Polar patterning of the spikelet is disrupted in the two opposite lemma mutant in rice

ABSTRACT

Angiosperms produce diverse flowers and the pattern of floral symmetry is a major factor for flower diversification. Bilaterally symmetric flowers have evolved multiple times in different angiosperm lineages from radially symmetric ancestors. Whereas most monocots produce radially symmetric flowers, grasses such as rice (Oryza sativa) and maize (Zea mays) generate bilaterally symmetric flowers and spikelets. In this paper, we focused on the two opposite lemma (tol) mutant, which displays a pleiotropic phenotype in the spikelet. Close morphological examination revealed that a typical spikelet phenotype of the tol mutant was principally based on the mirror image duplication of the lemma-side half of the spikelet. Other spikelet phenotypes can be explained as the derivation from the spikelet with this duplication. A polar pattern of organ formation along the lemma-palea axis was disrupted by this duplication. Accordingly, tol mutation seems to change the spikelet from bilateral symmetry (monosymmetry) to disymmetry. Thus, the tol mutant provides good genetic material to investigate the regulation of spikelet symmetry in rice.

INTRODUCTION

Elucidation of the molecular mechanisms underlying flower and inflorescence development is one of the central issues in plant developmental biology. The ABC model that explains the principle of floral organ specification is a great milestone in flower development research, and a number of studies to understand further the detailed mechanisms of flower development have been performed based on this model (Coen and Meyerowitz, 1991; Lohmann and Weigel, 2002; Prunet and Jack, 2014). Although floral shapes are diversified in angiosperms, the ABC model is applicable in principle to not only eudicot flowers but also monocot flowers. Among monocots, the molecular mechanisms of flower development are relatively well understood in rice (Yoshida and Nagato, 2011; Hirano et al., 2014; Tanaka et al., 2014). Although B class genes specify stamen and lodicule (petal homolog), the YABBY gene DROOPING LEAF (DL), which is not included in the ABC model, plays a crucial role in carpel specification in rice (Nagasawa et al., 2003; Yamaguchi et al., 2004; Yao et al., 2008). These findings suggest that some modifications to the original ABC model are required to explain rice flower development.

In grasses such as rice and maize, floral organs are formed in a unique inflorescence unit, the spikelet (Yoshida and Nagato, 2011; Hirano et al., 2014; Tanaka et al., 2014). The spikelet consists of floral organs such as lodicules, stamens and carpels, and of organs specific to the spikelets such as a lemma, a palea, sterile lemmas and rudimentary glumes. This is in contrast with Arabidopsis flowers, which are generated directly on the inflorescence axis. Development and morphogenesis of the spikelet organs are regulated by many genes (Hirano et al., 2014; Tanaka et al., 2014). For example, the DEPRESSED PALEA1 (DP1) gene encoding an AT-hook DNA binding protein regulates proper development of the central region of the palea (Jin et al., 2011), whereas the G1 gene encoding a plant-specific nuclear factor specifies sterile lemma identity by repressing lemma identity (Yoshida et al., 2009). The TRIANGULAR HULL1 (TH1) gene, belonging to the same gene family as G1, is involved in structural fine-tuning of the lemma and palea (Sato et al., 2014). Although Arabidopsis has homologs of these genes, there is no report that they are involved in flower development in Arabidopsis.

The arrangement of floral (spikelet) organs also differs between rice and Arabidopsis. Floral organs such as sepals, petals and stamens are formed in concentric whorls in Arabidopsis, and the size of each floral organ is similar (Lohmann and Weigel, 2002; Prunet and Jack, 2014). Thus, Arabidopsis generates radially symmetric flowers. By contrast, in the rice spikelet, the lemma and palea, which are different in their morphology, are formed on the opposite sides in an alternating manner (Yoshida and Nagato, 2011; Hirano et al., 2014; Tanaka et al., 2014). In addition, two lodicules are distributed asymmetrically on the lemma side of the spikelet (see below for details; Fig. 1). Thus, the rice spikelet displays bilateral symmetry with a single plane (transverse plane) along the lemma-palea axis. In other words, the rice spikelet has polarity along the lemma-palea axis, which is associated with the potential to form different glumes (lemma or palea) and to initiate (or not to initiate) lodicules. Genetic regulation of floral symmetry is well studied in the flower of Antirrhinum majus, and several key genes responsible for dorsoventral asymmetry of the flower have been identified (Busch and Zachgo, 2009; Preston and Hileman, 2009). However, our understanding of the mechanism of floral symmetry in monocots is poor.

Fig. 1.

Wild-type spikelets. (A and B) A wild-type spikelet. Part of the lemma and palea is removed in (B). (C and D) A cross section of a wild-type spikelet. A close-up view of the marginal regions of the lemma and palea is shown in (D). The arrowhead indicates the thin margin of the palea. (E) Schematic representation of the wild-type spikelet. Green and orange indicate the lemma and the palea, respectively. The horizontal broken line indicates a plane of symmetry. Gray broken lines indicate concentric whorls. Bars = 2 mm in (A) to (C), and 0.5 mm in (D). ca, carpel; le, lemma; lo, lodicule; pa, palea; sl, sterile lemma; st, stamen.

In this paper, we focused on a new spikelet mutant, two opposite lemma (tol). Although the tol mutant generated various abnormal spikelets, close examination suggests that these abnormalities are associated with the disruption of the polarity along the lemma-palea axis (transverse axis), and that the tol spikelet has two symmetry planes. Thus, the tol mutant should be good genetic material to elucidate the mechanism of spikelet symmetry in rice.

MATERIALS AND METHODS

Plant materials

The tol mutant was identified among mutants that had been roughly selected from the TOS17 mutant panel (Miyao et al., 2007). Nipponbare was used as the wild-type strain for comparing phenotypes and in situ expression analysis.

Morphological observation

To observe spikelet phenotype, photographs of spikelets were taken using an SZX10 stereomicroscope and DP21 camera (Olympus, Tokyo, Japan). A BX50 optical microscope (Olympus) and an Axio Cam HRc camera (Carl Zeiss, Oberkochen, Germany) were used for observation of cross sections of spikelets.

In situ hybridization

A chimeric plasmid carrying part of the DL cDNA (Yamaguchi et al., 2004) was used to make a DL probe. RNA was transcribed with T7 RNA polymerase and labeled with digoxigenin (Roche, Mannheim, Germany). Mature spikelets before anthesis were fixed with PFA solution for about 16 h at 4 ℃. After dehydration and substitution with xylene, the samples were embedded in Paraplast Plus (McCormick Scientific, St. Louis, MO) and sectioned at 8-μm thickness by a rotary microtome. Hybridization and immunological detection of the hybridized transcripts were performed according to the procedure described by Kouchi and Hata (1993).

RESULTS

Structure of the wild-type spikelet in rice

In the wild-type spikelet, a lemma and a palea enclose the floral organs, such as lodicules, stamens and carpels (Fig. 1, A) (Yoshida and Nagato, 2011; Hirano et al., 2014; Tanaka et al., 2014). Two lodicules are formed asymmetrically only on the lemma side (Fig. 1, B and E). Although the lemma and palea are similar to each other at a glance, several differences are observed in their morphology (Ohmori et al., 2009). First, marginal regions are thick and curved inward in the lemma, while they are thin and mostly straight in the palea (Fig. 1, C and D). Second, the abaxial surface of the margin is rough in the lemma, while it is smooth in the palea. Third, the palea is formed in the same whorl as the lodicules, whereas the lemma is formed outside the palea whorl (Fig. 1, E). The lemma and palea are formed in an alternate phyllotactic pattern such that lemma primordium initiates earlier than palea primordium. Because the lemma and palea thus differ in shape and the lodicules are located asymmetrically, the rice spikelet has only a single plane of symmetry (bilateral symmetry), along its transverse axis (Fig. 1, E).

Characteristic features of the spikelets in the tol mutant

The tol mutant showed several abnormalities in the lemma and palea. Because tol spikelet phenotypes are pleiotropic, we roughly classified them into three types: twin lemma (TL)-type, reduced lemma (RL)-type and loss of lemma (LL)-type.

The TL-type spikelet generated an extra lemma-like organ, which replaced the palea (Fig. 2, A, G and K). Observation of a cross section of the TL-type spikelet showed that this extra organ had thick and curled margins with a rough surface, suggesting that it corresponded to the lemma (Fig. 2, G and H). Two thin glume-like organs, named lateral glumes, were detected in the lateral region of the spikelet (Fig. 2, B, G and K). The lateral glumes had thin and straight margins, the surface of which was smooth (Fig. 2, H), and they were formed in the lodicule whorl (Fig. 2, K). These characteristics of the lateral glume are similar to those of the palea, suggesting that the lateral glume has partial identity with the palea. Two extra lodicules were also formed on the side of the extra lemma (Fig. 2, B and K). Thus, the number of lemmas and lodicules doubled. These observations suggest that, in the TL-type spikelet, the palea-side half of the spikelet was replaced by the lemma-like half of it, and two palea-like organs were ectopically formed in the lateral region. The overall arrangement of the spikelet organs (two lemmas and two lateral glumes) and floral organs suggests that the TL-type spikelet has two planes of symmetry, that is, is disymmetric (Fig. 2, K).

Fig. 2.

Spikelet phenotypes in the tol mutant. (A and B) TLc-type spikelet of the tol mutant. (C and D) RL-type spikelet of the tol mutant. (E and F) LL-type spikelet of the tol mutant. Part of the lemma is removed in (B, E and F). Close-up views of the basal region of the spikelets in (C and E) are presented in (D and F). (G and H) A cross section of the TLc-type spikelet. Close-up view of the marginal regions of the lemma and lateral glume is shown in (H). The arrowhead indicates a thin and straight margin of the lateral glume. (I) A cross section of the ectopic lemma of TLi-type spikelet. The arrowhead indicates a thin and straight margin of the ectopic lemma. (J) A cross section of the RL-type spikelet. The region of the original lemma side is shown. (K–N) Schematic representation of each spikelet type. Green and orange indicate organs with lemma and palea identity, respectively. Horizontal and vertical broken lines indicate planes of symmetry. Gray broken lines indicate concentric whorls. (O) Frequency of the number of stamens in wild type and tol mutant spikelets. “Normal” means tol spikelets without any abnormality in the lemma or palea. “tol-type” includes TL-, RL- and LL-type spikelets. Bars = 2 mm in (A) to (C), (E) and (G), 1 mm in (D), (F) and (I and J), and 0.5 mm in (H). le, lemma; lg, lateral glume; lo, lodicule; rl, reduced lemma; sl, sterile lemma.

We found that some TL-type spikelets had a thin and straight margin in the extra lemma (Fig. 2, I and L). This phenotype suggests that transformation of the palea into the lemma is incomplete. Thus, the TL-type spikelet was divided into two subtypes: a TLc-subtype spikelet with a complete extra lemma and a TLi-subtype spikelet with an incomplete extra lemma.

In the RL-type spikelet, a normal-looking lemma and a small organ opposite it were formed inside the sterile lemma (Fig. 2, C and D). The small organ had curled margins and a rough surface at the abaxial side, suggesting that it had characteristics of the lemma (Fig. 2, J). A close-up view indicated that this small organ was initiated outside the normal lemma (Fig. 2, D). Therefore, the original lemma seems to be reduced to a small lemma-like organ (reduced lemma), whereas the normal-looking lemma is likely to be an ectopic lemma. In addition, two extra lodicules and two lateral glumes were formed in the RL-type spikelet, like the TL-type spikelet (Fig. 2, M). These observations suggest that the RL-type spikelet is a modified version of the TL-type spikelet, in which the size of the original lemma is reduced.

In the LL-type spikelet, only one lemma-like organ was formed (Fig. 2, E and N). This organ was ectopically formed at the side where the palea would be in wild type, judging from the phyllotactic pattern. However, the original lemma and lodicules, which were formed opposite the ectopic lemma in the TL-type spikelet, were absent. Instead of the original lemma, the LL-type spikelet generated an extra sterile lemma (Fig. 2, F). Two lateral glumes were formed in the LL-type spikelet, as they were in the TL- and RL-type spikelets (Fig. 2, E and N). These phenotypes suggest that the LL-type spikelet is also a modified version of the TL-type spikelet, in which the development of the organs at the original lemma side was inhibited.

Taking these observations together, three spikelet types in the tol mutant showed partial similarity to each other. We propose that the RL- and LL-type spikelets are derived from the TL-type spikelet, arising from the duplication of the lemma-side half of the spikelet and extra formation of the palea-like lateral glumes. The frequency of each spikelet type is indicated in Table 1.

Table 1. Frequency of spikelet types in the tol mutant

Phenotype	No. of spikelets	%
TL-type	78	31.8
TLc-subtype	(56)
TLi-subtype	(22)
RL-type	60	24.5
LL-type	7	2.9
Normal*	100	40.8
Total	245	100

*tol  spikelet without abnormality in the lemma or palea.

Analysis of the identity of the extra lemma by in situ hybridization

To check the identity of the extra lemma in TL-type spikelets and the putative reduced lemma in RL-type spikelets, we analyzed expression of the DL gene in these organs. DL is expressed in the central region of the leaf primordia and induces midrib formation in the leaf (Yamaguchi et al., 2004; Ohmori et al., 2011). In the spikelet, this gene was expressed in the region surrounding the midvein of the lemma (Fig. 3, A), as described previously (Toriba and Hirano, 2014). No expression was detected in the palea. Thus, DL is a good molecular marker to check lemma identity.

Fig. 3.

In situ hybridization analysis of DL gene expression in the spikelet. (A) A wild-type spikelet. (B) TLc-type spikelet of the tol mutant. (C) TLi-type spikelet of the tol mutant. (D) RL-type spikelet of the tol mutant. Bars = 80 μm. le, lemma; lg, lateral glume; lo, lodicule; pa, palea; rl, reduced lemma.

In situ analysis clearly indicated that DL was expressed in the extra lemma in TL-type spikelet (both TLc- and TLi-subtypes) (Fig. 3, B and C). DL was also expressed in the central region of the reduced lemma in RL-type spikelet (Fig. 3, D). These observations confirmed that the extra lemma of the TL-type spikelet and the reduced lemma of the RL-type spikelets had lemma identity.

Analysis of floral organs

The above analyses indicated that the number of lemmas and paleas/palea-like organs was increased in the tol spikelets except for the LL-type spikelet. We next examined the number of stamens and carpels in tol mutants. The proportion of spikelets exhibiting increased stamen number was more than 60% among spikelets showing the tol phenotype, whereas such spikelets were about 10% in normal spikelets of the tol mutant (Fig. 2, O). This suggests that the regulation of normal development of the lemma/palea is associated with that of stamen number. In contrast to stamen number, carpel number was less affected in the tol mutant: the proportion of spikelets having more than two carpels was only a few percent in both the tol-type and normal spikelets (3.5 and 2%, respectively), suggesting that the tol mutation affects stamen and carpel development independently.

DISCUSSION

In this paper, we analyzed spikelet phenotypes in the tol mutant. tol spikelets showed pleiotropic phenotypes not only in the shape and arrangement of the lemma and palea (spikelet organs) but also in those of the floral organs such as stamens and carpels. We here mainly focused on the spikelet phenotypes and classified them into three types. According to this classification, the morphological analysis suggests that pleiotropic spikelet phenotypes are continuous and are derived from a primary defect observed in the TL-type spikelet. This primary defect can be regarded as the duplication of the lemma-side half of the spikelet, with successive reduction in the number of cells on the original lemma side giving rise to reduction or loss of the original lemma and lodicules. Because of the duplication of the lemma-side half, the TL-type spikelet acquires a new plane of symmetry, the medial plane, which is perpendicular to the transverse plane (Fig. 2, K). This suggests that the putative gene responsible for the tol mutation is involved in polar pattern formation along the transverse axis.

In general, the number of floral organs increases in the mutant, in which the floral meristem is enlarged. The number of lemmas, paleas/palea-like organs and lodicules was increased in the tol spikelets. Stamen number was also increased to some extent. Thus, it is plausible that defects in the regulation of the number of spikelet and flower organs in the tol mutant result from an enlargement of the spikelet/flower meristem.

A mutation in the rice FLORAL ORGAN NUMBER1 (FON1) or FON2 gene causes an enlargement of the meristem, resulting in an increase in the number of floral organs, such as carpels and stamens (Suzaki et al., 2004, 2006). FON1 and FON2 are orthologs of Arabidopsis CLAVATA1 (CLV1) and CLV3 (Clark et al., 1997; Fletcher et al., 1999; Ha et al., 2010). Both the FON and CLV genes negatively regulate stem cell proliferation, and plants having defects in these genes show a stronger increase in carpel number than in stamen number. By contrast, carpel number was less affected in the tol mutant than stamen number. Therefore, roles in meristem size regulation may be different between the tol and fon mutants: a failure in meristem size regulation seems to occur at an early stage of spikelet development in the tol mutant and may be caused by a defect in an unknown mechanism, which is unrelated to the regulation of stem cell proliferation. Apart from this difference, the fon1 and fon2 mutants exhibit spikelets similar to the TL-type spikelet at a low frequency (Suzaki et al., 2004; our unpublished observations). This fact supports the idea that the formation of the TL-type spikelet is associated with meristem size.

Based on the idea that tol spikelet formation is associated with meristem size, we propose a model that accounts for the three types of tol spikelets (Fig. 4). We hypothesize that the spikelet meristem is enlarged from initial stages of spikelet development. The enlargement of the meristem may be associated with the loss of polarity along the transverse axis and may change the spikelet shape from monosymmetry to disymmetry. Space for the initiation of extra organs would be sufficient in this enlarged meristem. Whereas the lemma and the lodicules could be formed on both sides, palea-like organs would be initiated in the lateral side of the meristem. Thus, the TL-type spikelet may be formed. In some cases, the activity of the meristem to supply cells for lemma and lodicule development may be reduced at the original lemma side. Thus, the RL- and LL-type spikelets would result from this partial inactivation of the meristem. Observation of spikelet meristems in the tol mutant by scanning electron microscopy should provide good evidence to test these inferences.

Fig. 4.

A model of tol spikelet formation. Green and orange areas in the meristem indicate the regions where organs with lemma and palea identity, respectively, are initiated. The blue area indicates the region where meristematic activity is partially reduced in the RL- and LL-type spikelets. Broken lines indicate planes of symmetry. L and P indicate the lemma side and the palea side, respectively.

The genetic mechanism underlying floral symmetry is well understood in Antirrhinum. Several genes responsible for either dorsal or ventral identity and their coordinated action specify the asymmetric flower in Antirrhinum (Busch and Zachgo, 2009; Preston and Hileman, 2009). Floral asymmetry in Antirrhinum is mainly related to the size and shape of petals. Although the number of petals and normal stamens is increased in the Antirrhinum cycloidea (cyc) mutant, rearrangement of the floral organs does not occur (Luo et al., 1996). By contrast, the number of flower and spikelet organs was increased in the rice tol mutant, and their arrangement was altered: the lemma and lodicule were formed on the side where the palea is formed in wild type, and the palea-like organs were initiated at the lateral regions. Therefore, disruption of asymmetric patterning of the flower/spikelet seems to be different between these rice and Antirrhinum mutants. Isolation of the gene responsible for the tol mutation and elucidation of its function should reveal a new mechanism underlying the regulation of flower/spikelet symmetry.

We previously reported the fickle spikelet1 (fsp1) mutant, which shows a pleiotropic phenotype in the spikelet (Suzuki et al., 2015). One of the spikelet types in fsp1 produces an extra lemma at the palea side and two lateral palea-like organs. This phenotype is similar to that of the TL-type of the tol mutant. However, because the phenotypes of other types of the fsp1 spikelet differ from those of the RL- and LL-type of the tol spikelet, the gene responsible in each mutant seems to be different. Unlike the tol mutant, the fsp1 mutant has an increased number of carpels, suggesting that the regulation of stem cell proliferation is compromised. Transverse polarity is reportedly disturbed in the retarded palea1 (rep1) mutant of rice, which has a mutation in a gene similar to Antirrhinum cyc. However, unlike the tol spikelet, the spikelet in the rep1 mutant is not disymmetric (Yuan et al., 2009).

We are trying to isolate a gene responsible for the tol mutant phenotypes by a map-based cloning strategy using F2 plants from a cross between the tol mutant and an indica strain, Kasalath. Although we could map the tol locus to the long arm of chromosome 1, we found that the frequency of plants showing the tol phenotype was very low (about 7%) in this F2 population. This suggests that the tol phenotypes are not caused by a single recessive gene. Alternatively, the effect of the tol mutation may be masked by the genetic background of the indica strain. Thus, it remains to be clarified whether the tol mutation affects a single gene.

ACKNOWLEDGMENTS

We thank Drs. A. Miyao and H. Hirochika for kindly providing us with mutant seeds. We thank A. Takahashi for technical assistance. We also thank Dr. K. Nonomura and Mr. M. Eiguchi at the National Institute of Genetics, and technicians at the Institute for Sustainable Agro-Ecosystem Services at the University of Tokyo, for cultivating the rice used in our study. This research was supported in part by a Grants-in-Aid for Scientific Research from MEXT (25113008 to H.-Y. H.) and by the NIG Collaborative Research Program (2014−A76, 2015-A1-60).

REFERENCES

•  Busch,  A., and  Zachgo,  S. (2009) Flower symmetry evolution: towards understanding the abominable mystery of angiosperm radiation. BioEssays 31, 1181–1190.
•  Clark,  S. E.,  Williams,  R. W., and  Meyerowitz,  E. M. (1997) The CLAVATA1 gene encodes a putative receptor kinase that controls shoot and floral meristem size in Arabidopsis. Cell 89, 575–585.
•  Coen,  E. S., and  Meyerowitz,  E. M. (1991) The war of the whorls: genetic interactions controlling flower development. Nature 353, 31–37.
•  Fletcher,  J. C.,  Brand,  U.,  Running,  M. P.,  Simon,  R., and  Meyerowitz,  E. M. (1999) Signaling of cell fate decisions by CLAVATA3 in Arabidopsis shoot meristems. Science 283, 1911–1914.
•  Ha,  C. M.,  Jun,  J. H., and  Fletcher,  J. C. (2010) Shoot apical meristem form and function. Curr. Top. Dev. Biol. 91, 103–140.
•  Hirano,  H.-Y.,  Tanaka,  W., and  Toriba,  T. (2014) Grass flower development. In: Flower Development - Methods and Protocols (eds.: Riechmann, J. L., and Wellmer, F.), pp. 57–84. Springer, New York.
•  Jin,  Y.,  Luo,  Q.,  Tong,  H.,  Wang,  A.,  Cheng,  Z.,  Tang,  J.,  Li,  D.,  Zhao,  X.,  Li,  X.,  Wan,  J., et al. (2011) An AT-hook gene is required for palea formation and floral organ number control in rice. Dev. Biol. 359, 277–288.
•  Kouchi,  H., and  Hata,  S. (1993) Isolation and characterization of novel nodulin cDNAs representing genes expressed at early stages of soybean nodule development. Mol. Gen. Genet. 238, 106–119.
•  Lohmann,  J. U., and  Weigel,  D. (2002) Building beauty: The genetic control of floral patterning. Dev. Cell 2, 135–142.
•  Luo,  D.,  Carpenter,  R.,  Vincent,  C.,  Copsey,  L., and  Coen,  E. (1996) Origin of floral asymmetry in Antirrhinum. Nature 383, 794–799.
•  Miyao,  A.,  Iwasaki,  Y.,  Kitano,  H.,  Itoh,  J.,  Maekawa,  M.,  Murata,  K.,  Yatou,  O.,  Nagato,  Y., and  Hirochika,  H. (2007) A large-scale collection of phenotypic data describing an insertional mutant population to facilitate functional analysis of rice genes. Plant Mol. Biol. 63, 625–635.
•  Nagasawa,  N.,  Miyoshi,  M.,  Sano,  Y.,  Satoh,  H.,  Hirano,  H.-Y.,  Sakai,  H., and  Nagato,  Y. (2003) SUPERWOMAN1 and DROOPING LEAF genes control floral organ identity in rice. Development 130, 705–718.
•  Ohmori,  S.,  Kimizu,  M.,  Sugita,  M.,  Miyao,  A.,  Hirochika,  H.,  Uchida,  E.,  Nagato,  Y., and  Yoshida,  H. (2009) MOSAIC FLORAL ORGANS1, an AGL6-like MADS box gene, regulates floral organ identity and meristem fate in rice. Plant Cell 21, 3008–3025.
•  Ohmori,  Y.,  Toriba,  T.,  Nakamura,  H.,  Ichikawa,  H., and  Hirano,  H.-Y. (2011) Temporal and spatial regulation of DROOPING LEAF gene expression that promotes midrib formation in rice. Plant J. 65, 77–86.
•  Preston,  J. C., and  Hileman,  L. C. (2009) Developmental genetics of floral symmetry evolution. Trends Plant Sci. 14, 147–154.
•  Prunet,  N., and  Jack,  T. P. (2014) Flower development in Arabidopsis: there is more to it than learning your ABCs. In: Flower Development - Methods and Protocols (eds.: Riechmann, J. L., and Wellmer, F.), pp. 3–33. Springer, New York.
•  Sato,  D.-S.,  Ohmori,  Y.,  Nagashima,  H.,  Toriba,  T., and  Hirano,  H.-Y. (2014) A role for TRIANGULAR HULL1 in fine-tuning spikelet morphogenesis in rice. Genes Genet. Syst. 89, 61–69.
•  Suzaki,  T.,  Sato,  M.,  Ashikari,  M.,  Miyoshi,  M.,  Nagato,  Y., and  Hirano,  H.-Y. (2004) The gene FLORAL ORGAN NUMBER1 regulates floral meristem size in rice and encodes a leucine-rich repeat receptor kinase orthologous to Arabidopsis CLAVATA1. Development 131, 5649–5657.
•  Suzaki,  T.,  Toriba,  T.,  Fujimoto,  M.,  Tsutsumi,  N.,  Kitano,  H., and  Hirano,  H.-Y. (2006) Conservation and diversification of meristem maintenance mechanism in Oryza sativa: function of the FLORAL ORGAN NUMBER2 gene. Plant Cell Physiol. 47, 1591–1602.
•  Suzuki,  C.,  Tanaka,  W., and  Hirano,  H.-Y. (2015) Analysis of rice fickle spikelet1 mutant that displays an increase in flower and spikelet organ number with inconstant expressivity. Genes Genet. Syst. 90, 181–184.
•  Tanaka,  W.,  Toriba,  T., and  Hirano,  H.-Y. (2014) Flower development in rice. In: The Molecular Genetics of Floral Transition and Flower Development (ed.: Fornara, F.), pp. 221–262. Elsevier, Amsterdam.
•  Toriba,  T., and  Hirano,  H.-Y. (2014) The DROOPING LEAF and OsETTIN2 genes promote awn development in rice. Plant J. 77, 616–626.
•  Yamaguchi,  T.,  Nagasawa,  N.,  Kawasaki,  S.,  Matsuoka,  M.,  Nagato,  Y., and  Hirano,  H.-Y. (2004) The YABBY gene DROOPING LEAF regulates carpel specification and midrib development in Oryza sativa. Plant Cell 16, 500–509.
•  Yao,  S.-G.,  Ohmori,  S.,  Kimizu,  M., and  Yoshida,  H. (2008) Unequal genetic redundancy of rice PISTILLATA orthologs, OsMADS2 and OsMADS4, in lodicule and stamen development. Plant Cell Physiol. 49, 853–857.
•  Yoshida,  A.,  Suzaki,  T.,  Tanaka,  W., and  Hirano,  H.-Y. (2009) The homeotic gene long sterile lemma (G1) specifies sterile lemma identity in the rice spikelet. Proc. Natl. Acad. Sci. USA 106, 20103–20108.
•  Yoshida,  H., and  Nagato,  Y. (2011) Flower development in rice. J. Exp. Bot. 62, 4719–4730.
•  Yuan,  Z.,  Gao,  S.,  Xue,  D.-W.,  Luo,  D.,  Li,  L.-T.,  Ding,  S.-Y.,  Yao,  X.,  Wilson,  Z. A.,  Qian,  Q., and  Zhang,  D.-B. (2009) RETARDED PALEA1 controls palea development and floral zygomorphy in rice. Plant Physiol. 149, 235–244.