Agilent 2100 bioanalyser instrument

Microsatellite DNA fragments were amplified from spheroid DNA samples using Multiplex PCR Kit (Qiagen) and primer pairs for the Bethesda panel markers [2 (link), 34 (link)] (Supplementary Table 4). Three electrophoretic runs were performed for each sample (BAT25/D2S123, D5S346/D17S250, and BAT26) according to the standard protocol [16 (link)] (Supplementary Table 5). Amplified DNA fragments were purified with QIAquick PCR Purification Kit (Qiagen), applied on the DNA LabChip of the Agilent DNA 1000 Kit (Agilent Technologies, Santa Clara, CA, USA), and analyzed with an Agilent 2100 Bioanalyser instrument (Agilent Technologies) according to the previous protocol [16 (link)]. The electropherogram of each cancer spheroid line was overlaid with that of the normal epithelial spheroid line derived from the same patient. A marker was diagnosed as unstable if any of the cancer peaks for the marker shifted more than 0.5 seconds. The MSI status of each case was determined as follows: MSI-high (MSI-H), with more than one unstable markers; MSI-low (MSI-L), with one unstable marker; and MSS, with no unstable markers. If only dinucleotide repeat markers as D2S123, D5S346 and/or D17S250 were mutated, a secondary panel of markers with mononucleotide repeats (BAT40 and MYCL) was tested as recommended [1 (link)].

Total RNA from RASF and OASF (n = 9 each, see Table 1 for patients characteristics) was isolated with the miRNeasy Mini kit and treated with RNase-free DNase (Qiagen). The RNA quality was assured using the Agilent RNA 6000 Nano kit with Agilent 2100 Bioanalyser instrument (Agilent Technologies). RNA with RNA integrity number (RIN) values ≥9.5 was considered acceptable for the study. Small RNA library preparations and sequencing using HiSeq2500 (Illumina Inc., CA) were performed at the Functional Genomic Center Zurich according to the Illumina TruSeq® Small RNA protocol. After removal of the adaptor sequences the small RNA-seq reads were mapped to the human genome by the alignment program Bowtie [21 (link)] and aligned to the piRNABank–database of 23`439 known human piRNA sequences.[5 (link)] Reads Per Kilobase per Million mapped reads (RPKM) of piRNAs identified in RA were compared to RPKM in OA using the SARTools pipeline based on edgeR package.[22 ] piRNAs were considered differentially expressed when the adjusted p-value was ≤ 0.05.