Expression of a <Emphasis Type="Italic">Bacillus subtilis</Emphasis> protoporphyrinogen oxidase gene in rice plants reduces sensitivity to peroxidizing herbicides

Protoporphyrinogen oxidase (Protox) in the porphyrin pathway is the target site of the peroxidizing herbicides such as carfentrazone-ethyl and oxyfluorfen. In an attempt to develop herbicide-resistant plants, transgenic rice plants were generated via expression of herbicide-insensitive Bacillus subtilis Protox gene fused to the transit sequence for targeting to the plastid using Agrobacterium-mediated gene transformation. Homozygous transgenic rice lines of T₃ generation selected by hygromycin resistance test were examined if they are resistant to the herbicides carfentrazone-ethyl and oxyfluorfen. The homozygous transgenic lines had single copy insertion of B. subtilis Protox gene into their genomes and express its mRNA. Compared to wild-type rice, the transgenic lines were less susceptible to the herbicides when examined with respect to growth, electrolyte leakage, chlorophyll loss and lipid peroxidation. The in vitro Protox activities in transgenic lines were about 56 % higher than those in wild-type rice. With 10 µM concentration of the herbicides in the enzyme assays, Protox activities in transgenic lines were similar to those in non-inhibited wild-type rice. Less amount of protoporphyrin IX was accumulated in transgenic lines than in wild-type rice upon the treatment of the herbicides at 10 µM concentration. Our results indicated that expression of B. subtilis Protox gene was stably transmitted into T₃ rice plants and reduced their sensitivity to carfentrazone-ethyl and oxyfluorfen.This work was supported by Ministry of Agriculture and Forestry of Korea and Agricultural Plant Stress Research Center (grant No. R11-2001-09203000-0) funded by Korea Science and Engineering Foundation.     

The plastidic Arabidopsis protoporphyrinogen IX oxidase gene, with or without the transit sequence, confers resistance to the diphenyl ether herbicide in rice

Agrobacterium‐mediated gene transformation was used to introduce plastidic protoporphyrinogen IX oxidase (Protox) genes from Arabidopsis, with and without the transit sequence, into the rice genome. They were placed under the control of the constitutive and ubiquitous maize ubiquitin promoter, and their abilities to confer resistance to the diphenyl ether‐type herbicide, oxyfluorfen were compared. The integration and expression of the transgene in the T₁ generation was examined by Southern, northern and western blot analyses. Surprisingly, as judged by an in vivo seed germination assay and an in vitro cellular leakage assay, both lines were similarly resistant to oxyfluorfen. The tolerance to cellular damage (lipid peroxidation and electrolyte leakage) was higher in transgenic plants than in wild‐type plants. In transgenic plants, the degree of herbicide resistance varied directly with the absolute amount of Protox protein expression. Both the intact protein and the protein with the transit sequence deleted were accumulated in plastids.     

Expression of recombinant protoporphyrinogen oxidase influences growth and morphological characteristics in transgenic rice

Transgenic rice plants expressing a Bacillus subtilis protoporphyrinogen oxidase (Protox), the last shared enzyme of the porphyrin pathway, in the cytoplasm (C89) or the plastids (P72) were compared with wild-type rice plants in their growth characteristics. Production of tiller buds 18 d after seeding was more profuse in transgenic plants than in wild-type plants, especially in plastid-targeted plants. Transgenic plants had 12–27% increase in tiller number and 17–33% increase in above-ground biomass compared with wild-type plants 4 and 8 weeks after transplanting of 2-week-old rice seedlings, demonstrating that tiller production and above-ground biomass correlate with each other. Cytoplasm-expressed and plastid-targeted transgenic plants also had a distinct phenotypic characteristic of narrower and more horizontal leaves than wild-type plants. Phenotypic and anatomical characteristics of the transgenic plants were clearly different from wild-type plants, indicating that regulation of porphyrin biosynthesis by expression of B. subtilis Protox in rice influences morphological characteristics of plant growth as well as biomass.     

Resistance mechanisms in protoporphyrinogen oxidase (PROTOX) inhibitor-resistant transgenic rice

We investigated the mechanism for conferring herbicide resistance in transgenic rice. Plants from Line M4 were resistant to PROTOX inhibitors and had yields similar to those from wild-type (WT) rice.Myxococcus xanthus PROTOX mRNA was abundantly expressed in the transgenic leaf tissues, and theM. xanthus PROTOX gene was stably transmitted into the T₄ generation. We detected a protein with a predicted molecular mass of 50 kD, equal to the weight ofM. xanthus PROTOX, in M4 but not WT plants. Furthermore, several PROTOX-inhibitor herbicides — acifluorfen, oxyfluorfen, carfentrazone-ethyl, and oxadiazon — caused significant cellular leakage and lipid peroxidation in the WT, but not in the transgenics. Total PROTOX activity in untreated transformed rice was 17-fold higher than in the WT, with activity being inhibited in the latter genotype by 55%, 59%, 53%, or 60% as a result of treatment with acifluorfen, oxyfluorfen, carfentrazone-ethyl, or oxadiazon, respectively. However, PROTOX activities in transgenic rice were similar to their corresponding, untreated controls. The accumulation of Proto IX was 15-to 21-fold higher in the WT than in M4 when plants were treated with PROTOX inhibitors. In the former, their epicuticular wax and chloroplasts were severely damaged after oxyfluorfen treatment The lack of damage in transformed plants suggests that M4 does not accumulate Proto IX, probably due to the production of herbicide-resistant chloroplastic and mitochondria PROTOX.     

Transgenic rice plants expressing a Bacillus subtilis protoporphyrinogen oxidase gene are resistant to diphenyl ether herbicide oxyfluorfen

Protoporphyrinogen oxidase (Protox), the penultimate step enzyme of the branch point for the biosynthetic pathway of Chl and hemes, is the target site of action of diphenyl ether (DPE) herbicides. However, Bacillus subtilis Protox is known to be resistant to the herbicides. In order to develop the herbicide-resistant plants, the transgenic rice plants were generated via expression of B. subtilis Protox gene under ubiquitin promoter targeted to the cytoplasm or to the plastid using Agrobacterium-mediated gene transformation. The integration and expression of the transgene were investigated at T0 generation by DNA and RNA blots. Most transgenic rice plants revealed one copy transgene insertion into the rice genome, but some with 3 copies. The expression levels of B. subtilis Protox mRNA appeared to correlate with the copy number. Furthermore, the plastidal transgenic lines exhibited much higher expression of the Protox mRNA than the cytoplasmic transgenic lines. The transgenic plants expressing the B. subtilis Protox gene at T0 generation were found to be resistant to oxyfluorfen when judged by cellular damage with respect to cellular leakage, Chl loss, and lipid peroxidation. The transgenic rice plants targeted to the plastid exhibited higher resistance to the herbicide than the transgenic plants targeted to the cytoplasm. In addition, possible resistance mechanisms in the transgenic plants to DPE herbicides are discussed.     

Effective selection and regeneration of transgenic rice plants with mannose as selective agent

A new method for the selection of transgenic rice plants without the use of antibiotics or herbicides has been developed. The phosphomannose isomerase (PMI) gene from Escherichia coli has been cloned and consitutively expressed in japonica rice variety TP 309. The PMI gene was transferred to immature rice embryos by Agrobacterium-mediated transformation, which allowed the selection of transgenic plants with mannose as selective agent. The integration and expression of the transgene was confirmed by Southern and northern blot analysis and the activity of PMI indirectly proved with the chlorophenol red assay. The results of genetic analysis showed that the transgenes were segregated in a Mendelian fashion in the T₁ generation. The establishment of this selection system in rice provides an efficient way for producing transgenic plants without using antibiotics or herbicides with a transformation frequency of up to 41%.     

Overexpression of the Escherichia coli catalase gene, katE, enhances tolerance to salinity stress in the transgenic indica rice cultivar, BR5

Salinity stress is a major limiting factor in cereal productivity. Many studies report improvements in salt tolerance using model plants, such as Arabidopsis thaliana or standard varieties of rice, e.g., the japonica rice cultivar Nipponbare. However, there are few reports on the enhancement of salt tolerance in local rice cultivars. In this work, we used the indica rice (Oryza sativa) cultivar BR5, which is a local cultivar in Bangladesh. To improve salt tolerance in BR5, we introduced the Escherichia coli catalase gene, katE. We integrated the katE gene into BR5 plants using an Agrobacterium tumefaciens-mediated method. The introduced katE gene was actively expressed in the transgenic BR5 rice plants, and catalase activity in T₁ and T₂ transgenic rice was approximately 150% higher than in nontransgenic plants. Under NaCl stress conditions, the transgenic rice plants exhibited high tolerance compared with nontransgenic rice plants. T₂ transgenic plants survived in a 200 mM NaCl solution for 2 weeks, whereas nontransgenic plants were scorched after 4 days soaking in the same NaCl solution. Our results indicate that the katE gene can confer salt tolerance to BR5 rice plants. Enhancement of salt tolerance in a local rice cultivar, such as BR5, will provide a powerful and useful tool for overcoming food shortage problems.     

Inheritance of gusA and neo genes in transgenic rice

Inheritance of foreign genes neo and gusA in rice (Oryza sativa L. cv. IR54 and Radon) has been investigated in three different primary (T₀) transformants and their progeny plants. T₀ plants were obtained by co-transforming protoplasts from two different rice suspension cultures with the neomycin phosphotransferase II gene [neo or aph (3) II] and the -glucuronidase gene (uidA or gusA) residing on separate chimeric plasmid constructs. The suspension cultures were derived from callus of immature embryos of indica variety IR54 and japonica variety Radon. One transgenic line of Radon (AR2) contained neo driven by the CaMV 35S promoter and gusA driven by the rice actin promoter. A second Radon line (R3) contained neo driven by the CaMV 35S promoter and gusA driven by a promoter of the rice tungro bacilliform virus. The third transgenic line, IR54-1, contained neo driven by the CaMV 35S promoter and gusA driven by the CaMV 35S.Inheritance of the transgenes in progeny of the transgenic rice was investigated by Southern blot analysis and enzyme assays. Southern blot analysis of genomic DNA showed that, regardless of copy numbers of the transgenes in the plant genome and the fact that the two transgenes resided on two different plasmids before transformation, the introduced gusA and neo genes were stably transmitted from one generation to another and co-inherited together in transgenic rice progeny plants derived from self-pollination. Analysis of GUS and NPT II activities in T₁ to T₂ plants provided evidence that inheritance of the gusA and neo genes was in a Mendelian fashion in one plant line (AR2), and in an irregular fashion in the two other plant lines (R3 and IR54-1). Homozygous progeny plants expressing the gusA and neo genes were obtained in the T₂ generation of AR2, but the homozygous state was not found in the other two lines of transgenic rice.     

Comparative analysis of transgenic rice plants obtained by Agrobacterium-mediated transformation and particle bombardment

We compared rice transgenic plants obtained by Agrobacterium-mediated and particle bombardment transformation by carrying out molecular analyses of the T₀, T₁ and T₂ transgenic plants. Oryza sativa japonica rice (c.v. Taipei 309) was transformed with a construct (pWNHG) that carried genes coding for neomycin phosphotransferase (nptII), hygromycin phosphotransferase (Hyg^r), and -glucuronidase (GUS). Thirteen and fourteen transgenic lines produced via either method were selected and subjected to molecular analysis. Based on our data, we could draw the following conclusions. Average gene copy numbers of the three transgenes were 1.8 and 2.7 for transgenic plants obtained by Agrobacterium and by particle bombardment, respectively. The percentage of transgenic plants containing intact copies of foreign genes, especially non-selection genes, was higher for Agrobacterium-mediated transformation. GUS gene expression level in transgenic plants obtained from Agrobacterium-mediated transformation was more stable overall the transgenic plant lines obtained by particle bombardment. Most of the transgenic plants obtained from the two transformation systems gave a Mendelian segregation pattern of foreign genes in T₁ and T₂ generations. Co-segregation was observed for lines obtained from particle bombardment, however, that was not always the case for T₁ lines obtained from Agrobacterium-mediated transformation. Fertility of transgenic plants obtained from Agrobacterium-mediated transformation was better. In summary, the Agrobacterium-mediated transformation is a good system to obtain transgenic plants with lower copy number, intact foreign gene and stable gene expression, while particle bombardment is a high efficiency system to produce large number of transgenic plants with a wide range of gene expression.     

Use of bar as a selectable marker gene and for the production of herbicide-resistant rice plants from protoplasts

We have used the bar gene in combination with the herbicide Basta to select transformed rice (Oryza sativa L. cv. Radon) protoplasts for the production of herbicide-resistant rice plants. Protoplasts, obtained from regenerable suspension cultures established from immature embryo callus, were transformed using PEG-mediated DNA uptake. Transformed calli could be selected 2–4 weeks after placing the protoplast-derived calli on medium containing the selective agent, phosphinothricin (PPT), the active component of Basta. Calli resistant to PPT were capable of regenerating plants. Phosphinothricin acetyltransferase (PAT) assays confirmed the expression of the bar gene in plants obtained from PPT-resistant calli. The only exceptions were two plants obtained from the same callus that had multiple copies of the bar gene integrated into their genomes. The transgenic status of the plants was varified by Southern blot analysis. In our system, where the transformation was done via the protoplast method, there were very few escapes. The efficiency of co-transformation with a reporter gene gusA, was 30%. The T_o plants of Radon were self-fertile. Both the bar and gusA genes were transmitted to progeny as confirmed by Southern analysis. Both genes were expressed in T₁ and T₂ progenies. Enzyme analyses on T₁ progeny plants also showed a gene dose response reflecting their homozygous and heterozygous status. The leaves of T_o plants and that of the progeny having the bar gene were resistant to application of Basta. Thus, the bar gene has proven to be a useful selectable and screenable marker for the transformation of rice plants and for the production of herbicide-resistant plants.     

Inheritance of a co-transferred foreign gene in the progenies of transgenic rice plants

The integration pattern and the inheritance of exogenous DNA in transgenic rice plants were analysed. Plasmid pCH (4.8 kb), that contains chimaeric cauliflower mosaic virus 35S promoter-hygromycin phosphotransferase structural gene, and plasmid pGP400 (7.2 kb), possessing oat phytochrome promoter and structural gene of bacterial -glucuronidase, were co-transferred into protoplasts of rice (Oryza sativa L.) plants via electroporation. Primary transformants (T₀ generation) and their progenies (T₁, T₂ and T₃) were selected by hygromycin B. Southern blot analysis of inserted genes in transgenic rice plants suggests the integration of an intact hygromycin phosphotransferase gene and non-functional DNA fragments into host genome. Co-inheritance of the hygromycin phosphotransferase gene and -glucuronidase gene was also observed. There were no significant differences in terms of the morphology and size of seeds between untransformed and transgenic plants (T₃ generation).     

Excision of a selectable marker in transgenic rice (Oryza sativa L.) using a chemically regulated Cre/loxP system

Removal of a selectable marker gene from genetically modified (GM) crops alleviates the risk of its release into the environment and hastens the public acceptance of GM crops. Here we report the production of marker-free transgenic rice by using a chemically regulated, Cre/loxP-mediated site-specific DNA recombination in a single transformation. Among 86 independent transgenic lines, ten were found to be marker-free in the T₀ generation and an additional 17 lines segregated marker-free transgenic plants in the T₁ generation. Molecular and genetic analyses indicated that the DNA recombination and excision in transgenic rice were precise and the marker-free recombinant T-DNA was stable and heritable.The first two authors contributed equally to the work     

The expression of a maize glutathione S-transferase gene in transgenic wheat confers herbicide tolerance,both in planta and in vitro

Maize (Zea mays), in common with a number of other important crop species, has several glutathione S-transferase (GST) isoforms that have been implicated in the detoxification of xenobiotics via glutathione conjugation. A cDNA encoding the maize GST subunit GST-27, under the control of a strong constitutive promoter, was introduced into explants of the wheat (Triticum aestivum L.) lines cv. Florida and L88-31 via particle bombardment, using the phosphinothricin acetyltransferase (pat) gene as a selectable marker. All six independent transgenic wheat lines recovered expressed the GST-27 gene. T₁ progeny of these wheat lines were germinated on solid medium containing the chloroacetanilide herbicide alachlor, and tolerance to this herbicide was correlated with GST-27 expression levels. In glasshouse sprays, homozygous T₂ plants were resistant not only to alachlor but also to the chloroacetanilide herbicide dimethenamid and the thiocarbamate herbicide EPTC. These additional GST-27 activities, demonstrated via over-expression in a heterologous host, have not been described previously. T₂ plants showed no enhanced tolerance to the herbicides atrazine (an s-triazine) or oxyfluorfen (a diphenyl ether). In further experiments, T₂ wheat plants were recovered from immature transgenic scutella cultured on medium containing 100 mg/l alachlor, a concentration which killed null segregant and wild-type scutella. These data indicate the potential of the maize GST-27 gene as a selectable marker in wheat transformation.     

Over-expression of a <Emphasis Type="Italic">LEA</Emphasis> gene in rice improves drought resistance under the field conditions

Late embryogenesis abundant (LEA) proteins have been implicated in many stress responses of plants. In this report, a LEA protein gene OsLEA3-1 was identified and over-expressed in rice to test the drought resistance of transgenic lines under the field conditions. OsLEA3-1 is induced by drought, salt and abscisic acid (ABA), but not by cold stress. The promoter of OsLEA3-1 isolated from the upland rice IRAT109 exhibits strong activity under drought- and salt-stress conditions. Three expression constructs consisting of the full-length cDNA driven by the drought-inducible promoter of OsLEA3-1 (OsLEA3-H), the CaMV 35S promoter (OsLEA3-S), and the rice Actin1 promoter (OsLEA3-A) were transformed into the drought-sensitive japonica rice Zhonghua 11. Drought resistance pre-screening of T₁ families at anthesis stage revealed that the over-expressing families with OsLEA3-S and OsLEA3-H constructs had significantly higher relative yield (yield under drought stress treatment/yield under normal growth conditions) than the wild type under drought stress conditions, although a yield penalty existed in T₁ families under normal growth conditions. Nine homozygous families, exhibiting over-expression of a single-copy of the transgene and relatively low yield penalty in the T₁ generation, were tested in the field for drought resistance in the T₂ and T₃ generations and in the PVC pipes for drought tolerance in the T₂ generation. Except for two families (transformed with OsLEA3-A), all the other families (transformed with OsLEA3-S and OsLEA3-H constructs) had higher grain yield than the wild type under drought stress in both the field and the PVC pipes conditions. No significant yield penalty was detected for these T₂ and T₃ families. These results indicate that transgenic rice with significantly enhanced drought resistance and without yield penalty can be generated by over-expressing OsLEA3-1 gene with appropriate promoters and following a bipartite (stress and non-stress) in-field screening protocol.     

Resistance to Alternaria brassicicola in transgenic broccoli expressing a Trichoderma harzianum endochitinase gene

Transgenic broccoli plants expressing a Trichoderma harzianum endochitinase gene were obtained by Agrobacterium tumefaciens-mediated transformation. PCR and Southern blot analysis confirmed the presence of the gene in plants initially selected via resistance to kanamycin. Primary transformants (T₀) and selfed progeny (T₁) were examined for expression of the endochitinase gene using a fluorometric assay and for their resistance to the fungal pathogens Alternaria brassicicola and Sclerotinia sclerotiorum. All transgenic plants with elevated endochitinase activity had the expected 42 kDa endochitinase band in western blot analysis, whereas no such band was detected in the non-transgenic control. Leaves of most mature T₀ plants had 14–37 times higher endochitinase activity than controls; mature T₁ plants had higher endochitinase activity (100–200 times that in controls), in part because of lower control values. T₀ plantlets in vitro or young plants in soil had higher absolute and relative endochitinase activity. When detached leaves of T₀ plants were inoculated with A. brassicicola, lesion size showed a significant negative correlation with endochitinase levels. After inoculation of two-month old T₀ plants with A. brassicicola, all 15 transgenic lines tested showed significantly less severe disease symptoms than controls. In contrast, lesion size on petioles of T₀ and T₁ plants inoculated with S. sclerotiorum was not statistically different from controls.     

Excision of a selective marker in transgenic rice using a novel Cre/<Emphasis Type="Italic">loxP</Emphasis> system controlled by a floral specific promoter

Based on the Cre/loxP system, we have developed a novel marker-free system mediating a direct auto-excision of loxP-flanked marker genes from T₁ transgenic rice without any treatment or further offspring crossing. To achieve this, the floral-specific promoter OsMADS45 was isolated from rice and the expression patterns of OsMADS45 promoter was characterised by using the pOs45:GUS transgenic plants. Furthermore, the binary vector with Cre recombinase under the control of OsMADS45 promoter was constructed and introduced into rice by Agrobacterium-mediated transformation and transgenic rice plants were generated. Southern blot analysis showed that auto-excision of the selective markers occurred in some T₁ progeny of the transgenic plants, suggesting that a high auto-excision frequency can be achieved with our Cre/loxP system. This auto-excision strategy provides an efficient way of removing the selectable marker gene from transgenic rice. Xianquan Bai and Qiuyun Wang contributed equally to the work.     

Combined expression of chitinase and β-1,3-glucanase genes in indica rice (Oryza sativa L.) enhances resistance against Rhizoctonia solani

Agrobacterium-mediated transformation of rice was done using the binary vector pNSP3, harbouring the rice chitinase (chi11) gene under maize ubiquitin promoter and the tobacco β-1,3-glucanase gene under CaMV 35S promoter in the same T-DNA. Four of the six T₀ plants had single copies of complete T-DNAs, while the other two had complex integration patterns. Three of the four single-copy lines showed a 3:1 segregation ratio in the T₁ generation. Northern and western blot analyses of T₁ plants revealed constitutive expression of chitinase and β-1,3-glucanase genes. Homozygous T₂ plants of the single-copy lines CG20, CG27 and CG53 showed 62-, 9.6- and 11-fold higher chitinase activity over the control plants. β-1,3-Glucanase activity was 1.1- to 2.5-fold higher in the transgenic plants. Bioassay of homozygous T₂ plants of the three single-copy transgenic lines against Rhizoctonia solani revealed a 60% reduction in sheath blight Disease Index in the first week. The Disease Index increased from 61.8 in the first week to 90.6 in the third week in control plants, while it remained low (26.8–34.2) in the transgenic T₃ plants in the corresponding period, reflecting the persistence of sheath blight resistance for a longer period.     

Co-bombardment,integration and expression of rice chitinase and thaumatin-like protein genes in barley (<Emphasis Type="Italic">Hordeum vulgare</Emphasis> cv. Conlon)

Pathogenesis-related (PR) proteins associated with degradation of structural components of pathogenic filamentous fungi were overexpressed in the two-rowed malting barley (Hordeum vulgare L.) cultivar Conlon. Transgenes were introduced by co-bombardment with two plasmids, one carrying a rice (Oryza sativa L.) chitinase gene (chi11) and another carrying a rice thaumatin-like protein gene (tlp). Each gene was under the control of the maize ubiquitin (Ubi1) promoter. Fifty-eight primary transformants from three independent transformation events were regenerated. T₁ plants with high rice chi11 and tlp protein expression levels were advanced to identify T₂ homozygotes by herbicide spray and subjected to further molecular analyses. T₃ progeny from one event (E2) had stable integration and expression of the rice chi11 and tlp while those from the other events (E1 and E3) showed stable integration only of tlp. The successful production of these lines overexpressing the antifungal chi and tlp proteins provides materials to test the effects of these genes on a variety of fungal diseases that attack barley and to serve as potential additional sources of disease resistance.     

Characterization of sunlight-grown transgenic rice plants expressing Arabidopsis phytochrome A

Transgenic rice (Oryza sativa) overexpressing Arabidopsis phytochrome A (phyA) was cultivated up to the T₃ generation in paddy to elucidate the role of phyA in determining the plant architecture and the productivity of sunlight-grown rice plants. PhyA is light-labile and controls plant growth in response to the far-red light-dependent high-irradiance response as well as the very low fluence response. The Arabidopsis phyA gene linked to the rice rbcS promoter was transformed into embryogenic rice calli, and the calli were regenerated to whole plants. Compared to wild-type seedlings, the rbcS::PHYA transgenic seedlings contained more phyA when grown in the dark, and at least 10-fold more phyA when exposed to white light. When grown in paddy, the phyA transgenic plants in general exhibited reduced plant height (dwarfing), larger grain size, higher chlorophyll content, smaller tiller number, and low grain fertility compared to wild-type plants. The heading stage was not significantly changed. However, it is likely that a certain level of phyA is a prerequisite for induction of such changes. It is suggested that phyA overproduction in rice could be a useful tool to improve rice grain productivity by the larger grain size that increases grain yield and the dwarfing that tolerates lodging-associated damage.