Plant architecture is determined by the initiation and outgrowth of axillary structures at different positions along the stem. Different axillary structures are initiated during each phase of growth; in pea (Pisum sativum L.) vegetative buds are formed in the vegetative phase, and axillary inflorescences in the reproductive phase. Hence, because it determines the number of vegetative nodes, the timing of the floral transition has a large impact on shoot architecture. Further to the vegetative-reproductive phase change, the newly identified sαx (suppressed axillary meristem) mutants of pea appear to highlight developmental zones within the vegetative phase. The sαx mutants lack axillary buds at many nodes in a central zone of the stem. This thesis reports the characterisation of the sαx mutants and the investigation of the regulation of vegetative zones in relation to the floral transition.
All six sαx mutant lines developed axillary buds at basal nodes and at aerial nodes immediately prior to the node of flower initiation (NFI), leaving a central zone of "bare" axils. Allelism tests resolved that these lines represented three new loci. The sαx1 mutants appeared non-pleiotropic whereas the sαx2 mutants displayed a "bunched tendrils" leaf phenotype and the sαx3 mutant appeared to have a subtle pleiotropic effect on stipule shape and colour. The SΑX1 locus was mapped to linkage group HI and the SΑX3 locus to linkage group IV on the pea molecular map (Laucou et al., 1998).
The phenotypic characterisation of the sαx mutants focused primarily on the sαx1-1 mutant because it displayed a non-pleiotropic phenotype. Histological investigation of the bare axils in sαx1 plants suggested that axillary bud initiation was blocked at an early step of axillary bud development. Decapitation and exogenous cytokinin application could not force axillary bud outgrowth from bare axils in sαx1 plants. Likewise, grafting to WT rootstocks could not restore sαx1 scions to a WT phenotype. Unlike the rev mutant of Arabidopsis thaliana, radial stem structure and the rate and amount of auxin transported by sαx1 and sαx3 plants was not significantly different from that of WT plants.
To determine whether the axillary buds developed in the basal zone were altered in the sαx1 and sαx2 mutants, the initiation and outgrowth of buds at node 2 were investigated. WT plants pre-formed four buds at node 2 and could be induced to initiate at least another four following decapitation and successive removal of outgrowing buds. Treatments that stimulated bud outgrowth did not alter the number of axillary buds initiated in WT plants. sαx1 mutants pre-formed a similar number of axillary buds to WT at node 2. Outgrowth of these buds, stimulated by SD conditions or the rms4 mutation, was similar to that of WT buds. However, even when bud outgrowth was stimulated, sαxl mutants rarely initiated further axillary buds at node 2. sαx2 mutants pre-formed fewer axillary buds at node 2 than WT and, like the sαx1 mutant, did not initiate further axillary buds at node 2. The length of the axillary buds at basal nodes in sαx2 mutants appeared reduced early in their development compared with WT and sαx1 mutants.
The regulation of the first aerial node with an axillary bud (FAN) in relation to the floral transition was investigated in sαx1 mutants. The veg1 mutation, which completely inhibits flowering, did not affect the FAN. Further, the det and veg2 mutations, which affect inflorescence meristem identity, also did not affect the FAN, suggesting that this aerial zone represents a level of commitment towards flowering rather than a consequence of flowering.
However, some genes which determine the NFl also appeared to regulate the FAN. Grafting of sαx1 scions to sn mutant rootstocks, the phenotype of 5ax7 sn double mutants, and comparisons of plants grown under long and short photoperiods indicated that the FAN is regulated by the STV-controlled mobile flowering inhibitor. The HR allele delayed the NFI in sαx1 plants but did not delay the FAN. If the FAN is regulated by the inhibitor, this indicates that HR may affect the level of flowering inhibitor after the FAN threshold is met. Double mutants between sαxl and mutant alleles of LF, a gene which appears to determine shoot sensitivity to the flowering signals and thus the NFI, suggested that LF also regulates the FAN. Grafting of the double mutants between sαx1 and the LF series to sn mutant rootstocks suggested that the roles of SN and LF in the regulation of FAN are additive. The GI-controlled flowering promoter did not appear to regulate vegetative development because in sαx1 gi double mutants, where flowering was delayed or prevented, the FAN was not significantly altered.
Changes in meristem identity that may be associated with the FAN and the NFI were tested using the putative inflorescence marker gene DET. DET expression was clearly up-regulated in WT reproductive apices. In comparable vegetative gi-1 apices, DET expression mirrored the up-regulation observed in reproductive GI apices despite their lack of flower initiation. This indicates a change in apical meristem state to a pre-inflorescence identity prior to the initiation of flowers.
This thesis uses the sαx1 mutation to emphasise the roles of the pea flowering genes in vegetative development and suggests that the aerial zone represents a pre-inflorescence apical meristem identity. Future molecular analyses will stem from preliminary studies in this thesis and will allow developmental states underlying the current morphological description of the floral transition and inflorescence structure to be investigated.