Genetic linkage maps correspond to the linear order of molecular markers in a specific genome. Linkage maps are constructed by following the segregation of molecular markers in a population and placing them in linear order based on pair-wise recombination frequencies. The occurrence of many polymorphisms over the total genome is a highly desirable characteristic for a mapping population. Backcross, F2, and recombinant inbred lines are the most commonly used populations for molecular mapping. In outcrossing and self-incompatible species, pseudo F1 populations are used to construct genetic linkage maps.

Constructing genetic linkage maps of tall fescue is impaired by genome complexity, heterozygosity within clones, a high level of self-incompatibility, and lack of morphological genetic markers (Xu et al., 1991). The two-way pseudo-testcross procedure is considered an efficient way to construct genetic linkage maps in outcrossing species like tall fescue (Ritter et al., 1990; Hemmat et al., 1994). In the pseudo-testcross procedure, a mapping population is developed by hybridizing two unrelated highly heterozygous parents to produce a set of F1 progeny. High levels of marker segregation are observed in most of these populations. However, construction of linkage maps is complicated because one or both parents may be heterozygous at a certain locus, markers may be dominant or codominant, and the linkage phase of marker alleles usually is unknown (Maliepaard et al., 1997). Most of the forage grass species are polyploids and outcrossing; thus, genomic information for many forage species is quite limited when compared with other major crop species.


The first genetic linkage map of tall fescue was generated from an F2 population using 108 RFLP markers (Xu et al., 1995). The mapping population was constructed by crossing HD28-56, a plant with high forage quality, to a selected plant of KY-31, the most widely grown cultivar in the United States. A high level of marker polymorphism was observed in the population. Polymorphic probes detected multiple segregating fragments. Two- and three-point linkage analysis was performed using MAPMAKER software. The RFLP marker data in the population indicated clear and strong evidence of disomic inheritance, meaning that tall fescue behaved genetically like a diploid organism that contained two sets of chromosomes. Earlier cytogenetic (Sleper, 1985) and isozyme studies (Lewis et al., 1980) also suggested disomic inheritance in tall fescue. All these results supported the conclusion that tall fescue is an allohexaploid.

Ninety-five markers were used to construct a tall fescue genetic linkage map (Xu et al., 1995). The map covered 1274 centimorgans (cM) on 19 LGs with an average of five loci per LG, and a marker density of 17.9 cM per marker. (Map distances are calculated on the basis of recombination events; one cM equals 1% recombinant offspring.) Thirteen markers remained unlinked. Markers that segregated at more than one locus allowed identification of linkage groups that belonged to the same homeologous (i.e., base chromosome) group. Five of seven homeologous groups were identified. Linkage groups 15 to 19 were not placed in any of the homeologous groups. Several LGs were assigned to the three genomes (PG1G2) using genome-specific probes. The genome-specific probes hybridized with one or occasionally two of the three genomes in tall fescue.

AFLP and SSR Map

The first PCR-based genetic linkage map of tall fescue was constructed with AFLP and SSR markers (Saha et al., 2005). Conserved grass EST-SSR markers were developed from the EST sequences conserved across different cereal grass species (Kantety et al., 2002). Tall fescue and conserved grass EST-SSRs, genomic SSRs from Festuca × Lolium hybrids, and AFLP markers were used to construct the parental maps, followed by biparental consensus maps (Fig. 21-1). A two-way pseudo-testcross mapping strategy was followed for map construction. Seven hundred seventy-three AFLP and 343 microsatellite markers were used to construct the linkage maps. The AFLP markers contributed considerably to genome coverage and to providing links to associate distantly located SSR loci. The majority of markers segregated from either parent and showed a 1:1 Mendelian segregation ratio. However, 12% of the markers segregated from both parents and showed a 3:1 segregation pattern. Markers present in both parents and showing a 3:1 segregation ratio were useful for identifying homologous groups between maps (Maliepaard et al., 1997). The female (HD28-56) map included 558 loci placed in 22 LGs and covered 2013 cM of the genome. The male (R43-64) map comprised 579 loci grouped in 22 LGs with a total map length of 1722 cM. The marker density in the two maps varied from 3.61 cM (female parent) to 2.97 (male parent) cM per marker. A biparental consensus map was constructed on the basis of common markers in parental maps. The consensus map covered 1841 cM on 17 LGs, with an average of 54 loci per LG, and had an average marker density of 2.0 cM per marker. Six of the seven predicted homeologous groups were identified. The female parental map length was greater than the male parental map. A distinctly reduced level of recombination was found in the male parent compared to the female parent. Markers in general were evenly distributed throughout the genome. However, clustering of markers in some regions and gaps of 10 to 21 cM in some LGs were also evident. These results indicated that recombination events were not evenly distributed throughout the genome.

High levels of segregation distortion (23% of total markers) were observed in the cross between two distantly related, heterozygous tall fescue genotypes. Segregation distortion has been reported frequently in tall fescue (Xu et al., 1995; Saha et al., 2005) and other grass species (Warnke et al., 2004; Brummer et al., 1993; Wang et al., 1994). Distorted markers segregated from both parents of the HD 28-56 ´ R43-64 mapping population and were present in most of the linkage groups. The distribution of skewed markers in both parental maps indicated that both the male and female gametophytes and/or sporophytes were involved in segregation distortion. Clustering of markers showing significant segregation distortion was observed in four of the LGs of the consensus map. The AFLP and microsatellite based maps (Saha et al., 2005) were a substantial improvement over the RFLP-based map (Xu et al., 1995) in both map coverage and marker density.

Microsatellite Map

Recently, a substantial number of microsatellite and sequence tagged site (STS) primers (>1800 primer pairs) has been developed for tall fescue. The same mapping population used for the construction of the AFLP-SSR map has been used to construct the biparental and consensus maps. This map has been used to identify molecular markers associated with traits of interest and to map quantitative trait loci (QTL), regions of the genome that contain a cluster of genes influencing a quantitative trait. This is helpful because it indicates that these traits are conditioned by a large number of genes.


Fig. 21-1. Parental linkage groups (LG) of HD-8 (HD28-56 map) and R-7 (R43-64 map) along with the integrated group 2-C of a tall fescue genetic linkage map (Saha et al., 2005). The cM distance scale is at the extreme left. Bars indicate length of each LG. Marker names are placed at the right side of each bar. The AFLP markers are prefixed with a P or E. Tall fescue and conserved grass EST-SSRs are indicated as NFFA and CNL, respectively. The Festuca × Lolium genomic SSRs begin with a B. The last three digits of each marker indicate the size in base pairs.




<--Previous       Back to Top      Next-->