Introduction

Depression is a prevalent mental health disorder affecting millions worldwide1. Persistent sadness, hopelessness, and loss of interest in daily activities44. Another reason is the urgent need to find alternative therapies to overcome the limitations of the current treatment for depression45. Molecular studies on the potential of Longya Lilium as a therapeutic option could open new areas of development for effective treatment modalities46. Finally, the potential application of this study goes beyond the treatment of depression, as it could provide essential insights into the underlying molecular mechanisms of natural compounds that improve mental health47. Thus, this investigation may be critical for develo** new therapeutic options to improve the quality of life of depressed patients and advance mental health research.

Combined Chinese and Western medicine therapies have been widely used in human diseases due to their superior efficacy48. Our meta-analysis showed that gentian herb combined with fluoxetine had more pronounced antidepressant effects and fewer adverse effects49. Oral administration of Ziziphi spinosae lily powder suspension exhibited a specific antidepressant-like effect due to a reduction in tail suspension and swimming immobility time50. The results of a previous study showed that gentianoside, in combination with bailey saponin, had better antidepressant activity compared to bailey saponin or gentianoside alone in a rat CUMS model51. Fluoxetine, the first selective 5-hydroxytryptamine reuptake inhibitor, has been approved for the clinical treatment of depression because it improves depression-like behavior and monoamine neurotransmitter levels52. In addition, fluoxetine can effectively alleviate CUMS-induced depression-like behavior in mice by modulating the expression of lncRNAs in the hippocampus53. There is growing evidence that fluoxetine combination therapy has better antidepressant effects than treatment alone; for example, the standardized commercial ginseng extract G115® significantly enhanced the antidepressant-like effects of fluoxetine in FST54,55. Combined treatment with fluoxetine and 7,8-dihydroxyflavone significantly inhibited sucrose preference and depression-like behavior in FST, accompanied by a significant increase in autophagy, neuronal nuclei and Iba1 expression56.

In the current study, subsequent network pharmacological findings suggest that the antidepressant therapeutic effect of gentiana combined with fluoxetine is associated with inhibition of COX-2 expression57. Consistently, pharmacological inhibition of COX-2 showed promise in preventing increases in anxiety-like behaviors induced by acute stress58. COX-2 is associated with the inflammatory response, and its decreased levels indicate a marked suppression of the inflammatory response59. Neuroinflammation is a significant risk factor for the pathophysiology of depression60. reduced COX-2 expression in the hippocampus contributes to the amelioration of depression-like behaviors and hippocampal neuroinflammation induced by the stress of chronic social defeat61. Together, these data reveal the potential of COX-2 inhibition as a target for the treatment of depression because of its inhibitory effect on neuroinflammation62. The combination of gentian and fluoxetine represents a promising alternative to traditional monotherapy for depression63. The current investigation highlights the potential advantages of this therapeutic combination over either agent alone because of its ability to significantly reduce depression-like behavior and neuroinflammation in the preclinical setting64. This study also provides valuable insight into the underlying molecular mechanisms involved in this combination therapy65.

Previous research has demonstrated the significant role of PGE2 as a downstream factor of COX-237. Moreover, it has been shown to play a crucial role in the neuroprotective effect observed in SY5Y cells38. Hence, we hypothesize that Longya Lily and fluoxetine may exert their effects through the COX-2/PGE2 pathway in depression. In addition, IL22 is a pivotal inflammatory factor regulated by PGE2 in immune cells39. Therefore, we selected the PGE2/IL22 axis as the downstream signaling pathway for this study. Specifically, this therapy appears to reduce COX-2 expression and inactivate the PGE2/IL-22 pathway, leading to reduced microglial activation, inflammation, and neuronal damage66. The importance of this combination therapy lies in its ability to provide more effective and tolerable treatment options for depressed patients and improve their quality of life67. In addition, the study’s focus on natural compounds, such as gentian, highlights the potential for alternative therapies derived from traditional medicinal sources68. The potential applications of such integrative therapies are numerous and reflect the growing need for novel and effective treatments in the mental health field69. Ultimately, this investigation represents an important step toward develo** new treatment strategies for depression70.

Furthermore, further mechanistic investigations in the present study suggest that COX-2 activates microglia, thereby inducing neuroinflammation by activating the PGE2/IL-22 axis66. Indeed, COX-2 produces the pro-inflammatory factor PGE2 and the subsequent glial cell activation, leading to the core symptoms of depression in a rat model71. The synthesis of PGE2 is induced in the occurrence of neuroinflammation-associated depression72. Also, PGE2 mRNA expression has been elevated in CUMS-induced rat hippocampus and frontal cortex, and its inhibition can help ameliorate CUMS-induced depression-like behavior73. A previous study showed that PGE2 acts directly on type 3 innate lymphocytes and promotes their production IL-2274. Recently, paroxetine was reported to disrupt depression-like behavior by downregulating IL-22 expression in combination with chemotherapeutic agents75.

Interestingly, both lily and fluoxetine inhibited COX-2 expression76. Therefore, it can be concluded that the combination of Lobelia and fluoxetine significantly inhibited the neuroinflammatory response and ameliorated the resulting depression by inhibiting the COX-2/PGE2/IL-22 axis77. The COX-2/PGE2/IL-22 axis is an essential inflammatory pathway implicated in the pathogenesis of several neuropsychiatric disorders, including depression78. The current investigation provides essential insights into the role of this pathway in the antidepressant effects of gentian and fluoxetine combination therapy79. Studies have shown that this combination reduces COX-2 expression, thereby reducing microglia activation, inflammation and neuronal apoptosis through inactivation of the PGE2/IL-22 pathway78. This is important because the COX-2/PGE2/IL-22 axis is known to contribute to inflammation-associated neuronal damage and to promote the production of other pro-inflammatory mediators involved in the pathophysiology of depression. Therefore, understanding the molecular mechanisms underlying the activity of this inflammatory pathway in depression is crucial for the development of more effective treatments to reduce neuroinflammation and alleviate psychiatric symptoms80. The results of this study represent an essential contribution to this emerging field of research and may be vital in paving the way for novel therapeutic strategies targeting the COX-2/PGE2/IL-22 axis in depression and other neuropsychiatric disorders.

Overall, the present study confirmed the antidepressant effect of the combination of gentian and fluoxetine. This effect may be related to the inhibition of the COX-2/PGE2/IL-22 axis (Fig. 7). The results of this study suggest that the combination of gentian herb and fluoxetine may be a potential alternative antidepressant treatment with fewer side effects than conventional monotherapy. Further studies are needed to elucidate the optimal dose and duration of this agent and examine the safety, tolerability, and efficacy of this combination in humans. Clinical trials could use a combination of behavioral and neuroimaging measures to understand better the effects of Longya Lilium and fluoxetine on the brain and clinical symptoms of depression. At the same time, the current study provides some insight into the underlying mechanisms of this combination therapy; the involvement of multiple pathways and components of the immune system warrants further investigation. In addition, studies evaluating the effects of Gentiana and fluoxetine in different populations, such as those with treatment-resistant or severe forms of depression, are needed to determine the potential efficacy of this combination in more challenging situations. However, these findings provide promising evidence for develo** new treatment strategies to address the ongoing challenges in managing depression.

Fig. 7: Schematic illustration of the mechanism of Longya Lilium combined with fluoxetine in depression.
figure 7

Longya Lilium combined with fluoxetine can inhibit the PGE2/IL-22 axis by inhibiting the expression of COX-2 and reducing neuroinflammation, thus exerting an antidepressant therapeutic effect.

Full size image

Although this study has explored in-depth the mechanisms of the combination of Longya Lily and Fluoxetine in the treatment of depression from multiple perspectives, there are still some limitations. Firstly, our study primarily focuses on the role of the COX-2/PGE2/IL-22 axis in neuroinflammatory response in depression, while investigating other possible inflammatory regulatory pathways is not comprehensive enough. Secondly, this study mainly observes the activation state of microglial cells and their associated inflammatory response, without delving into the role of astrocytes. Astrocytes are essential supportive cells in the central nervous system, and they may play a crucial role in the occurrence and development of neuroinflammation and depression. The lack of detailed investigation on astrocytes may result in an incomplete understanding of the overall neuroinflammatory regulatory mechanisms. In the future, we will further explore the role of Longya Lily and Fluoxetine in other neuroinflammatory regulatory pathways, especially in investigating the role of astrocytes in this process. Simultaneously, in-depth analysis of various biomarkers associated with depression and the effects of drugs on these biomarkers will provide a theoretical basis for individualized treatment and better outcomes for patients with depression.

Methods

Ethics statement

The current study was performed with the approval of the Ethics Committee of Youjiang Medical University for Nationalities and performed strictly with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health.

Literature retrieval

From the inception to December 2020, PubMed, Embase, Web of Science, China National Knowledge Infrastructure (CNKI), and Wanfang databases were used to obtain relevant literature by retrieving depression, fluoxetine, Lilium, baihe, Lilii Bulbus; Lily Bulb, integrated Chinese and western medicine, and randomized controlled study. To screen the target documents accurately, we used the term NOT (review, animal experiment, meeting, Case report).

Literature screening and data extraction

The inclusion criteria were: (1) publicly published documents at home and abroad; (2) randomized controlled study; (3) patients with depression according to the diagnostic criteria for depression; (4) intervention measures including Longya Lilium combined with fluoxetine. The exclusion criteria were: (1) duplicated publications; (2) studies without sufficient data; (3) the included patients had taken antidepressants before. Two reviewers independently extracted information from the selected studies. Any disputes regarding data extraction were resolved by agreement among several investigators.

Meta-analysis

Meta-analysis was performed using R (v4.2.3), and the combined effect was evaluated using odds ratio (OR), weighted mean difference (WMD) and 95% confidence interval (CI). The Cochran Q test evaluated the heterogeneity test among studies. The Ph < 0.05 indicated that there is heterogeneity. I2 was used to evaluate the magnitude of heterogeneity. The I2 value ranged from 0 to 100%. The larger the I2 value was, the more pronounced the heterogeneity was. The p < 0.05 and I2 > 50% indicated significant heterogeneity among the studies, which the random effect model further analyzed. Otherwise, the fixed-effect model was used for analysis. Sensitivity analysis: the method of elimination one by one was used to evaluate the stability of the results. Stata software was used to conduct the Egger test to evaluate whether the included literature has publication bias and to clarify the reliability of the original analysis results.

Drug and disease target screening of network pharmacology

Previous studies have shown that saponins are the main component responsible for the antidepressant effects of Bulbus Lilii, a traditional Chinese medicinal herb81. Hence, this study focuses on screening target genes directly related to saponins. Active ingredients and targets of Lilium saponins were retrieved from the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP) database, and the targets related to the active ingredients of saponins were screened. The UniProtKB database retrieved the official gene symbol corresponding to the target, with the species set as “Homo sapiens”. Next, fluoxetine-related target genes were retrieved from the Comparative Toxicogenomics Database (CTD) database (Interaction Count ≥ 3). At the same time, the target genes related to depression were retrieved using the CTD database (Inference score ≥ 100). A Venn diagram of target genes related to Longya Lilium, fluoxetine, and depression was plotted using the Draw Venn Diagram tool.

Drug-component-target network construction and enrichment analysis

Cytoscape 3.6.0 software was applied to construct the Longya Lilium-fluoxetine-gene regulatory network. ClueGO plug-in in Cytoscape software was used to perform function enrichment analysis to construct a gene-function-pathway regulatory relationship network. Different functions were clustered according to different colors, and a pie chart of gene function enrichment was drawn to display the main Kyoto Encyclopedia of Genes and Genomes (KEGG) and gene ontology (GO) functions.

Visualization of single-cell sequencing data analysis results

The specific cell types involved in the regulation axis of COX-2/PGE2/IL-22 in brain tissue were retrieved from the scRNASeqDB database82, available at https://bioinfo.uth.edu/scrnaseqdb/. The distribution of COX-2 (PTGS2) in various types of brain tissue cells is shown in the following link: https://bioinfo.uth.edu/scrnaseqdb/index.php?r=site/geneView&id=PTGS2&set=GSE67835&csrt=9002072548059807797. Similarly, the distribution of PGE2 (PTGER2) in various types of brain tissue cells can be found at: https://bioinfo.uth.edu/scrnaseqdb/index.php?r=site/geneView&id=PTGER2&set=GSE67835&csrt=9002072548059807797. Lastly, the distribution of IL-22 (IL22) in different types of brain tissue cells can be viewed here: https://bioinfo.uth.edu/scrnaseqdb/index.php?r=site/geneView&id=IL22&set=GSE67835&csrt=9002072548059807797.

Depression mouse model establishment

120 SPF C57BL/6J male mice (aged 9–12 weeks old, weighing 18–21 g) were used to develop a model of depression based on the CUMS method. The stress paradigm consisted of the following stressors: swimming in ice water (5 min), food and water deprivation (24 h for each), tail pinch (1 min), shaking (once/s, 15 min), reversal of day and night and bondage (5 min each time) for a total of 24 days. One of these stressors was randomly arranged every day, and each stimulus was performed 21 times so that the mice could not predict the occurrence of the stimulus. Ten mice were used as normal controls, and 110 were stimulated to establish a depression model. After 5 weeks, the behavioral test was performed, and the serum sample was collected for biochemical testing.

The overall grou** was as follows: Control group (normal mice, no treatment), Model group (model mice, intravenous injection of empty vector), Fluoxetine group (model mice, oral gavage of Fluoxetine at a dose of 20 mg/kg), Lilium saponins group (model mice, oral gavage of Lilium saponins at a dose of 50 mg/kg), Lilium saponins+Fluoxetine group (model mice, oral gavage of Lilium saponins at a dose of 25 mg/kg and Fluoxetine at a dose of 10 mg/kg). The gavage volume was 0.3 mL, administered once daily for a continuous period of 21 days. In this study, we used Lilium saponins (Shaanxi Feiste Biotechnology Co., Ltd., product code: 904) to treat the animals without using the whole or parts of Lilium brownii. This study employed lily saponins (**

BV-2 (ATL03001) purchased from National Infrastructure of Cell Line Resource (Bei**g, China) were cultured in DMEM (SH30022.01B, Shanghai Suer Biotechnology Co., Ltd., Shanghai, China) supplemented with 10% FBS (SH30070.03, HyClone Laboratories, Logan, UT), 1% penicillin and streptomycin in a 5% CO2 incubator at 37 °C. Three siRNA sequences were designed for mouse-derived COX-2 CDS and IL-22 CDS and synthesized by Guangzhou RiboBio Co., Ltd. (Guangzhou, Guangdong, China). BV-2 cells were plated in a 24-well plate at 1 × 105 cells/healthy density and cultured with 1 mL DMEM at 37 °C with 5% CO2 for 24 h. Lipofectamine 2000 reagent (#11668027, Invitrogen) was used to transduce 50 nM negative control (NC), COX-2 small interfering RNA (siRNA, si), or IL-22 siRNA into BV-2 cells for 48 h. The silencing efficiency of COX-2 and IL-22 in BV-2 cells was detected by Western blot.

The lentivirus packaging system was constructed by LV5-GFP (lentivirus gene overexpression vector) and pSIH1-H1-copGFP (lentivirus short hairpin RNA (sh RNA) fluorescent expression vector gene silencing vector). sh-COX-2, sh-IL-22, overexpression (oe)-COX-2, oe-IL-22, sh-NC, and oe-NC were completed by Shanghai GenePharma Co. Ltd. (Shanghai, China). BV-2 cells were co-transfected with packaged lentivirus and target vectors. After incubation for 48, cell supernatant was collected. The lentivirus particles in the supernatant after centrifugation were filtered to detect the lentivirus titer. G418 was used to screen the stably transfected cells for over 2 weeks, and the mRNA or protein levels were identified by reverse transcription-quantitative polymerase chain reaction (RT-qPCR) or Western blot analysis.

The CUMS-induced mice were intravenously injected with lentivirus (10 μL) harboring sh-NC, sh-COX-2, sh-IL-22, Longya Lilium + fluoxetine + oe-NC, Longya Lilium + fluoxetine + oe-COX-2 and Longya Lilium + fluoxetine + oe-IL-22 with a final titer of 1 × 109 TU/mL. One day before CUMS, body weight, depression-like behaviors, and sucrose consumption were evaluated. On the second day, the mice underwent CUMS, and at the same time, were injected with lentiviruses and administrated with drugs via the tail vein. After 28 days, body weight, depression-like behaviors and sucrose consumption were assessed. The mice were anesthetized with sodium pentobarbital and euthanized; after that, the blood and brain tissue were collected, and the brain tissue was stored at −80 °C.

Behavioral assessments

In this study, the body weight of mice was measured on the day before the experiment and again on the 28th day. The weight gain of each group of mice was calculated by subtracting the weight on the first day of the experiment from the weight on the 28th day.

An opaque cube measuring 80 cm × 80 cm × 40 cm was constructed for the open field test with 25 equal-sized squares marked on the floor. Mice were placed in the center square, and their activity was recorded for 3 min. Both horizontal and vertical activity were recorded, with horizontal activity being measured as the number of squares crossed and vertical activity as the number of times both forepaws left the ground. Each animal was tested only once, and to ensure the reliability of the experiment, a double-masked procedure was employed whereby different researchers monitored the activity without knowledge of the group assignment of the mice88.

In the forced swim test, mice were placed individually in a water tank (80 cm high, 30 cm in diameter, and at a temperature of 25 °C) and forced to swim for 6 min. After 24 h, the mice were placed in the same tank for 5 min, and their swimming and immobility times were recorded for 3 min. Immobility time was defined as floating with minimal movement to keep the head above water36.

In the tail suspension test, mice were subjected to a series of parameters while hanging by their tails for 360 s with their heads facing downwards, with the tail being fixed to support 1.5–2 cm from the base. The immobility time was recorded as the time the mouse spent motionless while in this state. The measurement was taken using a single-masked procedure89.

In the sucrose preference test, mice were individually housed in a cage containing two bottles of sucrose solution (1%, w/v) for 24 h during an adaptation stage. During the second 24-h period, one bottle of sucrose solution was replaced with tap water. In the test phase, mice were deprived of food and water for 24 h and then placed in a cage containing two bottles, one containing 100 mL of 1% sucrose solution and the other 100 mL of tap water, for 3 h. Sucrose preference was defined as the sucrose consumption divided by the total consumption (sucrose solution and water) multiplied by 100%. Eight mice were used for each group36.

Hematoxylin-eosin (HE) staining

Paraffin-embedded sections of hippocampal tissues were prepared and cut into 4-μm-thick sections. The sections were stained with hematoxylin (C0007, Baoman Biotechnology Co., Ltd., Shanghai, China) for 10 min at room temperature. Next, the sections were counterstained with eosin at room temperature for 5–10 min. Before observation, the samples were mounted with neutral gum under an optical microscope (XSP-36, Boshida Optical Instrument Co., Ltd., Shenzhen, China).

Immunohistochemistry

The tissue sections were incubated with 3% H2O2 (84885, Sigma-Aldrich, St Louis, MO) at 37 °C for 30 min and boiled in 0.01 M citrate buffer at 95 °C for 20 min. The sections were then blocked with normal goat serum at 37 °C for 10 min and probed with primary antibody rabbit anti-mouse COX-2 (1:100, ab15191) at 4 °C overnight. The following day, the sections were re-probed with secondary antibody goat anti-rabbit immunoglobulin (IgG; 1:1000, ab6721, Abcam Inc., Cambridge, UK) at room temperature for 30 min, developed with diaminobenzidine (DAB; ab64238, Abcam), counterstained with hematoxylin and mounted. The primary antibody was substituted with PBS as NC. Five high-power fields were randomly selected from each section, with 100 cells counted in each field, and the rate of positive cells was calculated.

Enzyme-linked immunosorbent assay (ELISA)

Levels of IL-1β, IL-6, tumor necrosis factor-α (TNF-α) and interferon-gamma (IFN-γ) in peripheral blood of mice were measured using the IL-1β (ab197742, Abcam), IL-6 (ab222503, Abcam), TNF-α (ab208348, Abcam) and IFN-γ (ab100689, Abcam) ELISA kits. The optical density (OD) value was measured at 450 nm.

Immunofluorescence staining

The mouse hippocampal tissue sections were fixed with 4% paraformaldehyde, permeabilized in 0.3% Triton X-100, and blocked with 1% bovine serum albumin. After that, the sections were probed with primary antibody Iba1 (A12391, 1:100, Abclonal Technology, Inc.) overnight at 4 °C. Following PBS washing, the sections were re-probed with corresponding fluorescent secondary antibodies against AS011 and AS007 (1:100, Abclonal). Finally, the sections were observed under a confocal laser scanning microscope (FluoView FV10i, Olympus Optical Co., Ltd, Tokyo, Japan). Five sections were randomly selected from each mouse, and three visual fields were randomly selected from each section for a photograph. ImageJ software (National Institutes of Health, Bethesda, Maryland) was used for fluorescence intensity analysis.

TUNEL assay

TUNEL kit (Boehringer Mannheim, Germany) was used for TUNEL staining as the instructions described. The 20-µm-thick tissue cross-sections were prepared in a constant cold box cutter at −22 °C, and 10 sections were taken from each group for TUNEL staining. Under an optical microscope (Leica DM4 P, Shanghai Mei**g Electronics Co., Ltd., Shanghai, China), the dark particles indicated apoptotic cells. Five high-power fields were randomly selected from each section for observation and photographing, and 100 positive stained cells were calculated.

RNA isolation and quantitation

Total RNA was extracted with RNA Extraction Kit (D203-01, GenStar Biosolutions Co., Ltd., Bei**g, China) and then reversely transcribed as per the instructions of TaqMan MicroRNA Assays Reverse Transcription Primer (4366596, Thermo Fisher Scientific Inc., Waltham, MA). RT-qPCR was conducted using SYBR® Premix Ex TaqTM II kit (RR820A, Action-award Biological Technology Co., Ltd, Guangzhou, China) on the ABI PRISM® 7300 system (Prism® 7300, Shanghai Kunke Instrument Equipment Co., Ltd., Shanghai, China). Takara Biotechnology Ltd. (Dalian, China) designed and synthesized the primers, with sequences shown in Supplementary Table 3. The fold changes were calculated using relative quantification (the 2−ΔΔCt method), with GAPDH as a loading control.

Western blot analysis

Total protein was extracted from tissues with tissue lysis buffer containing phenylmethylsulphonyl fluoride (PMSF) (Boster Biological Technology Co., Ltd., Wuhan, China). The concentration of the total protein was determined by a bicinchoninic acid (BCA) kit (20201ES76, YEASEN Biotechnology Co., Ltd., Shanghai, China). The protein was separated by SDS-PAGE using 10% separation gel and spacer gel and transferred onto the nitrocellulose membrane. The membrane was then blocked using 5% skimmed milk powder at 4 °C overnight and underwent overnight incubation at 4 °C with the diluted primary rabbit antibodies (Abcam) against COX-2 (1:500, ab62331, Abcam,), PGE2 (1:200, #PA5-77694, Thermo Fisher Scientific Inc., Waltham, MA), IL-22 (1:1000, #PA5-115408, Thermo Fisher Scientific), TNF-α (1:1000, ab215188, Abcam), and IFN-γ (1:100, ab24780). The next day, the immunocomplexes on the membrane were visualized using enhanced chemiluminescence (ECL) reagent (Pierce Biotechnology Inc., Rockford, IL) at room temperature for 1 min and band intensities were quantified using ImageJ software, with GAPDH serving as a loading control. All original western blot images can be found in Supplementary Figs. 3–25.

Cell culture and grou**

BV-2 cells were cultured with high-glucose DMEM containing 10% FBS, 100 units/mL penicillin, and 100 μg/mL streptomycin in a 5% CO2 incubator at 37 °C, with the medium changed every 24 h. Cells were passaged after 48 h. After recovery, the growth status of BV-2 cells in the culture flask was observed under a microscope. When cells reached about 80–90% confluence, the medium was discarded, and 5 mL 0.01 M PBS was used to wash the culture flask twice to remove the residual medium. Next, 1 mL of 0.25% trypsin was added to the culture flask to cover the cells. The digestion was halted when the gap between the BV-2 cells under a microscope increased. The cells were gently pipetted until complete suspension and then passaged for another culture.

BV-2 cell suspension was seeded in a 96-well culture plate at a density of 1 × 104 cells/well (100 μL) and incubated in a cell incubator for 12 h to allow cell adherence to the wall. Cells were treated with PBS and served as the control. After 1 h, the cells were administered 1 mg/L lipopolysaccharide (LPS) for 24 h and further treated with Longya Lilium (10 μM) + fluoxetine (15 μM), and lentivirus containing si-NC, si-COX-2, si-COX-2 + oe-PGE2, oe-NC, oe-PGE2, and oe-PGE2 + si-IL-22.

MTT assay

Cells were seeded in a 96-well plate at a density of 5 × 104 cells/well, with 6 parallel wells set in each group. After reoxygenation, 20 μL of MTT solution (Sigma-Aldrich) was added to the cells and incubated in a 37 °C incubator for 4 h. Each well was supplemented with 150 μL of methyl sulfoxide (Sigma-Aldrich. Afterward, each well’s optical density (OD) value was measured at 450 nm using a microplate reader.

Flow cytometry

A flow cytometer assessed the cell apoptosis after 48 h of transfection. Following the instructions of Annexin-V-FITC Cell Apoptosis Detection Kit (CA1020, Bei**g Solarbio Science & Technology Co., Ltd., Bei**g, China), Annexin-V-FITC, PI, and HEPES buffer were prepared into Annexin-V-FITC/PI dye solution at a ratio of 1:2:50. A total of 1 × 106 cells were resuspended per 100 μL dye solution.

Statistical analysis

Statistical analysis was performed using SPSS 21.0 (IBM Corp., Armonk, NY). The measurement data were described as mean ± standard deviation. Data between the two groups were compared using an unpaired t-test. Data among multiple groups were assessed by one-way analysis of variance (ANOVA) or two-way ANOVA, followed by Tukey’s post hoc tests with corrections for multiple comparisons. A value of p < 0.05 was statistically significant.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.