Background Understanding developmental processes requires the unambiguous identification of cells and tissues, and the selective manipulation of the properties of those cells and tissues. Both requirements can most efficiently be satisfied through the use of GAL4/GFP enhancer‐trap lines. No such lines, however, have been characterized for the study of early leaf development in the Columbia‐0 reference genotype of Arabidopsis. Results Here we address this limitation by identifying and characterizing a set of GAL4/GFP enhancer‐trap lines in the Columbia‐0 background for the specific labeling of cells and tissues during early leaf development, and for the targeted expression of genes of interest in those cells and tissues. Conclusions By using one line in our set to address outstanding questions in leaf vein patterning, we show that these lines can be used to address key questions in plant developmental biology.

Unlike in animals, in plants vein patterning does not rely on direct cell-cell interaction and cell migration; instead, it depends on the transport of the plant hormone auxin, which in turn depends on the activity of the PIN-FORMED1 (PIN1) auxin transporter. The current hypotheses of vein patterning by auxin transport propose that in the epidermis of the developing leaf PIN1-mediated auxin transport converges to peaks of auxin level. From those convergence points of epidermal PIN1 polarity, auxin would be transported in the inner tissues where it would give rise to major veins. Here we tested predictions of this hypothesis and found them unsupported: epidermal PIN1 expression is neither required nor sufficient for auxin-transport-dependent vein patterning, whereas inner-tissue PIN1 expression turns out to be both required and sufficient for auxin-transport-dependent vein patterning. Our results refute all vein patterning hypotheses based on auxin transport from the epidermis and suggest alternatives for future tests.

AbstractUnderstanding developmental processes requires the unambiguous identification of cells and tissues, and the selective manipulation of the properties of those cells and tissues. Both requirements can most efficiently be satisfied through the use of GAL4/GFP enhancer-trap lines. No such lines are available, however, for the study of leaf development in the Columbia-0 reference genotype of Arabidopsis. Here we address this limitation by identifying and characterizing a set of GAL4/GFP enhancer-trap lines in the Columbia-0 background for the specific labeling of cells and tissues during leaf development, and for the targeted expression of genes of interest in those cells and tissues. By using one line in our set to resolve outstanding problems in leaf vein patterning, we show that these lines can be used to address key questions in plant developmental biology.

Unlike in animals, in plants vein patterning does not rely on direct cell-cell interaction and cell migration; instead, it depends on the transport of the plant signal auxin, which in turn depends on the activity of the PIN-FORMED1 (PIN1) auxin transporter. The current hypotheses of vein patterning by auxin transport propose that in the epidermis of the developing leaf PIN1-mediated auxin transport converges to peaks of auxin level. From those convergence points of epidermal PIN1 polarity, auxin would be transported in the inner tissues where it would give rise to major veins. Here we tested predictions of this hypothesis and found them unsupported: epidermal PIN1 expression is neither required nor sufficient for auxintransport-dependent vein patterning, whereas inner-tissue PIN1 expression turns out to be both required and sufficient for auxintransport-dependent vein patterning. Our results refute all vein patterning hypotheses based on auxin transport from the epidermis and suggest alternatives for future tests.

No abstract