Sugar signaling regulation by arabidopsis SIZ1-driven sumoylation is independent of salicylic acid

ABSTRACT

SUMO is a modifying peptide that regulates protein activity and is essential to eukaryotes. In plants, SUMO plays an important role in both development and the response to environmental stimuli. The best described sumoylation pathway component is the SUMO E3 ligase SIZ1. Its mutant displays inefficient responses to nutrient imbalance in phosphate, nitrate and copper. Recently, we reported that siz1 also displays altered responses to exogenous sugar supplementation. The siz1 mutant is a salicylic acid (SA) accumulator, and SA may interfere with sugar-dependent responses and signaling events. Here, we extended our previous studies to determine the importance of SA in the SIZ1 response to sugars, by introducing the bacterial salicylate hydroxylase NahG into the siz1 background. Results demonstrate that siz1 phenotypes involving delayed germination are partially dependent of SA levels, whereas the sugar-signaling effect of sugars is independent of SA.

Abbreviations

PGGA

post-germination growth arrest

PTM

post-translational modification

SA

salicylic acid

SIZ1

SAP and Miz 1

SnRK

SNF1-Related Kinase

SUMO

Small Ubiquitin-like Modifier

Sugars are vital molecules, acting as both structural and protective components, and sources of cellular energy. Consequently, sugar signaling and homeostasis is tightly regulated by many essential sugar-sensing proteins. Given their importance, these sensors are subjected to multiple points of regulation.¹ Post-translational modifications (PTMs) are well-documented strategies to regulate protein activity in a fast and reversible way.² A specific PTM, sumoylation, consists of modification by a small peptide, designated Small Ubiquitin-like Modifier (SUMO). SUMO is covalently attached to a lysine in a consensus motif of a target protein, via a 3-step enzymatic cascade (E1-E2-E3).³ The SUMO E3 ligase SAP and Miz 1 (SIZ1) is the major sumoylation enhancer in plants during the response to environmental stimuli.⁴ SUMO conjugation can have different effects on target proteins, and is especially involved in the modulation of nuclear-associated functions.⁵ Previously, we uncovered that known arabidopsis SUMO-targets were enriched in carbohydrate-responsive genes,⁶ and have since reported the reciprocal regulation of SUMO and sugar signaling in plants.⁷ Increasing amounts of sugar were shown to induce SUMO-conjugates in a SIZ1-dependent pathway.⁷ Also, the siz1 mutant displayed constitutively lower sugar and starch levels,^7-9 correlated with the up-regulation of sucrose and starch degradation genes.⁷ In response to exogenous glucose, siz1 seedlings presented post-germination growth arrest (PGGA), lower root growth rate and altered root hair morphology. Ultimately, we uncovered that SIZ1 is important for germination via osmotic-related signaling, whereas subsequent involvement of SIZ1 in post-germination growth is dependent of sugar signaling events.⁷

A key issue that remains to be answered is whether these phenotypes depend on salicylic acid (SA) levels. To clarify, it has been demonstrated that knocking-out of SIZ1 leads to the accumulation of SA, an outcome that can also be observed in other sumoylation pathway mutants.^10-12 Some of the siz1 phenotypes involving stress-responses, e.g. cold and chilling tolerance, are SA-dependent.¹³ SA is normally associated with pathogen defense responses, but several reports have highlighted its involvement in both plant development^14,15 and sugar signaling.^16,17 Mutants with constitutively high levels of SA are morphologically dwarfed, and this phenotype can be reverted by knocking-out the enzyme ICS1, or by introducing the transgene NahG, which encodes for a bacterial salicylate hydroxylase. Most significantly, the siz1 mutant's dwarf phenotype is greatly reversed by the transgene NahG.¹⁸ To investigate whether SA-accumulation in siz1 correlated with sugar signaling events, we repeated the previously reported phenotypic characterization of siz1 under exogenous sugar supplementation,⁷ but this time introducing the siz1 NahG genetic background. To this effect, the T-DNA insertion mutant siz1-2 (SALK_065397) was introgressed into the transgenic line NahG, kindly provided by Dr. Miguel Botella (University of Malaga, Spain). Homozygous lines for siz1-2 NahG were determined by siz1-2 phenotype reversion of F3 seedlings.¹⁰ Primers used for genotyping siz1-2 were SIZ1-2 RP, 5′-CACGACAGATGAAGCATTGTG-3′, SIZ1-2 LP, 5′-GAGCTGAAGCATCTGGTTTTG-3′, and LBb1.3, 5′-ATTTTGCCGATTTCGGAAC-3′. Presence of NahG was determined using primers NahG FW, 5′-ACTGGAACTCTGCCGCTA-3′, and NahG RV, 5′-TGAGTTACTAGGGCGTCG-3′.

First, we analyzed whether germination time was affected in response to glucose. Synchronized seeds were sown onto 0.8% agar-solidified half-strength MS medium supplemented with different types and concentrations of sugars.⁷ The seeds germinated horizontally in culture rooms with a 16 h light/8 h dark cycle under cool white light (80 µE m⁻² s⁻¹ light intensity) at 22–24°C. Each replica plate contained > 30 seeds per genotype. Germination was quantified by scoring double green cotyledon appearance for 10 d using a stereomicroscope, and root morphology was observed in an optical fluorescence microscope. As depicted in Fig. 1A, presence of glucose or mannitol generated a delay in germination across the different genotypes. Simultaneously, the siz1 mutant displayed a constitutive germination delay in the control condition (no sugar added), that was also observed in both glucose and mannitol (the latter acting as an osmotic control). Results demonstrate that siz1 germination delay in response to sugar is due to an osmotic effect. Here, we showed that insertion of NahG in the siz1 background partially alleviated the delay in germination time in all the different experimental conditions, thus suggesting that the germination phenotype of siz1 is at least partially related with SA. Subsequently, we analyzed two phenotypes previously associated with glucose signaling but not with an osmotic effect, i.e. PGGA and the formation of root hair developmental defects (basal bulges). As previously demonstrated,⁷ PGGA in the siz1 mutant increased in glucose but not mannitol, indicating the sugar-dependency of this phenotype (Fig. 1B). Additionally, PGGA levels were identical in siz1and siz1 NahG. Earlier results also established the presence of a morphological defect in glucose-grown siz1 roots, characterized by the presence of basal bulges resulting from abnormal root hairs.⁷ In the present report, these defects in root hair development were observable in both siz1 and siz1 NahG (Fig. 1C). These findings indicate that, during early developmental stages, SA influences the SIZ1-dependent germinative and/or osmotic response to sugars, but not SIZ1 involvement in glucose signaling (Fig. 1D).

Figure 1.

Sugar supplementation affects germination rate and induces developmental arrest of siz1 mutant seedlings in a salicylic-dependent and independent manner, respectively. The transgenic line NahG was introgressed into the siz1 mutant background to create the double mutant siz1 NahG. (A) Time-course analysis of germination levels in MS medium supplemented with no sugar (Control), 4% glucose (Glc) or 4% mannitol (Man). (B) Percentage of post-germination growth arrest (PGGA) in different sugar-supplemented media for each genotype. (C) Representative root tip morphology of 2-week-old seedlings grown on media with 4% glucose; boxes highlight defective root hairs. (D) In response to high exogenous sugar supplementation, SIZ1 is implicated in different events of early development: germination time, via an osmotic pathway that is partially dependent of salicylic levels (SA); post-germination development and root hair morphology, via a glucose signaling-dependent pathway. Error bars represent SEM (n ≥ 3 ); a and b represent statistically different populations in a 1-way ANOVA with Tukey's multiple comparisons test.

Collectively, results rule out the influence of SA on the glucose-dependent phenotypes displayed by siz1, suggesting that SIZ1 may be directly regulating components of glucose signaling pathways. A recent report by Crozet and co-workers¹⁹ reinforces this notion, by evidencing that SIZ1 is a direct regulator of the carbon sensing SnRK1 complex in arabidopsis, but SA treatment does not seem to affect SnRK1 signaling. SIZ1-dependent sumoylation targets SnRK1 to ubiquitination and subsequently to its turnover. The accumulation of higher levels of SnRK1 in siz1 may result in SnRK1 pathway over-activation, which may explain glucose oversensitive phenotypes.^7,19 However, considering the high levels of SUMO-conjugates in plants treated with sugars,⁷ it is likely that other targets exists that are involved in SUMO-sugar interplay.

Funding Statement

Work was funded by the Spanish Ministerio de Ciencia y Tecnologia [AGL2013-48913-C2-2-R. Work was funded by FEDER funds through the Operational Program for Competitiveness Factors – COMPETE, and by National Funds through FCT - Foundation for Science and Technology, within the scope of project “SUMOdulator” [FCOMP-01-0124-FEDER-028459 and PTDC/BIA-PLA/3850/2012]. P.H.C. was supported by FCT [SFRH/BD/44484/2008 and PTDC/BIA-PLA/3850/2012]. H.A. was supported by FEDER through COMPETE, and by FCT, for Rede de Investigação em Biodiversidade e Biologia Evolutiva [UID/BIA/50027/2013 and POCI-01-0145-FEDER-006821].

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.