UBC Faculty Research and Publications

Genomic sequence of a mutant strain of Caenorhabditis elegans with an altered recombination pattern Rose, Ann M; O'Neil, Nigel J; Bilenky, Mikhail; Butterfield, Yaron S; Malhis, Nawar; Flibotte, Stephane; Jones, Martin R; Marra, Marco; Baillie, David L; Jones, Steven J Feb 23, 2010

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12864_2009_Article_2725.pdf [ 1.77MB ]
JSON: 52383-1.0223856.json
JSON-LD: 52383-1.0223856-ld.json
RDF/XML (Pretty): 52383-1.0223856-rdf.xml
RDF/JSON: 52383-1.0223856-rdf.json
Turtle: 52383-1.0223856-turtle.txt
N-Triples: 52383-1.0223856-rdf-ntriples.txt
Original Record: 52383-1.0223856-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessGenomic sequence of a mutant strain ofCaenorhabditis elegans with an alteredrecombination patternAnn M Rose1*, Nigel J O’Neil1, Mikhail Bilenky2, Yaron S Butterfield2, Nawar Malhis2, Stephane Flibotte2,Martin R Jones1, Marco Marra2, David L Baillie3, Steven JM Jones2AbstractBackground: The original sequencing and annotation of the Caenorhabditis elegans genome along with recentadvances in sequencing technology provide an exceptional opportunity for the genomic analysis of wild-type andmutant strains. Using the Illumina Genome Analyzer, we sequenced the entire genome of Rec-1, a strain that altersthe distribution of meiotic crossovers without changing the overall frequency. Rec-1 was derived fromethylmethane sulfonate (EMS)-treated strains, one of which had a high level of transposable element mobility.Sequencing of this strain provides an opportunity to examine the consequences on the genome of altering thedistribution of meiotic recombination events.Results: Using Illumina sequencing and MAQ software, 83% of the base pair sequence reads were aligned to thereference genome available at Wormbase, providing a 21-fold coverage of the genome. Using the softwareprograms MAQ and Slider, we observed 1124 base pair differences between Rec-1 and the reference genome inWormbase (WS190), and 441 between the mutagenized Rec-1 (BC313) and the wild-type N2 strain (VC2010). Themost frequent base-substitution was G:C to A:T, 141 for the entire genome most of which were on chromosomes Ior X, 55 and 31 respectively. With this data removed, no obvious pattern in the distribution of the base differencesalong the chromosomes was apparent. No major chromosomal rearrangements were observed, but additionalinsertions of transposable elements were detected. There are 11 extra copies of Tc1, and 8 of Tc2 in the Rec-1genome, most likely the remains of past high-hopper activity in a progenitor strain.Conclusion: Our analysis of high-throughput sequencing was able to detect regions of direct repeat sequences,deletions, insertions of transposable elements, and base pair differences. A subset of sequence alterations affectingcoding regions were confirmed by an independent approach using oligo array comparative genome hybridization.The major phenotype of the Rec-1 strain is an alteration in the preferred position of the meiotic recombinationevent with no other significant phenotypic consequences. In this study, we observed no evidence of a mutatoreffect at the nucleotide level attributable to the Rec-1 mutation.BackgroundCaenorhabditis elegans is an animal model widely usedin biomedical and biological research. C. elegans was thefirst animal to have its genome completely sequenced[1] and the compiled and annotated sequence is avail-able at WormBase http://www.wormbase.org. The readyavailability of genomic sequence information along withan extensive body of knowledge about gene function inthis species provides an exceptional opportunity toexamine the consequences of mutational change on thecomposition of the genome. High-throughput sequen-cing of wild type [2,3] and mutant strains [4] hasdemonstrated the diverse benefits of examining genomicsequence. Not only is genome-wide sequencing valuablefor finding the mutational basis of phenotypic change,but also for understanding evolutionary processes. Den-ver et al. [3] identified and characterized base-substitu-tion mutations that arose spontaneously in 10 lines of* Correspondence: ann.rose@ubc.ca1Department of Medical Genetics, University of British Columbia, 419 - 2125East Mall, Vancouver, BC, V6T 1Z4, CanadaRose et al. BMC Genomics 2010, 11:131http://www.biomedcentral.com/1471-2164/11/131© 2010 Rose et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction inany medium, provided the original work is properly cited.C. elegans, providing us with a fuller understanding ofthe nature of genome-wide base-substitution events.A question that has long been debated is the relation-ship of mutational patterns to biological processes suchas meiotic recombination [5,6]. In C. elegans, the centralportions of the five autosomes are relatively gene densecompared to the arms [1]. Furthermore, traditionalgenetic approaches using forward mutational screens torecover lethal alleles of essential genes have showngenes in the central clusters of chromosomes I and V tobe more mutable to lethality than the arms [7]. How-ever, the most striking feature of the C. elegans auto-somes is the recombination suppression associated withthe central gene clusters, [8] reviewed in [6]. In wild-type, the frequency of crossing over per length of DNAvaries as much as ten-fold between the cluster and anarm of chromosome I [9], making this species an excel-lent model for studying the relationship betweensequence variation and recombination rate.The recombinational suppression of the gene clustersis eliminated in the mutant Rec-1 [10], resulting inincreased crossing over in the autosomal central regionsand a compensatory decrease in the arms [9]. The con-sequence is an altered distribution of meiotic exchangeevents while retaining the same overall number. In Rec-1, the genetic recombination map resembles more clo-sely the physical length of the chromosome than it doesthe wild-type pattern of crossovers. The phenotype wasoriginally identified as a recessive mutation in a strainheterozygous for morphological markers in the centralcluster of chromosome I, dpy-5(e61) unc-15(e73) +/++unc-13(e51). A three-fold increase in crossing over wasobserved in the central region of the autosomes [10].The visible markers were eventually eliminated byrecombination resulting in a wild-type appearing strain,BC313, for which the major phenotype is an altered dis-tribution of recombination, affecting both exchange offlanking markers and apparent intragenic gene conver-sion [11]. There are no detrimental effects on growth,progeny number or spontaneous mutation rate. Nondis-junction of the X-chromosome is elevated somewhat,but not dramatically. The rec-1(s180) mutation is inher-ited as a Mendelian recessive, and crossover distributionis altered for the entire genome [10], including the X-chromosome, despite the fact that it has a more uniformdistribution of recombination events (V. Vijayaratumand AMR, unpublished data). The consequence of themutation is that the recombination map in Rec-1 moreclosely reflects the physical map than the genetic mapin wild type [9] (Figure 1).In this paper, we used the high-throughput Solexaplatform (Illumina) to sequence the genome of the Rec-1 strain. This study provides the first opportunity toexamine the consequences on a genome of altering thedistribution of meiotic recombination events.ResultsBase Pair composition of Rec-1 compared to WormBaseand VC2010There were a total of 60,601,198 forty-two base pairsequence reads, of which 50,595,466 (83%) were alignedto the WormBase reference genome WS190 using MAQsoftware [12] with a maximum of two mismatches perread resulting in approximately 21-fold redundantsequence coverage. Base pair differences were calledusing both MAQ [12] and Slider [13] software.Figure 1 Comparision of the Genetic and Physical Maps of chromosome I. The top line shows the wild-type (N2) genetic map of autosomeI of C. elegans using genetic distances measured by Zetka and Rose, 1995. Line 2 is the position of the gene markers on the physical map asannotated in WormBase http://www.wormbase.org. The bottom line is the position of markers in the Rec-1 mutant (data taken from Zetka andRose, 1995).Rose et al. BMC Genomics 2010, 11:131http://www.biomedcentral.com/1471-2164/11/131Page 2 of 12We observed 1124 base pair differences between Rec-1(BC313) and WormBase WS190, and 441 between Rec-1and the wild-type strain VC2010. Fourteen of theobserved differences were tested by either PCR or directsequencing and all fourteen confirmed. In this paper, weanalyzed those differences that were identified by bothMAQ and Slider compared to VC2010. The canonicalsequence of C. elegans, archived in WormBase, is valu-able because it is compiled, annotated and readily acces-sible. The WormBase reference sequence [1] wasobtained from cloned cosmids and Yacs, which for tech-nical reasons came from different strains, and is not thegenome sequence of any one existing strain. Thus, wecould not experimentally determine the allelic status ofrec-1 for the WormBase reference sequence. Since manystrains of C. elegans carry the s180 allelic variant of rec-1, we could not simply assume the reference sequencewas wild type. Thus, in this paper we examine the dif-ferences between Rec-1 and VC2010, a strain that weconfirmed by measuring meiotic crossing over to bewild-type for rec-1.The C. elegans genome is approximately 100 millionbase pairs (Mbp) in size. We observed a base pair differ-ence approximately every 225,000 bps on average. Thenumber of base pair differences for each of the six non-strand-specific base substitution mutation types (Table1) per Mbp of aligned sequence was plotted for eachchromosome (Figure 2). The most frequent change is G:C to A:T on chromosomes I and X, 55 and 31 respec-tively. Chromosome I has as many G:C to A:T substitu-tions as chromosomes II, III, IV and V together. Rec-1was originally observed in strains mutagenized withEMS, a mutagen known to generate G to A changes.The gene is linked to chromosome I markers, and dueto difficulty in scoring the recombination phenotype, themutation has not been outcrossed extensively. Mostlikely many of the A:T differences are attributable tomutational changes retained in the Rec-1 strain. Thepredominance of G:C to A:T substitutions on chromo-somes I and X is also reflected in the ratio of transitionto transversions (Ts/Tv) (Table 1). Ignoring those twochromosomes, the Ts/Tv ratio is very close to randomexpectation of 0.5.Although our study is not designed to follow muta-tional accumulation over generations, we have analyzedthe distribution of base changes along the chromo-somes. Table 1 shows that the number and distributionof base pair differences on chromosome I (61%) and toa lesser extent the X-chromosome (33%) are predomi-nantly G:C to A:T. In an attempt to separate thesechanges from what may be due to de novo mutation inRec-1, we have plotted them separately along the chro-mosomes (Figure 3). Examination of the distribution ofchange along the chromosomes does not reveal anydramatic pattern, either for the distribution of G:C toA:T changes (upper red crosses) or for the distributionof the other types (lower blue crosses). In Denver et al.[3], after several generations of accumulated mutationin wild-type strains under relaxed selection no distinc-tive pattern of base substitutions was seen along thechromosomes. Neither do we see any dramatic differ-ence in distribution that might correlate with theabsence of a recombinational pattern. In an attempt toinvestigate the distribution numerically, we calculatedthe number base pair differences in the autosomalarms and in the central clusters as defined in [14] permegabasepair(Mbp). In the arms there are approxi-mately 4.16 differences per Mbp (251/60) compared to4.04 (97/24) in the cluster. A histogram of the numberbase changes per Mbp along chromosome I is shownin Additional file 1, Figure S1. When plotted this way,none of the chromosomes show any distinctive pattern(data not shown). In the absence of any detectable pat-tern of mutational distribution, it seems most likelythat Rec-1 has had no significant affect on mutationrate.We plotted the composition of base pair differencesboth for the entire genome and for the genome minuschromosome I and the X (Additional file 2, Figure S2).When chromosomes I and X are removed from the ana-lysis, the relative frequency of the different types of basepair differences is similar to that observed by Denver etal. [3], with the exception that we see considerablyfewer G:C to T:A differences.The 441 base pair differences are shown in Additionalfile 3, Table S1. Ninety-five of these are in exons (22%compared to the 27% of the genome reported to be inexons [1]). Approximately half of these (51/95) werenon-synonymous changes. Fifteen of the changes are inuntranslated regions (UTRs), 155 in introns and the restin intergenic regions.Table 1 Base Pair (bp) Differences between BC313 andVC2010 by ChromosomeBase-differenceI II III IV V X TotalG:C to A:T 55 11 14 16 14 31 141A:T to T:A 10 16 17 17 18 12 90G:C to T:A 9 7 12 14 17 18 77A:T to G:C 6 6 6 9 12 17 56A:T to C:G 5 8 5 4 7 10 39G:C to C:G 5 3 6 7 12 5 38Total 90 51 60 67 8 93 441Ts/Tv 2.1 0.5 0.5 0.6 0.48 1.07 0.81Size inMbp15.072 15.279 13.783 17.494 20.924 17.719 100.27No. alignedbp14.674 14.878 13.428 16.857 20.077 17.394 97.310Rose et al. BMC Genomics 2010, 11:131http://www.biomedcentral.com/1471-2164/11/131Page 3 of 12Rec-1 is neither caused by nor causes detectablechromosomal rearrangementsThe genome of Rec-1 was sequenced by whole genomeshotgun sequencing (WGSS) using paired end tags(PETs) and aligned to the reference genome available inWormBase. Using the alignment to the WS190 refer-ence genome, small insertions can be characterized byclusters of PETs that are shorter than the average size(Figure 4), whereas deletions in the sample can bedetected by PETs that are longer. Although this is coun-ter-intuitive to geneticists familiar with interpretinggenetic maps, it is true because the sequence reads fromthe ends (paired end tags) of the genomic fragments arefurther away in the sequenced DNA if there is an inser-tion of unannotated material than they appear on theWormBase map (reference DNA), which lacks thatinsertion. In the example shown in Figure 4, the top ofthe figure shows the size of sequence reads aligned toFigure 2 The number of each of the nonstrand- specific types of base pair differences by chromosome. The chromosomes are identifiedon the horizontal axis. The number of changes per million base pairs of aligned sequences are plotted on the vertical axis. Data from Table 1.Figure 3 Distribution of base pair differences between BC313 and VC2010 along the chromosomes. Red crosses (upper) indicate thephysical location of G:C to A:Ts in BC313 but not in VC2010. Blue crosses (lower) indicate the physical location of the remaining nonstrand-specific base differences. The chromosome number is shown on the X axis and the distance in Mbp along the Y axis.Rose et al. BMC Genomics 2010, 11:131http://www.biomedcentral.com/1471-2164/11/131Page 4 of 12the WormBase sequence. Although the inserted frag-ments are actually longer, they appear shorter on thereference genome. In an analogous way, deletionsappear longer. Translocations will have links connectingclusters on different chromosomes with a loss of readcoverage at the breakpoints. In this way, the Rec-1sequence was analyzed for chromosomal rearrange-ments, insertions, deletions, inversions and transloca-tions. No large chromosomal rearrangements wereobserved. Confirmational data was obtained using oligoarray Comparative Genome Hybridization (aCGH). Anexon-centric array design that covered the entiregenome revealed no major sequence copy numberchanges relative to the reference DNA.Direct repeat sequences can appear as longer paired endreadsExamination of fragment sizes in the Rec-1 strainrevealed a number of paired end tags (PETs) longerthan average, an indication of potential deletions. Thesequence and position of these PETs were examined indetail. An example from the right arm of chromosome Iis shown in Figure 5. In this case, the observed longsizes resulted from one of the paired ends aligning withFigure 4 Insertions of DNA can result in paired end tags (PETs) shorter than expected. PETs that fall outside the normal size range can bean indication of DNA insertions. Top: In the case of a small insertion, paired end reads can cover a region in the alignment to the reference (Ref)that does not include the inserted sequence in the sample. Lower Right: Insertions in the sample are characterized by multiple PETs with ashorter than average fragment size based on the alignment.Rose et al. BMC Genomics 2010, 11:131http://www.biomedcentral.com/1471-2164/11/131Page 5 of 12an imperfect direct repeat sequence in the sample DNA.The observed longer tag length is the consequence ofone of the paired ends aligning to different componentsof a repeat sequence. In Rec-1 as well as in other gen-omes analyzed, regions like this one that containedimperfect direct repeats gave poor coverage of sequencereads as is illustrated by the absence of normal sizePETs. The analysis illustrates how long paired-end tagscan be used to identify direct repeat sequences.Analysis of the long PETs and alignment to the refer-ence genome revealed the presence of two deletions inthe Rec-1 strain that were not in the VC2010 wild type.One of these affected an exonic region on the X-chromo-some near 11,285,000 bp and was confirmed by aCGH.The other removed approximately 100 base pairs (bp) ofintergenic sequence in a region of chromosome I around2,233,500 bp, a region of DNA not on the exon array.Additional transposable elements exist in Rec-1In a progenitor strain of BC313, Tc1 was observed toactively transpose [15], although in the original CB51strain Tc1 was apparently inactive. Blot hybridizationpatterns of the high-hopper strains have been publishedpreviously [15]. In this paper, we examined the numberand position of the transposable elements, Tc1, Tc2,Tc3, Tc4, Tc5, Tc7 and Cemar1 compared to the posi-tions reported in WormBase and reviewed by [15]. Inthe wild-type strain of C. elegans there are 30 copies ofthe transposable element Tc1 and four of Tc2 [16]. Inthe Rec-1 strain, there are 11 additional copies of Tc1,and 8 novel locations for Tc2 (Table 2; Figure 6). Anexample of how Tc1 insertion was analyzed is shown inFigure 7. Reads from within unique sequence pairedwith a read from the terminus of Tc1 identified theinsertion. All the full length Tc1 and Tc2’s had TA ter-mini. We were not able to uniquely identify the progeni-tor Tc1 since none of the Tc1’s analyzed had a uniquebase pair change in the portion of the elementsequenced. As might be expected, most of the newinsertions were in either introns or intergenic regions.One Tc1 that inserted into a coding region of a gene onthe X was detectable also by aCGH. No empty sites,that is, sites vacated by Tc1, were found by searchingunmapped reads for the DNA sequence TA or TATA.There were no new locations for Tc3, Tc4, Tc5, Tc7.These elements had positions identical to those reportedFigure 5 Imperfect direct repeats can result in paired end tags (PETs) longer than expected. PETS for bases 10,941,077-10,960,123 ofchromosome I are shown. Below the line in green are the normal sized PETs. Above the line longer, potentially aberrant, PETs are shown. Thelonger PETs are in regions lacking normal sequence coverage. Typically, the left end of the longer tags detects unique sequence and the rightend is aligned with one of a group of imperfect direct repeats producing longer than normal PETs of differing sizes.Rose et al. BMC Genomics 2010, 11:131http://www.biomedcentral.com/1471-2164/11/131Page 6 of 12in WormBase. There is an additional copy of Cemar1reported in WormBase that correlates with a duplicationof a portion of chromosome V, which is present in sev-eral wild-type strains, but not in VC2010 [17] or Rec-1.aCGH high density chip analysis agrees with the genomicsequencingIn addition to the genome-wide aCGH, a speciallydesigned high density array was used to examine thecentral portion of the gene cluster of chromosome I.The array identified five base pair differences relative tothe reference DNA, an example of one is shown in Fig-ure 8. All five of these differences were also identified inthe sequence analysis (see below) of the Rec-1 genomicDNA and confirmed by either restriction enzyme analy-sis followed by PCR or direct sequencing across the siteusing primers.DiscussionThe sequence of the Rec-1 genome was obtained bywhole genome shotgun sequencing (WGSS) with theIllumina Genome Analyzer and compared to both thereference genome available at WormBase http://www.wormbase.org and a laboratory wild type (VC2010)using MAQ [12] and Slider [13] software. The base paircomposition of Rec-1 was more similar to the wildstrain VC2010 than to the reference genome, WS190.VC2010 is a line of N2, separated from the originalBrenner strain at some time in the past. There are actu-ally two major N2 derivative lines distinguishable by thepresence or absence of a duplication of a portion of theleft arm of chromosome V [17]. Although the Rec-1strain, BC313, is not directly derived from VC2010, bothstrains lack the chromosome V duplication. Rec-1 wasoriginally detected in strains that along with a wild-typemale strain from Brenner’s original collection weretransported to the BC laboratory and maintained onplates for approximately two years before being pre-served by freezing in liquid nitrogen. The major detect-able phenotype of Rec-1 is its alteration in crossoverfrequency between markers, a phenotype that is difficultand time consuming to follow through genetic crosses.For this reason, once the strain was constructed andconfirmed, it was maintained without additional out-crossing.The number, type and location of the base pair differ-ences detected by both MAQ [12] and Slider [13] soft-ware have been analyzed for the genome. The Slidersoftware was developed to enhance the quality of align-ment possible from a low read number and improve theaccuracy in base pair change prediction [13] and indeedthe number of differences detected was considerablyhigher with the Slider software (data not shown). Weconfirmed the existence of a subset of the differencesdetected by restriction enzyme analysis of base pair dif-ferences that created new cut sites. In addition, in oneregion of 3 million bp chromosome I, five differencesthat were observed by Slider were confirmed by aCGH.aCGH has been proposed as a method for detection ofsingle nucleotide mutations in homozygous C. elegansstrains [18].A large fraction of the base pair differences in Rec-1were potentially G to A changes (141 of the 441 differ-ences were G:C to A:T) and may represent remnants ofEMS mutagenesis. Thus, we attempted to separate thesefrom other types of changes, which may be more repre-sentative of spontaneous changes having occurred in theRec-1 background. When plotted and compared to themutations accumulated under relaxed selection in tenwild-type lines observed by Denver et al [3], we see asimilar pattern of nonstrand-specific base substitutionmutation types, with the exception that there werefewer G:C to T:A changes in Rec-1 than in their MA-lines.The number and location of transposable elementswas examined. Tc3, Tc4, Tc5, Tc7 and Cemar1 wereunchanged. However, in addition to the ones reportedTable 2 Sequences flanking the sites of the new Tc1 andTc2 insertionsChr Position Type Flanking Sequences GeneI 6,440155 Tc1 TGCACATATATATTTGAATAGT snt-4 intronI 6,992279 Tc1 TAAAAAAATATATGTAAAATTT C30F12.5intronI 11,633804 Tc1 AAAATGTACATATATGTACATA IntergenicI 12,872161 Tc1 TGCTCTCAATTAGTACGTATCA IntergenicIV 11,191078 Tc1 Ambiguous insertion point IntergenicX 237866 Tc1 CTCCGTCAATTACAACACATGG AC8.10X 621137 Tc1 CATATACATATATATATATATT unc-96 intronX 827813 Tc1 CACGGAAATGTAGTTGGGTTCT IntergenicX 14,10838 Tc1 GGCTAACACATATATCCACTCA IntergenicX 8,571364 Tc1 GCCCAAGAAGTATGTCATTGGT tag-279 intronX 8,669552 Tc1 ATCATTTAGATAGATTCAAAAC rig-1 intronI 2,429485 Tc2 Ambiguous insertion point intergenicI 3,211994 Tc2 TTGTAGTTCATATTTAAAAAAG fog-1 intronI 3,215076 Tc2 AGATTTTAGCTATTTAGAATCA fog-1 intronI 6,920634 Tc2 AAAAATGATTTATCCTGATACT bbs-9 intronI 6,954054 Tc2 TGTTTACAATTAGCTTTCCGAA T10B11.8intronI 10,588754 Tc2 GAAACTGACCTATTTTTTGTCA ist-1 intronI 12,963952 Tc2 AAAATTCATTTATATAAATAAA C47B2.2intronI 13,832650 Tc2 AAAAATGGGTTAGTTTATTATT intergenicI 13,867194 Tc2 Ambiguous insertion point taf-1 intronIII 130778 Tc2 CAAATAGGTATATATAGTTGTT Nhr-280X 51825 Tc2 Ambiguous insertion point intergenicRose et al. BMC Genomics 2010, 11:131http://www.biomedcentral.com/1471-2164/11/131Page 7 of 12Figure 6 Locations of Tc1s and Tc2s specific to Rec-1. A. The red (Tc1) and blue (Tc2) crosses show the position of the elements inWormBase. Triangles show the positions of new full-length insertion sites for Tc1 (Red) and Tc2 (Blue). Three insertions of Tc2 on chromosome Iare close together, at 3.2 Mb, 6.9 Mb, and 13.8 Mb (Table 2), and each is shown as a single triangle in the Figure. B. The aCGH data for aninsertion of Tc1 into a coding region on the X-chromosome is shown.Figure 7 A Tc1 insertion with the TATA at position 6992277 on chromosome I. On the left there are four reads adjacent to the 5’ end ofthe Tc1 sequence and on the right are four reads adjacent to 3’ end. The ends of the Tc1 sequence are shown in blue. The TA is inserted atposition 6,992,279. Lower case indicates read sequences that are partial Tc1’s that are unmapped in the genome and shown as mismatches inthe alignment. Upper case indicates read sequences that are mapped to the genome. All the reads shown are paired with another read thatmaps to Tc1 internal sequence.Rose et al. BMC Genomics 2010, 11:131http://www.biomedcentral.com/1471-2164/11/131Page 8 of 12in the reference genome, 11 Tc1 and 8 Tc2 wereobserved. These are most likely the remnants of trans-position events that occurred in a progenitor strain [15].The results emphasize the advantage over previous tech-nologies of having the genomic sequence. We identifiednot only the location of the new Tc1 insertions but alsothe number and location of a large number of Tc2insertions. Tc1 is 1,610 bp long and contains two 54-bpterminal inverted repeats and transposes by excision andreinsertion into target DNA containing a TA dinucleo-tide, leaving behind a double-strand DNA break whichis repaired by the cellular machinery. Tc2 is a 2,074base pair element that has perfect terminal invertedrepeats of 24 bp and like Tc1, insertions are flanked bya TA dinucleotide at either end. It may not be obviouswhy Tc1 and Tc2, which are members of differenttransposable element superfamilies [16], would havebeen the two elements to have transposed in the highhopper strain [15], although both elements have beenobserved to transpose in Bristol Bergerac hybrids andproposed to move together possibly by a mechanisminvolving mut-4 [19]. The original strain in which mobi-lity was first observed [15] is maintained as a frozenarchive and available for further characterization withregard to aspects of Tc1 and Tc2 mobilization.The Rec-1 strain is unique in that it alters the patternof meiotic exchange events without affecting the totalnumber of crossovers [9] and has little other phenotypiceffects [11]. Rattray and Rose [11] investigated fitness ofRec-1 relative to wild type in a short-term experimentperformed in a laboratory setting. No difference wasobserved under those conditions. In addition, mutationaldamage as measured by capture of lethal events using agenetic balancer did not differ from wild type [11]. Inthe present study, no major chromosomal rearrange-ments, which might reduce the fitness of the strain,were observed. There were however, a large number ofbase pair differences from wild type, and many of thesewere in coding regions. These changes are presumablynon-detrimental based on their benign effect on thephenotype of the strain.The genetic maps of many sexually reproducing spe-cies reveal that relative to physical distance recombina-tion occurs more frequently in some regions than inothers. Furthermore, the position of the crossover eventcan be influenced by a number of factors, includingtreatment with ionizing radiation. In Drosophila, ioniz-ing radiation increases crossing over [20], primarily inregions of centric heterochromatin [21], a region knownto have low recombination relative to the amount ofFigure 8 High density aCGH. An example of aCGH hybridization data identifying a C to T change in Rec-1. The Y-axis is the normalized log2ratio of fluorescent intensities (Rec-1 versus reference). Each bar represents one 50 mer oligo probe on the oligo array chip. Below the plot is aschematic of the exon structure of predicted gene R05D11.9 and the position of the C-T base pair change identified at position 8,595,578 bp onchromosome I in the Rec-1 strain.Rose et al. BMC Genomics 2010, 11:131http://www.biomedcentral.com/1471-2164/11/131Page 9 of 12DNA [22]. Similarly in C. elegans gamma radiation hasbeen shown to increase crossing over across the recom-bination-poor central region of autosome I [23]. Inyeast, DNA damaging agents have been shown to stimu-late homologous recombination between ectopic repeatsresulting in translocations [24]. In addition, induceddouble-strand breaks within dispersed small repeats cangenerate rearrangements resulting in genome reshapingand are a potential source for evolutionary change [25].C. elegans, which is a self-fertilizing hermaphrodite without-crossing, a relatively rare situation in nature, pro-vides a useful model for the study of genomic character-istics as they relate to recombination and short termevolution in a self-fertilizing hermaphrodite [5,6].In both the Denver et al. [3] analysis of ten mutation-accumulating wild-type strains and in our study of Rec-1, there is no dramatic pattern of mutational alterationsalong the chromosome. These results may be interestingin the context of whether or not recombination is muta-genic. Both the wild-type MA-lines, which presumablyhave regions of both high and low recombination, andRec-1, in which the variation in recombination is rando-mized, have very similar patterns of base pair differ-ences. The data are compatible with a model in whichmutational events are independent of the meioticrecombination processes.ConclusionOur analysis of high-throughput sequencing was able todetect regions of direct repeat sequences, deletions, inser-tions of transposable elements, and base pair differences.A subset of sequence alterations affecting coding regionswere confirmed by an independent approach using oligoarray comparative genome hybridization. The major phe-notype of the Rec-1 strain is an alteration in the preferredposition of the meiotic recombination event with noother significant phenotypic consequences. In this study,we observed no evidence of a mutator effect at thenucleotide level attributable to the Rec-1 mutation.MethodsGenetic strainsTwo strains of Caenorhabditis elegans were used in thisanalysis. The BC313 strain carrying the s180 allele ofthe gene rec-1 was constructed in 1977 and maintainedfrozen in liquid nitrogen. BC313 was derived from CB51[unc-13(e51)], CB73 [unc-15(e73)] and CB61 [dpy-5(e61)], all of which were generated using 0.05 M ethyl-methane sulfonate (EMS) in the CB laboratory of S.Brenner, Cambridge University UK [8] and transportedto the BC laboratory of D. Baillie, Simon Fraser Univer-sity CA where they were maintained on agar cultureplates streaked with E. coli OP-50 at 15°C for approxi-mately two years before being frozen. These rec-1progenitor strains and the subsequent BC313 strain areestimated to have been maintained on plates forapproximately 100 generations prior to sequencing.The VC2010 strain is a wild type N2 strain subcul-tured in the Knock-out Consortium laboratory of D.Moerman, University of British Columbia CA from thewild-type N2 strain, VC196. It was received from theCaenorhabditis Genetics Centre in October 2008.VC2010 carries the wild-type allele of rec-1 and wassequenced by the Genome Sciences Centre, VancouverCA prior to this analysis.Brenner’s wild-type N2 strain gave rise to bothVC2010 and the N2 strain used in the BC313 construc-tion, but BC313 was not derived directly from VC2010.aCGHThe two C. elegans arrays used for oligo-array compara-tive genome hybridization (aCGH) were designed by S.Flibotte at the Genome Sciences Centre, Vancouver CA.The whole genome array consisted of overlapping 50-mer probes targeting primarily annotated exons andmicro-RNAs. Both it and the high density array wereproduced by NimbleGen Systems Inc. http://www.nim-blegen.com. Sample preparation, hybridization and ana-lysis was done as previously described [26].Copy number aberrations were detected by visualinspection using the SignalMap™ browser software[NimbleGen Systems Inc. http://www.nimblegen.com.DNA preparation and High-Throughput SequencingDNA preparation for whole genome shotgun sequencing(WGSS) was done by shearing approximately 10 ugDNA for 10 min using Sonic Dismembrator 550 (cuphorn, Fisher Scientific, Canada) with a power setting of“7” in pulses of 30 seconds interspersed with 30 secondsof cooling, and analyzed on a 8% PAGE gel. A 180-220bp DNA fraction was excised and eluted from the gelslice overnight at 4°C in 300 μl of elution buffer (5:1,LoTE buffer (3 mM Tris-HCl, pH 7.5, 0.2 mM EDTA)-7.5 M ammonium acetate), and was purified using aSpin-X Filter Tube (Fisher Scientific), and by ethanolprecipitation. The WGSS library was prepared using amodified paired-end protocol supplied by Illumina Inc.(USA). This involved DNA end-repair, formation of 3’ Aoverhangs using klenow fragment (3’ to 5’ exo minus)and ligation to Illumina PE adapters. Adapter-ligatedproducts were purified on Qiaquick spin columns (Qia-gen) and PCR-amplified using Phusion DNA polymerasefor 10 cycles using the PE primer 1.0 and 2.0 (Illumina).PCR products of the desired size range were purifiedusing a 8% PAGE gel. DNA quality was assessed andquantified using an Agilent DNA 1000 series II assayand Nanodrop 7500 spectrophotometer (Nanodrop,USA), and DNA was subsequently diluted to 10 nM.Rose et al. BMC Genomics 2010, 11:131http://www.biomedcentral.com/1471-2164/11/131Page 10 of 12The final concentration was confirmed using a Quant-iTdsDNA HS assay kit and Qubit fluorometer (Invitrogen).For sequencing, clusters were generated on the Illuminacluster station and paired end reads were generatedusing an Illumina GAII platform following the manufac-turer’s instructions. Image analysis, basecalling and errorcalibration was performed using the V1.0 Illumina Gen-ome Analyzer analysis pipeline. The BC313 genomicsequence was aligned to the annotated sequence of C.elegans available at WormBase WS190 http://www.wormbase.org and compared with the sequence of thewild-type strain VC2010.Identification of Transposable Element InsertionsThe alignment of pair-end reads was done by finding allPETs with at least one read matching 300 bp of the 5’ endor 3’ end of the canonical transposon sequence. Thematching reads were clustered and those that had morethan five reads were analyzed further. Each transposon wascharacterized by two clusters, one containing reads alignedto the forward strand and one containing reads aligned toreverse strand. If two clusters identified a location for atransposon that is present in the reference genome, theclusters were separated by a distance approximating thelength of the transposon plus 200 bp. All novel locationshas two clusters separated by about 200-300 bp. Examina-tion of the reads flanking novel transposon locationsallowed us to identify the point of insertion.The SRA accession# is SRA009755.Additional file 1: Figure S1: Histogram of the number basedifferences per Mbp along chromosome I. Data from Additional file 3,Table S1 was used to plot the number of base changes alongchromosome I, revealing no obvious difference for different regions ofthe chromosome.Click here for file[ http://www.biomedcentral.com/content/supplementary/1471-2164-11-131-S1.DOC ]Additional file 2: Figure S2: The total number of each of thenonstrand- specific types of base pair differences. Blue bars indicatethe total base pair differences between Rec-1 and VC2010 for thegenome. Red bars indicate the differences for chromosomes II, III, IV andV summed together.Click here for file[ http://www.biomedcentral.com/content/supplementary/1471-2164-11-131-S2.PDF ]Additional file 3: Table S1: Base Differences between BC313 andVC2010. All the base differences identified by both MAQ and Slider arelisted and annotated. Each difference is identified by a unique ‘h’ alleledesignation.Click here for file[ http://www.biomedcentral.com/content/supplementary/1471-2164-11-131-S3.XLSX ]AcknowledgementsWe thank Donald Moerman for providing us with the strain and sequenceinformation for wild-type VC2010; Sanja Tarailo and Shir Hazhir for technicalassistance. This work was supported in part by a grant from the NaturalSciences and Engineering Research Council (Canada) to AMR and DLB.Author details1Department of Medical Genetics, University of British Columbia, 419 - 2125East Mall, Vancouver, BC, V6T 1Z4, Canada. 2Genome Sciences Centre, BritishColumbia Cancer Research Centre, 600 West 10th Avenue, Vancouver, BC,V5Z 4E6, Canada. 3Molecular Biology and Biochemistry, Simon FraserUniversity, Burnaby, BC, V5A 1S6, Canada.Authors’ contributionsAMR, SJMJ, DLB, MM, SF, NON conceived and designed experiments. NON,MB, NM, YB and SF performed experiments. AMR, YB, MB, NON, DLB, MRJand SJMJ analyzed data. AMR, NON, MRJ, DLB YB and SJMJ wrote themanuscript. All authors read and approved the final manuscript.Received: 9 September 2009Accepted: 23 February 2010 Published: 23 February 2010References1. C. elegans Sequencing Consortium: Genome sequence of the nematodeC. elegans: a platform for investigating biology. Science 1998,282(5396):2012-2018.2. Hillier LW, Marth GT, Quinlan AR, Dooling D, Fewell G, Barnett D, Fox P,Glasscock JI, Hickenbotham M, Huang W, Magrini VJ, Richt RJ, Sander SN,Stewart DA, Stromberg M, Tsung EF, Wylie T, Schedl T, Wilson RK,Mardis ER: Whole-genome sequencing and variant discovery in C.elegans. Nat Methods 2008, 5(2):183-188.3. Denver DR, Dolan PC, Wilhelm LJ, Sung W, Lucas-Lledo JI, Howe DK,Lewis SC, Okamoto K, Thomas WK, Lynch M, Baer CF: A genome-wide viewof Caenorhabditis elegans base-substitution mutation processes. ProcNatl Acad Sci USA 2009, 106(38):16310-116314.4. Sarin S, Prabhu S, O’Meara MM, Pe’er I, Hobert O: Caenorhabditis elegansmutant allele identification by whole-genome sequencing. Nat Methods2008, 5(10):865-867.5. Cutter AD, Payseur BA: Selection at linked sites in the partial selferCaenorhabditis elegans. Mol Biol Evol 2003, 20(5):665-673.6. Cutter AD, Dey A, Murray RL: Evolution of the Caenorhabditis elegansgenome. Mol Biol Evol 2009, 26(6):1199-1234.7. Johnsen RC, Jones SJM, Rose AM: Mutational accessibility of essentialgenes on chromosome I(left) in Caenorhabditis elegans. Mol Gen Genet2000, 263:239-252.8. Brenner S: The genetics of Caenorhabditis elegans. Genetics 1974,77(1):71-94.9. Zetka MC, Rose AM: Mutant rec-1 eliminates the meiotic pattern ofcrossing over in Caenorhabditis elegans. Genetics 1995, 141(4):1339-1349.10. Rose AM, Baillie DL: A mutation in Caenorhabditis elegans that increasesrecombination frequency more than threefold. Nature 1979,281(5732):599-600.11. Rattray B, Rose AM: Increased intragenic recombination and non-disjunction in the Rec-1 strain of Caenorhabditis elegans. GeneticalResearch 1988, 51:89-93.12. Li H, Ruan J, Durbin R: Mapping short DNA sequencing reads and callingvariants using mapping quality scores. Genome Res 2008,18(11):1851-1858.13. Malhis N, Butterfield YS, Ester M, Jones SJ: Slider–maximum use ofprobability information for alignment of short sequence reads and SNPdetection. Bioinformatics 2009, 25(1):6-13.14. The Nematode Caenorhabditis elegans [Edited by W.B. Wood and theCommunity of C. elegans Researchers], Appendix 4. Cold Spring HarborLaboratory Press 1988.15. Babity JM, Starr TV, Rose AM: Tc1 transposition and mutator activity in aBristol strain of Caenorhabditis elegans. Mol Gen Genet 1990, 222(1):65-70.16. Bessereau JL: Transposons in C. elegans. WormBook The C. elegansResearch Community, WormBook 2006, 1-13, (January 18, 2006).17. Vergara IA, Mah AK, Huang JC, Tarailo-Graovac M, Johnsen RC, Baillie DL,Chen N: Polymorphic segmental duplication in the nematodeCaenorhabditis elegans. BMC Genomics 2009, 10:329.18. Maydan JS, Okada HM, Flibotte S, Edgley ML, Moerman DG: De Novoidentification of single nucleotide mutations in Caenorhabditis elegansRose et al. BMC Genomics 2010, 11:131http://www.biomedcentral.com/1471-2164/11/131Page 11 of 12using array comparative genomic hybridization. Genetics 2009,181(4):1673-1677.19. Levitt A, Emmons SW: The Tc2 transposon in Caenorhabditis elegans.Proc Natl Acad Sci USA 1989, 86:3232-3236.20. Muller HJ: The Regionally Differential Effect of X Rays on Crossing over inAutosomes of Drosophila. Genetics 1925, 10(5):470-507.21. Robbins LG: Genetically induced mitotic exchange in theheterochromatin of Drosophila melanogaster. Genetics 1981, 99(3-4):443-459.22. Grell RF: High frequency recombination in centromeric and histoneregions of Drosophila genomes. Nature 1978, 272(5648):78-80.23. Kim JS, Rose AM: The effect of gamma radiation on recombinationfrequency in Caenorhabditis elegans. Genome 1987, 29(3):457-462.24. Fasullo M, Dave P, Rothstein R: DNA-damaging agents stimulate theformation of directed reciprocal translocations in Saccharomycescerevisiae. Mutat Res 1994, 314(2):121-133.25. Argueso JL, Westmoreland J, Mieczkowski PA, Gawel M, Petes TD: Double-strand breaks associated with repetitive DNA can reshape the genome.Proc Natl Acad Sci USA 2008, 105(33):11845-11850.26. Maydan JS, Flibotte S, Edgley ML, Lau J, Selzer RR, Richmond TA, Pofahl NJ,Thomas JH, Moerman DG: Efficient high-resolution deletion discovery inCaenorhabditis elegans by array comparative genomic hybridization.Genome Res 2007, 17(3):337-347.doi:10.1186/1471-2164-11-131Cite this article as: Rose et al.: Genomic sequence of a mutant strain ofCaenorhabditis elegans with an altered recombination pattern. BMCGenomics 2010 11:131.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitRose et al. BMC Genomics 2010, 11:131http://www.biomedcentral.com/1471-2164/11/131Page 12 of 12


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items