Principal directions of growth and the generation of cell patterns in wild-type and gib-1 mutant roots of tomato (Lycopersicon esculentum Mill.) grown in vitro

Abstract

The patterns of cell growth and division characteristic of the apex of tomato roots grown in vitro were simulated by computer using a growth tensor (GT). The GT was used to clarify the basis of the altered cell patterns found within apices of roots whose gibberellin levels had been depressed by mutation (at the GIB-1 locus) or through application of the gibberellin-biosynthesis inhibitor, 2S,3S paclobutrazol. At the pole of wild-type roots, where the cell files of the cortex converge, there are commonly only one or two tiers of cortical cells sandwiched between the pole of the stele and the cap initials. By contrast, root apices of the gib-1 mutant contain additional tiers in this region. The development of these additional tiers is suppressed when roots of the mutant are grown in the presence of gibberellic acid (GA3), but could be induced in wild-type roots when they are grown in 2S,3S paclobutrazol. The wild-type cell pattern can be simulated using the GT and by the application of appropriate rules that govern cell growth and division. The induced variations in cell pattern are interpreted as being due to displacements, within the apex, of the principal directions of growth (PDGs), which are represented, in part, by the set of periclines and anticlines seen in the cell wall network; these, in turn, are utilized in the specification of the GT. During normal (wild-type) root growth, the PDGs maintain a stable pattern and the corresponding cell pattern is also stable. However, in order to interpret the cellular behaviour found in wild-type roots grown in 2S, 3S paclobutrazol, simulation using the GT shows that, if the pattern of PDGs is destabilized and displaced distally along the root axis, the cell pattern reorganizes into that typical of gib-1 mutant roots. Conversely, the cell pattern of gib-1 roots, which reverts to wild-type upon exposure to GA3, can be simulated if the PDGs are displaced proximally to the inside of the apex whereupon the number of cortical tiers at the root pole decreases. These results suggest a link between endogenous gibberellin level and the specification of the PDGs in the growing tomato root apex. Furthermore, the evidence of cell patterns from gib-1 roots suggests that, in order to achieve stability of PDGs with concomitant stable cellular patterning, an optimal gibberellin level is necessary. In practice, this can be attained by culturing the mutant roots in medium containing 1 μM GA3.