Salmo salar

ChIP assays were performed using a kit following the manufacturer’s instructions (#17-295; Millipore). Accordingly, cells were first treated with 1% formaldehyde and incubated at 37°C for 10 min. Next, cells were washed twice with ice-cold PBS containing protease inhibitors. Cells were then scraped and pelleted by centrifugation at 2,000 RPM for 4 min at 4°C. Then, cells were resuspended in SDS Lysis Buffer (Millipore, #20-163) and incubated for 10 min on ice. After samples were centrifuged for 10 min at 13,000 RPM at 4°C, the supernatants were diluted 10 times by adding ChIP Dilution Buffer (#20-153; Millipore) containing protease inhibitors. The diluted supernatants were then treated with 75 μl of a 50% slurry of Protein-A Agarose/Salmon Sperm DNA (#16-157C; Millipore) at 4°C for 30 min with agitation. After centrifugation, supernatants were immunoprecipitated with antibodies against Smad4, Smad2, Smad3, Prdm16, GAPDH or isotype-matched control IgG and 60 μl of a 50% slurry of Protein-A Agarose/Salmon Sperm DNA at 4°C for 1 h with rotation. Agarose was pelleted using centrifugation at 1,000 RPM for 1 min at 4°C. The pellets were washed for 5 min in Low Salt Immune Complex Wash Buffer (#20-154; Millipore) once, High Salt Immune Complex Wash Buffer (#20-155; Millipore) once, LiCl Immune Complex Wash Buffer (#20-156; Millipore) once, and TE Buffer (#20-157; Millipore) twice. To amplify DNA bound to the immunoprecipitates, elution buffer (1% SDS, 0.1 M NaHCO₃) was added to each sample followed by agitation and incubation for 15 min with rotation at room temperature. Eluates were then mixed with NaCl (final concentration of 0.2 M) and incubated for 4 h at 65°C followed by adding EDTA (0.01 M), Tris-HCl, pH 6.5 (0.04 M), and Proteinase K (0.04 mg/ml). Samples were then incubated for 1 h at 45°C, and DNA was recovered using phenol/chloroform extraction coupled with ethanol precipitation. Pellets were washed with 70% ethanol and air-dried. Lastly, pellets were resuspended in an appropriate buffer for PCR, and PCR products were analyzed on a 2% agarose gel. The immunoprecipitated DNA was also analyzed by qPCR using locus specific primers and normalized to input DNA. Relative fold enrichment in each locus was quantified relative to the control as described above (qRT-PCR) as well as in our published studies (Parajuli et al., 2018 (link); Zhang et al., 2015 (link)). The following primers were used: PRDM16-For 5′-CAT​CTC​CCC​AGC​ATT​GTC​AGT-3′; PRDM16-Rev 5′-GGA​GCG​CCG​AAC​ACG​GAA​TG-3′; JUNB-For 5′-GGC​AAA​GCC​CAG​GGT​CAA​TA-3′; JUNB-Rev 5′-AAA​GCT​AGT​AAG​CGG​CCT​GG-3′; GAPDH-For 5′-CGG​GAT​TGT​CTG​CCC​TAA​TTA​T-3′; GAPDH-Rev 5′-GCA​CGG​AAG​GTC​ACG​ATG​T-3′.

Total RNA was extracted separately from testis (n = 4) and ovary (n = 4) tissues using TRIzol (Invitrogen). For each sample, RNA quality and concentration were assessed using agarose gel electrophoresis, a NanoPhotometer spectrophotometer (Implen, CA), a Qubit 2.0 Fluorometer (ThermoFisher Scientific), and an Agilent BioAnalyzer 2,100 system (Agilent Technologies, CA), requiring an RNA integrity number (RIN) of 8.5 or higher; one ovary sample failed to meet these quality standards and was excluded from downstream analyses. Sequencing libraries were generated using the NEBNext Ultra RNA Library Prep Kit for Illumina following the manufacturer’s protocol. After cluster generation of the index-coded samples, the library was sequenced on one lane of an Illumina Hiseq 4,000 platform (PE 150). Transcriptome sequences were filtered using Trimmomatic-0.39 with default parameters (Bolger et al., 2014 (link)). 30, 848, 170 to 39, 695, 323 reads were retained for each testis or ovary sample, and in total, 290, 925, 984 reads remained, with a total length of 42, 385, 060,050 bp. Remaining reads of all testis and ovary samples were combined and assembled using Trinity 2.12.0 (Haas et al., 2013 (link)), yielding 573,144 contigs (i.e., putative assembled transcripts). Contigs were clustered using CD-hit-est (95% identity). Completeness of this final de novo transcriptome assembly were assessed using the BUSCO pipeline (Simao et al., 2015 (link)).
Expression levels of contigs in each sample were measured with Salmon (Patro et al., 2017 (link)), and contigs with no raw counts were removed. To annotate the remaining contigs containing autonomous TEs, BLASTp and BLASTx were used against Repbase with an E-value cutoff of 1E-5 and 1E-10, respectively. The aligned length coverage was set to exceed 80% of the queried transcriptome contigs. To annotate contigs containing non-autonomous TEs, RepeatMasker was used with our Ranodon-derived genomic repeat library of non-autonomous TEs (LARD-, TRIM-, MITE-, and SINE-annotated contigs) and the requirement that the transcriptome/genomic contig overlap was >80 bp long, >80% identical in sequence, and covered >80% of the length of the genomic contig. Contigs annotated as conflicting autonomous and non-autonomous TEs were filtered out.
To identify contigs that contained endogenous R. sibiricus genes, the Trinotate annotation suite (Bryant et al., 2017 (link)) was used with an E-value cutoff of 1E-5 for both BLASTx and BLASTp against the Uniport database, and 1E-5 for HMMER against the Pfam database (Wheeler and Eddy, 2013 (link)). To identify contigs that contained both a TE and an endogenous gene (i.e., putative cases where a TE and a gene were co-transcribed on a single transcript), all contigs that were annotated both by Repbase and Trinotate were examined, and the ones annotated by Trinotate to contain a TE-encoded protein (i.e., the contigs where Repbase and Trinotate annotations were in agreement) were not further considered. The remaining contigs annotated by Trinotate to contain a non-TE gene (i.e., an endogenous Ranodon gene) and also annotated either by Repbase to include a TE-encoded protein or by blastn to include a non-autonomous TE were filtered out for the expression analysis.