Résumé
A single plant contains at least thirty different MIP genes, most of which appear to be expressed under normal growth conditions. The expression of some may be linked to specific physical stresses such as water deficit. The proteins are found in the tonoplast (vacuolar membrane) and the plasma membrane. It is not known whether all these genes encode aquaporins (or aqua-glyceroporins) because the activities of many have not yet been tested. On a phylogenetic tree we can distinguish the tonoplast aquaporins (TIPs) from the plasma membrane aquaporins (PIPs) and among the latter there are different sub-families. The activity of putative aquaporins has been tested almost exclusively in heterologous systems, primarily in Xenopus oocytes. In planta function must be inferred from such tests. TIPs appear to always be very active when expressed in oocytes, and among the PIPs, the PIP2 family is active, but the PIP1 family is largely inactive. This inactivity is not caused by the failure of these proteins to be transported to the plasma membrane (see Chaumont et al., this volume) but could be due to the inability of oocytes to activate PIPs. Such findings point up the need to assay aquaporin activity in planta. Because plants have so many aquaporin genes, single knockouts are probably not a particularly effective way to understand function. However, using an antisense construct, it is possible to downregulate an entire family of sequence related genes.