Direct ETTIN-auxin interaction controls chromatin state in gynoecium development

Background: Not all auxin responses can be explained by canonical auxin signalling

The plant hormone auxin is a master regulator of plant morphology as it controls cell proliferation, elongation and differentiation at many developmental stages from embryogenesis to senescence. Auxin perception has been thoroughly investigated for decades, which led to the discovery of what is now considered to be the canonical auxin signalling pathway (Fig. 1): In the absence of auxin, AUXIN/INDOLE-3-ACETIC ACID (Aux/IAA) repressor proteins, together with their co-repressor TOPLESS (TPL), inhibit the transcriptional regulatory function of AUXIN RESPONSE FACTORS (ARFs). Auxin promotes interaction of Aux/IAAs with F-box proteins of the TRANSPORT INHIBITOR RESPONSE 1 (TIR1)/AUXIN-BINDING F-BOX (AFB) family. These proteins are part of a ubiquitin ligase complex that ubiquitinates Aux/IAAs and thereby targets them for degradation in the 26S proteasome. This revokes repression of ARFs, which can in turn regulate auxin-responsive genes (Leyser, 2018).

Figure 1: The canonical auxin signalling pathway. At low auxin concentrations, Aux/IAA and TPL repress activity of ARFs. At high auxin concentrations, IAA is bound by a TIR-Aux/IAA co-receptor complex, which triggers ubiquitination and subsequent degradation of the Aux/IAA repressor, thereby enabling ARF activity.

The nuclear TIR/AFB-Aux/IAA pathway however cannot account for all responses triggered by auxin, and several non-canonical signalling pathways have been identified (Kubeš and Napier, 2019). One of these centres around ETTIN (ETT)/ARF3, a repressive ARF that regulates flower development, lateral root formation and leaf polarity in Arabidopsis. ARF3 is an atypical ARF that does not possess an interaction domain for Aux/IAA repressors, and therefore is not regulated by the canonical pathway. Instead, auxin disrupts the interaction between ETT and other transcription factors such as INDEHISCENT (IND) and thereby interferes with ETT’s repressive activity (Simonini et al., 2016). Whether this effect of auxin on ARF3 is direct is however unknown. In their preprint, Kuhn et al. set out to answer this question and to further analyse the consequences of auxin sensing by ETT.

Key findings: Direct auxin binding by ETT triggers de-repression of auxin-regulated genes

A 207 amino acid fragment of the so-called ETT-specific (ES) domain is sufficient to confer auxin sensitivity to ETT function (Simonini et al., 2018). This fragment is intrinsically disordered, hampering classic structural approaches such as X-ray crystallography to observe potential auxin binding. Instead, the authors used heteronuclear single quantum coherence (HSQC) NMR spectroscopy to probe for chemical shifts of amide-NH bonds in a recombinantly expressed ES fragment. Multiple residues changed their position in the presence of the naturally occurring auxin indole-3-acetic acid (IAA), but not in response to the related benzoic acid, showing that this fragment indeed binds IAA. IAA-ETT interaction was further confirmed by isothermal titration calorimetry (ITC).

In addition to its DNA binding domain and the disordered ES domain, ETT contains a conserved EAR motif, which is also found in Aux/IAA proteins, where it mediates interaction with the TPL co-repressor. Yeast-2-hybrid and co-immunoprecipitation experiments revealed that ETT interacts with TPL and the related TOPLESS-RELATED 2 (TPR2) and TPR4, and that this interaction requires the EAR motif. IAA had previously been shown to disrupt interactions between ETT and other transcriptional regulators, and this also proved to be the case for ETT’s interaction with TPL/TPRs.

TPL is known to recruit HISTONE DEACETYLASE 19 (HDA19), which keeps chromatin in a repressed state. The authors found that both tpl tpr2 and hda19 mutants show strong defects in gynoecium development (i.e. aberrant formation of “female” flower organs), similar to ett mutants. Expression of two established ETT target genes, PINOID (PID) and HECATE 1 (HEC1), is elevated in gynoecia not only in ett, but also in tpl tpr2 and hda19 mutants. Using chromatin immunoprecipitation (ChIP), the authors demonstrated that ETT, TPL and HDA19 associate with the same regions of the PID and HEC1 promoters, and H3K27 acetylation at these loci increases upon IAA treatment and in an ett mutant background.

Kuhn et al. suggest a model in which ETT recruits TPL/TPR and HDA19 to its target genes, where HDA19-induced histone deacetylation keeps chromatin in a repressed state. Direct binding of IAA to ETT causes dissociation of TPL/TPR-HDA19, allowing histone acetylation to occur and the respective genes to be expressed (Fig. 2).

Figure 2: Model for ETT-controlled gene expression. At low auxin concentrations, ETT recruits TPL and HDA19 to keep its target genes in a repressed state. Direct auxin binding disrupts interaction between ETT and TPL, de-repressing transcription (reproduced from Kuhn et al., Fig. 5d, under a CC-BY 4.0 license).

Why I chose this preprint

This preprint provides exciting insight into the mechanisms of a previously somewhat obscure auxin signalling pathway. It seems the auxin field is well underway to establish ETT as a non-canonical auxin receptor alongside the canonical TIR/AFB-Aux/IAA pathway.

Open questions/future directions

1. HSQC NMR did not allow positional changes to be assigned to specific ETT residues with the exception of W505 (as it is the only tryptophan residue in the tested ETT fragment). Does mutating W505 affect IAA binding and/or downstream signalling?
2. The authors’ model predicts ETT to recruit TPL/TPR and HDA19 to DNA. Is association of these proteins with the PID and HEC1 promoters reduced in the ett mutant?
3. In which time frame do ETT-mediated changes in gene expression occur and how does their timing compare to TIR/AFB-mediated responses?
4. De-repression is not the same as activation. Can the authors speculate how ETT-regulated genes are induced once repression is revoked in response to auxin and could ETT be involved in this process?

References/further reading

Kubeš, M. and Napier, R. (2019). Non-canonical auxin signalling: fast and curious. J Exp Bot 70, 2609–2614.

Leyser, O. (2018). Auxin Signaling. Plant Physiology 176, 465–479.

Simonini, S., Deb, J., Moubayidin, L., Stephenson, P., Valluru, M., Freire-Rios, A., Sorefan, K., Weijers, D., Friml, J. and Østergaard, L. (2016). A noncanonical auxin-sensing mechanism is required for organ morphogenesis in Arabidopsis. Genes Dev. 30, 2286–2296.

Simonini, S., Mas, P. J., Mas, C. M. V. S., Østergaard, L. and Hart, D. J. (2018). Auxin sensing is a property of an unstructured domain in the Auxin Response Factor ETTIN of Arabidopsis thaliana. Scientific Reports 8, 1–11.

Tags: arabidopsis, auxin, chromatin, histone, hormone, receptor

doi: https://doi.org/10.1242/prelights.15806

Read preprint (No Ratings Yet)