The timing of flowering is determined by endogenous genetic components as well as various environmental factors, such as day length, temperature, and stress. The genetic elements and molecular mechanisms that rule this process have been examined in the long-day-flowering plant Arabidopsis thaliana and short-day-flowering rice (Oryza sativa). However, reviews of research on the role of those factors are limited. Here, we focused on how flowering time is influenced by nutrients, ambient temperature, drought, salinity, exogenously applied hormones and chemicals, and pathogenic microbes. In response to such stresses or stimuli, plants either begin flowering to produce seeds for the next generation or else delay flowering by slowing their metabolism. These responses vary depending upon the dose of the stimulus, the plant developmental stage, or even the cultivar that is used. Our review provides insight into how crops might be managed to increase productivity under various environmental challenges.

In much of the tropics and subtropics, rice (Oryza sativa L.) is grown under long days (LDs). Therefore, LD must play a major role in inducing flowering signal in rice. However, little is known on LD-dependent flowering signal in the species. We previously reported that OsMADS50, which is highly homologous to Arabidopsis SOC1, functions as a positive regulator for flowering. However, its detailed photoperiodic mechanism was not yet elucidated. Here, we report the functional analysis of OsMADS50 and its closely related gene OsMADS56. Knock-out of OsMADS50 caused a late-flowering phenotype only under LD conditions. Overexpression of OsMADS56 (56OX) also resulted in delayed flowering under LD. In the osmads50 mutants and 56OX transgenic plants, transcripts of Ehd1, Hd3a and RFT1 were reduced, although that of OsLFL1 increased. On the other hand, mRNA levels of OsGI, Hd1, OsId1, OsDof12, Ghd7, Hd6 and SE5 were unchanged. These observations imply that OsMADS50 and OsMADS56 function antagonistically through OsLFL1-Ehd1 in regulating LD-dependent flowering. Yeast two-hybrid and co-immunoprecipitation analyses indicated an interaction between those two proteins as well as their formation of homodimers. These results suggest that OsMADS50 and OsMADS56 may form a complex that regulates downstream target genes.

SUMMARYSeed shattering is an important trait that influences grain yield. A major controlling quantitative trait locus in rice is qSH1. Although the degree of shattering is correlated with the level of expression of qSH1, some qSH1-defective cultivars display moderate shattering while others show a non-shattering phenotype. Os05 g38120 (SH5) on chromosome 5 is highly homologous to qSH1. Although we detected SH5 transcripts in various organs, this gene was highly expressed at the abscission zone (AZ) in the pedicels. When expression of this gene was suppressed in easy-shattering 'Kasalath', development of the AZ was reduced and thereby so was seed loss. By contrast, the extent of shattering, as well as AZ development, was greatly enhanced in moderate-shattering 'Dongjin' rice when SH5 was overexpressed. Likewise, overexpression of SH5 in the non-shattering 'Ilpum' led to an increase in seed shattering because lignin levels were decreased in the basal region of spikelets in the absence of development of an AZ. We also determined that two shattering-related genes, SHAT1 and Sh4, which are necessary for proper formation of an AZ, were induced by SH5. Based on these observations, we propose that SH5 modulates seed shattering by enhancing AZ development and inhibiting lignin biosynthesis.

SUMMARYPlants recognize environmental factors to determine flowering time. CONSTANS (CO) plays a central role in the photoperiod flowering pathway of Arabidopsis, and CO protein stability is modulated by photoreceptors. In rice, Hd1, an ortholog of CO, acts as a flowering promoter, and phytochromes repress Hd1 expression. Here, we investigated the functioning of OsCOL4, a member of the CONSTANS-like (COL) family in rice. OsCOL4 null mutants flowered early under short or long days. In contrast, OsCOL4 activation-tagging mutants (OsCOL4-D) flowered late in either environment. Transcripts of Ehd1, Hd3a, and RFT1 were increased in the oscol4 mutants, but reduced in the OsCOL4-D mutants. This finding indicates that OsCOL4 is a constitutive repressor functioning upstream of Ehd1. By comparison, levels of Hd1, OsID1, OsMADS50, OsMADS51, and OsMADS56 transcripts were not significantly changed in oscol4 or OsCOL4-D, suggesting that OsCOL4 functions independently from previously reported flowering pathways. In osphyB mutants, OsCOL4 expression was decreased and osphyB oscol4 double mutants flowered at the same time as the osphyB single mutants, indicating OsCOL4 functions downstream of OsphyB. We also present evidence for two independent pathways through which OsPhyB controls flowering time. These pathways are: (i) night break-sensitive, which does not need OsCOL4; and (ii) night break-insensitive, in which OsCOL4 functions between OsphyB and Ehd1.