Agriculture Reference
In-Depth Information
CURRENT PROGRESS IN QTL ANALYSIS
AND DEPLOYMENT OF MAS
before planting. The actual plants used in cross-
ing or backcrossing can be directly assayed for
desirable traits (Kuchel et al., 2005, 2007a). In
some programs, breeders have combined MAS
and DH techniques to improve wheat. Parents are
screened before DH production, and the haploid
plants are screened prior to chromosome dou-
bling to generate populations of DH plants that
are fi xed for the gene(s) of interest. Parents can
be assayed for a single-gene trait (e.g.,
Lr39
, Gold
et al., 1999; Glu-1Bx7, Radovanovic and Cloutier
2003) to ensure the DH population will include
the trait(s) of interest.
The effectiveness of early-generation selection
with markers could at times be compromised due
to poor marker reliability (i.e., RAPDs) or exces-
sive linkage distances (Kuchel et al., 2007a). At
present, microsatellite markers are most com-
monly used in molecular mapping studies. Mic-
rosatellite markers are highly polymorphic and
are reliable and repeatable. Microsatellite markers
are amenable to high-throughput data collection,
including 96-well-plate DNA extraction, 384-
well PCR assays, as well as capillary electropho-
resis systems. Microsatellite linkage maps provide
a reliable framework that can be used in an ongoing
fashion to map traits of interest, including both
Mendelian and quantitatively inherited traits
(McCartney et al., 2005b, 2006; Quarrie et al.,
2005; Fofana et al., 2007).
It is sometimes stated that wheat breeding is as
much art as science. Through controlled hybrid-
ization followed by selfi ng, wheat breeders gener-
ate new genetic variability. Selection of inbred
lines derived from crosses can lead to novel germ-
plasm and new cultivars. However, traditional
breeding methods are laborious, time-consuming,
and costly. In Canada, approximately 10 genera-
tions are required to develop wheat lines suitable
for cultivar evaluation. Selection for desirable
traits is required during pureline development to
ensure that cultivars possess the traits necessary
for adoption by producers. The modern era of
molecular linkage mapping began in the 1980s
(Botstein et al., 1980). Twenty years later, high-
density molecular linkage maps have been devel-
oped for most major cereal crops, including wheat
(Somers et al., 2004). Molecular mapping and
marker development for agronomic, disease, and
quality traits (i.e., McCartney et al., 2005a, 2006)
now provide robust tools that can be used to
effectively and effi ciently select for traits of
interest.
Single-gene traits and complex traits
Robust markers have been developed for many
important traits such as resistance to leaf rust,
stem rust (caused by
Puccinia graminis
Pers.:Pers.
f. sp.
tritici
Eriks. & E. Henn.), or common bunt
(caused by
Tilletia tritici
[syn.
T. caries
]), and
insect resistance, protein content, and glutenin
alleles. A list of genes and traits is provided in
Table 14.4, which is not intended to be exhaus-
tive. The use of MAS for important single traits
is desirable, because phenotypic selection for
certain traits requires destructive tests. Other
traits are expensive to measure such as grain
cadmium content (Penner et al., 1995). The eval-
uation of quality traits requires a minimum grain
sample that impedes quality analyses in early gen-
erations. Molecular marker assays are largely free
of such constraints (Radovanovic and Cloutier
2003; Kuchel et al., 2007a). Selection can be con-
ducted on single seeds to ensure traits are present
Recurrent selection
Since SSRs are abundant on the wheat genetic
map and high-throughput genotyping technology
is available, the prospect of performing genome-
wide recurrent genotype selection is made possi-
ble. For example, in a backcross breeding strategy,
the restoration of the recurrent parent could be
accelerated by selection of the desired alleles at a
modest number of loci across the genome. Somers
et al. (2005) demonstrated this approach by select-
ing at 40 to 70 loci in different backcross popula-
tions and restoring 95% of the recurrent genetic
background in two cycles of backcrossing. Without
marker assistance, the same level of restoration
would require at the minimum three to four back-
crossing cycles.
