Effects of CRISPR/Cas9 generated drooping leaf (dl) alleles on midrib and carpel formations in Oryza sativa Nipponbare.

MAIN CONCLUSION: We generated drooping leaf rice mutants by CRISPR/Cas and identified two novel alleles with specific editing that allow underpinning of the function of the DL protein domain towards midrib and carpel formations. The DROOPING LEAF (DL) gene plays an essential role in regulating midrib formation and carpel specification in rice and other grass species, but the specific function of DL protein domains in different developmental processes is unclear. Analysis of different dl mutant alleles will allow dissecting the function of DL. Here, we generated Nipponbare rice dl mutants using CRISPR/Cas gene editing and identified two novel dl alleles with different effects on midrib formation and carpel development. Phenotypic and genotypic analysis of T0 and segregated T1 edited lines showed that while dl-51S allele (a 3 bp deletion and a serine deletion at position 51) reduces midrib sizes and produces normal carpels, the dl-50LS allele (a 6 bp deletion and a leucine-serine deletion at position 50-51) causes the lack of midribs and abnormal stigma. This result indicates that the 51-serine is important for midrib formation and the 50-leucine is essential for midrib and carpel development. These dl mutant alleles contribute to the DL gene functional analysis and to gain insights into possible modifications of leaf architecture of rice and other grass species.

Identification of cis-localization elements of the maize 10-kDa delta-zein and their use in targeting RNAs to specific cortical endoplasmic reticulum subdomains.

The RNAs for the storage proteins of rice (Oryza sativa), prolamines and glutelins, which are stored as inclusions in the lumen of the endoplasmic reticulum (ER) and storage vacuoles, respectively, are targeted by specific cis-localization elements to distinct subdomains of the cortical ER. Glutelin RNA has one or more cis-localization elements (zip codes) at the 3' end of the RNA, whereas prolamine has two cis-elements; one located in the 5' end of the coding sequence and a second residing in the 3'-untranslated region (UTR). We had earlier demonstrated that the RNAs for the maize zeins ('prolamine' class) are localized to the spherical protein body ER (PB-ER) in developing maize endosperm. As the PB-ER localization of the 10-kDa delta-zein RNA is maintained in developing rice seeds, we determined the number and proximate location of their cis-localization elements by expressing GFP fusions containing various zein RNA sequences in transgenic rice and analyzing their spatial distribution on the cortical ER by in situ RT-PCR and confocal microscopy. Four putative cis-localization elements were identified; three in the coding sequences and one in the 3'-UTR. Two of these zip codes are required for restricted localization to the PB-ER. Using RNA targeting determinants we show, by mis-targeting the storage protein RNAs from their normal destination on the cortical ER, that the coded proteins are redirected from their normal site of deposition. Targeting of RNA to distinct cortical ER subdomains may be the underlying basis for the variable use of the ER lumen or storage vacuole as the final storage deposition site of storage proteins among flowering plant species.

Control of starch synthesis in cereals: metabolite analysis of transgenic rice expressing an up-regulated cytoplasmic ADP-glucose pyrophosphorylase in developing seeds.

We had previously demonstrated that expression of a cytoplasmic-localized ADPglucose pyrophosphorylase (AGPase) mutant gene from Escherichia coli in rice endosperm resulted in enhanced starch synthesis and, in turn, higher seed weights. In this study, the levels of the major primary carbon metabolites were assessed in wild type and four transgenic CS8 rice lines expressing 3- to 6-fold higher AGPase activity. Consistent with the increase in AGPase activity, all four transgenic CS8 lines showed elevated levels of ADPglucose (ADPglc) although the extent of increases in this metabolite was much higher than the extent of increases in starch as measured by seed weight. Surprisingly, the levels of several other key intermediates were significantly altered. Glucose 1-phosphate (Glc 1-P), a substrate of the AGPase reaction, as well as UDPglucose and Glc 6-P were also elevated to the same relative extent in the transgenic lines compared with the wild-type control. Analysis of metabolite ratios showed no significant differences between the wild type and transgenic lines, indicating that the reactions leading from sucrose metabolism to ADPglc formation were in near equilibrium. Moreover, glucose and fructose levels were also elevated in three transgenic lines that showed the largest differences in metabolites and seed weight over the wild type, suggesting the induction of invertase. Overall, the results indicate that the AGPase-catalyzed reaction is no longer limiting in the transgenic lines, and constraints on carbon flux into starch are downstream of ADPglc formation, resulting in an elevation of precursors upstream of ADPglc formation.

Dual regulated RNA transport pathways to the cortical region in developing rice endosperm.

Prolamine and glutelin RNAs are localized to two subdomains of the cortical endoplasmic reticulum (ER), the protein body ER and the cisternal ER, in developing rice seeds. The addition of nearly full-length prolamine sequences at the 3' untranslated region of a reporter RNA redirects its localization from the cisternal ER to the protein body ER. Deletion analysis of prolamine RNA sequences indicates the presence of two partially redundant cis elements required for protein body ER targeting. The addition of glutelin 3' untranslated region to protein body ER cis sequences, however, redirects RNA localization to the cisternal ER. These results indicate that there are at least two regulated RNA transport pathways as well as a constitutive pathway to the cortical ER.