Most instrumental in identifying molecular components of PAT were genetic approaches, especially in Arabidopsis thaliana. Various screening strategies have been successfully applied to identify mutants affected in PAT. Some mutants have been selected on the basis of abnormal responses to auxin transport inhibitors or were identified fortuitously in screens for developmental alterations and only later was the connection to PAT discovered (Friml & Palme, 2002).
A mutant called auxin1 (aux1), which confers a root agravitropic and auxin-resistant phenotype, was instructive for identification of a gene possibly encoding an auxin influx carrier. The AUX1 gene encodes a 485 amino acid long protein sharing significant similarity with plant amino acid permeases consistent with the role for AUX1 in the uptake of the tryptophan-like IAA (Bennett et al., 1996). So far the definitive biochemical proof of AUX1 function as an auxin uptake carrier is lacking, but several lines of evidence (mainly based on detailed analysis of the aux1 phenotype) strongly support that AUX1 is required for auxin influx. Strikingly, the aux1 root agravitropic phenotype can be restored by treatment with a membrane permeable auxin NAA in contrast to less permeable 2,4-D. Moreover, this rescue coincides with restoration of basipetal auxin transport, which is defective in aux1 (Yamamoto & Yamamoto, 1998; Marchant et al., 1999). In addition, the main features of the aux1 phenotype can be mimicked by growing seedlings on inhibitors of auxin influx (Parry et al., 2001). Other evidence that AUX1 participates in auxin influx came from auxin uptake assays in aux1 and wild-type roots. They revealed that aux1 roots accumulated significantly less radioactively labeled 2,4-D than did wild-type, and that this difference was not found when the membrane-permeable 1-NAA or the IAA-like amino acid tryptophan were assayed (Marchant et al., 1999). Recently, the AUX1 protein was localized within Arabidopsis root tissue (Swarup et al., 2001). The AUX1 protein was detected in a subset of stele, columella, lateral root cap and epidermal cells exclusively in root tips. Considering the localized expression of AUX1 only in root tips, it is surprising that aux1 mutant root tips contain lower auxin levels, which rather suggests defects in long-distance supply to the root tip (Swarup et al., 2001). This paradox, taken together with localization of AUX1 at the upper side of protophloem cells (see Plate 1.1A, following page 146), suggests a role of the AUX1 protein in unloading of the phloem flow via the protophloem to the root apical meristem (Swarup et al., 2001). Thus, AUX1 would appear to provide a molecular connection between nonpolar and polar auxin transport routes. AUX1 is a member of the small gene family in Arabidopsis. However, the characterization of the three other LIKE AUX1 (LAX) genes has not yet been reported.
Another Arabidopsis mutant, pin-formed (pin1), with its characteristic needle-like stem had already been functionally associated with auxin efflux on the basis of its dramatic morphological aberrations, which can be phenocopied by inhibition of auxin efflux. In addition, pin1 inflorescences show a drastic reduction in basipetal auxin transport (Okada et al., 1991). The PIN1 gene was cloned by transposon tagging and found to encode a 622 amino acid protein with up to 12 putative transmembrane segments with similarity to a group of transporters from bacteria (Galweiler et al, 1998). Simultaneously, a homologous gene was identified independently by several groups -the PIN2/EIR1/AGR1 gene (Chen etal., 1998; Luschnig etal., 1998; Muller et al., 1998; Utsuno et al, 1998) - and analysis of additional homologs (PIN3, PIN4 and PIN7) followed (Friml etal., 2002a,b, 2003). In total, the Arabidopis PIN gene family consists of eight members and homologous genes were found in other plant species, e.g. maize, rice, soybean and others. The proposed function for PIN proteins as efflux carriers has not ultimately been proven; nonetheless, several lines of evidence strongly support their role in PAT:
1. Topology and localization of PIN proteins: The PIN proteins share more than 70% similarity and have almost identical topology - a large hy-drophilic loop is symmetrically flanked by two conserved, highly hy-drophobic domains with five to six transmembrane segments. Transporters of the major membrane facilitator class display similar topology (Chen et al., 1998; Luschnig et al, 1998; Muller et al, 1998; Utsuno et al, 1998). When localized in planta, most PIN proteins show asymmetric cellular localization (Plate 1.1B-D), impressively correlating with the known direction of PAT in these tissues. This polar localization was predicted by chemiosmotic hypothesis for auxin efflux proteins (Rubery & Sheldrake, 1974; Raven, 1975).
2. Heterologous expression of PIN proteins: To date, the only experimental system used to address PIN transport activity is yeast assay (Luschnig etal., 1998). Yeast carrying a mutation in the GEF1 gene (resulting in an altered ion homeostasis) shows enhanced resistance to the yeast toxin fluoroindole, when overexpressing PIN2/EIR1/AGR1. Fluoroindole shows some (albeit limited) structural similarity to auxin (Luschnig et al., 1998). The yeast also retains less radioactively labeled auxin than does control yeast (Chen et al., 1998). Nonetheless, measurements of auxin efflux instead of auxin retention have not been demonstrated so far.
3. PAT is defected in pin mutants: All defects observed so far in pin mutants occur in processes known to be regulated by PAT and they can be pheno-copied by treatment of wild-type plants with AEIs. One of the strongest arguments for the involvement of PIN proteins in auxin transport is a reduction of PAT in pin mutants, which directly correlates with loss of PIN expression in corresponding tissue, as was demonstrated for basipetal auxin transport in stem of pin1 mutant or in root of pin2 mutant (Okada et al., 1991; Rashotte et al., 2000). In addition, local distribution and accumulation of auxin monitored both by the activity of an auxin responsive construct (e.g. DR5::GUS; Sabatini et al., 1999) and by direct measurements of auxin content (Friml et al., 2002b) or using radioactive IAA preloaded root tips (Chen et al., 1998) have been shown to be affected in pin mutants.
The data accumulated so far provide an extensive body of evidence to argue that PIN proteins are involved in some important aspects of auxin transport, probably in auxin efflux. Nevertheless, the central question of whether PIN proteins represent transport or regulatory component of auxin efflux machinery still remains a topic for future investigations.
Recently, another protein family has been implicated in auxin transport - MDR proteins, a subfamily of the ABC transporters (Noh et al., 2001). Members of this family are known to enhance the export of chemotherapeutic substances in mammalian systems. Two of them, AtMDR1 and AtPGP1, were isolated by NPA-affinity chromatography and were also able to bind NPA in vitro or when expressed in yeast cells. Moreover, the corresponding mutants and double mutants show lower rate of PAT and other phenotypic aberrations somewhat resembling defects in PAT. Nevertheless, Atmdr1 mutants still exhibited up to 60% of the NPA binding found in wild type and NPA is effective in reducing PAT to background levels in the mutant (Noh et al., 2001). So the question whether MDR/PGPs represent the elusive NPB, which regulates PAT, remains open. A mechanism how the MDR/PGP proteins modulate PAT has not been identified so far, but a direct auxin transport function or a positive regulation of a PIN-type auxin efflux carriers has been suggested. Another possibility is that MDR proteins are involved in correct localization of auxin carriers. Indeed, it has been reported that in all cell types of Atmdr1 and Atpgp1 hypocotyls the usual basal localization of PIN1 is replaced with more punctuate pattern (Noh et al., 2003). However, previous reports identified PIN expression associated only with vascular tissue (Friml et al., 2002a), and so the biological meaning of this observation remains unclear.
Was this article helpful?