2.1. Reporters Based on Auxin Signalling

Visualisation of auxin in plants, direct or indirect, has attracted a lot of interest in phytohormone research for many years. The first auxin reporters were made of promoters of auxin inducible genes that were fused to a β-glucuronidase (GUS) reporter gene, such as SAUR:GUS transformed into tobacco [37] or soybean GRETCHEN HAGEN 3 (GH3)-derived GH3:GUS used in white clover (Trifolium repens) [38]. Both of the reporters were able to show an asymmetric pattern of the auxin action during gravitropism or phototropism.

Going more into details on the DNA sequence, a 183-bp auxin-responsive region (AuxRR) of the PsIAA4/5 promoter was identified in Pisum sativum containing two auxin-responsive domains (AuxRD) A and B defined by linker scanning mutagenesis [39,40]. AuxRD A possesses a conserved sequence ^T/_GGTCCCAT and has been described as an auxin switch, while AuxRD B was hypothesised to have an enhancer-like activity, with ^C/_AACATGGN^C/_A^A/_GTGT^T/_C^T/_C^C/_A nucleotide sequence [39]. Domains A and B were cloned to control GUS expression in a BA:GUS construct and tested in Arabidopsis for their functionality [41]. In the root elongation zone, the expression of BA:GUS was induced by active auxins such as IAA, NAA or 2,4-dichlorophenoxyacetic acid (2,4-D); and, less by indole-3-butyric acid (IBA). Moreover, other tested compounds, such as inactive auxin analogue, IAA metabolic precursors, IAA transport inhibitors, or phytohormones, were unable to induce GUS expression. In planta, the inducibility of the BA:GUS reporter gene by IAA was increased from 10⁻⁷ M to 10⁻⁴ M, but was inhibited at 10⁻³ M. In addition, BA:GUS expression pattern was confirmed by introducing the second reporter gene, encoding the green fluorescent protein (GFP), under the control of BA sequence. BA:GFP expression displayed a similar pattern to that of BA:GUS, and was inducible by auxin as well [41]. Using chemical genetics in Arabidopsis, BA:GUS reporter has been successfully used as bait for the identification of inhibitors of auxin transcriptional activation [42].

The most popular auxin reporter to indirectly visualise auxin in plants is the artificial auxin-response promoter DR5 [43], whose activity reflects an auxin response maximum [44]. Among several auxin inducible genes, GH3 from a soybean was identified as rapidly and specifically induced by auxins [45]. Transcriptional activation of this gene was observed within 5 min after auxin application [46]. Within the GH3 promoter, the smallest composite natural auxin response element (AuxRE) with strict auxin specificity was identified and named D1-4 element [47]. The D1-4 represents an 11 bp 5′-CCTCGTGTCTC-3′ sequence, and contains a coupling element that overlaps with the TGTCTC motif required for auxin inducibility [48]. The TGTCTC sequence occurs in many promoters of early auxin responsive genes, bound by ARFs and responding rapidly to active auxins only [47] (Figure 1a). Together with the GGTCCCAT sequence that was identified in a pea [39], it is also present as a TGTCTCtcatttGGTCCCAT sequence in SAUR promoters [49].

Indirect auxin reporters. (a) DR5 reporters were derived from auxin response element (ARE) sequence for binding of ARF transcription factors in auxin responsive promoters. (b) The expression of DR5rev:GFP, DR5rev:3xVenus-N7 and DR5rev:erRFP reflects similar auxin signalling output in Arabidopsis root tip. Degradation based reporters DII and R2D2 contain degron domain from Aux/IAA repressors leading to ubiquitination and degradation in the presence of auxin. They represent auxin signalling input. 35S, CaMV35S minimal promoter; ARE, auxin response element; ARF, AUXIN RESPONSE FACTOR; Aux/IAA, AUXIN/INDOLE-3-ACETIC ACID; GFP, green fluorescent protein; RFP, red fluorescent protein; Venus, yellow fluorescent protein; and, Ω, tobacco mosaic virus leader sequence.

Thymidine substitutions in the natural D1-4 AuxRE (CCTCGTGTCTC) provided the synthetic DR5 AuxRE 5′-CCTttTGTCTC-3′, with an exceptionally strong auxin response when cloned upstream of a minimal −46 cauliflower mosaic virus (CaMV) 35S promoter [43]. Eight repeats of the synthetic DR5 (8x) AuxRE displayed up to 10-fold higher inducibility by NAA when compared with the eight repeats of natural D1-4 (8x) AuxRE. In addition, the spacing between TGTCTC elements and nucleotide composition upstream of TGTCTC elements was suggested to be important for the auxin inducibility in the DR5 construct [43]. Several variants of DR5 element were prepared to monitor auxin signalling action in plants (Figure 1). Seven tandem repeats of the 11 bp sequence 5′-CCTTTTGTCTC-3′ fused to a −46 bp CaMV35S minimal promoter and driving the GUS gene gave a rise to the DR5:GUS reporter [50]. Nine inverted repeats of the 11 bp element, a CaMV35S minimal promoter and a TMV leader sequence were used to create a DR5rev version of the auxin responsive promoter. Different reporter genes were combined with DR5rev promoter, such as phosphonate monoester hydrolase PEH A gene in DR5rev:PEHA [51], an endoplasmic reticulum-targeted green fluorescent protein in DR5rev:GFP [44] (Figure 1b), three tandem copies of Venus, a fast maturating variant of the yellow fluorescent protein, fused to a nuclear localization signal (NLS) in DR5rev:3xVenus-N7 [52] (Figure 1b), a red fluorescent protein (RFP) targeted to the endoplasmic reticulum in DR5rev:mRFPer [53] and DR5rev:erRFP [54] (Figure 1b), or a luciferase coding region in DR5:Luciferase [55]. Overall, transgenic Arabidopsis plants that were carrying these reporters displayed a similar pattern, with visible staining in root quiescent centre (QC), columella cells, protoxylem, the most distal domain of developing shoot primordia with an incipient leaf vein and in root primordia tips. It has been shown that the activity of DR5 correlates with auxin accumulation detected by immunolocalisation in Arabidopsis [56].

To create a more sensitive auxin responsive promoter, two bases in the original DR5 binding sequence TGTCTC were exchanged to make a TGTCGG with higher binding affinity to ARF, as identified by protein binding microarrays [57]. Interestingly, the TGTCGG sequence occurs also in a promoter of Agrobacterium tumefaciens T-DNA of Ach5 Ti plasmid [58]. Nine original AuxREs in the DR5rev promoter were replaced with new binding site elements producing a DR5v2 promoter [59] (Figure 1a). The expression pattern of DR5v2 matches more precisely the auxin accumulation sites, as predicted from the localisation of the polar auxin transporters [60]. Moreover, DR5v2 showed a weak activity in the dividing cells of the embryo, leaf, or shoot meristem corresponding to an auxin function in cell division processes [61]. When comparing the activity of DR5 and DR5v2 in a DR5v2:ntdTomato-DR5:n3EGFP double reporter [59], all of the expression sites of DR5 were overlapped by a DR5v2 expression and the additional DR5v2 signal appeared in other cell types (cotyledons and vasculature during embryogenesis, in metaxylem, pericycle, lateral root cap, epidermal cells of root, and in the cells surrounding the shoot primordia and the L1 layer of the shoot apical meristem). The difference in DR5 and DR5v2 sensitivity and localisation can be useful for the identification of unique regulatory factors, preferring specific AuxRE binding sequences in both promoters.

In addition to DR5, another type of auxin responsive promoter was constructed to monitor auxin signalling input [62]. The auxin interacting domain II (DII) [63] of IAA28 protein was cloned under a constitutive promoter and was fused to Venus with a NLS sequence [64] to generate the DII-Venus auxin sensor (Figure 1). The DII domain is the Aux/IAA domain that is ubiquitinated and induces degradation of the protein in response to the auxin dose-dependent presence. Therefore, DII-Venus monitors the input into the auxin signalling pathway by the degradation of fusion protein, thus switching off the signal in the presence of auxin, in an opposite manner to DR5 principle. Two promoter variants were used for the sensor: a CaMV35S promoter [64] or a RPS5A promoter [59]. The need of “auxin input” quantification led to the development of an innovated reporter. The combination of DII-Venus and mDII-ntdTomato, a mutated auxin insensitive variant of DII, into one construct gave a rise to the ratiometric version of the auxin input—R2D2 [59] (Figure 1a). Two fluorophores allow for a semiquantitative measurement of auxin accumulation as a ratio of yellow and red signal. Auxin sensitive DII and R2D2 reporters enable the observation of fast changes in auxin accumulation at cellular resolution in real-time [59,64,65,66,67]. Based on DII degradation, another quantitative ratiometric sensor for analysis of auxin dynamics in real-time was developed and optimised for the use in single cell systems combining a luminescent reporter with an internal normalization element [68].

Interestingly, DII and R2D2 reporters showed partial auxin insensitivity in the root tip, particularly in the epidermis, cortex, and endodermis cell files that are close to the QC [59,64,65]. After the gravistimulation or exogenous auxin application, the DII-Venus signal of both the reporters is not switched off completely in these cells, suggesting a distinct type of regulation when compared to cells without signal. Moreover, the comparison of DR5 and DII signals revealed discrepancies between the auxin signalling response input and output, suggesting the presence of the auxin, but the absence of a signalling response in particular parts of the growing plant [59]. It would be useful to combine DR5v2 and R2D2 in a single three-colour reporter to inspect the auxin input and output in one plant.

To follow the specificity of the auxin signalling, a set of Aux/IAA and ARF reporters were fused with GUS or GFP tag to report signalling pathways with particular sets of Aux/IAA and ARF proteins. An ARF collection using transcriptional fusion with nuclear localised 3xGFP mapped their different, as well as overlapping expression pattern in embryo and in the root tip [69]. Analogically, members of Aux/IAA family possess a wide range of localization patterns in Arabidopsis, suggesting their spatiotemporal specificity [70,71,72,73,74,75,76,77,78]. When combining the members of Aux/IAA and/or ARF families provides a huge set of possible mutual interactions pointing to variability and complexity of the auxin signalling in plant development [62,79] and waiting to be revealed.