR in GC tissues and adjacent normal tissues.B: quantitative PCR to detect circLDLR in human normal gastric mucosa cells GES-1 and human GC cells AGS and MKN-45.C: circLDLR gene information.D: RNAse R experiment to verify the ring structure of circLDLR.E: actinomycin D experiment to verify the stability of circLDLR.F: Subcellular isolation experiment to detect circLDLR localization in MKN-45 cells.Data were expressed as mean ± SD (repetitions = 3)

Figure 2 .
Figure 2. circLDLR induces proliferation and aerobic glycolysis of GC cells.MKN-45 cells were transfected with si-circLDLR or circlDLR-OE.A: quantitative PCR to detect circLDLR and LDLR.B: CCK-8 assay to detect the proliferation of MKN-45 cells.C: EdU assay to detect the proliferation ability of MKN-45 cells.D: Flow cytometry to detect apoptosis of MKN-45 cells.E: glucose uptake.F: lactic acid production.G: ATP/ADP ratio.H: NAD + /NADH ratio.Data were expressed as mean ± SD (repetitions = 3).

Figure 5 .
Figure 5. miR-449b-5p degrades CHD1.A: TNMplot.comdatabase to analyze the expression patterns of target genes in GC and normal tissues.B: quantitative PCR and immunoblot analysis to detect CHD1 in GC cells.C: luciferase reporter assay to detect the interaction between CHD1 and miR-449b-5p.D: immunoblot analysis to detect CHD1 in MKN-45 cells.Data were expressed as mean ± SD (repetitions = 3).

Figure 7 .
Figure 7. Tumor growth mediated by circLDLR in vivo.A: Representative image of the tumor.B: Growth curve of tumors.C: Tumor weight.D: immunoblot analysis to detect CHD1 in xenograft tumors.E: Immunohistochemical analysis to detect Ki-67 in xenograft tumors.Data were expressed as mean ± SD (n = 4).
animal experiment was approved by the Animal Ethics Committee of Affiliated Hospital of Traditional Chinese Medicine of Xinjiang Medical University.BALB/c female nude mice (6 weeks old; obtained from Shanghai Experimental Animal Research Center) were randomly divided into two groups, with 4 mice in each group.MKN-45 cells (5 × 10 6 cells) transfected with si-circLDLR were injected subcutaneously into the right abdominal muscle layer of the mice.Tumor diameter was measured every 7 days to calculate tumor volume (1/2 × tumor length × tumor width 2

Table 2 .
Association between circLDLR expression and clinicopathological characteristics of gastric cancer.