SUMO E3 ligase SIZ1 negatively regulates arsenite resistance via depressing GSH biosynthesis in Arabidopsis

Arsenic is a metalloid toxic to plants, animals and human beings. Small ubiquitin-like modifier (SUMO) conjugation is involved in many biological processes in plants. However, the role of SUMOylation in regulating plant arsenic response is still unclear. In this study, we found that dysfunction of SUMO E3 ligase SIZ1 improves arsenite resistance in Arabidopsis. Overexpression of the dominant-negative SUMO E2 variant resembled the arsenite-resistant phenotype of siz1 mutant, indicating that SUMOylation plays a negative role in plant arsenite detoxification. The siz1 mutant accumulated more glutathione (GSH) than the wild type under arsenite stress, and the arsenite-resistant phenotype of siz1 was depressed by inhibiting GSH biosynthesis. The transcript levels of the genes in the GSH biosynthetic pathway were increased in the siz1 mutant comparing with the wild type in response to arsenite treatment. Taken together, our findings revealed a novel function of SIZ1 in modulating plant arsenite response through regulating the GSH-dependent detoxification.

Rapid and dynamic SUMO conjugation of cellular proteins is known to be crucial in plant adaption to environmental changes (Morrell and Sadanandom, 2019). In Arabidopsis, the SUMO E3 ligase SIZ1 is essential for SUMOylation of the substrates mediating stress responses (Augustine and Vierstra, 2018). To investigate whether the SIZ1-mediated SUMOylation is involved in plant response to arsenic toxicity, we first performed a phenotypic assay of the Arabidopsis T-DNA insertion mutants siz1-2 (SALK_065397) and siz1-3 (SALK_ 034008) in response to arsenite treatments. Under normal growth conditions without arsenite, the siz1 mutant plants had reduced fresh weight but similar root length comparing with the wild type. However, when treated with 25 μM arsenite, the mutant plants showed increased fresh weight and root elongation when compared with the wild type ( Fig. 1a-c). These results indicated that SIZ1 negatively regulates arsenite resistance in Arabidopsis. Since natural inorganic arsenic compounds that can be absorbed by and toxic to plants mainly consist of trivalent arsenite [As (III)] and oxidative pentavalent arsenate [As(V)] (Ashraf et al., 2020), we also tested the response of siz1 mutants to arsenate stress. The results showed that the siz1 mutants and Col-0 wild type responded similarly to arsenate treatment ( Fig. 1d and e), indicating that SIZ1 plays an important role in the detoxification of arsenite but not arsenate in Arabidopsis. We further detected the SUMOylation profiles in Col-0 wild type and siz1-2 mutant seedlings with or without sodium arsenite treatment. The immunoblot assay using anti-AtSUMO1 antibody showed dynamic changes of SUMO-conjugated products in Col-0 wild type with an increase in SUMOylation after As (III) treatment for 2 h and a decline after a Arsenite resistance assay of Col-0, siz1-2, and siz1-3 seedlings. Five-dayold seedlings grown in 0.5× MS medium were transferred to medium with or without 25 μM As (III) for additional 12 days. Bar, 1 cm. b, c Fresh weight (b) and root length (c) of Col-0, siz1-2, and siz1-3 seedlings shown in (a). d, e Fresh weight (d) and root length (e) of Col-0, siz1-2, and siz1-3 seedlings treated with 0, 25, 50, 100 or 250 μM As(V). Five-day-old seedlings grown in 0.5× MS medium were transferred to medium with or without As(V) for additional 12 days. Values are means ± SD of three replicates, each replicate containing 9 plants per genotype. *** P < 0.001, Student's t test. f SUMOylation profiles of wild type Col-0 and siz1-2 seedlings after As (III) treatment. Ten-day-old Col-0 and siz1-2 seedlings grown in 0.5× MS medium were treated with 100 μM As (III) for the indicated time periods. The SUMOylation profiles were determined by Western blot using anti-SUMO1 antibody. The Coomassie blue (CCB) staining was used as a loading control Fig. 2 Molecular complementation and mimicking siz1-2 arsenite-resistance phenotype by dominant-negative SCE1 overexpression. a Transcript levels of SIZ1 determined by qRT-PCR in 18-day-old Col-0, siz1-2, Com-1, and Com-2 seedlings. Com-1 and Com-2 are two independent complementation lines of siz1-2 plants expressing the SIZ1pro:3×HA-SIZ1 transgene. ACT2 was used as an internal control. b Immunoblot analysis of 3×HA-SIZ1 in Col-0, Com-1, and Com-2 plants. The Coomassie blue (CCB) staining was used as a loading control. c Arsenite resistance assay of Col-0, siz1-2, Com-1, and Com-2 plants. Five-day-old seedlings grown on 0.5× MS medium were transferred to medium with or without 25 μM As (III), and the photograph were taken after treatment for 12 days. Bar, 1 cm. d Fresh weight and root length of Col-0, siz1-2, Com-1, and Com-2 shown in (c). Values are means ± SD of three replicates, each replicate containing 9 plants per genotype. *** P < 0.001, Student's t test. e Arsenite sensitivity assay of Col-0, siz1-2 and transgenic lines expressing the E2 conjugation enzyme SCE1. Five-day-old seedlings grown on 0.5× MS medium were transferred to the medium containing 0 or 25 μM As (III) and treated for 12 days. SCE1 WT -1 and SCE1 WT -2, two independent lines of 35S:SCE1 WT -3×FLAG in Col-0 plants. SCE1 C94S -1 and SCE1 C94S -2, two independent lines of 35S:SCE1 C94S -3×FLAG in Col-0 plants. Bar, 1 cm. f Fresh weight and root length measured from the seedlings shown in (e). Values are means ± SD of three replicates, each replicate containing 9 plants per genotype. The letters a and b above columns indicate significant difference relative to Col-0 and siz1-2 mutant, respectively (P < 0.01, Student's t test) longer time treatments, while this pattern was clearly altered in the siz1-2 mutant (Fig. 1f), which further supports that SIZ1-mediated SUMO conjugation is involved in arsenite response in Arabidopsis.
The role of SIZ1 in arsenite response was further consolidated by the molecular complementation of the siz1-2 mutant. The SIZ1 gene with its native promoter and coding sequence was amplified and cloned into pCam-bia1305 vector to generate the SIZ1pro:3×HA-SIZ1 construct, which was then introduced into siz1-2 mutant. Two independent transgenic lines, which fully rescued the dwarf-like phenotype of siz1-2 under normal conditions, were designated as Com-1 and Com-2 and used for further analysis. Quantitative reverse transcription (qRT)-PCR analysis showed that the SIZ1 transcript level was recovered in the complementation lines (Fig. 2a). The immunoblot assay using anti-HA antibody confirmed the expression of the 3×HA-SIZ1 fused proteins in these two complementation lines (Fig. 2b). Phenotypic analysis indicated that the expression of the native SIZ1 gene rescued the arsenite response phenotype of the siz1-2 mutant ( Fig. 2c and d), revealing that the arsenite-resistant phenotype of siz1-2 mutant is resulted from the loss of function of SIZ1 gene. Moreover, we tested the arsenite response of the overexpression line of SUMO E2, designed as SCE1 WT , and the dominantnegative line of SCE1, named SCE1 C94S , which was reported in several previous studies (Tomanov et al., 2013). Two independent SCE1 WT lines (SCE1 WT -1 and SCE1 WT -2), two independent SCE1 C94S lines (SCE1 C94S -1 and SCE1 C94S -2), siz1-2, and Col-0 wild type were used in the analysis. Under normal growth conditions, the SCE1 C94S lines showed a dwarf-like phenotype that resembled the phenotype of siz1-2 mutant, while the phenotype of SCE1 WT lines is similar with the Col-0 wild type. Interestingly, the SCE1 C94S plants also displayed arsenite resistant phenotype with increased fresh weight and root length, which was similar to the siz1-2 mutant, while the SCE1 WT plants were comparable to the Col-0 wild type under As (III) stress condition ( Fig.  2e and f). The similar arsenite resistant phenotype between the siz1 mutant and the dominant-negative SCE1 plants manifests the role of SUMOylation in arsenite response in Arabidopsis.
The uptake of pentavalent arsenate is mediated by the phosphate transporters and the cellular arsenate is then converted into trivalent arsenite by the function of arsenate reductases (LeBlanc et al., 2013;Chao et al., 2014). The cytotoxic arsenite is either extruded from the cytoplasm or complexed with thiol(−SH)-rich peptides, and the formation of arsenite-SH is conducive to reducing the translocation of the harmful arsenite in plants (Tripathi et al., 2007). To explore the molecular mechanism of arsenite-resistance conferred by the siz1 mutations, we measured the contents of arsenic in shoots and roots of siz1-2, siz1-3 and Col-0 wild type seedlings after As (III) treatment to determine whether SIZ1 controls arsenite uptake and accumulation. The result showed no significant differences in arsenic contents in shoot or root of siz1 mutants and Col-0 wild type seedlings ( Fig. 3a and b), suggesting that siz1 mutations did not affect the uptake and accumulation of arsenite. Heavy metals and metalloids lead to excessive production of reactive oxygen species (ROS) which is detoxified by reductive glutathione (GSH) in plants (Yadav, 2010). In addition, glutathione results in the synthesis of thiol(−SH)-rich metal-binding peptides, the phytochelatins (PCs) that are involved in heavy metal tolerance (Angulo-Bejarano et al., 2021). To determine whether these detoxification mechanisms are involved in arsenite tolerance of siz1 mutant, we measured the contents of GSH and its precursor cysteine in siz1 mutants and Col-0 wild type seedlings under normal and As (III) treatment conditions. The siz1 mutants had significantly higher GSH contents than the Col-0 wild type plants under both normal and As (III) treatment conditions, whereas the contents of cysteine were significantly lower in siz1 mutants than Col-0 wild type ( Fig. 3c and d). To further evidence the contribution of GSH accumulation to the As (III) tolerance of siz1 mutants, we tested the As (III) sensitivity of siz1 mutants in the presence of the GSH biosynthesis inhibitor L-buthionine sulfoximine (L-BSO) (Schnaubelt et al., 2015). When supplemented with 100 μM L-BSO, seedling growth was not affected under normal conditions, while the arsenite resistance phenotype of siz1 mutants completely disappeared in the medium with 25 μM As (III) (Fig. 3e and f). These results indicate that increased GSH biosynthetic accumulation is responsible for the arsenite resistance of siz1 mutants. We therefore determined the transcript levels of the key genes involved in GSH biosynthesis and metabolism. The results showed that the expression of GSH1, GSH2, PCS1 and PCS2 were increased in siz1-2 mutant compared to Col-0 wild type under As (III) stress ( Fig. 3g-j), which further supports that increased biosynthesis of GSH results in arsenite resistance of siz1 mutants.
Increased expression of the GSH biosynthetic genes in siz1 mutant suggested a transcriptional regulation conferring SIZ1-mediated arsenite response. The transcription factor PHR1 is a master regulator for phosphate uptake and implicated in arsenic stress response (Navarro et al., 2021). PHR1 was also shown to be a SUMOylation target of SIZ1 (Miura et al., 2005). We therefore tested whether PHR1 gene is involved in GSH biosynthetic regulation and arsenite tolerance in siz1 mutant. The expression levels of GSH1, GSH2, PCS1 and PCS2 were comparable between the phr1 mutant and Col-0 wild type (Fig. 4a-d), indicating that SIZ1mediated arsenite response is unlikely through the function of PHR1. The dwarf-like phenotype of siz1 mutant is caused by increased accumulation of salicylic acid (SA), which can be rescued by the expression of nahG, a bacterial salicylate hydroxylase that catabolizes SA (Miura et al., 2010). We tested whether SA accumulation is associated with arsenite response in siz1 mutant by using the siz1 mutant expressing nahG. The siz1-2 nahG showed a s response to As (III) similar to siz1-2 mutant ( Fig. 4e and f), suggesting that SIZ1 modulates arsenite response via an SA-independent pathway in Arabidopsis. Since SIZ1 activates COP1 (CONSTITUTIVE PHOTOMORPHOGENIC 1), an ubiquitin E3 ligase promoting the degradation of the bZIP transcription factor HY5 (ELONGATED HYPOCOTYL 5), and HY5 is a central positive regulator in sulfur assimilation that provides the thiol group for GSH biosynthesis (Lee et al., 2011;Lin et al., 2016), we speculate that the transcription factor HY5 may be involved in SIZ1-mediated gene regulation in GSH biosynthesis. However, this requires further experimental validation in the future.
In conclusion, we identified a novel function of SIZ1 in modulating arsenite response in Arabidopsis. Overexpression of the dominant-negative SCE1 C94S mimicking the siz1 mutant suggests that protein SUMOylation Fig. 3 Arsenite-resistant phenotype of siz1 mutants is suppressed by the inhibition of GSH accumulation. a, b Arsenic contents measured by ICP-MS analysis in shoot (a) and root (b) of 7-day-old Col-0, siz1-2, and siz1-3 seedlings after As (III) treatment for 4 days. c, d The contents of GSH (c) and cysteine (d) in 7-day-old Col-0, siz1-2, and siz1-3 seedlings after As (III) treatment for 4 days. e The effect of L-BSO on arsenite-resistance of siz1 mutants. Five-day-old seedlings grown on 0.5× MS medium were transferred to the medium supplemented with 25 μM As (III), 100 μM L-BSO, or combination of 25 μM As (III) and 100 μM L-BSO for 12 days. Bar, 1 cm. (f) Fresh weight and root length from the seedlings shown in (e). Values are means ± SD of three replicates, each replicate containing 9 plants per genotype. *** P < 0.001, Student's t test. gj The transcript abundance of GSH1 (g), GSH2 (h), PCS1 (i), and PCS2 (j) determined by qRT-PCR analysis in Col-0 and siz1-2 seedlings after 100 μM As (III) treatment for the indicated time. The ACT2 was used as an internal control. Values are means ± SD (n = 3). * P < 0.05, ** P < 0.01, Student's t test negatively regulates arsenite resistance in Arabidopsis. Our results reveal that SIZ1-mediated SUMOylation modulates arsenite response through the control of GSH biosynthetic genes and thus the accumulation of GSH and cellular detoxification. Rapid industrialization and urbanization have accelerated arsenic pollution in agricultural land and water which adversely affects crop production and human health (Zhao et al., 2010). Our findings provide important genetic insights into plant adaption to heavy metal and metalloid stress and a possible target for gene editing to improve arsenite resistance in crops.

Plant materials and growth conditions
In this study, all the Arabidopsis (Arabidopsis thaliana) genetic materials are in Columbia-0 background. The T-DNA insertion mutants of SIZ1, SALK_065397 and SALK_034008, were obtained from the Arabidopsis Biological Resource Center (ABRC). The phr1, nahG, siz1-2 nahG, transgenic lines had been reported in our previous study (Dong et al., 2019;Kong et al., 2020). After surface-sterilization and stratification at 4°C for 48 h, the seeds were sown on 0.5× Murashige and Skoog medium (pH 5.8) containing 1% (w/v) sucrose and 0.6% (w/v) agar and grown in a growth room at 22°C with 16 h light / 8 h dark condition. To generate the complementation lines, a 2 kbp promoter of SIZ1 was amplified and cloned into the upstream of 3×HA in pCambia1305 vector. The CDS of SIZ1 was then cloned into the downstream of 3×HA to generate the SIZ1pro:3×HA-SIZ1 construct. The construct was introduced into siz1-2 plants by Agrobacterium tumefaciens GV3101 using the floral dip method. The homozygous T4 plants were used for the analyses.

Phenotype assays
For As (III) and As(V) resistance assay, five-day-old seedlings grow on 0.5× MS medium were transferred to 0.5× MS medium containing As (III), As(V), and/or L-BSO. After growth for 12 days, the plates were photographed and the fresh weight and root length were measured. The experiments were performed three times, each containing nine plants per genotype.

Gene expression analysis
12-d-old seedlings were treated with exogenous 100 μM As (III) for indicated times. Total RNA was extracted Fig. 4 The effect of phr1 mutation and SA accumulation on arsenite resistance of siz1 mutants. ad The transcript levels of GSH1 (a), GSH2 (b), PCS1 (c), and PCS2 (d) determined by qRT-PCR analysis in Col-0 and phr1 seedlings after 100 μM As (III) treatment for the indicated time. The ACT2 was used as an internal control. Values are means ± SD (n = 3). ** P < 0.01, Student's t test. e Arsenite resistance assay of Col-0, siz1-2, nahG, and siz1-2 nahG. The seedlings grown on 0.5× MS medium were transferred to the medium containing 25 μM As (III) for 12 days. Bar, 1 cm. f The fresh weight and root length measured from the seedlings shown in (e). Values are means ± SD of three replicates, each replicate containing 9 plants per genotype. The letter a and b above columns indicate significant difference (P < 0.05, Student's t test) among the values using TRIzol reagent (Invitrogen). Reverse transcription was carried out using One-Step gDNA Removal and cDNA Synthesis Supermix (TransGen Biotech), followed by quantitative PCR on a CFX96™ Real-Time system (BIO-RAD) with ChamQ SYBR qPCR Master Mix (Vazyme Biotech co., ltd). Each analysis included three biological replicates. ACT2 was used as an internal control.

Immunoblot analysis
Immunoblot analysis was conducted as described previously . In brief, 10-day-old seedlings were collected and total protein was extracted using the extraction buffer (50 mM Tris-HCl, pH 8.0; 400 mM NaCl; 0.5% (v/v) Nonidet P-40; 10% (v/v) glycerol; 1 mM EDTA; 1 mM dithiothreitol; and 1 mM phenylmethylsulfonyl fluoride). Total proteins were separated in a 10% SDS-PAGE gel and electroblotted to NC membrane (Millipore), and the abundances of 3×HA-SIZ1 were then determined using anti-HA antibody (Roche). For profiling of SUMO1-cojugated proteins after As (III) treatment, 10-day-old seedlings were collected and subjected to 100 μM As (III) for indicated hours. Total protein was extracted and used to determine SUMOylation profiles using an anti-SUMO1 antibody (ab5316, Abcam).

Elemental and metabolites analysis
Ten-day-old seedlings grow on 0.5× MS medium plates were transferred to 0.5× MS medium with or without 25 μM As (III) for additional 4 days. Arsenic contents were measured using inductively coupled plasma mass spectrometry (ICP-MS) as described in previous studies (Chao et al., 2014;Wang et al., 2020). Briefly, the shoots and roots of the seedlings were sampled separately and diluted to 10.0 mL with deionized water after digesting with 0.90 mL nitric acid. Elemental analysis was performed with an ICP-MS (NexION 350D; PerkinElmer) coupled to an Apex desolation system and an SC-4 DX auto sampler (Elemental Scientific Inc., Omaha, NE, US). The content of cysteine and GSH were measured by using cysteine and glutathione assay kits (NJJCBIO, China) following the manufacturer's instructions.