. 13: 233–240 (1997)

Rapid Amplification of Uncharacterized Transposon-tagged DNA Sequences from Genomic DNA

KRISTIN T. CHUN1*, HOWARD J. EDENBERG2, MARK R. KELLEY3 AND MARK G. GOEBL1

1Department of Biochemistry and Molecular Biology and The Walther Oncology Center, Indiana University Schoolof Medicine, Indianapolis, Indiana 46202, USA2Departments of Biochemistry and Molecular Biology and of Medical and Molecular Genetics, Indiana UniversitySchool of Medicine, Indianapolis, Indiana 46202, USA3Department of Pediatrics, Section of Pediatric Endocrinology, and Biochemistry and Molecular Biology andThe Herman B Wells Center for Pediatric Research, Indiana University School of Medicine, Indianapolis,Indiana 46202, USA

Received 12 June 1996; accepted 9 September 1996

Although the entire DNA sequence of the yeast genome has been determined, the functions of nearly a third of theidentified genes are unknown. Recently, we described a collection of mutants, each with a transposon-taggeddisruption in an essential gene in Saccharomyces cerevisiae. Identification of these essential genes and characteriza-tion of their mutant phenotypes should help assign functions to these thousands of novel genes, and since eachmutation in our collection is physically marked by the uniform, unique DNA sequence of the transposable element,it should be possible to use the polymerase chain reaction (PCR) to amplify the DNA adjacent to the transposon.However, existing PCR methods include steps that make their use on a large scale cumbersome. In this report, wedescribe a semi-random, two-step PCR protocol, ST-PCR. This method is simpler and more specific than currentmethods, requiring only genomic DNA and two pairs of PCR primers, and involving two successive PCR reactions.Using this method, we have rapidly and easily identified the essential genes identified by several of our mutants.? 1997 by John Wiley & Sons, Ltd.

Yeast 13: 233–240, 1997.No. of Figures: 4. No. of Tables: 2. No. of Refs: 27.

— PCR; Saccharomyces cerevisiae; transposon

INTRODUCTIONWith the completion of the international effortto determine the entire DNA sequence of theSaccharomyces cerevisiae genome, all of its poten-tial genes are known. However, about 30% of these

genes have no known function (Williams, 1996).Recently, we described a collection of S. cerevisiaemutants in which essential yeast genes are mutatedand physically marked by a modified bacterialtransposable element (mTn3; Chun and Goebl,1996). Identification of these mutated genes andcharacterization of their phenotypes will advancethe effort to assign functions to the thousands ofnewly identified genes.The mTn3 transposable element provides a

unique, physical marker for each mutated gene.However, the large-scale characterization of these

*Correspondence to: Kristin T. Chun, Department of Biochem-istry and Molecular Biology, 635 Barnhill Drive, IndianaUniversity School of Medicine, Indianapolis, IN 46202-5122,USA.Contract grant sponsor: National Institutes of Health.Contract grant sponsor: American Cancer Society.Contract grant sponsor: Showalter Charitable Trust.

CCC 0749–503X/97/030233–08 $17.50? 1997 by John Wiley & Sons, Ltd

mutations requires a reliable, rapid, and simplemethod to identify the site of insertion. Thepolymerase chain reaction (PCR) enables re-searchers to identify and analyse specific DNAsequences quickly and easily, but in its originalform, this process requires knowledge of the DNAsequences at each end of the DNA being amplified(Mullis et al., 1986). For our collection of mutants,the DNA sequence of the transposon is known,but the DNA sequence of the surrounding geneof interest is not. Numerous modifications ofPCR that allow one to amplify unknown DNAsequences next to known sequences have beendescribed, but each includes steps that would makeits use on a large scale unwieldy. RACE (Frohmanet al., 1988), one-sided PCR (Ohara et al., 1989)and anchored PCR (Loh et al., 1989) all requirepolyA mRNA to generate the PCR template.When seeking DNA sequences that are not at the3* end of a cDNA (and lack the polyA tail), theseprocedures all require terminal deoxynucleotidyl-transferase to attach to the end of the unknownsequence a polynucleotide tail. Inverse PCR(Ochman et al., 1988), inverted PCR (Triglia et al.,1988), ligation mediated single-sided PCR(Mueller and Wold, 1989), single specific primerPCR (SSP-PCR; Shyamala and Ames, 1989) andpanhandle PCR (Jones and Winistorfer, 1992), allrequire the generation of a DNA fragmentcontaining the known DNA sequence as wellas the adjacent unknown sequences of interest(for example, restriction enzyme digestion). Inaddition, these five methods require the ligation ofa known DNA sequence to the unknown end. Inthese types of modified PCR, one primer annealsto the DNA of known sequence, a second annealsto the sequences added to the other end, and theintervening, unknown sequences are amplified.Several other PCR protocols (e.g. targeted gene

walking; Parker et al., 1991) employ a primer thatanneals specifically at or near the end of a regionof known sequence and a collection of primers thatanneal semi-specifically to a nearby, unknownsequence. Non-stringent annealing conditionsallow the second primers to anneal. A similarmethod, PCR mediated by a single primer (Parkset al., 1991), uses only one primer and relieson conditions where it anneals specifically toits complementary sequence as well as non-specifically to a nearby sequence. These methodsrequire none of the enzymatic steps describedabove, but they often yield multiple PCR products,since in addition to the expected DNA fragment,

they yield products generated from two primersannealing semi-specifically. The multiple PCRproducts generated must be screened by Southernanalysis to identify the desired fragment.In this report, we describe a novel PCR protocol

(semi-random, two-step (ST)-PCR) with which wecan identify the mutated gene in each of ourmutants with a haploid-lethal mutation. Involvingonly two successive PCR reactions and two pairsof PCR primers, this method is simpler and morespecific than the methods described above. For thefirst reaction, one primer anneals to the end of thetransposable element, while the other contains aspecific 20-nucleotide sequence followed by tenbases of degenerate sequence and a specific five-nucleotide sequence. A subset of these degenerateprimers anneals to an unknown DNA sequencenear the transposable element and allows initialamplification. The second pair of primers annealsto specific DNA sequences, one from the transpos-able element and the other from the 20-nucleotidesequence in the semi-random primer. Using thismethod, we have rapidly and easily identifiedessential genes present in our collection of yeastmutants with transposon-tagged, haploid-lethalmutations.

MATERIALS AND METHODS

ReagentsReagents for yeast and Escherichia coli media

were purchased from Difco, and chemicals werepurchased from Sigma. Agarose, restriction en-zymes, T4 DNA ligase, and Taq polymerase werepurchased from Boehringer Mannheim. Meta-Phor= agarose was purchased from FMC Bio-Products. The CircumVent= Thermal Cycle DNASequencing kit was purchased from New EnglandBiolabs, and the Sequenase> (version 2.0) kitwas purchased from United States BiochemicalCorporation. The CloneAmp> pAMP1 kit waspurchased from Gibco BRL, and the Geneclean>

kit was purchased from Bio 101. DNA primerswere synthesized by the Department of Bio-chemistry and Molecular Biology, BiochemistryBiotechnology Facility and the Herman B WellsCenter for Pediatric Research, both of the IndianaUniversity School of Medicine.

Media and growth conditionsYeast strains were grown in yeast extract, pep-

tone, dextrose (YPD) medium (Rose et al., 1990)

234 . . .

at 30)C. To identify auxotrophic markers, yeastwere grown on synthetic complete ‘drop-out’medium (Rose et al., 1990) in which all of thecommonly encountered auxotrophies were sup-plemented except for those being used in thegenetic selection. This medium was also sup-plemented with 2% glucose (SD medium). Liquidcultures were grown with aeration at 30)C.E. coli strains were grown at 36)C in LB medium

(Davis et al., 1980) and, to select for plasmids ofinterest, supplemented with ampicillin (100 ìg/ml).

Construction of pcdc53::mTn3 and thecorresponding mutant yeast strainTo construct a mTn3 disruption in CDC53, the

2·8 kb EcoRI/HindIII fragment containing CDC53(Mathias et al., 1996) was ligated into the corre-sponding sites in pHSS21 (Hoekstra et al., 1991).Using previously described stains and methods(Hoekstra et al., 1991), the mTn3(URA3) transpos-able element was then inserted into this gene togenerate the plasmid pcdc53::mTn3. To constructa yeast strain that is heterozygous for this dis-ruption mutation (KC108), the mutated gene wasexcised from the plasmid with NotI and used totransform yeast strain KC100 (Schiestl and Gietz,1989; Chun and Goebl, 1996). Ura+ transformantswere selected for on synthetic complete mediumlacking uracil. Transformants heterozygous for thedisruption mutation were identified by PCRand tetrad analysis (Kassir and Simchen, 1991;Sherman and Hicks, 1991).

PCR reactionsGenomic DNA was isolated (Sherman et al.,

1986) from wild-type and mutant yeast strains,and 1 ìl (approximately 100 ng) was used as thetemplate in each 20 ìl ST-PCR reaction with50 m-KCl, 10 m-Tris–HCl pH=8·3, 200 ìg/mlgelatin, 3 m-MgCl2, 200 ì of each dNTP,20 pmol each of PCR primers 1 and 2 (5* TAATTTAAAAGGATCTAGGTGAAGATCCTTTTTGA3* and 5* GGCCACGCGTCGACTAGTAC (N)10GATAT 3*, respectively) and 2·5 units of Taqpolymerase (Boehringer Mannheim) (Mathiaset al., 1994; Chun and Goebl, 1996). If ten-fold lesschromosomal DNA template is used, this PCRprotocol fails to produce a detectable product(data not shown). Reactions were carried out usingthe PCR1 program described in Table 1. Aftercompletion of the first PCR, the 20 ìl reaction wasdiluted with 80 ìl of water, and 1 ìl was used for

the second PCR (essentially the same as the first,except the PCR2 program (Table 1), and primers 3and 4 (5* (CAU)4 TGATAATCTCATGACCAAAATCCC 3* and 5* (CUA)4 GGCCACGCGTCGACTAGTAC 3*, respectively) are used instead).Except for the gel shown in Figure 4, PCR

products were separated in 1·0% agarose gelsin 0·5# TAE buffer (1# is 40 m-Tris-acetate,1·0 m-EDTA (disodium ethylenediaminetetra-acetate)). The gel shown in Figure 4 was composedof 2·0% MetaPhor= agarose (FMC Bioproducts,prepared according to the manufacturer’s instruc-tions) and 1·0# TAE buffer.

Cloning PCR products and DNA sequenceanalysisBecause primer 3 contains the sequence CAU

CAUCAUCAU and primer 4 contains CUACUACUACUA at their respective 5* ends, the Clone-Amp> pAMP1 System for Rapid Cloning ofAmplification Products (Gibco BRL) can be usedto clone the resulting PCR products into thepAMP1 bacterial plasmid.If the PCR products were cloned into pAMP1,

the DNA sequence was determined using theSequenase> kit (version 2.0, United States Bio-chemical Corporation). Otherwise, PCR products(5 ìl of the 20 ìl reaction) were separated in a 1·0%agarose, 0·5# TAE gel and visualized with ethid-ium bromide. The DNA band was excised andpurified using the Geneclean> kit. An aliquot ofthis DNA (5 ìl of the 20 ìl total) was analysedusing the CircumVent= Thermal Cycle DNA

Table 1. PCR programs.

PCR1 program PCR2 program

1. 94)C, 2 min 1. 94)C, 30 s2. 94)C, 30 s 2. 65)C, 30 s3. Initial temp.=42)C, 30 s, 3. 72)C, 3 min

"1·0)C for eachsubsequent cycle

GOTO 1 29#more

4. 72)C, 3 min 4. 4)C5. GOTO 2 5# more 5. End6. 94)C, 30 s7. 65)C, 30 s8. 72)C, 3 min9. GOTO 6 24# more10. 4)C11. End

235 -

Sequencing kit (New England Biolabs) andprimer 3.

RESULTS AND DISCUSSION

PCR protocolThe mutation in each of our mutant yeast

strains results from the insertion of a bacterialtransposable element (mTn3) (Chun and Goebl,1996; Hoekstra et al., 1991). Consequently, thetransposon physically marks each mutation. OurPCR protocol involves two successive PCR reac-tions (Figure 1), which together, are intendedto amplify the chromosomal DNA next to theinserted transposon. In the first reaction, chromo-somal DNA from one of our yeast mutants isincubated with a primer (primer 1) that anneals tothe transposable element and a degenerate primer(primer 2) composed of 20 bases of definedsequence, followed by ten random bases, followedby the bases GATAT (Figure 1 and Materials andMethods). During the synthesis of primer 2, amixture of all four nucleosides was incorporatedat each of the degenerate base positions, so in thisregion, the resulting collection of primers shouldinclude every possible permutation. Alternatively,nucleosides containing 2*-deoxyinosine or auniversal base, such as 3-nitropyrrole or 5-nitroindole, could be used as degenerate bases(Knoth et al., 1988; Nichols et al., 1994; Loakeset al., 1995).The sequence GATAT should occur approxi-

mately every 600 bp of the yeast genome, whichconsists of approximately 40% G+C. Assuming theoccurrence of this sequence follows a Poisson dis-tribution, there is an 80% chance that at least one‘GATAT’ occurs within 1 kb of any position in thegenome and a 90% chance that one occurs within1·5 kb (Figure 2). Therefore, at least one ‘GATAT’should be found near each transposon-tagged mu-tation. Consequently, at least one of the primer 2s islikely to anneal to the complement of the GATATsite (ATATC) near the transposon mutation and,

along with primer 1, promote the amplification ofthe DNA sequence in and next to the insertionmutation. The annealing temperatures for the firstsix rounds of amplification are relatively low toallow primer 2s with low melting temperatures aswell as those with some mismatches to anneal(Table 1). If the 15 3*-most bases of primer 2contribute the most to its melting temperature (Tm),a primer with ten As and/or Ts in the degenerateregion will have a predicted Tm of about 32)C(Thein and Wallace, 1986). Alternatively, a primerwith ten Gs and/or Cs will have a Tm of about 52)C.To favour primers that anneal under relativelystringent conditions, the first annealing tempera-ture is 42)C, and for each of the next five PCRcycles, the annealing temperature is one degreecooler.A portion of the resulting PCR product is

incubated in a second reaction with a second pairof primers (Figure 1 and Materials and Methods).Primer 3 anneals downstream of primer 1 andprimer 4 anneals to the complement of the 20 basesof defined sequence at the 5* end of primer 2. Thisreaction is intended to amplify the PCR pro-ducts from the first reaction that were specificallygenerated from primers 1 and 2.

Amplification of DNA sequences adjacent to amTn3 disruption in CDC53To test the accuracy of this method, we first

amplified the DNA sequences next to a known

Figure 1. Summary of ST-PCR scheme.

Figure 2. Probability that the sequence ‘GATAT’ will occurwithin a given interval of yeast genomic DNA. This graphassumes that the occurrence of ‘GATAT’ occurs randomly andconforms to a Poisson distribution and that the yeast genomeconsists of 40% G+C base pairs.

236 . . .

insertion. A mTn3 transposon insertion mutationwas generated in a plasmid-borne copy of CDC53(pcdc53::mTn3), and this mutated allele was usedto replace one of the wild-type copies in a diploidyeast strain (see Materials and Methods). Theplasmid and the chromosomal DNA from theyeast mutant (KC108) were each used as templatesfor ST-PCR. As shown in Figure 3, an approxi-mately 300 bp fragment was generated fromeach template. As expected, no DNA bands wereobserved when chromosomal DNA from the wild-type strain (KC10) without the transposable ele-ment was used as the template (lane C). Uponinspection of the CDC53DNA sequence, we foundthe nearest GATAT sequence is 248 bp upstreamof the mTn3 insertion mutation. Taking intoaccount the location of the primer 3 sequence inthe transposable element, we expected a PCRproduct of 338 bp, which agrees with the observedsize. The PCR product from the chromosomalDNA template was cloned and its DNA sequencewas determined to confirm that the sequence of thePCR-amplified insertion mutation is the same asthat of the original plasmid-borne mutation.

In theory, the second PCR might generate, inaddition to a single PCR product amplified withprimers 3 and 4, one or more PCR productsamplified with a pair of primer 4s (that hadannealed to a PCR1 product generated from twoprimer 2s). Surprisingly, we did not observe suchPCR products. It is possible that this is not acommon occurrence and would only be observedafter analysis of many more mutants. However, ifthe product(s) from the second PCR reaction arecloned into a vector in a way that requires that theends of the insert contain primers 3 and 4, theseextraneous PCR products are eliminated. In ourcase, specific DNA sequences were incorporatedinto the 5* ends of primers 3 and 4, and these wereused to insert the PCR product asymmetricallyinto a compatible plasmid (see Materials andMethods for details).

Identification of the mutated gene in anuncharacterized mTn3 disruption mutantTo identify a previously unknown mutation,

chromosomal DNA from an uncharacterizedmTn3 disruption mutant was used as the templatefor ST-PCR. As shown in Figure 3, a nearly 1·0 kbPCR product results. The sequence of this DNAwas determined directly using cycle sequencingand found to be from URA4, which encodes thethird enzyme of the pyrimidine pathway, a di-hydroorotase subunit (Materials and Methods;Guyonvarch et al., 1988). Examination of theDNA sequence in this region of chromosome XIIconfirms that the DNA sequence, GATAT, occursapproximately 1 kb away from where the mTn3transposable element inserted.

Identification of mutations in a novel gene,YIL106To assess the efficiency of this method, we used

ST-PCR to analyse the transposon-tagged muta-tion in a collection of 25 previously characterizedmutants, each with a disruption in open readingframe YIL106 (Chun and Goebl, 1996). The PCRproducts resulting from each mutant were sepa-rated in an agarose gel (Figure 4) that shows that aband was produced from 80% of the samples. TheDNA sequences from five samples were examined(Materials and Methods), and all five mutationswere confirmed to be in YIL106.However, unlike the first two mutations de-

scribed above, for these five, primer 2 does notanneal at the expected GATAT sequence. Instead,

Figure 3. ST-PCR products generated using plasmid or yeastgenomic DNA as templates. For the plasmid template,pcdc53::mTn3 (Materials and Methods) was linearized andsubjected to ST-PCR. For genomic cdc53::mTn3, #28 and ‘C’,approximately 100 ng of genomic DNA from yeast strainsKC108, KC028 and KC100, respectively (Materials andMethods) was subjected to ST-PCR. One-quarter of eachresulting PCR product was then separated by electrophoresis ina 1·0% agarose gel. The strain KC100 does not contain themTn3 transposable element.

237 -

the primer anneals to the complement of thesequence, GACAT. The sequence, GATAT, doesappear further from the transposable element.Apparently, even though there is a mismatch be-tween the 3* end of primer 2 and the region whereit was observed to anneal, the synthesis of thesmaller PCR product is more efficient than if theprimer annealed at the GATAT, which is 167 bpfarther away. Furthermore, the cloned sequencesin the degenerate portion of these primers do notmatch the wild-type sequence (Table 2). This resultis surprising, because it suggests that this region ofthe primer contributes little to determining whereit anneals. There is a possible explanation for this.During the first several amplification cycles, thisregion of the primer may be important for anneal-ing. However, the products generated from these

initial cycles all have the same 20 nucleotides ofdefined sequence from the 5* end of primer 2. Inthis situation, new primer 2s may anneal tightly tothese 20 nucleotides of exact match, and synthesiscould occur if the 3* end transiently anneals,whether or not the intervening ten nucleotidesmatch. Thus, virtually any primer 2 might func-tion, and since only about 10"6 of the primer 2smatch exactly, it is much more likely that a primer2 with mismatches will be incorporated.

Advantages of ST-PCRThis ST-PCR protocol for identifying the un-

known DNA sequences adjacent to a knownsequence overcomes many of the drawbacks ofpreviously described methods. Several of theseother methods require an additional step (e.g.terminal deoxynucleotidyl-transferase tailing orligation) to attach a known DNA sequence to theunknown sequences of interest. These steps can beinefficient and certainly require additional time.Other methods do not require these additionalenzymatic steps, but they employ less stringentannealing conditions to allow one of the PCRprimers to anneal semi-specifically in the regionof unknown sequences. These methods yieldPCR products of various different sizes that mustbe screened by Southern analysis to determinewhich contains the desired known DNA sequence.ST-PCR requires none of these additional enzy-matic steps and usually produces a single PCRproduct.A similar PCR protocol, thermal asymmetric

interlaced (TAIL)-PCR, was described (Liu andWhittier, 1995). Like ours, it does not requireadditional steps before PCR, and it does notrequire involved screening afterwards. How-ever, this method requires three successive PCRreactions, while ours involves only two.ST-PCR is a simple, efficient, and rapid method

to identify novel DNA sequences next to pre-viously known sequences. In addition to analysingsingle sequences of interest, it should also beapplicable to large-scale analyses. Although wehave used this method to analyse mutations inyeast, it should be possible to modify it for exam-ining more complex genomes, including those ofmammals.

Modifications of ST-PCR for other applicationsTo adapt this method for use where the known

DNA sequence is different from the mTn3

Figure 4. ST-PCR products generated using DNA from 25mutants, each with a transposon-tagged disruption in YIL106.Genomic DNA from each of 25 mutants with disruptions inYIL106 was used separately as the template for ST-PCR, andthe resulting products were separated on a 2·0% MetaPhor=

agarose gel (Materials and Methods). For ‘C’, DNA from thestrain KC100, which does not contain the mTn3 transposableelement, was used as the template.

Table 2. Comparison of cloned primer 2 DNA sequenceswith the YIL106 wild-type sequence.

Strain DNA sequence

Wild type AACACATCAA GACATMutant 6 CCAGTGCGAG GATATMutant 7 TTCGAATGCG GATATMutant 9 CTGTGTGGCG GATATMutant 10 TCGAGTCGTC GATATMutant 12 GGGTGACGCG GATAT

Bold letters denote bases which match the wild-type sequence.

238 . . .

transposable element, one would merely substituteprimers specific for the sequence of interest. If thetemplate DNA is from an organism other thanS. cerevisiae, one could change the sequence ofprimer 2. The specific 3*-most nucleotides couldbe chosen based upon the %G+C of the genomicDNA or on known biases in dinucleotide fre-quencies. In addition, one might increase thelength of this region to decrease the frequency ofthis primer annealing to the template and to in-crease its Tm, which would allow the use of higherannealing temperatures during the first six cyclesof the first PCR. This change would be expected toincrease the specificity of the first PCR. Alterna-tively, if the conditions described above yield noPCR product, one might use a mixture of two ormore types of primer 2s, each with a differentsequence at its 3*-end. Combining these modi-fications should allow one to tailor the specificityand the efficiency of ST-PCR, as well as theaverage product length.

ACKNOWLEDGEMENTS

We thank Prianto Moeljadi for excellent technicalassistance. This work was supported by NationalInstitutes of Health grant GM-45460, AmericanCancer Society grant IN-161-D, and a grant fromthe Showalter Charitable Trust to M.G.G. and apostdoctoral fellowship from the American CancerSociety (PF-4227) to K.T.C.

REFERENCES

Chun, K. T. and Goebl, M. G. (1996). The identificationof transposon-tagged mutations in essential genes thataffect cell morphology in Saccharomyces cerevisiae.Genetics 142, 39–50.

Davis, R. W., Botstein, D. and Roth, J. R. (1980). AManual for Genetic Engineering: Advanced BacterialGenetics. Cold Spring Harbor Laboratory, ColdSpring Harbor, NY.

Frohman, M. A., Dush, M. K. and Martin, G. R.(1988). Rapid production of full-length cDNAs fromrare transcripts: amplification using a single gene-specific oligonucleotide primer. Proc. Natl. Acad. Sci.USA 85, 8998–9002.

Guyonvarch, A., Nguyen-Juilleret, M., Hubert, J. C.and Lacroute, F. (1988). Structure of the Saccharo-myces cerevisiae URA4 gene encoding dihydroorotase.Mol. Gen. Genet. 212, 134–141.

Hoekstra, M. F., Seifert, H. S., Nickoloff, J. andHeffron, F. (1991). Shuttle mutagenesis: bacterialtransposons for genetic manipulations in yeast.Meth.Enzymol. 194, 329–342.

Jones, D. H. and Winistorfer, S. C. (1992). Sequencespecific generation of a DNA panhandle permits PCRamplification of unknown flanking DNA. Nucl. AcidsRes. 20, 595–600.

Kassir, Y. and Simchen, G. (1991). Monitoring meiosisand sporulation in Saccharomyces cerevisiae. Meth.Enzymol. 194, 94–110.

Knoth, K., Roberds, S., Poteet, C. and Tamkun, M.(1988). Highly degenerate, inosine-containing primersspecifically amplify rare cDNA using the polymerasechain reaction. Nucl. Acids Res. 16, 10932.

Liu, Y. G. and Whittier, R. F. (1995). Thermal asym-metric interlaced PCR: automatable amplification andsequencing of insert end fragments from P1 and YACclones for chromosome walking. Genomics 25, 674–681.

Loakes, D., Brown, D. M., Linde, S. and Hill, F. (1995).3-Nitropyrrole and 5-nitroindole as universal bases inprimers for DNA sequencing and PCR. Nucl. AcidsRes. 23, 2361–2366.

Loh, E. Y., Elliott, J. F., Cwirla, S., Lanier, L. L. andDavis, M. M. (1989). Polymerase chain reaction withsingle-sided specificity: analysis of T cell receptor deltachain. Science 243, 217–220.

Mathias, N., Bayes, M. and Tyler-Smith, C. (1994).Highly informative compound haplotypes for thehuman Y chromosome. Human Molecular Genetics 3,115–123.

Mathias, N., Johnson, S. L., Winey, M., et al. (1996).Cdc53 acts in concert with Cdc4 and Cdc34 to controlthe G1 to S phase transition and identifies a conservedfamily of proteins. Mol. Cell. Biol. 16, 6634–6643.

Mueller, P. R. and Wold, B. (1989). In vivo footprintingof a muscle specific enhancer by ligation mediatedPCR. Science 246, 780–786. [Published erratumappears in Science (1990) 248, 802.]

Mullis, K. B., Faloona, F. A., Scharf, S. J., Sakai, R. K.,Horn, G. T. and Erlich, H. A. (1986). Specific enzy-matic amplification of DNA in vitro: the polymerasechain reaction. Cold Spring Harbor Symp. Quant.Biol. 51, 263–273.

Nichols, R., Andrews, P. C., Zhang, P. and Bergstrom,D. E. (1994). A universal nucleoside for use at am-biguous sites in DNA primers. Nature 369, 492–493.

Ochman, H., Gerber, A. S. and Hartl, D. L. (1988).Genetic applications of an inverse polymerase chainreaction. Genetics 120, 621–623.

Ohara, O., Dorit, R. L. and Gilbert, W. (1989). One-sided polymerase chain reaction: the amplification ofcDNA. Proc. Natl. Acad. Sci. USA 86, 5673–5677.

Parker, J. D., Rabinovitch, P. S. and Burmer, G. C.(1991). Targeted gene walking polymerase chain reac-tion. Nucl. Acids Res. 19, 3055–3060.

Parks, C. L., Chang, L. S. and Shenk, T. (1991). Apolymerase chain reaction mediated by a singleprimer: cloning of genomic sequences adjacent to aserotonin receptor protein coding region. Nucl. AcidsRes. 19, 7155–7160.

239 -

Rose, M. D., Winston, F. and Hieter, P. (1990).Methodsin Yeast Genetics: A Laboratory Course Manual.Cold Spring Harbor Laboratory Press, Cold SpringHarbor, NY.

Schiestl, R. H. and Gietz, R. D. (1989). High efficiencytransformation of intact yeast cells using singlestranded nucleic acids as a carrier. Curr. Genet. 16,339–346.

Sherman, F. and Hicks, J. B. (1991). Getting startedwith yeast. Meth. Enzymol. 194, 3–21.

Sherman, F., Fink, G. R. and Hicks, J. B. (1986).Laboratory Course Manual for Methods in Yeast

Genetics. Cold Spring Harbor Laboratory, ColdSpring Harbor, NY.

Shyamala, V. and Ames, G. F. (1989). Genome walkingby single-specific-primer polymerase chain reaction:SSP-PCR. Gene 84, 1–8.

Triglia, T., Peterson, M. G. and Kemp, D. J. (1988). Aprocedure for in vitro amplification of DNA segmentsthat lie outside the boundaries of known sequences.Nucl. Acids Res. 16, 8186.

Williams, N. (1996). Yeast Genome Sequence FermentsNew Research. Science 272, 481.

240 . . .