Dissecting the Biological Functions of ARF and Aux/IAA Genes

The plant hormone auxin affects almost every aspect of plant development, from embryogenesis to flower development and seed set, and is also an important regulator of the interaction between the plant and its environment. An intriguing question is how the plant translates the presence of auxin into many different developmental outputs according to the internal and external contexts. One possible way of attaining this specificity is by using unique signaling components for each context. Auxin signaling involves the regulation of gene expression by Auxin Response Factors (ARFs) and their inhibition by Aux/IAA proteins. Both of these components are encoded by gene families with multiple members in most flowering plants. Elucidating the importance of this complexity and its contribution to the specificity of auxin output is central to the understanding of the regulation of plant development by auxin. However, the ability of ARF and Aux/IAA proteins to compensate for each other challenges the identification of the biological roles for the different family members.

In their pioneering studies, Overvoorde et al. (2005) and Okushima et al. (2005) took up the challenge by taking a systematic approach to identify and characterize single and multiple loss-of-function mutants in Arabidopsis (Arabidopsis thaliana) ARF and Aux/IAA genes, utilizing emerging reverse genetic tools. The thorough analysis demonstrated the degree of redundancy in these gene families and identified new biological functions for redundant pairs of ARF genes. The analysis of the arf7 arf19 double mutant laid the foundation for understanding how auxin promotes lateral root development.

The most informative evidence for a function of a gene is the phenotypic effect of a loss-of-function mutation in the gene. Okushima, Overvoorde, and their collaborators assisted in generating one of the first populations of insertion mutants in Arabidopsis and utilized several such populations to identify as many loss-of-function mutations in ARF and Aux/IAA genes as possible at the time. They characterized single, double, and in one case a triple mutant. They then went on to assess the effect of selected mutant lines on global auxin-regulated gene expression.

Okushima et al. (2005) identified 27 insertion mutants in 18 of the 23 Arabidopsis ARF genes. Most single mutants did not have a gross phenotypic effect, and the 4 that did had been characterized before. This showed both the strength of forward genetics approaches and their limitation in identifying the function of redundant genes. They then went on to generate double mutants in pairs of closely related genes. This led to the identification of several novel roles, showing the redundant activity of these gene pairs and the importance of generating targeted mutations in multiple members of gene families.

Focusing on the arf7 and arf19 single and arf7 arf19 double mutants, the authors identified unique as well as redundant functions for the closely related genes ARF7 and ARF19. arf19 mutants had a nearly wild-type phenotype, while the arf7 mutant showed mild phenotypic effects. In the double mutant, a subset of the arf7 abnormalities became much more severe, and new redundant roles of both genes were exposed. For example, phototropism was affected similarly in the single arf7 mutant and the arf7 arf19 double mutant, suggesting that it is regulated mainly by ARF7. Root gravitropism was normal in both single mutants but severely impaired in the double mutant, indicating that it is redundantly regulated by both genes. Hypocotyl gravitropism and lateral root formation were affected slightly in arf7 single mutants and severely in the double mutant. Transcriptome analysis of auxin-regulated gene expression revealed a similar picture with respect to the molecular effects of the mutants: some genes were affected in one single mutant and the double mutant, and others only in the double mutant. The authors concluded that this gene pair mediates some auxin-regulated processes redundantly or partially redundantly, while other processes are mediated by ARF7 only. Interestingly, ARF19 expression was induced by auxin, and this induction was dependent on the existence of intact ARF7, suggesting that some of the phenotypic effects of the single arf7 mutant may actually result from reduced functions of both genes. These results started to unravel how the existence of multiple ARF genes can be utilized to tune the developmental output of auxin.

The transcriptome analysis pointed to the possible function of a few transcription factors from the LBD family as targets of ARF7 and ARF19 in lateral root formation. In follow-up research, a group including some of the same authors identified LBD16 and LBD19 as prominent regulators of lateral root formation downstream of auxin and as direct targets of ARF7 and ARF19 (Okushima et al., 2007). This set the foundations for the elucidation of the pathway by which auxin promotes the formation of lateral roots. Interestingly, ARF7 and ARF19 pair with other ARFs in other processes (Hardtke et al., 2004).

Overvoorde et al. (2005) identified mutations in 12 of the 29 Arabidopsis Aux/IAA genes. They then generated several double mutant and one triple mutant combination. Strikingly, none of the mutant combinations had an obvious phenotypic difference when compared with the wild type. Furthermore, auxin-regulated gene expression was nearly normal in the triple mutant. This indicated the existence of a high degree of compensation among Aux/IAA genes. In contrast to loss-of-function mutations, dominant mutations that render Aux/IAAs resistant to auxin-mediated degradation show severe phenotypic aberrations. The authors analyzed the effect of one of these dominant mutants, axr3/arf17 (Leyser et al., 1996), on auxin-regulated gene expression, and this time they found a dramatic effect. Interestingly, almost 15 years later, only mild phenotypes were reported for only a few loss-of-function aux/iaa mutants in Arabidopsis (Tian and Reed, 1999). While gain-of-function mutations provided important hints on Aux/IAA functions, their phenotypes may represent artificial effects, and gain-of function mutations in different Aux/IAAs show similar phenotypes. Therefore, the scarcity of reported loss-of-function phenotypic effects still hinders the understanding of the authentic biological functions of Aux/IAAs and ARF proteins. One possible approach is to investigate additional species outside of Arabidopsis. In tomato (Solanum lycopersicum), a single loss-of-function mutation in IAA9/ENTIRE was found to have a prominent effect on leaf patterning and fruit set, first identified by an RNAi approach (Wang et al., 2005). The basal lineages Marchantia and Physcomitrella have one and three Aux/IAAs, respectively, and their loss leads to dramatic auxin-related phenotypes (Flores-Sandoval et al., 2015; Kato et al., 2015; Lavy et al., 2016). New reverse genetics technologies such as synthetic microRNA and CRISPR can be used to target gene families and can be applied to a broad range of plant species. Therefore, now the time may be ripe to get back to this important and challenging question to gain new insights on the specific roles of Aux/IAA and ARF genes.

In summary, Overvoorde et al. (2005) and Okushima et al. (2005) shed light on the importance of the complexity in auxin response and exemplified how reverse genetics approaches can help in dissecting the contribution of this complexity to the understanding of how auxin regulates plant development. They identified important partially overlapping roles for a pair of ARF genes, paving the way to understanding the mechanisms by which auxin exerts its many roles. These remain important questions in development and auxin biology.