The Lotus japonicus LjSym4 gene is required for the successful symbiotic infection of root epidermal cells.

The role of the Lotus japonicus LjSym4 gene during the symbiotic interaction with Mesorhizobium loti and arbuscular mycorrhizal (AM) fungi was analyzed with two mutant alleles conferring phenotypes of different strength. Ljsym4-1 and Ljsym4-2 mutants do not form nodules with M. loti. Normal root hair curling and infection threads are not observed, while a nodC-dependent deformation of root hair tips indicates that nodulation factors are still perceived by Ljsym4 mutants. Fungal infection attempts on the mutants generally abort within the epidermis, but Ljsym4-1 mutants allow rare, successful, infection events, leading to delayed arbuscule formation. On roots of mutants homozygous for the Ljsym4-2 allele, arbuscule formation was never observed upon inoculation with either of the two AM fungi, Glomus intraradices or Gigaspora margarita. The strategy of epidermal penetration by G. margarita was identical for Ljsym4-2 mutants and the parental line, with appressoria, hyphae growing between two epidermal cells, penetration of epidermal cells through their anticlinal wall. These observations define a novel, genetically controlled step in AM colonization. Although rhizobia penetrate the tip of root hairs and AM fungi access an entry site near the base of epidermal cells, the LjSym4 gene is necessary for the appropriate response of this cell type to both microsymbionts. We propose that LjSym4 is required for the initiation or coordinated expression of the host plant cell's accommodation program, allowing the passage of both microsymbionts through the epidermis layer.

A plant regulator controlling development of symbiotic root nodules.

Symbiotic nitrogen-fixing root nodules on legumes are founded by root cortical cells that de-differentiate and restart cell division to establish nodule primordia. Bacterial microsymbionts invade these primordia through infection threads laid down by the plant and, after endocytosis, membrane-enclosed bacteroids occupy cells in the nitrogen-fixing tissue of functional nodules. The bacteria excrete lipochitin oligosaccharides, triggering a developmental process that is controlled by the plant and can be suppressed. Nodule inception initially relies on cell competence in a narrow infection zone located just behind the growing root tip. Older nodules then regulate the number of nodules on a root system by suppressing the development of nodule primordia. To identify the regulatory components that act early in nodule induction, we characterized a transposon-tagged Lotus japonicus mutant, nin (for nodule inception), arrested at the stage of bacterial recognition. We show that nin is required for the formation of infection threads and the initiation of primordia. NIN protein has regional similarity to transcription factors, and the predicted DNA-binding/dimerization domain identifies and typifies a consensus motif conserved in plant proteins with a function in nitrogen-controlled development.

Symbiotic mutants deficient in nodule establishment identified after T-DNA transformation of Lotus japonicus.

Nitrogen-fixing root nodules develop on legumes as a result of an interaction between host plants and soil bacteria collectively referred to as rhizobia. The organogenic process resulting in nodule development is triggered by the bacterial microsymbiont, but genetically controlled by the host plant genome. Using T-DNA insertion as a tool to identify novel plant genes that regulate nodule ontogeny, we have identified two putatively tagged symbiotic loci, Ljsym8 and Ljsym13, in the diploid legume Lotus japonicus. The sym8 mutants are arrested during infection by the bacteria early in the developmental process. The sym13 mutants are arrested in the final stages of infection, and ineffective nodules are formed. These two plant mutant lines were identified in progeny from 1112 primary transformants obtained after Agrobacterium tumefaciens T-DNA-mediated transformation of L. japonicus and subsequent screening for defects in the symbiosis with Mesorhizobium loti. Additional nontagged mutants arrested at different developmental stages were also identified and genetic complementation tests assigned all the mutations to 16 monogenic symbiotic loci segregating recessive mutant alleles. In the screen reported here independent symbiotic loci thus appeared with a frequency of approximately 1.5%, suggesting that a relatively large set of genes is required for the symbiotic interaction.

Gene targeting approaches using positive-negative selection and large flanking regions.

We report here on strategies aimed at improving the frequency of detectable recombination in plants by increasing the efficiency of selecting double-recombinants in transgenic calli. Gene targeting was approached on the Gln1 and the Pzfloci of Lotus japonicus, using Agrobacterium tumefaciens T-DNA replacement vectors. Large flanking regions, up to 22.9 kb, surrounding a positive selection marker were presented as substrates for homologous recombination. For easier detection of putative recombinants the negative selectable marker cytosine deaminase was inserted at the outside borders of the flanking regions offered for cross-over. A combination of positive and negative selection allowing double-recombinants to grow, while counter-selecting random insertions, was used to select putative targeting events. The more than 1000-fold enrichment observed with replacement vectors designed to minimize gene silencing demonstrated the efficiency of the negative selection. Using five different replacement vectors an estimated total of 18,974 transformation events were taken through the positive-negative selection procedure and 185 resistant calli obtained. Targeting events could not be verified in the survivors by PCR screening and Southern blot analysis. With this approach the frequency of detectable gene targeting in L. japonicus was below 5.3 x 10(-5), despite the large flanking sequences offered for recombination.