Ex vivo microRNA and gene expression profiling of human Tr1‐like cells suggests a role for miR‐92a and ‐125a in the regulation of EOMES and IL‐10R

Ex vivo gene expression and miRNA profiling of Eomes+Tr1‐like cells suggested that they represent a differentiation stage that is intermediate between Th1‐cells and cytotoxic CD4+ T‐cells. Several microRNAs were downregulated in Eomes+Tr1‐like cells that might inhibit Tr1‐cell differentiation. In particular, miR‐92a targeted Eomes, while miR‐125a inhibited IFN‐g and IL‐10R expression.

sFigure 2: Validation of gene expression and miRNA targeting in CD4 + T-cells A. Validation of selected expressed and control genes (EOMES, GZMK, GZMB, IL10, IFNG) by RT-qPCR in independent donors (n=4), statistical analysis was done by with a Friedman Test. B. Alignment of miR-92a with the human EOMES 3'UTR. The miRNA seed sequence and its target sequence are indicated in red. C. Representative ex vivo Eomes expression in conventional (CD4 w/o Tregs, see sFig 1A)) CCR5 + T-cells D. Eomes protein expression levels in purified human CD4 + CCR5 + T-cells that were transfected with miR-92a or a scrambled control. Data shows 3 different donors analysed in 3 independent experiments. E. Histogram Overlay showing ex vivo IL-10R expression in gated Tr1-cells. The red line shows the negative Control (Fluorescence minus One), whereas the blue line shows IL-10R staining. F. Memory T-cells (CD4 + CD45RA -) were transfected with a miR-125a mimic or antagomir oligonucleotide, or controls. 48 h after transfection, the expression of miR-125a was measured by RT-qPCR. 4 different Donors analysed in 2 independent experiments. Error bars report standard deviations. G. Same as D., except that expression of IL-10R and IFN-g were measured both at mRNA (top) and protein (bottom) level. Error bars report standard deviations. sTable 1: List of 424 differentially expressed genes Differentially expressed genes in Tr1-like cells were identified by One-way ANNOVA (p<0.01) followed by a post-hoc analysis (Tukey HSD) in order to compare every mean to every other mean. Genes of interest are highlighted in yellow.

sTable 2: Differentially expressed genes in Tr1-cells compared to CTL, Th1EM or Th1CM
Differentially expressed genes in Tr1-like cells identified by One-way ANNOVA (p<0.01) as compared to the individual subsets (CTL, Th1EM or Th1CM). Genes identified in two subsets are highlighted with the indicated colours.

RNA isolation and miRNA-mRNA expression profiling
Total RNA was isolated using mirVana miRNA Isolation Kit (Ambion) following standard protocols. Briefly, the cell lysates were extracted once with acid phenol-chloroform and further purified to yield total RNA with specific miRNA retention. Extracted RNA was quantified with RiboGreen Quantitation Kit (Molecular Probes) on an Infinite F200 plate reader (Tecan Trading AG). All extracted RNA samples were quality controlled for integrity with a 2100 Bioanalyzer (Agilent Technologies) and samples with RIN (RNA Integrity Number) lower than 8 were discarded. The standard Megaplex protocol (Applied Biosystems) was performed starting from 10 ng of total RNA for each sample with pre-amplification.
Gene expression of whole transcriptome was performed using Illumina Direct Hybridization Assays according to the standard protocol (Illumina Inc.). Total RNA was isolated, quality controlled and quantified; for each sample 80 ng of total RNA were reverse transcribed according to the Illumina TotalPrep RNA Amplification kit (Ambion) and cRNA was generated after 14 h of in vitro transcription. Washing, staining and hybridization were performed according to the standard Illumina protocol: briefly, 750 ng of cRNA of each sample in a final volume of 15 μl were hybridized onto Illumina HumanHT-12 v3 Expression BeadChip arrays. Hybridization and scanning were performed according to the manufacturer's indications on an Illumina iScan System and data were processed with BeadStudio v.3.
For miRNA analysis, total RNA was reversed transcribed using the Megaplex RT stem-loop primers in a 7.5 μl reaction volume through the protocol's default 40 cycles run; 2.5 μl of each RT product were pre-amplified in a 25 μl reaction volume with Megaplex PreAmp primers to increase detection sensitivity according to manufacturer's specifications. Expression analysis of mature miRNAs was assessed using TaqMan Low Density Arrays (Applied Biosystems), enabling specific detection of 664 human miRNAs (miRBase v.10.1) and 3 small RNA controls common to all plates (RNA44, RNA48, MammU6). 9 μl of 4-fold diluted pre-amplified RT product were 100-fold diluted in the PCR reaction mix and amplified using TaqMan Low Density Arrays (TLDAs) on a 7900HT (Applied Biosystems) according to the standard protocol.

Data filtering and statistical analysis
Gene expression profiling. Gene expression arrays were quantile normalized with background subtraction, and average signals were calculated on gene-level data for genes whose detection p-value was lower than 0.001 in at least one of the cohorts considered (Th1CM, Th1EM, CTL and Tr1-like cells). Normalized data were log2 transformed. A one-way ANOVA (p-value< 0.01) was used to select for all four cell subsets profiled. Unsupervised hierarchical clusters on significant genes were performed with euclidean distance and ward.D2 linkage. In order to survey possible groupings in the data, principal component analysis (PCA) was performed. Statistical test, hierarchical clustering and PCA were carried out using R 4.0.3 software. miRNA profiling. Raw CT values were calculated using the SDS software v.2 and exported as "Amplification Data" text files for each plate. TLDAs affected by globally poor or anomalous amplification were discarded. After this quality control step, 16 samples belonging to the four distinct cell subsets entered the final dataset. A one-way ANOVA (p-value< 0.001) was used to select miRNA classifiers for all four cell subsets profiled. In order to minimize the biological variability for the classifiers selection we considered miRNAs expressed in at least 2/3 (66%) of the profiled samples analyzed in each cell subset. Unsupervised hierarchical clusters on significant miRNAs were performed with Pearson distance and complete linkage. Statistical test and hierarchical clustering were performed on MeV software version 4.5.
For assessment of miR-31, miR-125a-5p, miR-125b-5p, miR-92a and RNU48 (used as normalizer) levels on specific human lymphocyte subsets single Taqman MicroRNA assays were used (Applied Biosystems). 5 ng of total RNA were used for reverse transcription with specific primers for miR-31, miR-125a-5p, miR-92a and RNU48 and Taqman MicroRNA Reverse Transcription Kit followed by real-time PCR reaction on a 7900HT Real Time PCR System according to the manufacturer's standard protocol.