Differential screens of cDNA libraries in the 1980s led to the identification of the first auxin responsive genes (Theologis et al. 1985; Hagen and Guilfoyle 1985; Key et al. 1986; van der Zaal et al. 1987). Most of these genes are activated within minutes after auxin stimulation in a process independent of de novo synthesis of proteins. Several auxin responsive elements (AuxREs) have been identified in the promoters of these primary auxin response genes (Ul-masov et al. 1997; Guilfoyle et al. 1998a; Guilfoyle et al. 1998b), and Auxin Response Factors (ARFs) were shown to bind to these elements and to activate or to repress transcription (Ulmasov et al. 1999a).
ARFs in general contain three well-defined domains: a DNA binding domain (DBD) that binds AuxREs, a middle region domain, and a third domain that was found to mediate homo- or heterodimerization. Whether an ARF is an activator or repressor depends on the structure of its middle region domain. For example, ARFs with Q-rich middle regions activate transcription, while ARFs with P/S/T-rich middle region repress transcription (Ulmasov et al. 1997, 1999a).
Some of the primary auxin response genes were found to encode small short-lived proteins, named Aux/IAA proteins, that resemble bacterial repres-sors (Abel and Theologis 1996; Abel et al. 1994). Like ARFs, also Aux/IAA proteins contain three distinct domains, of which domain I has been shown to have transcription repression activity (Tiwari et al. 2004), domain II is involved in Aux/IAA protein stability and may be a target for ubiquitination (Ramos et al. 2001), and the third domain that allows Aux/IAA proteins to homo- or heterodimerize with ARFs or other Aux/IAA proteins (Ulmasov et al. 1999b). The interaction between Aux/IAAs and ARFs was shown to result in repression of the ARF-stimulated gene expression (Ulmasov et al. 1997), and the repression of ARF action can only be released upon degradation of the Aux/IAAs (Fig. 1) (Weijers and Jürgens 2004).
Apart from being identified in screens for auxin responsive genes, the Aux/IAA encoding genes have also been identified through gain-of-function mutations that lead to auxin insensitivity. Most Aux/IAA proteins are extremely short lived, and all gain-of-function mutations in the Aux/IAA genes result in a specific amino acid change in domain II that stabilizes the encoded protein, and thus represses auxin responses leading to auxin insensitivity phenotypes (Liscum and Reed 2002).
Unfortunately, loss-of-function mutants in the Aux/IAA genes provide very little information about the functions of the individual genes. All of the mutant plant lines analyzed to date display no or only very subtle phenotypes, suggesting that there is functional redundancy between Aux/IAAs. By contrast, loss-of-function mutations have been very informative for the ARF family of transcription factors. For example, based on mutant phenotypes, ARF3/ETTIN has been associated with gynoecium development (Sessions et al. 1997), ARF5/MONOPTEROS with early embryogenesis and vascular development (Hardtke and Berleth 1998), ARF7/NON PHOTOTROPIC HYPOCOTYL4 with differential growth responses in aerial tissues (Liscum and Reed 2002; Liscum and Briggs 1995) and ARF8/FRUIT WITHOUT FERTILIZATION with fruit initiation (Goetz et al. 2006).
The Arabidopsis genome encodes 29 AUX/IAA proteins and 23 ARFs, which can combine to translate the auxin signal into a gene expression response. For example, it has been shown in yeast two-hybrid assays that specific combinations of ARFs and Aux/IAAs are preferred interaction partners (Kim et al. 1997; Hardtke et al. 2004). The specificity of these interactions seems essential to differentiate auxin responses in different cell types, and is further supported by the observation that both ARFs and Aux/IAAs are at best only partially interchangeable in each others expression domain (Weijers and Jürgens 2004; Weijers et al. 2005a).
Was this article helpful?