Molecular Plant Breeding 2016, Vol.7, No.12, 1
-
19
2
associated with grain width, and 167 QTLs reported to be associated with grain weight. The QTLs for grain
thickness, however, was relatively less reported (Huang et al. 2013).
Among the above-mentioned QTLs, 13 genes associated with grain size and shape, including several major
QTLs, such as
GS3
(Fan et al., 2006, 2009; Mao et al., 2010),
GW2
(Song et al., 2007),
GW5
(Shomura et al.,
2008),
GS5
(Li et al., 2011), and
GW8
(Wang et al., 2012) have been isolated by map-based cloning strategy.
Some of them have been functionally characterized at genetic, biochemical, and cell biological levels. A major
QTL for grain length and weight,
GS3
, was first fine mapped and isolated using a population derived from
crossing between MH63 and Chuan 7, and further was functionally characterized as negative regulator for grain
length (Fan et al., 2006, 2009; Mao et al., 2010). A mutation in MH63
GS3
allele caused a 178 aa-truncation in
the C-terminus of its predicted protein results in the formation of longer grain rice (Fan et al. 2006, 2009; Mao
et al., 2010).
GW2
, a major QTL for grain width initially identified in cross between a large grain cultivar, WY3, and a
small-grain cultivar, FENGAIZHAN-1, encodes a RING-type E3 ligase of proteasomes, and involves in
targeting its substrates to proteasome-regulated proteolysis. A 1-bp deletion in the allele of
GW2
in WY3
resulting in a premature stop in its exon 4 in protein translation, causes large and width grain in WY3. In cell
biological level,
GW2
negatively regulating cell division. Loss function of
GW2
accelerates cell division and
increases cell number in spikelet hull of grains (Song et al., 2007).
GW5
, another gene for grain width, encodes
a 144-aa nuclear protein with an arginine-rich domain. It has been shown that
GW5
physically interacts with
polyubiquitin and it is likely to act as a component of proteasome pathway to regulate the cell division of outer
glume of rice spikelets (Shomura et al., 2008). The deletion in this gene may have been the target for selection
of large grain rice during rice domestication (Weng et al., 2008; Shomura et al., 2008; Wan et al., 2008).
GS5
, a
major QTL for grain width, grain filling, and grain weight, encodes a serine carboxypeptidase and acts as a
positive regulator of grain size (Li et al., 2011).
GW8
, third major QTL for grain width and grain yield, was
isolated from a cross between HJX74 and Basmati385, and being characterized as SQUAMOSA
promoter-binding protein-like 16, referred to as
OsSPL16
, which belongs to the family of SPB-domain
transcription factor. A mutation in the promoter of Basmati385
GW8
allele results in the slender grain shape of
Basmati385 (Wang et al., 2012).
The molecular genetic studies of the rice grain size in the literature show that rice grain size traits are controlled
by complicated genetic networks. Many different biochemical and physiological pathways involve in the
formation of the diverse grain sizes and shapes (Heang and Sassa, 2012). Although more than 10 genes have
been isolated and functionally characterized in the past two decades, many genes associated with rice grain size
still poorly understood. Also it is expected that many QTLs still need to be identified. Thus, in this study, a F
2
population derived from the cross between a large grain
indica
rice variety, Nanyangzhan (hereafter, referred to
as NYZ) and a medium grain
indica
rice variety, Ce253 was used to identify QTLs for grain length, width,
thickness, length to width ratio, and grain weight of rice. The objective of this study was to identify novel QTL
loci for grain size in the NYZ and Ce253 genetic background to facilitate the understanding of complex genetic
regulatory network of grain size comprehensively.
1 Results
1.2 Phenotypic performance of the parents and F
2
population
In phenotypic evaluation, twenty randomly chosen fully filled grains from each rice plant were used for measuring
grain length, grain width, grain thickness, and calculating length to width ratio (hereafter, referred to as GL, GW,
GT, LWR, respectively). One hundred grains from each genotype was weighted and converted to 1000-grain
weight, and hereafter referred to as TGW. The performances of grain size related traits showed significant
differences between two parents respectively (Figure 1). Although the grains of two parents are exhibited long
shape, the length, width, thickness, and TGW of grains of NYZ were distinct from those of Ce253 (Figure 1; Table
1). There was no significant difference observed in LWR between NYZ (3.68) and Ce253 (3.97) in student T-test
