Hydractinia symbiolongicarpus assembled transcriptome

Hydractinia symbiolongicarpus colonies were collected in Woods Hole, MA (USA), and were cultured in 40 L aquaria with recirculating artificial sea water (Instant Ocean Sea Salt) at of 20 ± 2°C and a relative density of 1,022, and water replacement of 25% every three days. In order to increase the representation of both inducible and repressible immune-related transcripts, two types of samples were used to construct cDNA libraries for sequencing; one sample was challenged with live bacteria whereas the other was left without immune challenge. For the bacterially challenged sample, a male colony (HWB103) was starved for three days and subsequently incubated with 108 cells/ml of a LB stationary phase co-culture of a Gram-positive (Microbacterium sp.) and a Gram-negative (Vibrio sp.) bacterium, for one hour at room temperature. After the bacterial challenge, the colony was washed with sterile artificial seawater, and tissue samples were collected in TRIzol Reagent (Invitrogen) for total RNA extraction. In order to increase the chances of sequencing cDNAs from inducible transcripts with different transcriptional times, samples were collected at 1, 3, 6, 12, 24 and 36 hours post-exposure.

Total RNA was isolated from approximately 0,2 g of each tissue sample using TRizol Reagent (Invitrogene). Three types of RNA samples were prepared to construct the cDNA libraries as follows: i) early time post-challenge (ET) RNA, by pooling RNA samples from 1, 3 and 6 hours after bacterial challenge, ii) late time post-challenge (LT) RNA, by pooling RNA samples from 12, 24 and 36 hours after challenge, and iii) RNA from a non-challenged sample (C). cDNAs were synthesized from poly-A transcripts in each RNA class. They were subsequently fragmented by sonication, and fragments of about 300 bp long were gel-purified to add Illumina adaptors, tags, and primers. The resulting three libraries were sequenced in a single HiSeq 2000 Illumina PET indexed lane at the Canada’s Michael Smith Genome Sciences Center - British Columbia Cancer Agency, Canada.

Quality of raw sequence data was evaluated using the FastQC software V. 0.10.1. Illumina adaptors and bases with a Phred score under 20 at the 3’ ends were trimmed out using the FastXToolkit (http://hannonlab.cshl.edu/ fastx_toolkit/index.html), and only reads longer than 30 bp were selected for further analysis. Ribosomal sequence depletion was performed using the RiboPicker software (Schmieder et al., 2011), based on the ribosomal sequence databases Silva (Quast et al., 2013), GreenGenes (DeSantis et al., 2006), RDP (Maidak et al., 2000), Rfam (Griffiths-Jones et al., 2003), and the NCBI. To reduce redundancy and the noise created by sequencing errors the data was digitally normalized using the Diginorm tool. The resulting raw sequences from the three cDNA libraries (ET, LT, and C) were combined to assemble the H. symbiolongicarpus transcriptome using the Trinity software (Grabherr et al., 2011).