Reagents
RPMI-1640 medium, Dulbecco′s Modified Eagle′s Medium (DMEM), Minimum Essential Medium Eagle (EMEM), Stemline Keratinocyte Medium II, Stemline Keratinocyte Growth Supplement, Fetal Bovine Serum (FBS), Penicillin-streptomycin (PEST), sodium pyruvate (SP) and Trypsin-EDTA solution were purchased from Sigma Aldrich (St. Louis, MO, USA).
RNA extraction was performed with the RNeasy Mini Kit from QIAGEN including DNA digestion with DNase set (Hilden, DE) and reverse transcription PCR (RT-PCR) with the High Capacity cDNA Reverse Transcription Kit from Applied Biosystems (Foster City, CA, USA). Both RNA and DNA samples were eluted in nuclease free water from VWR (Radnor, PA, USA).
For the droplet digital polymerase chain reaction (ddPCR): ddPCRTM Supermix for Probes (No dUTP), ddPCRTM Droplet Generation Oil for Probes, Droplet Reader Oil and two different Prime PCR Assays were all purchased from Bio-Rad Laboratories (Hercules, CA, USA). Prime PCR Assay for PDIA3-Normal (PDIA3) consist in a forward primer (5´-GTGTGGCGCTGCTTCTTG- 3´), a reverse primer (5´-AAGAACTCGACGAGCATGAG-3´) and a FAM probe or internal primer FAM (5´-GCCTCGCCGCTGCCTCCGAC-3´). Prime PCR Assay for PDIA3-Novel (PDIA3N) consist in a forward primer (5´-GGCAGTGGATTTTAGTCCCA-3´), a reverse primer (5´-CAAGTCTCTTGCAGTGTCCA-3´) and a HEX probe or internal primer HEX (5´- ACACACACACCTGGTTGCTCCCCAGA -3´). Positive target controls for the ddPCR experiment consisted in DNA fragments or oligos with the same sequence as the target (PDIA3 and PDIA3N respectively).
Cell culture
Cells cultured for further experiments were ranging in the severity of the prostate cancer disease. PNT2 (Control), normal prostatic epithelial cells well differentiated; P4E6 (Early stage), immortalized human prostate cell line derived from a biopsy of a well-differentiated early stage prostate cancer. DU145, prostate derived from metastatic site in the brain (Prostate adenocarcinoma, grade II, AR negative). PC3, prostate cell line derived from bone metastasis (Prostate adenocarcinoma, grade IV, AR negative). LNCaP from a lymph node metastasis (carcinoma metastatic, AR positive).
PNT2, P4E6 and LNCaP were obtained from Sigma Aldrich. PNT2 and LNCaP cell lines were cultured in RPMI-1640 complete medium supplemented with 10% FBS and 1% PEST. The cell lines DU145 and PC3 were obtained from ATCC (Manassas, VA, USA). P4E6 cells were cultured in Stemline Keratinocyte Medium II with Stemline Keratinocyte Growth Supplement, 2mM Glutamine and 2% of Foetal Bovine Serum (FBS). Addition of 2% serum was found to increase cell viability. DU145 cells were cultured in EMEM supplemented with 10% FBS and 1% PEST. PC3 cells were cultured in DMEM containing 10% FBS, 5% of pyruvate sodium and 1% PEST. All the cells were cultured in T25 flasks and culture adherent cells were detached with Trypsin-EDTA solution and maintained at 37°C in 5% CO2.
RNA extraction and RT-PCR
Total RNA was extracted from one million cells for each cell line: PNT2, P4E6, DU145, PC3 and LNCaP by following the protocol for the RNeasy Mini Kit from Qiagen including genomic DNA digestion with DNase. Samples were eluted in RNase-Free Water and stored at -80ºC. 20 µl of each cell line sample that fulfilled the quality standards (A260/280= 1.9-2.0, A260/230= 1.5-1.8, RNA integrity number ≥ 8) was sent to The National Genomic Infrastructure (NGI) in The Science Life Lab in Stockholm, Sweden.
From the PNT2, DU145, PC3 and LNCaP samples, RNA (up to 2 µg) was reversed transcribed to cDNA according to the protocol for the High Capacity cDNA Reverse Transcription Kit from Applied Biosystems. The obtained cDNA samples were used as a reaction template in the ddPCR absolute quantification experiment.
NGS
Five RNA samples (PNT2, P4E6, DU145, PC3 and LNCaP) were sequenced with Illumina HiSeq 2500 High Output v4, 2x125bp in one lane giving >37.6M read pairs/sample. The library used was TruSeq Stranded Total RNA Illumina RiboZero and was also performed in the NGI. Data delivered from the NGS included demultiplexing, quality control and raw data.
The Software TopHat/2.0.4 was used for mapping reads to the Human genome assembly, build GRCh37 (hg19). The output of the mapping was BED files, which comprises the splice junctions reported by the TopHat algorithm; and “accepted hits” BAM files, which contains a list of the reads aligned to the reference genome. Both files were sorted and indexed with Samtools. Quantification of normalized expression values were obtained as FPKM (Fragments per Kilobase of Exon per Million Fragments Mapped) values generated by cufflinks/2.1.1. FPKMs values were obtained from genes and their different transcript isoforms.
Droplet digital TM PCR (ddPCR)
Droplet digital polymerase chain reaction (ddPCR; QX200, Bio-Rad, Hercules, CA, USA) was performed according to the manufacturer´s instructions. A total of 178 cDNA samples were analyzed to quantify the amount of PDIA3-Normal and PDIA3-Novel target. Each sample was partitioned into 8000-15000 droplets, there every droplet can be considered as one PCR reaction, with target and background DNA randomly, but uniformly, distributed among the droplets. The reactions were performed in 25 µl reaction volumes that consisted of up to 10 µl and 330ng of template cDNA, 12.5 µl of ddPCR supermix for probes (No dUTP), 1.25 µl of PDIA3-Normal Prime PCR Assay, 1.25 µl of PDIA3-Novel Prime PCR Assay and DNase free water (variable volume). The non-template controls (NTC) contained 8 μl of purified water instead of cDNA template and each positive control contained a complementary single DNA sequence or oligo to the PDIA3-Normal isoform and to the PDIA3-Novel isoform respectively. 20 µL of each reaction mix was loaded into the DG8 Cartridge and 70 µL of droplet generation oil onto QX200 Droplet Generation system for droplet generation. Total amount of droplets of each reaction was transferred to a 96 well plate to perform the PCR amplification. In order to determine the optimal annealing temperature for the PCR, thermal gradient experiments were performed with one sample (Additional file 1). After evaluating which temperature that showed the best separation between positive droplets with the target and negative droplets, the PCR temperature conditions were set to 95 °C for 10 min (1 cycle), 94 °C for 30 s, 58.1 ° C for 60 s (40 cycles), 98 °C for 10 min and infinite hold at 4 °C. After that the PCR 96-well plate was read by the QX200 Droplet Reader and the type of experiment was set to absolute quantification. The Poisson-corrected determination of template concentration (copies/ µl) and the ration between both isoforms (PDIA3-Novel/PDIA3-Normal) was calculated using QuantaSoft™ Analysis Pro Software (v1.0.596, Bio-Rad).
Statistical analysis
Differences between sample groups and template concentrations were assessed by non-parametric Kruskal-Wallis multiple comparisons test (p<0.05) in GraphPad Prism version 7.00 for Windows (GraphPad Software, La Jolla California USA). Groups pairwise comparisons were assessed by Mann-Whitney U test (p>0.05). These tests were selected assuming that the data did not follow a normal (Gaussian) distribution (Kolmogorov-Smirnov test, P value <0.0001).
Functional analysis
Protein sequences were retrieved from UniprotKB (P30101) for PDIA3 and from UniParc (UPI000066D935) for PDIA3N (24). Both protein sequences were analysed by the I-TASSER (Iterative Threading ASSEmbly Refinement server) for generating the protein structure model (25). Differences in the protein sequences were highlighted with UCSF Chimera 1.0 (26). A prediction of the damage or pathogenicity of the 56 variations (amino acid substitutions and deletions) that PDIA3N present in comparison with PDIA3 was carried out with PolyPhen-2 (27) and PROVEAN (Protein Variation Effect Analyzer) v1.1 (28) software tools. The threshold score used to determine that the variant was pathogenic in PolyPhen-2 (HumDiv and HumVar algorithms) was 0.403 and in PROVEAN was -2.5. In PolyPhen-2 an amino acid substitution was considered “Possibly Damaging” if the score was higher than 0.403 and “Probably Damaging if it was higher than 0.975. In PROVEAN an amino acid substitution/deletion was considered “Deleterious” in PROVEAN if the score was under -2.5. A prediction of the subcellular localization of PDIA3 and PDIA3N was performed by DeepLoc-1.0 (29).