UBC Faculty Research and Publications

TILLING is an effective reverse genetics technique for Caenorhabditis elegans Gilchrist, Erin J; O'Neil, Nigel J; Rose, Ann M; Zetka, Monique C; Haughn, George W Oct 18, 2006

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

Item Metadata


52383-12864_2006_Article_645.pdf [ 853.68kB ]
JSON: 52383-1.0223183.json
JSON-LD: 52383-1.0223183-ld.json
RDF/XML (Pretty): 52383-1.0223183-rdf.xml
RDF/JSON: 52383-1.0223183-rdf.json
Turtle: 52383-1.0223183-turtle.txt
N-Triples: 52383-1.0223183-rdf-ntriples.txt
Original Record: 52383-1.0223183-source.json
Full Text

Full Text

ralssBioMed CentBMC GenomicsOpen AcceMethodology articleTILLING is an effective reverse genetics technique for Caenorhabditis elegansErin J Gilchrist*1, Nigel J O'Neil2, Ann M Rose2, Monique C Zetka3 and George W Haughn1Address: 1Department of Botany, 6270 University Blvd, University of British Columbia, Vancouver, BC, V6T 1Z4, Canada, 2Department of Medical Genetics, University of British Columbia, Vancouver, BC, V6T 1Z3, Canada and 3Department of Biology, McGill University, Stewart Building N5/16, 1205 Avenue Docteur Penfield, Montreal, QC, H3A 1B1, CanadaEmail: Erin J Gilchrist* - ering@interchange.ubc.ca; Nigel J O'Neil - noneil@gene.nce.ubc.ca; Ann M Rose - arose@gene.nce.ubc.ca; Monique C Zetka - monique.zetka@mcgill.ca; George W Haughn - haughn@interchange.ubc.ca* Corresponding author    AbstractBackground: TILLING (Targeting Induced Local Lesions in Genomes) is a reverse genetictechnique based on the use of a mismatch-specific enzyme that identifies mutations in a target genethrough heteroduplex analysis. We tested this technique in Caenorhabditis elegans, a modelorganism in which genomics tools have been well developed, but limitations in reverse geneticshave restricted the number of heritable mutations that have been identified.Results: To determine whether TILLING represents an effective reverse genetic strategy for C.elegans we generated an EMS-mutagenised population of approximately 1500 individuals andscreened for mutations in 10 genes. A total of 71 mutations were identified by TILLING, providingmultiple mutant alleles for every gene tested. Some of the mutations identified are predicted to besilent, either because they are in non-coding DNA or because they affect the third bp of a codonwhich does not change the amino acid encoded by that codon. However, 59% of the mutationsidentified are missense alleles resulting in a change in one of the amino acids in the protein productof the gene, and 3% are putative null alleles which are predicted to eliminate gene function. Wecompared the types of mutation identified by TILLING with those previously reported fromforward EMS screens and found that 96% of TILLING mutations were G/C-to-A/T transitions, arate significantly higher than that found in forward genetic screens where transversions anddeletions were also observed. The mutation rate we achieved was 1/293 kb, which is comparableto the mutation rate observed for TILLING in other organisms.Conclusion: We conclude that TILLING is an effective and cost-efficient reverse genetics tool inC. elegans. It complements other reverse genetic techniques in this organism, can provide an allelicseries of mutations for any locus and does not appear to have any bias in terms of gene size orlocation. For eight of the 10 target genes screened, TILLING has provided the first geneticallyheritable mutations which can be used to study their functions in vivo.Published: 18 October 2006BMC Genomics 2006, 7:262 doi:10.1186/1471-2164-7-262Received: 12 May 2006Accepted: 18 October 2006This article is available from: http://www.biomedcentral.com/1471-2164/7/262© 2006 Gilchrist et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Page 1 of 16(page number not for citation purposes)BMC Genomics 2006, 7:262 http://www.biomedcentral.com/1471-2164/7/262BackgroundCaenorhabditis elegans is a well-established model system(reviewed by Hodgkin [1]) that is increasingly being usedfor genetic and molecular investigations into conservedbiological processes, including those involved in humandisease [2-5]. Although simple in structure, C. elegans iscomparable to higher animals in development and formsmost of the major tissue types that are important to verte-brate physiology. Indeed, in a comparison of 18,452 C.elegans protein sequences against human EST databases,83% (15,344 sequences) of the C. elegans sequences werefound to have human homologues [6].Because the sequence of the complete C. elegans genomehas been available since 1998, bioinformaticians havebeen presented with ample opportunity to mine the data,and a plethora of genomic and proteomic information isaccessible to researchers wishing to build upon this infor-mation [7]. Powerful in silico techniques have also beendeveloped for the analysis of genome sequence informa-tion and are used in the prediction of gene function,expression and interaction [5,8,9]. Despite the excitingpossibilities flowing from these studies, the testing of pre-dictions made in silico relies largely on the existence of effi-cient reverse genetic approaches that target specific genesor classes of genes in vivo. In vitro techniques such as yeasttwo-hybrid analysis [10] and microarray analysis [11]have also been used to generate an abundance of valuabledata about gene expression and protein interactions but,like the data generated in silico, these data need to be ver-ified in vivo.C. elegans has approximately 19,800 protein-coding genesand 12,000 of these have been conserved over the 100million years since this species has diverged from therelated nematode Caenorhabditis briggsae, indicating thatthey are likely important functional genes [12]. In spite ofthis fact, however, only about 3,400 genes in C. eleganshave mutant alleles available for genetic and biochemicalanalysis to ascertain their function and importance [13].High-throughput reverse genetics is an ideal way of gener-ating mutations in the remaining 16,400 genes and sev-eral such approaches have been developed for thenematode each of which has advantages and drawbacksthat affect the applicability or efficiency of the techniqueas a tool for probing gene function on a genomic scale.Currently, the most efficient and popular method to dis-rupt the activity of a gene in C. elegans is the technique ofRNA interference (RNAi) [14]. Large-scale RNAi screenshave demonstrated that the function of a diverse popula-tion of genes with roles in many biological processes canbe disrupted by the injection of double-stranded RNAtodes bacteria expressing dsRNA [17,18]. These samestudies, however, have also documented that the pheno-types resulting from the RNAi treatment often depend onthe method of delivery. In addition, the RNAi techniquecannot replace classical genetic analysis because the phe-notypic effects are transient and not heritable, makingclassical genetic interaction studies impossible.Another effective reverse-genetic technique that is beingused successfully in C. elegans is mutagenesis with tri-methylpsoralen and ultraviolet radiation (TMP/UV) fol-lowed by detection of gene knockouts by PCR. This iscurrently the method of choice for obtaining heritableloss-of-function mutations in C. elegans but there are alsodrawbacks to this approach. First, the limitations of thedetection method necessitate using a high dosage ofmutagen which requires multiple rounds of outcrossingto remove accompanying background mutations. In addi-tion, missense alleles cannot be isolated and large dele-tion events may result in the loss of function from morethan one locus simultaneously. Finally, although the rea-son for this is unclear, mutations in certain genes havebeen more difficult to obtain than in others.Transposon-insertion mutagenesis is another tool that isavailable to the C. elegans community [19,20] but it sharesmany of the limitations discussed for the previous tech-niques in addition to some that are specific to thisapproach such as the fact that small genes are less likely tobe targets of transposon insertion and certain regions ofthe genome may vary in the frequency at which trans-posons insert. The mutagenic effect of Tc1 insertions canalso sometimes be circumvented by innate compensationmechanisms that allow spicing around the transposon.A recently reported study of biolistic transformation in C.elegans indicates that homologous recombination ofintroduced DNA is also possible in this species [21] but,in spite of the potential of this technique to provide thelong-sought ability to perform site-directed mutagenesisin C. elegans, the low success rate and the fact that an elab-orate microparticle bombardment set-up is required,make it unlikely that this procedure will soon become effi-cient enough for high-throughput reverse genetics.As a result of drawbacks in currently used reverse genetictechniques, the pace of research into biological processesin C. elegans is still largely dictated by the probability ofobtaining a mutant of any given gene and, thus, new tech-niques are needed to complement those previouslydescribed. TILLING (Targeting Induced Local Lesions inGenomes) is a relatively novel reverse genetics techniquebased on the use of a mismatch-specific enzyme that willPage 2 of 16(page number not for citation purposes)(dsRNA) directly into the gonad [15], by soaking the nem-atodes in a dsRNA solution [16], or by feeding the nema-identify mutations in any target gene through heterodu-plex analysis [22]. The technique involves PCR amplifica-BMC Genomics 2006, 7:262 http://www.biomedcentral.com/1471-2164/7/262tion of a target gene or region of DNA using fluorescentlylabelled primers, followed by digestion with an enzymethat specifically cleaves at the site of a mismatch such asthat induced by ethylmethanesulfonate (EMS) mutagene-sis (see Figure 1). The sizes of the cleavage fragmentsresolved on polyacrylamide gels reveal the approximateposition of the mutation within the amplicon. We reporthere on a pilot project to test the use of this technology inC. elegans: we have constructed and arrayed a mutagenisedpopulation and used it to isolate mutations in 10 differentgenes. On the basis of these data we conclude that TILL-ING is as effective and cost-efficient in C. elegans as it hasbeen shown to be in other species in which it has beentested [23-29].Results and DiscussionEfficiency of TILLING in C. elegansTo determine whether TILLING represents an effectivereverse genetic strategy for C. elegans we generated andarrayed an EMS-mutagenised population of approxi-mately 1500 individual animals (see below) and screenedfor mutations in 10 genes varying in size from 788 basepairs (bp) to 9112 bp. A region of approximately 1500 bpfrom each gene was examined and a total of 71 novelmutations were identified by TILLING, thus providingmultiple mutant alleles for each gene (Table 1). Some ofthe mutations we identified are predicted to be silent,either because they are in non-coding DNA or becausethey affect the third bp of a codon which does not changethe amino acid encoded by that codon. However, 59% ofthe mutations we identified are missense alleles resultingin a change in one of the amino acids in the protein prod-uct of the gene, and 3% are nonsense alleles resultingfrom the insertion of a premature stop codon into thecoding region of the gene, or in the elimination of a con-served splice junction site. These data demonstrate theefficacy of TILLING in C. elegans.Comparison of forward and reverse genetics with EMS mutagenesisA C. elegans population was constructed for this TILLINGproject using the mutagen EMS (Figure 2). EMS was cho-sen because it has been shown to be an effective mutagenin this species and because it is known to generate prima-rily single bp mutations which can be identified usingTILLING [23,30-34].A survey of Wormbase [7] was performed in order toexamine the type of molecular lesion induced by EMS inC. elegans to ensure that the majority of these mutationsare, indeed, small intergenic lesions of the type that can beidentified by TILLING (see Additional File 1 for a list ofalleles). Two hundred and thirteen alleles from 51 differ-ent genes whose molecular sequence was known wereselected randomly from the database. All of the mutationswere reported to be identified in screens of EMS-treatedanimals. Ninety three percent of the 213 alleles examinedfrom Wormbase were found to be single bp mutations.Eighty seven percent were G/C-to-A/T transitions, six per-cent were other single bp mutations, and seven percent ofthe mutations reported were deletions that ranged in sizefrom 88 bp to 2.3 kb.In our TILLING experiment, 68 of the 71 independentmutations identified (96%) were G/C-to-A/T transitionsand the remaining three mutations were A/T-to-T/A or G/Table 1: List of TILLING targets, sizes of amplicons and number and type of mutations identified for each gene.Gene Name Description GeneSize (bp)PCRSize (bp)1Prev.alleles2Prev.strains3Mis-sense3Null 3Silent 3Total*C05C10.5 Hypothetical protein 788 1175 0 0 2 1 3mel-32 C05D11.11 Serine hydroxyl-methyl-transferase 1600 1500 16 1 4 1 2 7mus-81 C43E11.2 Endonuclease MUS81 2530 1171 1 0 4 3 7xpf-1 C47D12.8 Structure-specific endonuclease ERCC1-XPF 9112 1452 0 0 5 3 8*F25H2.13 Helicase of the DEAD superfamily 4985 1499 1 0 5 4 9htp-3 F57C9.5 HIM-3 paralogue 2598 1452 1 1 5 5 10*M03C11.2 Helicase of the DEAD superfamily 5943 1490 1 0 4 1 5cki-2 T05A6.2 Hypothetical protein 1555 1569 0 0 7 1 2 10mdf-2 Y69A2AR.30 Spindle assembly checkpoint protein 4461 1466 1 0 1 5 6htp-2 Y73B6BL.2 HIM-3 paralogue protein 2 1199 1451 0 0 5 1 6Totals 14225 42 2 27 71* Gene name not assigned1 Number of mutant alleles listed in Wormbase [7] as existing prior to this study.2 Number of mutant strains available from the Caenorhabditis Genetic Stock Center.3 Number of mutations of this type identified in this TILLING study.Missense mutations alter the amino acid sequence of the encoded protein. Null mutations refer to mutations that convert an amino acid codon into Page 3 of 16(page number not for citation purposes)a premature stop codon, or that alter a conserved splice junction and result in premature truncation of the protein product of the gene. Silent mutations are changes that do not affect the protein product of the gene. These include mutations in introns or intergenic sequences, and mutations that alter the third bp of a codon in such a way that it does not change the amino acid encoded by that codon.BMC Genomics 2006, 7:262 http://www.biomedcentral.com/1471-2164/7/262Page 4 of 16(page number not for citation purposes)Overview of the TILLING procedureFigure 1Overview of the TILLING procedure. Pooled DNA is amplified using fluorescently tagged, gene-specific primers. The forward and reverse primers are labelled with different fluorophors that label both ends of the fragment. The amplified products are denatured by heating and then allowed to cool slowly so that they randomly re-anneal. Heteroduplex molecules form when mutant and wild-type PCR products anneal together, and these then become targets for a single-strand-specific nuclease found in Celery Juice Extract (CJE). The nuclease cleaves these heteroduplex fragments at one of the two strands, 3' to the site of the mismatch in the DNA. The PCR products that retain one of the labelled primers can then be detected on polyacrylamide dena-turing LI-COR gels. Individuals with a mutation in the gene of interest are identified by the smaller cleavage fragment seen on the gel as well as the wild-type product. Because the nuclease cleaves either of the two strands randomly, cleavage products can be detected in both the IRD700 and IRD800 channels of the gel image. The position of the mutation within the PCR ampli-con can be calculated from the size of the two fragments carrying the forward, IRD700-labeled primer, and the reverse, IRD800-labeled primer. Grey bands on the gel are thought to result from partial PCR products and aid in sizing of mutant bands.denature, re-annealdigest with CJEdenatureresolve on LI-COR gelPCR with labelled primersWildtype Mutantfull-length productcleaved product labeled with forward primer700 Channel Image 800 Channel Image1.5 kb0.9 kb0.6 kbfull-length productcleaved product labeled with reverse primerca3’5’ 5’3’cg3’5’5’3’ gt3’5’5’3’cg3’5’5’3’ at3’5’5’3’cg3’5’5’3’at3’5’ 5’3’cg3’5’5’3’ca3’5’5’3’at3’5’5’3’ ca3’5’5’3’g3’5’5’3’tg3’5’5’3’tg3’5’5’3’tt5’c5’ 3’c5’ 3’t5’ 3’3’a3’ 5’g3’ 5’a5’g5’BMC Genomics 2006, 7:262 http://www.biomedcentral.com/1471-2164/7/262C-to-T/A transversions (Table 2). This percentage of G/C-to-A/T transitions is significantly higher than that foundin forward genetic screens and probably reflects the factthat many EMS-induced lesions are not identified usingclassical genetic screens because they do not have an effecton phenotype. The frequency of transversions seen withTILLING in our screen is similar to the frequency found inforward genetic screens using EMS in C. elegans (6%), butsignificantly lower than the number of transversionsmutations reported in a small Drosophila TILLING project(16%) [29], and significantly higher than the numberfound in Arabidopsis where greater than 99% of mutationssequenced were G/C-to-A/T transitions [23].Three of the mutant strains we generated in this study,CN579, CN1162 and CN1574 carried two point muta-tions within the target gene. In each case, one of the sec-ond-site mutations was in non-coding DNA and sopresumably would not affect the phenotype of the ani-mals carrying the linked mutation. Second-site mutationshave been reported in forward genetic screens of EMS-treated worms as well. A case in point is the molecularanalysis of unc-52 mutations and intragenic suppressoralleles [35]. Two of the 19 mutations sequenced in theunc-52 study carried a second site mutation less than 400bp upstream of the primary mutation. One of these sec-ond-site mutations was a single bp transition, and theother was a 311 bp deletion. In both cases, the second-sitemutations were not found to affect the phenotype of the[34], and so presumably reflect some common DNArepair mechanism that is induced upon EMS damage.Deletions are another type of mutation that has beenreported in forward genetic screens using EMS in severaldifferent species [30,32,34]. In C. elegans approximately7% of all mutations from forward genetic screens of EMS-treated animals are deletions (Additional File 1). We didnot identify any deletions in this TILLING project but weare confident (based on previous studies [23,36]) that ifEMS does generate this type of mutation in C. elegans,TILLING will be able to detect these events as well as thesingle base pair changes that are more common. The rea-son that deletions have been identified more frequently inforward genetic screens than in our reverse genetic screenis probably because these mutations are much more likelyto produce a phenotype than the single bp mutationsmore commonly produced by EMS.Mutagen dose and mutation rateThe dose of EMS used for our TILLING project (0.025M)was lower than that used in many forward genetic screensbecause studies have shown that this lower dose simpli-fies the identification of mutant phenotypes caused by thegene of interest while limiting confounding backgroundphenotypes or lethality [37]. In two strains, however(CN843 and CN1643), we identified mutations in twodifferent genes in the same strain. While this might seemto indicate a very high overall mutation frequency, we doOutline of C. elegans TILLING procedureFigur 2Outline of C. elegans TILLING procedure. Animals are mutagenised with EMS, picked individually to plates, and allowed to self. One third of the worms are used for DNA and the remaining two thirds are frozen for future analysis. DNA is pooled 8-fold to reduce time and expense. TILLING is performed in order to determine which individuals carry mutations in the gene of interest. Mutations are sequenced and individuals from lines carrying mutations that have an effect on the gene product are thawed and genotyped to isolate heterozygous or homozygous mutants.Freeze for libraryIsolate DNA TILLING Genotype to find heterozygotes or homozygotesMutagenizewith EMSP0selfF18-fold poolsF2 +row poolcolumn poolOutcross, analysephenotypes, and balance mutationsSequence mutationsthawPage 5 of 16(page number not for citation purposes)animals carrying them. Second-site mutations have alsobeen found in EMS screens of yeast [33] and Drosophilanot believe that this is the case since the mutagenised ani-mals seem healthy and fertile, and since the overall muta-BMC Genomics 2006, 7:262 http://www.biomedcentral.com/1471-2164/7/262Table 2: Mutations identified by TILLING.Gene Strain Allele Change EffectC05C10.5 CN556 vc21 C112T P23SC05C10.5 CN1688 vc40 G178A G45RC05C10.5 CN1746 vc41 G230A Non-codingmel-32 C05D11.11 CN843 vc11 G361A V106Imel-32 C05D11.11 CN1181 vc68 C1339T Q416*mel-32 C05D11.11 CN1621 vc69 G289A G82Rmel-32 C05D11.11 CN1665 vc70 G373A D110Nmel-32 C05D11.11 CN1738 vc71 G384A K113=mel-32 C05D11.11 CN1805 vc72 G972A K293=mel-32 C05D11.11 CN1856 vc73 G373A D110Nmus-81 C43E11.2 CN1162 vc42 C1830T L368Fmus-81 C43E11.2 CN1162 vc43 G1231A Non-codingmus-81 C43E11.2 CN1211 vc44 A955T Non-codingmus-81 C43E11.2 CN1456 vc45 G972A Non-codingmus-81 C43E11.2 CN1604 vc46 G1897A G390Emus-81 C43E11.2 CN1766 vc47 C1687T T320Imus-81 C43E11.2 CN568 vc48 G1313A D214Nxpf-1 C47D12.8 CN665 vc18 C602T L183Fxpf-1 C47D12.8 CN720 vc19 G930A R292Hxpf-1 C47D12.8 CN1286 vc62 G1036A R278Qxpf-1 C47D12.8 CN1475 vc63 G446A E100Kxpf-1 C47D12.8 CN1574 vc64 C101T Non-codingxpf-1 C47D12.8 CN1574 vc65 G818T S205=xpf-1 C47D12.8 CN1751 vc66 C902T V233=xpf-1 C47D12.8 CN1798 vc67 G942A D247NF25H2.13 CN838 vc10 A1200T I277FF25H2.13 CN1245 vc52 G1206A E279KF25H2.13 CN1326 vc53 G421A Non-codingF25H2.13 CN1742 vc54 G285A E95=F25H2.13 CN1812 vc55 G399A Non-codingF25H2.13 CN1838 vc56 G1167A A266TF25H2.13 CN579 vc7 C649T Non-codingF25H2.13 CN579 vc8 C1165T S265FF25H2.13 CN48 vc9 G1242A E291Khtp-3 F57C9.5 CN646 vc1 G2224A E616Khtp-3 F57C9.5 CN823 vc13 G1785A E469=htp-3 F57C9.5 CN727 vc2 G2029A E551Khtp-3 F57C9.5 CN1362 vc23 C2048T P557Lhtp-3 F57C9.5 CN1369 vc24 G1905A S509=htp-3 F57C9.5 CN1425 vc25 G2181A Q601=htp-3 F57C9.5 CN1630 vc26 G2230A V618Ihtp-3 F57C9.5 CN1723 vc27 C1245T P306Lhtp-3 F57C9.5 CN1735 vc28 C2331T Y651=htp-3 F57C9.5 CN825 vc3 G1333A R335=M03C11.2 CN1246 vc57 G4804A E680KM03C11.2 CN1479 vc58 G5097A G725DM03C11.2 CN1543 vc59 C4755T I663=M03C11.2 CN1643 vc60 C4739T P658LM03C11.2 CN1712 vc61 C5740T H782YCki-2 T05A6.2 CN843 vc20 C413T T123ICki-2 T05A6.2 CN1157 vc31 G869A Non-codingCki-2 T05A6.2 CN1231 vc32 C338T T98ICki-2 T05A6.2 CN1254 vc33 G214A G57RCki-2 T05A6.2 CN1309 vc34 G524A E146KCki-2 T05A6.2 CN1364 vc35 G876A Non-codingCki-2 T05A6.2 CN1575 vc36 G370A V109MCki-2 T05A6.2 CN1643 vc37 C170T S42FCki-2 T05A6.2 CN1672 vc38 G148A E35KPage 6 of 16(page number not for citation purposes)Cki-2 T05A6.2 CN1787 vc39 G76A Splice JunctionBMC Genomics 2006, 7:262 http://www.biomedcentral.com/1471-2164/7/262tion rate was calculated to be one mutation every 293 kb(71 mutations in 14225 bp of DNA from 1464 animals).This is not significantly higher than the rate of one muta-tion in 300 kb seen for TILLING in Arabidopsis [23], and islower than the rate of one mutation in 156 kb reported forDrosophila [29]. Hence, the 0.025M dose of EMS appearsto be an adequate dose for TILLING based on comparisonwith these other systems in which this technique is cur-rently being used.Effects of mutations identified through TILLINGThe spectrum of mutations identified through forwardand reverse screens using EMS, although similar at thelevel of DNA sequence is much different when the effectsof the mutations are compared. Of the 194 single bpmutations we analysed from Wormbase (Additional File1) 50% resulted in missense alleles and the remaining50% in nonsense or splice junction mutations. In ourTILLING screen, because the selection of mutants was notbased on phenotype, 38% of the mutations we identifiedare predicted to be silent. These include mutations inintrons and intergenic regions and mutations that alterthe third bp of a codon such that it still encodes thewildtype amino acid. The majority of the mutations weidentified (59%) were missense alleles that alter theamino acid sequence of the protein encoded by the targetgene (Table 2). Of the 42 missense mutations identified inour screen, 17 of these may not have a significant effect onphenotype since the amino acid mutated was replaced byan amino acid of similar charge and polarity, but the 25remaining missense mutations are predicted to signifi-cantly affect the structure of the protein product of the tar-get gene by changing the charge or hydrophobicity of thisregion of the protein.Two of the mutations identified in our TILLING screen,phenotype because they truncate the protein product ofthe gene. One of these introduces a premature stop codoninto the third exon of gene mel-32 C05D11.11, and theother is a splice junction mutation that eliminates thesplice donor site of first intron of the gene cki-2 T05A6.2(Figure 3). The proportion of putative null mutationsidentified in our screen was 3% which is not significantlydifferent than the frequency of 2% seen in the DrosophilaTILLING study published [29] or than the 5% reportedfrom the much larger Arabidopsis TILLNG project [23].This frequency is higher than would be expected usingother chemical mutagens in C. elegans such as ENU [38]which cause a different spectrum of mutagenic events thatare more likely to result in missense than nonsense muta-tions.Pooling and library constructionA frozen library of approximately 1500 individual EMS-treated lines of C. elegans was constructed for this study(Figure 2), and DNA was isolated, purified, and arrayed inpools of eight as has been shown to work for TILLING inother diploid species such as Arabidopsis thaliana [22],Lotus japonicas [24], Zea maize [27] and Brassica oleracea[Gilchrist and Haughn, unpublished]. Purification of theDNA was found to be necessary both because the TILLINGreactions did not work on unpurified DNA samples andbecause accurate quantitation of the samples is essentialbefore pooling so that DNA from mutant animals is suffi-ciently represented in the 8-fold pools [27]. Both 4-foldand 12-fold pools were tested in C. elegans in order to con-firm that 8-fold pooling would be efficient (see Materialsand Methods for details). With 4-fold pooling all of themutant bands were detected, as they were with the 8-foldpools, whereas with 12-fold pooling only a subset of themutant bands were seen on the TILLING gel. Although 10-fold pools might be possible in C. elegans given that themdf-2 Y69A2AR.30 CN711 vc15 G243A D65Nmdf-2 Y69A2AR.30 CN902 vc17 C1083T Non-codingmdf-2 Y69A2AR.30 CN1613 vc49 C838T Non-codingmdf-2 Y69A2AR.30 CN1703 vc50 C838T Non-codingmdf-2 Y69A2AR.30 CN1865 vc51 C520T Non-codingmdf-2 Y69A2AR.30 CN1114 vc74 G76A Non-codinghtp-2 Y73B6BL.2 CN750 vc14 C507T A139Vhtp-2 Y73B6BL.2 CN50 vc22 C756T T222Ihtp-2 Y73B6BL.2 CN1271 vc29 C712T S207=htp-2 Y73B6BL.2 CN1540 vc30 G345A R85Qhtp-2 Y73B6BL.2 CN574 vc5 G49A D17Nhtp-2 Y73B6BL.2 CN901 vc6 G878A G263ROne letter nucleotide and amino acid codes follow IUPAC-IUB nomenclature. The first letter in the Nucleotide Change column indicates the wildtype nucleotide at this site, followed by the position of the mutation from the start codon in the genomic DNA and then the mutant nucleotide. The first letter in the Effect column indicates the wildtype amino acid at this site, followed by the position of the mutation within the predicted protein sequence and then mutated amino acid. An equal sign after the amino acid position means no change in the amino acid encoded by that codon, and an asterisk indicates a stop codon. Mutations in introns and intergenic regions are designated "Non-coding".Table 2: Mutations identified by TILLING. (Continued)mel-32 C05D11.11(vc68) and cki-2 T05A6.2(vc39), are genome size of this organism is slightly smaller than Ara-Page 7 of 16(page number not for citation purposes)predicted to result in a complete loss-of-function, or null, bidopsis, the construction of libraries for screening in 96 orBMC Genomics 2006, 7:262 http://www.biomedcentral.com/1471-2164/7/262384-well plates dictates that pooling is only efficient inmultiples of eight or 12, and since mutations are missedwith 12-fold pooling in C. elegans, libraries constructedfor this TILLING study were pooled in 8-fold aliquots.Our first library consisted of 696 rather than the planned768 (8 × 96) mutagenised worms because the DNA qual-ity in 72 of the 768 lines established was too poor to beused for TILLING. Only 8-fold column pools were con-structed for this library and when a mutation was detectedin a column pool well, each of the eight individuals thatmade up that pool was examined by TILLING in order todetermine which strain carried the mutation. For the sec-ond library of 768 animals, however, both 8-fold rowpools and 8-fold column pools were constructed andscreened. The DNA from individual worms was arrayed inples from one row or one column into a single pool,resulting in a total of 96 pools. The identity of the straincarrying a mutation detected on the TILLING gels wascomputed automatically by cross-referencing the datafrom the row and column pools.In theory, the method used for screening library #1 onlyrequired an average of 120 reactions: one 8-fold columnpool, plus eight individuals for each of three mutationsdetected (24 additional reactions). However, the 24 reac-tions done to determine which individual from a pool car-ried the mutation had to be set up manually and oftenneeded to be repeated in order to ensure that all of thereactions worked. Thus, this pooling strategy was moretime-consuming and often required almost as many TILL-ING reactions as the screening of both row and columnGene models depicting the distribution of different types of mutations within the genesFigure 3Gene models depicting the distribution of different types of mutations within the genes. The figure was designed from PARS-ESNP [42] output files. Blue lines indicate the extent of the amplified region that was used for TILLING. Orange open boxes denote exons. Purple up arrows indicate a change in the DNA sequence that does not affect the amino acid product. Purple down arrows indicate a change in non-coding DNA. Black up arrows indicate a change that induces a missense mutation in the predicted protein product. Red up arrows indicate a premature stop codon or splice junction error.T05A6.2aC05C10.5aC43E11.2aM03C11.2C47D12.8C05D11.11F25H2.13F57C9.5Y69A2AR.30aY73B6BL.2Page 8 of 16(page number not for citation purposes)12, eight-by-eight grids so as to simplify pooling in eitherdirection. Then the DNA was pooled by combining sam-pools simultaneously. In addition, the row and columnpool strategy allowed false positive bands to be excludedBMC Genomics 2006, 7:262 http://www.biomedcentral.com/1471-2164/7/262from our study since a mutation was never followedunless it appeared in at least one channel in both row andcolumn pool gels.Selection of targeted regionsMutations identified through TILLING are randomly dis-tributed in the genome, thus making it possible to targetgenes of any size and at any location [23]. The average sizeof the region that is tested in one TILLING screen is usu-ally about 1500 bp and this is the size that was used formost genes in this study (Table 1). For the gene C05C10.5whose genomic sequence is only 788 bp we designedprimers that amplified a region of approximately 1200 bpto avoid screening excess intergenic DNA upstream ordownstream of the locus where mutations would have ahigher probability of being silent. In addition, for genemus-81 C43E11.2, the primer sets that we designed toamplify a product of 1500 bp gave multiple amplificationbands when used with our standard PCR conditions. Thuswe were forced to use a primer set that amplified a smallerregion in order to obtain a single, clean PCR product fromthis locus.For genes that are much larger than 1500 bp, twoapproaches have been used for TILLING in other systems.The web-based programme, CODDLE [39] was originallydesigned for use in the Arabidopsis thaliana TILLINGproject and assists researchers in designing primers toselect regions of a gene of interest that are most likely toprovide loss-of-function or deleterious alleles. Researchersregion of the gene in which they are most interested, forexample, a specific domain that is known to interact withanother gene or protein. A different approach was used fora recent Drosophila melanogaster TILLING project [29]. Inthis study, multiple primers were designed to amplifyoverlapping fragments of a gene so that the entire genecould be screened by TILLING. For our TILLING project inC. elegans, as mentioned previously, the average ampliconsize was 1500 bp and primers were designed to amplifythe region of the gene predicted by CODDLE to be mostsusceptible to EMS-induced mutations. Primer sequencesused in this study are listed in Table 3. For C. elegans,where the average gene size is 3000 bp and introns aregenerally small [40], TILLING should prove to be evenmore efficient than in other species where larger genesand intervening sequences are the norm. Indeed, 62% ofmutations recovered in this C. elegans screen are predictedto have an effect on the protein product (Table 2),although long-term studies are required to determineexactly how many of these mutations will result in amutant phenotype.Identification of individual mutant animalsEach F1 mutagenised line was frozen individually in orderto ensure that identified mutations could be recoveredeven if a sample was frozen and thawed multiple times(see Figure 2). Except in the case of deleterious mutationsthat affect viability or reproduction, each mutant allelethat is heterozygous in the F1 parent is expected to bepresent in 3/4 of the progeny of this individual. If theTable 3: Primers used to amplify target genes in pilot C. elegans TILLING projectGene Oligo Name Oligo SequenceF57C9.5 ce0001Lb GTGCTGAGAATCCTGAACTTGACGce0001R TCTACTTGGCATGTTCGGCGACTGY73B6BL.2 ce0002L GGGTTCGCGAATTTCACTTGCATTce0002R CGGCTCCTCTGCGAGTAGTTGGTCT05A6.2 ce0003L GCGGCGCACTCACATTTTTCTCTTce0003R CTGTGCGGACTTTGGCACATTTGAC05C10.5 ce0004L GAACTATTTGTGCGCGCGCGTTTce0004R TCAATGAGTGGGGTGGATTCAAGAAGAC43E11.2 ce0006-3L CTCCGAAATGAGAACTGTCCGACCAATce0006-3R AAAGCTGAAGAAGTCGAATCGGTGCATY69A2AR.30 ce0007L CGCGATTTCCCTCAAAGGATCTGCce0007R AGAGCACCATCACACCACCTGACGF25H2.13 ce0008L TCAAAAAGAGACGAAGCCGCTGGAce0008R GCAGCAGCAACATCTTGAGCGTGTM03C11.2 ce0009-2L CAGCTCAGCTTCTCGTGGAGACCCTATce0009-2R AGGAATCTTTAGAGCAACCGGGCAAAAC47D12.8 ce0010L CCGGAATCGCATTGATTCCAAAAGce0010R TGCAGCGAAATCACTTACAATCGTTTCCC05D11.11 ce0011L CGCCACAAGTACACCAACAACGAGAAce0011R GCGAGATCAGCGACGTCTTTCTTGAPage 9 of 16(page number not for citation purposes)also have the option of requesting that CODDLE designprimers that will amplify a fragment within a specificmutant allele is recessive lethal, then the frequency ofprogeny carrying this allele should be 2/3. In this study weBMC Genomics 2006, 7:262 http://www.biomedcentral.com/1471-2164/7/262were able to recover individual descendants of F2 wormscarrying the mutant alleles in 19 of the 20 strains thawedfor testing. The remaining strain proved to be a false pos-itive isolated from our screening of our first library. Theprobability of this type of error occurring with our currentmethodology of screening both row and column pools isvery low.There are several methods that can be used to follow thesegregation of point mutations in F2 and subsequent pop-ulations. We used three different strategies for comparisonin this study: TILLING, cleaved amplified polymorphicsequences (CAPS) [41], and direct sequencing. TILLINGwas the most expensive and labour-intensive methodbecause of the requirement to purify the extracted DNAfor PCR and because detection of homozygous individu-als necessitated duplexing the DNA from the putativemutant with wildtype DNA (since fragments are onlycleaved with CJE if there is a mismatch in the DNA). Inaddition, multiple rounds of TILLING were sometimesnecessary if one of the two reactions (duplexed and non-duplexed DNA) failed, because in such a case the zygosityof the animal in question could not be determined. CAPSwas tried if the PARSESNP programme ([42] indicatedone or more restriction enzyme polymorphisms betweenthe mutant and wildtype sequences (Figure 4). Sixty of the71 mutations we identified (85%) in this screen are of thistype. For the samples where there were no restrictionenzyme polymorphisms either TILLING or directsequencing was used to detect mutant individuals becauseof time constraints, although studies have shown that thesive method for following mutations in the long-term[43].Results from direct sequencing of DNA from F2 descend-ants of mutant animals showed that the mutation of inter-est was present in an average of two out of three of thethawed progeny (43 out of 64 samples), although this var-ied from a low of one out of eight with F25H2.13(vc9),and a high of 13 out of 16 with htp-3 F57C9.5(vc2). Bothof these alleles with non-typical segregation patterns aremissense mutations and, although the vc2 allele does nothave an obvious phenotype in homozygous mutants, wespeculate that it may be incompatible with the wildtypeallele of this locus since heterozygous individuals carryingboth mutant and wildtype alleles together were neverrecovered. The high frequency of homozygous vc2mutants compared to wildtype individuals may just be astatistical anomaly or may indicate that the mutant pro-tein confers some type of fitness advantage upon animalscarrying it.Two other mutations, F25H2.13(vc10) and mel-32C05D11.11(vc11), were apparently homozygous in theparent F1 strains in which these alleles were detectedbecause all of the thawed progeny from the F2 plates werefound to be homozygous for these mutations. All othermutations were heterozygous in the F1 generation, asexpected, because EMS-induced damage in C. elegans usu-ally occurs in either the P0 egg or the P0 sperm before fer-tilization. Neither of the homozygous mutations waspresent as a background mutation in the wildtype strainRestriction enzyme digests of DNA from heterozygous and homozygous mutantsFigure 4Restriction enzyme digests of DNA from heterozygous and homozygous mutants. A) CAPS analysis of sibling lines for CN646 htp-3(vc1) using the restriction enzyme Taq1. The lanes labelled N2 are wildtype controls. Lane marked 4 exhibits additional bands when digested with this enzyme indicating this line is heterozygous for the vc1 mutation. B) CAPS analysis of sibling lines for CN711 mdf-2(vc15), using the restriction enzyme Hinf1. The lanes labelled N2 are wildtype controls. Lanes marked 4, 5 and 6 show additional cleavage bands and are missing the wildtype band indicating that they are homozygous for the vc15 mutation.A) B)Page 10 of 16(page number not for citation purposes)derived cleaved amplified polymorphic sequences(dCAPS) technique works well and would be a less expen-used for mutagenesis, or in any of the other mutant strainswhose DNA was sequenced. The F1 homozygosity of vc10BMC Genomics 2006, 7:262 http://www.biomedcentral.com/1471-2164/7/262and vc11 may be the result of an early mitotic crossover orother heritable event that occurred in the developingzygote rather than in the maternal germ cells, during orafter EMS mutagenesis. The fact that these mutations weredetectable at all using TILLING is curious since a DNAmismatch is needed for cleavage with CJE. In the pooledsamples, adequate wildtype DNA from other sampleswould have been amplified and paired with the mutantDNA for cleavage to occur, but when testing the DNAfrom the individual F1 animals no wildtype DNA shouldhave been present since we did not detect any wildtypeprogeny from these animals. It is possible that only thegerm cells of the F1 animal carried the mutations identi-fied in our screen and that wildtype DNA from thesomatic cells of this animal was sufficient to allow forcleavage with CJE when amplified using PCR. It is alsopossible that some wildtype DNA contamination waspresent in these reactions and was amplified along withthe homozygous DNA from the mutant. The fact that thecleavage bands were very faint on the TILLING gels is con-sistent with either of these ideas.An additional polymorphism in gene F25H2.13 was iden-tified when sequencing other alleles of this gene andfound to be homozygous in all TILLING strains and in theN2 P0 strain utilised for mutagenesis in this study, makingit clear that this strain is different from the N2 strain usedin the C. elegans genome sequencing project. Althoughthis polymorphism induces an amino acid change in theprotein sequence, it has no obvious effect on gene func-tion since animals carrying the polymorphism are seem-ingly wildtype.Analysis of mutants identified through TILLINGSome of the loci chosen as candidates for this pilot projectwere genes that are thought to play roles in chromosomesegregation, recombination or genome maintenance. Insome cases, RNAi constructs of these genes had beenshown to induce varying phenotypes, and we reasonedthat null and missense alleles of these targets would allowus to better identify the true function of these genes. Inother cases, the genes we targeted had no known RNAiphenotype, but had been implicated in meiotic functionsthrough two-hybrid or bioinformatics studies.Examination of the sequence of the 71 alterations shownin Table 2 revealed that 27 of the changes resulted in noamino acid change (silent mutations) and these strainsare unlikely to have visible phenotypes. Of the remaining44 mutations resulting in amino acid changes, 24 affectedcharged or conserved residues. These are the alleles thatare most likely to affect the functioning of the gene prod-uct and thus most likely to have phenotypic changes. Wefor 16 of these alleles. Although further studies are neededto confirm these data, the TILLING alleles reported hereclearly allow characterisation of these genes in a mannerthat was not previously possible.mel-32 C05D11.11The gene, mel-32 C05D11.11 was selected simply as a con-trol because many mutations at this locus have been iden-tified through forward genetic screens and sequencing ofthese indicates that most of the amino acids in theencoded protein are essential for normal gene function[46]. Mutations in this gene result in a maternal-effectlethal phenotype (Mel). The homozygotes are viable andfertile, but produce eggs that fail to hatch. The putativenull allele isolated by TILLING (vc68) does indeed have aMel phenotype. The three remaining missense alleles arecurrently being examined, although preliminary evidenceindicates that one of them (vc11) is a conservative changethat does not appear to confer any mutant phenotype.C05C10.5A gene name has not yet been assigned for this locus (indi-cated by * in Table 1). Little is known about this gene orthe function of the gene product. RNAi treatment pro-duces variable results ranging from embryonic lethality tohigh incidence of males (Him). We have two TILLed alle-les (vc21 and vc40) that were isolated as heterozygotes andare being maintained as such. As a consequence we havenot yet determined whether or not these alleles will causea mutant phenotype.mus-81 C43E11.2We have isolated the first genetic mutations in this genethrough TILLING. Three of the missense mutations (vc42,vc46 and vc47) have a radiation-sensitive (Rad) pheno-type. The mutant strains, which can be maintained ashomozygotes, vary in the severity of their response to radi-ation, and thus will provide valuable resources for dissect-ing the molecular characteristics of this gene.Homozygous mus-81(vc46) animals are severely radiationsensitive, while animals carrying either mus-81(vc42) ormus-81(vc47) exhibit less severe phenotypes. These phe-notypes are consistent and continue to segregate with themolecular mutations even after multiple outcrossings.xpf-1 C47D12.8This gene is the C. elegans orthologue of the essentialnucleotide excision repair gene XPF/ERCC4 [48]. We haveidentified three mutations in this gene (vc18, vc19 andvc67). All of these are missense alleles, and we have deter-mined that two of them (vc19 and vc67), have a Rad phe-notype as would be predicted. None of the alleles is lethalsince they can be maintained as homozygotes.Page 11 of 16(page number not for citation purposes)have done preliminary phenotypic analysis on 25 of theTILLING alleles we identified and observed phenotypesBMC Genomics 2006, 7:262 http://www.biomedcentral.com/1471-2164/7/262F25H2.13A gene name has not yet been assigned for this locus (indi-cated by * in Table 1), but it is predicted to encode aDEAD helicase closely related by sequence to DOG-1.RNAi treatment reveals no detectable phenotype and thedeletion allele (tm1866) is listed in Wormbase [7] ashomozygous viable. We have identified five new missensealleles through TILLING, and strains carrying these allelesare also viable although they appear to have reducedbrood sizes.htp-3 F57C9.5Excellent antibodies are available for studying the proteinproduct of this HIM-3 paralogue in vivo, but the deletionallele htp-3(gk26) is associated with a complex rearrange-ment that includes a wild-type copy of the gene, makingphenotypic analysis impossible. RNAi experimentsrevealed that the animals exhibit no phenotype when theworms are fed a dsRNA construct for htp-3 F57C9.5 [44],but injection of the dsRNA results in severe embryoniclethality as a consequence of chromosome nondisjunc-tion [45]. We have isolated five new missense alleles ofthis gene by TILLING, and observed varying levels ofembryonic lethality that segregate with the mutant allelefor three of these mutations (vc1, vc75 and vc77).M03C11.2A gene name has not yet been assigned for this locus (indi-cated by * in Table 1), but the gene is a member of theDEAD helicase family related to DOG-1. RNAi treatmentproduces arrested embryos, and a deletion allele (tm2188)has been shown to be sterile, but we have not yet deter-mined whether any of the missense alleles we have iden-tified by TILLING have phenotypes.cki-2 T05A6.2The knockout allele of this gene cki-2(ok741) causes steril-ity in homozygous animals as does the TILLING mutationcki-2(vc39) which affects a conserved splice junction site.The strain is easily maintained as a heterozygote, however,and can used for genetic analysis in this way.mdf-2 Y69A2AR.30This gene was studied previously using RNAi and shownto have reduced brood size and increased incidence ofmales [47]. Using TILLING, we have successfully identi-fied the first genetic mutation in this gene. The mdf-2(vc15) mutation can be maintained homozygouslydespite exhibiting reduced brood size and high incidenceof males, and thus provides a valuable tool for the studyof metaphase to anaphase checkpoint signalling.htp-2 Y73B6BL.2produces a Him phenotype. We have five TILLed alleles ofthis locus that are viable as homozygotes and for whichwe have observed no obvious phenotype.One of the major advantages of TILLING is that it can notonly identify null mutations which eliminate the functionof the gene product entirely, but also missense alleles thatresult in a partial loss or change of gene function andwhich can allow disruption of specific domains within agene and are especially useful for suppressor screenswhich can be used to identify interacting genes. Thereverse genetic techniques that are presently being used inC. elegans are all more likely to result in complete-loss-of-function alleles which, if the effect is lethal or detrimental,may limit the analysis that can be done. With TILLING,however, it is possible to use missense mutations in differ-ent regions of the gene for the dissection of multiple func-tions and interactions of a given gene product. While thepoint mutations that TILLING identifies can result incomplete loss-of-function alleles that are as effective asdeletions in knocking out gene function, this techniquecan also identify partial loss-of-function alleles or otheralterations of gene function that can be extremely usefulfor investigating the function of essential genes or genesencoding proteins with multiple domains. A well-knownexample illustrating the value of an allelic series of muta-tions is the elucidation of the functions of the let-60 geneof C. elegans which encodes a member of the GTP-bindingRAS proto-oncogene family involved in signalling(reviewed in [49]. Different mutations in this gene canhave recessive, semi-dominant, or dominant phenotypesthat define functions for the protein in developmentalprocesses as diverse as vulval induction, migration of thesex myoblasts, function of chemosensory neurons, pro-gression through pachytene in meiosis I, and differentia-tion of the excretory cell. Thus, a comprehensiveunderstanding of the biology of a given gene is oftenrevealed using non-null mutations. In this study we haveidentified 42 new missense mutations and two nonsensemutations that are available for genetic studies, and pre-liminary analysis indicates that at least some of these havedeleterious effects on phenotype.ConclusionsWe have used TILLING in C. elegans to determine the spec-trum of mutations induced by EMS in this species andfound that 96% of the mutations we identified were G/C-to-A/T transitions. In this pilot project we identified 71point mutations in 10 genes, of which 44 or more mayhave an effect on gene function. For seven of the genes wetargeted no mutant strains were previously available fromthe Caenorhabditis Stock Centre. One of the remaining tar-get genes had a deletion mutation, but the strain carryingPage 12 of 16(page number not for citation purposes)This gene encodes a paralogue of HIM-3 and has beenshown to play a role in meiotic function. RNAi treatmentthis mutation was shown to also carry a wildtype copy ofthe gene. Hence, for eight of the 10 target genes screened,BMC Genomics 2006, 7:262 http://www.biomedcentral.com/1471-2164/7/262TILLING has provided the first genetically heritable muta-tions which can be used to study their functions in vivo.A frozen library of more than 1500 EMS-mutagenisedworms was constructed, and enough DNA has beenextracted and purified to screen for mutations in morethan 5000 genes. The initial construction of the mutantlibrary is labour-intensive but, if well-constructed, itshould only be necessary to perform this step once.Approximately one gene, per Li-cor sequencer, per weekcan be screened after library construction is complete. Thecost and rate of TILLING is dependent partially on thequality of the DNA being screened (how reliably the reac-tions work) and the mutation rate (how many alleles areidentified per pooled population). Current estimates ofcost-per-gene vary from $1500 to $2500 USD, includingequipment, labour and consumables.TILLING appears to provide a reasonably high proportionof missense mutations in C. elegans probably, in part,because of the small size of C. elegans introns compared tosome other species. With EMS, the expected frequency ofnonsense and splice junction mutations from TILLINGscreens is approximately five percent in most species. Thetwo putative null alleles of this type that we have identi-fied (out of 71 mutations) represent a frequency not sig-nificantly different from the expected five percent. At theCAN-TILL facility TILLING is successfully being used inmany species for the detection of both induced and natu-ral variation (reviewed in [50]) and there appears to be nospecies bias in terms of performance. It would seem, how-ever, to be especially useful for genetically tractable organ-isms such as C. elegans where genomics tools are welldeveloped, but where reverse genetics techniques that canprovide heritable mutations suitable for genetic analysislag far behind.MethodsMutagenesis and library constructionN2 wild type hermaphrodites were exposed to 0.025 MEMS for 4 h (as in Brenner, 1974 [51], but at a lower EMSconcentration). After exposure to mutagen, the wormswere washed and allowed to recover for 1–2 h beforeyoung gravid hermaphrodites were picked to fresh plates(10 per plate). After 24 h, the mutagenised P0 animalswere transferred to fresh plates for a second 24 h brood. F1progeny were picked 1 per plate and allowed to self-ferti-lize. The F1 progeny of mutagenised worms were set upindividually rather than in pooled populations so as toultimately simplify the isolation of F3 animals carryingany mutations. This was considerably more work thanwould have been required with pooled F1 samples but,because worm libraries were frozen, the extra work was ations had exhausted the bacteria on the plate, worms werewashed off each plate with 2 ml of M9 buffer. One thirdof the population from each F1 line was used for DNA andthe other two thirds was frozen for future use as follows:from each plate, 0.5 ml of worms were mixed with 0.5 mlof 2X freeze solution (100 mM NaCl, 50 mM KPO4 pH6.0, 0.3 mM MgSO4, 30% glycerol) and frozen at – 80°Cto generate a frozen library stock.The remaining 1 ml of worms in M9 buffer for each linewas centrifuged at 14,000 g to pellet worms. Excess bufferwas aspirated leaving approximately 50 μl of buffer andworms in each tube. 50 μl of worm lysis buffer (50 mMKCl, 10 mM Tris (pH 8.3), 2.5 mM MgCl2, 0.45% Noni-det P-40, 0.45% Tween 20, 0.01% (w/v) gelatin, and 10mg/ml proteinase K) was added to each tube which wasthen frozen at -80°C for at least 1 h. DNA was extracted byproteinase K lysis at 57°C for 4 h with occasional vortex-ing, and then 100 μl of phenol:chloroform:isoamyl alco-hol (24:24:1) solution was added to the crude DNAextracts and tubes were vortexed for 5 minutes. Phaseswere separated by centrifugation at 14,000 g for 5 minutesand then aqueous layer was removed to a fresh tube con-taining 100 μl of chloroform, vortexed for 5 minutes andthe phases separated by centrifugation at 14,000 g. Theaqueous layer was again removed to a fresh tube contain-ing 400 μl of isopropanol, and then the DNA was precip-itated by centrifugation at 14,000 g for 10 minutes. DNApellets were washed once with 70 % ethanol, resuspendedin 50 to 100 μl ddH2O and quantified using a ND-1000Spectrophotometer (NanoDrop Technologies, Wilming-ton, DE, USA). Samples were diluted to 1 ng/ul in 10 mMTris, 1 mM EDTA pH 7.4, and 1 ml aliquots were distrib-uted in plates of 64 samples (8 rows by 8 columns) beforepooling. The DNA samples were then pooled 8-fold, inboth column and row directions, and then distributedinto 96-well plates of 8-fold column pools and 96-wellplates of 8-fold row pools for each library of 768 individ-uals.Primer design and PCR AmplificationPrimers were designed, using the web-based programmeCODDLE [39] and selecting "EMS (not TILLING)" as theMutation Method since a C. elegans option was not avail-able and we did not know how similar the spectrum ofmutations caused by EMS was in C. elegans compared toother organisms. Primers for C05C10.5 were designed toamplify a fragment of approximately 1200 bp since thegenomic size of this gene is only 788 bp. The other primersets were designed to amplify fragments as close to 1500bp as possible, given the structure of the DNA in theregion. Primers were purchased from MWG Biotech, Inc.(High Point, NC, USA), suspended to a concentration ofPage 13 of 16(page number not for citation purposes)one time occurrence and greatly simplified identificationof thawed mutants. When the mutagenised worm popula-100 uM in 10 mM Tris, 1 mM EDTA pH 7.4 and used at afinal concentration of 0.2 mM in a mixture of 3:2BMC Genomics 2006, 7:262 http://www.biomedcentral.com/1471-2164/7/262(labeled:unlabeled) for the forward (IRD700-labeled)primers and 4:1 (labeled:unlabeled) for the reverse(IRD800-labeled) primers as per Colbert et al., 2001 [52].PCR was also performed according to Colbert et al.,2001[52]: 10 ul PCR reactions with 2.5 ng – 5 ng of genomicDNA were used for amplification in 96-well or 384-wellPCR plates using ExTaq polymerase (Takara Bio Inc,Japan), but with 0.6 times the recommended concentra-tion of ExTaq buffer and 2 mM MgCl2. PCR cycles were asfollows: 95°C for 2 min; eight cycles of [94°C for 20 sec,73°C for 30 sec (decrementing 1°C per cycle), 72°C for 1min]; 45 cycles of: [94°C for 20 sec, 65°C for 30 sec, and72°C for 1 min]; 72°C for 5 min; 99°C for 10 min (dena-turation and inactivation of taq enzyme); and 70 cycles of20 sec at 70°C (decrementing 0.3°C per cycle for randomreannealing to allow hybridisation of mutant andwildtype molecules), hold at 4°C.Preparation of celery juice extractCrude celery juice extract (CJE) was prepared as describedby Till et al. [53]. Briefly, 0.5 kg of celery was processed ina kitchen-quality juicer until liquefied. Tris HCl (pH 7.7)was added to 0.1 M along with Phenylmethylsulpho-nylfluoride (PMSF) to 100 mM. The solution was spun at2600 G for 20 minutes and the supernatant removed,brought to 25% saturation in (NH4)2SO4, mixed for 30minutes at 4°C, and spun at 15,000 G for 40 minutes at4°C. The supernatant was removed again and adjusted to80% saturation in (NH4)2SO4, mixed for 30 minutes at4°C, and spun at 15,000 G for 1.5 hours at 4°C. The pelletfrom this cut was resuspended in 1/10 the starting volumeof 0.1 M Tris HCl (pH 7.7), 100 mM PMSF. The suspen-sion was dialysed against 8 L of the same buffer, fourtimes, for one hour each time at 4°C using Spectraporedialysis tubing (10,000 MW cut-off). Aliquots were storedat -70°C and were spun at approximately 2000 G for oneminute before use to remove any tissue debris.CJE digestion, sequence analysis and identification of mutantsPCR products were digested with CJE by adding 20 μl ofextract and buffer mix (100 mM MgSO4, 100 mM HEPES,300 mM KCl, 0.02% Triton X-100, 0.002 mg/ml BSA, and0.2% to 0.3% crude CJE) directly to the PCR reactions andincubating at 45°C for 15 minutes. Reactions werestopped by adding 2.5 μl of 0.5 M EDTA. The DNA waspurified by passage through G50 Sephadex in 96-well Mil-lipore Multiscreen® filtration plates (Millipore Corpora-tion, Billerica, MA) and concentrated for 30 minutes at90°C before running on a 25 cm long LI-COR acrylamidegel with a 0.4 mm wide, 96-well sharkstooth comb. Anal-ysis of the gel images was done using GelBuddy [54] todefine lanes and estimate sizes of cleavage products. Thewere sequenced in both directions using either the sameforward or reverse primers as for PCR or an internalprimer designed for sequencing. Sequence analysis wasperformed using Sequencher 4.2 (Gene Codes Corpora-tion, Ann Arbor, MI, USA) and the potential effect of themutations was predicted using PARSESNP [42].PoolingDNA from 4 strains previously shown to carry a singlemutation each was screened to test the efficiency of differ-ent pooling depths. Four bands were expected in each ofthe two channels of the Li-cor sequencing image whenamplified DNA from these individuals was run on a gel.With 4-fold pooling all eight mutant bands were detected(four in the 700 channel image and four in the 800 chan-nel image), as they were with the 8-fold pools, whereaswith 12-fold pooling, six of the eight mutant bands wereseen on one test gel (three in the 700 channel image andthree in the 800 channel image), and only three bands onthe second gel (two in the 700 channel image and one inthe 800 channel image). Based on these data and datafrom other studies, we conclude that 8-fold pooling wasthe best option for C. elegans.Identification of mutant individualsFrozen worms were thawed and 20 individuals wereplated 1 per 60 mm NGM plate and allowed to self-ferti-lize. Two approaches were taken to identify lines contain-ing the target mutation. In the first approach, the 20individuals from the strain carrying the mutation of inter-est were plated individually, allowed to grow until plateswere starved, and DNA was prepared as described for theDNA library construction. In the second approach, indi-vidual animals from the strain carrying the mutation ofinterest were allowed to lay eggs for 2–3 days, and thenthe parent was picked into 5 μl of lysis buffer and lysed at57 °C for 1 h followed by 95 °C for 15 m to inactivate theproteinase K. In both approaches the extracted DNA wasPCR amplified and the mutation detected either by TILL-ING, or by direct sequencing of the amplified DNA, or bydigestion of the PCR product with restriction enzymesresulted in a banding pattern different from wild type.Restriction enzyme site changes were found by PARSESNPto occur in 60 of 71 mutations.If homozygous lines were not identified, 20 more individ-uals were set up from one line that had been shown to beheterozygous and the progeny from each of these lineswas screened for evidence of deleterious mutations thatmight result in inviability. If sterile adults, dead embryosor larval lethals were observed, DNA was prepared fromthese and tested for the presence of the targeted mutation.Mutations that segregated with inviable phenotypes werePage 14 of 16(page number not for citation purposes)correlation of row and pool columns indicated whichindividual F1 worm carried the mutation. Most mutationsbalanced with genetic balancers to prevent the loss of themutation.BMC Genomics 2006, 7:262 http://www.biomedcentral.com/1471-2164/7/262Statistical comparison of results from TILLING in different organismsFrequencies of EMS-induced mutations identified duringdifferent TILLING experiments have been based on differ-ent samples sizes in different species. When comparingour data with others we used an on-line Proportions Test[55] to compare the frequencies we observed with thosereported in other species and determine whether or notobserved differences were significant (at the 90% confi-dence level) or were simply likely to be the result of sam-ple size differences.List of abbreviations usedbp: base pair(s);CAPS: cleaved amplified polymorphic sequences;CJE: celery juice extract;dCAPS: derived cleaved amplified polymorphicsequences;dsRNA: double stranded RNA;EMS: ethylmethanesulfonate;RNAi: RNA interference;TILLING: Targeting Induced Local Lesions in Genomes;TMP/UV: trimethylpsoralen and ultraviolet radiationAuthors' contributionsEJG and MCZ conceived of the study. GWH participatedin its design and coordination. AMR and NJO carried outthe mutagenesis and culturing of nematode strains. EJGperformed the TILLING and data analysis and wrote themanuscript. MCZ and NJO analysed mutant phenotypes.All authors read and approved the final manuscript.Additional materialAcknowledgementsThis work was supported by Canadian Institutes for Health Research authors would like to especially acknowledge the efforts of Sanja Tarailo and Shir Hazhir during library construction and sibbing analysis. We are also grateful to Quentin Cronk and the UBC Botanical Garden and Centre for Plant Research for laboratory space provided to CAN-TILL.References1. Hodgkin J: Introduction to genetics and genomics (September6, 2005).  WormBook 2005 [http://www.wormbook.org]. The C. ele-gans Research Community2. Barr MM: Super models.  Physiol Genomics 2003, 13:15-24.3. Hariharan IK, Haber DA: Yeast, flies, worms, and fish in thestudy of human disease.  N Engl J Med 2003, 348:2457-2463.4. Link CD: Invertebrate models of Alzheimer's disease.  GenesBrain Behav 2005, 4:147-156.5. Tamas I, Hodges E, Dessi P, Johnsen R, Vaz Gomes A: A combinedapproach exploring gene function based on worm-humanorthology.  BMC Genomics 2005, 6:65.6. Lai C, Chou C, Ch'ang L, Liu C, Lin W: Identification of NovelHuman Genes Evolutionarily Conserved in Caenorhabditiselegans by Comparative Proteomics.  Genome Res 2000,10:703-713.7. Chen N, Harris TW, Antoshechkin I, Bastiani C, Bieri T, Blasiar D,Bradnam K, Canaran P, Chan J, Chen CK, Chen WJ, Cunningham F,Davis P, Kenny E, Kishore R, Lawson D, Lee R, Muller HM, NakamuraC, Pai S, Ozersky P, Petcherski A, Rogers A, Sabo A, Schwarz EM, VanAuken K, Wang Q, Durbin R, Spieth J, Sternberg PW, Stein LD:WormBase: a comprehensive data resource for Caenorhab-ditis biology and genomics.  Nucleic Acids Res 2005, 33:D383-9[http://www.wormbase.org].8. Hillier LW, Coulson A, Murray JI, Bao Z, Sulston JE, Waterston RH:Genomics in C. elegans: so many genes, such a little worm.Genome Res 2005, 15:1651-1660.9. Wei C, Lamesch P, Arumugam M, Rosenberg J, Hu P, Vidal M, BrentMR: Closing in on the C. elegans ORFeome by cloning TWIN-SCAN predictions.  Genome Res 2005, 15:577-582.10. Li S, Armstrong CM, Bertin N, Ge H, Milstein S, Boxem M, VidalainPO, Han JD, Chesneau A, Hao T, Goldberg DS, Li N, Martinez M, RualJF, Lamesch P, Xu L, Tewari M, Wong SL, Zhang LV, Berriz GF, Jaco-tot L, Vaglio P, Reboul J, Hirozane-Kishikawa T, Li Q, Gabel HW,Elewa A, Baumgartner B, Rose DJ, Yu H, Bosak S, Sequerra R, FraserA, Mango SE, Saxton WM, Strome S, Van Den Heuvel S, Piano F,Vandenhaute J, Sardet C, Gerstein M, Doucette-Stamm L, GunsalusKC, Harper JW, Cusick ME, Roth FP, Hill DE, Vidal M: A map of theinteractome network of the metazoan C. elegans.  Science2004, 303:540-543.11. Reisner K, Asikainen S, Vartiainen S, Wong G: Developmental andBiological Insights Obtained from Gene Expression Profilingof the Nematode Caenorhabditis Elegans.  Curr Genomics 2005,6:97-107.12. Stein LD, Bao Z, Blasiar D, Blumenthal T, Brent MR, Chen N, Chin-walla A, Clarke L, Clee C, Coghlan A, Coulson A, D'Eustachio P, FitchDH, Fulton LA, Fulton RE, Griffiths-Jones S, Harris TW, Hillier LW,Kamath R, Kuwabara PE, Mardis ER, Marra MA, Miner TL, Minx P,Mullikin JC, Plumb RW, Rogers J, Schein JE, Sohrmann M, Spieth J, Sta-jich JE, Wei C, Willey D, Wilson RK, Durbin R, Waterston RH: Thegenome sequence of Caenorhabditis briggsae: a platform forcomparative genomics.  PLoS Biol 2003, 1:E45.13. Schwarz EM: Genomic classification of protein-coding genefamilies (September 23, 2005).  WormBook 2005 [http://www.wormbook.org]. The C. elegans Research Community14. Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE, Mello CC:Potent and specific genetic interference by double-strandedRNA in Caenorhabditis elegans.  Nature 1998, 391:806-811.15. Gonczy P, Echeverri C, Oegema K, Coulson A, Jones SJ, Copley RR,Duperon J, Oegema J, Brehm M, Cassin E, Hannak E, Kirkham M, Pich-ler S, Flohrs K, Goessen A, Leidel S, Alleaume AM, Martin C, Ozlu N,Bork P, Hyman AA: Functional genomic analysis of cell divisionin C. elegans using RNAi of genes on chromosome III.  Nature2000, 408:331-336.16. Maeda I, Kohara Y, Yamamoto M, Sugimoto A: Large-scale analysisof gene function in Caenorhabditis elegans by high-through-put RNAi.  Curr Biol 2001, 11:171-176.Additional File 1Table in MS Word that shows a list of mutations identified in forward genetic screens of EMS-treated C. elegans (obtained from Wormbase [7]).Click here for file[http://www.biomedcentral.com/content/supplementary/1471-2164-7-262-S1.doc]Page 15 of 16(page number not for citation purposes)(CIHR) grants to GWH, MCZ, and AMR, and by Natural Sciences and Engi-neering Research Council of Canada (NSERC) funding to AMR. The 17. Kamath RS, Martinez-Campos M, Zipperlen P, Fraser AG, Ahringer J:Effectiveness of specific RNA-mediated interferencePublish with BioMed Central   and  every scientist can read your work free of charge"BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime."Sir Paul Nurse, Cancer Research UKYour research papers will be:available free of charge to the entire biomedical communitypeer reviewed and published immediately upon acceptancecited in PubMed and archived on PubMed Central BMC Genomics 2006, 7:262 http://www.biomedcentral.com/1471-2164/7/262through ingested double-stranded RNA in Caenorhabditiselegans.  Genome Biol 2001, 2:RESEARCH0002.18. Timmons L, Fire A: Specific interference by ingested dsRNA.Nature 1998, 395:854.19. Zwaal RR, Broeks A, van Meurs J, Groenen JT, Plasterk RH: Target-selected gene inactivation in Caenorhabditis elegans byusing a frozen transposon insertion mutant bank.  Proc NatlAcad Sci USA 1993, 90:7431-7435.20. Williams DC, Boulin T, Ruaud AF, Jorgensen EM, Bessereau JL: Char-acterization of Mos1-mediated mutagenesis in Caenorhabdi-tis elegans: a method for the rapid identification of mutatedgenes.  Genetics 2005, 169:1779-1785.21. Berezikov E, Bargmann CI, Plasterk RHA: Homologous gene tar-geting in Caenorhabditis elegans by biolistic transformation.Nucl Acids Res 2004, 32:e40.22. Till BJ, Reynolds SH, Greene EA, Codomo CA, Enns LC, Johnson JE,Burtner C, Odden AR, Young K, Taylor NE, Henikoff JG, Comai L,Henikoff S: Large-Scale Discovery of Induced Point MutationsWith High-Throughput TILLING.  Genome Res 2003,13:524-530.23. Greene EA, Codomo CA, Taylor NE, Henikoff JG, Till BJ, ReynoldsSH, Enns LC, Burtner C, Johnson JE, Odden AR, Comai L, Henikoff S:Spectrum of Chemically Induced Mutations From a Large-Scale Reverse-Genetic Screen in Arabidopsis.  Genetics 2003,164:731-740.24. Perry JA, Wang TL, Welham TJ, Gardner S, Pike JM, Yoshida S, Parni-ske M: A TILLING reverse genetics tool and a web-accessiblecollection of mutants of the legume Lotus japonicus.  PlantPhysiol 2003, 131:866-871.25. Slade AJ, Fuerstenberg SI, Loeffler D, Steine MN, Facciotti D: Areverse genetic, nontransgenic approach to wheat cropimprovement by TILLING.  Nat Biotechnol 2005, 23:75-81.26. Smits BMG, Mudde J, Plasterk RHA, Cuppen E: Target-selectedmutagenesis of the rat.  Genomics 2004, 83:332-334.27. Till BJ, Reynolds SH, Weil C, Springer N, Burtner C, Young K, BowersE, Codomo CA, Enns LC, Odden AR, Greene EA, Comai L, HenikoffS: Discovery of induced point mutations in maize genes byTILLING.  BMC Plant Biol 2004, 4:12.28. Wienholds E, van Eeden F, Kosters M, Mudde J, Plasterk RHA, Cup-pen E: Efficient Target-Selected Mutagenesis in Zebrafish.Genome Res 2003, 13:2700-2707.29. Winkler S, Schwabedissen A, Backasch D, Bokel C, Seidel C, BonischS, Furthauer M, Kuhrs A, Cobreros L, Brand M, Gonzalez-Gaitan M:Target-selected mutant screen by TILLING in Drosophila.Genome Res 2005, 15:718-723.30. Anderson P: Mutagenesis.  Caenorhabditis elegans: Modern BiologicalAnalysis of an Organism 1995. none31. Burns PA, Allen FL, Glickman BW: DNA sequence analysis ofmutagenicity and site specificity of ethyl methanesulfonatein Uvr+ and UvrB- strains of Escherichia coli.  Genetics 1986,113:811-819.32. Klungland A, Laake K, Hoff E, Seeberg E: Spectrum of mutationsinduced by methyl and ethyl methanesulfonate at the hprtlocus of normal and tag expressing Chinese hamster fibrob-lasts.  Carcinogenesis 1995, 16:1281-1285.33. Kohalmi SE, Kunz BA: Role of neighbouring bases and assess-ment of strand specificity in ethylmethanesulphonate and N-methyl-N'-nitro-N-nitrosoguanidine mutagenesis in theSUP4-o gene of Saccharomyces cerevisiae.  J Mol Biol 1988,204:561-568.34. Pastink A, Heemskerk E, Nivard MJ, van Vliet CJ, Vogel EW: Muta-tional specificity of ethyl methanesulfonate in excision-repair-proficient and -deficient strains of Drosophila mela-nogaster.  Mol Gen Genet 1991, 229:213-218.35. Rogalski TM, Gilchrist EJ, Mullen GP, Moerman DG: Mutations inthe unc-52 gene responsible for body wall muscle defects inadult Caenorhabditis elegans are located in alternativelyspliced exons.  Genetics 1995, 139:159-169.36. Oleykowski CA, Bronson Mullins CR, Godwin AK, Yeung AT: Muta-tion detection using a novel plant endonuclease.  Nucl Acids Res1998, 26:4597-4602.37. Rosenbluth RE, Cuddeford C, Baillie DL: Mutagenesis inCaenorhabditis elegans : I. A rapid eukaryotic mutagen testsystem using the reciprocal translocation, eTI(III;V).  Mutation38. De Stasio EA, Dorman S: Optimization of ENU mutagenesis ofCaenorhabditis elegans.  Mutat Res 2001, 495:81-88.39. CODDLE   [http://www.proweb.org/coddle/]40. Spieth J, Lawson D: Overview of gene structure (in press).WormBook 2005 [http://www.wormbook.org]. The C. elegansResearch Community41. Konieczny A, Ausubel FM: A procedure for mapping Arabidop-sis mutations using co-dominant ecotype-specific PCR-basedmarkers.  Plant J 1993, 4:403-410.42. Taylor NE, Greene EA: PARSESNP: a tool for the analysis ofnucleotide polymorphisms.  Nucl Acids Res 2003, 31:3808-3811.43. Neff MM, Neff JD, Chory J, Pepper AE: dCAPS, a simple tech-nique for the genetic analysis of single nucleotide polymor-phisms: experimental applications in Arabidopsis thalianagenetics.  Plant J 1998, 14:387-392.44. Fraser AG, Kamath RS, Zipperlen P, Martinez-Campos M, SohrmannM, Ahringer J: Functional genomic analysis of C. elegans chro-mosome I by systematic RNA interference.  Nature 2000,408:325-330.45. Piano F, Schetter AJ, Mangone M, Stein L, Kemphues KJ: RNAi anal-ysis of genes expressed in the ovary of Caenorhabditis ele-gans.  Curr Biol 2000, 10:1619-1622.46. Vatcher GP, Thacker CM, Kaletta T, Schnabel H, Schnabel R, BaillieDL: Serine hydroxymethyltransferase is maternally essentialin Caenorhabditis elegans.  J Biol Chem 1998, 273:6066-6073.47. Kitagawa R, Rose AM: Components of the spindle-assemblycheckpoint are essential in Caenorhabditis elegans.  Nat CellBiol 1999, 1:514-521.48. Park HK, Suh D, Hyun M, Koo HS, Ahn B: A DNA repair gene ofCaenorhabditis elegans: a homolog of human XPF.  DNARepair (Amst) 2004, 3:1375-1383.49. Sternberg PW, Han M: Genetics of RAS signaling in C. elegans.Trends Genet 1998, 14:466-472.50. Gilchrist EJ, Haughn GW: TILLING without a plough: a newmethod with applications for reverse genetics.  Current Opinionin Plant Biology 2005, 8:211-215.51. Brenner S: The genetics of Caenorhabditis elegans.  Genetics1974, 77:71-94.52. Colbert T, Till BJ, Tompa R, Reynolds S, Steine MN, Yeung AT, McCa-llum CM, Comai L, Henikoff S: High-Throughput Screening forInduced Point Mutations.  Plant Physiol 2001, 126:480-484.53. Till BJ, Burtner C, Comai L, Henikoff S: Mismatch cleavage by sin-gle-strand specific nucleases.  Nucl Acids Res 2004, 32:2632-2641.54. Zerr T, Henikoff S: Automated band mapping in electro-phoretic gel images using background information.  Nucl AcidsRes 2005, 33:2806-2812.55. Lieberman Research Worldwide   [http://www.lrwonline.com/stattest/index.htm]yours — you keep the copyrightSubmit your manuscript here:http://www.biomedcentral.com/info/publishing_adv.aspBioMedcentralPage 16 of 16(page number not for citation purposes)Research/Fundamental and Molecular Mechanisms of Mutagenesis 1983,110:39-48.


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