Genomic analysis of uncultured marineviral communitiesMya Breitbart*, Peter Salamon†, Bjarne Andresen†‡, Joseph M. Mahaffy†, Anca M. Segall*, David Mead§, Farooq Azam¶,and Forest Rohwer*储*Department of Biology, San Diego State University, San Diego, CA 92182-4614;†Department of Mathematical Sciences, San Diego State University,San Diego, CA 92182-7720;‡Ørsted Laboratory, University of Copenhagen, Universitetsparken 5, DK-2100 Copenhagen Ø, Denmark;§Lucigen, Middleton, WI 53562; and¶Marine Biology Division, Scripps Institution of Oceanography, La Jolla, CA 92093Communicated by Allan Campbell, Stanford University, Stanford, CA, August 14, 2002 (received for review February 22, 2002)Viruses are the most common biological entities in the oceans byan order of magnitude. However, very little is known about theirdiversity. Here we report a genomic analysis of two unculturedmarine viral communities. Over 65% of the sequences were notsignificantly similar to previously reported sequences, suggestingthat much of the diversity is previously uncharacterized. The mostcommon significant hits among the known sequences were toviruses. The viral hits included sequences from all of the majorfamilies of dsDNA tailed phages, as well as some algal viruses.Several independent mathematical models based on the observednumber of contigs predicted that the most abundant viral genomecomprised 2–3% of the total population in both communities,which was estimated to contain between 374 and 7,114 viral types.Overall, diversity of the viral communities was extremely high. Theresults also showed that it would be possible to sequence theentire genome of an uncultured marine viral community.Marine viruses, the majority of which are phages, haveenormous influences on global biogeochemical cycles (1),microbial diversity (2, 3), and genetic exchange (4). Despite theirimportance, virtually nothing is known about marine viral biodi-versity or the evolutionary relationships of marine and nonma-rine viruses (5–7). Addressing these issues is difficult becauseviruses must be cultured on hosts, the majority of which cannotbe cultivated by using standard techniques (8). In addition,viruses do not have ubiquitously conserved genetic elementssuch as rDNA that can be used as diversity and evolutionarydistance markers (9). To circumvent these limitations, we de-veloped a method to shotgun clone and sequence unculturedaquatic viral communities.Materials and MethodsIsolation of Viral Community DNA. Marine viruses were isolatedfrom 200 liters of surface seawater from Scripps Pier (SP, LaJolla, CA; May 2001) and the channel side of Fiesta Island inMission Bay (MB, San Diego; June 2001) by using a combinationof differential filtration and density-dependent gradient centri-fugation. The water at the MB site is exchanged with each tidalcycle. Both the SP and MB sites experience increased levels ofpollution during the rainy season, because of runoff from thesurrounding city. The MB site routinely has more eukaryoticalgae than does the SP site. Once collected, the water sampleswere initially filtered through a 0.16-m Centramate tangentialflow filter (TFF; Pall) to remove bacteria, eukaryotes, and largeparticles. Approximately 90% of the viruses, as determined byepifluorescent microscopy (10), and most of the water, passedthrough the filter and were collected in a separate tank. Subse-quently, the viruses in the filtrate were concentrated by using a100-kDa TFF filter until the final sample volume was ⬍100 ml(⬇5,000⫻ concentration). Recovery of viruses during this stepwas essentially 100%. After the TFF, the phage concentrate wasloaded onto a cesium chloride (CsCl) step gradient, ultracen-trifuged, and the 1.35–1.5 g兾ml fraction was collected. Thisfraction contains the majority of the viral DNA as determinedby pulse field gel electrophoresis (11); however, this method willnot recover all viruses (e.g., large eukaryotic viruses and ssRNAphages). After CsCl purification, the viruses were lysed by usinga formamide extraction, and the DNA was recovered by anisopropanol precipitation and a cetyltrimethylammonium bro-mide (CTAB) extraction (12).Construction of the Shotgun Library. The amount of viral DNA inan environmental sample is very low (⬇10g兾100 liters). Viralgenomes often contain modified nucleotides that cannot bedirectly cloned into Escherichia coli. Additionally, because viralgenomes contain genes (e.g., holins, lysozyme) that must bedisrupted before cloning, we have not been able to create arepresentative cosmid library from these communities. We havecircumvented these problems by randomly shearing the totalmarine viral community DNA (HydroShear, GenMachine, SanCarlos, CA), end-repairing, ligating dsDNA linkers to the ends,and amplifying the fragments by using the high-fidelity VentDNA polymerase. The resulting fragments were ligated into thepSMART vector and electroporated into MC12 cells (Lucigen,Middleton, WI). We call these libraries LASLs for linker-amplified shotgun libraries. This method has been checked toensure randomness as described (ref. 13, and our web site atwww.sci.sdsu.edu兾PHAGE兾LASL兾index.htm). A test librarywas constructed of coliphageDNA, and 100 fragments weresequenced without observing any chimeras. Additionally, wehave recently sequenced two phage genomes, Vibriophage 16Tand 16C, from a mixed lysate by using the LASL approach. Nochimeric molecules were observed in this mixed library. To-gether, these three libraries represent ⬎1,000 sequences. There-fore, it is highly unlikely that we are observing a significantnumber of chimeric sequences in our library (F.R. and A.M.S.,unpublished results).Analysis of Sequences: Composition Analyses. Clones from the SPlibrary were sequenced with both forward and reverse primers,yielding a total of 1,061 sequences. Eight hundred and seventy-three clones from the MB library were sequenced only with theforward primer. These sequences were compared against Gen-Bank by usingTBLASTX (14, 15). A hit was consider significantif it had an E value of ⬍0.001. Significant hits to GenBank entrieswere classified into the groups described in the text, based onsequence annotation. In cases where multiple significant hitswere observed for a single query sequence, the sequence waspreferentially classified as a phage or virus if these occurredwithin the top five hits. Mobile elements consisted of trans-posons, plasmids, insertion
View Full Document
Unlocking...