Marker-Assisted Backcrossing Selection for High O/L Ratio in Cultivated Peanut - Translational Genomics for Crop Breeding: Abiotic Stress, Yield and Quality - page 188

and modification of the system depending on the
situation of the breeding program is important.
Another remarkable difference between the
two studies appeared in the breeding scheme. In
this study, backcrossing and selfing were alter-
nately performed, while in the study performed
by Chu and colleagues (2011), three backcrosses
were continuously done first, then two selfings
were performed. From the point of view of short-
ening the breeding period, the scheme used by
Chu and colleagues (2011) was better than that
used in our study. The main reason for alternat-
ing backcrossing and selfing in this study was the
difficulty of performing backcrossing with the
lower temperatures prevalent during off-season
in the greenhouse. In order to compensate for the
disadvantage, GW genotyping was performed to
decrease ARv values across the genome. The
ideal scheme of MABS for high O/L would be
integration of the schemes used in the two stud-
ies: two backcrossings per year with genotyping
of GW and FAD2 loci markers in each gener-
ation, then selfing with the genotyping of GW
markers.

benefits of MABS were not fully tested in this
breeding program for the following reasons. One
was the lack of a sufficient number of GW poly-
morphic markers and information on their posi-
tion on a linkage map. Large differences in the
genomic backgrounds were observed across LGs
in the BC 2 F 1 generation. Although seven LGs
were almost completely replaced by homozy-
gous 'Nakateyutaka' genomes, four LGs still
kept the 'YI-0311' genomes. The utilization of
graphical genotypes would enhance selection
pressure and improve the strategic selection in
each generation. The MABS was very effective
in many cereal crops such as barley for disease
resistance, maize for earliness and yield, rice for
bacterial blight, wheat for powdery mildew, and
so forth (Jefferies et al. 2003; Bouchez et al.
2002; Chen et al. 2001; Zhou et al. 2005). More-
over, the yield of soybean was increased by
using MABs to introgress a yield QTL from a
wild accession into commercial genetic back-
grounds (Concibido et al. 2003). Those species
have now developed thousands of GW polymor-
phic markers and are able to break precisely a
linkage between target genes and unfavorable
genomic regions. As with those crops, recent
advances in molecular genetics in peanut, espe-
cially the increasing number of developed mark-
ers and constructed linkage maps, should enable
a more strategic selection than was attainable in
the present study.
The other factor that limited the efficiency
of MABS was the amount of breeding material
used. Because selection can be done in the early
growth period or in seeds, larger numbers of indi-
viduals can be used in MAS than in conventional
breeding. The higher ratio of genomic regions
derived from the donor genome on the four LGs
in the BC 2 F 1 generation indicates the need for
an increased number of progenies in order to
obtain more recurrent parent specific homozy-
gous genomic regions in the generation. Increas-
ing the amount of breeding material used is a
promising solution for that problem. However,
limitations in facilities and the labor required
for backcrossing and plant management often

Merits of MABS and Future Prospects

In this breeding program, we paid approximately
US$5 per sample for the chemicals used in the
determination of fatty acid content, and two days
of labor were required for the determination.
On the other hand, the cost of the chemicals
and the labor required for the investigation of
genotypes on the two FA D 2 loci were approxi-
mately USD$1 per sample and one day, respec-
tively. Although it was not used in the breeding
program, a direct PCR approach for genotyp-
ing would drive the cost of materials and the
required labor to less than US20 and half a day,

respectively. These figures indicate the efficiency
of MAS for high O/L ratio breeding in peanut,
from the standpoint of breeding costs.
With regard to backcrossing with MAS, the
results in Table 11.1 suggest a higher efficiency
of marker-assisted selection compared to ran-
dom selection. However, we recognize that the

¢