We recently reported rice promoters that are active in late stages of pollen development. However, rice promoters that allow manipulation of gene expression at earlier stages of pollen development are still very limited to date. In this study, we have chosen 10 putative microspore promoters, OsMSP1 through OsMSP10, based on publicly available transcriptomic datasets in rice (Oryza sativa L.). Sequence analysis of these promoter regions revealed some cis regulatory elements involved in pollen-specific expression. We also examined promoter activities using the promoter-GUS reporter constructs in both transgenic rice and Arabidopsis. In rice, all of the 10 promoters directed GUS signals from the microspore stage throughout the all stages of pollen development. In addition, while GUS signals from 4 promoters, OsMSP2, OsMSP7, OsMSP9 and OsMSP10, seem to be expressed preferentially during pollen development, those from other six promoters were observed in vegetative tissues such as leaves, stems, and roots of seedlings. Similarly, in Arabidopsis, all of the 10 promoters directed GUS signals during pollen development. In detail, 8 promoters, OsMSP1 ~ OsMSP8 directed GUS signals from the microspore stage, whereas 2 promoters, OsMSP9 and OsMSP10, exhibited GUS signals from tricellular stage. Furthermore, seven promoters, except for OsMSP1, OsMSP2 and OsMSP10, showed GUS signals in shoot apical region or root tissues of seedlings. Furthermore, we verified microspore activity of four promoters, OsMSP1, OsMSP2, OsMSP3 and OsMSP6, by complementation analysis of the sidecar pollen (scp) mutant which displays microspore-specific defects. Currently, further analyses are underway for GUS expression of T2 generation in transgenic rice and scp complementation with remaining promoters.

Tissue-specific promoters are a very useful tool for manipulating gene expression in a target tissue or organ; however, their range of applications in other plant species has not been determined, to date. In this study, we identified two late pollen-specific rice promoters (ProOsLPS10 and ProOsLPS11) via meta-anatomical expression analysis. We then investigated the expression of both promoters in transgenic rice (a homologous system) and Arabidopsis (a heterologous system) using ProOsLPS10 or ProOsLPS11::GFP-GUS constructs. As predicted by microarray data, both promoters triggered strong GUS expression during the late stages of pollen development in rice, with no GUS signals detected in the examined microspores and sporophytic tissues. Interestingly, these promoters exhibited different GUS expression patterns in Arabidopsis. While in Arabidopsis, the OsLPS10 promoter conferred GUS expression at the uni- and bi-cellular macrospore stages, as well as at the shoot apical region during the seedling stage, the OsLPS11 promoter was not active in the pollen at any stage, or in the examined sporophytic tissues. Furthermore, by performing a complementation analysis using a sidecar pollen (scp) mutant that displays developmental defects at the microspore stage, we found evidence that OsLPS10, which can be an applied promoter expressed in Arabidopsis, is useful for directing gene expression in the early stages of pollen development. Our results indicate that the OsLPS10 and OsLPS11 promoters can drive the expression of target genes during the late stages of pollen development in rice, but not in Arabidopsis. Our results also emphasize the necessity of confirming the applicability of an established promoter to heterologous systems.

The correct development of male gametophytes (pollen grains) in flowering plants is essential for proliferate in gamete production. Here we have taken a map-based cloning approach using Arabidopsis male gametophytic mutant, named gemini pollen3 (gem3) to identify and characterize key gene that is expressed gametophytically for the completion of microgametogenesis focusing on genes which control cell division and cell fate determination. Previously reported gem1 and gem2 mutants with similar characteristics to gem3 that are disturbed at asymmetric division and cytokinesis at pollen mitosis I (PMI) in Arabidopsis. However, gem3 was mapped to a different genetic locus, and pollen developmental analysis revealed that gem3 exert an effect at meiosis and mitosis causing complete sterility. We also discovered that gem3 homozygous lines produce aberrant pollen grains, arising from incomplete cytokinesis during male meiosis with sporophytic phenotypes of twisted-shape leaves, large flowers. This mutation shows reduced genetic transmission of gem3 allele through male gametophyte. In previous results, the gem3 locus was confirmed by mapping to the region located on chromosome 5. To further confirm strong candidate gene, we performed sequencing and genetic complementation analysis. Currently, we are performing functional studies of the gem3 gene for the better understanding of molecular mechanisms that control asymmetric division at meiosis and mitosis during pollen development.

OsLPS is pollen specific gene that express at late stage of pollen development in rice. Based on microarray database, promoter region of two genes Os03g0106900 and Os03g0106500 were identified. The sequence of 2287bp and 2468bp upstream region of these genes were amplified and designated as OsLPS10 and OsLPS11. These promoters were fused with GUS-GFP reporter gene in a destination vector, pKGWFS7 and introduced into rice (Dongjin cultivar) and Arabidopsis (Col-0). The results of GUS assay showed different pattern of gene expression in pollen of rice and Arabidopsis. In Arabidopsis, the OsLPS10 gene strongly activated in young anther and not expressed in mature pollen. Pollen development analysis revealed GUS expression was detected at unicellular stage and strongest at the bicellular pollen developmental stage. No GUS signal was recorded in mature pollen. In case of OsLPS11, no GUS signal was detected in during pollen development of inflorescent. By contrast, in rice, the GUS expression pattern of OsLPS10 and OsLPS11 exhibited similar. GUS expression was first detectable in the anthers of spikelets at the bicellular stage and intensity increased in tricellular and mature pollen. The GUS signal was not detected in the anthers in unicellular microspores in both genes, OsLPS10 and OsLPS11. The results suggested that these genes were different activity in heterologous plant system, monocot and dicot. Complementation analysis and Cis-regulatory elements will be examined to illuminate the characteristic of these genes

Based on the results of microarray analysis we selected ten candidate genes that express in pollen at the early pollen developmental stage. By PCR amplification, the promoter region of these genes were amplified from rice genomic DNA (Nipponbare) and cloned into the destination pKGWFS7 vector via an entry vector, pDONR201. The characteristic of promoters were evaluated in Arabidopsis thaliana (Col-0) through GUS expression analysis. Fifty T2 plants respectively from each promoter were tested. Whole inflorescence of individual plant was stained with 1mM X-Gluc solution to observe tissue-specific GUS expression patterns. The results showed that all 10 promoters activated in pollen tissues. Among them six promoters expressed at the early developmental stage (unicellular) of pollen and the others expressed at both early (unicellular) and late pollen developmental stage (mature pollen). The results indicated that these promoters would be potential applicable for the studies of pollen function. Currently, we are performing these promoters analysis in rice transgenic plants as well as molecular characterization.

Organ size control is a fundamental developmental processes for higher plants as well as a promising target trait for molecular breeding in crop plants. Genetic mechanisms how plant organs grow to a certain size remains still unclear. Here we present the identification and characterization of a genetic mutant, big flower1-1 (bif1-1) in Arabidopsis that exhibits bigger organ size primarily due to increased cell size. Genetic analysis indicated that it is a single, semi-dominant mutation. Phenotypic analysis showed that bif1-1 exerts pleiotropic effects: it caused bigger seed size, bigger seedling, bigger leaf, thicker stem, increased trichome branching, smaller fruit, and bigger pollen. Microscopic analysis suggested that the bigger organ size in bif1-1 mutant is primarily attributed to increased cell size. Gene expression analysis indicated that most of growth-control genes tested were not altered in bif1-1 mutant. Instead, expression of ARGOS and auxin-inducibility of ANT were reduced in bif1-1 mutant. Our ongoing positional on the corresponding gene would not only shed light on the molecular mechanisms how plants adopt final organ size but also provide a promising genetic resource for genetic engineering of flower- and seed-size in crop plants.

Toward molecular understanding of flower senescence/abscission, we have identified a mutant, designated as dea1-1D (dealyed abscission1-1D), with delayed flower senescence/abscission syndrome from activation-tagged pools. Phenotypic analysis revealed pleiotropic effects of dea1-1D mutation including delayed flowering as well as smaller serrated leaves. Genetic analysis showed that it is a dominant mutation. Molecular analysis on the flower senescence syndrome indicated that dea1-1D might define novel regulatory branch of flower abscission, controlling expression of ethylene-responsive AP2 transcription factor. On the contrary, triple responses was not affected by dea1-1D mutant. Though the penetrance was not complete, the mutant phenoytpes was shown to be tightly linked with the T-DNA selection marker, BASTA-resistance. We identified the T-DNA insertion site through molecular cloning of the T-DNA flanking genomic DNA and found that a neighboring gene was overexpressed in the dea1-1D mutant. Together with gene expression analysis, we will discuss possible function of DEA1 during flower senescence and abscission.