The GSE43300 array dataset, based on the GPL7723 platform, was downloaded from the Gene Expression Omnibus (GEO; http://www.ncbi.nlm.nih.gov/geo/) database. These data were deposited by Chen et al (18). A total of two LPS-treated cell groups and two negative control groups were included in the present study. Raw data and annotation files were used for subsequent analysis.

The R statistical software (version 3.5.0) program in Bioconductor (version 3.8; http://www.bioconductor.org/) was used, as it provides a robust multiarray average algorithm to preprocess raw expression data. A total of 579 miRNA expression values were obtained and the limma package (19) in Bioconductor was used to analyze differential miRNA expression between the LPS-treated and control groups. The t-test in the limma package was used to calculate P-values for differential miRNA expression. Log₂-fold change ≥1.2 and P<0.05 were used as the cut-off criteria. The pheatmap package (version 1.0.12) was used to generate the heat map presented in Fig. 1.

The RAW264.7 murine macrophage leukemia cell line and 293T cell line were purchased from the American Type Culture Collection (Manassas, VA, USA) and cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum (Biological Industries USA, Cromwell, CT, USA) and antibiotics (100 U/ml penicillin and 100 µg/ml streptomycin) at 37°C with 5% CO₂. Lipofectamine^® 3000 (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) was used to transfect cells with miR-494-3p mimics (5′-UGAAACAUACACGGGAAACCUC-3′; 20 µM), miR-494-3p inhibitor (5′-GAGGUUUCCCGUGUAUGUUUCA-3′; 20 µM), negative control (5′-UUCUCCGAACGUGUCACGUTT-3′; 20 µM), inhibitor negative control (5′-CAGUACUUUUGUGUAGUACAA-3′; 20 µM), short interfering RNA (siRNA)-PTEN (sense, 5′-CAAAUCCAGAGGCUAGCAGUU-3′ and antisense, 5′-CUGCUAGCCUCUGGAUUUGTT-3′; 20 µM), siRNA negative control (sense, 5′-UUCUCCGAACGUGUCACGUTT-3′ and antisense, 5′-ACGUGACACGUUCGGAGAATT-3′; 20 µM; all from Shanghai GenePharma Co., Ltd., Shanghai China) and pmiRGLO Dual-Luciferase miRNA Target Expression Vector plasmids (500 ng/µl; Promega Corporation, Madison, WI, USA), according to the manufacturer's instructions. Following transfection for 48 h, cells were continuously stimulated with LPS at a concentration of 1 µg/ml for 24 h at 37°C in the presence of 5% CO₂. Total RNA and protein was subsequently extracted for reverse transcription-quantitative polymerase chain reaction (RT-qPCR) and western blotting analysis.

RAW264.7 cells were seeded in 6-well plates and incubated overnight, following which cells were treated with various concentrations of LPS (0, 1, 2, 3 and 6 µg/ml) for 24 h at 37°C in the presence of 5% CO₂. The expression levels of IL-1β and TNF were detected by RT-qPCR. TRIzol^® (Invitrogen; Thermo Fisher Scientific, Inc.) was used to isolate total RNA from RAW264.7 cells. cDNA was synthesized using a PrimeScript™ RT reagent kit (RR037A; Takara Biotechnology Co., Ltd., Dalian, China), according to the manufacturer's protocols. qPCR was performed using the SYBR-Green qPCR Master Mix (2X RealStar Green Power Mixture; A311-05; GenStar, Beijing, China) on an FTC-3000 thermal cycler (Funglyn Biotech Inc., Richmond Hill, ON, Canada). qPCR was performed as follows: 95°C for 10 min, followed by 40 cycles of 95°C for 15 sec and 60°C for 1 min. U6 and GAPDH were used as the internal reference genes. The 2^−ΔΔCq method was used to calculate the relative expression of genes (20). All primers used in the study are listed in Table I. The RNA extraction and RT-qPCR experiments were repeated three times.

Radioimmunoprecipitation assay buffer supplemented with protease and phosphatase inhibitors (Beyotime Institute of Biotechnology, Shanghai, China) was used to lyse RAW264.7 cells. Protein concentration was determined using a bicinchoninic acid kit (E162-01; GenStar). The nuclear protein was obtained using a Cytoplasmic and Nuclear Extract kit (KGP1100; Nanjing KeyGen Biotech Co., Ltd., Nanjing, China). Protein (20 µg/lane) was then added to 5X loading buffer, separated by 15% SDS-PAGE and electrophoretically transferred to a polyvinylidene difluoride membrane (EMD Millipore, Billerica, MA, USA). The membrane was were blocked with 5% non-fat milk at room temperature for 1 h and incubated with primary antibodies overnight at 4°C: β-actin (1:1,000; cat. no. 4970; Cell Signaling Technology, Inc., Danvers, MA, USA); lamina (1:1,000; 10298-1-AP; ProteinTech Group, Inc., Chicago, IL, USA); IL-1β (1:1,000; cat. no. 31202; Cell Signaling Technology, Inc.); TNF-α (1:2,000; 60291-1-Ig; ProteinTech Group, Inc.); PTEN (1:1,000; cat. no. 9552; Cell Signaling Technology, Inc.); AKT serine/threonine kinase 1 (Akt1; 1:1,000; cat. no. 2967; Cell Signaling Technology, Inc.), phosphorylated (p)-Akt1 (1:1,000; cat. no. 9018; Cell Signaling Technology, Inc.) and p65 (1:1,500; ab16502; Abcam, Cambridge, UK). The membranes were subsequently incubated for 1 h at room temperature with donkey anti-rabbit immunoglobulin G secondary antibodies (1:2000; cat. no. 711-005-152; Jackson ImmunoResearch, West Grove, PA, USA). The protein bands were visualized using Immobilon™ chemiluminescent substrate (WBKLS0050; EMD Millipore). β-actin and lamina were used as the loading controls. Blots were scanned using a Bio-Rad ChemiDoc™ XRS (Bio-Rad Laboratories, Inc., Hercules, CA, USA) and the band densities were quantified using Quantity One^® analysis software version 4.6.2 (Bio-Rad Laboratories, Inc.).

Putative miRNA target genes were predicted by application of TargetScan 7.2 (http://www.targetscan.org/mmu_72/), picTar (https://pictar.mdc-berlin.de/), PITA version 6 (https://genie.weismann.ac.il/pubs/mir07/mir07_data.html) and miRanda version 3.3a (http://www.microrna.org/microrna/home.do). Venn diagram analysis (Venny; version 2.1; http://bioinfogp.cnb.csic.es/tools/venny/) was used to identify the intersection between the four target predictions; PTEN was identified as a candidate mRNA associated with miR-494-3p. To validate that miR-494-3p can target PTEN, a reporter plasmid containing the 3′-UTR sequence of PTEN was generated. Primers were designed based on the National Center for Biotechnology Information reference sequence NM_008960.2, and specific primers were used for the amplification of wild-type and mutant 3′-UTR sequences from PTEN mRNA (Table I). The wild-type and mutated 3′-UTR fragment were cloned downstream of the luciferase reporter gene of the pmiRGLO-reporter vector, using the XhoI and SacI restriction sites. One day prior to transfection, 293T cells were seeded into a 96-well plate at a density of 1×10⁴ cells/well. Then, cells were co-transfected with miR-494-3p mimics and PTEN luciferase reporter constructs using Lipofectamine 3000. Blank pmiRGLO dual-luciferase was used as a positive control. After 24 h, a Dual-Luciferase^® Reporter Assay system (Promega Corporation) was used to measure reporter activity. The ratio of the Renilla fluorescence value to the firefly fluorescence value was calculated.

To determine the effect of miR-494-3p on the protein level of PTEN, immunocytochemistry was performed using the PTEN antibody. Cells were seeded 24 h prior to treatment. Following transfection with miR-494-3p mimics or negative controls for 48 h, RAW264.7 cells were stimulated with LPS (1 µg/ml) for 24 h at 37°C with 5% CO₂, and PTEN expression was determined. The transfected RAW264.7 cell lines were fixed with 4% formaldehyde for 30 min at room temperature, then incubated with 0.5% Triton X-100 for 10 min at room temperature. Cells were washed three times with PBS and blocked with goat serum (5%; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) for 1 h at room temperature. The cells were incubated with anti-PTEN antibody (1:100) at 4°C overnight, washed three times with PBS, and then incubated with goat anti-rabbit secondary antibody (Cy3 AffiniPure Goat Anti-Rabbit Immunoglobulin G; 1:1,000; E031640-01, EarthOx Life Sciences, Milbrae, CA, USA) for 1 h at 37°C in the dark. The cells were incubated with DAPI (1:10,000; cat. no. 4083; Cell Signaling Technology, Inc.) for 5 min at room temperature. Images of PTEN expression were acquired by confocal microscopy (magnification, ×10; A1; Nikon Corporation, Tokyo, Japan).

SPSS software version 13.0 (SPSS, Inc., Chicago, IL, USA) was used to perform statistical analyses. Data are presented as the mean ± standard deviation of at least three independent experiments. One-way analysis of variance followed by Bonferroni post hoc test was used to compare the means of multiple groups. Student's t-test was used to assess differences between two groups. P<0.05 was considered to indicate a statistically significant difference.

Following preprocessing of the raw expression data from the GSE43300 array dataset, 44 upregulated and three downregulated miRNAs were identified in the LPS-treated samples compared with in the control samples. The number of upregulated genes was higher than the number of downregulated genes. A hierarchical heatmap of the 47 differentially expressed miRNAs is shown in Fig. 1. Notably, miR-494-3p was upregulated by LPS in RAW264.7 cells.

It was previously demonstrated that RAW264.7 cells can release inflammatory cytokines in response to LPS stimulation (21). To verify this, RAW264.7 cells were treated with different concentrations of LPS, and IL-1β and TNF-α expression were measured. The IL-1β and TNF-α mRNA and protein expression levels were increased following LPS stimulation, with peak expression occurring with 1 µg/ml LPS (Fig. 2A and B). To determine whether LPS treatment would affect miR-494-3p expression in RAW264.7 cells, RT-qPCR was performed. As shown in Fig. 2C, peak expression of miR-494-3p occurred following stimulation with 1 µg/ml LPS. A previous study also selected 1 µg/ml LPS to induce inflammatory responses in RAW264.7 cells (22). The expression miR-494-3p, IL-1β and TNF-α following LPS treatment (1 µg/ml) was upregulated; however, the expression of all the molecules notably decreased following stimulation with >2 µg/ml LPS compared with 1 µg/ml LPS. This may be due to high concentrations of LPS inducing apoptosis in cells instead of inflammatory responses (23). Further investigation into the underlying mechanisms is required. The results indicated that miR-494-3p expression may be closely associated with LPS-induced inflammatory responses in RAW264.7 cells.

To explore the regulatory function of miR-494-3p during LPS-induced inflammation, RAW264.7 cells were transfected with miR-494-3p mimics or negative control (Fig. 2D), and RT-qPCR was performed to measure IL-1β and TNF-α mRNA levels. Overexpression of miR-494-3p downregulated IL-1β and TNF-α expression levels in LPS-treated RAW264.7 cells (Fig. 2E and F). To further confirm these results, an inhibitor was used to suppress miR-494-3p expression (Fig. 2G), and the results demonstrated that the protein expression levels of IL-1β and TNF-α were increased in RAW264.7 cells (Fig. 2H and I). These results suggested that miR-494-3p may regulate proinflammatory cytokine production via a negative feedback loop in RAW264.7 cells.

Next, to explore the molecular function of miR-494-3p in suppressing inflammatory responses, bioinformatics analysis was used to identify potential target genes. Putative miRNA target genes were identified by employing TargetScan, picTar, PITA and miRanda. The miRNA databases predicted a seed match for miR-494-3p in the 3′-UTR region of PTEN (Fig. 3A and B). Sequences from the predicted miR-494-3p target site in the 3′-UTR of PTEN, and mutant variants, were successfully cloned into the pmiRGLO vector. RAW264.7 cells were co-transfected with miR-494-3p mimics and the pmiRGLO vector. Transfection with miR-494-3p mimics inhibited the luciferase activity of pmiRGLO-PTEN-wild-type, but not that of pmiRGLO-PTEN-mutant (Fig. 3C). This indicated that miR-494-3p bound specifically to the PTEN 3′-UTR. Confocal microscopy and western blotting was used to determine whether miR-494-3p inhibited the protein expression levels of PTEN in RAW264.7 cells. PTEN protein expression levels were decreased in RAW264.7 cells transfected with miR-494-3p mimics. Conversely, it was increased with a miR-494-3p inhibitor (Fig. 3D).

TLR4, a member of the TLR family, interacts all four Toll-interleukin receptor domain-containing adaptor proteins to induce inflammatory responses; stimulation of TLR activates the NF-κB pathway (24). Based on a previous study, Akt1 activation is enhanced in PTEN-/-cells (25). This suggests that PTEN might augment TLR4-induced inflammatory responses by suppressing Akt1 activation (17). In addition, NF-κB pathways have been demonstrated to enhance LPS-induced inflammatory responses (26). Using miR-494-3p or negative control mimics to transfect RAW264.7 cells, it was revealed that miR-494-3p enhanced LPS-induced p-Akt expression, and suppressed the nuclear expression of its downstream molecule, p65 (Fig. 3E and G), which may induce decreased levels of the inflammatory mediators, IL-1β and TNF-α. Taken together, these results indicated that miR-494-3p enhanced Akt1 phosphorylation and suppressed the NF-κB pathway.

To determine the role of PTEN in miR-494-3p-mediated inhibition of LPS-induced inflammatory responses, RAW264.7 cells were transfected with PTEN siRNA or control siRNA. Western blotting was performed to measure the siRNA efficiency; PTEN siRNA significantly reduced the expression of PTEN (Fig. 3F). Akt1 phosphorylation was upregulated in RAW264.7 cells following transfection with PTEN siRNA compared with control siRNA (Fig. 3F and H). The protein expression levels of proinflammatory cytokines and p65 were also measured (Fig. 3F). The results suggested that miR494-3p targets PTEN and suppresses LPS-induced inflammation via the Akt/NF-κB pathway.