Cell expansion coordinates leaf vein and stomatal density

The efficiency with which water is delivered through leaf tissue to sites of evaporation (the leaf hydraulic conductance) is an important determinant of photosynthetic capacity. This is because inadequate water supply forces stomata to close due to plant desiccation. The resulting reduction in stomatal conductance restricts carbon dioxide (CO2) uptake. An oversupply of water, on the other hand, results in unprofitable investment in vascular tissue. Accordingly, water supply is strongly correlated with the demand for water generated by stomata. Recent work suggests that this balance is achieved in a tropical angiosperm tree by a simple developmental mechanism in which changes to leaf size co-regulate vein and stomatal density (major determinants of water supply and demand, respectively). However, the generality of this mechanism is unknown. Thus, in this thesis I set out to establish whether there is a general developmental mechanism that allows plants to maintain a constant ratio between leaf vein and stomatal density. In doing so the following specific research questions were investigated: 1. Does acclimation to high and low VPD (vapour pressure difference) modify the relationships between vein density, stomatal density and leaf size? (Chapter 2) 2. Does leaf or epidermal cell expansion coordinate vein and stomatal density in herbs and woody angiosperms? (Chapter 3) 3. Does epidermal cell expansion coordinate vein and stomatal density in older plant lineages like ferns? (Chapter 4) 4. Is vein density directly regulated by epidermal cell expansion? (Chapter 5) The difference in vapour pressure between leaf tissue and the atmosphere determines how much water is transpired at any given stomatal conductance. Thus, I hypothesised in Chapter 2 that plants grown under high VPD would exhibit a modified relationship between vein and stomatal density resulting in maintenance of homeostasis in leaf water potential such that there is either a decrease in stomatal density and no change in vein density, or no change in stomatal density and an increase in vein density. This would also disrupt previously observed relationships with leaf size. However, contrary to my hypothesis, I found a small but coordinated increase in vein and stomatal density in plants grown under high VPD compared with those grown under low VPD. Furthermore, densities of veins and stomata were independent of large VPD-induced changes to leaf size and were instead limited by epidermal cell size (which was fairly insensitive to VPD). This suggests that significant modification of epidermal cell size is required to produce large changes in vein and stomatal density. Thus, to further investigate whether leaf expansion or epidermal cell expansion coordinates vein and stomatal density more generally among plants I examined relationships between vein density, stomatal density, leaf size and epidermal cell size across a diverse range of woody and herbaceous angiosperms grown under sun and shade conditions (Chapter 3). Contrary to the original premise that differential leaf expansion coordinates vein and stomatal density, I found that leaf size was independent of epidermal cell size in most cases and that relationships between vein density, stomatal density and epidermal cell size were well described by modelled relationships that assume veins and stomata are passively 'diluted' by epidermal cell expansion. These results demonstrate that coordination of vein and stomatal density in angiosperms is driven by their co-variation with epidermal cell size. While this 'passive dilution' mechanism seems to be common among angiosperms, it is not known when it appeared in the vascular plant phylogeny. Furthermore, older plant lineages like ferns may employ different mechanisms to coordinate vein and stomatal density because they differ from angiosperms in both current water relations physiology and evolutionary history. Contrary to this expectation, I found that relationships between vein density, stomatal density and epidermal cell size across four ferns species were very similar to those previously observed across angiosperm species (Chapter 4). However, there was little plasticity in these traits within fern species and changes to stomatal density across species were actively regulated by stomatal index, as well as epidermal cell size. Despite this, epidermal cell size was a strong determinant of vein and stomatal density in ferns (explaining 55.5 % of the variation in stomatal density versus the 44.5 % explained by stomatal index). Thus, ferns (like angiosperms) appear to use the co-variance of vein and stomatal density with epidermal cell expansion to maintain a constant ratio between the abundance of veins and stomata in the leaf. This suggests that the 'passive dilution' mechanism may be an ancient feature of vascular plants that co-regulates these tissues. In Chapters 3 and 4 it was proposed that coordination of vein and stomatal density is achieved through a 'passive dilution' mechanism in which densities of veins and stomata are co-regulated by epidermal cell size. However, unlike stomata, leaf veins are spatially isolated from the epidermis and it is not known whether they are directly regulated by epidermal cell expansion. To investigate this I tested whether relationships between vein density and epidermal cell size in a wild type genotype of Arabidopsis (Col-0) and seven other genotypes with modified forms of genes that affect both stomatal development and epidermal cell size were the same as modelled relationships that assume veins are passively 'diluted' by epidermal cell expansion (Chapter 5). Vein density in wild type plants was correlated with abaxial epidermal cell size in a way that was consistent with the 'passive dilution' mechanism (despite some deviation from modelled relationships). However, vein density was independent of variation in epidermal cell size among mutant and transgenic genotypes. This suggests that epidermal cell size in these genotypes was modified independently from the rest of the leaf in spite of prior evidence that cell sizes are correlated within leaves. Thus, vein density is not causally linked to epidermal cell expansion. Instead the relationship between vein density and epidermal cell size in wild type plants reflects developmental factors that affect both mesophyll and epidermal cells, suggesting that adaptation favours coordination of veins and stomata over independent development of these tissues. Thus, the overarching finding of this thesis is that the 'dilution' of veins and stomata by differential epidermal cell expansion (and perhaps mesophyll cell expansion) appears to be a general mechanism capable of maintaining a constant ratio between vein and stomatal density both within and across a diverse range of vascular plant species. Leaf cells expand more in the shade than the sun, and more in some species compared with others, which increases the space between veins and stomata concomitantly reducing their density. These relationships provide an insight into how plants construct leaves that can efficiently replace transpired water and maintain maximum carbon assimilation for the minimum investment in vein and stomatal infrastructure. Achieving this maximises the energetic return for investments made during leaf construction increasing the energy available for growth and reproduction.