THE SHRIMP EPIGENOME (shrimpENCODE) PROJECT: EFFORTS TO SEQUENCE THE GENOME OF TRUE SPECIFIC PATHOGEN-FREE (SPF) Litopenaeus vannamei FROM THE UNITED STATES
As part of The Shrimp Epigenome (shrimpENCODE) Project, initially funded by the FUCOBI Foundation of Quito, Ecuador, we aim to characterize the epigenetic mechanisms associated with Acute Hepatopancreatic Necrosis Disease (AHPND), also known as Early Mortality Syndrome (EMS). AHPND is caused by plasmid-containing Vibrio parahameolyticus (Vp) strains with insecticidal toxins PirAB that affect the stomach and hepatopancreas of susceptible SPF Litopenaeus vannamei. DNA transposition of insertion sequence (IS) mobile elements has been proposed as a potential epigenetic mechanism involved in AHPND susceptibility (1). We also aim at identifying and cataloguing methylation variable positions in control and AHPND-challenged animals using the large and complex transcriptome of SPF L. vannamei available in Genbank. However, due to the complexity of the SPF L. vannamei transcriptome, leaders of the 'Ag ENCODE Workshop' held Jan 10-11, 2014 in San Diego at the PAG meeting recommended that we focus first on finishing the assembly of L. vannamei genome sequence. We present here preliminary results from two of our collaborators, a story of their failures attempting to sequence the genome of the first and only stocks of L. vannamei that can truly be called 'SPF'. These stocks were from the breeding program of the U.S. Marine Shrimp Farming Program (USMSFP), a congressional initiative administered by the Cooperative State Research Service of the United States Department of Agriculture (2) of which Tufts University was a member. The consortium goal was to help American shrimp producers to develop the shrimp industry in the USA, and help reduce the yearly ~US$4.5 billion in trade deficit. Shrimp is the favorite seafood of Americans, and 95% of the shrimp consumed is still imported mostly from Southeast Asia and Latin America.
The EGI team will summarize their 'pervasive' transcriptome of the SPF L. vannamei which is susceptible to all the diseases in the OIE list, including Taura Syndrome Virus (TSV) that originated in Ecuador and spread quickly, but SPF stocks grow very fast, with good survival rates, and became the most economically important species farmed worldwide, until AHPND/EMS appeared. SPF stocks also bioaccumulate the EDC cadmium, increasing 16X when grown in closed recirculation systems. To begin to understand mechanisms of HGT transposition and virulence/transmission of genes in plasmids of EMS-causing Vp strains, the epigenetic modifications for shrimp genome need to be first characterized. Working towards this end, we need access to (a) the assembled L. vannamei genome sequence, (b) more cDNAs in the Unigene database, from studies that use shrimp from the coastal ecosystems where shrimp farms are also located, or grown in water containing different concentrations of EDCs, and (c) whole genome sequencing (WGS) assays recommended for epigenetics research by leading epigeneticists of the ENCODE program of US and EU who attended the "Ag ENCODE' workshop at the PAG meeting on January 10-11, 2014 in San Diego. Once this information is available, the epigenetic modifications associated with EMS and emerging diseases can be properly examined. In Vp pathogenic strains, the virulence genes are present in distinct 'pathogenicity islands' characterized by the presence of flanking inverted repeats, insertion sequences (IS) or tRNA genes that act as sites for recombination into the genome. Similar genomic signatures have been identified in the transcriptome of SPF L. vannamei exposed to cadmium and TSV, confirming the highly repetitive nature of the shrimp genomes, with large number of non-coding RNAs, inverted repeats and genetic signatures similar to those of TEs including the target site duplication motif (AGTT/CGTC) of non-LTR retrotransposons and the LINE-1 direct repeat (AAATACT). Some of our published SSRs (microsatellites) appear to be generated from non-LTR retrotransposon RTE. Also found: nuclear mitochondrial pseudogenes (NUMTs), pseudogenes of 16s rRNA+tRNA-Leu, 16s rRNA-like with (A)nCATGT motif that is not in shrimp mtDNA, a few 16s rRNA cDNAs are in Unigene clusters for hemocyanin (1 of the 12 new shrimp allergens of banana shrimp from Thailand), many chimeras of 16s rRNA with 99nt of DNaseI, and chimeric 12s rRNA containing small fragments of 16s rRNA, suggesting a rolling-circle mechanism similar to plant TEs and bacterial IS91 transposon. Putative conserved domains of chloramphenicol acetyltransferase (CAT) were found in ESTs from cadmium-exposed SPF L. vannamei postlarvae only, suggesting either contamination or caution for antibiotic resistance. Exposure to cadmium also induced expression of genes with unique genetic signatures, remnants of R2 repeats, known to insert into 28s RNA.
The Amplicon Express team started working on shrimp in late 2002 when Dr. Alcivar-Warren hired us to make cDNA library resources and a large insert genomic DNA BAC library from Penaeus vannamei (now called Litopenaeus vannamei). Amplicon scientists built several cDNA libraries, performed Sanger sequencing on a variety of cDNA clones. It wasn't until late 2003 that funding was obtained to begin work on the BAC library. In +18 years, Amplicon scientists led by Quanzhou Tao, have built more than 1,750 custom BAC libraries. In all this time, we only encountered two organisms from which we were unable to build a high quality BAC library: tetrahymena (possibly due to the ~78% AT rich sequence content) and L. vannamei. We have been trying to make a BAC library from this shrimp ever since (for several clients).
Over the course of several years, Amplicon scientists performed a suite of R&D experiments, made many empirical observations and tested a wide variety of hypothesis in hopes to identify the reason BAC library construction was failing and fix it. Following is a short recap of some of the work performed and the results. HMW DNA: we were able to obtain beautiful HMW DNA from several different tissue types (sperm and hemolymph were the best). These HMW DNA preparations looked suitable for BAC library construction. We performed partial digestions and obtained excellent partial digestion results and Pulsed Field Gel (PFG) showed fragments of high quality DNA at appropriate sizes, several ranges were perfect for a BAC library (so we selected a series of different sizes, to determine the maximum insert size and still maintain a high transformation frequency). After ligation with appropriate BAC vector, the DNA was electroporated into DH10b E. coli host cells and plated on agar media with antibiotics. We expected between 1,500 colonies and 10,000 colonies and we obtained ZERO white colonies. This was a shocking result. We tested all reagents with controls and repeated the same results many times.
We performed myriad experiments over the next 10 years and obtained the same results: zero white colonies growing on the agar plates. We were brainstorming with clients, colleagues and vendors. We knew the DNA cut right, we knew it ligated properly, but it would not grow inside the E. coli host. We tried and tested many hypotheses: trying various endonuclease enzymes that were not sensitive to methylation, ligating into a variety of BAC vectors, maybe there is SAP (Shrimp Alkaline Phosphatase) active in the DNA prep and preventing ligation so we tried processing insert DNA with kinase to insure phosphorylation, transforming into various E. coli host strains with different properties, we tried using additional DNA cleanup steps in hopes to remove any unusual protein complexes that might be binding/blocking the DNA from replicating in a host cell. Along these same lines, we tried experiments to remove any RNA molecules that might be bound to the DNA (and blocking replication of the BAC plasmid inside the E. coli host). We tried using Whole Genome Amplification enzymes (i.e. Templify) and this did not give us white colonies. There was one set of experiments, from which we he had a few hundred BAC clones total, however it turned out these were bacterial in origin. For this experiment, HMW DNA was extracted from the entire shrimp and we suspect the clones obtained were from bacterial organisms growing inside the shrimp gut.
The only way this story has a "Hollywood ending": after 12 years, we discovered that the HMW DNA we made for the BAC library construction was ideal for making 20Kb PacBio SMRTCell libraries and running on the Pacific Bioscience RSII platform with P5 chemistry!
Preliminary review of this data by Pacific Biosciences and TGI-WU collaborators confirmed the complexity of the shrimp genome, full of repetitive sequences, and AT-rich. Although multiple duplications of large DNA segments and a complex transcriptome is reported for SPF L. vannamei, the genome of 2.89 pg/nucleus does not appear to be duplicated and none of the tissues were found to contain endopolyploid cells (3). In order to generate a good assembly of this difficult genome, collaborators at TGI-WU recommend a two-prong approach that combines Fosmid Pooling with PBJelly gap-filling, and 2X250 sequencing and DiscoVar assembly of the genome, and then merge with the fosmid assembly. The exact details of the ideal Whole Genome Sequencing strategy for Litopenaeus vannamei are still being formulated as additional high quality NGS data are currently being generated and analyzed.
References
(1) Acacia Alcivar-Warren and collaborators of The Shrimp Epigenome Project (shrimpENCODE). 2014. ONE HEALTH: Is EMS/AHPND a potential risk to human health? - a review of epigenetic mechanisms associated with emerging shrimp diseases and antibiotic resistance, and a call for increased funding for basic research on the shrimp epigenome. Proceedings of the Aquaculture America 2014 meeting, Seattle, WA, abstract 28: https://www.was.org/meetingabstracts/ShowAbstract.aspx?Id=32080
(2) U.S. Marine Shrimp Farming Program. 1992. An annotated bibliography of Consortium Sponsored Publications. Compiled by Gary Pruder. A publication of the Gulf Coast Research Laboratory Consortium. The Oceanic Institute, P.O.Box 25280, Honolulu, HI, USA 96825.
(3) Jeffery, NW, Gregory T. Ryan, A. Alcivar-Warren. 2014. Synergy between aquaculture programs, genome size and polyploidy assessments: benefits for both fields. Proceedings of the Aquaculture America 2014 meeting, abstract 489: https://www.was.org/meetingabstracts/ShowAbstract.aspx?Id=31961