Study of the microRNA expression profile of foreskin derived mesenchymal stromal cells following inflammation priming

This study was conducted in accordance with the Declaration of Helsinki (1964) and approved by the local ethics committee of the “Institut Jules Bordet” (Belgium). All donors and/or their parents or legal guardian were voluntary for giving the foreskin (FSK) samples as obtained following a circumcision procedure from healthy subjects. They gave written informed consent before specimen collection for using the samples and for publishing any associated scientific data. However, all personal information are kept confidential.

Briefly, after circumcision, foreskin was aseptically collected into a sterile specimen container containing sterile phosphate-buffered saline (PBS) buffer supplemented with penicillin/streptomycin. The samples were processed as previously published [8]. After a sterile wash with PBS, the sample was transferred into a petri dish where epidermis was manually removed from the skin and the dermis was cut into small pieces. By using Liberase Research Grade solution (Roche Diagnostics, Belgium), the tissue was digested and the resulting cell suspension was washed by centrifugation (800g, 5 min) in Dulbecco’s modified Eagle’s medium with low glucose (DMEM-LG; Lonza, Belguim) and containing 10% fetal bovine serum (FBS; Sigma-Aldrich, Belgium). The subsequent cell pellet was seeded in culture flasks with DMEM-LG (Lonza) supplemented with 10% FBS (Sigma-Aldrich), 2 mM l-glutamine and 50 U/ml penicillin (both from Lonza) and the cultures were incubated at 37 °C in a 5% CO2 humidified atmosphere. Non-adherent cells were removed when the medium was changed (once a week). When sub-confluence (80–90%) was achieved, adherent cells were harvested by TrypLE Select (Lonza) and expanded until the desired passage.

The cells were characterized according to the ISCT criteria. Thus, the immunophenotype was analysed by flow cytometry (MacsQuant analyzer (Miltenyi Biotec, Netherlands)) using fluorochrome labelled monoclonal antibodies: anti-CD45-FITC and anti-HLA-Dr-PE (Exalpha Biologicals, Maynard, MA), anti-CD34-PE and anti-CD73-PE (BD Biosciences, San Diego, CA, USA), anti-CD14-PE, anti-CD19-PE, anti-CD105-FITC and anti-CD90-PE (R&D systems, Minneapolis, MN, USA). Their multilineage potential was confirmed by inducing specific lineage commitment (adipogenic, osteogenic and chondrogenic differentiation) using appropriate induction culture medium (NH media, Miltenyi Biotec). Specific lineage commitment was highlighted by using lineage-specific cell staining techniques and microscopic examination.

Foreskin–mesenchymal stromal cells were analyzed under both constitutive and inflammatory conditions. Inflammation priming of FSK–MSCs was performed as previously described [16]. Briefly, cells were treated (overnight) with a pro-inflammatory cytokine cocktail containing IL-1β (Peprotech, Rocky Hill, NJ, USA) (25 ng/ml), TNF-α (50 ng/ml), IFN-α (3000 U/ml or 10 ng/ml) and IFN-γ (1000 U/ml or 50 ng/ml) (all from Prospec Inc., Rehovot, Israel). After priming, the medium was removed, and the cells were washed and became available for analysis.

Total RNA was extracted from cells using TRIzol total RNA isolation reagent (Roche Applied Science). The concentration was quantified using a NanoDrop spectrophotometer. A three-step procedure was performed to profile the miRNAs. First, for cDNA synthesis from the miRNAs, 30 ng of total RNA was subjected to RT (reverse transcription) using a TaqMan® microRNA Reverse Transcription Kit (#4366596; Applied Biosystems) and Megaplex RT primers (Human Pool A, #4399966; Applied Biosystems) following the manufacturer’s protocol, allowing simultaneous reverse transcription of 380 mature human miRNAs to generate a miRNA cDNA library corresponding to each plasma sample. RT was performed on a Mastercycler Epgradient thermocycler (Eppendorf) with the following cycling conditions: 40 cycles at 16 °C for 2 min, 42 °C for 1 min and 50 °C for 1 s followed by a final step of 80 °C for 5 min to inactivate reverse transcriptase. Thereafter, to generate enough miRNA cDNA template for the following real-time PCR, the cDNA libraries were pre-amplified using Megaplex PreAmp primer (Humam Pool A, #4399233; Applied Biosystems) and PreAmp Master Mix (#4384266; AppliedBiosystems) following the manufacturer’s instructions. The PreAmp primer pool used here consisted of forward primers specific for each of the 380 human miRNAs and a universal reverse primer. The pre-amplification cycling conditions were as follows: 95 °C for 10 min, 55 °C for 2 min, 72 °C for 2 min followed by 12 cycles at 95 °C for 30 s and 60 °C for 4 min; the samples were then held at 99.9 °C for 10 min. After the pre-amplification step, the products were diluted with RNase-free water, combined with TaqMan gene expression Master Mix and then loaded into TaqMan Human MicroRNA Array A (#4398965; Applied Biosystems), which is a 384-well formatted plate and real-time PCR-based microfluidic card with embedded TaqMan primers and probes in each well for the 380 different mature human miRNAs. Real-time PCR was performed on an ABI PRISM 7900HT sequence detection system (Applied Biosystems) with the following cycling conditions: 50 °C for 2 min, 94.5 °C for 10 min followed by 40 cycles at 95 °C for 30 s and 59.7 °C for 1 min. The Ct (cycle threshold) was automatically given by SDS 2.4 software (Applied Biosystems) and is defined as the fractional cycle number at which the fluorescence passes the fixed threshold of 0.2. RNU48 embedded in the TaqMan human microRNA arrays was used as an endogenous control. The relative expression levels of miRNAs were calculated using the comparative ΔΔCt method as described previously [17, 18]. The fold changes in miRNAs were calculated by the equation 2^−ΔΔCt.

Gene-specific reverse transcription was performed for each miR using 10 ng of purified total RNA, 100 mM dNTPs, 50 U MultiScribe reverse transcriptase, 20 U RNase inhibitor, and 50 nM of gene-specific RT primer samples using the TaqMan MicroRNA Reverse Transcription kit (Applied Biosystems, Gent, Belgium). 15 μl reactions were incubated for 30 min at 16 °C, 30 min at 42 °C, and 5 min at 85 °C to inactivate the reverse transcriptase. Real time RT-PCR reactions (5 μl of RT product, 10 μl TaqMan 2 × Universal PCR master Mix, (Applied Biosystems, Gent, Belgium), and 1 μl TaqMan MicroRNA Assay Mix containing PCR primers and TaqMan probes) were carried out on ABI Prism 7900HT Sequence Detection System (Applied Biosystems, Gent, Belgium) at 95 °C for 10 min followed by 40 cycles at 95 °C for 15 s and 60 °C for 1 min. The qRT-PCR reactions were performed in triplicate, and the signal was collected at the end of every cycle. The Ct (cycle threshold) was automatically given by SDS 2.4 software (Applied Biosystems) and is defined as the fractional cycle number at which the fluorescence passes the fixed threshold of 0.2.

RNU48 was used as an endogenous control. The relative expression levels of miRNAs were calculated using the comparative ΔΔCt method as described previously [31, 32]. The fold changes in miRNAs were calculated by the equation 2^−ΔΔCt.

Pathway enrichment analysis was employed to investigate the regulatory mechanisms of significantly differentially expressed miRNAs. DIANA-miRPath v.3.0 analysis web server was used for pathway enrichment analyses. The enriched pathways were defined by their enrichment of significantly differentially expressed miRNA target genes.

Data analyses were performed using the SDS RQ Manager 1.2 software and DataAssist v2.0 software (Applied Biosystems). Statistical significance of miRNA expression between control and treated cells was determined according to unpaired Mann–Whitney U test. p Values <0.05 (*), <0.01 (**) were considered significant.

In order to obtain FSK–MSCs, the foreskin sample was first processed to separate the epidermis from the dermis (see "Methods" section) and then scraping was applied to obtain small tissue pieces from the dermis. Following enzymatic digestion, the tissue mixture was centrifuged and the resulting cell pellet was incubated for culture. After removing non adherent cells, colonies of cells with fibroblastic morphology started to appear. When sub-confluency (80–90%) was reached, adherent cells were harvested using TrypLE Select solution and expanded by successive passages.

These adherent cells were then characterized according to the ISCT criteria. Flow cytometry analysis demonstrated that those cells are positive (>95%) for the MSC markers CD73, CD90 and CD105 whilst negative (<5%) for CD45, CD34, CD14, CD19 and HLA-DR markers. It is noteworthy that inflammation priming had no significant impact on the expression level of those markers (data not shown).

To comply with the ISCT criteria, we further confirmed that the obtained cells exhibited a multi-lineage potential since they were able to differentiate into cells of adipogenic, osteogenic or chondrogenic lineage after being cultured with appropriate induction media (data not shown). After 21 days of osteogenic differentiation, a mineralization matrix following calcium deposits was observed by Alizarin red staining demonstrating the generation of osteoblasts. After 10 days of adipogenic differentiation, cells with cytoplasm containing several lipid vacuoles were revealed by Oil Red O staining indicating the formation of adipocytes. After 21 days of chondrogenic differentiation, the production of proteoglycans-based extracellular matrix was shown by Alcian blue staining illustrating the generation of chondrocytes.

Aware that MSCs can serve as sensors and switchers of inflammation, and given the importance of miRNAs in regulating inflammatory responses, we aimed at characterizing the expression level of different miRNAs in FSK–MSCs and investigating the impact of inflammation on this expression. FSK–MSCs were derived from 5 independent donors and total RNA was prepared from those cells being either untreated (control cells) or treated with a pro-inflammatory cytokine cocktail (treated cells). TaqMan low density array (TLDA) was then carried out to assess miRNA expression level in those cells. Before measuring miRNAs’ levels and in order to identify suitable endogenous controls, three different candidate miRNAs (RNU44, RN48, and U6 snRNA) were analyzed for variance in gene expression using Data Assist software v2.0. The statistical method ranked the candidate endogenous control genes with an excellent correlation of raw stability values (data not shown). The most stably expressed RNU48 was chosen as the endogenous control, and relative miRNA expression was normalized against RNU48. Our TLDA analysis revealed that the miRNA expression profile in treated cells is not totally similar to that of untreated cells. In fact, we show the expression level of 25 miRNAs that were modulated in treated vs. control cells (Fig. 1). Among these, 20 miRNAs (miR-145, -149, -182, -194, -199a, -221, -27a/b, -328, -330-5p, 345, -34c, -361, -369-5p, -423-5p, 485-3p, 485-5p, -494, -615-5p and-758) were downregulated whilst 5 miRNAs (miR-107, -155, -183, -363 and -886-3p) were upregulated. Details of the results are shown in Table 1. A clustergram of the samples as well as the significantly differentially expressed miRNAs in control and treated cells are illustrated in Fig. 1 in the form of a heat map generated using ΔCt values. Heat map is a typical method used in gene-expression analysis where the expression level of many genes across a number of tested samples can be displayed in a two dimensional image with a color code representing the expression intensity of each gene.

The numbers corresponding to these colors are the ΔCt values. The dendrogram on the left side of the heat map classifies miRNAs into groups based on the divergence of miRNA expression values among the different samples. The dendrogram presented at the top indicates the relatedness of the samples based on overall miRNA expression values and separates the control from the treated group of samples.

In a second step, and in order to validate their differential expression, miRNAs that appeared to be upregulated or downregulated in treated vs. control cells were further examined using individual quantitative Real Time PCR (qRT-PCR). Interestingly, out of the 25 miRNAs that showed altered expression (Table 1), 16 miRNAs were confirmed to exhibit such differential expression in treated vs. control cells (Fig. 2). Those 16 miRNAs fall in two groups. Group 1 contains 13 miRs that were downregulated (ratio between 0.1and 0.005) in treated cells in comparison to control cells and includes miR-27a, -145, -149, -194, -199a, -221, -328, -345, -423-5p, -485-3p, -485-5p, -615-5p and -758 (Fig. 2). Observing that those 16 miRs are not equal in terms of their downregulation rate led us to further classify them into subgroups. Group 1A corresponds to miRNAs that were most strikingly downregulated and consists of miR-27a, -145 and -221 that decreased 10, 13.7 and 15 folds, respectively. Group 1B consists of miRNAs that were less strikingly downregulated and includes miR-149, -194, -615-5p and -758 that exhibited decreased rates of 7, 8.4, 5 and 5.3 folds, respectively. Group 1C contains the least strongly downregulated miRNAs and includes miR-199a, -328, -345, -423-5p, 485-3p and -485-5p that showed downregulation rates of 3.8, 2, 4.8, 2.5, 3.4 and 3.7 folds, respectively.

On the other hand, group 2 contains 3 miRNAs (miR-155, -363 and -886-3p) that were upregulated (ratio greater than 3) in treated vs. control cells (Fig. 2). Among these, miR-155 was the most strikingly upregulated miR exhibiting a 9.4 fold increase whilst miR-363 and -886-3p showed increased rates of 4.7 and 4.5 folds, respectively. Altogether, these observations demonstrate a clear difference in the miRNA expression profile in FSK–MSCs exposed to inflammatory signals vs. control cells suggesting a potential role for miRNAs in modulating FSK–MSCs’ transcriptional programs in response to inflammatory conditions.

Since each miRNA can regulate the expression of many target genes but multiple miRNAs can modulate specific pathways, we explored the pathways that were potentially regulated by the miRNAs observed to be altered in inflammation primed FSK–MSCs in comparison to untreated control MSCs.

For the identification of these pathways, DIANA-miRPath v.3.0 analysis web server was used and each of the statistically differentially expressed miRNAs was used as a query. The output Kyoto encyclopedia of genes and genomes (KEGG) pathways obtained for each miR included many results and for reasons of simplicity and specificity we focused only on pathways linked to inflammation. These pathways were previously summarized in [19] and included Adhesion-Extravasation-Migration, apoptosis signaling, calcium signaling, cytokine signaling, leukocyte signaling, innate pathogen detection, MAPK signaling, NF-κB signaling, PI3 K-AKT signaling, TNF superfamily signaling, NK cell signaling, GPCR signaling, ROS/Glutathione/cytotoxic granules, phagocytosis-antigen presentation, complement cascade, Eicosanoid signaling and Glucocorticoid signaling/PPAR signaling. The pathways targeted by upregulated and downregulated miRs are shown in Table 2. It is noteworthy that no KEGG pathways linked to inflammation were identified for miR-149, -328, -345, -363, -423-5p, -615-5p, -758 and -886-3p.