No abstract

Title page 1 INTRODUCTIONThe description of plant form and shape is useful to understand evolutionary relationships and quantify differences between species, as well as to define the processes of plant development over space and time. The scientific discipline of Plant Morphology (Kaplan, 2001), however, needs to be mathematically and not solely descriptively based. Seed spacing in a sunflower head, or rings swirls on a pineapple fruit, whose distribution follows Fibonacci series, are good examples, while others remain awaiting discovery. Computational modeling of plant development has been applied to legume lateral root development (Han et al, 2011) as well as nodule pattern (Han et al, 2010b). Such mathematical and computational basis permits the in silico testing of thousands of experimental conditions to achieve outcomes which then need to be verified in "real world" experiments.Recently we described a mathematical-geometrical model to quantify the shape of Arabidopsis seeds. It consisted in the comparison of the outline of the longitudinal equatorial section of the seed with a transformed cardioid (Cervantes et al., 2010). The cardioid is the trajectory described by a point of a circle that rolls around another fixed circle with the same radius, and transforming it by multiplying the x-axis by the scaling factor  (  1.61803399), the so-called "Golden Ratio", the figure obtained resembled the image of an Arabidopsis dry seed. The similarity was evaluated by a factor, the G index, allowing the precise morphological description of genotypes and opening new ways for the association between genetic changes and alterations in the morphology, i.e., a classical genome-phenome analysis. (Cervantes et al., 2010). The functional connection between ethylene perception and seed shape is unclear but may involve diverse aspects of seed development as well as drying rates.Legume seeds offer an excellent material to expand this type of analysis in search of accurate morphological descriptions. Legumes are a large family containing members of great agronomic interest (e.g., soybean, pea, common bean, medics, clovers as well as trees such as Acacias, Robinia and Pongamia; Biswas et al., 2011). In the context of seed morphology, legume species are characterized by remarkable variation, ranging from small seeds such as Lotus japonicus to large-sized seeds of Phaseolus or Vicia and gigantic ones (7-8 cm long) in Black Bean Tree (Castanospermum australe) in Australia.Interestingly, Lotus japonicus and Medicago truncatula, two model species intensively used in genomic analysis (Jiang and Gresshoff, 1997;Sato et al., 2008; Cannon et al., 2006; Young et al., 2005), are amenable to morphological analysis based on a simple model resembling the one described for Arabidopsis, i.e., the comparison with a cardioid. Whereas the seed dimensions of Lotus japonicus match well to a cardioid, in Medicago truncatula accurate adjustment is obtained by the simple transformation of scaling the vertical axis by a factor equal to the Golden Rati...

An ABA insensitive mutant, Beyma, was isolated in Lotus japonicus MG-20 from an EMS mutagenesis population using root growth inhibition to applied ABA as the screening criterion. (The name 'Beyma' was taken from the Australian Aboriginal language, Wagiman, beyma, meaning 'drying up'.) The stable mutant that segregates as a dominant Mendelian mutation is insensitive to ABA induced inhibition of germination, vegetative growth, stomatal opening, as well as nodulation. Tissue ABA levels were normal, suggesting a sensitivity rather than biosynthesis mutation. It is slow-growing (50-70% of wild-type MG-20) and has a near-constitutive wilty phenotype associated with its inability to regulate stomatal opening. Whilst showing a wide range of ABA insensitive phenotypes, Beyma did not show alteration of nodule number control, as, in the absence of added ABA, the number and patterning (but not size) of nodules formed in the mutant were similar to that of MG-20. Split root experiments on MG-20 showed that application of ABA on one side of the root inhibited nodulation locally but not systemically. We propose that ABA is not involved directly in systemic autoregulation of nodulation (AON).

Three independent ethylene insensitive mutants were selected from an EMS-mutagenized population of Lotus japonicus MG-20 (Miyakojima). The mutants, called 'Enigma', were mutated in the LjEIN2a gene from Lotus chromosome 1, sharing significant homology with Arabidopsis EIN2 (ethylene-insensitive2). All three alleles showed classical ethylene insensitivity phenotypes (e.g., Triple Response), but lacked the increased nodulation phenotype commonly associated with ethylene insensitivity. Indeed, all showed a marginal reduction in nodule number per plant, a phenotype that is enigmatic to sickle, an ethyleneinsensitive EIN2 mutant in Medicago truncatula. In contrast to wild type, but similar to an ETR1-1 ethylene ethylene-insensitive transgenic of L. japonicus, enigma mutants formed nodules in between the protoxylem poles, demonstrating the influence of ethylene on radial positioning. Suppression of nodule numbers by nitrate and colonisation by mycorrhizal fungi in the enigma-1 mutant were indistinguishable from the wild-type MG-20. However, reflecting endogenous ethylene feedback, the enigma-1 mutant released more than twice the wild-type amount of ethylene. enigma-1 had a moderate reduction in growth, greater root mass (and lateral root formation), delayed flowering and ripening, smaller pods and seeds. Expression analysis of ethylene-regulated genes, such as ETR1, NRL1 (neverripe-like 1), and EIL3 in shoots and roots of enigma-1 and MG-20 illustrated that the ethylene-insensitive mutation strongly affected transcriptional responses in the root. These mutants open the possibility that EIN2 in L. japonicus, a determinate nodulating legume, acts in a more complex fashion possibly through the presence of a duplicated copy of LjEIN2.