Biology Reference
In-Depth Information
appears to have a different QTL peak at 31-36
Mb, and indeed this is thought to be derived
from the sensitive parent in that population, IR29
(Thomson et al. 2010). Comparison of this puta-
tive QTL with other populations tends to sup-
port the hypothesis of two QTLs in the region,
as QTLs for multiple traits that were not fully
overlapping were also observed in Nona Bokra,
Cheriviruppu, and one FL478 population, while
the “upper” QTL alone was observed in Kala
Rata
chromosome 1 from Pokkali, FL478, and IR64
may be caused by a 1-bp deletion in the cod-
ing region of
OsHKT2.3
, a gene related to other
genes known to increase shoot Na
+
content. The
presence of a functional copy of this gene would
thus counteract the effects of tolerant alleles at
the
Saltol
locus. Extending this further, it can
be hypothesized that specific combinations of
genes may mask the effects of a gene responsi-
ble for a QTL, and thus a QTL may not be seen in
certain genetic backgrounds even when present.
This becomes important when considering the
use of marker-assisted backcrossing, especially
if “anonymous,” non-gene-specific markers such
as flanking SSRs (simple sequence repeats) are
used for this purpose. Recipient lines for such
marker-assisted backcrossing are chosen due to
their lack of the QTL phenotype, i.e., salinity
tolerance, and it is assumed that they lack sig-
nificant loci conferring this phenotype. Markers
are then chosen for their polymorphism between
the donor and recipient lines, and the assumption
is made that the marker haplotype (forward and
flanking markers) from recipient lines denotes
a sensitive haplotype. However, this is not nec-
essarily true. Only those markers polymorphic
between parents are chosen, and the combina-
tion of marker genotypes (the marker haplo-
type) is specific for one or the other parent. But
unless one or more markers is gene-specific, the
marker haplotype of the recipient parent does
not necessarily indicate that the recipient parent
(or progeny that have that haplotype) lacks the
QTL; this is an assumption based on the pheno-
type. It is thus perfectly possible to use MABC
to introgress a QTL into a recipient line that
already contains that QTL, but whose effects
are masked due to genetic background. This
may explain some of the difficulties encountered
in the MABC of
Saltol
into the variety BR28:
the MABC was performed without any prob-
lem, but the resulting BR28-Saltol lines have
limited improvement in salinity tolerance. Pre-
liminary sequence evidence of
OsHKT1.5
from
BR28 suggests that BR28 already contains an
allele identical to that from Nona Bokra, and that
×
×
ZYQ8 populations
(The et al. unpubl.; Gong et al. 2001). This com-
parison also tends to suggest that the lower QTL
is primarily responsible for affecting biomass-
related traits, with possibly secondary effects on
Na
+
concentrations. However, even after taking
into account information from all available popu-
lations, most QTL intervals are still on the order
of 7-10 Mb, relative to the Nipponbare refer-
ence genome. This appears to be largely due to
an inherent limitation in QTL studies: that QTL
limits are defined by both population size and
marker density, and both are typically minimized
to save time and reduce costs. In addition, some
marker types are difficult to map onto the refer-
ence genome, making precise definition of QTL
limits difficult or impossible. Thus, comparison
of QTLs from multiple populations is very pow-
erful in helping to determine the prevalence and
effect of a QTL, but typically cannot narrow QTL
limits sufficiently to allow unambiguous deter-
mination of the underlying genes, aside from the
expense and time required to develop these mul-
tiple populations.
Another complicating factor in the use of
QTLs in plant breeding is the phenomenon of
epistasis, often attributed to and described as
the “genetic background” effect. Epistasis is the
case in which the effect of an allele (
A
) at one
locus (A) masks the effect of genetic variation at
another locus (B) (
B/b
); a different allele at the
first locus (
a
) may unmask the effects at locus
(B). Consensus is growing that epistasis plays a
very important role in determining the extent of
salinity tolerance displayed by rice plants. For
example, a minor QTL around the centromere of
Azucena and JX17
Search WWH ::
Custom Search