Characterisation of the expression of a novel constitutive maize promoter in transgenic wheat and maize.

A novel constitutive promoter from the maize histone H2Bgene was recently identified. In this study, we characterised H2B promoter activity in both wheat and maize tissues using the gusA reporter gene and two synthetic versions of the pat (phosphinothricin acetyl transferase) selectable marker gene, namely mopat and popat. Analyses of transgenic plants showed that the H2B promoter is able to drive the expression of gusA to strong, constitutive levels in wheat and maize tissues. Using an H2B:mopat construct and phosphinothricin selection, we recovered transgenic wheat plants at efficiencies ranging from 0.3% to 7.4% (mean 1.6%), and the efficiency of selection ranged from 40% to 100% (mean 77.7%). In another application, H2B was combined with the maize Ubi-1 or the maize Adh-1 intron to drive the expression of mopat and popat. Transformation efficiencies with the Ubi-1 intron were between 1.4- to 16-fold greater than with the Adh-1 intron. However, the use of either of the introns was necessary for the recovery of transgenic plants. Mopat gave higher transformation efficiencies and induced higher levels of PAT protein in maize tissues than popat.

Gene discovery and product development for grain quality traits.

The composition of oils, proteins, and carbohydrates in seeds of corn, soybean, and other crops has been modified to produce grains with enhanced value. Both plant breeding and molecular technologies have been used to produce plants carrying the desired traits. Genomics-based strategies for gene discovery, coupled with high-throughput transformation processes and miniaturized, automated analytical and functionality assays, have accelerated the identification of product candidates. Molecular marker-based breeding strategies have been used to accelerate the process of moving trait genes into high-yielding germplasm for commercialization. These products are being tested for applications in food, feed, and industrial markets.

Aspartic proteinase genes in the Brassicaceae Arabidopsis thaliana and Brassica napus.

Active aspartic proteinase is isolated from Brassica napus seeds and the peptide sequence is used to generate primers for PCR. We present here cDNA and genomic clones for aspartic proteinases from the closely related Brassicaceae Arabidopsis thaliana and Brassica napus. The Arabidopsis cDNA represents a single gene, while Brassica has at least 4 genes. Like other plant aspartic proteases, the two Brassicaceae enzymes contain an extra protein domain of about 100 amino acids relative to the mammalian forms. The intron/exon arrangement in the Brassica genomic clone is significantly different from that in mammalian genes. As the proteinase is isolated from seeds, the same tissue where 2S albumins are processed, this implies expression of one of the aspartic proteinase genes there.

Modified 2S albumins with improved tryptophan content are correctly expressed in transgenic tobacco plants.

Brazil nut 2S albumins lack the essential amino acid tryptophan. In order to improve the protein's nutritional value and create a basis for structural investigations, three separate modified Brazil nut 2S albumin genes were constructed. The first mutant contains five consecutive tryptophan codons, while the other two modified genes encode proteins carrying single tryptophan residues at sites that will allow confirmation of the predicted protein structure through fluorescence quenching techniques. The modified genes, under the regulation of the CaMV 35S promoter, were introduced into Nicotiana tabacum. All three modified genes were correctly transcribed and the 2S albumin accumulated in the seeds of transgenic plants.

Unexpectedly higher expression levels of a chimeric 2S albumin seed protein transgene from a tandem array construct.

In wild-type Arabidopsis seeds the 2S albumin seed protein gene family members are differentially expressed. In this work it is shown that as predicted by the wild type situation, the at2S2 promoter is much more effective than that of the at2S1 gene in the expression of a transgene. However, unexpectedly high expression levels were obtained using a construct in which the transgene was present as a tandem duplication in the T-DNA. Neither in this case nor in homozygous plants with either construct was epigenetic silencing observed. While transgene mRNA levels were of the same order of magnitude as the endogenous at2S2 gene, protein levels were much lower.

A cotyledon regulatory region is responsible for the different spatial expression patterns of Arabidopsis 2S albumin genes.

The 2S albumin genes of Arabidopsis thaliana are a model system to study gene expression during late embryogenesis. The at2S1 gene has previously been shown to be expressed essentially in the embryo axis, unlike at2S2, which is expressed throughout the embryo. Hybrid promoter constructs between at2S1 and at2S2 were introduced into Arabidopsis and used to identify a cotyledon regulatory region necessary for 2S albumin expression in palisade parenchyma and specific epidermal cells. Other promoter sequences flanking this tissue-specific promoter element were shown to control mRNA expression levels independently of the mRNA distribution throughout the embryos. Certain hybrid promoters resulted in the alteration of the time course of expression in cotyledons. Differential expression of 2S albumin genes is discussed in terms of layered cellular organization and mitotic activity throughout the embryo.

An aspartic proteinase present in seeds cleaves Arabidopsis 2 S albumin precursors in vitro.

The Arabidopsis thaliana 2 S albumins are examples of vacuolar proteins which undergo intensive posttranslational processing. An in vitro processing assay to screen for processing enzymes present in seeds was developed using an in vitro synthesized 2 S albumin precursor as the substrate. A protease was characterized which cleaved the substrate into two fragments with molecular weights (as determined from their migration distance on SDS-polyacrylamide gel) corresponding to those of the small and large subunits of Arabidopsis 2 S albumin. The pH optimum of this protease activity, its inhibition by pepstatin A, and partial sequence data led to the conclusion that the protease under study was an aspartic proteinase. Synthetic peptides representing two 2 S albumin propeptides allowed the determination of the in vitro cleavage sites, and suggested that the protease activity is capable, in vitro, of cleaving the amino-terminal propeptide as well as the internal propeptide linking the two subunits. Alterations of the amino acids in and around the cleavage sites, made to study the specificity of the protease activity, suggest that both structural and sequence determinants are important in cleavage site recognition.

Studies of the role of the propeptides of the Arabidopsis thaliana 2S albumin.

To investigate the possible roles of the Arabidopsis thaliana 2S albumin propeptides with respect to sorting, processing, and stability of the protein in plant cells, five gene constructions deleting or modifying the propeptides were made based on one of the genes encoding the Arabidopsis 2S albumin. These constructions were introduced into tobacco (Nicotiana tabacum) plants. Using subcellular fractionation and immunocytochemistry on ripe seeds, it was demonstrated that none of the propeptides was necessary for the sorting of the protein. Detailed protein-chemical analysis of the mature gene products indicated that, for all of the modified 2S albumin precursors made, the proteins were stably folded and correctly processed. However, the latter is less efficient when the internal fragment between the small and the large subunit is missing or when this internal fragment is changed. In an attempt to establish a rapid assay system for modified 2S albumin precursors, yeast cells were transformed with the same gene constructs. It was demonstrated that the processing machinery in yeast cells differs from that in plants, and, in a perhaps related observation, differences in stability of a particular modified protein were observed.

Arabidopsis rbcS genes are differentially regulated by light.

Individual members of the Arabidopsis thaliana ribulose-1,5-bisphosphate carboxylase/oxygenase small subunit (rbcS) gene family are differentially regulated by light of different qualities. In 10-d-old etiolated seedlings, the expression of only three of the four genes is under inductive phytochrome control. rbcS mRNA levels reach a maximum (3- to 5-fold higher than the dark level) about 6 h after a red light pulse, but the rate of decay differs among the genes. Moreover, rbcS 2B requires a higher fluence for induction. At early stages of development, rbcS 1A, 2B, and 3B are highly expressed in the dark and cannot be further induced by red light, indicating a developmental component in the overall regulatory mechanism. Continuous light experiments indicate that high-irradiance responses may play a role in the induction of at least three of the four rbcS genes. Under conditions of phytochrome saturation, rbcS 1A is insensitive to blue light pulses, whereas among the three B locus genes, at least rbcS 3B appears to respond to a blue-light photoreceptor. These results add to the data suggesting that individual members of rbcS gene families in higher plants may be subject to a variety of differing regulatory mechanisms.

Stable Accumulation of Modified 2S Albumin Seed Storage Proteins with Higher Methionine Contents in Transgenic Plants.

We present the results of two sets of experiments designed to express high methionine proteins in transgenic seeds in three different plant species. In the first approach, two chimeric genes were constructed in which parts of the Arabidopsis 2S albumin gene 1 (AT2S1) were fused at different positions to a Brazil nut 2S albumin cDNA clone. Brazil nut 2S albumin was found to accumulate stably in transgenic Arabidopsis, Brassica napus, and tobacco seeds. In the second approach, methionine-enriched AT2S1 genes were constructed by deleting sequences encoding a region of the protein which is not highly conserved among 2S albumins of different species and replacing them with methioninerich sequences. Introduction of the modified AT2S1 genes into three different plant species resulted in the accumulation of the methionine-enriched 2S albumins in all three species at levels reaching 1 to 2% of the total high salt-extractable seed protein.

Expression of the 2S albumin from Bertholletia excelsa in Brassica napus.

The methionine rich 2S albumin seed storage protein of Bertholletia excelsa has been expressed in seeds of Brassica napus (rapeseed). A chimeric gene driven by the soybean lectin 5' flanking regions was used to produce a fusion protein consisting of the soybean lectin signal peptide and the propeptide of the Brazil nut 2S albumin. Several transgenic plants were studied at the RNA and protein levels; in each case the chimeric gene was expressed and the protein detected at levels ranging from 0.02% to 0.06% of total protein. Transcriptional studies in a particular transgenic plant show that expression of the gene is tissue specific and developmentally regulated during seed maturation. The endogenous napin genes and the introduced gene are regulated differently, with expression of the chimeric gene paralleling that seen when the soybean lectin gene is expressed in other plant species. Western analysis using antibodies to Brazil nut 2S albumins resulted in the detection of a protein whose size is consistent with correct processing of the precursor.

Differential Expression of the Arabidopsis 2S Albumin Genes and the Effect of Increasing Gene Family Size.

We studied the expression of the four genes encoding 2S albumin seed storage proteins (at2S1 to at2S4) in Arabidopsis thaliana. All four genes followed similar temporal profiles throughout development, but at2S2 and at2S3 were expressed at significantly higher levels than at2S1 or at2S4. In situ hybridization showed that at2S2 to at2S4 mRNAs were present throughout the embryo, whereas at2S1 was expressed at levels similar to at2S2 and at2S3 in the embryo axis but at only insignificant levels in the cotyledons. The different members of the gene family are, thus, likely to be regulated by different combinations of cis-acting elements, but it cannot be ruled out that post-transcriptional factors play a role. We studied the effect of enlarging the gene family by introducing an extra, nearly identical gene driven by the promoter of at2S1. The data were consistent with a model in which the expression of at2S2 to at2S4 is not affected by that of at2S1, and in which, at least at low copy numbers of the introduced gene, there is no limit on the overall amount of RNA that the at2S gene family can produce.

Expression and Processing of an Arabidopsis 2S Albumin in Transgenic Tobacco.

2S albumin seed storage proteins undergo a complex series of posttranslational proteolytic cleavages. In order to determine if this process is correctly carried out in transgenic plants, the gene AT2S1 encoding an Arabidopsis thaliana 2S albumin isoform has been expressed in transgenic tobacco. Initial experiments using a reporter gene demonstrated that the AT2S1 promoter directs seed specific expression in both transgenic tobacco and Brassica napus plants. The entire AT2S1 gene was then transferred into tobacco plants, where it showed a tissue specific and developmentally regulated expression. Arabidopsis 2S albumin accumulates up to 0.1% of the total high-salt extractable seed protein. Protein sequencing demonstrated that the amino termini of the two Arabidopsis 2S albumin subunits were correctly processed, suggesting that the protease(s) necessary for posttranslational processing of 2S albumin precursors may display common specificities among different dicot plant species. Immunocytochemical studies showed that the Arabidopsis 2S albumin is localized in the protein body matrix of tobacco endosperm and embryo. Correct processing and targeting of the 2S albumin in transgenic plants suggests that modified versions could be expressed, allowing the study of 2S albumin processing and in particular the possible roles of the processed fragments in protein stability and/or targeting.

Determination of the Processing Sites of an Arabidopsis 2S Albumin and Characterization of the Complete Gene Family.

The most abundant isoform of the 2S albumin present in seeds of Arabidopsis thaliana has been sequenced and the corresponding gene isolated. Examination of the protein and DNA sequences allows the determination of the exact proteolytic cleavage sites during posttranslational processing. Like other 2S albumins, that of Arabidopsis is made as a prepropeptide. After removal of the signal peptide, the propeptide is cleaved at four other points, giving two subunits linked by a disulfide bridge(s). Comparison of these cleavage sites with those of 2S albumins of Brassica napus and Bertholletia excelsa suggests that while individual cleavage sites between species are conserved, the four processing sites within a species are not similar, suggesting that up to four different proteases may be involved in processing 2S albumins. The Arabidopsis 2S albumin gene was used to isolate the entire gene family. There are four genes, tightly linked in a tandem array. None of the genes contains an intron. Comparison of the predicted protein sequences shows that only one of the genes can encode the isoform determined by protein analysis to be the most abundant, and therefore this gene is certain to be expressed. It is possible that some or all of the other three genes are also active.

Four genes in two diverged subfamilies encode the ribulose-1,5-bisphosphate carboxylase small subunit polypeptides of Arabidopsis thaliana.

The multigene family encoding the small subunit polypeptides of ribulose-1,5-bisphosphate carboxylase/oxygenase in the crucifer Arabidopsis thaliana has been isolated and the organization and structure of the individual members determined. The family consists of four genes which have been divided into two subfamilies on the basis of linkage and DNA and amino acid sequence similarities. Three of the genes, designated ats1B, ats2B, and ats3B, reside in tandem on an 8 kb stretch of the chromosome. These genes share greater than 95% similarity in DNA sequence and encode polypeptides identical in length and 96.7% similar in amino acid sequence. The fourth gene, ats1A, is at least 10 kb removed from, or completely unlinked to the B subfamily. The B subfamily genes are more similar to each other than to ats1A in nucleotide and amino acid sequence. All four genes are interupted by two introns whose placement within the coding region of the genes is conserved. The introns of the B subfamily genes are similar in length and nucleotide sequence, but show no similarity to the introns of ats1A. Comparison of the DNA sequences within the immediate 5' and 3' flanking sequences among the genes revealed only limited regions of homology. S1 analysis shows that all four genes are expressed.

Molecular analysis of paramutant plants of Antirrhinum majus and the involvement of transposable elements.

Paramutation is observed when the Antirrhinum majus lines 44 and 53 are crossed. These two lines both have insertions at the nivea locus, which encodes chalcone synthase (chs). The allele niv-53 carries the transposable element Tam1 in the promoter region of the chs gene; niv-44 carries the element Tam2 within the gene. The Tam1 element has previously been extensively characterised. Here the Tam2 element is further characterised, and the arrangement of the nivea locus in paramutant plants is analysed. The complete sequence of Tam2, and that of a partial cDNA complementary to it, have been determined. The cDNA is probably transcribed from a different copy of Tam2 from that present at the nivea locus, and does not encode a functional protein. Genomic Southerns of F1 plants from the 53/44 cross show that no major rearrangements are consistently associated with paramutation at the nivea locus of A. majus. The isolation from a paramutant plant arising from a 53/44 cross of an allele (niv-4432) resulting from the excision of Tam2 is reported. The excision of Tam2 resulted in a 32 bp deletion of chs gene sequences. Plants homozygous for the new niv-4432 allele have white flowers and are still paramutagenic, demonstrating that Tam2 need not be present at the nivea locus for paramutation to occur. Different interactions between Tam1 and Tam2 are discussed, and a possible model for paramutation is presented.

DNA sequences for the Zea mays tRNA genes tV-UAC and tS-UGA: tV-UAC contains a large intron.

The chloroplast genome contains genes for a large and probably complete set of tRNAs. These genes are unique in sharing attributes of both nuclear and bacterial tRNA genes. Two chloroplast tRNA genes from Zea mays are described here. tV-UAC, encoding a valine tRNA with the anticodon UAC, contains a 603 bp intron and is highly homologous, both in coding regions and in the intron, to the analogous gene from tobacco described by Deno et al. (Nucleic Acids Res 10:7511-7520, 1982). It is located near the gene for the beta and epsilon subunits of the CF1 complex. (Krebbers et al.: Nucleic Acids Res 10:4985-5002, 1982). The gene tS-UGA, encoding a serine tRNA with the anticodon UGA, is located 41 kbp 3' to tV-UAC. Both genes contain promoter-like sequences in their 5' flanking regions.