Activation tagging and insertional mutagenesis in barley.

The process of activation tagging in plants involves the random distribution of plant regulatory sequences throughout the genome. The insertion of a regulatory sequence in the vicinity of an endogenous gene can alter the transcriptional pattern of this gene resulting in a mutant phenotype that arises from excess functional gene product. Activation tagging has been undertaken extensively in a number of dicot plants and also in rice. This has been achieved primarily by high-throughput plant transformation using T-DNA sequences that encode regulatory elements. Apart from rice, most cereals do not have a suitably efficient transformation system for high-throughput transformation. In this article, we detail an activation tagging system in barley that exploits the mobility of the maize Ac/Ds transposable element system to distribute a highly expressed promoter throughout the barley genome. The advantage of this approach in this species is that a relatively small number of primary transgenics are required to generate an activation tagging population. Insertion of this transposable element into genes can also generate insertional inactivation mutants enabling both gene overexpression and gene knockout mutants to be identified in the same population.

Suppression of the barley uroporphyrinogen III synthase gene by a Ds activation tagging element generates developmental photosensitivity.

Chlorophyll production involves the synthesis of photoreactive intermediates that, when in excess, are toxic due to the production of reactive oxygen species (ROS). A novel, activation-tagged barley (Hordeum vulgare) mutant is described that results from antisense suppression of a uroporphyrinogen III synthase (Uros) gene, the product of which catalyzes the sixth step in the synthesis of chlorophyll and heme. In homozygous mutant plants, uroporphyrin(ogen) I accumulates by spontaneous cyclization of hydroxyl methylbilane, the substrate of Uros. Accumulation of this tetrapyrrole intermediate results in photosensitive cell death due to the production of ROS. The efficiency of Uros gene suppression is developmentally regulated, being most effective in mature seedling leaves compared with newly emergent leaves. Reduced transcript accumulation of a number of nuclear-encoded photosynthesis genes occurs in the mutant, even under 3% light conditions, consistent with a retrograde plastid-nuclear signaling mechanism arising from Uros gene suppression. A similar set of nuclear genes was repressed in wild-type barley following treatment with a singlet oxygen-generating herbicide, but not by a superoxide generating herbicide, suggesting that the retrograde signaling apparent in the mutant is specific to singlet oxygen.

Transfer of plastid DNA to the nucleus is elevated during male gametogenesis in tobacco.

In eukaryotes, many genes were transferred to the nucleus from prokaryotic ancestors of the cytoplasmic organelles during endosymbiotic evolution. In plants, the transfer of genetic material from the plastid (chloroplast) and mitochondrion to the nucleus is a continuing process. The cellular location of a kanamycin resistance gene tailored for nuclear expression (35SneoSTLS2) was monitored in the progeny of reciprocal crosses of tobacco (Nicotiana tabacum) in which, at the start of the experiments, the reporter gene was confined either to the male or the female parental plastid genome. Among 146,000 progeny from crosses where the transplastomic parent was male, 13 transposition events were identified, whereas only one atypical transposition was identified in a screen of 273,000 transplastomic ovules. In a second experiment, a transplastomic beta-glucuronidase reporter gene, tailored to be expressed only in the nucleus, showed frequent stochastic expression that was confined to the cytoplasm in the somatic cells of several plant tissues. This gene was stably transferred in two out of 98,000 seedlings derived from a male transplastomic line crossed with a female wild type. These data demonstrate relocation of plastid DNA to the nucleus in both somatic and gametophytic tissue and reveal a large elevation of the frequency of transposition in the male germline. The results suggest a new explanation for the occurrence of uniparental inheritance in eukaryotes.

A barley activation tagging system.

Activation tagging, as the result of random genomic insertion of either promoter or enhancer sequences, can produce novel, dominant mutations by over-expression of endogenous genes. This powerful genomics tool has been used extensively in dicot species such as Arabidopsis, while rice is the only cereal for which an equivalent system exists. In this study we describe an activation tagging system in barley based upon the maize Ac/Ds transposable element system. A modified Ds element (UbiDs) containing two maize polyubiquitin promoters, transposed in families derived from multiple independent UbiDs transformants and generated new Ds insertion events at frequencies ranging from 0% to 52% per family. The majority of transposed UbiDs elements activated high levels of adjacent flanking sequence transcription. Transposon-mediated expression was detected in all barley cell and tissue types analysed suggesting that this system is applicable to all aspects of plant development and biogenesis. In addition to transcriptional activation, this system is also capable of generating insertional knockout mutants and a UbiDs inactivated allele of the granule bound starch synthase I gene (waxy) was recovered that lead to reduced amylose accumulation. The recovery and analysis of dominant over-expression phenotypes generated by this system will provide a novel approach to understanding gene function in large cereal genomes where gene redundancy may mask conventional loss-of-function mutations.

A rust-inducible gene from flax (fis1) is involved in proline catabolism.

A gene fis1 from flax (Linum usitatissimum), which is induced in mesophyll cells at the site of rust (Melampsora lini) infection, is also expressed in vascular tissue, particularly in floral structures of healthy plants. This paper reports that the promoter controlling this expression is contained within 282 bp 5' to the coding region and that fis1 gene induction is specifically by the rust pathogen and not by other fungal pathogens or by wounding. The fis1 gene has 73% homology with an Arabidopsis gene which encodes delta-1-pyrroline-5-carboxylate dehydrogenase (P5CDH) which is a part of the proline degradation pathway. Transgenic flax plants that either over-express fis1 or show reduced fis1 expression due to RNA-mediated gene silencing have an unaltered morphology. However, plants with reduced fis1 expression have markedly increased sensitivity to exogenous proline and show alteration in epidermal cell morphology, callose deposition and the production of hydrogen peroxide during proline-induced death. These lines, which show a biologically significant level of fis1 suppression, have an unaltered reaction to either virulent or avirulent rust infections, as do fis1 over-expression lines. These data indicate that the fis1 gene plays a role in proline metabolism and most likely encodes for a P5CDH enzyme. However, the precise role of fis1 and P5C catabolism in the development of rust disease remains unclear.

Comparative analysis in cereals of a key proline catabolism gene.

Proline accumulation and catabolism play significant roles in adaptation to a variety of plant stresses including osmotic stress, drought, temperature, freezing, UV irradiation, heavy metals and pathogen infection. In this study, the gene Delta1 -pyrroline-5-carboxylate dehydrogenase (P5CDH), which catalyzes the second step in the conversion of proline to glutamate, is characterized in a number of cereal species. P5CDH genes from hexaploid wheat, Triticum turgidum (durum wheat), Aegilops tauschii, Triticum monococcum, barley, maize and rice were shown to be conserved in terms of gene structure and sequence, present as a single copy per haploid, non-polyploid genome and located in evolutionarily conserved linkage groups. A wheat cDNA sequence was shown by yeast complementation to encode a functional P5CDH activity. A divergently-transcribed rab7 gene was identified immediately 5' of P5CDH in all grasses examined, except rice. The rab7/P5CDH intergenic region in these species, which presumably encompasses 5' regulatory elements of both genes, showed a distinct pattern of sequence evolution with sequences in juxtaposition to each ORF conserved between barley, wheat, A. tauschii and T. monococcum. More distal 5' sequence in this intergenic region showed a higher rate of divergence, with no homology observed between these regions in the wheat and barley genomes. Maize and rice showed no similarity in regions 5' of P5CDH when compared with wheat, barley, and each other, apart from a 22 bp region of conserved non-coding sequence (CNS) that is similar to a proline response element identified in the promoter of the Arabidopsis proline dehydrogenase gene. A palindromic motif similar to this cereal CNS was also identified 5' of the Arabidopsis AtP5CDH gene showing conservation of this sequence in monocot and dicot lineages.

The gene product specified by a double stranded RNA in the flax rust (Melampsora lini) is expressed during rust growth.

Strain SP6 of the flax rust (Melampsora lini) contains 11 double-stranded RNAs (dsRNAs) of unknown function. A large open-reading frame (B3ORF1 in dsRNA B3 encodes a polypeptide of 614 amino acids, and using an antiserum raised against the B3ORF1-glutathione S-transferase fusion protein prepared from a bacterial expression system, we have detected the presence of a 67 kDa polypeptide in rust urediospores. This polypeptide, identical in size to that of the predicted translation product of B3ORF1 , was not detected in spores from either a fungal strain lacking the B3 dsRNA or an isogenic strain containing no dsRNA. These data indicate that B3ORF1 present in the flax rust B3 dsRNA is expressed in vivo which warrants farther investigation in search for its function during rust development.

Allelic and haplotypic diversity at the rp1 rust resistance locus of maize.

The maize Rp1 rust resistance locus is a complex consisting of a family of closely related resistance genes. The number of Rp1 paralogs in different maize lines (haplotypes) varied from a single gene in some stocks of the inbred A188 to >50 genes in haplotypes carrying the Rp1-A and Rp1-H specificities. The sequences of paralogs in unrelated haplotypes differ, indicating that the genetic diversity of Rp1-related genes is extremely broad in maize. Two unrelated haplotypes with five or nine paralogs had identical resistance phenotypes (Rp1-D) encoded in genes that differed by three nucleotides resulting in a single amino acid substitution. Genes in some haplotypes are more similar to each other than to any of the genes in other haplotypes indicating that they are evolving in a concerted fashion.

Aberrant mRNA processing of the maize Rp1-D rust resistance gene in wheat and barley.

The maize Rp1-D gene confers race-specific resistance against Puccinia sorghi (common leaf rust) isolates containing a corresponding avrRp1-D avirulence gene. An Rp1-D genomic clone and a similar Rp1-D transgene regulated by the maize ubiquitin promoter were transformed independently into susceptible maize lines and shown to confer Rp1-D resistance, demonstrating that this resistance can be transferred as a single gene. Transfer of these functional transgenes into wheat and barley did not result in novel resistances when these plants were challenged with isolates of wheat stem rust (P. graminis), wheat leaf rust (P. triticina), or barley leaf rust (P. hordei). Regardless of the promoter employed, low levels of gene expression were observed. When constitutive promoters were used for transgene expression, a majority of Rp1-D transcripts were truncated in the nucleotide binding site-encoding region by premature polyadenylation. This aberrant mRNA processing was unrelated to gene function because an inactive version of the gene also generated such transcripts. These data demonstrate that resistance gene transfer between species may not be limited only by divergence of signaling effector molecules and pathogen avirulence ligands, but potentially also by more fundamental gene expression and transcript processing limitations.