Patients. Gene Expression Profiling Was Performed on Muscle Biopsies from 3 Patients JDM

Appendix e-1

Patients. Gene expression profiling was performed on muscle biopsies from 3 patients JDM (mean age 9.3 ± standard deviation 3.2 years; all females), 3 with DM (44 ± 23.1 years; all females), 4 with PM (49.5 ± 20.4 years; all females) and 5 controls (43.2 ± 10.5 years; four males and one female). Further molecular and immunohistochemistry experiments were also done on muscles from an additional 6 JDM (mean age 8.6 ± 6.3 years; 3 males and 3 females), 13 DM (57.4 ± 10.6 years; 5 males and 8 females), 18 PM (54.2 ± 16.1 years; 8 males and 10 females), 10 dystrophinopathy patients with varying degrees of inflammation and muscle fiber degeneration (5 with Duchenne muscular dystrophy: 3.6 ± 1.8 years, 5 with Becker muscular dystrophy: 3.6 ± 2.2; all males) and 14 controls (29.7 ± 20.8 years; 8 males and 6 females).

Microarrays. Total RNA integrity was assessed by using a 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, US). cDNA was synthesized with SuperScript II reverse transcriptase (Invitrogen) using T7-(dT)24 primer (Affymetrix, Santa Clara, CA, US). cDNA was biotin-labeled and transformed into target cRNA in vitro using the BioArray High Yield RNA Transcript Labeling Kit (Enzo Life Sciences, Farmingdale, NY, US).

Fifteen µg of fragmented cRNAs was hybridized on human HG-U133A GeneChip® microarrays following the procedure described by the manufacturer (Affymetrix).

The Bioconductor software was used for data processing.e-1,e-2 The GC Robust Multi-array Analysis (GCRMA) algorithm was used to calculate probe set intensity.e-3 Normalization employed a quantile method. To assess differential expression, an empirical Bayes methode-4 was used with a false discovery rate correction (Benjamini-Hochberg method) applied to p values.

Thus, the list of differentially expressed genes was generated using an FDR ≤ 0.01.e-5 Functional annotation of DEGs was based on a subset of the annotations provided by the Bioconductor project. The most representative GO functional annotations for DEGs from each experimental condition were identified by determining the probability of random occurrence of functional terms (hyperGeometric distribution). Based on this probability, if p values are less than 10-5, the enrichment is considered significant. The microarray data have been submitted to ArrayExpress (http://.www.ebi.ac.uk/microarray), accession number E-MEXP-2681.

Real-time and semi-quantitative PCR. TLR3 primer/probe sequences were designed using the Primer Express software package (Applied Biosystems): TLR3 forward 5’-CCTGGTTTGTTAATTGGATTAACGA-3’; TLR3 reverse 5’-GAGGTGGAGTGTTGCAAAGG-3’; TLR3 probe 5’-FAM-ACCCATACCAACATCCCTGAGCTGTCAA-3’. Optimal concentrations of TLR3 primers and probe, efficiency and validation were determined following procedures previously described.e-6 For the analysis of MX1, TLR7, TLR9 and IFN-g transcripts, pre-designed functionally tested assays (all from Applied Biosystems) were used: MX1: Hs00182073_m1; TLR7: Hs00152971_m1; TLR9: Hs00152973_m1; IFN-g: Hs00174143_m1. The primers used for mRNA quantification of MyD88 were: forward: 5’-TGGGACCCAGCATTGAGGA-3’; reverse: 5’-CGCTGGGGCAATAGCAGA-3’. Primer sequences for STAT-1 mRNA, designed with Primer3 Software (Applied Biosystems), were: forward 5’-TTCAGAGCTCGTTTGTGGTG-3’; reverse 5’-GGTGCCAGCATTTTTCTGTT-3’ (fragment length 314 bp).

Immunohistochemistry. For diagnosis, cellular infiltrates were characterized by immunohistochemistry of acetone-fixed 6 mm-thick cryosections of muscle. The following mouse anti-human monoclonal antibodies were used: anti-CD3, anti-CD4, anti-CD8, anti-CD22, anti-CD163, anti-MHC class I and II (all from Dako). The antibodies were applied for 2 hours at ambient temperature in a humid chamber, followed by 60 min incubation with secondary antibodies (DakoCytomation EnVision + System Labelled Polymer-HRP Anti-Mouse or Anti-Rabbit, Dako). Sections were mounted with Bio Mount medium (Bio-Optica, Milan, Italy) and observed under a Zeiss microscope.

Quantitation of double positive cells. Frozen muscle sections from IIM and controls were stained with antibody against TLR3 in combination with antibodies against the angiogenic marker aVb3 or MHCdev for regenerating myofibers (see Methods Section). Single- and double-positive cells were counted by two operators blinded to the diagnosis on 3 to 8 randomly selected adjacent field areas per section at x40 magnification. The counts of the two operators never differed by more than 5%.