Environmental Engineering Reference
In-Depth Information
21.7 suGarcane ImProvement tools: marker-
assIsted BreedInG and transGenIcs
Sugarcane improvement has been achieved through the classical route for over 2 centuries but it is
a costly and slow process. Currently, it takes around 12 years for a new cultivar to be released. One
strategy to speed up this process is the use of genetic maps and molecular markers through marker-
assisted breeding. Another strategy is the use of transgenics.
21.7.1 g
EnEtic
m
apS
and
m
olEcular
m
arkErS
Most genetic designs used for the construction of genetic linkage maps use populations derived
from crosses between inbred lines (e.g., using backcross or selfing). The statistical and genetics
methods used in this case are already established and implemented in several softwares, such as
Mapmaker/EXP. However, to obtain such strains for sugarcane is impractical, especially because of
the large inbreeding depression that occurs when selfing occurs. In this case, the mapping popula-
tions are F1 generations obtained from crosses between non-inbred individuals (Lin et al. 2003).
For sugarcane (and several other species for which there are no available inbred lineages, such
as fruit and eucalyptus), an alternative that was widely used in the past is called
double pseudo-
testcross
. This strategy is the construction of two individual maps (one for each parental), by the
polymorphic identification of single dose markers for each parental (Grattapaglia and Sederoff 1994;
Porceddu et al. 2002; Shepherd et al. 2003; Carlier et al. 2004). On the basis of this approach, link-
age maps for
S. officinarum
(“LA Purple”) and
S. robustum
(“Mol 5829”) were constructed using
random amplified polymorphic DNA (RAPD), amplified fragment length polymorphism (AFLP)
and restriction fragment length polymorphism (RFLP) markers in a single dose (Guimarães et al.
1999). Other situations were also considered, but rarely the kind of crossing used involved commer-
cial parents (Al-Janabi et al. 1993; da Silva et al. 1993; Sobral and Honeycutt 1993; da Silva et al.
1995; Grivet et al. 1996; Hoarau et al. 2001). However, from both a biological and statistical point of
view, the integration of the information contained in these individual maps into a single integrated
map is desirable. This can only be done with the presence of heterozygote markers between parents,
which are used to establish linking relationships between the markers individually segregated in
each parental (Barreneche et al. 1998; Wu et al. 2002; Garcia et al. 2006; Oliveira et al. 2007).
The construction of integrated genetic maps, using different types of molecular markers with
different segregation, has great advantages because it allows the linking map to be saturated and
extends the characterization of polymorphic variation throughout the genome. Codominant markers
may be useful to bring together cosegregation groups in their respective homology groups, specifi-
cally for polyploid species such as sugarcane (da Silva et al. 1993; Grivet et al. 1996). Moreover,
the more precise localization of QTL is helped by the availability of an integrated genetic map
(Maliepaard et al. 1997). However, there may be different numbers of alleles, in heterozygote
parentals for each segregating loci, turning linkage analysis more complex, because the linkage
phases in the parentals may be unknown a priori, making it difficult to detect recombination events
(Maliepaard et al. 1997; Wu et al. 2002).
Wu et al. (2002) proposed a statistical method on the basis of maximum likelihood analysis that
allows the simultaneous estimation of recombination fractions and linking phases between loci in
mapping populations derived from crossings between non-inbred individuals (F
1
generation). This
method works toward the construction of an integrated genetic map, which is the result of the com-
bination of various pieces of information generated by different types of molecular markers, whose
information content varies. Garcia et al. (2006) and Oliveira et al. (2007) constructed integrated
genetic maps, consisting of 357 markers distributed throughout 131 co-segregation groups from
the crossing between two precommercial sugarcane cultivars (SP80-180 × SP80-4966). The results
were better than those obtained when the same data were analyzed using the JoinMap program,
which indicates, in this case, better efficiency in the estimation of linking and linking phases of the
