GEO (http://www.ncbi.nlm.nih.gov/geo/) is an international public repository that archives and freely distributes high-throughput gene expression and other functional genomics datasets (11). A gene expression dataset, namely, GSE83233 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE83233) ‘An inhaled dose of budesonide induces genes involved in transcription and signaling in human airways’, was downloaded from GEO [GPL15207 platforms, (PrimeView) Affymetrix Human Gene Expression Array]. In the study (9), healthy male, nonsmoker, nonallergic volunteers (age 18-50 years) with normal lung function were recruited into this prospective, double-blind, placebo-controlled, randomized, two-period crossover study involving an initial screening visit, followed by two intervention visits. Participants were screened at visit one to fulfil the study eligibility criteria. At visit two, volunteers were randomized to receive inhaled Bud (1,600 µg) or placebo, both via Turbuhaler. Then, at two to three weeks later, at visit three, participants received either Bud or placebo, as appropriate, to complete both study arms. At 6 h after placebo or Bud inhalation, bronchial biopsies were obtained via bronchoscopy (9). The study protocol and consent form were approved by the Conjoint Health Research Ethics Board at the University of Calgary and Alberta Health Services and the ethical approval number was 23241.

The DEGs in the Bud and placebo specimens were selected from GEO2R (http://www.ncbi.nlm.nih.gov/geo/geo2r), which is a platform for examining DEGs across experimental conditions by comparing multiple datasets in GEO series. The genes with multiple probes were averaged and the probes that lacked gene symbols were removed. The DEGs with a |logFC (fold change)| >0 and P<0.01 were screened out and represented statistical significance.

The database for annotation, visualization and integrated discovery (DAVID), which can be freely accessed (http://david.ncifcrf.gov), is a web-based online bioinformatics resource that aims to provide tools for the functional interpretation of large lists of genes and proteins (12). In addition, GO, a significant bioinformatics tool, enables the annotation of genes according to biological processes (13). The KEGG is a knowledge-based platform for the systematic analysis of gene functions, linking genomic information with higher-order functional information (14,15). The DAVID online database was used to study the functions of DEGs. P<0.05 was considered to indicate a statistically significant difference. ImageGP (http://www.ehbio.com/ImageGP/) was used to draw enrichment plots for GO and KEGG.

A mouse macrophage RAW264.7 cell line was purchased from Procell Life Science & Technology Co., Ltd. According to the instructions, the cells were maintained in RPMI-1640 medium (Beijing Solarbio Science & Technology Co., Ltd.) supplemented with 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.), 100 U/ml penicillin and 100 µg/ml streptomycin (Shanxi MiniBio Technology Co., Ltd.) in a 5% CO₂ incubator at 37˚C. The medium was replaced the next day.

A conditionally immortalized mouse MPC-5 podocyte cell line was purchased from Fuheng Biology Company and cultured in RPMI-1640 medium supplemented with 10% FBS and recombinant IFN-γ (cat. no. G1021; APeXBIO Technology LLC) at 33˚C. Podocytes were reseeded and cultured in RPMI-1640 medium with 10 mg/ml type-I collagen (BD Biosciences) at 37˚C without IFN-γ for 7-15 days before testing.

RAW264.7 cells and MPC-5 cells were seeded into 96-well plates separately at a density of 1x10⁶ cells/ml and cultured in 10% FBS RPMI-1640 medium for 24 h at 37˚C. Following another 24 h treatment with Bud (cat. no. 200721527; Chiatai Tianqing Pharmaceutical Group Co., Ltd.) at 5, 10, 20, 40 and 80 µM (16,17), LPS at 0.02, 0.10, 0.50, 2.50 and 12.5 µg/ml at 37˚C (18-20). The supernatants were removed and each well was washed with PBS before the addition of 10% FBS RPMI-1640 medium and 10 µl CCK-8 reagent (Shanxi MiniBio Technology Co., Ltd.). Cell viability was determined by measuring the absorbance at 450 nm using a microporous plate reader (Model 550; Bio-Rad Laboratories, Inc.) after an incubation period of 2 h at 37˚C. The average optical density was determined by examining six wells per group.

Based on the cytotoxicity results of LPS on RAW264.7 cells. LPS (0.10 µg/ml) was used to induce macrophage polarization to the M1 subtype for 24 h at 37˚C. The nitric oxide (NO) level was tested with the Griess assay (21). Nitrite, a stable end-product of NO metabolism, was measured using the Griess reaction. Culture media of the RAW 264.7 cells (100 µl) was mixed with an equal volume of Griess reagent (Yantai Science & Biotechnology Co. Ltd.), followed by spectrophotometric measurement at 540 nm. Nitrite concentrations in the culture media were determined by comparison with a sodium nitrite standard curve. All experiments were repeated three times.

The protein expression levels of iNOS, TNF-α, Arg-1 and CD206 at different concentrations of Bud (5, 10, 20 µM) in 0.10 µg/ml LPS-cultured macrophages were tested by western blotting. RPMI-1640 was used as the normal control, LPS was used as the model control and 10 µM curcumin (cat. no. C7090; Beijing Solarbio Science & Technology Co., Ltd.; purity: 95.0%) was used as the positive control (22). The treated cells (1x10⁶ cells/ml) were removed from the culture media and lysed with RIPA lysis buffer from Beijing Solarbio Science & Technology Co., Ltd. for 30 min. The protein concentrations were determined using a BCA Protein Assay kit from Beijing Solarbio Science & Technology Co., Ltd. Samples containing 50 µg of protein were resolved by 10% SDS-PAGE electrophoresis and transferred to polyvinylidene fluoride membranes (MilliporeSigma) in a buffer tank with platinum wire electrodes. After immersing the membranes into 5% nonfat dried milk (diluted in 0.1% (v/v) Tween-20 PBS) for 2 h at room temperature to block nonspecific binding, the membranes were incubated overnight with a primary antibody against iNOS at a 1:2,000 dilution (cat. no. 18985-1-AP; ProteinTech Group, Inc.), a primary antibody against TNF-α (cat. no. bs-2081R; Bioss) at a 1:1,000 dilution, a primary antibody against CD206 (cat. no. bs-21473R; Bioss) at a 1:1,000 dilution and a primary antibody against Arg-1 (cat. no. 16001-1-AP; ProteinTech Group, Inc.) at a 1:5,000 dilution at 4˚C. The membranes were then washed three times (10 min each) and incubated with the corresponding secondary IgG conjugated to HRP antibody (cat. no. SA00001-2; ProteinTech Group, Inc.) at room temperature for 1 h. The results were finally analyzed by the Quantity One analysis system (Bio-Rad Laboratories, Inc.). GAPDH at a dilution of 1:5,000 (cat. no. 10494-1-AP; ProteinTech Group, Inc.) was used as the internal loading control.

RAW264.7 cells (1x10⁶ cells/ml) were treated with 0.10 µg/ml LPS for 2 h and then treated with Bud (5, 10, 20 µM) in LPS-cultured macrophages for another 22 h at 37˚C. RPMI-1640 was designated the normal control for all the above-tested groups and the curcumin group was designated the positive control (22). Cell supernatants were then harvested and centrifuged at 1,500 x g for 10 min at 4˚C. TNF-α levels were determined using an ELISA kit (cat. no. MM-0132M1; Jiangsu Meimian Industrial Co., Ltd.). IL-1β levels were tested with an ELISA kit (cat. no. MM-0040M1; Jiangsu Meimian Industrial Co., Ltd.). The absorbance was measured using a microplate reader. Each sample underwent repeated testing four times.

The NO levels of the normal control, curcumin control and Bud at 5, 10 and 20 µM were also tested with the Griess assay (21). All experiments were repeated three times.

RAW264.7 cells were seeded into six-well plates and cultured in normal control, LPS model and curcumin positive, Bud 5, 10, 20 µM at 37˚C, respectively. The supernatants were collected 24 h later, centrifuged at 1,500 x g for 15 min at 4˚C and labeled as (Nor)MS, (LPS)MS, (Cur)MS, (Bud5)MS, (Bud10)MS and (Bud20)MS. The collected supernatants were centrifuged for the second time at 1,500 x g for 10 min at 4˚C, filtered with a 0.22 µm sterile membrane and stored at -80˚C for further use.

Podocytes (5x10⁵ cells/ml) were cultured with normal medium, 0.1 µg/ml LPS, 10 µM curcumin, 5 µM Bud, 10 µM Bud, or 20 µM Bud and (Nor)MS, (LPS)MS, (Cur)MS, (Bud5)MS, (Bud10)MS, or (Bud20)MS for 22 h at 37˚C. Podocytes were then seeded into 96-well plates. The cells were subsequently treated with bromodeoxyuridine (BrdU) reagent during the final 2 h of incubation. The BrdU cell proliferation assay kit (cat. no. 2750; MilliporeSigma) was performed according to the manufacturer's protocol. The survival rate of the normal group was 100% and the survival rate of cells in the other groups was compared with that of the normal group. The mean was obtained by examining six wells per group.

SPSS 19.0 software (IBM Corp.) was used for statistical analysis. All the data were expressed as the mean ± standard deviation (SD) of the mean. A two independent sample unpaired Student's t-test was used to analyze differences between two groups. One-way analysis of variance with Bonferroni's posttest was used when more than two groups were present. P<0.05 was considered to indicate a statistically significant difference.

After standardization of the microarray data, all DEGs between the Bud group and placebo group in GSE83233 were identified, as shown in Fig. 2. The dataset consisted of a total of 832 significant DEGs [|logFC (fold change)| >0 and P<0.01], of which 311 DEGs were downregulated and 521 were upregulated.

Functional and pathway enrichment analyses were conducted on DEGs for further analyses of biological classification. Based on total DEGs, the numbers of significant cell component (CC), molecular function (MF), biological process (BP) and KEGG pathways were 22, 34, 142 and 37, respectively (P<0.05). The top 20 CC, MF, BP and KEGG terms are shown separately in Fig. 3. The results indicated that the main CC terms included the nucleus, cytoplasm and plasma membrane. The MF terms were mainly related to protein binding, transcription factor activity, sequence-specific DNA binding, etc. The main BP terms included positive/negative regulation of transcription from RNA polymerase II promoter and positive/negative regulation of transcription, DNA-templated and inflammatory response. KEGG results showed that a number of pathways were involved in the inflammatory response, such as the forkhead box subgroup O (FoxO) signaling pathway and the TNF signaling pathway.

Based on upregulated DEGs, there were 18 significant KEGG items (P<0.05) among the 31 items in Table II. The FoxO signaling pathway had the lowest P-value and the second highest gene ratio in KEGG. Based on downregulated DEGs, further KEGG study showed that there was a total of 16 items in which the 10 had significant differences (P<0.05) in Table I. The TNF signaling pathway had the lowest P-value and the second highest gene ratio.

Fig. 4A shows that Bud inhibited the growth of RAW 264.7 cells in the range of 40-80 µM (cell viability at 40 µM: 90.5±4.5 and 80 µM: 83.8±8.0%) and had significant differences compared with 0 µM Bud (100.3±4.0%; P=0.009 and P=0.006). Fig. 4A also shows that the viability of MPC-5 cells treated with 80 µM Bud was (86.3±7.5)%, which was significantly different from that of cells treated with 0 µM Bud (101.4±9.7%; P=0.005). The above results suggested that Bud at 5, 10 and 20 µM could be used to test its effects on macrophage polarization and podocyte injury.

Fig. 4B shows that LPS inhibited the growth of RAW 264.7 cells at 2.50 µg/ml and 12.5 µg/ml (cell viability: 91.1±4.0 and 85.1±7.2%) and had a significant difference compared with 0 µg/ml LPS (P=0.010 and P=0.008). According to the above results, LPS at 0.02, 0.10 and 0.50 µg/ml could be used on RAW 264.7 cells to induce M1 polarization.

The cell survival rates of MPC-5 cells treated with different concentrations of LPS are shown in Fig. 4B. The results showed that compared with 0 µg/ml LPS, LPS had significant cytotoxicity on MPC-5 cells at 0.10, 0.50 µg/ml, 2.50 and 12.5 µg/ml (four: P<0.01). In this case, 0.10, 0.50, 2.50 and 12.5 µg/ml LPS could be used on MPC-5 cells to establish a podocyte injury model.

Compared with NO production in 0 µg/ml LPS (normal control; 3.3±0.3 µM), NO production significantly increased after macrophages were cultured with 0.10 µg/ml LPS for 24 h (77.6±1.5 µM; P<0.001).

When Bud was added to RAW264.7 cells and cultured for 24 h separately, the results showed that 5, 10 and 20 µM Bud had no significant effects on NO production compared with the normal control (4.0±0.1, 4.5±0.6, 4.0±0.2 µM; three: P=NS). Additionally, compared with the normal control, curcumin did not increase NO production significantly (4.3±0.2 µM; P=NS).

Fig. 5A and B shows the CD206, Arg-1, TNF-α and iNOS protein bands of the normal, LPS model, curcumin and Bud at three different concentrations.

Compared with the normal control, LPS significantly increased TNF-α and iNOS protein expression (P=0.025; P=0.025) while decreasing Arg-1 and CD206 protein expression (CD206: P<0.001). As the positive control, curcumin significantly decreased TNF-α and iNOS protein expression compared with the LPS model (P=0.004; P=0.013). The above results are shown in Fig. 5.

The results also showed the effects of 5, 10 and 20 µM Bud on CD206, Arg-1, TNF-α and iNOS protein expression, as shown in Fig. 5C-F. Compared with the LPS model, 5 µM Bud significantly increased Arg-1 and CD206 expression (P<0.001; P=0.002) and decreased iNOS expression (P=0.021) and 10 µM Bud significantly increased Arg-1 expression (P=0.030) and decreased TNF-α and iNOS expression (P=0.024; P=0.002). Among the three concentrations, Bud at 20 µM not only showed a significant effect on the above four protein expression levels when compared with the LPS model (CD206: P<0.001, Arg-1: P<0.001, TNF-α: P<0.001 and iNOS: P=0.004) but also showed significantly improved effects on CD206, TNF-α and iNOS than the other two concentrations (all: P<0.05).

The 0.10 µg/ml LPS significantly stimulated TNF-α levels (677.5±23.5 pg/ml vs. 530.8±27.0 pg/ml; P=0.003), IL-1β levels (188.8±14.6 pg/ml vs. 89.4±6.7 pg/ml; P<0.001) and NO production (77.6±1.5 µM vs. 3.3±0.3 µM; P<0.001) compared with the normal group. When compared with the LPS group, the positive curcumin significantly decreased the TNF-α level to (627.2±23.3 pg/ml; P=0.039), significantly decreased the IL-1β level to (122.6±6.0 pg/ml; P=0.006) and significantly decreased NO production to (51.2±1.0 µM; P<0.001).

Compared with the LPS group, Bud at 5, 10 and 20 µM all showed significant decreasing effects on IL-1β levels and NO production (six: P<0.05). The decreasing effects of Bud on NO production occurred in a significant dose-dependent manner (40.8±1.2 µM NO production at 5 µM vs. 35.4±0.7 µM NO production at 10 µM vs. 8.8±0.3 µM NO production at 20 µM; three: P<0.05). Additionally, compared with the LPS group, Bud at 10 and 20 µM showed significant effects on decreasing TNF-α levels (P=0.026; P<0.001).

Bud at 20 µM showed a significantly improved decreasing effect on NO production and TNF-α levels than curcumin (P<0.001 and P=0.001), but Bud at 20 µM and curcumin did not show a significant difference in decreasing IL-1β levels (P=NS). All of the above results are shown in Fig. 6.

Fig. 7 shows that the MPC-5 cell survival rates between normal control and LPS model were significantly different (P<0.001). Compared with LPS, (LPS)MS further decreased the cell survival rate significantly (54.6±5.2% vs. 66.1±7.5%; P=0.001).

Additionally, compared with LPS, curcumin significantly increased the cell survival rate (84.1±5.0%; P=0.003), but Bud at 5, 10 and 20 µM did not show any significant influence on the cell survival rate under LPS stress (three: P=NS).

Compared with (LPS)MS, (Cur)MS (80.6±4.9%; P=0.001) and (Bud5)MS (66.0±5.0%; P=0.027), (Bud10)MS (74.0±4.5%; P=0.002), (Bud20)MS (79.8±5.8%; P=0.001) all showed significantly increased effects on cell survival rates. The increasing effect on the cell survival rate of (Bud20)MS was significantly improved compared with that of (Bud10)MS and (Bud5)MS (P=0.046; P=0.004), but there was no significant difference compared with that of (Cur)MS (P=NS).