"Frontmatter". In: Plant Genomics and Proteomics

CHAPTER
8
I
D E N T I F I C AT I O N A N D
M
A N I P U L AT I O N O F
C
O M P L E X
T
R A I T S
O
VERVIEW
The aim of most of the previous chapters has been to provide knowledge, in
general terms, relating to plant genome structure and function and to
provide a basic understanding of how transcription and translation prod-
ucts are characterized. 
In this chapter problems associated with the identification, isolation, and
manipulation of the genes underlying complex traits are considered. Single
genes that have a very large effect on the phenotype are generally relatively
easily manipulated, even if they are not necessarily as easily isolated. In con-
trast, some traits are genetically very complex and the traits themselves are
difficult to evaluate. Other traits are not difficult to evaluate, but numerous
genes can be involved in their control. In the latter two cases, it may not even
be possible to uncover all those genes by using a single segregating popu-
lation because the full range of alleles underlying the variation in the trait
may not be present in the population. 
Many of the genes in higher plants may not have an obvious phenotypic
effect. This is apparent from the data on insertional mutagenesis in Ara-
bidopsis. Many of the individual plants that have an insertion in a gene do
not show dramatic phenotypic alterations. Part of the explanation may lie in
the presence of multigene families providing redundant functions, in which
case all the copies of that gene family must be silenced before any pheno-
typic effect will be seen (see Chapter 6). Alternatively, the effect of 
Plant Genomics and Proteomics, by Christopher A. Cullis
ISBN 0-471-37314-1 Copyright © 2004 John Wiley & Sons, Inc.
1 4 7

---

many genes may be subtle and require dissection in an appropriate way
before their roles can be adequately described. For example, the genes under-
lying stress tolerance, such as those for cold and drought stress, can be genet-
ically mapped. Many of the loci that contribute to these responses have been
mapped and manipulated in crosses. These loci, called quantitative trait loci
(QTLs), are difficult to evaluate, and the genes that condition their effects are
difficult to isolate. The most popular method for isolating these genes is by
a genome walking strategy after their initial mapping to a specific region of
a chromosome. However, this strategy depends on the availability of many
genomic resources to be successful. Despite the wealth of positional infor-
mation for QTLs in many plant species, very few have actually been isolated.
Therefore, the nature of these QTLs remains largely unknown.
It is possible to detect and locate the loci affecting quantitative traits by
the joint analysis of the segregation of marker genotypes and the phenotypic
values of individuals in appropriate segregating populations. However,
these QTLs are difficult to identify both because of the lack of discrete phe-
notypic segregation and because the phenotypic effects of each gene associ-
ated with a complex trait are usually relatively small. In practice, QTL
analysis involves the selection and hybridization of parental lines that differ
in one or more different quantitative traits (for example, tomato fruit shape,
size, and solid content) (Grandillo et al., 1999) followed by analyses of the
segregating progeny resulting in linkages between the quantitative trait loci
and known DNA markers.
To characterize and manipulate such complex characters a number of
resources are necessary. These include:
∑ High-quality molecular maps
∑ Appropriate crosses and the subsequent generations of segregating
populations to allow mapping of the traits of interest
∑ Resources to determine genotype-by-environment effects of the
various loci, namely, the variation in phenotype of a single genotype
grown under various environmental conditions
∑ Transformation systems to reintroduce the genes, to test whether or
not they have the function expected
∑ Tissue- and/or organ-specific promoters to get the appropriate
expression of the genes in cases where inappropriate expression
might interfere and/or enhance effects. An example is that of the
expression of trehalose in rice, where different expression in various
species gives conflicting results as to the stress protection provided
by this molecule. 
The generation of high-quality molecular maps is also a necessary step
in marker-assisted selection (MAS). MAS is based on the information
retrieved through the application of molecular markers to segregating pop-
ulations that are simultaneously measured for a variety of phenotypic char-
1 4 8
8. I
D E N T I F I C AT I O N A N D
M
A N I P U L AT I O N O F
C
O M P L E X
T
R A I T S

---

acters. The markers can be used to enhance plant breeding efforts and to
speed up the creation of cultivars. Of particular importance is the unmask-
ing and incorporation of interesting wild alleles into elite germplasm.
However, to be effective, MAS requires the availability of many markers so
that the interval between the marker and the gene of interest is small. When
this distance is sufficiently small, subsequent testing for the actual presence
of the gene itself may not be necessary. Most of the currently useful markers
are ones that can be converted into some kind of sequence-tagged sites so
that PCR-based methods for genotyping are available.