Abstract
Genotypic differences in transpiration rate responses to high vapour pressure deficit (VPD) was earlier reported. Here we tested the hypothesis that this limitation could relate to different degrees of dependence on the apoplastic (spaces between cells), and symplastic water transport pathways (through cells via aquaporin-facilitated transport), which are known to have different hydraulic conductivities. The low transpiration rate (Tr) genotype PRLT 2/89/33 either restricted its transpiration under high VPD, or was more sensitive to VPD than H77/833-2, when grown hydroponically or in soil. The slope of the transpiration response to an ascending series of VPD was lower in whole plants than in de-rooted shoots. In addition, the transpiration response of detached leaves to moderately high VPD (2.67kPa), normalised against leaves exposed to constant VPD (1.27kPa), was similar in low and high Tr genotypes. This suggested that roots hydraulics were a substantial limitation to water flow in pearl millet, especially under high VPD. The dependence on the apoplastic and symplastic water transport pathways was investigated by assessing the transpiration response of plants treated with inhibitors specific to the AQP-mediated symplastic pathway (AgNO3 and H2O2) and to the apoplastic pathway (precipitates of Cu(Fe(CN)6) or Cu(CuFe(CN)6)). When CuSO4 alone was used, Cu ions caused an inhibition of transpiration in both genotypes and more so in H77/833-2. The transpiration of high Tr H77/833-2 was decreased more by AQP inhibitors under low VPD (1.8kPa) than in PRLT 2/89/33, whereas under high VPD (4.2kPa), the transpiration of PRLT 2/89/33 was decreased more by AQP inhibitors than in H77/833-2. The transpiration rate of detached leaves from H77/833-2 when treated with AgNO3 decreased more than in PRLT 2/89/33. Although the root hydraulic conductivity of both genotypes was similar, it decreased more upon the application of a symplastic inhibitor in H77/833-2. The transpiration of low Tr PRLT 2/89/33 was decreased more by apoplastic inhibitors under both low and high VPD. Then the hydraulic conductivity decreased more upon the application of an apoplastic inhibitor in PRLT 2/89/33. In conclusion, both pathways contributed to water transport, and their contribution varied with environmental conditions and genotypes. Roots were a main source of hydraulic limitation in these genotypes of pearl millet, although a leaf limitation was not excluded. The similarity between genotypes in root hydraulic conductivity under normal conditions also suggests changes in this conductivity upon changes in the evaporative demand. The low Tr genotype depended more on the apoplastic pathway for water transport, whereas the high Tr genotype depended on both pathway, may be by 'tuning-up' the symplastic pathway under high transpiration demand, very likely via the involvement of aquaporins.