Hypersensitivity of Arabidopsis TAXIMIN1 overexpression lines to light stress is correlated with decreased sinapoyl malate abundance and countered by the antibiotic cefotaxime

ABSTRACT

Peptide signaling in plants is involved in regulating development,^1,2 ensuring cross pollination through initiation of self-incompatibility⁴ and assisting with recognition of beneficial (nitrogen fixing bacteria⁵) or unfavorable organisms (pathogens⁶ or herbivores⁷). Peptides function to help plants to respond to a changing environment and improve their chances of survival. Constitutive expression of the gene encoding a novel cysteine rich peptide TAXIMIN1 (TAX1) resulted in fusion of lateral organs and in abnormal fruit morphology. TAX1 signaling functions independently from transcription factors known to play a role in this process such as LATERAL ORGAN FUSION1 (LOF1). Here, we report that the TAX1 promoter is not induced by the LOF1 transcription factor and that the TAX1 peptide neither interferes with transcriptional activation by LOF1.1 or transcriptional repression by LOF1.2. Furthermore, we found that TAX1 overexpressing lines were hypersensitive to continuous light, which may be reflected by a decreased accumulation of the UV-B protecting compound sinapoyl-malate. Finally, adding the antibiotic cefotaxime to the medium surprisingly countered the light hypersensitivity phenotype of TAX1 overexpressing seedlings.

Abbreviations

BY-2

bright yellow-2

CaMV

Cauliflower Mosaic Virus

Col-0

Columbia-0

ctf

constricted fruit

CUC

CUP SHAPED COTYLEDON

DBD

DNA Binding Domain

fLUC

firefly Luciferase

Ler

Landsberg erecta

LOF

LATERAL ORGAN FUSION

OE

overexpression

PCA

Principal Component Analysis

PLS-DA

Partial least squares Discriminant Analysis

rLUC

Renilla Luciferase

TEA

Transient Expression Assay

TF

transcription factor

UAS

Upstream Activator Sequence

LC-ESI-IT-MS

Liquid Chromatography Electrospray Ionization Ion Trap Mass Spectrometry

Introduction

Development of organs is tightly regulated by a network of transcription factors (TFs), hormones and sRNA molecules. Coordination of these pathways is necessary to ensure optimal growth and development and to allow for adequate responses to a changing environment. Recently, we identified a novel signaling peptide, encoded by TAXIMIN in Taxus baccata⁸ and TAX1 (At2g31090) in Arabidopsis thaliana.³ This peptide has an N-terminal signal peptide of 29 amino acids which directs it to the plasma membrane and a 46 amino acid C-terminal peptide with six conserved cysteines and three conserved prolines. Constitutive expression of TAX1 in Arabidopsis resulted in fusion of lateral organs, indicating that this peptide plays a role in organ boundary formation.³ Our results suggested that TAX1 functions independently of known boundary regulating TFs such as LATERAL ORGAN FUSION (LOF) and CUP SHAPED COTYLEDONS 3 (CUC3). LOF1 (At1g26780) is a MYB transcription factor⁹ with two splice variants (Fig. 1A). Mutants for LOF1 and LOF2 (At1g69560) displayed fusion of lateral organs⁹ similar to TAX1-OE lines. A LOF1-OE line generated through insertion of Cauliflower Mosaic Virus (CaMV) 35S enhancers in the promoter region of LOF1 in the Landsberg erecta background resulted in more compact plants with an altered fruit phenotype.¹⁰ The fruits of this line were narrower than wild-type Ler plants and therefore this line was called constricted fruit (ctf).¹⁰

Figure 1.

TAX1 and LOF1 function independently of each other. (A) Splice variants of LOF1 (LOF1.1 and LOF1.2). Bars represent the exons and lines represent the introns for the LATERAL ORGAN FUSION (At1g26780) transcription factor. (B) TAX1 does not interfere with the regulatory activity of the two splice variants of the MYB transcription factor LOF1. Transactivation assay in tobacco protoplasts co-transfected with a pUAS::fLUC reporter construct, effector constructs overexpressing LOF1.1 or LOF1.2 fused to GAL4DBD, TAX1 or GUS (as a control) and an rLUC construct for normalization.⁹ Values are fold-changes relative to protoplasts transfected only with a GUS expression construct instead of LOF1 effector constructs and are the mean (±SE) of eight biological repeats. Significant differences (Student's t-test): *, P < 0.05. (C) Promoter region of TAX1 989 bp upstream of the start codon for TAX1 (adapted from Athena promoter website¹²). (D) LOF1 variants do not transactivate pTAX1. Transactivation assay in tobacco protoplasts co-transfected with a pTAX1::fLUC reporter construct, effector constructs overexpressing LOF1.1 or LOF1.2 and an rLUC construct for normalization¹¹. Values are fold-changes relative to protoplasts transfected only with a GUS expression construct instead of LOF1 effector constructs and are the mean (±SE) of eight biological repeats. Significant differences (Student's t-test): *, P < 0.05.

TAX1 expression is not induced by LOF1

To confirm that TAX1 functions independently of the LOF1 transcription factor we performed transient expression assays (TEAs) in tobacco protoplasts.¹¹ First, a fusion of each of the two LOF1 splice variants to a GAL4 DNA Binding Domain (DBD) was co-transfected in tobacco Bright Yellow 2 (BY-2) protoplasts with the firefly Luciferase (fLUC) gene fused to an Upstream Activator Sequence (UAS). The two splice variants had opposite transcription regulatory activity with LOF1.1 and LOF1.2, increasing and reducing reporter gene expression, respectively (Fig. 1B). To determine if TAX1 can interfere with LOF1 regulatory activity, we co-transfected tobacco BY-2 protoplasts and included a construct which overexpresses TAX1. However, TAX1 did not alter the regulatory activity of either variant of LOF1 (Fig. 1B).

Analysis of the TAX1 promoter using the Athena promoter platform¹² indicated that it contains several MYB transcription factor binding sites (Fig. 1C). To determine if the MYB transcription factor LOF1 can regulate TAX1 expression, we repeated the TEA and co-transfected BY-2 protoplasts with a construct in which fLUC is fused to the promoter of TAX1 (pTAX1::fLUC) and a construct which constitutively expresses the LOF1 splice variants. The LOF1 splice variants did not affect reporter gene expression considerably (Fig. 1D), suggesting that they do not regulate TAX1 expression. These results are in agreement with our previous findings that constitutive expression of TAX1 did not result in changes in LOF1 expression in nodes or seedlings of A. thaliana and no change in LOF1 expression was observed in tax1tax2 mutants either. TAX1 expression in nodes of the lof1lof2 mutants was also similar to the wild type.³

Cefotaxime reduces light stress response of TAX1-OE lines

The three TAX1-OE lines described previously were cultivated on MS¹⁴ medium in a growth room with continuous light or a 16-h/8-h light/dark photoperiod. Notably, TAX1-OE plants grown in continuous light were smaller than wild-type (Col-0) seedlings and turned yellow (Fig. 2A), whereas when grown in a 16-h/8-h light/dark regime they looked similar to wild-type plants (Fig. 2A). Surprisingly, cultivation of TAX1-OE seedlings on medium containing the antibiotic cefotaxime fully restored the light-sensitivity of the TAX1-OE lines (Fig. 2A). Cefotaxime is an antibiotic that is regularly used to remove Agrobacterium growth after transformation and that is mainly selected for its limited toxicity for plant cells.¹⁵ Cefotaxime is a cephalosporin that belongs to the β-lactam group and functions to bind to penicillin binding proteins and interferes with peptidoglycan synthesis in bacteria.¹⁵ It had been reported previously that, depending on its concentration, it may influence development of the plant. For example, addition of low concentrations of cefotaxime stimulated the growth, regeneration and organogenesis of wheat callus in culture,¹⁵ whereas they inhibited regeneration from Arabidopsis root explants after Agrobacterium transformation.¹⁶

Figure 2.

(A) Cefotaxime alters the light stress response of TAX1-OE Arabidopsis thaliana lines. Arabidopsis plants cultivated in continuous (24-h; panel Ai and Aiii) or long day (16-h light/ 8-h dark; panel Aii and Aiv) light conditions on Basal MS (Ai and Aii) or MS supplemented with cefotaxime (Aiii and Aiv) for 21 days are shown. Each plate contains wild-type Columbia (Col-0) plants (left) and TAX-1 overexpression (TAX1-OE) lines on the right. Figure panels in Ai are reproduced from Colling et al.(2015).³ (B-D) LC-ESI-IT-MS analysis of Arabidopsis wild-type (Col-0) and TAX1-OE (TAX) lines grown on Basal MS medium (MS) or MS medium supplemented with cefotaxime (CEF) in continuous light (24H) or a 16-h/8-h light/dark regime (16H). (B) PCA projecting the first (t[Comp. 2]) and second (t[Comp. 2]) principal components separates the plants growing in continuous light (red encircled) from those growing in a 16-h/8-h light/dark regime (blue encircled). (C) S-plot for correlation (p(corr)[Comp. 1]) and covariance (w*c[Comp. 1]) for a PLS-DA that separates the samples of plants growing in continuous light from those growing in a 16-h/8-h light/dark regime. (D) Average total ion current (TIC) of the peak corresponding to sinapoyl malate. The error bars designate SE of the mean (n = 3).

Although no experimental mechanistic evidence for their reported growth effect has been described, it is possible that observed effects involve plant esterases that could degrade cefotaxime to products which could influence development.¹⁵ Likewise, antibiotics may also be sensitive to light and higher temperature and the degradation products of these compounds, or the compounds themselves, may display plant hormone like properties and thereby influence plant growth and development.¹⁷

Hypersensitive TAX1-OE plants accumulate less sinapoyl malate

To assess if metabolic changes could account for the observed stress responses, extracts of wild-type and TAX1-OE seedlings grown in a 16-h/8-h light/dark regime or in continuous light were analyzed by Liquid Chromatography Electrospray Ionization Ion Trap Mass Spectrometry (LC-ESI-IT-MS). A principal component analysis (PCA) with the 4,234 detected m/z peaks clearly separated the plants that were cultivated in continuous light from those that were cultivated in a 16-h/8-h light/dark regime (Fig. 2B). Using a Partial least squares Discriminant Analysis (PLS-DA), the m/z peaks responsible for the separation between the two conditions were identified (Fig. 2C). Based on the observed MSⁿ fragmentation, these m/z peaks were shown to be flavonoids, glucosinolates and the phenylpropanoid sinapoyl malate, all of which were higher abundant in the plants growing in the continuous light regime. To identify the metabolites that were different between the TAX1-OE and control lines, PLS-DAs were carried out using only samples grown in a certain light regime. These analyses revealed that sinapoyl malate was the major differential compound and was lower abundant in the TAX1-OE lines grown in the 16-h/8-h light/dark conditions (Fig. 2D). Sinapoyl malate was also observed to be more abundant in the continuous light samples and was shown to play a role in UV-B protection.¹⁸ The addition of cefotaxime to the growth medium did not lead to drastic changes on metabolite level, suggesting that other mechanisms may be involved in the light-stress alleviating role of this antibiotic.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Funding

This work has been supported by funding from the Research Foundation-Flanders through the project G005312 and the Special Research funds from Ghent University and the National Research Foundation (NRF) from South-Africa for a North-South “Sandwich”-type predoctoral scholarship to J.C. L.P. and J.P. are a postdoctoral fellows of the Research Foundation-Flanders.