IL-1β-induced modulation of gene expression profile in human dermal fibroblasts: the effects of Thai herbal Sahatsatara formula, piperine and gallic acid possessing antioxidant properties

Background Pain is the main symptom of most musculoskeletal disorders and can be caused by inflammation in association with oxidative stress. Thai herbal Sahatsatara formula (STF), a polyherbal formula, has been traditionally used for relieving muscle pain and limb numbness. This study aimed to investigate biologically active compounds of STF and its pharmacological effects related to antioxidant and anti-inflammatory activities. Methods The identification of possibly active compounds of STF was performed by high performance liquid chromatography (HPLC). Moreover, this study also assessed the free radical scavenging activities of STF and its components using DPPH radical scavenging assay and their inhibitory effects on IL-1β-induced intracellular reactive oxygen species (ROS) formation in primary human dermal fibroblasts (NHDFs) using DCFDA-flow cytometry analysis. Modulation of human gene expression by STF and its active compounds was investigated by microarray analyzed through Gene Ontology (GO) classification and pathway enrichment analysis. Results HPLC analysis has revealed the presence of gallic acid (GA) and piperine (PP) as the major compounds in STF extracts. Our finding discovered that STF and its active compounds (GA and PP) yielded free radical scavenging activities and abilities to inhibit IL-1β-induced cellular ROS formation in NHDFs. Furthermore, microarray analysis demonstrated that a total of 84 genes (54 upregulated and 30 downregulated) were significantly affected by IL-1β involved in inflammatory cytokines, chemokines, transcription factors, cell adhesion molecules and other immunomodulators participating in NF-κB signaling. The significantly upregulated genes in IL-1β-treated in NHDFs participate in interleukin and cholecystokinin (CCRK) signaling pathways. The GO analysis of the target genes showed that all test compounds including indomethacin, STF and its active compounds, can downregulate the genes involved in NF-кB signaling pathway in IL-1β-treated NHDFs compared to the cells treated with IL-1β alone. Conclusions STF and its active compounds possessing antioxidant actions can modulate the effects of IL-1β-mediated alteration of gene expression profiles associated with inflammatory signaling in NHDFs. Electronic supplementary material The online version of this article (doi:10.1186/s12906-016-1515-0) contains supplementary material, which is available to authorized users.


Background
Pain is a non-specific symptom that can be found in several musculoskeletal disorders such as myofascial pain syndrome, fibromyalgia, tendinitis and arthritis [1]. Pathogenesis of pain is attributed to initiation of cellular damage and connective tissue damage through a number of mechanisms related to inflammation and oxidative stress. Some of secreted chemotactic molecules and mediators including cytokines can also trigger excessive reactive oxygen species (ROS) formation [2,3].

Materials and reagents
The STF composes of 21 components (Table 1). The crude powder of STF and its ingredients was supported by Center of Applied Thai Traditional Medicine, Faculty of Medicine Siriraj Hospital, Mahidol University. The characteristic of individual material was circumspectly authenticated by two experienced applied Thai traditional practitioners. The formula and ingredient powder was processed by the Manufacturing Unit of Herbal Medicines and Products Ayurved Siriraj under Good Manufacturing Practice (GMP) certification. Normal human dermal fibroblasts (NHDFs) and 2% fibroblast basal medium (FBM) were purchased from LONZA (LONZA, USA). The analytical grade ethanol and methanol reagents were purchased from Scharlau (Scharlau, Spain). All chemical reagents were obtained from Sigma-Aldrich (MO, USA).

Preparation of STF and its component extracts
The dried powder of STF and its ingredients were extracted 3 times by ethanol 80% at ratio of 1:10 (w/v) using ultrasonic sonicator machine (BANDELIN, Germany) controlled the ultrasonic frequency at 30 kHz and the temperature between 25 and 35°C for 30 min and then filtrated through filter paper no. 4 (Whatman, England). The extraction solution was concentrated by rotary evaporator under 50°C and at a pressure between 150 and 180 mbar (Buchi, Switzerland) and each sample was stored at −80°C prior to lyophilization. The calculated % yield (w/w) was shown in Table 1. The lyophilized powder (10 mg) was reconstituted with 80% ethanol, vortexed and centrifuged at 22136.4 g for 10 min at 4°C. The extraction solution was then filtered through 0.22 μm polyvinylidene difluoride membrane (PVDF) (Vertical Chromatography, Thailand) to remove the extraction residues.

Identification of active compounds by HPLC analysis
HPLC analysis was applied for identification and quantification of the STF extracts and its possible active compounds. Briefly, phenolic compounds (including gallic acid (GA), p-coumaric acid, kojic acid, caffeic acid, ferulic acid, and vanillic acid) and piperine (PP) were selected as reference markers from literature-based review of STF components. Chromatographic profiles of STF extract and reference compounds were detected by the Waters 2998 Photodiode Array (PDA) and the analytical procedure was performed using Alliance e2695 (Water, USA) and Waters Empower 2 software. Phenumenix C18 5 μm (4.6 mm × 250 mm) and Sunfire C18 5 μm (4.6 mm × 150 mm) (Water, Ireland) were used as a stationary phase for analysis of phenolic compounds and piperine, respectively. The mobile phases at a flow rate 1 mL/min composed of 1% acetic acid in Milli-Q water or 10 mM ammonium acetate (pH 6.8) (solvent A) for phenolics and PP, respectively, and 100% methanol (solvent B) for both phenolics and PP. For analysis of phenolic substances, a gradient condition was used as follows: 0 min, 98% A/2% B; 3 min, 90% A/10% B; 5 min, 80% A/20% B; 7 min, 70% A/30% B; 13 min, 85% A/15% B; 14 min, 100% B; 16-18 min, 98% A/2% B. The peaks were detected at wavelengths 210-400 nm. For analysis of PP, an isocratic elution with 25% A/75% B was used and chromatograms were detected at 340 nm. The identification of phenolic compounds and PP in STF extracts was presented in Fig. 1a and b. The GA and PP contents in STF were 0.175 (W/W) % and 0.602 (W/W) %, respectively.

Cell culture and treatments
NHDFs divided from adult female skin were cultured in 2% FBM supplemented with 1% penicillin (100 units/ To induce inflammation and oxidative stress of NHDFs, IL-1β was used in this study and diluted with serum-free medium and indomethacin applied for a positive control was dissolved in 1.80% dimethyl sulfoxide (DMSO) solution.

Cell viability assay
MTT assay was performed to determine non-toxic concentrations of vehicles (80% ETOH and DMSO) and test compounds for further cellular studies. Briefly, cells were pre-incubated with vehicle or test compounds at vary concentrations for 24 h and 200 μL of MTT reagent (0.2 mg/mL) was then added into each well and incubated for 4 h. To dissolve formazan product, 200 μL of DMSO reagent were added into each well followed by measuring the absorbance at 595 nm using a spectrophotometer (SpectraMAX M5, Molecular Devices, CA).

Gene expression profiling study
To assess the effect of STF and its active compounds on human-whole genome expression. NHDFs were treated with indomethacin (50 μM), GA (0.3, 1, 3 μg/mL), PP (1, 30 μg/mL) and STF extracts (3, 10, 30 μg/mL) for 30 min prior to treatment with IL-1β (1 ng/mL) for 4 h. Briefly, total mRNA of each treatment was isolated by Illutra RNA spin Mini RNA isolation Kit (GE Healthcare, UK). The concentration and purity of isolated mRNA were determined using NanoVue UV/visible spectrophotometer (GE Healthcare, UK) and Qubit® quantitation assay kit (Life Technologies, USA) followed by the instructions. To estimate RNA integrity, RNA samples were analyzed using Agilent RNA 6000 Nano kit (Agilent Technologies, USA) with Agilent 2100 Bioanalyzer (Agilent Technologies, USA). The 28S/ 18S ratio of total RNA was equal and greater than 2.0 and the RNA Integrity Number (RIN) was greater than 7.0. Then, the total mRNA was converted into complementary RNA (cRNA), amplification and biotin labelling were carried out according to the Illumina TotalPrep™ RNA Amplification kit (Life Technologies, USA). Biotin-labelled cRNA (750 ng) was hybridized with Human HT-12 V4 BeadChip (Illumina, USA) following the whole genome

Microarray data analysis
The signal intensities of samples were normalized and analyzed using Microsoft Excel program version 2013. Data analysis was performed as described in Additional file 1: Figure S1. Twelve human housekeeping genes (Additional file 1: Table S1) were applied for normalization of each gene intensity using the normalization factor to adjust the unequal distribution of signal intensity. The normalized and non-normalized microarray data had deposited in the Gene Expression Omnibus (GEO) databases with accession number GSE86798 (https://www.ncbi.nlm.nih.gov/ geo/query/acc.cgi?acc=GSE86798). The normalized values were used to calculate fold change, which was converted into logarithmic base 2. To determine the up-regulated or down-regulated genes, the Student's t-test was performed to estimate significant difference of gene expression based on P < 0.05 with fold change that cut off ≥ 2 or ≤ 0.5 times compared with untreated control group or IL-1β-treated group without compound treatment. To investigate the effects of STF extract and its active compounds on gene expression analyzed by Gene Ontology (GO) classification and pathway enrichment analysis using PANTHER webdatabase followed a protocol [11].
Quantitative real-time polymerase chain reaction (qRT-PCR) qRT-PCR was used to validate differential gene expression from IL-1β stimulation. The list of primer sequences were described in (Additional file 2: Table S1). The relative normalized genes expression were estimated using ΔΔCt method and each gene expression result was normalized with GADPH being housekeeping gene and internal control in the experiment. The experiment was carried out in triplicate. The results of qRT-PCR analysis were shown in Additional file 2: Figures S1 and S2.

Statistical analysis
Data were expressed as means and standard deviation (SD) or standard error of the mean (SEM). The significantly statistical values concerned at P <0.05 were estimated by Unpaired t-test and one-way analysis of variance (ANOVA) followed by Dunnett's post hoc test using GraphPad Prism version 5 for Windows (GraphPad Software Inc., San Diego, CA, USA).

The antioxidant activities of STF and its ingredients
The free radical scavenging activities of STF extracts, its components and possible active compounds including GA and PP were examined by DPPH assay. As shown in Table 1, STF extracts, GA and L-ascorbic acid had abilities to scavenge DPPH radical in a dosedependent manner, whereas PP had no free radical scavenging properties. Based on IC 30  Cytotoxicity assessments: inflammatory inducer, solubilizing vehicles, STF and its active compounds The results showed that treatment of NHDFs with vehicles and test compounds including STF extracts, GA and PP (3.75 to 30 μg/mL), indomethacin (1-100 μM) and IL-1β (0.1,1,10 ng/mL) for 24 h did not affect NHDFs viability compared to untreated control cells (Fig. 2).
The inhibitory effects of STF extracts and its active substances on intracellular ROS formation induced by IL-1β in NHDFs As shown in Fig. 3a, IL-1β (1 ng/mL and 10 ng/mL) significantly enhanced ROS generation in NHDFs (P <0.01 and <0.001) determined by flow cytometry. The results of oxidized-DCFDA fluorescence indicated that STF extracts (10 and 30 μg/mL) and the highest dose of GA (3 μg/mL) and PP (30 μg/mL) significantly suppressed ROS production in NHDFs compared with IL-1β-induced cells (Fig. 3b-d). Additionally, treatment with vehicle control (Fig. 3e), STF extracts or its active compounds alone did not affect ROS formation in NHDFs.

IL-1β modulated gene expression profile in fibroblasts: the effects of STF and its active compounds
To investigate the role of IL-1β on transcriptional profiling of fibroblasts, the volcano plot of whole human gene expressions (47,231 probes) demonstrated the differential expression of IL-1β-treated cells compared with untreated cells. Following the analytical criteria, the volcano plot was shown total of 84 genes with significantly upregulated expression (54 genes) and down-regulated expression (30 genes) detected in IL-1β-treated cells compared to untreated control cells ( Fig. 4a and Additional file 3: Table S1 and S2). The gene GO analysis of up-and down-regulated genes was shown in Fig. 4b. The upregulated genes in IL-1β-treated cells mainly involved signaling pathways such as inflammation, apoptosis, heterotrimeric G-protein activation, and adaptive immune activation. However, only two pathways from PANTHER pathway enrichment analysis comprising BICR3, CXCL1, CXCL2, NFKBIA, IL-8 and IL-6 were significantly moderated by IL-1β (Table 3). 22 genes of the 54 upregulated genes were targeted genes associated with NF-кB transcription, including related transcription factors, chemokines, cytokines and immunoreceptors regulated by NF-кB ( Table 2). The 84 genes which were significantly altered (P < 0.05, fold change ≥2 or ≤ 0.5) were considered to be the target genes. To determine the effects of test compounds on IL-1β-treated cells, the expression profiling of 84 genes modulated by STF extracts, GA, PP and indomethacin were observed compared to IL-1β -treated cells, as shown in Fig. 5. In upregulated genes induced by IL-1β, treatment with STF (3 μg/mL) provided greater effects on downregulation of inflammatory and immune genes involved in NF-κB components (NFKB1,NFKBIA,NFKBIZ), chemokines (CCL2,CXCL1,CXCL2,IL8), transcriptional regulator (IRF1), prostaglandin E synthase (PTGES) and TNF alpha induced protein family (TNFAIP2, TNFAIP3, TNFAIP6) than indomethacin in IL-1β-treated cells. Furthermore, Meanwhile, gallic acid (3 μg/mL) appeared to give the greatest inhibition of IL-1β-induced MSC and ZC3H12A genes (Fig. 5a). Additionally, STF reversed the effects of IL-1β-mediated downregulation of some genes such as early B-cell factor 3 (EBF3), LIM domain binding 2 (LDB2) and PHD finger protein 13 (PHF13). Moreover, STF enhanced the effects of IL-1β-mediated downregulation of SLC38A2 and STC genes (Fig. 5b). The overall of gene expression profile modulated by IL-1β, indomethacin, STF and its active compounds were demonstrated in Additional file 4: Table S1.
Furthermore, Fig. 6 demonstrated the effect of all test compounds on gene expression pattern following the findings showing 1,653 significantly altered genes in IL-1β-treated cells compared to untreated control cells (Fig. 4a). Treatment with indomethacin (Fig. 6a), PP ( Fig. 6b and c), GA (Fig. 6d-f ) and STF differently modulated gene expression patterns (Fig. 6g -i).

Discussion
In this study, we first investigated biologically active compounds of STF using LC separation and its pharmacological effects related to the anti-inflammatory and antioxidant actions using primary human fibroblast culture model and microarray analysis. The HPLC analysis revealed that PP and phenolic compounds including GA were present in STF. In a previous study using reversephase HPLC [12], PP was also present in long pepper (Piper retrofractum Vahl.) and black pepper (Piper nigrum L.). In addition, other previously reported study cited GA and PP as having anti-inflammatory and antioxidant activities and thus both compounds were chosen as reference markers for investigating the effects of STF extracts in our study [13][14][15].
Since ROS play a role in initiation and progression of inflammatory disorders. Our results indicated that STF and its components were able to scavenge DPPH free radicals, although their abilities to scavenge DPPH were less than ascorbic acid base on IC 30 values. Among all STF's components, Terminalia chebula Retz. (gall and fruit), showed the strongest free radical scavenging Fig. 3 The effects of STF extracts and its active compounds on ROS formation was assessed by DCFDA-flow cytometry. IL-1β significantly enhanced ROS formation in a dose-dependent manner in IL-1β-treated NHDFs (a). The pretreatment of STF extracts (b), GA (c) and PP (d) can attenuate IL-1βinduced intracellular ROS levels whereas, treatment with each compound alone (b-d) and vehicle reagents (ETOH 80%) (e) didn't affect ROS generation. DCFDA intensity data were presented as Mean ± SD at least three experiments. The statistical significant between control and IL-1β-treated cells was evaluated by Unpaired t-test and between IL-1β-treated and test compound-treated cells was assessed by one-way ANOVA followed by Dunnett's post hoc test. **P < 0.01, *** P < 0.001 compared with control. # P < 0.05, ## P < 0.01 and ### P < 0.001 compared with IL-1β-treated cells activities. Previous studies also reported that tannins and phenolic compounds, including GA, punicalagin, chebulagic acid, casuarinin and chebulanin found in Terminalia chebula Retz. were accountable for its antioxidant potentials [16]. Concurrently, PP standard failed to yield free radical scavenging activities while black pepper and long pepper, which were the major constituents in STF extracts, had minimal ability to scavenge the DPPH radicals. It is possible that free radical scavenging actions of black pepper and long pepper may be attributed to other phytochemical substances [17]. Thus, the DPPH results suggested that GA and other unidentified compounds present in the STF's components may be responsible for the free radical scavenging activity of the formula.
Inflammatory and oxidative stress events are associated with the production of pro-inflammatory cytokines through NADPH oxidase complex [18] which subsequently produce ROS such as superoxide and hydrogen peroxide in human fibroblasts. The flow cytometry analysis supported that pro-inflammatory cytokine IL-1β in NHDFs can induce ROS formation. Furthermore, this study also explored antioxidant effects of STF extracts, GA and PP on IL-1β-treated NHDFs. Our study demonstrated that pretreatment of the cells with STF and its active compounds can suppress IL-1β-induced intracellular ROS levels. Previous reports also demonstrated that GA and PP showed abilities to promote mRNA levels and activities of antioxidant enzymes and inhibition of lipid peroxidation [14,19]. Additionally, the concentrations at which STF demonstrated its inhibitory effects on ROS formation were much higher than those of GA. Other unidentified substances in the formula may also contribute to antioxidant actions of the STF extracts.
IL-1β plays critical roles in regulating inflammatory and antioxidant genes related to cytokines, chemokines, adhesion molecules, immunomodulators, apoptotic regulators, and growth factors involved in physiological, pathological and oxidative stress events. In our study, total 84 genes were significantly altered; 54 genes were up-regulated and 30 genes were down regulated in IL-1β-treated NHDFs and, in consistent with previous studies, similar pattern of altered genes or gene expression profiling was observed fibroblast model of inflammation [20,21]. In addition, it is well known that IL-1 can activate NF-кB in various cell types through canonical NF-кB-activating cascades [22,23]. Our results indicated that the member of NF-кB transcription components, NFKB1, NFKBIA (IκBα) and NFKBIZ genes, were significantly upregulated in IL-1β-treated NDHFs. Additionally, treatment with IL-1β upregulated the mRNA of SOD2, which serves as the primary antioxidant defense in mitochondria that changes superoxide molecules into Fig. 4 The differential expression of genes in NHDFs treated with IL-1β when compared with non-treated cells. The significantly altered genes (30 down-regulated and 54 up-regulated) induced by IL-1β were indicated by yellow dots and red dots, respectively (a). The Gene Ontology (GO) classification of 84 altered genes was performed using PANTHER classification system analysis (b).
hydrogen peroxide and oxygen and could be regulated by NF-кB [24,25]. The microarray results suggesting altered expression of genes involved in mitochondrial dysfunction were in consistent with the flow cytometric findings showing that the presence of IL-1β enhanced ROS production in NHDFs as augmented ROS generation can be associated with mitochondrial injury [26].
Thus, it is possible that upregulation of antioxidant defenses may reflect an adaptive mechanisms in response to inflammatory challenge.
Currently, it has been suggested that more than 150 target genes including transcription factors, cytokines, chemokines and immune mediators can be regulated by NF-кB [27]. The crosstalk of inflammatory and Table 2 The summery of 22 target genes involved in transactivation of NF-кB induced by IL-1β in NHDFs. All gene description and functional annotation were referred from PANTHER and NCBI databases  immunomodulatory pathways with multiple other pathways unrelated to NF-кB signaling is therefore possible [28]. Interestingly, it is well recognized that NOD2 is involved in host defense against bacterial infections and various inflammatory diseases. Previous studies showed that pro-inflammatory cytokines such as IFN-γ, TNF-α are able to upregulate NOD mRNA and protein levels [29]. Furthermore, NOD2 signaling pathway also participates in NF-кB activation [30] resulting in active IL-1β secretion [31]. Additionally, BIRC3 encodes a member of the inhibitor of apoptosis protein (IAP) family, which plays the potential role in caspase-1 modulation in association with inflammasome activation, which could be mediated by IL-1β [32]. Furthermore, IAP encoded by BIRC3 gene is suggested to serve as a regulator of NF-кB signaling [33]. Our study showing upregulation of NOD2 and BIRC3 gene by IL-1β in human fibroblasts supports the role of IL-1β in acute inflammation; moreover, it also contributes to immune activation in fibroblasts through secreting multiple biochemical mediators such as cytokines, chemokines, transcription factors, proteases, and apoptosis regulators involved in the downstream of NF-кB activation.
Interestingly, PANTHER pathway analysis indicated that IL-1β could upregulate interleukin and CCKR signaling in NHDFs (Table 3). Previous evidence showed that IL-1β had the potential to upregulate important genes including NFKBIA and IL-8 involved in activation of NF-кB cascade associated with CCKR signaling [34,35], as well as IL-6 and IL-8, the critical pro-inflammatory mediators, involved in interleukin signaling. Our findings also indicated the effects of IL-1β on CCKR and prostaglandins (Additional file 3: Table 1) postulated to be involved in pain sensitization [36]. In addition, increased expression of IL-6 and IL-8 genes by IL-1β in our study support the inflammatory role of IL-1β as previous in vitro study observing enhanced IL-6 and IL-8 levels in rheumatoid arthritis synovial fibroblasts [37].
We also investigated the effects of indomethacin used as the positive control, STF extracts and its active compounds (GA and PP) on 84 target genes induced by IL-1β ( Fig. 5a and b). All test compounds can downregulate the expression of genes involved in NF-кB signaling pathway through three NF-кB modulated pathways [38]. NFKB1 and NFKB1A genes which were significantly down-regulated more than 2-fold by STF (Fig. 5a) participate in the pathways that can counteract the NF-кB upstream signaling including the inhibition of phosphorylation of IKK complexes and IKB degradation, the inhibition of mobilization of NF-кB into cell nucleus and binding with specific DNA sequence.
In addition, our study showed that STF extracts led to a significant down-regulation (>1.5-fold) of IL-1β-induced IRF1 gene involved in COX-2 expression [39]. Moreover, the STF extracts also attenuated many inflammatory chemokine and interleukin genes (Fig. 5a) such as CXCL1, CXCL2, CCL2 IL-6 and IL-8 enhanced by IL-1β which were previously suggested to associate with clinical analgesia [40] and inflammatory hypernociception [41]. On the contrary, all test compounds except STF significantly downregulated (>2-fold) IL-1β-induced expression of CD83, and NOD2 genes. CD83 is associated with the activation of NF-кB binding and increased level of PGE 2 [42].  Nevertheless, the mechanism by which STF enhances the overexpression of NOD2 gene involved in innate immune and inflammatory responses [43] needs further investigation. Furthermore, STF reversed the effects of IL-1β on downregulation of various genes (Fig. 5b), such as, EBF3 associated with B lymphocyte differentiation [44], bone and neuronal genesis [45] and tumor suppression [46]. Moreover, STF remarkably downregulated expression of STC2 gene involved in endoplasmic reticulum-stress through PERK pathway in nonalcoholic fatty liver disease [47]. Our results revealed for the first time that STF and its active compounds including PP and GA can modulate various genes involved in intracellular signaling cascades including inflammatory and immunomodulatory signaling pathways. However, our work provided only the evidence-based data on gene expression profiles affected by STF using primary fibroblast model of inflammation. Thus, further studies are warranted to elucidate their therapeutic targets integrating diverse signaling pathways and other pharmacological fields such as pharmacokinetics, pharmacodynamics and herb-drug interactions relating with therapeutic efficacy for development of STF as a novel anti-inflammatory agent.

Conclusions
In summary, HPLC analysis has revealed the presence of GA as the major phenolic in STF extracts and the presence of PP as the major compound in pepper, the main component herb in STF, indicating that both GA and PP could be possible active ingredients in STF extracts. Pathway enrichment analysis of microarray data indicated that IL-1β can enhance various genes involved in interleukin and CCRK signaling pathways in NHDFs. In targeted gene analysis, all test compounds (indomethacin, STF, GA and PP) provided diverse gene expression profiling involving multiple transcriptional factors including NF-кB as well as inflammatory signaling in IL-1β-pretreated NHDFs.