cenetic analysis with random amplified

8/3/2019 Cenetic Analysis With Random Amplified

1/4

Plant Physiol. (1993) 101: 349-352

Cenetic Analysis with Random Amplified PolymorphicDNA Markers

Scott V. Tingey* and Joseph P. de1 TufoDu Pont Agricultura1 Products, E.I. du Pont de Nemours & Company, Wilmington, Delaware 19880-0402

For many years, the principles of genetics have been ap-plied to crop variety improvement with great success. Severa1crop species, notably corn, wheat, and tomato, have beenused as model genetic systems because of their central im-portance to food production. Until recently, virtually a11 pro-gress in both breeding and model genetic systems has reliedon a phenotypic assay of genotype. Because the efficiency ofa selection scheme or genetic analysis based o n phenotype isa function of the heritability of the trait, factors like theenvironment, multigenic and quantitative inheritance, or par-tia1 and complete dominance often confound the expressionof a genetic trait. Many of the complications of a phenotype-based assay can be mitigated through direct identification ofgenotype with a DNA-based diagnostic assay. For this reason,DNA-based genetic markers are being integrated into severalplant systems and are expected to play an important role inthe future of plant breeding.

The utility of DNA-based diagnostic markers is determinedto a large extent by the technology that is used to revealDNA-based polymorphisms. Currently, the technology ofchoice for many species is the RFLP assay. RFLP assays detectDNA polymorphisms through restriction endonuclease diges-tions, coupled with DNA blot hybridizations, and are, ingeneral, time consuming and labor intensive. Over the lastfew years, polymerase chain reaction technology has led tothe development of several nove1 genetic assays based onselective DNA amplification (Krawetz, 1989; Innis et al.,1990). One of the strengths of these new assays is that theyare more amenable to automation 'than conventional tech-niques. They are also simple to perform and are preferablein experiments where the genotype of a large number ofindividuals is to be determined at a few genetic loci. Unfor-tunately, because of a prerequisite for DNA sequence infor-mation, these assays are limited in their application.

Nearly 2 years ago, a new genetic assay was developedindependently by two different laboratories (Welsh andMcClelland, 1990; Williams et al., 1990). This procedure,which we have called the RAPD assay, detects nucleotidesequence polymorphisms in a DNA amplification-based as-say using only a single primer of arbitrary nucleotide se-quence. In this reaction, a single species of primer binds tothe genomic DNA at two different sites on opposite strandsof the DNA template. If these priming sites are within anamplifiable distance of each other, a discrete DNA product

* Corresponding author; fa x 1-302-695-4296.

is produced through thermocyclic amplification (Fig. 1).Thepresence of each amplification product identifies complete orpartia1 nucleotide sequence homology, between the genomicDNA and the oligonucleotide primer, at each end of theamplified product. On average, each primer will direct theamplification of several discrete loci in the genome, makingthe assay an efficient way to screen for nucleotide sequencepolymorphism between individuals. For example, the fre-quency of finding RAPD polymorphisms has been shown tobe 0.3 per primer in Ar a b i d o p s i s thaliana, 0.5 per primer insoybean, 1 per primer in corn, and 2.5 per primer in Neuro-s p o r a c r a s s a . The major advantage of this assay is that thereis no requirement for DNA sequence information. The pro-toco1 is also relatively quick and easy to perform and usesfluorescence in lieu of radioactivity (Williams et al., 1992).Because the RAPD technique is an amplification-based assay,only nanogram quantities of DNA are required, and auto-mation is feasible.

APPLICATIONS OF THE RAPD ASSAYThere are several applications of the RAPD assay that have

been developed over the past 2 years. Each of these tech-niques exploits the efficiency of detection of DNA sequence-based polymorphisms in the RAPD assay. The RAPDtechnology has quickly gained widespread acceptance andapplication because it has provided a tool for genetic analysisin biological systems that have not previously benefitted fromthe use of molecular markers.Development of Genetic Maps

One of the first practical uses of RAPD markers was in thecreation of high-density genetic maps. By using a moreefficient assay, Reiter et al. (1992) were able to place over250 new genetic markers on a recombinant inbred populationof A . thaliana in only 4 person-months, clearly demonstratingthe utility of RAPD markers for quickly saturating both aglobal and local genetic map.

Historically, many important crop systems have sufferedfrom a lack of genetic markers. For example, genetic linkageanalysis in conifers has been slow primarily due to the largesize of the genome and the inherent difficulty involved inproducing a segregating F2 population (Carlson et al., 1991).

Abbreviations: RAPD, random amplified polymorphic DNA;RFLP, restriction fragment-length polymorphism.349


2/4

35 0 Tingey and del Tufo Plant Physiol. Vol. 101, 1993

Primer spacing 200-2000 bpPrimer

ooooooooooi s - i 1 1 1 1 1 1 1 1 1 1 1 1 n \ i i i i 1 1 1 1 1 1 1 1 1 1 1 1 1 3 11 3 1 1 1 1 1 1 1 1 1 1 1 1 1 o n i M 1 1 1 1 1 M 1 1 1 MooooooooooIGIAITIAICICITIGIAIC^

Primer

-1.35

Figure 1. A, Schemat ic representat ion of the amplification of DNA with a single oligodeoxynucleotide pr imer. B,Amplification products from F 2 individuals segregating from a cross between Clycine max cv Bonus and Glycine so/aPI81762. The arrow points to a segregating locus, and the s ize s tandards are shown to the right of the gel. The RAPDreact ion was performed as described in Williams et al. (1990), and the products of the amplification react ion werevisualized by separation on a 1.4% agarose gel stained with ethidium bromide.

Carlson et al. (1991) and C haparro et al. (1992) hav e recentlyshown that th e speed an d efficiency of RA PD analysis hasmade m apping in conifers a reasonable e ndeavo r. For ex-ample, Chaparro et al. (1992) were able to create a 191-marker RAPD map of loblolly pine in only 6 person-months.Both th e A. thaliana and loblolly pine genetic maps weresynthesized with unprecedented speed, taking advantage ofthe unique reproductive biology of each system.Because RA P D polymorphisms are the result of either anucleotide base change tha t alters the primer binding site, oran insertion or deletion within the am plified region (Williamset al., 1990; Parks et al., 1991), polymorphisms are usuallynoted by the presence or absence of an amplifi cation productfrom a single locus. This also means that the R A PD techniquetends to provide only domin ant m arkers. Indi vidu als contain-ing two copies of an allele are not distinguished quantitativelyfrom those containing only one copy of the allele. T hedisadvantage of mapping with dominant markers is thatmarkers linked in repulsion, for example ma rkers residing onseparate chromatids, such as could be found in an F2 popu-lation, provide little information for the estimate of geneticdistance (Allard, 1956). Th erefore, when mapping with dom -inant markers, it is necessary to work with markers that areonly linked in coupling, i.e. markers residing on a singlechromatid as can be found in a backcross or recombinantinbred population, in haploid or gametophytic tissue, oralternatively in an F2 population where only R A PD markersamplified from a single parent are mapped (Williams et al.,1992). Genetic simulations show that dominant markerslinked in coup ling are as efficient for mapping as codom inantmarkers on a per gamete basis (Hanafey Maize GeneticsMeeting).

Targeting Genetic MarkersSeveral groups have used the RAPD assay as an efficienttool to ident ify molecular markers that lie within regions ofa genome introgressed during th e development of near iso-genic lines (Klein-Lankhorst et al., 1991; Martin et al., 1991;Paran et al., 1991). By defini t ion, an y region of the genomethat is polymorphic between two near-isogenic plants ispotentially linked to the introgressed trait. Thus, K lein-La nk-horst et al. (1991) were able to identify RA PD markers specificto chromosome 6 of tomato by screening a Lycopersicon

esculentum substitution line, an d Martin et al. (1991) wereable to confirm linkage of RA PD markers to the Pfo locus intomato after screening tw o near-isogenic lines. Paran et al.(1991) used tw o different sets of near-isogenic lettuce linesto ident ify RA P D markers linked to the Dml, Dm3, an d Dm 11locus. RA PD markers were 4 to 6 times more efficient, on aper assay basis, than was screening for these polymorphismsusing RFLP technology, an d were over 10-fold more efficientin time an d labor. Another advantage of this technology isthat a genetic map of the entire genome is not required toidentify markers linked to a trait of interest; instead, specificregions of the genome can be focused on.There are, however, two disadvantages to using near-isogenic lines to identify markers linked to a genetic trait. T hefirst is that it takes several generations of backcrossing tocreate a near-isogenic line. T he second is that frequentlythere are several regions of the donor genome that areinadvertently co-introgressed into the near-isogenic line(Young and Tanksley, 1989). This results in the identificationof marker polymorphisms between near-isogenic lines thatare not necessarily linked to the trait that is being studied.


3/4

Random Amplified Polymorphic DNA 35 1

Population CeneticsThe area of research that has shown the most growth with

respect to the use of RAPD technology is that of populationgenetics (Hedrick, 1992). RAPD markers have been used tocreate DNA fingerprints for the study of individual identityand taxonomic relationship in both eukaryotic and prokar-yotic organisms (Caetano-Anolls et al., 1991a; Hu and Qui-Tos, 1991; Welsh et al., 1991; Wostemeyer et al., 1991; Hadryset al., 1992; Kresovich et al., 1992; Lark et al., 1992; Stiles etal., 1992; Wilde et al., 1992). An important question iswhether RAPD bands of equal mo1 wt that are shared be-tween individuals are homologous characters (characters in-herited from a common ancestor) or homoplastic characters(characters that arise independently within a population). Itseems likely that closely related individuals would co-inherita shared character state from a common ancestor and unlikelythat they would acquire the same character independently.Williams et al. (1992) demonstrated this to be the case byusing single RAPD bands as hybridization probes to detecthomologous characters on a DNA blot of RAPD products.Within the limits of the resolution of an agarose gel, RAPDbands that were amplified from different species of the genusGlycine, and scored as homologous by relative mobility, werealso shown to be homologous by hybridization.

Severa1 groups have reported on the utility of RAPD mark-ers as a source of phylogenetic information. Arnold et al.(1991) were successful in using RAPD markers to test forinterspecific nuclear gene flow between lris fulva and 1.hexagona, and to study the presumed hybrid origin of 1.nelsonii. Hu and Quiros (1991) were able to show that theamplification products from only four random primers weresufficient to discriminate between 14 different broccoli and12 different cauliflower cultivars (Brassica oleracea L.). RAPDmarkers have also been used effectively to assess the amountof genetic diversity in germplasm collections. Using only 25different decamer oligonucleotide primers, Kresovich et al.(1992) collected information on 140 different polymorphiccharacters in a "test array" of individuals representing B.oleracea L. and B. rapa L. They showed the utility of the assayfor discriminating between different individuals in a germ-plasm collection and that the ability to distinguish betweenclosely related individuals was simply a function of thenumber of RAPD bands that were observed. RAPD markersprovided an efficient technology for discovering these poly-morphic characters.Using an arbitrary primer as short as five nucleotides,combined with silver staining to increase the sensitivity ofDNA band detection, Caetano-Anolls et al. (1991a) pro-duced a detailed and relatively complex DNA fingerprint forsevera1 different species. This approach, termed DNA ampli-fication fingerprinting, has been reviewed recently (Caetano-Anolls et al., 1992b) and promises to generate more geneticinformation from each amplification.

Pooling StrategiesRecently, another technology has been developed that is

designed to identify genetic markers linked to very specificregions of the genome. Arnheim et al. (1985) outlined agenome pooling strategy that allows RFLP markers to betargeted to a region of the genome that is not in linkageequilibrium as a result of selection at a particular locus. Thestrategy requires pooling genomic DNA from individuals thatare known to be genetically fixed at a particular locus. Mark-ers linked to that locus are identified by their linkage dis-equilibrium with respect to the rest of the population. Thelimitation of this approach is that it relies on RFLP technol-ogy, which is relatively inefficient for identifying polymor-phic regions of a genome, and, more importantly, on thepremise that linkage disequilibrium exists at the locus ofinterest within the source population.

Recently, Michelmore et al. (1991) have described the useof RAPD markers to screen efficiently for markers linked tospecific regions of the genome. This method, called bulksegregant analysis, uses two bulked DNA samples gatheredfrom individuals segregating in a single population. Eachbulk is composed of individuals that differ for a specificphenotype or genotype, or individuals at either extreme of asegregating population. For simple genetic traits, a11 loci inthe genome should appear to be in linkage equilibrium exceptin the region of the genome linked to the selected locus.Markers linked to this locus should appear polymorphicbetween the pools for alternate parental alleles. Becausemany segregating individuals are used to generate the pools,there is only a minimal chance that regions of the genomeunlinked to the target locus will also be polymorphic betweenthe pools. Random primers can then be used efficiently toamplify loci from each pool and to identify RAPD polymor-phisms linked to the trait of interest. Michelmore et al. (1991)have successfully used this technique to target markers to theDm5/8 locus in lettuce.

The advantage of this technology is that markers are tar-geted to a much smaller locus within the genome, and thelikelihood of identifying false positive markers is small (Mich-elmore et al., 1991) compared with near-isogenic line analy-sis. Selections made from an FZpopulation will always be inlinkage disequilibrium with respect to selected regions of thegenome, and markers can be targeted to any locus where anyform of selection can be applied, either phenotypic or geno-typic.

Giovannoni et al. (1991) demonstrated the use of a poolingstrategy, based on known RFLP genotypes from existingmapping populations, to create pools of DNA from individ-uals homozygous for opposite parental alleles in a targetedchromosomal interval. This method was used to target RAPDmarkers to regions of the tomato genome responsible for fruitripening and pedicel abscission. Reiter et al. (1992) used thispooling strategy to identify 100 RAPD markers specific tochromosome 1 of A . thaliana. As genetic maps approachsaturation, pooling on phenotype or genotype will allowresearchers to move away from a random approach to mapsaturation and focus more efficiently on specific regions ofthe genome.

CONCLUSIONDNA-based diagnostics are now well established as a

means to assay diversity at the locus, chromosome, and wholegenome levels. As technology has advanced, DNA sequence-


4/4

35 2 Tingey and de1 Tufo Plant Physiol. Vol. 101, 1993

based assays have become easier to use, more efficient atscreening for nucleotide sequence-based polymorphisms, a ndavailable to a wider cross-section of the genetics researchcommunity. The ultimate genetic assay would be based onthe determination of the complete D N A sequence at an ylocus of interest. Increased genetic resolution would be ob-tained simply by sequencing a larger contiguous segment ofD N A a t the locus. As a prelude to this, there ar e now severa1good examples of DNA sequence-based diagnostic assaysthat are designed to identify the presence or absence of aspecific nucleotide bp a t a discrete locus (Landegren et al.,1988; Korner and Livak, 1989; Newton et al., 1989; Wu etal., 1989; Barany, 1991; Dockhorn-Dworniczak et al., 1991;Kuppuswamy et al., 1991; Suzuki et al., 1991). Today theseassays are limited in their application only by the high costof DN A sequence determination. As DNA sequencing tech-nology becomes m ore cost efficient and automated, geneticassays may be based directly on DN A sequence analysis.Received September 3, 1992; accepted October 11, 1992.Copyright Clearance Center: 0032-0889/93/101/0349/04.

LITERATURE ClTEDAllard RW (1956) Formulas and tables to facilitate the calculationof recombination values in heredity. Hilgardia 2 4 235-278Arnh eim N, S trange C, Erlich H (1985) Use of pooled DNA samplesto detect linkage disequilibrium of polymorphic restriction frag-ments and human disease: studies of the HLA class I1 loci. ProcNatl Acad Sci USA 82: 6970-6974Arnold ML, Buckner CM, Robinson JJ (1991) Pollen-mediatedintrogression and hybrid speciation in Louisiana irises. Proc NatlAcad Sci USA 88: 1398-1402Barany F (1991) Genetic disease detection and DNA amplification

using cloned thermostable ligase. Proc Natl Acad Sci USA 88:Caetano-Anolls G, Bassam BJ, Gresshoff PM (1991a) High reso-lution DNA amplification fingerprinting using very short arbitraryoligonucleotide primers. Biotechnology9 53-557Caetano-Anolls G, Bassam BJ, Gresshoff PM (1991b) DNA am-plification fingerprinting: a strategy for genome analysis. PlantMo1 Biol Rep 4 294-307Carlson JE, Tulsieram LK, Glaubitz JC, Luk V, Kauffeldt C,Rutledge R (1991) Segregation of random amplified DNA markersin f l progeny of conifers. Theor Appl Genet 8 3 194-200Chaparro J, Wilcox P, Grattapaglia D, O M all ey D, M cCord S,Sederoff R, McIntyre L, Whetten R (1992) Genetic mapping ofpine using RAPD markers: construction of a 191 marker map anddevelopment of half-sib genetic analysis. l n Advances in Gene

Technology: Feeding the World in the 21st Century. Miami WinterSymposium, Miami, FLDockhorn-Dworniczak B, Dworniczak B, Brommelkamp L, BiillesJ, Horst J, Bocker WW (1991) Non-isotopic detection of singlestrand conformation polymorphism (PCR-SSCR):a rapid and sen-sitive technique in diagnosis of phenylketonuria. Nucleic AcidsRes 1 9 2500Giovannoni JJ, Wing RA, Gana1 MW, Tanksley SD (1991) Isolationof molecular markers from specific chromosomal intervals using

189-193

DNA pools from existing mapping populations. Nucleic Acids Reg19:6553-6558Hadrys H, Balick M, Schierwater B (1992) Applications of RAPDfingerprinting in molecular ecology. Mo1 Eco1 (in press)Hedrick P (1992) Shooting the RAPDs. Nature 355:679-680Hu J, Quiros CF (1991) Identification of broccoli and cauliflowercultivars with RAPD markers. Plant Cell Rep 1 0 05-511Innis MA, Gelfand DH, Sninsk y JJ, White TJ (1990) PCR Protocols,Ed 1. Academic Press, San Diego, p 482Klein-Lankhorst RM, Vermunt A, Weide R, Liharska T, Zabel P(1991) lsolation of molecular markers for tomato (Lycopersicon

esculentum) using random amplified polymorphic DNA (RAPD).Theor Appl Genet 83: 108-114Korner JS, Livak KJ (1989) Mutation detection using nucleotideanalogs that alter electrophoretic mobility. Nucleic Acids Res 17:7779-7783Krawetz SA (1989) The polymerase chain reaction: opportunities foragriculture. AgBiotech News lnfo 1: 897-902Kresovich S,William s JGK, McFerson JR, Routman EJ, Schaal BA(1992) Characterization of genetic identities and relationships ofBrassica oleracea L. via random amplified polymorphic DNA assay.Theor Appl Genet (in press)Kuppuswamy MN, Hoffman JW, Kasper CK, Spitzer SG, GroceSL, Baja SP (1991) Single nucleotide primer extension to detectgenetic disease: experimental application to hemophilia B (factorIX) and cystic fibrosis genes. Proc Natl Acad Sci USA 88:Landegren U, Kaiser R, Sanders J, Hood L (1988) A ligase-mediatedgene detection technique. Science 241: 1077-1080Lark KG, Evans J, Basha F, Bogden R, Copeland R, Ellison R,Horne D, Lee K, McDonald K, Pierson J, Schuster W, WilhelmP, Yu Y (1992) Molecular phylogeny as a genetic tool for soybeanbreeding. Soybean Genetic Newsletter (in press)Martin GB, William s JGK, Tank sley S D (1991) Rapid identificationof markers linked to Pseudomonas resistance pene in tomato bv

1143-1 147

using random primers and near isogenic lines. Troc Natl Acad SiiUSA 88: 2336-2340Michelmore RW, Paran I, Kesseli RV (1991) Identification of mark-ers linked to disease resistance genes by bulked segregant analysis:a rapid method to detect markers in specific genomic regions usingsegregating populations. Proc Natl Acad Sci USA 88: 9828-9832Newton CR, Graham A, Heptinstall LE, Powell SJ, Summers C,Kalsheker N, Smith JC, Markham AF (1989) Analysis of anypoint mutation in DNA. The amplification refractory mutationsystem (ARMS). Nucleic Acids Res 17: 2503-2515Paran I, Kesseli R, Michelmore R (1991) Identification of restriction-fragment-length-polymorphism nd random amplified polymor-phic DNA markers linked to downy mildew resistance genes inlettuce, using near-isogenic lines. Genome 34: 1021-1027Parks C, Chang, L-S, Shenk T (1991) A polymerase chain reactionmediated by a single primer: cloning of genomic sequences adja-cent to a serotonin receptor protein coding region. Nucleic AcidsRes 1 9 7155-7160Reiter RS, Williams J, Feldmann KA, Rafalski JA, Tingey SV,Scolnik PA (1992) Global and local genome mapping in Arabidop-si s thaliana by using recombinant inbred lines and random ampli-fied polymorphic DNAs. Proc Natl Acad Sci USA 8 9 1477-1481Stil es JI, Lemm e C, Sondur S, Morshidi MB, Manshardt R (1992)Using randomly amplified polymorphic DNA for evaluating ge-netic relationships: analysis of the relationship among Solo andother papaya cultivars. Theor Appl Genet (in press)Suzuki Y, Sekiya T, Hayashi K (1991) Allele-specific polymerasechain reaction: a method for amplification and sequence determi-nation of a single component among a mixture of sequence var-iants. Ana1 Biochem 192: 82-84Welsh J, McClelland M (1990) Fingerprinting genomes using PCRwith arbitrary primers. Nucleic Acids Res 18: 7213-7218Welsh J, Petersen C, McClelland M (1991) Polymorphisms gener-ated by arbitrarily primed PCR in the mouse: application to strainidentification and genetic mapping. Nucleic Acids Res 1 9 303-306Wilde J, Waugh R, Powell W (1992) Genetic fingerprinting ofTheobroma clones using randomly amplified polymorphic DNAmarkers. Theor Appl Genet 8 3 871-877Williams JGK, Kubelik AR, Livak KJ, Rafalski JA, Tingey SV(1990) DNA polymorphisms amplified by arbitrary primers areuseful as genetic markers. Nucleic Acids Res 18: 6531-6535Williams JGK, Rafalski JA, Tingey SV (1992) Genetic analysisusing RAPD markers. Methods Enzymol (in press)Wostemeyer J, Schafer C, Kellner M, Weisfeld M (1991) DNApolymorphisms detected by random primer dependent PCR as apowerful tool for molecular diagnostics of plant pathogenic fungi.ln Advances in Molecular Genetics. Hiithig-Verlag, HeidelbergWu DY, Ugozzoli L, Pal BK, Wallace RB (1989) Allele-specificenzymatic amplification of P-globin genomic DNA for diagnosisof sickle cell anemia. Proc Natl Acad Sci USA 8 6 2757-2760Young ND, Tanksley SD (1989) RFLP analysis of the size ofchromosomal segments retained around the T m -2 locus of tomatoduring backcross breeding. Theor Appl Genet 77:353-359

cenetic analysis with random amplified

Documents