| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Plant Sciences, Loftsgard Hall, North Dakota State University, Fargo, ND 58105
b USDA-ARS, Cereals Crops Research Unit, Northern Crop Science Laboratory, Fargo, ND 58105
* Corresponding author (farisj{at}fargo.ars.usda.gov).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: AFLP, amplified fragment length polymorphism • BC, backcross • BSA, bulked segregant analysis • HRL, homozygous recombinant line • PCR, polymerase chain reaction • RFLP, restriction fragment length polymorphism • RSL, recombinant substitution line
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Tan spot of wheat and durum is an economically important disease worldwide. Most commercial cultivars are susceptible to tan spot, and although some genotypes with good levels of resistance have been identified, genotypes possessing complete resistance are rare (Rees and Platz, 1990). In addition to common wheat and durum, disease susceptibility extends to several of the wheat relatives (Krupinsky, 1992).
Eight major races of the tan spot fungus, Pyrenophora tritici-repentis, have been classified on the basis of their virulence and ability to produce the symptoms of tan necrosis and/or chlorosis (Lamari and Bernier, 1989; De Wolfe et al., 1998; Lamari et al., 2003). Inheritance of resistance to the tan spot fungus has been well documented (Lamari and Bernier, 1989; Sykes and Bernier, 1991; Lamari et al., 1991; Gamba and Lamari, 1998; Gamba et al., 1998; Anderson et al., 1999). The genetically independent symptoms of tan necrosis and extensive chlorosis were found to be controlled by a single recessive gene and a single dominant gene, respectively (Lamari and Bernier, 1991; reviewed in De Wolfe et al., 1998). Faris et al. (1997) identified a quantitative trait locus (QTL) with major effects for resistance to chlorosis on chromosome 1AS. Resistance to tan necrosis was thought to be controlled solely by Tsn1, which conditions sensitivity to Ptr ToxA (see below); however, Friesen et al. (2003) suggested that in some host backgrounds resistance to tan necrosis might be quantitatively inherited with Tsn1 playing a major role.
Pyrenophora tritici repentis is known to produce at least three host selective toxins (HSTs) that are toxic to certain genotypes of wheat and its close relatives (Ciuffetti and Tuori, 1999). Ptr ToxA, a proteinaceous 13.2 kDa HST, is a well-characterized necrosis toxin (Ballance et al., 1989; Tomas et al., 1990; Tuori et al., 1995; Zhang et al., 1997; Ciuffetti and Tuori, 1999), which causes extensive necrosis when infiltrated into the leaves of sensitive genotypes.
Host sensitivity to Ptr ToxA is associated with disease susceptibility (Friesen et al., 2003). The Tsn1 gene, located on the long arm of wheat chromosome 5B, conditions sensitivity to Ptr ToxA (Faris et al., 1996). The absence of Tsn1 results in insensitivity (Anderson et al., 1999). Functional studies of the action of Ptr ToxA using an electrolyte leakage assay have shown that de novo mRNA and protein synthesis are required for toxin activity (Kwon et al., 1998). Recently, it has been demonstrated that an arginyl-glycyl-aspartic (RGD) motif in the Ptr ToxA protein is responsible for eliciting the necrotic reaction (Meinhardt et al., 2002). This amino acid sequence is involved in the binding of effector molecules to transmembrane receptors called integrins (reviewed in D'Souza et al., 1988) in animals; however, it remains unclear what function integrin-like molecules have in plants. This biochemical and genetic data suggest active host processes may contribute to the virulence of the fungus.
The haploid DNA content of the wheat genome is approximately 16 billion base pairs, of which approximately 12% is composed of single copy sequences (Smith and Flavell, 1975). The wheat genome was once thought to be intractable to positional cloning because of its large genome size and complexity. It is now known that over 85% of the expressed genes in wheat are located in small chromosomal segments of high gene density and frequent recombination (Gill et al., 1996; Faris et al., 2000; Sandhu et al., 2001). Assembly of contiguous bacterial artificial chromosome (BAC) sequences and comparisons with corresponding genetic linkage maps have provided more precise estimates of physical to genetic distance ratios within gene-rich regions. Stein et al. (2000) assembled a 450-kb contig spanning the Lr10 leaf rust resistance gene and found recombination frequencies ranged from 400 kb/cM to 12 000 kb/cM with an average of 1.4 Mb/cM. Faris et al. (2003) constructed a 300-kb contig spanning the Q gene on chromosome 5AL where the estimates ranged from 130 kb/cM to >600 kb/cM with an average of 330 kb/cM.
Saturation mapping on chromosome 5B was performed to determine the recombination frequency of the gene-rich region between fraction breakpoints 0.75 and 0.79 where the Tsn1 gene is located (Faris et al., 2000). The recombination frequency was estimated to be 400 kb/cM, an 11-fold increase in recombination compared to the genomic average. These data suggest positional cloning could be used to isolate the Tsn1 gene.
The objectives of this research were to (i) use methylation-sensitive amplified fragment length polymorphism (AFLP) (Vos et al., 1995), traditional AFLP, and cDNA-AFLP to target markers to the genomic region containing the Tsn1 locus and (ii) develop a high-resolution map of the region.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Nullisomic-tetrasomic (NT) lines (Sears, 1966) and chromosome deletion lines (Endo and Gill, 1996) involving group 5 chromosomes were used to identify fragments within the targeted region on 5B. The NT lines and chromosome deletion lines were obtained from the Wheat Genetics Resource Center (WGRC), Kansas State University, Manhattan, KS. The CS x CS-DIC 5B RSL population (hereafter referred to as the CS RSL population) and a backcross population derived from LDN x LDN-DIC 5B (Israel A) (hereafter referred to as the LDN BC population) were used for preliminary genetic mapping of candidate markers and consisted of 117 and 85 individuals, respectively. High-resolution mapping was performed on 633 insensitive F2 plants derived from a cross between the synthetic hexaploid W-7976 and the cultivar Kulm (hereafter referred to as the WK F2 population) and on 930 insensitive F2 plants from LDN x LDN-DIC 5B (PI478742) (hereafter referred to as the LDN F2 population).
Ptr ToxA Screening
All plants were grown in the greenhouse at temperatures between 21 and 24°C with a 16-h photoperiod. The youngest fully expanded leaf at the third leaf stage was infiltrated with purified Ptr ToxA provided by Dr. Steven Meinhardt, North Dakota State University. The boundaries of the infiltration site were marked with a nontoxic felt pen, and the reactions were scored 3 d after infiltration as sensitive or insensitive (Fig. 1)
. All 85 individuals of the LDN BC population were screened with Ptr ToxA and used for mapping. For the high-resolution mapping populations, approximately 2500 WK F2s and 4000 LDN F2s were screened. Only insensitive plants were saved and used for mapping because they were homozygous for tsn1.
|
–33P]dATP in combination with selective displayPROBE primers. Amplified products were separated on a 50 cm 5% (w/v) polyacrylamide gel at 80 W for 3.5 h, dried on filter paper for 2 h at 80°C, and exposed to X-ray film for 3 to 7 d.
AFLP-BSA
Bulked DNA pools consisting of CS RSLs were constructed on the basis of genotypes of markers known to flank Tsn1 (Faris et al., 2000). Each bulk consisted of 10 individuals that were homozygous for either CS or DIC 5B within the targeted region. DNA extraction of CS, CS-DIC 5B, deletion line 5BL-14, and the bulks was done according to Faris et al. (2000). AFLP analysis using EcoRI and MseI restriction sites was performed with the AFLP Analysis System I (Life Technologies Inc.) according to the manufacturer's instructions, except that eight additional EcoRI and MseI primers were constructed. Selective amplification products were separated on a 50 cm 5% (w/v) polyacrylamide gel at 80 W for 3.5 h, dried on filter paper for 2 h, and either exposed to X-ray film for 2 to 7 d or exposed to a phosphorimaging screen for 2 to 5 h and scanned with a Typhoon 9410 Variable Mode Imager (Molecular Dynamics Inc., Sunnyvale, CA). A total of 256 primer combinations were screened for polymorphism between the bulks and the concomitant absence of the polymorphic fragment in deletion line 5BL-14.
Methylation-sensitive AFLP analysis was done according to the following method. Double-stranded adapters were prepared by mixing 1500 pmoles of each of the long and short strands of the MseI and PstI adapters to produce a 5 pmol/µL solution which was heated to 94°C for 3 min and allowed to slowly cool to room temperature for 2 h. The adaptor sequences are as follows: MseI long adaptor (5'-GACGATGAGTCCTGAG-3'), MseI short adaptor (5'-TACTCAGGACTCAT-3'), PstI long adaptor (5'-CTCGTAGACTGCGTACATGCA-3'), and PstI short adaptor (5'-TGTACGCAGTCTAC-3'). Genomic DNA from CS, the bulked segregants, and deletion line 5BL-14 was digested and ligated to adapters simultaneously as follows. Genomic DNA (250 ng) was mixed with 5 units of PstI and 2 units of MseI restriction enzymes, 5 pmol PstI adaptor, 50 pmol MseI adaptor, 1 µL 10 mM ATP, 1 unit T4 ligase, 0.5 µL 100x BSA (bovine serum album), 5 µL 10x One-Phor-All Buffer Plus (Amersham Pharmacia Biotech, Inc., Piscataway, NJ), and filled to a final volume of 50 µL with ddH20. This mixture was placed in the thermal cycler at 37°C for 2 h followed by 15 min at 70°C to inactivate the enzymes. The digestion/ligation reaction mixture was then diluted 10-fold with ddH20, and the diluted reaction mixture was used directly as template DNA for the preamplification.
To assemble the preamplification reaction (50 µL total volume), we mixed the following components: 5 µL diluted ligated DNA, 75 ng MseI-primer, 75 ng of the PstI one-base extension primer, 4 µL 2.5 mM dNTPs, 1 µL Advantage cDNA polymerase mix (Clontech Laboratories, Inc., Palo Alto, CA), 5 µL 10x PCR-buffer, and 32 µL water. The mix was then used for the preamplification step with the same thermal cycler program used for EcoRI/MseI preamplification.
Three microliters of the preamplification reaction was diluted with 147 µL of ddH20 for use in the selective amplifications. Selective amplification was performed according to the same procedures used for EcoRI/MseI AFLP.
Analysis of Positive Fragments
Positive AFLP and cDNA-AFLP fragments were excised from dried polyacrylamide gels, added to 40 µL dH20, and boiled for 10 min. The liquid was then purified and eluted to a final volume of 30 µL with the Qiagen QIAquick Gel Extraction Kit (Qiagen Inc., Valencia, CA) according to the manufacturer's instructions. Five microliters of this purified liquid was then used as template in PCR reactions to reamplify each fragment. cDNA-AFLP fragments were reamplified from their original selective primers, and AFLP fragments were reamplified from universal (EcoRI, MseI, or PstI) primers, which lacked selective nucleotides. PCR conditions were 94°C for 3 min followed by 30 cycles of 94°C for 30 s, 55°C for 30 s, and 72°C for 1 min, followed by 72°C for 7 min.
Ten nanograms of reamplified product was cloned and transformed with either the AdvanTAge PCR Cloning Kit (Clontech Laboratories Inc.) or the TA Cloning Kit (Invitrogen Corporation, Grand Island, NY) according to the manufacturer's instructions. Bacterial colonies were grown on Luria-Bertani (LB) plates containing carbenicillin and X-Gal (5-bromo-4-chloro-3-indolyl-ß-D-glactopyranoside) overnight at 37°C.
At least 12 white colonies were picked from each transformation event and grown in 150 µL of liquid LB-carbenicillin media for 6 to 18 h. One-half or 1 µL of each culture was used as template for PCR to reamplify the cloned inserts. Ten microliters of each sample was then separated on a 2% (w/v) agarose gel, and stained with ethidium bromide, visualized under UV light, and photographed.
Clones that contained the expected insert size were digested with the restriction enzyme RsaI and electrophoresed on a 2% (w/v) agarose gel for fingerprinting. Clones that showed unique restriction patterns were tested for map position by hybridization to Southern blots containing digested genomic DNA from CS, the group 5 nullisomic-tetrasomic lines, and deletion lines 5BL-9 and 5BL-14. Clones that were low-copy and mapped distal to the 5BL-14 breakpoint were used for genetic mapping.
RFLP Analysis
DNA isolation, Southern blotting, and hybridization procedures were performed as described in Faris et al. (2000). Fragments proven to hybridize within the targeted deletion interval were tested for polymorphism in the low-resolution mapping populations by hybridizing them to diagnostic blots containing the parents of each respective population digested with the restriction enzymes EcoRI, EcoRV, DraI, HindIII, and either XbaI or ScaI. Those markers showing polymorphism were used to construct low-resolution genetic linkage maps of the region. Markers found to detect loci closely linked to Tsn1 were then hybridized to the parents of the high-resolution mapping populations digested with restriction enzymes ApaI, BamHI, BglII, DraI, EcoRI, EcoRV, HindIII, KpnI, SacI, ScaI, and XbaI, and subsequently hybridized to individuals of the populations digested with the enzyme giving the clearest polymorphism.
Linkage Analysis
The computer program Mapmaker (Lander et al., 1987) version 2.0 for Macintosh was used to calculate linkage distances using the Kosambi mapping function (Kosambi, 1944) and a LOD of 2.0 for the LDN BC map, 3.0 for the CS RSL map, and 7.0 for the WK F2 and LDN F2 maps.
Sequencing
A sterile toothpick was used to pick a single bacterial colony and inoculate 3 mL of liquid LB media containing 20 mg/mL carbenicillin. Cultures were grown for approximately 16 h at 37°C while shaking at 225 rpm. Plasmid DNA was isolated from cultures with the Qiagen Plasmid Mini Kit (Qiagen Inc., Valencia, CA) according to the manufacturer's instructions. Sequencing was done by the Kansas State University Sequencing and Genotyping Facility (Manhattan, KS). Sequences were analyzed for similarity to known sequences in public databases by BLASTx and BLASTn searches against the NCBI non-redundant (nr) database (Altschul et al., 1997). Sequences were also analyzed with Gramene (Ware et al., 2002) to identify orthologous sequences in rice (Oryza sativa L.) BACs/PACs.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
|
|
Probes FCG9 and FCG12 hybridized to 5B chromosomes only. Probe FCG3 was also specific for group 5 chromosomes, but it detected loci on all three homeologs. Probes FCG1, FCG2, FCG7, FCG10, FCG13, FCG16, and FCG17 each hybridized to a single fragment on the long arm of chromosome 5B and no other group 5 chromosomes, but they did detect other fragments not located on group 5 chromosomes.
cDNA-AFLP-BSA
Sixty-four primer combinations were surveyed by cDNA-AFLP analysis. A total of 15 positive fragments were detected, isolated, and used as RFLP probes. Of these fragments, 8 were low-copy and 7 were high-copy sequences. Of the low-copy fragments, three detected loci on group 5 chromosomes, and two of these mapped within the targeted region (Table 1, Figure 3). Probes FCC1 and FCC2 detected loci on all three homeologs, and FCC3 detected loci on chromosomes 5B and 5D. None were group 5 specific.
Mapping of Clones
Physical Mapping
Initial mapping of newly identified (cDNA)-AFLP fragments consisted of conversion to RFLP markers and hybridization to diagnostic blots containing the group 5 NT lines and the deletion lines 5BL-9 and 5BL-14. A total of 18 clones derived from the three techniques used in this experiment were identified as low-copy and mapped to the region between deletions 5BL-14 and 5BL-16. In addition, clones FCG3, FCG5, and FCC3 were hybridized to a panel of 5B long arm deletion lines and were found to map between the breakpoints in deletion lines 5BL-1 and 5BL-11 (Fig. 3).
Low-Resolution Genetic Mapping
Low-resolution genetic mapping is necessary for the expedient ordering of genetic markers. For initial mapping, RFLP markers within the Tsn1 region were selected from maps presented by Faris et al. (1996)(2000) and used to construct skeleton maps in the low-resolution LDN BC and CS RSL populations. All markers that mapped within the Tsn1 region on the physical map and showed polymorphism between the parents of the respective mapping populations were used to construct the low-resolution maps.
Eight (cDNA)-AFLP derived clones could be mapped in the LDN BC population. Markers Xfcg1, Xfcg10, Xfcg16, and Xfcg17 cosegregated with Tsn1 in this population. Fourteen (cDNA)-AFLP clones could be mapped in the CS RSL population. Only one marker, Xfcg17, mapped at a LOD <3.0. Presumably, this is due to one plant having a double crossover flanking the marker (data not shown). Although this population does not segregate for Tsn1, the markers Xfcg1 and Xfcg17 did not cosegregate with each other or with Xfcg10 and Xfcg16, indicating reduced recombination within the Tsn1 region in the LDN BC population compared with the CS RSL population. Comparison of the genetic regions flanked by RFLP markers Xbcd1030 and Xcdo400 further delineates reduced recombination in the LDN BC population. The LDN BC genetic map is only 5.3 cM between markers Xcdo400 and Xbcd1030, while the same interval on the CS RSL genetic map accounts for 8.9 cM.
High-Resolution Genetic Mapping
Markers closely linked to Tsn1 were further mapped in 633 WK F2s and 930 LDN F2s that were genotyped for homozygosity at the Tsn1 locus. Because of low levels of polymorphism, relatively few clones could be mapped in the WK F2 population; however, markers Xfcg7 and Xfcg9 flanked Tsn1 at 0.4 and 0.2 cM, respectively, and the marker Xfcg17 cosegregated with Tsn1 (Fig. 4)
. The same probes were mapped in the LDN F2 population. Xfcg7 and Xfcg9 flanked Tsn1 at 0.4 and 0.6 cM, respectively, and Xfcg17 mapped 0.2 cM proximal to Tsn1.
The interval between markers Xfcc1 and Xfcg7 contains 4.4 cM in the WK F2 population, but the same interval consists of 3.6 cM in the larger LDN F2 population. The size difference for this genetic interval is primarily due to a difference in recombination frequency between Xfcc1 and Xfcg10, which is 2.7 cM on the WK F2 map compared to 1.7 cM on the LDN F2 map. The genetic interval between Xfcg10 and Xfcg7 is nearly identical at 1.7 cM on the WK F2 map and 1.9 cM on the LDN F2 map. The LDN F2 map had better resolution at the Tsn1 locus because three recombinants between Tsn1 and Xfcg17 were observed while no recombinants between Tsn1 and Xfcg17 were observed in the WK F2 population.
Sequence Analysis of Clones
All cDNA-AFLP and AFLP-derived clones that detected fragments on the long arm of chromosome 5B were sequenced. Three pairs of probes, FCC1/FCC2, FCG9/FCG12, and FCG10/FCG16, detected identical fragments when hybridized to Southern blots containing genomic DNA. Therefore, once sequence was obtained, we aligned the sequences of each of the three pairs to determine if we had cloned common fragments or alleles.
Clones FCC1 and FCC2 were amplified with different cDNA-AFLP primers, and FCC1 was cloned from LDN while FCC2 was cloned from DIC-5B. These two sequences aligned with 87% identity. FCG9 and FCG12 were amplified with different PstI/MseI primers, and FCG9 was cloned from CS DNA while FCG12 was cloned from DIC-5B. The sequences of these two clones aligned perfectly with 100% identity. Therefore, in the case of clone pairs FCC1/FCC2 and FCG9/FCG12, the clones represent alleles of the same fragments.
FCG10 and FCG16 were amplified with different PstI primers, but the same MseI primer. FCG10 was cloned from DIC-5B while FCG16 was cloned from CS. Sequence alignment of these two clones indicated they share identity only within the primer sequences, and there is no similarity within the segments between the primer sequences. Therefore, we do not consider these two clones as alleles even though they are physically close enough to each other to hybridize to the same fragments on genomic Southern blots.
The sequences of cDNA-AFLP and AFLP-derived clones within the Tsn1 genomic region defined by the deletion fraction breakpoints in 5BL-14 and 5BL-16 were subjected to BLASTn and BLASTx searches of the NCBI nonredundant (nr) database (Altschul et al., 1997). The most significant BLASTn and BLASTx hits are presented in Table 2. Of the 16 unique sequences tested, 10 sequences had no significant similarity to any sequences in the database. The remaining six unique sequences had significant similarities to known sequences in wheat, barley (Hordeum vulgare, L.), and rice.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
cDNA-AFLP analysis of RNA isolated from sensitive and insensitive bulked segregants allowed us to identify fragments of expressed genes within the targeted genomic region; however, this procedure may have shown bias toward high copy number RNAs (Bertoli et al., 1995). Approximately 2700 fragments were visualized by the cDNA-AFLP technique, and two of these resulted in low-copy markers that mapped within the targeted interval. However, these two clones were found to contain identical sequences; therefore, only one unique and useful marker was generated from this technique. Thus, 0.04% of the observed fragments yielded useful markers within the targeted region.
AFLP bands represent genomic sequences that may be intergenic or intragenic. The EcoRI/MseI AFLP technique shows no bias toward expressed regions of the genome, and fragments derived from the technique may therefore be from any genomic region. Over 20 000 fragments were visualized using the EcoRI/MseI AFLP technique. One of these resulted in the discovery of a low-copy marker that mapped within the targeted interval. For this procedure, 0.005% of the observed fragments yielded useful DNA markers.
The methylation-sensitive restriction enzyme PstI can be used in AFLP analysis to locate markers in the recombinationally active regions of chromosomes (Powell et al., 1997). Unlike EcoRI/MseI AFLP, the fragments amplified by this technique are likely to be located in regions of euchromatin. There is no bias toward highly expressed gene regions, which is an advantage of PstI/MseI AFLP over cDNA-AFLP. Over 20 000 fragments were visualized by the methylation-sensitive PstI/MseI AFLP technique, and 15 of these fragments produced 14 unique markers that mapped within the targeted interval. Therefore, the efficiency of this technique was 0.075%, which is 14 times more effective than the EcoRI/MseI AFLP technique and about twice as effective as the cDNA-AFLP technique.
A similar approach to gene targeting was employed by Faris and Gill (2002). Differential display analysis and AFLP were combined with the chromosome deletion lines known to flank the Q gene, which is largely responsible for the domestication of wheat. Here, the differential display technique had an efficiency of 0.04%, and several of the positives isolated from PAGE gels did not map to the targeted deletion interval. These were likely transcripts regulated by the Q gene or another closely linked regulatory gene. EcoRI/MseI AFLP on the deletion lines appeared to be much more effective than the differential display technique. With an efficiency of 0.14%, AFLP showed a 3.5-fold increase in efficiency for gene targeting as opposed to differential display on the same genetic material.
When comparing our results with those of Faris and Gill (2002), one must consider the size of the targeted genomic region. The genetic size of the segment targeted by Faris and Gill (2002) was about 20 cM, while our target was less than 10 cM. A target of reduced size will presumably lead to lower efficiency. The 0.14% efficiency of EcoRI/MseI AFLP reported by Faris and Gill (2002) would likely have been very similar to the 0.075% efficiency that we report for PstI/MseI AFLP had their targeted region been the same size.
Taken together, our results and those of Faris and Gill (2002) suggest the use of AFLP for the amplification of genomic DNA may be more efficient for genomic targeting in wheat than mRNA differential display or cDNA-AFLP. This is probably due to the amplification of transcripts derived from genes that do not map to the targeted interval but are, instead, controlled by regulatory genes within the region, or a bias toward the amplification of high-copy transcripts. In the current study, methylation-sensitive AFLP was much more efficient for genomic targeting than was traditional methylation-insensitive EcoRI/MseI AFLP or cDNA-AFLP. This may be due to its preferential targeting of recombinationally active (gene rich) regions without the regulatory and transcript bias problems associated with differential display (Bertioli et al., 1995), and possibly cDNA-AFLP, techniques.
Physical and Genetic Mapping
The premise underlying positional cloning is that a gene's location can be pinpointed with great enough precision to allow for its sequencing or its subjugation to complementation or transformation studies. Genetic mapping accomplishes the localization of a gene locus by the determination of its closest flanking crossover events. The likelihood of obtaining the best estimate of a gene's position by genetic mapping relies upon several factors inherent to the mapping population including the relatedness of the parental lines used for crossing and population size.
Population size has greatly influenced the estimation of the recombination fraction in this research. Resolution differences due to population size are apparent in comparisons among the smaller and larger mapping populations. Only in the largest population of 1860 gametes did we find that the marker Xfcg17 and Tsn1 could be resolved to 0.2 cM. These results substantiate the importance of population size for positional cloning purposes, but the size of population necessary for positional cloning depends on the frequency of recombination that occurs within the genomic region of interest. The Lr10 leaf rust resistance locus in hexaploid wheat was spanned by a T. monococcum BAC contig following the analysis of 6240 gametes (Stein et al., 2000). Within a 350-kb segment, they found recombination frequencies to range from 400 kb/cM to more than 12 000 kb/cM. However, Faris et al. (2003) found that 930 gametes were sufficient to do chromosome walking and assemble a contiguous BAC sequence spanning the Q gene on the long arm of chromosome 5A. Here, they found recombination frequencies ranged from 130 to 600 kb/cM. Therefore, depending on the target region and the frequency of recombination that occurs within it, a relatively small population may be sufficient for positional cloning.
The physical size of the haploid wheat genome is approximately 16 000 Mb. The total genetic length of the hexaploid wheat genome is approximately 3700 cM; therefore, the genome-wide genetic to physical distance ratio is about 4.4 Mb/cM. If we assume that each chromosome is of equal size and that the long arm of chromosome 5B accounts for two-thirds of the entire chromosome length, it would be composed of 510 Mb.
Faris et al. (2000) performed saturation mapping on chromosome 5B to determine the recombination frequency of the gene-rich region between fraction breakpoints 0.75 and 0.79 where the Tsn1 gene is located. Although the exact size of this chromosomal region is not known, cytological experiments indicate it is approximately 4% of the long arm of chromosome 5B. Because the recombination-based map of this region was about 50 cM, the recombination frequency was determined to be approximately 400 kb/cM, which is an 11-fold increase in recombination compared with the genomic average.
We have been able to estimate further the recombination frequency in the Tsn1 region between fraction breakpoints 0.75 and 0.76, a region consisting of approximately 1% of the length of the long arm of chromosome 5B. The physical size of this region is about 5.1 Mb, and there are 18.4 cM between breakpoints 0.75 and 0.76 in the CS RSL population. Therefore, the recombination frequency is about 272 kb/cM in this region, a 16-fold increase in recombination compared to the genomic average, and a 1.5-fold increase compared with previous estimates of the recombination frequency between fraction breakpoints 0.75 and 0.79 (Faris et al., 2000). These data suggest markers Xfcg17 and Xfcg9 will provide excellent starting points for chromosome walking and the eventual map-based cloning of Tsn1.
Wheat/Rice Colinearity within the Targeted Region
A remarkable level of colinearity between the rice and wheat genomes at the chromosome level has been well established. Using primarily RFLPs, the rice genome was divided into linkage blocks that correspond to homologous wheat chromosomes (Ahn et al., 1993; Van Deynze et al., 1995; Gale and Devos, 1998). These results implied that the relatively small genome of rice could be used as a tool for cloning genes in grasses with much larger genomes such as wheat. However, attempts to clone genes from barley or wheat with rice as a vehicle have met with varying success (Gallego et al., 1998; Han et al., 1999), and indeed, other experiments have shown that the degree of microcolinearity is limited (Feuillet and Keller, 1999; Dubcovsky et al., 2001; Li and Gill, 2002; Brunner et al., 2003). However, Brunner et al. (2003) showed that the availability of the extensive rice genome sequence data could be used to generate markers for saturation mapping to expedite chromosome walking to a gene of interest in a related species. But, the success of this approach relies on the presumption that some degree of microcolinearity exists at the targeted region.
In extensive in silico comparative analysis of bin-mapped wheat ESTs with rice BAC/PAC sequences, Sorrells et al. (2003) showed a remarkable level of conservation in gene order of rice linkage group 1 with wheat homeologous group 3 chromosomes. However, in comparison to rice, wheat homeologous group 5 chromosomes showed very low levels of colinearity with rice sequences and contained an abundance of rearrangements that would likely lead to complications when rice is used as a model for cross-species transfer of information involving these chromosomes. Indeed, comparative mapping of the wheat crossability locus on the short arm of chromosome 5B with rice showed a break in colinearity indicating the complexity of syntenic relationships at the fine level (Lamoureux et al., 2002).
Our comparative analysis of the Tsn1 locus on wheat chromosome 5B with rice agrees with Sorrells et al. (2003) that very low levels of colinearity exist between wheat group 5 chromosomes and rice. Not only was homology to three different rice chromosomes detected within a 15.6-cM interval, but nine of the 15 markers within the interval detected no homologous sequences in the rice databases. Of the nine sequences that had no similarity to rice, three (FCC1/FCC2, FCG1, and FCG7) represent coding sequences as indicated by their presence in the Triticeae EST databases. The remaining six sequences with no similarity to rice likely represent noncoding sequences. While these sequences are expected to be less conserved through evolution, the three coding sequences would be expected to retain some level of conservation. It is possible that the orthologous sequences of these markers have not yet been sequenced in rice, or they simply may not exist in the rice genome. Nevertheless, these results indicate that rice would not be useful as a vehicle for cloning the Tsn1 gene or other genes in this region.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
Received for publication May 21, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Related articles in Crop Science:
This article has been cited by other articles:
|
|
 |
|
  T. L. Friesen, Z. Zhang, P. S. Solomon, R. P. Oliver, and J. D. Faris Characterization of the Interaction of a Novel Stagonospora nodorum Host-Selective Toxin with a Wheat Susceptibility Gene Plant Physiology, February 1, 2008; 146(2): 682 - 693. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
