| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J.M. McGrath, USDA-ARS, Sugar Beet and Bean Research Unit, 494 PSSB, Michigan State Univ., East Lansing, MI 48824-1325; D. Trebbi, Keygene N.V., P.O. Box 216, 6700 AE Wageningen, The Netherlands; A. Fenwick and L. Panella, USDA-ARS, Crops Research Lab., 1701 Centre Ave., Fort Collins, CO, 80526; B. Schulz, KWS SAAT AG, Grimsehlstrasse 31, 37574 Einbeck, Germany; V. Laurent, UMR1281 Stress Abiotiques et Differencation des Vegetaux Cultives, Univ. des Sciences et Technologies de Lille, 59655 Villeneuve d'Ascq, France; and Florimond Desprez, 59242 Cappelle en Pévèle, France; S. Barnes, Sesvanderhave NV/SA, Industriepark, Soldatenplein Z2 nr 2, 3300, Tienen, Belgium; S.C. Murray; Institute for Genomic Diversity, Cornell Univ., Ithaca, NY 14853. Mention of a trademark or proprietary product does not constitute a guarantee or warranty of the product by the USDA or imply approval to the exclusion of other products that may also be suitable
* Corresponding author (mitchmcg{at}msu.edu).
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMaterials and MethodsResultsDiscussionREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionREFERENCES |
|---|
A number of molecular marker genetic maps in sugarbeet have been constructed (Barzen et al., 1992, 1995; Halldén et al., 1996; Hansen et al., 1999; Nilsson et al., 1997; Pillen et al., 1992, 1993; Rae et al., 2000; Schondelmaier et al., 1996, 1997; Schumacher et al., 1997; Uphoff and Wricke 1992, 1995). Each has been constructed from sugarbeet, and other crop types [table and fodder beet, chard, wild beet (B. vulgaris subsp. maritima)] are not yet represented with genetic maps. Although their fundamental genetic basis is unlikely to be vastly different, allele frequencies will likely vary, and fixation of crop-type-specific alleles might be expected. Many marker systems have been used, most are anonymous, including restriction fragment length polymorphisms (RFLPs), randomly amplified DNA polymorphisms (RAPDs), AFLPs, and SSRs, as well as a few morphological (e.g., color, seed type) and isozyme markers. Some single nucleotide polymorphisms (SNPs) within protein-encoding genes are available for mapping in sugarbeet (Möhring et al., 2004; Schneider et al., 2001), and Pillen et al. (1996) determined linkage relationships among 12 nuclear genes encoding chloroplast thylakoid proteins. In these published maps, the number of markers used ranged from 85 to 413 markers, and the total genetic distance summed across nine linkage groups ranged from 621 cM to 1057 cM. Most maps showed strong clustering of markers in one or two regions of each linkage group, suggesting restricted genetic recombination, and perhaps influenced by the type of marker used (Nilsson et al., 1997). Genes linked in Arabidopsis were co-located to beet and other species chromosomes (Dominguez et al., 2003), demonstrating the blocks of conserved synteny extend among unrelated eudicot plant families. Importantly, Schondelmaier and Jung (1997) defined molecular, isozyme, and morphological linkage groups based on the Butterfass (1964) trisomic series, thus establishing a common nomenclature for beet linkage groups. Inconsistencies persist in the literature regarding chromosome assignments, although many maps contain a few morphological markers in common.
The work described here represents a step toward a public set of markers for genetic analyses in beet. A significant aspect of this work is that the nine linkage groups have been delineated, and named according to the Butterfass chromosome nomenclature. The DNA of this mapping population has been amplified using rolling circle amplification (Dean et al., 2001; Brukner et al., 2005). This mapping resource can be, and has been, shared among laboratories for efficient marker placement on a common genetic framework. Genetic maps rely on recombination to locate markers to linkage groups, thus a marker must be polymorphic in this population to be mapped. To circumvent this limitation, a physical map is being constructed using large insert clones from a bacterial artificial chromosome (BAC) library (McGrath et al., 2004). Physical maps only rely on the absolute distance between two loci measured in base pairs. The physical mapping resource is also amplified using rolling circle amplification and represented in pools of BAC clones that allow convenient screening via polymerase chain reaction (PCR). Typically, the BAC library is used to recover genes of interest in their native genomic state. These large genomic fragments are a rich source of potentially informative genetic markers that can be screened in any mapping population, and provide a means to link the nascent physical and genetic maps.
|
Materials and Methods |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionREFERENCES |
|---|
AFLP Marker Mapping
Amplified fragment length polymorphism mapping followed the protocol of Vos et al. (1995) with modification (Myburg et al., 2001). Two different restriction enzyme pair combinations were used to generate the initial AFLP linkage map: the C-methylation-insensitive EcoRI/MseI (E/M) combination and the C-methylation-sensitive PstI/MseI (P/M) combination. Preamplification was performed with one (A or C for EcoRI) or zero (for PstI) selective nucleotide primers, and one selective nucleotide (A or C) for MseI primers. A total of 16 and 20 different selective primer combinations (PCs) were analyzed for E/M and P/M restriction enzyme pair combinations, respectively. Fluorescence-labeled primers (IRD700 or IRD800; LI-COR Biosciences, Lincoln, NE) with three (EcoRI, E+3) or two (PstI, P+2) selective nucleotides were used along with a single unlabeled MseI primer also with three selective nucleotides (M+3) to generate the scored AFLP fingerprints. Expected Mendelian segregation ratios for 1:2:1 and 3:1 were tested by 
2 analysis for codominant and dominant polymorphic amplified fragments, respectively, and marker clustering tendency was tested using the Poisson distribution.
Other Markers and Linkage Group Nomenclature Unification
A limited number of other marker types were used in the genetic map, including 25 RFLPs, 46 SSRs, 14 ESTs–UTRs (expressed sequence tags–untranslated regions), and three phenotypic markers (Table 1). The SSR sequences described but not mapped previously were tested (Cureton et al., 2002; Mörchen et al., 1996; Richards et al., 2004; Viard et al., 2002) and mapped if polymorphism was evident. Twenty-three SSR loci are newly described; those with the prefix FDSB were discovered and mapped in Florimond Desprez's population (Cappelle en Pévèle, France); SSRs prefixed USDA were identified and mapped in SESVANDERHAVE's population (Tienen, Belgium) from ESTs deposited in the NCBI database; and those named according to their GenBank numbers were mapped in this study from beet ESTs deposited in GenBank processed with SSR Primer software (Robinson et al., 2004). Fourteen SSR markers, owned by KWS SAAT AG (Einbeck, Germany), were run on the population used here to confirm and unify chromosome nomenclature according to Butterfass trisomics.
|
The procedure for RFLP was as described by McGrath et al. (1993), using 5 µg of DNA and one of four restriction enzymes (EcoRI, EcoRV, HindIII, or XbaI). Probes were generated from randomly selected cDNA clones from a sugarbeet leaf library (courtesy of Dan Bush, Ft. Collins, CO) or germinating seedlings (de los Reyes et al., 2003), and were amplified from purified plasmids followed by excision from an agarose gel, labeled with 32P-dCTP, and hybridized and detected as described.
The EST–UTR markers were generated from a single IRD700 labeled gene-specific primer in a pool of single enzyme digested genomic DNA ligated to unlabeled T7 adaptor sequences, detected as for AFLP. The EST–UTR markers were named with the GenBank Accession number followed by an E or D (for EcoRI or DraI, respectively; if scored dominant in this case), or cd (if codominant) and a sequential number. All EST–UTR loci reported here were developed from the calmodulin-like EST BI543691 (sequence in Supplementary Table 1).
Linkage analysis used JoinMap 3.0 software (Van Ooijen and Voorrips, 2001) with LOD grouping threshold of 4.0. Marker order was calculated using pairwise data estimated with the REC threshold function set to 0.35 and LOD threshold > 3.0. Genetic distances were corrected for double crossover events using the Kosambi function.
Mapping Population Replication and Immortalization
Genomiphi DNA Amplification (GE Healthcare Technologies, Waukesha, WI) was used exactly following manufacturer's directions, based on the methods of Dean et al. (2001) and Brukner et al. (2005). 100 ng of DNA (1 µL) was added to 9 µL of (proprietary) sample buffer containing random hexamer primers, and heated to 95°C for 3 min. To this was added 10 µL of Phi29 DNA polymerase and dNTP mix (proprietary concentrations consisting of 1 µL enzyme and 9 µL reaction buffer, based on Dean et al., 2001), incubated at 30°C for 20 h, followed by heat inactivation of Phi29 at 65°C for 10 min. The resulting DNA products were diluted with water to 50 ng µL–1, and 1 µL of this diluted sample was used for traditional PCR.
Markers to BACs
The sugarbeet BAC library SBA (Amplicon Express, Pullman, WA), constructed from the hybrid US H20 sugarbeet genome (McGrath et al., 2004), was matrix pooled (Stormo et al., 2004). This allowed a specific clone to be identified in two rounds of PCR. Initially, a signal was identified within one of eight 4608 BAC clone superpools. Each superpool has a corresponding matrix pool consisting of 36 PCR reactions designed to resolve an individual plate, row, and column within each superpool. The second round of PCR identified the specific desired clone from among these 36 matrix pools, each with 1152 BAC clones, constructed from one superpool. These pools are available for research purposes. Mapped markers and genes were identified to individual BAC clones via the pooling strategy by PCR using 1 x GoTaq Green master mix (Promega, Madison, WI), 0.375 µM each forward and reverse primer, and 50 ng DNA. The PCR conditions consisted of an initial denaturation at 94°C for 1.5 min, followed by 13 cycles of 94°C for 30 s, 58°C for 30 s (touchdown using –0.8 C per cycle), 72°C for 60 s, and an additional 31 cycles of 94°C for 30 s, 47°C for 30 s, 72°C for 60 s, and final extension of 72°C for 10 min.
|
Results |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionREFERENCES |
|---|
The map framework was primarily based on the segregation of AFLP markers. Two enzyme systems were used, the more traditional EcoRI/MseI combination and the PstI/MseI combination, whose details are elaborated here for beet. Similar numbers of fragments were scored for each combination between the parents (728 for EcoRI vs. 830 for PstI). Overall, the 36 different restriction enzyme PCs generated 1558 amplified fragments (43.3 bands PC–1), of which 316 (8.8 bands PC–1; 20.3%) were polymorphic between parents. The 16 E/M PCs yielded 15 to 79 bands PC–1, averaging 45.5 bands PC–1 (SD = 16.9), of which 10.4 bands PC–1 (SD = 7.1) showed polymorphism (22.9%). The 20 P/M PCs yielded 17 to 93 bands PC–1, averaging 41.5 bands PC–1 (SD = 19.7), of which 7.5 bands PC–1 (SD = 4.8) were polymorphic (18.0%). Percentages of polymorphisms for each PC ranged from 3.7 to 41.4% for E/M and from 5 to 29.7% for P/M combinations. The percentage A/T nucleotide content of the selective nucleotides was not statistically correlated with the total number of amplified fragments, or with the percentage of polymorphisms.
Initially, AFLP markers were used to define unnamed linkage groups. In this first iteration, the number of AFLP markers and the length of individual linkage groups varied from 26 to 47, and from 36.8 to 69.7 cM, respectively (data not shown). The Poisson distribution of AFLP-derived markers indicated that, at a density of >5 markers per 5 cM, 51.7% (61/118) of E/M markers significantly clustered (P < 0.001) on six linkage groups (1, 3, 5, 6, 8, and 9), while 14.4% (19/132) of P/M markers clustered on just two (3 and 9). In the second map iteration, AFLP-identified linkage groups were named according to the Butterfass nomenclature using known chromosome assignments of morphological loci and chromosome-specific SSR markers (including those coded KWS, Table 1), and the combined map was integrated with other markers listed Table 1. In the final iteration reported here (Table 1), AFLP markers with LOD scores < 4.0 were discarded. The final map retained 115 E/M and 127 P/M AFLP markers.
All SSR markers were mapped with genomic DNA that had been amplified using Phi29 polymerase mediated rolling circle replication. Successful placement of these markers with respect to AFLP markers, in particular, demonstrated the utility of this method to amplify DNA of the mapping population, and thus can be used to provide adequate DNA amounts for continued discovery and mapping new SSRs. Genomiphi amplified DNA proved very reliable for PCR-based markers, but not RFLP or other hybridization-based detection approaches, where complex band patterns or smears were seen, perhaps the result of strand switching during the rolling circle replication process or the single stranded nature of the replicated products.
Morphological traits were scored and showed the expected results. The R locus, which governs production of betalain pigments typically used as a hypocotyl color marker for hybrid seed identification, was mapped to Butterfass Chromosome 2. The M locus, which conditions monogerm seed present in most modern hybrids and obviates the need for thinning stands, was located to Chromosome 4. Nuclear male sterility (locus A), often used in facilitating crosses, has been recently assigned to Butterfass Chromosome 1 (Friesen et al., 2006) and that assignment is confirmed here. The RFLP loci were scored using cDNA clones as probes; as most of these have been sequenced, their Genbank accession numbers are indicated in Supplementary Table 1. The EST–UTR genetic markers are described here for the first time, and could show unique utility for simultaneously mapping genes in families with conserved motifs, as demonstrated here for a calmodulin-containing motif where 14 separate loci with this motif were mapped to eight of the nine linkage groups (Fig. 1 ).
|
Overall, markers per chromosome ranged from 26 (Chromosome 2) to 47 (Chromosome 4). Average distance between markers ranged from 1.1 cM for Chromosome 3 to 2.0 cM on Chromosomes 1 and 7, with an average across all linkage groups of 1.61 cM between markers (SD = 0.31). Segregation of individual markers was tested for consistency for expected Mendelian ratios using the Chi-square statistic, of which 72 of the 331 markers (21.8%) showed distorted segregation ratios. Markers were predominantly skewed in favor of the sugarbeet allele (53/72 = 74%), and 15 were skewed in favor of the table beet allele. Interestingly, three markers on Chromosome 9 and one of Chromosome 8 (Marker No. 291, 314, 321, 326; Table 1) showed an apparent heterozygote disadvantage, while one on Chromosome 7 (No. 245) showed an apparent heterozygote advantage.
Less than 10% of markers of Chromosomes 2, 3, 4, and 6 showed distorted segregation ratios. In contrast, the other chromosomes showed distorted segregations of nearly 20% or more of markers assigned to their respective chromosome [Chromosome 1 (25%); 5 (71.9)%; 7 (19.4%); 8 (29.5%); 9 (30.8%)]. The majority of skewed segregation on Chromosome 1 was toward the table beet allele (Marker No. 8, 13, 18, 19, 21, 22; Table 1) with two toward the sugarbeet parent (No. 2, 14). Interestingly, the R locus on Chromosome 2 (No. 54), which was scored here as a root trait and not hypocotyl color, showed a distorted ratio in favor of the recessive (rr), perhaps the result of unconscious selection due to our interest in sugarbeet improvement. The other Chromosome 2 marker (No. 58) with distortion was in favor of the table parent allele. All skewed segregation of markers on Chromosomes 3, 4, and 6 were in favor of the sugarbeet allele, with the exception of Marker No. 102. Similarly, all distortions on Chromosome 9 favored the sugarbeet allele (excepting the heterozygote disadvantages indicated above), as did all but two on Chromosome 8 (No. 249, 256, and also one heterozygote disadvantaged marker). Distortion on Chromosome 7 was evenly divided among sugarbeet and table beet alleles in excess (No. 223, 243, 248, and 215, 219, 246, respectively, not including No. 245 above). Interestingly, the sugarbeet gene-specific marker (No. 215) for a putative oxalate oxidase involved in enhanced germination (de los Reyes and McGrath, 2003) occurred less frequently than its presumed table beet allele, suggesting an advantage of the table beet allele that could be exploited for germplasm improvement. Distortion was extreme for Chromosome 5, with 23 of 32 markers showing a distorted segregation ratio. All but three skewed markers were in excess from the sugarbeet allele (i.e., No. 150, 151, and 168; Table 1).
Genetic maps have utility for examining transmission of alleles through generations, and associating molecular markers with trait genes; however, all alleles will rarely be segregating in any one desired population of interest. The map presented here is a framework by which additional markers can be located on a common map. Not all alleles mapped in other populations will be segregating here, as was the case for additional published SSR primer sequences available for mapping in this population (Supplementary Table 1), and for this reason large-insert clones may help in discovering cis-linked polymorphisms. To evaluate whether such a strategy could be readily implemented, a BAC library was pooled, and the pools were used to identify BAC clones containing mapped markers from this map (Table 2). In all cases, at least one specific BAC clone was identified that carried a sequence similar to the mapped marker in as few as 44 PCR reactions. Such clones serve as genetic marker anchor points for physical mapping (Fig. 1).
Large-insert clones may help in discovering cis-linked polymorphisms.
View this table:
[in this window]
[in a new window]
Table 2. Bacterial artificial chromosome clones containing mapped markers in the sugarbeet x table beet population, and RGA-containing BAC clones mapped by Hunger et al. (2003).
Conversely, interesting candidate genes may become apparent for which nucleotide sequence is available but no marker has been developed. Discovering a series of cis-linked polymorphisms from large insert clones may allow mapping of the candidate gene in multiple populations. Resistance gene analogs (RGAs) are putatively involved in host plant disease resistance, and primers amplifying 47 sugarbeet RGAs were reported and many were mapped by Hunger et al. (2003). We recovered 31 of these from the BAC library. Two showed genetic polymorphism and were mapped in the sugarbeet x table beet population here (Marker No. 1 and 245; Table 1), where No. 1 was previously unassigned and the location of No. 245 was reported on Chromosome 5 but mapped to Chromosome 7 here, which is plausible because of common sequence motifs and predicted functions. These BAC clones are available for further characterization (Table 2). Further refinement of such a reciprocal genetic–physical mapping strategy will be required to quickly discover useful polymorphisms physically located on such BAC clones for which a nonpolymorphic PCR product has been used, either by sequencing outwards from the PCR primers or by implementing various other mutation scanning methods.
|
Discussion |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionREFERENCES |
|---|
In sugarbeet, clustering is generally observed with anonymous RFLP and RAPD markers. Information is limited on the clustering behavior of AFLP markers, and specific information generated with PstI and MseI restriction enzymes for sugarbeet genetic mapping is anecdotal. In this study, the average number of E/M-derived AFLP scorable bands (45.5 bands PC–1) and the percentage polymorphism (22.9% = 10.4 polymorphic bands PC–1) are in agreement with the results of Hansen et al. (1999) in an exhaustive evaluation of E/M-derived AFLP markers in the genus Beta. Those authors reported an average of 44.3 bands PC–1, of which 27% were polymorphic (12 polymorphic bands PC–1). Schondelmaier et al. (1996) found an average of 61 amplified bands PC–1, of which 50% were polymorphic, only considering four selected E+3/M+3 PCs. Using 16 PCs of the HindIII/MseI restriction enzyme pair, Schafer-Pregl et al. (1999) observed an average of 11 polymorphic bands PC–1, where HindIII is similar to EcoRI in that it is also not senstive to 5-methyl-cytosine. In this study, similar numbers of amplified bands and polymorphisms were also obtained using the PstI/MseI pair using P+2/M+3 PCs. P+3/M+3 PCs were initially tested on the mapping population, but the lower number of amplified bands and polymorphism reduced the efficiency, and these combinations were discarded from further analysis. Overall, the sugarbeet x table beet cross appeared to show an average number of band amplification and polymorphism normally present between sugarbeet lines. Although we had not expected this, it may not be surprising considering table beet is a likely progenitor of sugarbeet (through fodder beet) and that heterozygosity has been maintained in the beet germplasm through out-crossing as open-pollinated populations. Sugarbeet and table beet likely diverged as recently as the 17th century (Biancardi et al., 2005; Draycott, 2006). The effect of selection in sugarbeet during the past 100 yr has not acted to reduce overall genetic diversity, but rather to partition diversity among breeding populations (McGrath et al., 1999). Plus in this case, our table beet parent is perhaps not as widely diverged as most since it carries the self-fertility and CMS (cytoplasmic male sterile) restorer genes introgressed from sugarbeet early after their discovery (Goldman, 1996).
Early sugarbeet genetic maps had large total map lengths with a relatively low number of markers, resulting in a very high intermarker distance average (Barzen et al., 1995; Pillen et al., 1992, 1993). The integration of AFLP and SSR markers increased marker density but did not significantly increase the length of the genetic maps (Rae et al., 2000; Schondelmaier et al., 1996; Schumacher et al., 1997). In this study, the uniformity and relatively high density of markers found on each chromosome and the expected number of linkage groups observed indicate a general confidence in this coverage of the B. vulgaris genome. The relatively high proportion of unlinked AFLP markers may suggest that a fraction of the genome is not represented by this genetic map, and that protein-encoding genes are vastly underrepresented, and current efforts are geared to improving the density of markers in gene-rich regions. Other good indicators of the general quality of this genetic map is that each chromosome-specific marker was correctly mapped to one of the nine chromosomes, including the three previously located morphological markers. Marker coverage of each chromosome was also relatively uniform (less than 2x variation), with the number of markers per chromosome roughly equivalent perhaps because of the similarity in size of B. vulgaris karyotyped chromosomes (Bosemark and Bormotov, 1972; de Jong et al., 1985; Nakamura et al., 1991).
Segregation distortion is common in sugarbeet. Wagner et al. (1992) and Pillen et al. (1992, 1993) found that approximately 15% of their markers showed distorted segregation ratios, which were attributed to lethal loci present on six linkage groups. Barzen et al. (1992, 1995) found that 19.3% of markers distributed on eight linkage groups showed segregation distortion, and attributed the causes for this high proportion to the presence of lethal loci, of structurally abnormal chromosomes, and to gametic self-incompatibility (SI), for which four loci have been described in sugarbeet (Larsen et al., 1977). These two maps were combined, extended, and correlated with the Butterfass chromosome nomenclature by Schumacher et al. (1997). Segregation distortion was not uniform between the maps; only Chromosome 3 showed no distortion, and our Chromosome 3 had only a single distorted marker (No. 96) distal on one end where, in general, distortions appear more frequently. For all other chromosomes, except in two cases, the proportion of distorted segregation ratios ranged from 0 to 21.1% in those mapping populations and generally occurred in clusters within a linkage map, but the clusters were not shared between maps, excepting perhaps one end of Chromosome 7 where all three maps show linkage of segregation distortions. Two exceptions showed entire linkage group segregation distortions; 77.3% of Chromosome 1 markers deviated from expectation in the Barzen-derived map and 96.2% of Chromosome 5 markers deviated in the Pillen-derived map (Schumacher et al., 1997). In our population, four chromosomes had distorted marker proportions > 25%, (i.e., Chromosomes 1, 5, 8, and 9).
Distribution of linkage distortion may be more instructive than an overall level of distorted segregation value. For instance, a modestly high frequency of marker segregation distortion (22%) was observed in this study, and overall, segregation distortion showed a basic trend to favor the sugarbeet (female) parent's alleles. Unfortunately, the phase is not reported in previous maps, so the direction of allelic selection is not yet comparable. Reciprocal crosses would be particularly instructive in the case that gross segregation distortions were a consequence of maternally inherited states (e.g., cytosine methylation, imprinting, nuclear-cytoplasmic incompatibility). Evidence suggests that Chromosome 5 is particularly vulnerable to distortion in sugarbeet x table beet crosses. Our Chromosome 5 showed a strong tendency for preferential transmission of the sugarbeet configuration (female), and another sugarbeet x table beet population showed the opposite with the male (table beet) Chromosome 5 configuration preferentially inherited (V. Laurent, 2006, unpublished data). The Pillen-derived sugarbeet map showing extreme distortion of Chromosome 5 was derived by selfing a single F1 individual similar to these table beet maps (whereas the Barzen-derived mapping occurred in the F1 between heterozygous parents), suggesting selfing may play an undefined role. Curiously, Abe et al. (1993) described inheritance of an as-yet-unmapped isozyme locus that showed distorted segregation in progeny from self-incompatible x self-compatible sugarbeet crosses, but not in progeny from self-incompatible crosses.
Segregation distortion in our population is unlikely due to segregation of lethal or sublethal alleles, but rather due to an undefined genetic discordance between sugarbeet and table beet. The cross here was homozygous for the dominant self fertility (SF, Owen 1942) allele, so SI per se is unlikely to explain the segregation distortion we observed in the F2 generation. The table beet parent used here is homozygous at nearly 100% of its loci (data not shown) due to strong inbreeding during its development as a widely used CMS-maintainer line for table beet breeding, and the sugarbeet parent used here has reduced heterozygosity relative to most sugarbeets (McGrath et al., 1999). It should be noted that, due to SF, we have considered distortions as a favorable outcome of one parent or the other's allele, recognizing that it is equally probable for selection against a lethal allele, which may be more intuitive in many cases; however, such alleles likely have been purged from these parents. Positive selection for certain alleles in the homozygous state can be considered as a possibility, especially since four of five codominant loci showing distorted segregation appeared at a disadvantage for the heterozygote. Additional segregation analyses are needed for crosses involving each of the major crop types (chard, sugarbeet, fodder, table beet, wild beet) to test such hypotheses, and the availability of a common chromosome nomenclature and a set of common SSR markers will facilitate future comparisons.
The utility of this population, in conjunction with locating SSRs to BAC clones, will accelerate development of a physical map for beets, and will assist in developing SSR markers and other polymorphisms at physical locations in the genome. With these resources, a gene of interest needing confirmation via genetic co-segregation with a trait of interest, but lacking polymorphism in this population, can be mapped after recovering the gene-containing BAC clone, sequencing all or some of the clone until a putative SSR or SNP is identified, and testing that marker for segregation. End sequencing of BAC clones in this library is underway with the expectation that many of these sequences will carry useful SSR markers, and thus contribute to improving the coverage and resolution of this genetic mapping resource.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionREFERENCES |
|---|
Received for publication May 24, 2006.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and MethodsResultsDiscussionREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
