Systemic Wound Healing Associated with local sub-Cutaneous Mechanical Stimulation

Degeneration is a hallmark of autoimmune diseases, whose incidence grows worldwide. Current therapies attempt to control the immune response to limit degeneration, commonly promoting immunodepression. Differently, mechanical stimulation is known to trigger healing (regeneration) and it has recently been proposed locally for its therapeutic potential on severely injured areas. As the early stages of healing consist of altered intra- and inter-cellular fluxes of soluble molecules, we explored the potential of this early signal to spread, over time, beyond the stimulation district and become systemic, to impact on distributed or otherwise unreachable injured areas. We report in a model of arthritis in rats how stimulations delivered in the subcutaneous dorsal tissue result, over time, in the control and healing of the degeneration of the paws’ joints, concomitantly with the systemic activation of wound healing phenomena in blood and in correlation with a more eubiotic microbiome in the gut intestinal district.

20-40 probes was detected. For each comparison, differentially expressed miRNAs were selected using the same methodologies used for mRNA profiling. Data is available at the National Center for Biotechnology Information Gene Expression Omnibus (GSE58458).

Subcutaneous and Synovium Tissue (RAFLs cells) mRNA Differential Analysis
Quality assessment was confirmed by FastQC 12 and fastx toolkits 13 , including per-base quality score, nucleotide composition, N-and GC content, overrepresented sequences (frequency > 0.1%). Unique read frequency saturation curve was also calculated. Reads filtering was performed differently for subcutaneous tissue and synovium. For subcutaneous tissue data raw reads were trimmed and filtered by cutadapt 14 : reads beginning with the 5' adapter (P5) were discarded as PCR artifacts (Hiseq initiates reading after P5) and the trailing 3' adapters (P7) were trimmed (maximum mismatch rate 0.1, overlap length 5), low-quality ends (Phred quality score < 30) were trimmed and remaining reads shorter than 25nt were removed.
For synovium filtered reads include: 1) reads with adapter sequences, 2) reads in which the percentage of unknown bases (N) is greater than 10%, 3) reads in which more than 50% bases quality value ≤ 5, as processed by BGI Shenzhen Company.
Filtered reads were aligned to the rat reference genome (UCSC rn4) using bowtie2 15 for subcutaneous tissue and synovium. Gene expressions were quantified by HTSeq v0.5.4p3 16 at the transcript level using "intersection-nonempty" mode and the UCSC rn4 annotation (downloaded from iGENOME for rattus norvegicus). Uniformly to the 16S rRNA-seq processing, raw counts were normalized to cpm by TMM normalization 6 using edgeR v3.4. 2 7 and differential analyses were run with the limma-voom pipeline v3.18.9 3 . Prefiltering for limma was done by removal of transcripts with low (more than 20% of replicates < 1 cpm in both groups) or invariant expressions. DEs were selected with: 1) average expression in cpm ≥1; 2) nominal P value < 0.05; 3) FC ≥2 (output by limma as "logFC"). Data is available at the National Center for Biotechnology Information Gene Expression Omnibus (subcutaneous tissue batch1 GSE57983, synovium GSE58978).
Subcutaneous miRNA-Seq differential analysis. Raw sequences were obtained and de-multiplexed using the Illumina pipeline CASAVA v1.8.
Quality check were performed with FastQC 12 and FASTX 13 toolkits. The 3' adapters were trimmed by mapper.pl from miRDeep2 toolkit 17 , reads shorter than 18bp were discarded.
Clipped reads were aligned using mapper.pl to Rat genome (rn4) and miRBase v18 (rno sequences) 18 and quantified using quantifier.pl in miRDeep2. Expression of miRNA data (counts) from all samples were joined and miRNAs with at least 1 cpm in at least half of the total samples were kept. Based on the counts data, the limma-voom pipeline 9 was adopted to perform differential analysis, with the same parameters used for mRNA Affymetrix Each enrichment test t in each v in T has 3 possible outcomes xi, i=1,2,3: x1=up for upregulated genes enrichment, x2=down for downregulated genes enrichment, x3=both for up and  The coherence score C x  X / |{T}|, that multiplies the final logical score output in Eq.2 is used to visually grade the intensity of up (green) and down (red) enrichments as shown in Fig    Data S3-Subcutaneous tissue mRNA differential and enrichment analysis:  List of differential genes (after to before therapy)

 Enrichment analysis with David therapy-wise
Data S4-PBMC miRNA differential analysis and targets:  List of differential miRNAs (after to before therapy)  miRDB predicted targets of the differential miRNAs  miRBase validated targets of the differential miRNAs  miRNA&mRNA (includes predicted+validated differential miRNAs and differential mRNAs (from Supplementary Data S2) Data S5-Subcutaneous tissue miRNA differential analysis and targets:  List of differential miRNAs (after to before therapy)  miRDB predicted targets of the differential miRNAs  miRBase validated targets of the differential miRNAs  miRNA&mRNA (includes predicted+validated differential miRNAs and differential

mRNAs (from Supplementary Data 3)
Data S6-Spatiotemporal assays enrichment analysis:  List of genes sets used for enrichment (reference sets)

 Enrichment analysis results
Data S7-16S rRNA differential analysis, genera annotation, association tests:  OTUs abundances  Summary of differential fecal 16S rRNA analysis (Genus level) at 34 days (after to before)  Differential genera annotated as eubiotic or dysbiotic.
 Statistical test association of the eubiotic/dysbiotic microbiome varied composition between treatments Data S8-Synovial tissue mRNA differential and enrichment analysis:  List of differential mRNAs  Enrichment analysis with David  qRT-PCR validation on 3 markers