A series of chimeric promoters for higher-level expression of foreign genes in plants was constructed as fusions of a gene for beta-glucuronidase (GUS) with the terminator of a gene for nopaline synthase (nos) or of the cauliflower mosaic virus (CaMV) 35S transcript, and the strength of these promoters was assayed in transient and stable expression systems in tobacco and rice. As parts of these promoters, the CaMV 35S core promoter, three different 5'-upstream sequences of the 35S promoter, the first intron of a gene for phaseolin, and a 5'-untranslated sequence (omega sequence) of tobacco mosaic virus were used in various combinations. In tobacco and rice protoplasts, all three fragments of the 35S promoter (-419 to -90, -390 to -90 and -290 to -90, relative to the site of initiation of transcription), the intron, and the omega sequence effectively enhanced GUS activity. Some chimeric promoters allowed levels of GUS activity that were 20- to 70-fold higher than those obtained with the 35S promoter in pBI221. In tobacco protoplasts, the two longer fragments of the 35S promoter were more effective than the shortest fragment. In rice cells, by contrast, the shortest fragment was as effective as the two longer ones. The terminator of the 35S transcript was more effective than that of the nos gene for gene expression. In transgenic tobacco plants, a representative powerful promoter, as compared to the 35S promoter, allowed 10- and 50-fold higher levels of expression on average and at most, respectively, with no clear qualitative differences in tissue- and organ-specific patterns of expression. When the representative promoter was introduced into tobacco with a gene for luciferase, the autofluorescence of detached leaves after a supply of luciferin to petioles was great and was easily detectable by the naked eye in a dark room.

RNA-induced silencing complexes (RISCs) play central roles in posttranscriptional gene silencing. In plants, the mechanism of RISC assembly has remained elusive due to the lack of cell-free systems that recapitulate the process. In this report, we demonstrate that plant AGO1 protein synthesized by in vitro translation using an extract of evacuolated tobacco protoplasts incorporates synthetic small interfering RNA (siRNA) and microRNA (miRNA) duplexes to form RISCs that sequester the single-stranded siRNA guide strand and miRNA strand, respectively. The formed RISCs were able to recognize and cleave the complementary target RNAs. In this system, the siRNA duplex was incorporated into HSP90-bound AGO1, and subsequent removal of the passenger strand was triggered by ATP hydrolysis by HSP90. Removal of the siRNA passenger strand required the ribonuclease activity of AGO1, while that of the miRNA star strand did not. Based on these results, the mechanism of plant RISC formation is discussed.

No abstract

Blast fungus-induced accumulations of major rice diterpene phytoalexins (PA), momilactones A and B, and phytocassanes A through E were studied, focusing on their biosynthesis and detoxification. In resistant rice, all PA started to accumulate at 2 days postinoculation (dpi), at which hypersensitive reaction (HR)-specific small lesions became visible and increased 500- to 1,000-fold at 4 dpi, while the accumulation was delayed and several times lower in susceptible rice. Expression of PA biosynthetic genes was transiently induced at 2 dpi only in resistant plants, while it was highly induced in both plants at 4 dpi. Fungal growth was severely suppressed in resistant plants by 2 dpi but considerably increased at 3 to 4 dpi in susceptible plants. Momilactone A treatment suppressed fungal growth in planta and in vitro, and the fungus detoxified the PA in vitro. These results indicate that HR-associated rapid PA biosynthesis induces severe restriction of fungus, allowing higher PA accumulation in resistant rice, while in susceptible rice, failure of PA accumulation at the early infection stage allows fungal growth. Detoxification of PA would be a tactic of fungus to invade the host plant, and prompt induction of PA biosynthesis upon HR would be a trait of resistant rice to restrict blast fungus.

Pathogenesis-related (PR) proteins have been used as markers of plant defense responses, and are classiWed into 17 families. However, precise information on the majority members in speciWc PR families is still limited. We were interested in the individual characteristics of rice PR1 family genes, and selected 12 putatively active genes using rice genome databases for expressed genes. All were upregulated upon compatible and/or incompatible rice-blast fungus interactions; three were upregulated in the early infection period and four in the late infection period. Upon compatible rice-bacterial blight interaction, four genes were upregulated, six were not aVected, and one was downregulated. These results are in striking contrast to those among 22 Arabidopsis PR1 genes where only one gene was pathogen-inducible. The responses of individual genes to salicylic acid, jasmonic acid, and ethylene induced defense signaling pathways in rice are likely to be diVerent from those in dicot plants. Transcript levels in healthy leaves, roots, and Xowers varied according to each gene. Analysis of the partially overlapping expression patterns of rice PR1 genes in healthy tissues and in response to pathogens and other stresses would be useful to understand their possible functions and for use as characteristic markers for defenserelated studies in rice.