SummaryDirectional transport of the phytohormone auxin is required for the establishment and maintenance of plant polarity, but the underlying molecular mechanisms have not been fully elucidated. Plant homologs of human multiple drug resistance/P-glycoproteins (MDR/PGPs) have been implicated in auxin transport, as defects in MDR1 (AtPGP19) and AtPGP1 result in reductions of growth and auxin transport in Arabidopsis (atpgp1, atpgp19), maize (brachytic2) and sorghum (dwarf3). Here we examine the localization, activity, substrate specificity and inhibitor sensitivity of AtPGP1. AtPGP1 exhibits non-polar plasma membrane localization at the shoot and root apices, as well as polar localization above the root apex. Protoplasts from Arabidopsis pgp1 leaf mesophyll cells exhibit reduced efflux of natural and synthetic auxins with reduced sensitivity to auxin efflux inhibitors. Expression of AtPGP1 in yeast and in the standard mammalian expression system used to analyze human MDR-type proteins results in enhanced efflux of indole-3-acetic acid (IAA) and the synthetic auxin 1-naphthalene acetic acid (1-NAA), but not the inactive auxin 2-NAA. AtPGP1-mediated efflux is sensitive to auxin efflux and ABC transporter inhibitors. As is seen in planta, AtPGP1 also appears to mediate some efflux of IAA oxidative breakdown products associated with apical sites of high auxin accumulation. However, unlike what is seen in planta, some additional transport of the benzoic acid is observed in yeast and mammalian cells expressing AtPGP1, suggesting that other factors present in plant tissues confer enhanced auxin specificity to PGP-mediated transport.

These authors contributed equally to this work. SummaryTo facilitate glucocorticoid-inducible transgene expression from the pOp promoter in Arabidopsis the ligandbinding domain of a rat glucocorticoid receptor (GR LBD) was fused to the amino terminus of the synthetic transcription factor LhG4 to generate LhGR-N. Fusions bearing the GR LBD at other positions in LhG4 exhibited incomplete repression or inefficient induction. LhGR-N was stringently repressed in the absence of exogenous glucocorticoid but was fully activated by addition of 2 lM dexamethasone which resulted in 1000-fold increase in GUS reporter activity. Half maximal induction was achieved with 0.2 lM dexamethasone. Reporter transcripts were detectable within 2 h of dexamethasone application and peaked 4-10 h later. Neither LhGR-N nor dexamethasone affected seedling development although ethanol retarded development when used as a solvent for dexamethasone. The efficiency of the pOp target promoter was improved 10-to 20-fold by incorporating six copies of the ideal lac operator with sufficient inter-operator spacing to allow simultaneous occupancy. Introduction of the TMV X sequence into the 5¢UTR resulted in a further 10-fold increase in dexamethasone-inducible reporter activity and an increase in the induction factor to 10 4 . Although promoters containing the TMV X sequence exhibited slightly increased basal expression levels in the absence of dexamethasone, stringent regulation of the cytokinin biosynthetic gene ipt was achieved with all promoters. Despite the severity of the induced ipt phenotypes, transcripts for the KNOX homoeodomain transcription factors BREVIPEDICELLUS and SHOOTMERISTEMLESS were not significantly increased within 48 h of dexamethasone application to seedlings.