Cornell Small Grains Lab Protocols
This manual was put together in the summer of 1993 to be used in a wheat mapping workshop sponsored by the International Triticeae Mapping Initiative (ITMI), to be held at Cornell University. Much of the material in the sections entitled Theoretical background is a modified version of a manual produced for the 1992 workshop offered by the Australian Triticeae Mapping Initiative. The experimental protocols and recipes in that manual have been replaced with those currently in use in the laboratory of Dr. Mark Sorrells of the Department of Plant Breeding at Cornell. The exercise in linkage analysis using RFLP data is modeled on one that appears in a "RFLP Training Course Laboratory Manual" written by G. Kochert, S. D. Tanksley, and J. P. Prince for a 1989 course at Cornell sponsored by the Rockefeller Foundation, and some of the references in the bibliography have also been taken from this source. Like the above sources, this manual may be freely copied and distributed. James C. Nelson
DNA ISOLATION FROM CEREALS
APPENDIX A. Recommended procedure for developing oat populations for mapping traits
APPENDIX B. Protocol for submitting oat genome mapping data
APPENDIX C. Estimating genetic distances between varieties
List of Figures
Parental survey in which DNA of one of the three lines is degraded 15
Part of 30-lane mapping film with nonuniform quantities of DNA 18
Two parental survey films showing low- and high-copy probes 19
Defective hybridization resulting from a bubble in Southern blot 23
Darkroom door was opened before film entered the XOmat. 30
Static discharge between film and acetate sheet-protector 30
Probe did not label 31
Hessian-fly resistance (B) population probed with G-48, filter 1a 36
B population probed with G-48, filter 1b 37
B population probed with BCD 131, filter 1a 37
B population probed with BCD 131, filter 1b 38
**************************************************************************
Aims of this workshop:
1. Introduce participants to theoretical aspects of DNA work.
2. Provide practical experience in many aspects of DNA technology.
3. Provide discussion periods on the current and potential uses of DNA technology in plant breeding.
Significance
The genetic mapping of most important crop plants is now progressing as a result of the application of molecular techniques. Both the language and the technology are evolving rapidly and many new cultivars, produced as a result of the application of the new technology, can be expected to become available within the next decade. To assess new molecular techniques, plant breeders need to be familiar with the limitations and the strengths of the processes involved.
Course summary
The course will introduce participants to the nature of DNA and restriction endonucleases, the isolation of DNA, bacterial transformation and plasmid preparation, and the analysis of DNA using radioactive probes and polymerase chain reaction. The participants will carry out computer analysis of data to produce genetic maps. Time will be set aside on each day of the course to discuss theoretical aspects of the work, the interpretation of data from molecular analyses, and the application of molecular markers to plant breeding.
WHAT IS GENETIC MAPPING? #WHAT IS GENETIC MAPPING?
Rationale and overview
By analogy with geographical mapping, the aim of genetic mapping is to establish a reference framework of landmarks on a body to be mapped. The more such markers we are able to identify, the better able we are to navigate in a genome. What can be gained by such navigation? Some actual or prospective benefits are:
1) Answering basic genetic questions: What is dominance? What is heterosis? What are genes and how do eukaryotic genes work? What is the nature of chromatin? What causes interference? Many more questions could be added to this list.
2) Use of molecular markers in selecting for desirable traits and trait combinations in plant and animal breeding. Use in selecting parents for crossing. Use in determining the nature of pathogen and stress resistance and in applying this knowledge to produce healthy crops and herds.
3) Answering questions on a larger scale about evolution and speciation. For example, we are now using RFLP maps to examine the DNA-level resemblances between cultivated grain crops: barley, wheat, oats, rice, maize, and sugar cane.
Genetic mapping dates back at least to the 1920s, when it was based on segregation analysis for phenotypic traits in such organisms as Drosophila and maize. In these and other organisms the density of maps was limited by the numbers of single-gene, readily scorable traits expressed in the organism whose genome was to be mapped, and by the supply of mutant lines and other genetic stocks. Beginning in the 1970s, isozymes increased the number of loci that could be placed on plant genomes. However, by the 1980s the discovery of restriction enzymes and the efficient adaptation of gel electrophoresis to separating small DNA fragments had rendered effectively limitless the number of markers available for mapping. While we will discuss mainly restriction fragment length polymorphisms (RFLPs), it should be understood that there are many other kinds of polymorphisms amenable to linkage analysis in segregating populations; examples of these in wheat might be RAPDs (randomly amplified polymorphic DNA), minisatellite DNA, ribosomal RNA, or storage proteins. It should also be realized that physical mapping by techniques such as FISH (fluorescence in situ hybridization) can determine the order of many markers on individual chromosomes, if not their genetic (recombination) distances. In addition, YAC (yeast artificial chromosome) libraries and PFGE (pulsed-field gel electrophoresis) separation of very large pieces of DNA are being used in many organisms to construct physical maps. Lastly, aneuploids and chromosomal stocks, such as the many available in wheat, our experimental model for this manual, allow assignment of markers to individual chromosomes.
The techniques described in this manual, then, represent only a model subset of the resources now available for genetic mapping in wheat or any other crop. They are not the last word, but only one fairly primitive word, which new methods should make obsolete within a very few years. One witness to this assertion is the continuing use of an expensive and hazardous radionuclide, 32P, to locate specific sequences of DNA in a genome, often at the cost of a week's waiting time for an autoradiogram to be made.
How to make a map
Below is a summary of the steps typically involved in obtaining and analyzing RFLP mapping data for wheat. It is assumed that the reader is familiar with RFLPs; if not, several of the references in the list at the end of this manual will be helpful.
1) Taking into account the purpose for which mapping is being done, choose parents for a mapping population and cross them.
2) If we are trying to obtain information about the location of genes controlling a specific trait (like disease resistance), we must have reason to believe that the progeny will differ in their expression of the trait. For straightforward RFLP mapping, the "trait" means the response to a given RFLP probe, and so the parents must be polymorphic (that is, show different banding patterns) when probed with this RFLP sequence. Therefore we must do parental surveys (with RFLP probes, described below) to identify a set of probes that will be useful for mapping.
2) Grow out progeny of the desired generation (henceforth referred to as "lines" although in fact we may use highly heterozygous F2 progeny, backcrosses, or other generations). Extract DNA from the leaves of seedlings 3-4 weeks old. After wheat plants are cut (typically a few cm from the base) they will usually grow back and produce seed for the next generation.
3) Clean up the DNA. Excesses of proteins, pigments, and other undesired chemicals are removed by routine chemical and physical manipulations.
4) Test the quality of the DNA to assure that it has not undergone much random shearing during the extraction. RFLP mapping requires DNA that has been cut reproducibly, i. e. at specific sites.
5. Having established the relative concentration of the DNA extracted from each of the lines, digest the DNA of all the lines with a restriction enzyme chosen on the basis of the parental surveys. Run the digests out on an agarose gel in order to separate the fragments spatially according to size (measured in base pairs, bp).
6. Transfer the separated DNA from the gel to a nylon&emdash;nitrocellulose membrane. Gels are fragile and cannot be preserved for repeated use in mapping, while the membrane, once the DNA has been immobilized on it, can be reused many times. During the transfer, the DNA is denatured so that the double strands are separated into single.
7. Using the probes identified as polymorphic on the parents of this population, probe the mapping filters we have made. A RFLP "probe" is a piece of DNA generally around 1-3 kb in length. In order to use it in mapping, we commonly: a) denature it, b) add a DNA polymerase, some short DNA primer sequences, and free nucleotides with some kind of incorporated label, and c) incubate this mixture so that labeled double-stranded (ds) DNA will be produced. The resulting dsDNA is again denatured into ssDNA and incubated in a hybridization buffer with the membrane that contains the plant genomic DNA.
8) After some hours, wash the buffer and the excess labeled probe off the membrane. The probe ssDNA will have bound to the immobilized ssDNA that is homologous to it (has a complementary sequence) but not to nonhomologous sequences. The binding site must now be identified. If the label that we used was cytidine triphosphate, made with the b-emitting phosphorus isotope 32P, we now expose the membrane to photographic film. The emulsion will darken at the sites next to the label. Commonly a fluorescing screen is also placed on the other side of the film from the membrane. The b particles excite the fluor on the screen and the resulting photons are detected by the film.
Other labels, principally fluorescent or chemiluminescent ones, are frequently used.
9. Examine the mapping film and score each genotype for each polymorphism. By "score" is meant to determine which parent the genotype resembles. Additionally it must be determined whether a polymorphic band is dominant or codominant, and whether the genotype in question is heterozygous or homozygous.
10. Format the resulting data for entry into a statistical-analysis program such as Mapmaker in order to obtain a linkage map judged to represent the most likely order of the RFLP loci on the genome.
The DNA molecule
Cells have been studied for several hundred years and speciÞc nuclear structures were described as early as 1781. By the start of the twentieth century chromosomes had been observed to appear at certain stages of the cell cycle, often with distinctive morphologies. The constancy of their appearance and behavior in the cell cycle suggested they were structures of basic importance to cell biology.
Chromosomes established as the carriers of genes. During the early part of this century (1900&endash;1930) numerous studies were carried out on the chromosomes of plants, animals and insects and demonstrated that aneuploidy (loss of chromosomes), polyploidy (extra sets of chromosomes), translocations (the rearrangement of chromosome sections), inversions (the end-for-end swapping of a chromosome segment within the same chromosome), and deÞciencies (absence of chromosome segments) could occur. As a result of these studies cytological changes in chromosome structure were correlated with changes in the appearance of the organism (its phenotype). This work reached its height in the study of the salivary gland chromosomes of Drosophila and provided exquisite detail about the structure of chromosomes. As a result it became clear that chromosomes must carry the information that determines the size, shape and behavior of organisms, i. e. the carriers of genes.
The chemical components of chromosomes. During the same period that chromosomes were shown to carry heredity material it was established that they were composed of two main chemical substances: nucleic acid and protein. Nucleic acids were known to be relatively simple chemical compounds, while proteins were more complex, and thus for many years proteins were considered to be the heredity material in chromosomes because it was felt that the chemical nature of genes must be complex in order to determine the phenotype of an organism. In the 1940's this concept was tested as the two components of chromosomes were puriÞed. The nucleic acid component proved to be an extremely long molecule (hence very viscous in solution) and composed of four basic building blocks called bases. The deoxyribonucleic acid (DNA) attracted the attention of microbiologists studying the genetics of simple lifeforms such as bacteria and it is their experiments that proved that DNA was the component of chromosomes which carried the genetic information.
DNA as the source of genetic information. Experiments that are now considered classic involved the study of Pneumococcus, a bacterium associated with bacterial pneumonia in humans. Typically, these bacteria grow as smooth colonies on nutrient medium (agar plates). Altered forms of the bacteria (mutants) were known which formed colonies with a rough appearance, and when the DNA from these rough forms was mixed with the food for bacteria which usually gave smooth colonies, some rough colonies appeared. The rough phenotype derived from the stock of smooth colonies did not change upon further growth, and thus the bacteria had undergone a stable genetic transformation. Experiments of this type, with both bacterial and viral lifeforms, established that DNA was the genetic material. There was much conceptual resistance to accepting a chemically simple compound such as DNA as the genetic material and it was not until the 1950's that it dissipated. Today it is possible to transform with pure DNA not only lower lifeforms but also plants and animals. Genetic engineering has introduced pieces of DNA carrying novel genes into such animals as sheep and mice as well as into plants such as maize, tomato, alfalfa, and potato.
The physicochemical structure of DNA explains how genetic material is encoded in DNA. Biochemists had long accepted that the composition of DNA followed the rule that fraction of the base A approximately equaled the fraction of the base T. Similarly the proportion of the base G equaled the proportion of the base C. In the 1950s Watson and Crick took this chemical observation together with the detailed X-ray structure of puriÞed DNA to put forward the structure of DNA, thereby revolutionizing thinking about the mechanism of heredity.
The chemical composition of DNA
A nucleotide is the basic link in the DNA polymer and consists of a pentose sugar, deoxyribose, attached to a phosphate group and one of the four bases adenine (A), guanine (G), cytosine (C), and thymine (T). The alternating sugars and phosphates form the backbone of a single strand, while the complementary bases on either strand interact to hold two strands together in a helix. The double-stranded DNA molecule can be drawn thus, if we imagine the helix to be unwound and þattened out:
phosphate phosphate phosphate phosphate phosphate
sugar sugar sugar sugar sugar sugar
T A T C C G
A T A G G C
sugar sugar sugar sugar sugar sugar
phosphate phosphate phosphate phosphate phosphate
The process whereby biological traits are passed on from one generation to another requires that the active genetic material satisfy certain observable features of heredity in all living organisms. Some of these features include: accurate self-replication during the growth and division of cells, the potential of the genetic material to carry all the primary or essential biological information, and a structure sufÞciently stable to allow only rarely the occurrence of heritable changes (mutations) in the offspring of living organisms.
The signiÞcance of the discovery of the double-stranded structure of DNA was that it provided a focal point to examine the general patterns of the mechanism of heredity. The composition of bases along one strand of the DNA chain is exactly complementary to that on its partner strand, in essence allowing both strands to carry the same genetic information. The double-helical structure of DNA is held together by weak chemical bonds (hydrogen bonds) between pairs of speciÞc bases on the two strands. The complementarity of the bases along the intertwined strands suggested that either strand could serve as a template upon which the opposite strand is reproduced. In this way genetic information can be copied through mechanisms that cause the separation of strands on one hand, and yet allow each strand to serve as a template for reproducing new complementary strands during cell growth and division.
The speciÞcity of base-pairing in DNA implied that the biological/genetic information was carried in the linear arrangement of bases along the strand. The biological or genetic information can be envisaged as sentences written using the bases A, C, T, and G (bonded to sugar and phosphate compounds to form nucleotides) as alphabet letters or building blocks in forming a large DNA molecule. Thus for a given DNA molecule of n bases long there are 4n possible combinations of linear arrangements of nucleotides. The number of base pairs in living organisms varies considerably; for example, the estimated haploid DNA content in bacteria (E. coli ) is 4 ¥ 106 , in the fruit þy (Drosophila ) 1.2 ¥ 109, and in mammals 3 ¥ 109. The number of possible arrangements of bases in each of these organisms gives the potential variation that can occur in the species. The amount of biological variation that is feasible is very large and is contained in approximately one meter of DNA which is packaged up in chromosomes.
The separation of the double strands of DNA and subsequent reproduction of complementary strands is well orchestrated with various proteins to ensure accuracy in DNA replication. This highly regulated system ensures speciÞcity in the base pairing, the formation of chemical bonds linking new nucleotides to the growing strand, and the elimination of incorrectly placed nucleotides. Occasionally, mistakes in replication do occur, such as misplaced base pairing or the insertion or deletion of bases in the DNA strand. These alterations in the DNA sequence or genetic mistakes are called mutations. These changes can be passed along in subsequent cycles of DNA replication and on to future generations. The magnitude of these changes in altering the phenotype of an organism is inþuenced by the position of the mutation in the DNA sequence. A single base change in an essential region can have a major effect, whereas in a nonessential region (50&endash;90% of the DNA in many organisms) the mutation may remain unnoticed. The latter is not true in the case of RþP mapping, where changes in the DNA can be useful regardless of whether they are in a segment of the DNA coding for a visible phenotype or in a DNA segment with no apparent function.
Isolation of DNA
The problems in isolating DNA, especially from plants, are the presence of DNAse activities that degrade the DNA and the presence of other macromolecules that copurify with or polymerize to the DNA during the isolation procedure.
The nuclease problem is reduced by removal of cations such as Mg++ that are required for nuclease activity. Agents such as EDTA, EGTA, and phenanthroline have been used at a range of concentrations, in different protocols, depending on the plant or animal species being analyzed. Also detergents such as sodium dodecylsulphate (SDS) are often used to inhibit enzyme activities.
Plant researchers often encounter undesirable macromolecules other than DNA that create problems in the DNA isolation procedure. Problems that arise from the presence of phenolic compounds can be reduced by the addition of 1% (or more) polyvinylpyrrolidine (PVP) in the initial isolation buffer. Detergents such as SDS dissociate proteins from DNA and make them more accessible to degradation by proteinases used in the DNA isolation. A reagent used in several procedures is cetylmethylammonium bromide (CTAB) which binds strongly to DNA, displacing protein and preventing degradation. The CTAB itself is removed by chloroform extractions, leaving DNA in the aqueous phase to be ethanol-precipitated. In some plant species it has been found necessary to isolate a crude nuclear fraction, in the presence of PVP, to remove a large proportion of the cytoplasmic material before proceeding with the CTAB procedure.
Cetylmethylammonium bromide (CTAB) is CH3(CH2)15N+(CH3)3 Br-
Sodium dodecyl sulphate (SDS) is CH3(CH2)11 OSO3- Na+
Sarkosyl is CH3(CH2)11N(CH3)CH2COO- Na+
Restriction endonucleases
Restriction endonucleases occur in nature as part of a mechanism in bacteria to degrade foreign DNA. The proteins bind speciÞcally to double-stranded DNA and then cleave it. The vast majority of type II restriction endonucleases recognize speciÞc sequences that are four, Þve, six, or eight nucleotides in length and display twofold symmetry :
EcoRI (from Escherichia coli) cleaves GAATTC
PstI (from Providencia stuartii) cleaves CTGCAG
TaqI (from Thermus aquaticus) cleaves TCGA
The binding of the enzyme to DNA is very intimate to facilitate cleavage, and after this has occurred the enzyme dissociates from the DNA molecule. Some enzymes cleave exactly at the axis of symmetry, generating fragments of DNA that carry blunt ends. Other enzymes cleave each strand at similar locations on opposite sides of the axis of symmetry, creating fragments of DNA that carry protruding single-stranded termini.
When plant or animal DNA is digested with restriction endonucleases, relatively little substructure is revealed when the DNA fragments are separated by electrophoresis in an agarose gel. Staining with a þuorescent dye and observing under UV light reveals the array of fragments as a continuous smear of light along a gel lane, with some exceptions due to highly repeated sequences. If, however, the array of fragments is transferred to nitrocellulose or nylon membranes and assayed for speciÞc sequences, a limited number of size classes of fragments are visualized and these provide a glimpse into DNA structure surrounding the sequence that was used as a probe. Most importantly from the point of view of using this technology for genetic mapping, it has been shown empirically that individuals of a species are not identical with respect to the distance (in bases of DNA) between restriction endonuclease cleavage sites that þank the speciÞc sequence used in the assay. In this way a single probe may reveal fragment patterns that differ between different individuals of a species. These differences are what we refer to as polymorphisms, and the fragment patterns, being heritable, constitute artiÞcial genes (artiÞcial in the sense of having no phenotypic expression). Thus when these polymorphic individuals are crossed they provide genetic markers analogous to those provided by isozymes.
Cloning DNA in bacterial plasmids for probes and sequencing
Plasmids are naturally occurring, circular, DNA molecules that in bacteria often carry genes for resistance to antibiotics or heavy metals. The plasmids found in E. coli have been extensively "engineered" to produce cloning vehicles. The vectors currently used most widely are pUC118 and pUC119. The molecules carry the gene conferring resistance to ampicillin, a short "polycloning-site region" in the gene for the b-galactosidase enzyme, and the sequences required in cis for the initiation and termination of bacteriophage M13 DNA synthesis and for packaging into bacteriophage particles. When cells harboring these plasmids are infected with a suitable Þlamentous bacteriophage, copies of one strand of the plasmid DNA are synthesized and packaged into progeny bacteriophage particles. Single-strand DNA can then be isolated from the bacteriophage particles and used as a template to determine the nucleotide sequence of the foreign DNA.
Cloning DNA segments into a plasmid vector such as pUC19 involves the following steps:
1. Digest DNA from plant or animal sample with (for example) the restriction endonuclease Pst1. This enzyme is often used to recover clones from the less repetitive component of the eukaryote DNA (in plants, up to 90% of the DNA can be in a class not coding directly for gene products). The tendency to exclude repetitive sequences from cloning is observed empirically and is believed to be due to increased levels of DNA methylation observed in DNA that does not code for gene products (e. g. many repetitive DNA sequences). The net result is very useful because DNA segments that are cloned using Pst are more likely to generate probes for RþP analysis. The digested DNA is fractionated on a glycerol gradient (10&endash;40%) to separate three broad size classes, 0&endash;2 kb, 2&endash;10 kb, and greater than 10 kb in size. The ends of all the molecules are the same:
5' G CTGCA 3'
3' ACGTC G 5'
This results from the way Pst creates breakages in the DNA at its speciÞc recognition site.
2. The plasmid vector is similarly digested with Pst to produce a linear molecule with ends that are identical to those of the digested plant DNA. Mixing the two digested DNA samples in equimolar proportions then allows the ends to be ligated together:
5' CTGCAOH PG CTGCA 3'
3' GP OHACGTC GP OHACGTC 5'
The enzyme used to carry out the reaction of the OH group with the P group to reform the phosphate&endash;sugar backbone is bacteriophage T4 DNA ligase, in the presence of Mg++ and ATP.
3. The ligation mixture is then used to transform bacteria to recover plasmids carrying inserted DNA.
Nucleic acid hybridization
In the duplex state of DNA, the two strands are held together by H-bonds primarily through complementary base pairing. When the duplex molecule is subjected to conditions such as high temperature or alkali treatment, denaturation of the double strands results in partial or complete separation into single-stranded molecules. Under favorable conditions the single strands can be reannealed into a duplex state. Renaturation of nucleic acids is not limited to single strands of the same DNA molecule, but can occur between DNA strands from different genomes containing similar sequences. The formation of DNA:RNA hybrids is also possible. The incorporation of radioisotopes such as 32P&endash;CTP into a puriÞed nucleic acid sequence to make "probes" allows the detection of related nucleic acid sequences in similarly or distantly related organisms. In Þlter hybridizations the probe is usually in solution while the array of single-stranded DNA molecules to be probed is immobilized on a solid support such as nitrocellulose or nylon Þlters.
To ensure maximum detection of nucleic acid hybrids it is essential to optimize the factors affecting the reaction kinetics and maintenance of stable duplexes or hybrids. A useful measure of the stability of DNA duplexes or DNA:RNA hybrids is their melting temperature (Tm), the temperature at which 50% of the nucleic acids remain dissociated or denatured.
Hybridization in aqueous solutions is carried out usually at temperatures of 65&endash;68°C at which the stability of probe:nucleic acid hybrids is maintained at about 20&endash;25°C below their Tm. The addition of compounds such as formamide decreases the Tm of nucleic acid hybrids. Hybridization solutions containing 50% formamide allow a lower incubation temperature, usually 42°C which is less harsh on Þlters, and the probes are relatively more stable at lower temperatures. Concentrations of 80% formamide reduce the rate constant for hybridization in solution by a factor of about three for DNA:DNA duplexes and by a factor of twelve for DNA:RNA hybrids. Although the reaction kinetics in Þlter hybridizations may be slightly different, similar qualitative results are likely to occur.
Hybridization of nucleic acids occurs slowly at low ionic strengths, and increasing the ionic concentration to about 1.5M Na+ accelerates the reaction. The increased reaction rate is much greater at the lower range of the Na+ concentration (up to 0.1 M ). The stability of mismatched duplexes, for example in cross-hybridization of probes to different species, is maintained at high salt concentrations (e. g. 6¥ SSC; 1¥ SSC = 0.15 M NaCl, 0.015 M Na3C6H5O7O . 2H2O).
In hybridization solutions containing formamide a greater buffering capacity is obtained with 6¥ SSPE (NaCl, NaH2PO4, EDTA).
The inclusion of inert polymers such as polyethylene glycol or dextran sulfate leads to an increased rate of hybridization. The presence of 10% dextran sulphate in the hybridization buffer confers about a tenfold increase in the reaction rate. This effect is thought to be associated with an increase in the effective concentration of the probe solution due to its exclusion from the volume occupied by the polymer. This observation arises from the network of reassociated probes (concatenations) which through their single-stranded portions anneal to the immobilized nucleic acids on the Þlter. resulting in an overestimation of the extent of hybridization. The inclusion of these inert polymers increases the viscosity of the hybridization buffer (creating difÞculties in handling) and can lead to high backgrounds.
Regions of nonspeciÞc attachment of probes to the surface of Þlters do occur, and to eliminate this effect blocking agents are included in the prehybridization and hybridization steps. The most common blocking agent is Denhardt's reagent (which contains BSA, PVP, Þcoll 400); nonfat dried milk also serves a similar purpose. These agents are often used in combination with sheared or sonicated salmon-sperm or calf-thymus DNA. When nylon Þlters are used in hybridization it is recommended that blocking agents be omitted from the hybridization buffer since high concentrations of protein hinder the annealing of the probe to its target. The principles that apply to hybridizing labeled probes to DNA bound to membrane Þlters also apply to in situ hybridization where the DNA is present in cytological preparations of chromosomes or nuclei.
DNA sequencing
The most commonly used sequencing procedure involves the synthesis of a DNA strand (using the DNA of interest as a template) for further chemical analysis. The enzyme DNA polymerase (from either E. coli or T. aquaticus) carries out this function in the following way:
5' pCpCpGpApGOH 3'
3' pGpGpCpTpCpTpApTpCpGpAp 5'
Mg++, dATP, dTTP, dCTP, dGTP
DNA polymerase
5' pCpCpGpApGpApTpApGpCpTp 3'
3' pGpGpCpTpCpTpApTpCpGpAp 5'
The reaction catalyzes the covalent linking of the 3'&endash;OH on the chain being synthesized with the 5'&endash;phosphate (P) group on the incoming nucleoside triphosphate. If the incoming nucleoside triphosphate does not have a 3'&endash;OH (so-called dideoxy nucleoside triphosphate) on its sugar residue, the chain cannot be lengthened further after its incorporation. This circumstance forms the basis for the DNA-sequencing reaction. If the above DNA synthesis is carried out in the presence of, for example, a mixture of PAOH (dA) and PA (dideoxy, ddA), a mixture of DNA fragments is obtained:
5'. pCpCpGpApGpA 3'
3' pGpGpCpTpCpTpApTpCpGpAp 5'
5' pCpCpGpApGpApTpA 3'
3' pGpGpCpTpCpTpApTpCpGpAp 5'
The growing chain will occasionally terminate after incorporating a PA residue rather than a PAOH residue and if the growing chains are electrophoresed on a high-resolution polyacrylamide gel, a series of fragments of different size are obtained (depending on the position of the A's in the chain). A similar experiment can be carried out separately with each of the other triphosphates, and the electrophoresis of all the fragments next to each other will allow the sequence to be deduced. In the automated sequencer, the chain that is growing in length as a result of the synthetic activity of the DNA polymerase is tagged with a þuorescent marker. Four different tags are available for each of the reactions possible (i. e. DNA synthesized with PAOH/PA, PGOH, PCOH, PTOH or PAOH, PGOH/PG, PCOH, PTOH etc.).
The polymerase chain reaction (PCR)
The PCR reaction, described Þrst in 1985, is revolutionizing the way molecular biology is being carried out. The procedure enables small amounts of speciÞc DNA fragments (which may be mixed with large amounts of contaminating DNA) to be ampliÞed approximately a millionfold. PCR is an in vitro procedure for the enzymatic synthesis of DNA, using two oligonucleotide primers that hybridize to opposite strands and þank the region of interest in the target DNA.
A repetitive series of cycles involving template denaturation, primer annealing, and the extension of the annealed primers with DNA polymerase results in the exponential accumulation of a speciÞc fragment whose termini are deÞned by the 5' ends of the primers. The ampliÞcation is dramatic because the extension products of one cycle serve as templates for the subsequent reactions and thus the number of target copies doubles at every cycle. With 20 cycles of PCR a 106-fold ampliÞcation (220) is achieved.
The key factor in the widespread use of PCR was the introduction, in 1986, of the thermostable DNA polymerase (Taq polymerase) from Thermus aquaticus. Having this enzyme in commercial supply (Cetus) meant that the reaction components (template, primers, Taq polymerase, nucleoside triphosphates, and buffer) could be simply mixed and subjected to temperature cycling.
For RþP studies, PCR technology can replace a number of steps in the routine procedure. For example, in the production of a probe, the use of primers on either side of the cloning site of the vector can be used to produce large amounts of the probe for labeling. Inclusion of 32P or digoxigenin-labeled nucleoside triphosphates in the reaction leads to highly labeled probes. Perhaps the most signiÞcant way in which RFLP work will change will be in amplifying speciÞc sequences from the genomic DNA and analyzing them by ethidium bromide staining after agarose gel electrophoresis. If the fragment length is polymorphic the genetic typing can be performed without blotting or hybridization procedures, thus greatly speeding up the analyses. It seems likely that future typing of loci in plant breeding programs will be carried out in this way.
Nonradioactive procedures for detecting DNA sequences.
Although probes labeled with 32P are at present the basis for RþP work, they have two major disadvantages. First, the b-particles emitted by 32P are energetic and thus represent a potential health risk for personnel involved in the daily work. This requires careful screening to guard against radioactive contamination. Second, the short half-life of 32P (14 days) means that in almost all cases, probes cannot be stored and reused but must be labeled each time a hy-bridization is to be performed. This is especially so with single-copy probes, which give a signal close to the limit of detection with 32P.
Consequently, there has been considerable work aimed at developing nonradioactive probes and detection systems. They all rely on:
1. Incorporation of chemically modiÞed nucleoside triphosphates into the probe.
2. A system for identifying the chemically modiÞed probe (e.g. an antibody to the chemical).
3. A method for visualizing the identiÞed chemical group (e.g. an easily detected enzyme coupled to the antibody). ELISA technology has provided many of the methods used.
An early system involved using biotin-labeled nucleoside triphosphates in the labeling reaction, detecting the biotinylated probe with streptavidin (or anti-biotin antibodies) coupled with peroxidase or alkaline phosphatase, followed by histochemical staining for the enzymes. Systems such as this have been quite successful but have had two major drawbacks. Þrstly, they have never been quite sensitive enough for routine use with single-copy probes in complex genomes. Secondly, the histochemical detection systems rely on producing an insoluble colored precipitate on the membrane that is extremely difÞcult to remove, making it virtually impossible to strip and reprobe the membrane.
A recently developed system appears, however, to overcome some of the above difÞculties. It uses:
1. Nucleoside triphosphates labeled with digoxigenin (a steroid occurring only in digitalis).
2. Detection with an antibody&endash;enzyme conjugate (anti-digoxigenin alkaline phosphatase).
3. Detection by a chemiluminescent substrate AMPPD (lumi-phos) and exposure to X-ray or Polaroid Þlm.
This method appears to be sensitive enough for single-copy probes, with exposure time of 15&endash;60 minutes for human DNA studies. Because detection is via a Þlm and not on the membrane, stripping and reprobing are just as easy as with 32P.
DNA ISOLATION FROM CEREALS
Theoretical background
Plants contain three types of DNA: nuclear, mitochondrial and chloroplast DNA. Although elaborate methods exist for the isolation of each type of DNA, most experiments require only the rather simple preparation of total DNA. All DNA preparation methods involve the removal of the envelopes (cell wall and nuclear membrane) around the DNA, the separation of the DNA from all other cell components such as cell wall debris, proteins, lipids or RNA, and the maintenance of the integrity of the DNA during the procedure, i. e. the protection from nucleases and mechanical shearing. In the most common method, which is applicable to a wide range of plant material, cells are opened by grinding in liquid nitrogen. The low temperature also prevents nucleases from degrading the DNA. During the subsequent treatment with extraction buffer, sodium dodecyl sulfate (SDS) dissolves membranes and denatures proteins, while EDTA complexes Mg2+ ions, an essential cofactor of most nucleases. After an extraction with chloroform:isoamyl alcohol the aqueous phase contains DNA, RNA, polysaccharides, and some protein, while lipids are found in the organic phase and cell debris and most of the protein aggregate in the interphase. Ethanol precipitation concentrates the DNA and remaining macromolecules can, if desired, be separated on the basis of their different buoyant densities in CsCl&emdash;ethidium bromide equilibrium gradients. After removal of the ethidium bromide and CsCl, the DNA is ready to use.
Large-scale isolation of total DNA from cereals
1. Collect fresh tissue and place at &endash;20°C. It is convenient to quick-freeze tissue samples by placing them in a foam cooler and running liquid N2 over them. Mortar and pestle, paintbrush, an appropiate number of labeled 50-mL polypropylene tubes, and a plastic funnel should be chilled in liquid N2. Grind leaf tissue, adding liquid N2 as necessary, with the chilled mortar and pestle. Transfer the fine powder (15&endash;20 ml per tube) with the brush and funnel into the cold polypropylene tubes. Do not let the powder thaw. Cap loosely (in case some liquid N2 remains inside) and store at &endash;20°C.
2. Add sodium bisulfite to extraction buffer and adjust pH with NaOH to 7.8&endash;8.0. Heat extraction buffer to 65°C and add 20&endash;25 ml to each tube. Mix briefly with a spatula, then cap tubes.
3. Incubate at 65°C for 20&endash;30 min, inverting tubes every 5&endash;10 min.
4. Fume hood: Add chloroform:isoamyl alcohol (24:1) to about the 40-ml mark of tubes, cap, and shake vigorously to produce an emulsion. Gentle shaking on a shaker will not result in a good extraction. Do not overfill with this heavy organic solvent, as its weight tends to damage or even rupture the lower section of a tube during centrifugation.
5. Centrifuge 15 min at 2,800 rpm. Pour upper phase into a clean 50-ml tube through 2 layers of cheesecloth, or pipette off if interface doesn't look stable enough. It will usually be firm (and not follow the upper phase out of the tube) if at least 15 ml of tissue powder was extracted.
6. Add 2 volumes (fill tube near top) of cold (&endash;20°C) 95% EtOH, cap tubes, and invert several times to mix and precipitate the DNA. Place in &endash;20°C for 30&endash;60 min or overnight if desired.
7. Hook out DNA with a bent Pasteur pipette, place into cold 70% EtOH in 20-ml culture tubes, and invert a few times or leave on a lab shaker to wash out remaining color. Spin down at 1000&endash;2000 rpm in the centrifuge for 10&endash;15 min and pour off the EtOH.
8. If this EtOH is discolored, add fresh cold 70% EtOH. Finish by centrifuging, pouring off EtOH, inverting tubes a few min to let liquid drain out, then hooking out the DNA and blotting out excess 70% EtOH on a tissue.
9. Dissolve DNA in sterile TE (0.5 to 1 ml, depending on pellet size). Incubate in 65°C water bath until dissolved, with inversion or vortexing every 30 to 60 min or until dissolved (often several hours).
10. Centrifuge 10 min in microcentrifuge at 10,000 rpm to remove suspended debris. The supernatant may be pipetted off into clean labeled microcentrifuge tubes.
11. At the time of restriction-enzyme digestion, add 0.3% (v/v) of a 10 mg/ ml stock solution of RNAse to the dissolved DNA. For cleaner DNA (loss in quantity and time though), add RNAse separately, vortex, and incubate 30 min. at room temperature before proceeding to step 12.
To reprecipitate DNA (not necessary):
12. Add 1/10 volume 3 M sodium acetate and 2 vols 95% EtOH to reprecipitate DNA. Place in &endash;20°C for 30&endash;60 min.
13. Repeat steps 7 through 10.
• Extraction tubes may be aired out in the fume hood and then washed for multiple reuse.
DNA quantity/quality check (spectrophotometric)
Warning: spec is poor unless your DNA went through a cesium chloride gradient.
Spectrophotometer instructions
Prepare diluted DNA samples and quantify in the spec at 260 nm, 280 nm, and 310 nm.
Suggested starting dilution: 6.25 ml DNA + 493.75 ml TE (dilution factor = 80).
DNA conc. (mg/ml) = (Abs260 &endash; Abs310) ¥ 50 mg DNA/OD unit ¥ dilution factor.
Operation of spectrophotometer (Beckman DU&endash;7):
1. Turn on (Press "On/Idle" button)
2. Upon completion of self-test, press "UV" button, and wait for 20 min.
3. Press "multi" button and set desired wavelengths (260, 280, and 310).
4. Insert calibration sample (Pure sample of liquid used for dilution: distilled water or TE) into reading cell and press "start" button to start calibration process.
5. Upon completion of calibration, insert samples into reading cells and press "run" button.
6. When finished, press "On/Idle" button again to enter the idle mode.
Note: Wipe the faces of the sample containers carefully before inserting into the reading cells. Wash the containers under vacuum aspiration with a sequence of acid (1 N HCl) &endash; water (dH2O) &endash; alcohol (95% EtOH) washes.
Extraction buffer
Material For 1 liter Final conc.
5 M NaCl 100 ml 500 mM
1 M Tris&endash;HCl pH 8.0 100 ml 100 mM
0.25 M EDTA 200 ml 50 mM
20% SDS 62.5 ml 0.84% (w/v)
Fill to final volume with dH2O.
Note: Just before use, add 0.38 g sodium bisulfite /100 ml buffer and adjust pH to 7.8&endash;8.0 with NaOH.
TE buffer
Stock For 1 liter For 3 liters
1 M Tris, pH 8.0 10 ml 30 ml
0.25 M EDTA, pH 7.0 4 ml 12 ml
distilled water 986 ml 2958 ml
Autoclave before use.
1 M Tris pH 8.0
Use 121.1 g Tris for each liter of dH2O. Adjust pH to 8.0 with conc. HCl.
DNA DIGESTION WITH RESTRICTION ENZYMES
DNA quality-checking (on gel)
Since RFLP analysis depends on hybridization of a probe to a specific fragment of plant genomic DNA, generated by deliberate cutting with a chosen restriction enzyme, random shearing of DNA by mishandling during extraction or purification causes the problem illustrated in the figure below. Now the region of homology between a probe and the genomic DNA is shared by a multitude of fragments of varying molecular weights, migrating to varying distances on a gel and all hybridizing with the radiolabeled probe. We may not be able to see distinct bands or determine whether there is polymorphism between two genotypes. In the survey film here, the DNA extracted from the parent represented by the indicated lanes is degraded.
• •
Parental survey in which DNA of one of the three lines is degraded
Accordingly, before making mapping filters we check the quality of the genomic DNA by digesting a few ml of each sample and running it out next to an undigested bit of the same sample. Upon ethidium bromide (EtBr) staining and UV viewing, the lane of undigested good-quality DNA should show much less light than that of the digested sample; ideally all the DNA remains as a bright region near the origin. A photograph of a quality-check gel did not reproduce well enough on paper to be included here.
Quality-check digestion
1. Each partner: label 4 sets of 2 0.5-ml tubes, one for digested and one for undigested DNA of each of your 4 lines.
2. To each tube add:
ddw 8 ml for digested, 10 for undigested
DNA 4 ml for digested, 2 for undigested
RNAse 1.0 ml
For digested only:
spermidine (Spn) 2.0 ml
dithiothreitol (DTT) 2.0 ml
buffer C 2.0 ml
EcoRI* 2.0 ml
This is conveniently done with an enzyme cocktail to spare yourself pipetting. Calculate in advance of the lab the amounts of ingredients to add to it, including 5&endash;10% extra for insurance, and also the amount of cocktail to add to each digest. Calculate also how much you'll need to load into each well of the gel.
3. Cap tubes, vortex to mix, and incubate the digested sample at 37° for 5 h. The undigested, RNAse-treated sample may be left at RT. One hour is adequate time for RNA digestion.
4. Prepare a 0.9% agarose gel. When the digest is completed, add 10 ml of "blue juice" (stop dye) to each tube, load the gel with alternating lanes of digested and undigested DNA of the same line (for contrast), and run overnight.
5. Stain and photograph the gel as usual (see protocol below).
Spermidine 40 mM
10.184 mg/ml. 1000 mM = 254.6 mg/ml.
DTT 10 mM
30.8 mg in 20 ml H2O. Freeze in aliquots.
"Blue juice" (stop & loading dye)
Component Volume Final conc.
Glycerol 350 ml 70%
10¥ NEB 25 ml 0.5¥
0.25 M EDTA 40 ml 20 mM
dH2O 50 ml
20% SDS 5 ml 0.2%
2.5% Bromphenol blue 30 ml* 0.6 mg/ml
(25 mg/ml sodium salt B&endash;6131)
Total 500 ml
*More BPB may be added for a darker color.
Running gels (large)
1. For each gel prepare 250 ml of 1¥ NEB (Neutral Electrophoresis Buffer) solution by dilution from 10¥ stock NEB.
2. Put 2.25 g agarose (electrophoresis grade; not low-melting agarose) in a clean Erlenmeyer containing a stirring bar. Add the diluted 1¥ buffer. The final concentration of agarose will be 0.9%.
3. Prepare the gel mold to receive the gel solution by cleaning it and sealing both ends with tape. Press the tape firmly against the tray edges to assure a fluid-tight seal. Place the gel mold on a level surface. Select the combs to use. Bring the agarose solution to a boil in the microwave. Mix by swirling and re-boil for a few seconds. Cool the solution down to 50&endash;65°C with stirring. This goes much faster if done in a bath of ice and water.
4. Pour the gel and install the combs. Allow the agarose to gel completely, around 30 min.
5. Pour 2 L of 1¥ NEB buffer solution into the electrophoresis apparatus. Peel the tape from the mold ends to expose the edges of the solidified gel. Place the mold containing the gel in the apparatus by immersing one side first and then the whole mold, in order to avoid formation of bubbles under the mold. Remove the combs carefully when the mold is submerged.
7. Load samples into the appropriate wells.
8. Place the lid on top of the tank and connect power cords (DNA runs from negative (black) to positive (red)!). Turn power supply on and adjust voltage to level desired (e.g. 25&endash;30 volts for an overnight gel).
9. At the conclusion of the run, turn the power supply off, disconnect the power cords, and remove the mold containing the gel. Free the gel by sliding it onto a plexiglas plate in a washing tray. Half-fill with distilled water and add two drops of EtBr (Carcinogen. Wear rubber gloves), one in each opposite corner (or use a reusable stock EtBr solution for staining), and shake for 10 min. Pour off solution and wash gel for 15 min. in distilled water. Take picture of gel on the UV transilluminator.
Comments:
• A 40-well comb can hold 30 ml, others can hold 40 ml in a 250-ml gel.
Although economy suggests using as closely spaced lanes as possible, you may find that the 24-well gels give the best resolution for RFLP mapping. To obtain the strongest signal, we use 15&endash;20 mg of DNA per lane, and this quantity may result in misshapen and unevenly migrating bands in the narrow lanes of a 30-well gel. The film shown in the next figure may be compared with one of the 20-lane mapping films shown in the Data analysis section. Besides the difficulty of scoring the polymorphism (can you find it?), this film also shows that the lanes in the original gel were loaded with different amounts of DNA. An important use for the small preliminary digestion is to allow estimation of the relative concentrations of the DNA samples extracted from the mapping population. With this knowledge you can load similar quantities of DNA in all the lanes of a gel to be blotted for mapping, so that hybridization will give signal of the same strength from all lanes.
Part of 30-lane mapping film with nonuniform quantities of DNA
• If probing with lambda/HindIII (size standard), do not load more than 25 ng/lane (10 ml of 2.5 ng/ ml stock), or your lambda signal will overexpose in the time needed to detect the RFLP signal. To visualize with EtBr, however, add up to 10 times this.
• The running buffer can be reused for at least a week. To destroy the pH gradient after a run it is recommended to pour it out and back into the gel box.
Neutral Electrophoresis Buffer, 10¥
Component For 10 L For 20 L Final conc.
Tris (Sigma 7&endash;9) 1210 g 2420 g 1M
EDTA (disodium salt) 33.6 g 67.2 g 10 mM
NaAc . 3H2O 170.1 g 340.2 g 125 mM
Adjust to pH 8.1 with conc. acetic acid: for 10 L, 275&endash;350 ml; for 20 L, 600&endash;700 ml.
Ethidium bromide (EtBr)
Make a solution of 10 mg EtBr/ml dH2O. Always wear rubber gloves when handling ethidium bromide.
Parental surveys
In order to map with RFLPs, we must assemble a set of probes that reveal band polymorphisms (PMs) in the mapping population. Since the population will be segregating at a particular locus only if the two parents are polymorphic at that locus, we begin by screening the probes on the parents and keeping only those probes that show well-defined PMs.
We make parental survey filters (technically "membranes" but commonly called filters because made of the same materials used to make micropore filters) from digests of the two parents with an array of enzymes that experience has shown to reveal high levels of PM in wheat: EcoRI, EcoRV, HindIII, DraI, and XbaI. A set of mapping filters will be made with a single-enzyme digest of all the lines in the mapping population; thus we would make an EcoRI set to probe with RFLP probes that were polymorphic on the EcoRI-digested parents, an XbaI set for the XbaI PMs, and so on for the other enzymes. It often happens that more than one enzyme reveals a PM for a probe on the parents. In such a case we consider it likely that the two PMs represent the same locus and would cosegregate, so for mapping we choose only the enzyme for which the PM is easiest to score.
Below is shown a pair of parental survey films. In this case two populations were available for mapping, derived from the crosses M&endash;3/Opata85 and M&endash;6/Opata85. Hence PMs between the two parents of either population were sought. The legend shows the digestion and loading schemes used for the gels used to make the survey filters.
M-3 M-6 Opata
EcoRI EcoRV HindIII DraI
M-3 M-6 Opata
EcoRI EcoRV HindIII DraI
Two parental survey films showing low-(top) and high-copy probes
Often we encounter high-copy-number probes that show numerous bands, but the quality of the mapping filters or of the DNA extracted from the lines is not always high enough for the PMs to be easily scored. This should be evident from the figure above. The lower autoradiogram shows band differences between some of the lanes (as in the DraI digest, lightly circled on the film), but consistent scoring of these closely spaced differences will not be easy. The PMs between either pair of parents in the top film (in all digests but HindIII) will obviously be more straightforward to score. Sometimes, however, multiple PMs can be cleanly scored in a single probe and they can all be placed on a genetic map &emdash; not necessarily on the same chromosome &emdash; by linkage analysis.
For this lab, you and your partner will prepare one gel (two 20-lane half-gels), which will yield four identical 10-lane parental survey filters, each consisting of 2 parents digested with the 5 different restriction enzymes listed above.
Parental survey digestion
1. Label ten 0.5-ml tubes with the ten possible combinations of parent and enzyme. You will digest in each tube enough DNA to load 4 lanes. To each of the tubes add 60&endash;80 ml of the appropriate parental DNA and enough ddw to bring the total volume to 80 ml.
2. To make the enzyme cocktails, label five more tubes with the names of the five enzymes. Add to each tube:
spermidine (Spn)* 16 ml
dithiothreitol (DTT) 16 ml
restriction buffer** 16 ml
RNAse*** 8 ml
restriction enzyme 24 ml
These ingredients should be kept on ice. Except for the restriction enzyme, you may thaw them at room temperature.
Vortex to mix. Return the enzyme to the freezer as soon as you are finished with it.
* For the DraI cocktail, replace 12 ml of the Spn with ddw.
**Buffer A for HindIII and DraI, buffer C for the other three enzymes.
***RNAse is not strictly necessary for RFLP work. The RNA will all run down at the low-molecular-weight end of the gel, out of the way of any useful polymorphisms.
3. Pipette 40 ml of each cocktail into each of the 2 tubes of DNA labeled for it. If you were making cocktails for lots of digests, you would make 5&endash;10% extra to allow for inexactly calibrated pipettes, fluid remaining in tips, etc. Close tubes, vortex, and leave in 37°C incubator for about 5 h.
4. Make a 0.9% agarose gel, using the 24-well combs.
5. Add 40 ml of blue juice to each digest, vortex to mix, and load gel. The loading pattern for parents A and B should be:
EcoRI EcoRV HindIII DraI XbaI (repeat on right of top half of gel)
A B A B A B A B A B
(repeat on lower half of gel)
6. Run gels overnight, stain and photograph, and blot. We will later cut the filters so as to get four from each gel.
DNA TRANSFER TO NYLON MEMBRANES
Theoretical background
DNA hybridizations require the breakage of H&endash;bridges between the complementary strands of the DNA and therefore either high temperatures or the presence of denaturing chemicals like formamide. These requirements are not easily met in an agarose gel. Therefore, the DNA is transferred from the gel onto a synthetic membrane, in this case nylon coated with nitrocellulose. This membrane combines the physical strength of nylon membranes with the high resolution of nitrocellulose membranes. The banding pattern is preserved during the transfer. Since only single-stranded DNA binds to nitrocellulose, the DNA has to be denatured before transfer. With the system we will use, capillarity draws a NaOH solution through the gel, carrying the denatured DNA onto the surface of the adjacent membrane.
Hybond N+ blotting protocol
Treatment of gel
1. Depurinate in 0.25 N HCl on shaker for 10 min (no more) and rinse with dH2O. The dye line will turn yellow.
2. Neutralize a few minutes in 0.4 N NaOH until dye line is blue again.
3. Place gel on blotting apparatus.
• The neutralizing NaoH solution may be reused for that purpose.
2.5N HCl (10¥) stock
For 4 L: add 833 ml conc. HCl to 3167 ml dH2O.
Blotting apparatus
1. Wearing rubber gloves, Þll boxes with 0.4 N NaOH and saturate sponges by repeatedly pressing out bubbles and reÞlling with solution. Cut Hybond N+ to gel size, label on one side with an indelible pen, and presoak in 0.4 N NaOH.
2. Wet 2&endash;3 sheets of Whatman paper in 0.4 N NaOH and lay on sponges.
3. Cut off waste portion of gel and part the two half-gels with a razor blade, slide them onto the apparatus, and remove air bubbles by smoothing out with Þngers.
4. Place Hybond N+ membrane (labeled side up) on gel and remove air bubbles by rolling a plastic or glass pipette from middle to edges.
5. Wet 2&endash;3 sheets of Whatman paper in 0.4 N NaOH and lay on membranes, rolling again to remove bubbles. They result in regions of no DNA transfer to the membrane (see Þgure). Place a half-stack of cut newsprint or paper towels on top, lay a plexiglas gel support or tray on top of this, and weight with 0.5 kg per half-gel. Both halves of a large gel are conveniently blotted on the same apparatus, so a 1-kg weight such as a 1-L bottle of water is suitable.
6. Blot at least 3 hours. Overnight blotting is convenient but not necessary.
7. Remove membranes and rinse brieþy in 2¥ SSC, blotting excess between Whatman paper. Store moist, not wet, between acetate sheet-protectors in refrigerator. We do not UV-treat or bake the Þlters; this may even reduce the signal after hybridization.
Defective hybridization resulting from a bubble in Southern blot
Stripping
After exposure of hybridized Þlters, strip in a near-boiling (96&endash;100°C) solution of 0.5% SDS (25 ml of 20% SDS/liter dH2O) with gentle shaking for 1 min or more. Blot between Whatman paper and store as above.
Note: Stripping by this method is milder than stripping with NaOH, so the Þlters will last longer. However, high-copy probe will be more difÞcult to remove.
AMPLIFICATION AND ISOLATION OF RFLP PROBES
Theoretical background
We will not perform the ligation used in genomic library construction; rather, we will start from an available library and do the operations required to obtain enough copies of a probe sequence to use it in hybridization. An efÞcient system for maintaining a probe library involves sequences ligated to þanking PCR primer sites and inserted into plasmids. After transformation of competent E. coli cells with the plasmids, an overnight culture is extracted to give a "miniprep," which can be stored indeÞnitely at &endash;80°C. From this stock small aliquots can be withdrawn as desired and the probe sequence ampliÞed with PCR. When the miniprep stock runs low, another transformation is performed to amplify it once again.
Whole plasmids can be used as probes, provided the insert is the only sequence that hybridizes to the DNA being probed. If this is not the case, the plasmids are digested with the proper enzyme, the digest is run out on a gel, and the probe sequence is cut out and extracted from the gel. The PCR method is clearly an easier way to get the probe sequence in isolation.
Transformation
Note: The DH5a competent cells should be stored in the &endash;80°C freezer and not be refrozen. To avoid refreezing, make 20- and 100-ml aliquots and freeze them in dry ice/EtOH bath or in EtOH from the &endash;80°C freezer.
The term "ligation" in the following protocol refers to plasmids, whether constructed in the same lab (as by genomic library construction) or received from elsewhere.
Large reaction:
1. Warm ampicillin plates in 37°C incubator, thaw X&endash;Gal and IPTG.
2 . Dilute ligation with TE so that 1 ml contains 2.5 ng DNA (a 1:1 dilution of the 100 ng/20 ml ligation).
3. In 15-ml sterile Falcon culture tubes (or 1.5-ml Eppendorf tubes), add 1 ml of dilute ligate to 100 ml of DH5a competent cells.
4. Shake gently 5 sec.
5. Put on ice for 30 min.
6. Heat-pulse 90&endash;120 sec in 42°C water bath Do not disturb.
7. Place on ice 2 min.
8. Add 900 ml SOC (LB also works) and shake for 1 h at 225 rpm and 37°C.
9. During incubation, prepare XGal/IPTG stock solution containing:
500 ml XGal (stock 3% in N,N&endash;dimethyl formamide)
100 ml IPTG (100 mM).
10. Spread 60 ml stock/plate.
11. Plate out 80&endash;125 ml of transformed cells/plate.
Small reaction
1. Dilute ligation so that 1 ml contains 1 ng DNA (100 ng/20 ml ligation diluted 1:4)
2. In sterile 1.5 ml microfuge tubes, add 1 ml of dilute ligate to 20 ml of DH5 alpha competent cells.
3. Follow all other steps as described above except add only 80 ml of SOC.
SOC for transformation
Material Amount per 100 ml Final conc.
Bactotryptone 2 g 2%
Yeast extract 0.5 g 0.5%
NaCl 1 ml 1 M NaCl 10 mM
KCl 0.25 ml 1 M KCl 2.5 mM
MgCl2, MgSO4 1 ml 2 M Mg stock 10 mM in each
glucose 1 ml 2 M glucose 20 mM
dH2O to 100 ml
1. Bactotryptone, yeast extract, NaCl and KCl are added to 97 ml of dH2O, allowed to dissolve, and then autoclaved.
2. The medium is cooled to room temperature and then brought to 20 mM Mg++ with a 2 M Mg++ stock solution (1 M MgCl2 . 6H2O + 1 M MgSO4 . 7H2O, Þlter-sterilized).
3. A 2 M glucose stock (Þlter-sterilized) is used to make the medium 20 mM in glucose. The complete medium is then Þltered through a 0.2-mm Þlter unit. The pH should be 7.0 ± 0.1.
Source: BRl protocol: Library EfÞciency™ DH5 Alpha Competent Cells
Miniprep
1. Grow 6-ml cultures overnight with antibiotic in Falcon tubes at 37°C and shaking.
2. Take 1 ml and place in 2 ml cryovial along with 0.440 ml sterilized 50% glycerol, and store at &endash;80°C.
3. Spin the remaining 5 ml in a Beckman TJ&endash;6 at setting 8 for 5 minutes. Pour off supernatant and slightly dry pellet by inverting tubes on paper towels. (Pellets can now be frozen).
4. Resuspend pellet in 500 ml of freshly made LiCl buffer. Transfer to 1.5-ml Eppendorf tubes. Quickly add 50 ml fresh lysozyme solution (10 mg/ml dH2O) Incubate for 5&endash;7 min at room temperature.
Note: Keep lysozyme on ice.
5. Boil in water bath or heating block for 90 sec. Cool on ice for 5 min. Spin at 10,000 rpm for 15 min.
6. Immediately after the spin, remove (and discard) the pellet by stabbing it with a pipette and sliding it out of the tube. Spin at 10,000 rpm for 5&endash;8 min and transfer the supernatant to a fresh tube.
7. Add 2 volumes of cold 95% EtOH and freeze at &endash;80°C for 1 hour or more (or overnight at &endash;20°C).
8. Spin at 10,000 rpm, 4°C (or 25°C) for 15 min. Pour off supernatant. Add 1 ml 70% EtOH and spin for 5 min at 10,000 rpm, 4°C (or 25°C). Pour off EtOH and dry pellets slightly by inverting.
9. Resuspend in 200 ml TE. Vortex. Add 2 ml RNAse. "Hydrate" for about 30 min at 65°C. Spin for 8 min at 10,000 rpm and transfer the supernatant to a new tube.
Ampicillin
Dissolve 750 mg amp sodium salt in 10 ml sterile H2O. Filter-sterilize and aliquot 500 ml/microfuge tube. Store at &endash;20°C.
LB + Ampicillin Medium (for 1 L; halve quantities for 500 ml)
Bacto&endash;tryptone 10 g
Yeast extract 5 g
NaCl 10 g
For solid media: add 15 g of Bacto Agar/liter of solution.
1. Autoclave (30 min sterilization + 10 min liquid cooling).
2. 50 mg/ml ampicillin (or 1 ml stock amp/liter) when cool
(1 ml of prepared dilution of amp/liter of LB = 1/1000 dil. factor).
IPTG (100 mM)
Add 238 mg IPTG to 10 ml H2O, mix, aliquot into Eppendorf tubes, wrap in foil, freeze at &endash;20°C.
X&endash;Gal (3%)
Add 0.75 g (3 g) X&endash;Gal to 25 ml (100 ml) N,N,&endash;dimethylformamide. Store at &endash;20°C.
LiCl Buffer
Conc. Amount per liter
50 mM Tris 50 ml 1 M Tris pH 7.5
62.5 mM EDTA 2.33 g EDTA
0.4% Triton X&endash;100 0.4 ml Triton X&endash;100
2.5 M LiCl 10.6 g LiCl
Þll to 1000 ml ddw, pH to 7.5
• calibrate pH meter with pH 10.0 and pH 7.0 standards
• pH by adding NaOH 2.5 N; be aware of the buffer effect (EDTA)
HYBRIDIZATION OF DNA PROBES TO DNA BOUND TO MEMBRANES
Theoretical background
Hybridization between DNA bound to nitrocellulose and a radioactive DNA probe in solution is a function of temperature, time, salt concentrations, GC content, probe length, and hybridization volume. For DNA with a GC content of approx. 50% generally a temperature of 65°C is chosen, which is about 20°C below the actual melting temperature of double-stranded DNA at the standard concentration of 0.6 M NaCl. Under these stringent conditions, probes that are between 200 and 400 bp in length and have a very high homology to the blotted DNA (being in most cases from the same or a closely related organism), hybridize with an efÞciency of >90% within 8 h, while less homologous probes hybridize to a much lesser degree. Lower temperatures, longer hybridization times and higher salt concentrations allow more cross-hybridization to heterologous probes. The hybridization volume should be kept to a minimum to ensure maximum encounter rates between the probe and the bound DNA; at the same time this should not be too small or there will not be sufÞcient movement of the probe along the membrane. Therefore, inert polymers like dextran sulfate are sometimes added, since they have a large hydrated volume and take up most of the space in the hybridization solution. Before the hybridization, potential binding sites for the probe other than the homologous DNA have to be blocked to avoid background hybridization to the membrane. This is accomplished by prehybridization with hybridization buffer containing heat-denatured salmon-sperm DNA, which shows little homology with plant DNAs.
Hybridization
Prehybridization
1. Boil salmon-testicle DNA (stDNA) for at least 10 min. Put directly on ice to cool.
2. Heat hybridization buffer (HYB) to 65°C. Label hybridization boxes. Add a minimum of 50 ml HYB to large boxes, 30 to small boxes. If hybridizing more than one Þlter, add extra HYB. Add ST DNA to boxes at 1 ml/50 ml HYB.
3. Place Þlters in boxes one at a time, avoiding bubbles between them. Prehybridize at least 4 h if Þlters are new. If they have been previously used, 2 h is sufÞcient. Cover and incubate boxes at 65°C.
stDNA
Add 1 g salmon-sperm DNA to 200 ml dH2O. Stir at room temperature until DNA is in solution. Shear by sonication.
Hybridization buffer
Component Vol. to add for 4 L Final conc.
1 M Na2PO4 pH 7.2 2 L 0.5 M
Bovine serum albumin (BSA)
fraction 5 40 g in 600 ml dH2O 1%
20% SDS pH 7.0 1400 ml 7%
Total 4 L
Random hexamer labeling (random priming)
1. Label 0.5-ml Eppendorf tubes to match boxes. Add about 100 ng of probe DNA (typically 7&endash;14 ml) to each, making to 10 ml with ddw. Thaw 32P in the radioactive work area. Close tubes and heat probe DNA in block for 5&endash;10 min to denature. Place tubes in ice for about 5 min. If condensation forms on insides of tubes, it may be brieþy centrifuged down.
2. Add 11 ml LS (oligonucleotide mixture, thaw at RT) and 2 ml Klenow (large fragment of polymerase I, 1:5 dilution, keep on ice). Keep mixture on ice.
3. In 32P hood, add 5&endash;10 ml (depends on amount of HYB buffer and age of 32P) of 1:1 diluted 32P (make sure 32P is completely thawed). Incubate at 37°C for at least 1.5 h.
LS for random hexamer labeling
LS:25 parts HEPES:25 parts DTM:7 parts OL
How to make:
OL: Take 1 bottle of hexadeoxynucleotides (Pharmacia pd(N)6&endash;, #25.2166&endash;01, 50 Units)
Add 560 ml cold TE (Þnal conc. 90 U/ml)
HEPES: Ad 2.0 ml cold 1M HEPES, pH 6.6 (2.383 g HEPES/10.0 ml H2O, pH 6.6 with NaOH, store at 4°C)
DTM: Add 2 ml cold DTM:
1 M Tris&endash;HCl, pH 8.0 (= 250 mM) 485 ml
1 M MgCl2 (= 25 mM) 48.5 ml
b&endash;mercaptoethanol (= 50 mM) 6.8 ml
dH2O 1400 ml
old nick-translation dxTPs
(old dxTPs are 3 3mM
dATP, dTTP, dGTP) 30 ml
Store LS in 500-ml aliquots at &endash;80°C.
Klenow fragment
Dilute 20 ml of 6 U/ml Klenow with 80 ml of 50% glycerol.
No-column option
1. Add 25 ml of 0.4 N NaOH to probe to denature. Leave at RT 10 min. Add 150&endash;200k counts of l DNA to each probe if needed.
2. Add probes to boxes by lifting out all Þlters and mixing in probe. Replace Þlters one at a time (DNA side up) and remix to ensure good coverage of all Þlters. To reduce the risk of drying out, you may cover with acetate sheet if using large boxes. Hybridize at least 20 h at 65°C.
Column option (generally not necessary)
1. Cut off both ends of a 1-cc syringe: remove the narrower part of the tip and cut the top at 0.8 (need 1 column for each probe).
2. Make polyethylene Þlters for syringes with cork borer, and Þt snugly in the syringe; put syringe with Þlter in 1.5 ml Eppendorf tubes.
3. Fill syringe with G50&endash;80 Sephadex solution(1% SDS, 25 mM EDTA). Spin at 1000 rpm for 30&endash;60 sec. There should be 0.5&endash;0.8 cc of white Sephadex. If necessary, reÞll and repeat spin until most of liquid is out of the column. Transfer syringes to new labeled Eppendorfs and cover if not used.
4. Add 30&endash;35 ml of buffer (1% SDS, 25 mM EDTA) to probes to stop the reactions. Load into tops of columns. Add an additional 100 ml buffer to columns. Spin at 1000 rpm for 60 sec. Discard column in radiation waste.
5. Add 100 ml buffer to dilute the probe. Take out 2 ml (1:100), put into scintillation vial half full of water, and count (scintillation counter PRG 10: 15% efÞciency with 32P). Continue with steps 2 and 3 from no-column option.
Washing
1. Prepare proper stringency of SSC. Prepare 1 liter for each 20 Þlters. From 20¥ stock SSC, the 2¥, 1¥, and 0.5¥ solutions require respectively 100, 50, and 25 ml SSC per liter. Add 5 ml of 20% SDS to each liter of wash solution. Temperature and salt concentration of the third (0.5¥) wash are critical. Temperature should be exactly 65°C (±1°C).
2. Heat the 2¥ wash to 65°C and assemble washing boxes and separating screens in the "hot" room.
3. Pour HYB solution into liquid waste and place Þlters in wash solution, separating each with a mesh screen. Rinse out hybridization boxes with hot water into liquid waste, then put in sink to wash.
4. Shake Þlters at 65°C for 20 min for Þrst two (2¥ and 1¥) washes and 20+ min. for third (0.5¥) wash. Leaving the third wash longer will signiÞcantly reduce background. First two washes go into liquid waste, third goes down the drain with tap water.
5. After the last wash, place Þlters between Whatman paper to blot excess liquid, then place in acetate sheet-protectors cut to the size of the Þlmholders. Put into labeled Þlmholders, noting approximate radioactivity with the Geiger counter and recording your loading scheme. In the darkroom, write Þlmholder number on Þlm with pencil and insert Þlm between the Þlters and the intensifying screen. Leave in &endash;80°C freezer to expose; usually for 5 to 7 d for wheat, ranging from 1 d (3000 cpm) to 10 d (200 cpm). Develop Þlm.
• Shown below are a few undesired results of hybridization:
Darkroom door was opened before Þlm entered the XOmat. Wait for buzzer!
Static discharge between Þlm and acetate sheet-protector. Avoid sliding Þlm along acetate when removing from Þlmholder.
Probe did not label. The spots seem to be related to the hybridization buffer used.
SSC (20¥)
Component Per liter For 20 L Final conc.
NaCl 175.3 g 3506.4 g 3 M
Citric acid
(trisodium salt) 149.1 g 2941.0 g 0.5 M
Bring to Þnal volume with dH2O. Adjust pH to 7.0 with 7&endash;8 ml 4N HCl (for 20 L).
POLYMERASE CHAIN REACTION (PCR)
Theoretical background
The polymerase chain reaction (PCR) is a powerful, extremely sensitive technique with applications in the Þelds of molecular biology, medical diagnostics, population genetics and forensic analysis. PCR is based on the enzymatic ampliÞcation of a DNA fragment that is þanked by two oligonucleotide primers hybridizing to opposite strands of the target sequence. The primers are oriented with their 3' ends pointing towards each other. Repeated cycles of heat-denaturation of the template, annealing of the primers to their complementary sequences, and extension of the annealed primers with a DNA polymerase result in the ampliÞcation of the segment deÞned by the 5' ends of the PCR primers (see the Þgure on the next page). Since the extension products of each primer can serve as a template for the other primer, each cycle essentially doubles the DNA fragment produced in the previous cycle. This results in the exponential accumulation of the speciÞc target fragment, up to several millionfold in a few hours.
Note: PCR is capable of amplifying as little as a single molecule of DNA. Precautions should be taken to guard against contamination of the reaction with trace amounts of DNAs that could serve as templates. Disposable gloves should be worn throughout the preparation of the PCR. Always include a control that contains all the components of the PCR except the template DNA.
PCR for RAPDs
You will run a PCR reaction on each of your four DNA preparations. Each pair of lab partners will also run a control reaction with only the reagents and primers &emdash; no template DNA.
1. Label four 0.5-ml microfuge tubes for the RAPD reactions. Prepare dilutions of your plant DNAs containing 10&endash;20 ng/5 ml ddw. For example, if the DNA is about 1 mg/ml, dilute 1 ml in 0.33 ml ddw and aliquot 5 ml of this dilution into the appropriate reaction tube. Add 5 ml of ddw to a control tube.
2. Make a reagent cocktail consisting of the following ingredients added in order to a 0.5-ml tube. Note that the quantities are per reaction, so you will need to multiply each by 4, or 5 depending on who is running the control.
Component Volume per reaction (ml)
Cocktail ingredients:
dNTPs 1 ml
PCR buffer 2.5 ml
150 mM MgCl2 0.07 ml
Double-distilled H2O to total 25 ml
Primers 1.32 ml each
Taq 0.4 ml
Vortex brieþy, then spin down for 3 sec.
3. To each of the labeled 0.5-ml tubes add the amount of cocktail calculated to make a 25-ml reaction volume.
4. Overlay the reaction mixture with 20 ml of silicon oil. This prevents evaporation of the samples during repeated cycles of heating and cooling.
5. Carry out ampliÞcation with the program indicated on your machine. The conditions set up in the program are as follows.
Thirty-Þve PCR cycles each of denaturation at 94°C for 1.5 min, annealing at 55°C for 2 min, and extension at 72°C for 2 min. After the last cycle, the samples are incubated for an additional 7 min at 72°C.
6. Add 5 ml of blue marker dye to each tube including the control.
Note: If it is desired to run several different primers on the same template DNA, the template DNA is included in the cocktail and the different primers are added to the labeled reaction tubes instead. In either case the Þnal reaction volume should be 25.0 ml.
PCR for ampliÞcation of clone inserts
We will amplify inserts from minipreps of a cDNA library made from barley. These clones (BCDs) are useful in wheat mapping, owing to the generally high homology between genomes of grass species. After the PCR is complete, the product is passed through a Sephadex column to remove the free nucleotides by size-exclusion chromatography. Otherwise the unlabeled cytidine will compete with the 32P&endash;cytidine during probe labeling.
Reaction composition
Component Volume per 100-ml reaction Final conc.
Double-distilled H2O 77.5 ml
PCR buffer (10¥) 10 ml 1¥
dNTPs (2.5 mM) 8 ml 0.2 mM
Primers (10 mM) 1 ml each 0.1 mM
Plasmid DNA 2 ml @ 100 ng
Taq polymerase (5U/ml) 0.5 ml 2.5 U
Reaction Conditions
Temp (°C) Time (sec)
Denaturing 94 90
Annealing 50 90
Extension 72 120
1. Run for 25 cycles, then end with a 5-minute extension at 72°C.
2. Aliquot out 10 ml of the ampliÞed sample and mix with 5 ml of blue juice.
3. Run out on a day gel for speed; make a 2% gel (5 g agarose in 250 ml of 0.5¥ NEB; running buffer also 0.5¥ NEB). Include a size standard like f/HaeIII or l, 50 ng/ml; 10 ml gives 500 ng. Run at around 80 mA for 5 h or less. Use this gel to conÞrm the sizes of the inserts. The intensity of the EtBr-stained bands will also allow estimation of the concentration of DNA in the PCR products and thus the amounts to use in hybridizations.
4. Cut off both ends of a 1-cc syringe to make a column: remove the narrower part of the tip and cut the top at 0.8 (need 1 column for each probe).
5. Cut polyethylene Þlters out of sheets with a #2 cork-borer and Þt snugly in the columns; put each column with Þlter in a 1.5-ml Eppendorf tube.
6. Fill column with G&endash;20 Sephadex solution(5 g in 100 ml). Spin at 3500 rpm for 2 min. There should be 0.5&endash;0.8 cc of white Sephadex. If necessary, reÞll and repeat spin until Þnally little liquid emerges from column; each time take columns out of tubes and shake liquid out of tubes, then replace columns. Transfer columns to new labeled Eppendorfs.
7. Pipette the PCR product into tops of columns in the labeled tubes. Spin at 3500 rpm for 2 min. Discard column in an designated place; as these columns are of the same kind used in the radiolabeling protocol, they should not be put into the regular lab waste. The PCR product is now ready for use in hybridization and may be stored in the refrigerator.
Primers
• BCD and CDO plasmids use M13F and M13R primers
• WG plasmids use GF and GR primers
• Amount of plasmid will vary depending on concentration if known.
Theoretical background
The conversion of RFLP data into map distances begins with recording the alleles at all the RFLP loci that have been tested on the population. At any locus in a diploid species (or a polyploid like wheat whose separate genomes behave as self-pairing diploids) three classes of genotype are possible: homozygous for either of the two parental alleles, or heterozygous. Here an allele is represented by a band at a speciÞc molecular weight. Some wrinkles now arise. In some cases RFLPs are dominant; that is, the polymorphism consists of the presence or absence of a single band, not of a band at a different molecular weight for either parent. In this case the homozygote cannot be distinguished from the heterozygote since in either case we will just see one band.
Another wrinkle involves the kind of population being used for mapping. For mapping a single-gene, highly heritable trait such as are many pathogen-resistance genes, F2 or BCF1 populations furnish relatively quickly generated sources of data. Quantitative traits may be better analyzed with recombinant-inbred or doubled-haploid populations in which replicated trials (over year, location, etc.) are possible. Of course, such populations are more tedious to develop (see Appendix B).
If the mapping population has been made by doubling the chromosomes of a haploid gametophyte, as by anther cul-ture and colchicine treatment, there will be no heterozygotes to worry about. If it is a recombinant inbred (RI) population, made by several generations of selÞng or backcrossing following a planned cross between the mapping parents, heterozygos-ity will be reduced but is commonly still present. The points of this discus-sion are two:
1) The expected proportions of genotype classes will differ according to the type of population used for mapping, no matter whether we are considering molecular or phenotypic traits. For example, using two codominant markers Aa and Bb: an F1 backcross onto a aabb homozygote gives 4 expected classes,
AaBb Aabb aaBb aabb
in expected proportions of 1/4 each, while an F1 selÞng gives nine classes &emdash; all possible pairings of AA, Aa, and aa with BB, Bb, and bb, and in expected proportions of 1:1:2:2:4:2:2:1:1. The classes will be the same nine with a RI population, but the expected proportions of heterozygotes will be drastically reduced.
2) A notation for genotype classes is needed when we score RFLP data. The standard method employed in a widely used linkage-analysis program, Mapmaker, is as follows:
Genotype class Symbol
Parent A A
Parent B B
Heterozygote H
Dominant marker, parent A D
Dominant marker, parent B C
Missing data or can't score &endash;
A simple linkage analysis
Multilocus maps are tedious to construct by manual calculation, so it is ordinarily done with computer programs. Here we will look at data from two RFLP markers &emdash; G&endash;48 and BCD 131 &emdash; on wheat chromosome 3, used by Z. Ma to probe a F2 population of 38 plants segregating for a single Hessian-þy resistance gene. G&endash;48 was developed by B. S. Gill et al. (KSU) using Triticum tauschii ( = Aegilops squarrosa), the D-genome diploid ancestor of cultivated wheat, while BCD 131 belongs to a barley cDNA library developed in the Sorrells lab at Cornell. Ma did not do the hand analysis we are going to do; he simply fed the RFLP and resistance scores into the Mapmaker program to assess their linkages. We refer prospective users to the program manual for instructions on preparation and analysis of your own data. The following demonstration will serve to give you a feel for handling mapping data.
Our linkage analysis for two RFLP loci will be based upon the autoradiograms shown below. Ma's scoring on these Þlms is shown here, since they were not easy to score even from the original Þlms and are even less so from reproductions. The a and b Þlters for each probe contain the Þrst and second parts of the population.
Hessian-þy resistance (B) population probed with G&endash;48, Þlter 1a
B population probed with G&endash;48, Þlter 1b
B population probed with BCD 131, Þlter 1a
B population probed with BCD 131, Þlter 1b
The G&endash;48 marker shows 6 bands, or locations where bands sometimes appear, while BCD 131 shows 3. These numbers are not coincidental but almost certainly correspond to the three homoeologous genomes of hexaploid bread wheat; that is, G&endash;48 is hybridizing at two sites each, and BCD 131 at one site each, on the A, B, or D genomes. In fact, hybridizing such probes to Þlters made with sets of nullisomic or ditelosomic stocks (Anderson et al. 1992) allows assignment of the various bands to speciÞc genomes.
It is further evident that the markers being scored in these Þlms appear as present-or-absent bands. But G&endash;48 has been scored for two polymorphic bands (marked with the circled numbers 1 and 2 at the left of Þlter 1a); do they in fact represent alleles of the same RFLP locus? These mapping Þlters do not contain the parents, but we can deduce their genotypes from Ma's scoring of the progeny. Inspection of the scoring entries (using the code of 1 for the genotype of parent A, 3 for that of parent B, and 2 for a heterozygote) indicates that they in fact represent two different loci, since parent B possesses neither of the bands while parent A has both of them.
Heterozygotes are distinguished here from homozygotes by relative brightness of the polymorphic band. A genotype that is homozygotic (like parent A) for band 1 has two doses of the locus, while a heterozygote has only one and the B homozygote for band 1 has none. (The occasional importance of a distinction between band strengths is a good reason why all the lanes on a mapping Þlter should have the same quantity of DNA) Let us assume that the scoring was correctly done and determine whether the top (number 1) band of G&endash;48 is segregating independently of the bottom band.
We will address this question by constructing a 3¥3 contingency table for pairs of markers and using the c2 statistic to test whether the proportions of genotype classes for one marker are independent of those for the other marker. In the cells of the table appear the numbers of occurrences of the combination of scores at the heads of the row and the column, read from the mapping Þlm. That is, the entry of 5 in cell (1,1) means that 5 times in the 38 lanes we see the combination of 1 and 1 as scores of the two G&endash;48 polymorphisms.
Band #
2 1 1 2 3
1 5 2 3 10
2 4 5 3 12
3 1 10 5 16
10 17 11 38
For the c2 analysis, we Þrst calculate the expected value for each cell as the product of its row and column totals divided by the grand total; for example the expected value for cell (1,1) is 10 ¥ 10 / 38 = 2.63. We then calculate for each cell
(observed &endash; expected)2/expected
and sum over all 9 cells to get 7.42. When row and column data are independent, this statistic is known to be distributed as c2 with 4 (= (3&endash;1) ¥ (3&endash;1)) degrees of freedom. Consultation of a statistical table shows that the probability of a greater value for this sum is somewhere between 0.25 and 0.10; these data are not very likely under the hypothesis of independence, but neither are they very unlikely. The analysis with Mapmaker in fact found a map distance of 41 cM, or loose linkage. We will not discuss the estimation of actual recombination distance from F2 data. The c2 analyses for the other pairs of loci are left as exercises.
In the References section are listed numerous papers detailing the mathematical basis for inferring linkage distances from segregation data in various kinds of populations.
QTL mapping
An attractive prospect for plant breeders in possession of genetic maps with closely placed molecular markers is the identiÞcation of quantitative-trait loci (QTLs) conditioning important agronomic and quality traits. With this knowledge in hand, breeders will be able to make selections based on molecu-lar markers linked to QTLs. Such a practice will avoid phenotypic selection that relies on poorly heritable, quantitatively expressed traits such as yield, days to þowering, height, some disease resistances, etc. The advantage of this practice is expected to be accelerated progress &emdash; measured in time and ex-pense &emdash; in varietal improvement. In any breeding program, some desirable gene combinations may be improbable enough that the breeder would have to grow out several thousand F2 or BCF1 progeny merely to have a good chance of having these combinations. But it would still be unlikely that the desirable genotypes could be identiÞed by their phenotypes. If, however, the desired genes are linked with known molecular markers, selection is potentially much more efÞcient.
QTL analysis requires a cross between two parents, either which differ widely with respect to the trait of interest, or for which it is expected that their F2 and advanced progeny will show transgressive segregation for this trait. The latter might hold if the parents were not closely related, and the experimenter believed that their expression of the trait, although of similar magnitude, was conditioned by different sets of genes. In either case, the progeny must segregate for the trait to be mapped, and this trait must be assayed in each progeny. The analysis also requires a set of molecular polymorphisms such as RFLPs, and they also must be determined for each progeny.
It will be apparent that establishing correlations between RFLPs and the expression of a trait does not require a genetic map; it would sufÞce to show that the markers and the trait cosegregate to a degree improbable by chance alone. Such a result is obtained with a simple chi-squared test. But for the longer term, we would like not only to select QTLs by means of linked molecular markers, but also to place them on the genetic map.
A widely employed computer program written to do this mapping is Mapmaker&endash;QTL. Brieþy, this program takes the two kinds of data &emdash; RFLP and phenotypic &emdash; obtained for the mapping population, and does a maximum-likelihood analysis (described in Lande and Thompson 1989) to determine the sites in a genome at which QTLs would best explain the phenotypic data.
We presently face two barriers to using this information:
1) We know little about whether the molecular markers linked to a quantitative trait in one population can be used to select for that trait in another, unrelated, population.
2) RFLPs are impractical for use in routine breeding, which requires rapid and inexpensive processing of thousands of genotypes.
These two barriers represent challenges to the interested researcher. The Þrst one, it seems, may be addressed by well-planned experimentation. The second may be addressed with developments in basic genetics, molecular biology, chemistry, and applied sciences that will make molecular genotyping cheap and routine. Further reading on QTL mapping, to supplement this brief overview, is given in the bibliography at the end of this manual.
Instructions for acquiring Mapmaker may be obtained from
MAPMAKER
c/o Dr. Eric Lander
Whitehead Institute
9 Cambridge Center
Cambridge, MA 02142 USA
Internet: mapmaker@genome.wi.mit.edu
Bitnet: mapm@mitwibr
Allard, R. W. 1956. Formulas and tables to facilitate the calculation of recombination values in heredity. Hilgardia 24: 235-276.
Anderson, J. A.; Churchill, G. A.; Autrique, J. E.; Tanksley, S. D.; Sorrells, M. E. 1993. Optimizing parental selection for genetic linkage maps. Genome 36: 181-186.
Anderson, J. A.; Ogihara, Y.; Sorrells, M. E.; Tanksley, S. D. Development of a chromosomal arm map for wheat based on RFLP markers. Theor. Appl. Genet. 83: 1035-1043.
Asins, M. J.; Carbonell, E. A. 1988. Detection of linkage between restriction fragment length polymorphism markers and quantitative traits. Theor. Appl. Genet. 76: 623-626.
Beckmann, J. S. 1988. Oligonucleotide polymorphisms: a new tool for genomic genetics. BioTechnology 6: 1061-1064.
Beckmann, J. S.; Soller, M. 1986. Restriction fragment length polymorphisms and genetic improvement of agricultural species. Euphytica 35: 111-124.
Beckmann, J. S.; Soller, M. 1988. Detection of linkage between marker loci and loci affecting quantitative traits in crosses between segregating populations. Theor. Appl. Genet. 76: 228-236.
Bonierbale, M. W.; Plaisted, R. L.; Tanksley, S. D. 1988. RFLP maps based on a common set of clones reveal modes of chromosomal evolution in potato and tomato. Genetics 120: 1095-1103.
Botstein, D.; White, R. L.; Skolnick, M.; Davis, R. W. 1980. Construction of a genetic linkage map in man using restriction fragment length polymorphisms. Am. J. Hum. Genet. 32: 314-331.
Burr, B.; Burr, F. A.; Thompson, K. H.; Albertson, M. C.; Stuber, C. W. 1988. Gene mapping with recombinant inbreds in maize. Genetics 118: 519-526.
Cowen, N. M. 1988. The use of replicated progenies in marker-based mapping of QTLs. Theor. Appl. Genet. 75: 857-862.
Devos, K. M.; Gale, M. D. 1992. The use of randomly amplified polymorphic DNA markers in wheat. Theor. Appl. Genet. 84: 567-572.
Edwards, M. D.; Stuber, C. W.; Wendel, J. F. 1987. Molecular-marker-facilitated investigations of quantitative-trait loci in maize. I. Numbers, genomic distribution and types of gene action. Genetics 116: 113-125.
Gill, K. S.; Lubbers, E. L.; Gill, B. S.; Raupp, W. J.; Cox, T. S. 1991. Linkage map of Triticum tauschii (DD) and its relationship to the D genome of Triticum aestivum (AABBDD). Genome 34: 362-374.
Graner, J.; Jahoor, A.; Schondelmaier, J.; Seidler, H.; Pillen, K.; Fishbeck, G.; Wenzel, G. 1991. Herrmann, R. G. Construction of an RFLP map of barley. Theor. Appl. Genet. 83: 250-256.
Heun, M.; Kennedy, A. E.; Anderson, J. A.; Lapitan, N. L. V.; Sorrells, M. E.; Tanksley, S. D. 1991. Construction of an RFLP map for barley (Hordeum vulgare L.). Genome 34: 437-447.
Hospital, F.; Chevalet, C.; Mulsant, P. 1992. Using markers in gene introgression breeding programs. Genetics 132: 1199-1210.
Kam Morgan, L. N. W.; Gill, B. S.; Muthukrishnan, S. 1989. DNA restriction fragment length polymorphisms: a strategy for genetic mapping of D genome of wheat. Genome 32: 724-732.
Knapp, S. J. 1991. Using molecular models to map multiple quantitative trait loci: models for backcross, recombinant inbred, and doubled haploid progeny. Theor. Appl. Genet. 81: 333.
Knapp, S. J.; Bridges, W. C., Jr.; Birkes, D. 1990. Mapping quantitative trait loci using molecular marker linkage maps. Theor. Appl. Genet. 79: 583-592.
Lande, R.; Thompson, R. 1990. Efficiency of marker-assisted selection in the improvement of quantitative traits. Genetics 124: 743-756.
Lander, E. S.; Botstein, D. 1989. Mapping Mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics 121: 185-199.
Lander, E. S.; Green, P.; Abrahamson, J.; Barlow, A.; Daly, M. J.; Lincoln, S. E.; Newburg, L. 1987. MAPMAKER: an interactive computer package for constructing primary genetic maps of experimental and natural populations. Genomics 1: 174-181.
Landry, B.; Michelmore, R. W. 1987. Methods and applications of restriction fragment length polymorphism analysis to plants, in: Tailoring Genes for Crop Improvement, Bruening, G.; Harada, J.; Kosuge, T.; Hollaender, A. (Eds.); Plenum Press, New York.
Liu, B. H.; Knapp, S. J. 1990. GMENDEL: A program for Mendelian segregation and linkage analysis of individual or multiple progeny populations using log-likelihood ratios. J. Hered. 81: 407.
Luo, Z. W.; Kearsey, M. J. 1989. Maximum likelihood estimation of linkage between a marker gene and a quantitative trait locus. Heredity 63: 401-408.
Mackinnon, M. J. Georges, M. A. J. 1992. The effects of selection on linkage analysis for quantitative traits. Genetics 132: 1177-1185.
Melchinger, A. E. 1990. Use of molecular markers in breeding for oligogenic disease resistance. Plant Breed. Rev. 104: 1-19.
Michelmore, R. W.; Paran, I.; Kesseli, R. V. 1991. Identification of markers linked to disease-resistance genes by bulked segregant analysis: A rapid method to detect markers in specific genomic regions by using segregating populations. Proc. Nat. Acad. Sci. USA 88: 9828-9832.
Morton, N. E. 1988. Multipoint mapping and the emperor's clothes. Ann. Hum. Genet. 52: 309-318.
Muehlbauer, G. J.; Specht, J. E.; Thomas-Compton, M. A.; Staswick, P. E.; Bernard, R. L. 1988. Near-isogenic lines&emdash;a potential resource in the integration of conventional and molecular marker linkage maps. Crop Sci. 28: 729-735.
Paterson, A. H.; Damon, S.; Hewitt, J. D.; Zamir, D.; Rabinowitch, H. D.; Lincoln, S. E.; Lander, E. 1991. Mendelian factors underlying quantitative traits in tomato: comparison across species, generations, and environments. Genetics 127: 181-197.
Paterson, A. H.; DeVerna, J. W.; Lanini, B.; Tanksley, S. D.;. 1990. Fine mapping of quantitative trait loci using selected overlapping recombinant chromosomes, in an interspecies cross of tomato. Genetics 124: 735-742.
Ritter, E.; Gebhardt, C.; Salamini, F. 1990. Estimation of recombination frequencies and construction of RFLP linkage maps in plants from crosses between heterozygous parents. Genetics 125: 645-654.
Sharp, P. J.; Chao, S.; Desai, S.; Gale, M. D. 1989. The isolation, characterization and application in the Triticeae of a set of wheat RFLP probes identifying each homoeologous chromosome arm. Theor. Appl. Genet. 78: 342-348.
Simpson, S. P. 1989. Detection of linkage between quantitative trait loci and restriction fragment length polymorphisms using inbred lines. Theor. Appl. Genet. 77: 815-819.
Snape, J. W. 1988. Detection and estimation of linkage using doubled haploid or single-seed descent populations. Theor. Appl. Genet. 76: 185-188.
Soller, M.; Brody, T.; Genizi, A. 1979. The expected distribution of markerlinked quantitative effects in crosses between inbred lines. Heredity 43: 179-190.
Strauss, S. H.; Lande, R.; Namkoong, G. 1992. Limitations of molecular-marker-aided selection in forest tree breeding. Can. J. For. Res. 22: 1050-1061.
Suiter, K. A.; Wendel, J. F.; Case, J. S. 1983. LINKAGE-1: a PASCAL computer program for the detection and analysis of genetic linkage. J. Hered. 74: 203-204.
Tanksley, S. D.; Young, N. D.; Paterson, A. H.; Bonierbale, M. W. 1989. RFLP mapping in plant breeding: new tools for an old science. Bio/Technology 7: 257-264.
Watkins, P. C. 1988. Restriction fragment length polymorphism (RFLP): applications in human chromosome mapping and genetic disease research. Biotechniques 6: 310-320.
Weining, S.; Langridge, P. 1991. Identification and mapping of polymorphisms in cereal based on the polymerase chain reaction. Theor. Appl. Genet. 82: 209-216.
Weller, J. I. 1986. Maximum likelihood techniques for the mapping and analysis of quantitative trait loci with the aid of genetic markers. Biometrics 42: 627-640.
Weller, J. I.; Soller, M.; Brody, T. 1988. Linkage analysis of quantitative traits in an interspecific cross of tomato (Lycopersicon esculentum ¥ Lycopersicon pimpinellifolium) by means of genetic markers. Genetics 118: 329-339.
Williams, J. G. K. {DuPont}. 1990. DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Res. 18: 6531-6535.
Young, N. D. 1990. Potential applications of map-based cloning to plant pathology. Physiol. Mol. Plant Pathol. 37: 81-94.
Young, N. D.; Tanksley, S. D. 1989. RFLP analysis of the size of chromosomal segments retained around the Tm-2 locus of tomato during backcross breeding. Theor. Appl. Genet. 77: 353-359.
Young, N. D.; Tanksley, S. D. 1989. Restriction fragment length polymorphism maps and the concept of graphical genotypes. Theor. Appl. Genet. 77: 95-101.
Zehr, B. E.; Dudley, J. W.; Chojecki, J. 1992. Some practical considerations for using RFLP markers to aid in selection during inbreeding of maize. Theor. Appl. Genet. 84: 704-708.
APPENDIX A. RECOMMENDED PROCEDURE FOR DEVELOPING POPULATIONS FOR MAPPING TRAITS
Mark E. Sorrells, Cornell University
1) For parental lines, use only single-plant selections from the purest parental seed stocks available.
2) Save seed from the actual plants used in the hybridization.
3) Grow F1 seed in separate pots or thin to one plant per pot. Verify that each F1 is a true hybrid and discard selfs or abnormal plants. Save a few F1 seed for meiotic analysis. If possible, collect sporocytes from the actual F1 plant used to produce F2 seed.