Project Topics
Get Complete Project Material File(s) Now! »
Back to the water looking for ARF ancestors
Land plants evolution from algae happened hand in hand with the adaptation to terrestrial stresses and the development of a certain body complexity. Given the importance of auxin hormone in plant growth and architecture, the acquisition of the auxin signalling pathway must have been a breaking point in the transition from water to terrestrial life. However, not much is known about how and when this happened. ARF proteins have been traced back till bryophyte organisms but till now no sign of them has been found in charophyte algae, the proposed ancestors of land plants.
Swimming among the algae, we found by sequences homology and protein structure modelling two possible candidates of ARF ancestors in the charophyte algae Klebsormidium nitens and Chlorokybus atmophyticus. Both proteins presented a predicted DBD and a PB1 domain with high similarities to those found in plant ARF proteins. Furthermore, as it is the case for some ARF repressors in between both predicted domains we found potential motifs that might recruit TPL. Along the third and final chapter of this manuscript I will describe the structural and biochemical characterization of these different domains in terms of their DNA binding specificity, oligomerization potential and TPL interactions.
TPL in the specific context of auxin signalling
TPL interactions with several proteins of the Aux/IAA family have been demonstrated by several studies (Causier et al., 2012a, 2012b; Kagale et al., 2010b; Szemenyei et al., 2008). These interactions are known to happen through an EAR-motif type LxLxL present in the Domain I (DI) of Aux/IAAs. Except for Aux/IAA33, all A. thaliana Aux/IAA proteins have this motif. Aux/IAA18, 26 and 28 present a small variation of the EAR peptide (LxLxPP) also found in the Aux/IAAs and some ARF proteins from the moss Physcomitrella patens. Mutations of LxLxPP motif in these proteins abolished the interaction with TPL (Causier et al., 2012a).
Apart from Aux/IAAs proteins, some ARFs have also been found to interact with TPL (Causier et al., 2012a, 2012b). The TPL interactome analysis in A. thaliana revealed that ARF2, ARF9 and ARF18 proteins -all classed as repressors- can interact with TPL and they present a (K/M/R)LFGV EAR-like motif. Apart from this peptide, a LxLxL-type EAR motif was also described in ARF2 (Causier et al., 2012b). Whether ARFs interaction with TPL is necessary for auxin transcriptional repression or not remains an enigma (Chandler, 2016).
Whilst several studies have been done about TPL and its interactions with diverse transcriptional repressors that prove TPL involvement and importance in a variety of plant processes, not much is known about how TPL mediates chromatin compaction.
HDA6 and 19 have been frequently associated with TPL (Liu et al., 2014). Although a direct interaction between TPL and HDA has never been found, their importance in TPL-mediated repression has been demonstrated. hda mutants have similar phenotypes to tpl-1 or other TPL/TPR mutants (Long, 2006; Pi et al., 2015; Zhu et al., 2010) and ChIP-seq results prove that HDA and TPL co-localize in the promoters of the genes they are repressing (Krogan et al., 2012; Pi et al., 2015; Ryu et al., 2014; Wu et al., 2015). Understanding how HDA recruitment by TPL is made and the intermediate/s carrying it out remains to be answered.
Within auxin signalling, the hormone is thought to trigger a quick chromatin switch that allows the expression of auxin responsive genes. As previously mentioned (Fig 11), under low auxin concentrations auxin transcriptional response is inhibited by Aux/IAA proteins that block ARF activators and that recruit TPL. In agreement with this, TPL and HDA19 have been found to colocalize with the ARF activator, ARF5 (MONOPTEROS, MP), in the promoters of auxin-responsive genes. Auxin treatment led to Aux/IAA degradation and consequent disappearance of the co-repressor complex in these regions of the genome. The removal of the co-repressor complex allows the recruitment of chromatin remodellers by ARF5, that help opening chromatin and lead to auxin-responsive genes transcription. Due to the fast degradation of Aux/IAAs under auxin presence, this closed-to-opened switch in chromatin is thought to happen rapidly allowing a quick transcriptional response (Wu et al., 2015).
Recently, a similar “quick switch” mechanism has been described involving this time the Mediator complex. The Mediator complex has 4 modules (head, middle, tail and dissociable CDK8 Kinase module, CKM) with several components each. It affects transcription by direct interactions with the RNApol II that lead to activation or repression of transcription depending on the components of the complex that are intervening. ARF7/19 activators interact with MED25 and 8 (tail and head modules, respectively) whereas Aux/IAA14 was found to be associated with MED13 (CKM module) through TPL, that directly interacts with it. Auxin treatment led to the dissociation of the IAA14-MED13-TPL from auxin-responsive genes whereas ARF7/19-MED25 complex remained stable. CKM module is thought to function as a blocker of the interaction with RNApol II. Thus, in the absence of the hormone the Aux/IAA-TPL-CKM Mediator module could prevent the access of the transcriptional machinery, whereas auxin presence would induce Aux/IAA degradation and the consequent removal of the whole co-repressor complex in a fast way. RNApol II could now interact with the tail and head components of the Mediator complex associated to ARF activators inducing the transcription of auxin-responsive genes (Ito et al., 2016).
Native molecular mass determination
Molecular masses were determined by Size-Exclusion Chromatography-Multi Angle Light Scattering (SEC-MALLS) on an analytical Superdex-S200 increase (GE Healthcare) connected to an in-line MALLS spectrometer (DAWN HELEOS II, Wyatt Instruments). Analytical size exclusion chromatography was performed at 25ºC at a rate of 0.5 mL/min in buffer A for AtTPL184 and AtTPL202 (wt and mutants) and in buffer F (CAPS 20 mM pH 10.5, Tris-HCl 100 mM pH 8.8, NaCl 50 mM, TCEP 0.1 mM) for AtTPL202-EAR motifs complexes (fluorescence anisotropy conditions). The refractive index was measured with in-line refractive index detector (Optirex, Wyatt Instruments) was used to follow the differential refractive index relative to the solvent. Molecular masses calculation was done with the Debye model using ASTRA version 5.3.4.20 (Wyatt Instruments) and a theoretical dn/dc value of 0.185 mL/g.
Yeast two hybrid interaction tests
The vectors pGBKT7 and pACT2 were respectively transformed into yeast strains Y187 and AH109 (Clontech) using standard protocol (18). The analyses were performed after mating of the two yeast strains. Interactions were assessed using β-Gal activity (19). To do so, OD405nm were measured during exponential growth phase after incubation at 30°C in Buffer G (100 mM Phosphate buffer pH 7, 10 mM KCl, 1 mM Mg2SO4, β-mercaptoethanol 50 mM) with ONPG.
Homogeneous-Time Resolved Fluorescence (HTRF) interaction tests
His-tagged AtTPL202 wt/mutants and MBP-tagged IAA12 interactions were analysed by HTRF (20) using CisBio Bioassays Anti-His acceptor d2 and Anti-MBP donor Tb antibodies.
HTRF experiments were performed on Greiner 384 Flat bottom wells plate. Three simultaneous replicas were done for each binding mixture. After a 2h-incubation at room temperature in the dark, the binding reactions were excited at 337 nm and emission measurements were taken at 620 nm and 665 nm with a Tecan infinite M1000PRO. HTRF specific signal was calculated as follows: For HTRF competition assays, an initial GST-IAA12 (200 nM) – MBP-AtTPL202 (500 nM) complex was formed using CisBio Bioassays Anti-GST donor Tb and Anti-MBP acceptor d2 antibodies. The complex formed was competed by adding increasing amounts of His-tagged AtTPL202 wt and mutant proteins. Three independent replicas were done for each binding mixture and the measurements were done in the same conditions as before.
DNA-TFs interplay in the control of gene expression
The expression or repression of a certain set of genes is a game with two main players: the DNA regulatory sequences located inside the promoters of the regulated genes, and the transcription factors that can bind to them. A game that is played according to three main rules: DNA accessibility, DNA-TF specificity and TF activator or repressor identity.
In eukaryotes, chromatin state will be determinant for the access of a TF to its DNA binding site. Therefore, transcriptional regulation (activation or repression) is accompanied by chromatin remodelling that will render chromatin opened or closed. In this sense, transcription factors can act by recruiting co-regulator proteins that, as already reviewed in the previous chapter, can alter the state of chromatin. A very special group of TFs, pioneer TFs, can do both jobs: pioneers have not only the capacity to bind DNA, but also to make it more accessible by nucleosomes shifting (Zaret Kenneth S.; Mango, 2017).
Once allowed by chromatin state, the expression of a certain gene will be now controlled by the TFs bound to its promoter. TFs-DNA interaction is mediated by one or more DNA binding domains (DBDs). Additional regions normally accompany DBDs: domains that mediate homo or heterotypic interactions with other transcription factors (dimerization or oligomerization domains) or regions that can recruit co-regulator proteins (Smith and Matthews, 2016). These other domains, although they lack DNA-binding capacity, they can have some influence upon the interaction with the DNA (Sayou et al., 2016).
TF-DNA interactions take place between amino acids within the DBD and nucleotides (Crocker et al., 2016) in both a non-specific and specific manner (Smith and Matthews, 2016). Non-specific interactions occur with the DNA-backbone and provide stability rather than specificity. Specificity is driven by the so-called “base-readout”, base-amino acid interactions mainly with charged and polar amino-acidic residues (Luscombe et al., 2001). TFs can also recognize DNA structural features, such as 3D DNA conformation or flexibility of the DNA, that determine what is known as the “shape readout”.
According to the “base-readout” specificity, for a TF there will be one optimal Binding Site (BS) versus a diversity of less favourable nucleotides combinations that can also be bound by the same TF but with lower affinity. Genome-wide analysis indicates that whereas the most favourable DNA binding sequences are rare inside the genome, clusters or continuum of less favourable BSs are rather the rule (Crocker et al., 2016).
Frequently, eukaryotic TFs act in a cooperative manner to efficiently bind to these lower-affinity sites. They do so by binding as dimers or oligomers that allow simultaneous binding to multiple sites, increasing the length of the binding site and thus the specificity of the binding (Berg and von Hippel, 1987; Marianayagam et al., 2004; Morgunova and Taipale, 2017). The formation of these complexes can also take place in a heterotypic way with TFs involved in other processes. This, apart from conferring specificity to the binding, provides a hub for the integration of different signals and adds a possible extra level of regulation by the formation of complexes which activity can be easily modulated (Marianayagam et al., 2004; Smith and Matthews, 2016).
Under the stimulus “a” inside a cell, the combination of TFs present in it at a specific moment, their specificity for BSs and the accessibility of these BSs within chromatin will, altogether, determine the transcriptional response to “a”. What if “a” stands for “auxin”?
ARFs, transcription factors of the auxin response
Auxin signal converges towards a family of transcription factors, the ARFs, of which several copies are present in plants with different spatio-temporal expression patterns. The co-expression of different ARFs inside a cell is thought to contribute to the controlled, time and tissue-specific auxin transcriptional response.
As all transcription factors, ARFs present a DBD that mediates their interaction with DNA sequences named Auxin Response Elements, AuxREs, located in the promoters of auxin responsive genes. The DBD is localized in the N-ter of ARF proteins. In the C-terminal region, ARFs present a PB1 domain, also shared by Aux/IAA repressor proteins, that mediates homo and hetero-oligomerization (Chandler, 2016) (Fig7). Both ARFs DBD and PB1 domains can act autonomously: the DBD interacts with AuxREs without the PB1 being necessary for it, and isolated PB1s can mediate oligomerization (Korasick et al., 2014; Nanao et al., 2014; Tiwari et al., 2003; Ulmasov et al., 1997a). For these reasons, ARFs present what is called a “modular domains” structure, with the DBD and PB1 modules that must be fundamental for their function, since they have been extensively conserved along plants evolution (Finet et al., 2013).
Right at the beginning of land plants evolution, we can find the liverwort Marchantia polymorpha, a plant that is gaining more a more importance inside the plant physiology field due to the simplicity of
its genome. Indeed, M. polymorpha presents only 3 copies of ARF genes (Flores-Sandoval et al., 2015) versus the 23 that have been found in the flowering plant A. thaliana. This increase in the number of ARF genes is due to a series of gene duplications that on one hand, generated a partial functional redundancy within the ARF family (Okushima et al., 2005) (Fig 20).
ARFs DNA binding sites and mechanism
Auxin Response Elements, AuxREs, were first identified in the promoters of GH3 soybean gene, an early-auxin responsive gene implicated in IAA metabolism (Fig 5). By scraping this promoter into smaller sequences, Liu et al. in 1994 isolated the promoter regions that were auxin-inducible, in which the TGTCTC motif was first found as the consensus sequence of the AuxREs (Liu et al., 1994). 3 years later, an artificial AuxRE constructed with 4 tandem copies of TGTCTC tandem repeats (P3 4X) allowed, in a one-hybrid assay, the isolation of the first ARF transcription factor, AtARF1 (Ulmasov et al., 1997a).
Since, numerous researches have been done for trying to unravel the specificity of different ARFs for different AuxREs and how this specificity can lead to activation or repression of auxin responsive genes. Moreover, these findings have permitted the development of several auxin sensors that allow the quantification of the auxin transcriptional response (Fig 22) (Liao et al., 2015; Ottenschläger et al., 2002).
Table of contents :
General introduction to auxin, the hormone growth
Hormones in plants! Really? Why?
A little bit of history…
Hormones inside a plant cell
Auxins: the hormone that influences almost everything
Auxin synthesis
Auxin inactivation
Auxin transport
Auxin signalling
Objectives
Chapter I Introduction
Gene expression control in eukaryotic organisms
Co-repressors
Co-repression in plants
TOPLESS co-repressor mechanism
TPL in the specific context of auxin signalling
Article 1 Supplementary Information
SI Figures
SI Tables
SI Materials and Methods
SI References
Complementary results and discussion
Conclusions
Chapter II Introduction
DNA-TFs interplay in the control of gene expression
ARFs, transcription factors of the auxin response
ARFs DNA binding sites and mechanism
Results
Discussion
Conclusions
Chapter III Introduction
Charophyte organisms: from water to land
B3 family
Auxin signalling: from bryophytes to flowers
Auxin signalling clues in charophytes?
Results
Discussion
General discussion and conclusions
Materials & Methods
Bibliography
