Each IL generally contains only a small percentage of the donor genome (less than 5%), however these in-trogressions often influence several traits, including some undesirable ones. Therefore, it is often necessary to fine map QTL within an IL to reduce the length of the donor introgression. This allows not only the assessment of whether the effect on the phenotype is due to a single QTL or to several linked QTLs affecting the same trait, but also the verification of whether possible undesirable effects are caused by linkage drag of other genes or by pleiotropic effects of the selected QTLs (Eshed and Zamir 1995; Monforte and Tanksley 2000b; Monforte et al. 2001; Fridman et al. 2002; Frary et al. 2003b; Yates et al. 2004). Besides reducing linkage drag, the development of lines with smaller introgressions (sub-ILs) allows the identification of molecular markers more tightly linked to the QTL of interest, which can be used for MAS.
By substitution mapping it was possible, for example, to break linkages between poor yield, low fruit weight and high solids in a S. habrochaites IL mapping to the bottom of chromosome 1 (Monforte and Tanksley 2000b) or between orange fruit color and high sugars in a S. chmielewskii introgression (Frary et al. 2003b), and to distinguish between linkage drag and pleiotropy for several S. habrochaites QTL alle-les mapped to the bottom of tomato chromosome 4
(Monforte etal. 2001). This same region of chromosome 4 has been more finely mapped using a series of lines containing small overlapping introgressions from S. peruvianum and S. habrochaites (Yates et al. 2004). The results show that QTLs for soluble solids content, fruit weight and stem scar are not allelic between the two wild species, which suggests that it may be possible to combine the S. habrochaites and the S. peruvianum alleles in a single line with the potential of obtaining improved lines characterized by extremely high soluble solids content.
ILs have proven to be invaluable starting material for the positional cloning of key genes underlying quantitative traits (Frary et al. 2000; Fridman et al. 2000, 2004; Yano etal. 2000; El-Din-El-Assal etal. 2001; Takahashi et al. 2001; Kojima et al. 2002). In tomato, one such QTL is fw2.2, a major fruit-size QTL that is believed to have played a major role during the domestication of this crop (Alpert and Tanksley 1995; Frary et al. 2000; Tanksley 2004). Natural genetic variation at this locus alone can change the size of fruit up to 30%, with the cultivated tomato allele contributing to this increase in fruit size (Frary et al. 2000). Cloning offw2.2 has shown that the locus codes for a repressor of cell division mainly during the cell division phase of fruit development (Frary et al. 2000; Conget al. 2002). Large-fruit alleles offw2.2 are associated with a higher mitotic index during the cell division stage just after anthesis (Cong et al. 2002). FW2.2 is homologous to other plant proteins, but none of them has a known biological function. Interestingly, comparative sequencing offw2.2 locus in the genus Solanum showed that the fruit-weight phenotype was associated with variation in a few nucleotides in the promoter region rather than in the coding region (Nesbitt and Tanksley 2002), and the natural variants at the promoter of fw2.2 were correlated with subtle changes in transcript levels as well as in the timing of gene expression (heterochronic allelic variation) (Cong et al. 2002).
Another example is given by Brix9-2-5, a S. pen-nelli QTL that increases sugar yield of tomato, that was mapped to a SNP in a gene encoding a flower- and fruit-specific apoplastic invertase (LIN5), which operates in sugar transport to the developing fruit (Fridman et al. 2000). QTL analysis representing five different tomato species delimited the functional polymorphism of Brix9-2-5 to a single amino acid near the catalytic site of the invertase crystal, a point mutation that can alter enzyme kinetics and fruit sink strength (Fridman et al. 2004). Therefore, this study has also highlighted the power of using multi-species
IL resources for high-resolution analysis of complex phenotypes.
The use of ILs, which isolates a single QTL region, transformed the task of QTL cloning into one similar to that performed for simple Mendelian traits, with the exception that phenotyping requires more detailed measurements. Although the strategy of delimiting a QTL to a single gene using genetic approaches is extremely powerful and unbiased, it is still a time-consuming and technically demanding process (Frid-man et al. 2000, 2004). Any additional information that could be associated with the observed traits in the ILs would therefore be useful in identifying the allele(s) responsible for a particular phenotype. Gen-omic technologies and methodologies that enable integration of the genetic components of QTL variation in genomic databases can help to accelerate the rate of QTL discovery. Furthermore, integrated strategies can reduce the list of candidate genes for target QTL (Wayne and McIntyre 2002).
Along these lines, the S. pennellii IL population has been used to test the potential of the candidate gene approach to identify candidate genes for QTL influencing the intensity of tomato fruit color (Liu et al.
2003), tomato fruit size and composition (Causse et al.
2004), and ascorbic acid content in fruit (Stevens et al. 2007). In all studies QTLs were mapped for the quantitative traits of interest along with the mapping analysis of genes encoding, respectively, enzymes of the carotenoid biosynthetic pathway, enzymes involved in the fruit primary carbon metabolism, and enzymes for the ascorbic acid synthesis and turnover pathways. While Causse et al. (2004) found a number of clear links between the presence of S. pennellii alle-les of these genes and the observed trait, in the study conducted by Liu et al. (2003) on fruit color the number of QTLs that co-segregated within the same bins that contained the candidate gene was close to the number that was expected by chance alone. In a study conducted by Stevens et al. (2007), QTLs for ascorbic acid content were mapped not only on the S. pennel-lii IL population but also on two additional mapping populations. Among the candidate genes mapped, co-locations with mapped QTLs for ascorbic acid content were found between a monodehydroascorbate reductase (MDHAR) gene and a QTL mapped on bin 9-D, and a GDP-mannose epimerase (GME) gene and a QTL mapped on bin 9-J.
In order to further define the biochemical traits that are altered in each line, Overy et al. (2005) conducted an initial metabolomic profiling of the fruit pericarp of the two parents of the S. pennellii IL
population and of six selected ILs. Principal Component Analysis (PCA) of the metabolite profiles revealed subtle differences in metabolism of the ILs when compared to their parents. A more comprehensive metabolic profiling of the S. pennellii IL population was pursued by Schauer et al. (2006), using a high-throughput gas chromatography-mass spectrometry (GC-MS) metabolite protocol in parallel with whole-plant phenotypic characterization (see Sect. 1.16.2). This approach allowed the identification of 889 single-metabolite QTLs, in addition to many other metabolic QTLs that influenced numerous compounds in a metabolic pathway, and 326 loci that modified yield-associated traits. The analysis indicated that at least 50% of the metabolic loci were associated with QTLs that influenced whole-plant yield-associated traits, and harvest index was identified as a regulator of the metabolite content of the mature fruit pericarp. The observation that plant morphology is a major factor affecting the metabolic composition of fruit at harvest time, suggests that this phenotype might regulate biological processes at various molecular levels.
A transcriptional profiling approach via cDNA mi-croarray analysis was used by Baxter et al. (2005) on six non-overlapping S. pennellii ILs that share the common trait of increased ripe fruit soluble solids content and increased accumulation of fruit carbohydrate. This study provided evidence of genome-wide transcriptional changes and revealed links to mapped QTL and described traits (see Sect. 1.15.3).
Therefore, in the -omics era, ILs provide a new paradigm to improve the efficiency in discovery, candidate gene identification and cloning of target QTL. This can be achieved by combining the results derived from QTL position, DNA sequences, expression profiling data, and functional and molecular diversity analyses of candidate genes (Li et al. 2005b). However, in order to take advantage of the large amount of data that will be generated it is necessary to develop user-friendly bioinformatics management systems that will allow the integration of the entire range of statistical outputs derived from QTL analysis with genome information including gene content, expression and function (Gur et al. 2004).
Was this article helpful?