The Ficoll-Paque Plus was purchased from GE Healthcare (Chicago, IL, USA; cat. no. 17-1440-02). α-MEM was from Hyclone (GE Healthcare) and fetal bovine serum was from Gibco (Thermo Fisher Scientific, Inc., Waltham, MA, USA; cat. no. 10099141). Streptomycin was from Lonza Group, Ltd. (Basel, Switzerland; cat. no. 17-602E). Lipopolysaccharide (LPS) was from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany; cat. no. L6529). TRIzol (cat. no. 15596018), and Lipofectamine^® 2000 (cat. no. 11668019) were from Thermo Fisher Scientific, Inc. The PrimeScript™ RT reagent kit was from Takara Biotechnology Co., Ltd. (Dalian, China; cat. no. RR037A) and the SYBR Green Real-Time PCR Master mix was from Thermo Fisher Scientific, Inc., (cat. no. 4309155). miR-25 nucleotide fragments and primers were synthesized by Guangzhou RiboBio Co., Ltd., Guangzhou, China. The mouse anti-human HMGB-1 antibody was purchased from Cell Signaling Technology, Europe, B.V. (Leiden, The Netherlands; cat. no. 6893). Mouse anti-p-NF-κB p65 was purchased from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA; cat. no. sc-166748). Rabbit anti-GAPDH (cat. no. ab181602) and Lamin B1 (cat. no. ab133741) were purchased from Abcam (Cambridge, UK). Horseradish peroxidase-labeled anti-mouse (cat. no. 115-035-206) and anti-rabbit (cat. no. 323-005-021) secondary antibodies were purchased from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA, USA). Transwell chambers were purchased from EMD Millipore (Billerica, MA, USA; cat. no. PSET010R1). The pLUC Luciferase vector was purchased from Ambion; Thermo Fisher Scientific, Inc. The Dual-Luciferase^® Reporter Assay system was purchased from Promega Corporation (Madison, WI, USA; cat. no. E1910). The tumor necrosis factor α (TNF-α; cat. no. ELH-TNFa-1), interleukin 6 (IL-6; cat. no. ELH-IL6-1) and HMGB-1 (cat. no. ELH-HMGB1-1) ELISA detection kits were purchased from RayBiotech (Norcross, GA USA). Human recombinant M-CSF was purchased from R&D Systems, Inc. (Minneapolis, MN, USA; cat. no. 216-MC-025). Mouse anti-human CD68 was purchased from BD Pharmingen (BD Biosciences, Biosciences, Franklin Lakes, NJ, USA; cat. no. 562117). The BCA protein quantification kit was purchased from Beyotime Institute of Biotechnology (Haimen, China; cat. no. P0010).

A total of 39 patients with sepsis were admitted to the ICU of our hospital between July 2015 and April 2016. All patients met the American College of Chest Physicians/Society of Critical Care Medicine diagnostic criteria for sepsis (1992) (10). These criteria include definite or suspected infection, and (i) Body temperature >38°C or <36°C; (ii) Heart rate >90 beats/min; (iii) Breathing >20 times/min or pCO₂ <32 mmHg; (iii) White blood cell count >12×10⁹/l or <4×10⁹/l, or a proportion of immature rod-shaped cells >10%. Among the 39 cases of sepsis, 22 were male and 17 were female, aged 33–65 years (mean age 43.9±12.6 years). A total of 32 healthy subjects without signs of infection included 19 males and 13 females, aged 35–64 years (mean age, 44.6±11.8 years). There was no significant difference in age or sex ratio between the healthy and septic patients. All specimens were collected with the informed consent of all participants. The present study was reviewed and approved by Hefei Second People's Hospital Biomedical Ethics Committee (Hefei, China).

A total of 4 ml peripheral venous blood was collected from patients with sepsis after admission, or from healthy subjects in a fasting state, of which 2 ml was used to isolate peripheral blood mononuclear cells (PBMCs), and 2 ml was used to isolate serum. For serum isolation, blood was kept at room temperature for 1 h, then centrifuged for 10 min at 800 × g (21°C). The separated upper layer was then collected and stored at −80°C until further use.

A total of 2 ml blood was diluted in 2 ml PBS, and slowly dripped onto the surface of 5 ml Ficoll-Paque Plus. The mixture was centrifuged for 30 min at 300 × g at room temperature, then the cells in the upper white layer were transferred into another 15 ml centrifuge tube. PBS was added at a volume of 3:1 (PBS: cell solution) and mixed gently. The mixture was centrifuged for 5 min at 200 × g, the supernatant was discarded, and the cell pellet was resuspended in 1 ml ice-cold 1X PBS to obtain a PMBC cell precipitate.

PBMCs were placed in α-MEM medium containing 10% FBS and 1% streptomycin and cultured at 37°C in 5% CO₂. After 24 h, the culture supernatant was collected and non-adherent cells were removed. Medium containing 30 ng/ml M-CSF, 10% FBS and 1% streptomycin was replenished every 72 h for 1 week prior to detection of the macrophage marker, cluster of differentiation 68 (CD68).

Samples were prepared using a PrimeScript™ RT reagent kit according to the manufactures' instructions, (Takara Biotechnology Co., Ltd). The phenotype of macrophages was analyzed using an intracellular CD68 assay (cat. no. Y1/82A; BD Biosciences). Cells were fixed with 4% paraformaldehyde (room temperature, 20 min), permeabilized by FACS permeabilizing solution (BD PharMingen, San Diego, CA) and then stained with the aforementioned reagents for 30 min at room temperature. Flow cytometry analysis was performed using a FACS Aria II flow cytometer (BD Biosciences). Data were analyzed with FlowJo software (version 7.6.1; FlowJo LLC, Ashland, OR, USA).

The 293 cell genome was used as a template to amplify the HMGB1 3′-UTR full length fragment, or mutation-containing fragments, then cloned into the pLUC vector to transform DH5α competent cells (Thermo Fisher Scientific, Inc.). The efficient plasmids were sequenced, screened and named pLUC-HMGB1-3′-UTR-wt and pLUC-HMGB1-3′-UTR-mut. The pLUC-HMGB1-3′-UTR-wt or pLUC-HMGB1-3′-UTR-mut and miR-25 mimics were co-transfected into 293 cells (Thermo Fisher Scientific, Inc.) using Lipofectamine^® 2000, according to the manufacturer's instructions. Luciferase activity was detected after 48 h. Firefly and Renilla luciferase activities were measured for each sample using the Dual-Luciferase^® Reporter Assay System (cat. no. E1960; Promega Corporation).

The microRNA.org online target gene prediction tool was used to predict the relation between miR-25 and the 3′-UTR of HMGB1 mRNA (11). Therefore, the successfully differentiated macrophages were divided into 5 groups: miR-negative control (NC) transfection group, miR-25 mimic transfection group, short interfering RNA (si)-NC transfection group, si-HMGB1 group, and miR-25 mimic+si-HMGB1 group. The siRNA sequences are as follows: si-HMGB1 sense strand, 5′-CGGCCUCUGUUUGAUUUCUTT-3′, si-HMGB1 antisense strand, 5′-AGAAAUCAAACAGAGGCCGTT-3′; si-NC sense strand, 5′-UUCUCCGAACGUGUCACGUTT-3′, si-NC antisense strand, 5′-ACGUGACACGUUCGGAGAATT-3′; miR-25 mimic sense strand, 5-CAUUGCACUUGUCUCGGUCUGA-3; miR-25 mimic antisense strand, 5-UUCAAGUAAUCCAGGAUAGGCU-3; control sense strand, 5-UCACAAGUCAGGCUCUUGGGAC-3, control antisense strand, 5-ACGGGUUAGGCUCUUGGGAGCU-3. Lipofectamine^® 2000 was used to transfect the cells (5×10⁴ cell/well, 50 nM si-HMGB1 or si-NC in each well) according to the manufacturer's protocol. A total of 72 h after the transfection, cells were sequenced to confirm transfection and used for subsequent experiments. Cells in all transfection groups were treated with 100 ng/ml LPS for 48 h prior to collection of cells and culture supernatants.

Total RNA was isolated using TRIzol, according to the manufacturer's instructions, following extraction with 99% chloroform for 2 min at room temperature, precipitation with 95% isopropanol for 10 min at room temperature, and washing with 75% ethanol followed by centrifuging at 7,500 × g for 5 min at 4°C, then hydrolysis with RNAse free H₂O. cDNA was obtained using the PrimeScript™ RT reagent kit, according to the manufacturer's instructions, and the resulting cDNAs were stored in a −20°C refrigerator. PCR was performed using Taq DNA polymerase from the SYBR Green Real-Time PCR Master mix and the following primers: miR-25, forward, 5′-CGGCGGCATTGCACTTGGTCTC-3′, reverse, 5′-GTGCAGGGTCCGAGGT-3′; U6, forward, 5′-ATTGGAACGATACAGAGAAGATT-3′, reverse, 5′-GGAACGCTTCACGAATTTG-3′; HMGB1, forward, 5′-TATGGCAAAAGCGGACAAGG-3′, reverse, 5′-CTTCGCAACATCACCAATGGA-3′; GAPDH, forward, 5′-ACAACTTTGGTATCGTGGAAGG-3′, reverse, 5′-GCCATCACGCCACAGTTTC-3′. The PCR reaction contained 5.0 µl 2X SYBR Green Mixture, 0.5 µl 2.5 µm/l forward primer, 0.5 µ of 2.5 µm/l reverse primer, 1 µl cDNA and ddH₂O to make the volume up to 10 µl. The thermocycling conditions were as follows: 40 cycles of 95°C for 15 sec, 60°C for 30 sec and 74°C for 30 sec in an ABI 7500-type fluorescence quantitative PCR system (Thermo Fisher Scientific, Inc.). Results were expressed by using the 2^−ΔΔCq method (12), where the number of cycles (Ct) at which the fluorescence signal exceeded a defined threshold for GAPDH was subtracted from Ct values for target genes (Ct_targets-Ct_GAPDH=ΔCP), and values were calculated as 2^−ΔCP and normalized to each other.

Protein was extracted using radioimmunoprecipitation assay buffer (Thermo Fisher Scientific, Inc.). The protein concentration was determined using a BCA protein quantification kit, according to the manufacturer's protocol. A total of 50 µg of protein was loaded for separation by 8% SDS-PAGE, then transferred into polyvinylidene difluoride membranes by electroporation for 60 min. The membranes were blocked with 5% skimmed milk for 60 min at room temperature, followed by incubation with the primary antibodies (HMGB1, phosphorylated (p-)p65, GAPDH and Lamin B1 antibodies were used at dilutions of were 1:100, 1:100, 1:800 and 1:300, respectively) overnight at 4°C, then washed 3 times PBS-Tween (PBST) prior to incubation with the HRP-labeled anti-mouse or anti-rabbit secondary antibodies (dilution, 1:8,000 dilution) for 60 min at room temperature. The membranes were washed 3 times in PBST, and visualized using an enhanced chemiluminescent system (Thermo Fisher Scientific, Inc.).

All ELISA procedures strictly followed the protocols provided by the ELISA kit manufacturer.

Collagen IV added to the upper chamber of Transwell inserts with 8-µm pore size (Corning Incorporated, Corning, NY, USA). Macrophages were plated into the upper chamber (5×10⁵) and medium (α-MEM without FBS) containing 100 ng/ml LPS was added to the lower chamber. After 48 h of culture, the culture solution was discarded, and the insert was washed twice with PBS, and fixed with methanol for 30 min at room temperature, stained by 0.1% crystal violet for 20 min at room temperature. The number of migratory cells was counted under a light microscope at ×400 magnification.

All statistical analysis was performed using SPSS ver. 18.0 (SPSS, Inc., Chicago, IL, USA). Data are expressed as mean ± standard deviation. Differences between 2 groups were compared using Student's t-test. One-way analysis of variance was used to compare differences between multiple groups, followed by Bonferroni correction. Pearson correlation coefficient analysis was used to observe the relation between miR-25 level in serum and sepsis. P<0.05 was considered to indicate a statistically significant difference.

ELISA assay results demonstrated that the level of HMGB1 in the serum of peripheral blood in patients with sepsis was significantly increased compared with the healthy control group (Table I). Western blotting results revealed that the expression level of HMGB1 protein the PBMCs of patients with sepsis was significantly increased compared with that of healthy controls (Fig. 1A). RT-qPCR demonstrated that the expression of miR-25 in the serum and PBMCs of patients with sepsis was significantly decreased compared with that of healthy controls (Fig. 1B). Pearson correlation coefficient analysis demonstrated that the expression of miR-25 in serum of patients with sepsis was negatively correlated that of HMGB1 (r=−0.591, P=0.041). A significant positive correlation was identified between serum HMGB1 levels and HMGB1 protein expression in PBMCs (r=0.537, P<0.001), and a significant negative correlation was identified between serum miR-25 expression and HMGB1 protein content (r=−0.622, P<0.001), as presented in Fig. 1C and D.

MicroRNA.org online target gene prediction results indicated that there was a targeted binding site between miR-25 and the 3′-UTR of HMGB1 mRNA (Fig. 2A). Transfection of miR-25 mimics significantly reduced the relative luciferase activity, suggesting that miR-25 could target the 3′-UTR region of HMGB1 and inhibit its expression (Fig. 2B).

The results of flow cytometry demonstrated that the macrophage-specific marker, CD68, was increased by >80% following a 7-day induction with M-CSF, indicating that the differentiation of macrophages was induced successfully, allowing subsequent experiments to be performed (Fig. 3A). ELISA assays demonstrated that, following treatment with 100 ng/ml LPS for 48 h, the protein levels of HMGB-1, TNF-α and IL-6 in the culture supernatant were significantly increased (Fig. 3B). The results of RT-qPCR demonstrated that LPS treatment significantly upregulated the expression of HMGB1 mRNA and significantly decreased the expression of miR-25 in macrophages (Fig. 3C). Western blotting results revealed that LPS treatment significantly enhanced the transcriptional activity of NF-κB and upregulated the expression of HMGB1 protein in the nucleus and cytoplasm of macrophages (Fig. 3D).

The transfection of miR-25 mimics and/or si-HMGB1 significantly reduced HMGB1 mRNA expression (Fig. 4A) and attenuated protein expression in macrophages (Fig. 4B), resulting in a reduction in p-p65 protein expression (Fig. 4B). In addition, the levels of inflammatory cytokines, HMGB1, TNF-α and IL-6, were significantly decreased in culture supernatant (Fig. 4C), and the migration ability of macrophages was also significantly attenuated (Fig. 4D) in cells transfected with si-HMGB1 or miR-25 mimic + si-HMGB1 compared with control.