Molecular Breeding - Genetics, Genomics and Breeding of Crop Plants - page 238

QTL-NILs with those of the recurrent parent allowed the authors to obtain
an accurate evaluation of the effects of the target QTL in an adapted
background. On the other hand, TC-RIL populations may contribute to
identifying QTL without the confounding effect of diverse segregating
backgrounds. Draeger et al. (2006) developed a TC-RIL population fixed for
the branching (controlled by the B gene) character (i.e., an unbranched
population), and were able to identify a QTL affecting seed oil concentration
on LG 10 ( soc10.2* ) linked in repulsion to another B -linked QTL affecting
the same trait ( soc10.1 ). The soc10.2* QTL had not been identified previously
in the RIL analysis per se (Tang et al. 2006a) because its effect was apparently
masked by soc10.1 , or by pleiotropic effects of the B gene. Finally, Huang et
al. (2007) increased the resolution of QTL mapping by performing the
analyses on F 2 segregating populations from crosses between RILs selected
for having favorable alleles for somatic embryogenesis at three QTLs on
LGs 5, 10, and 13, and another RIL having alleles with negative effects at
these QTLs. The number of cross-over events and bins in the QTL regions
was higher in the new F 2 maps, making it possible to localize the QTLs to
smaller intervals.
The ultimate goal, as for traits controlled by major genes, is to develop
molecular markers based on functional genes underlying the QTLs
controlling complex traits. Reverse genetic approaches appear to be
promising for the identification of markers based on genes underlying QTLs
and for enabling the dissection of the genetic basis of complex traits. As a
first approach, functional genomic analyses of candidate genes can be
performed. For example, Poormohammad Kiani et al. (2007) identified QTLs
for water status traits as well as net photosynthesis in a sunflower RIL
population. Four RILs were selected for contrasting response to water stress
and QTL complement and their differential gene expression was studied
for four water-stress associated candidate genes (aquaporin, dehydrin, leafy
cotyledon1-like protein, and fructose-1,6 biphosphatase) under well-watered
and water-stressed conditions. RILs carrying different genomic regions for
some QTLs also had different gene expression patterns for some of the four
genes. This candidate gene approach has also been used to identify genes
underlying QTLs for salt tolerance in wild sunflower hybrids ( H. annuus x
H. petiolaris ). Salt tolerance candidate genes were identified in expressed
sequence tag (EST) libraries of sunflower and 11 were mapped to an existing
QTL framework map from an interspecific BC 2 population. One EST that
coded for a Ca-dependent protein kinase mapped to a salt tolerance QTL
(Lexer et al. 2003a).
For some traits of interest, a priori candidate-gene approaches are not
feasible because a biological model does not exist or the number of candidates
is too numerous that individual follow-ups are prohibitively expensive.
Quantitative expression studies, such as microarray analyses, can reveal