Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice

Please cite this article in press as: Ma et al., Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice, Molecular Plant (2018), https://doi.org/10.1016/j.molp.2018.12.005

Molecular Plant Research Article

Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice Siqi Ma1,3, Ning Tang1,2,*, Xu Li1, Yongjun Xie1, Denghao Xiang1, Jie Fu1, Jianqiang Shen1,4, Jun Yang1, Haifu Tu1, Xianghua Li1, Honghong Hu1 and Lizhong Xiong1,* 1

National Key Laboratory of Crop Genetic Improvement and National Center of Plant Gene Research (Wuhan), Huazhong Agricultural University, Wuhan 430070, China

2

BPMP, CNRS, INRA, Montpellier SupAgro, Universite´ de Montpellier, 34060 Montpellier, France

3Present

address: Tobacco Research Institute of Chinese Academy of Agricultural Sciences, Qingdao 266101, China

4Present

address: Department of Biology, University of Pennsylvania, Philadelphia, PA 19104, USA

*Correspondence: Ning Tang ([email protected]), Lizhong Xiong ([email protected]) https://doi.org/10.1016/j.molp.2018.12.005

ABSTRACT Histone H2B monoubiquitination (H2Bub1) plays important roles in several physiological and developmental processes, but its roles in the regulation of plant stress responses remain elusive. Here, we report that H2Bub1 is crucially involved in abscisic acid (ABA) signaling and drought response in rice. We found that rice HISTONE MONOUBIQUITINATION2 (OsHUB2), an E3 ligase for H2Bub1, interacted with OsbZIP46, a key transcription factor regulating ABA signaling and drought response in rice. Genetic analyses suggest that OsHUB2, upregulated by drought and ABA, positively modulates ABA sensitivity and drought resistance. The H2Bub1 levels were increased in the target genes of OsbZIP46 under the drought stress and ABA treatments, which were positively correlated with their increased expression levels. Interestingly, MODD, a reported suppressor of ABA signaling and drought resistance by mediating OsbZIP46 deactivation and degradation, could reduce the H2Bub1 levels in the target genes of OsbZIP46 by recruiting a putative deubiquitinase OsOTLD1. Suppression of OsOTLD1 in vivo resulted in increased H2Bub1 levels and expression of OsbZIP46 target genes. Collectively, these findings established an elaborate mechanism of histone monoubiquitination in the fine-turning of ABA signaling and drought response by balancing H2Bub1 deposition and removal. Key words: ABA signaling, transcriptional regulation, chromatin remodeling, histone monoubiquitination, rice Ma S., Tang N., Li X., Xie Y., Xiang D., Fu J., Shen J., Yang J., Tu H., Li X., Hu H., and Xiong L. (2019). Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice. Mol. Plant. --, 1–15.

INTRODUCTION Drought is one of the main abiotic stresses affecting plant growth and yield. Sessile plants have evolved various effective mechanisms to cope with drought stress (Hu and Xiong, 2014). Most of the regulatory mechanisms pertaining to drought responses are closely related to the abscisic acid (ABA)-dependent pathway (Xiong et al., 2002; Cutler et al., 2010; Vishwakarma et al., 2017), and ABA has been widely considered as a stress hormone. The ABA signaling pathway has been intensively studied especially after the discovery of the Pyrabactin resistance1 (PYR1)/PYR1-like (PYL) proteins as ABA receptors. Under drought stress conditions, ABA accumulates in plants and forms a complex with PYR/PYL. The ABA–PYR/PYL complex inhibits the phosphatase activities of clade A protein phosphatase 2Cs (PP2Cs), leading to the activation of SNF1-related

type 2 protein kinases (SnRK2s), which further phosphorylate and activate the downstream substrates such as transcription factors (Schweighofer et al., 2004; Ma et al., 2009; Nakashima et al., 2009; Park et al., 2009; Umezawa et al., 2009; Cutler et al., 2010; Lynch et al., 2012; Rodrigues et al., 2013; Zhang et al., 2013). The bZIP (basic region and leucine zipper) transcription factors family, especially members of the ABA-responsive elementbinding factor (ABF) subfamily, are major targets of SnRK2s in ABA core signaling (Nakashima et al., 2009; Banerjee and Roychoudhury, 2017). In rice, many members of the ABF

Published by the Molecular Plant Shanghai Editorial Office in association with Cell Press, an imprint of Elsevier Inc., on behalf of CSPB and IPPE, SIBS, CAS.

Molecular Plant --, 1–15, -- 2019 ª The Author 2018.

1

Please cite this article in press as: Ma et al., Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice, Molecular Plant (2018), https://doi.org/10.1016/j.molp.2018.12.005

Molecular Plant

Histone Monoubiquitination in Rice Drought Response

subfamily have been identified and are predicted to be associated with drought resistance regulation. One of the ABF subfamily members, OsbZIP23, has been reported as a positive regulator of ABA signaling and drought resistance in rice (Xiang et al., 2008). SAPK2, a homolog of SnRK2-type protein kinases, can phosphorylate and activate OsbZIP23 for its activation, which thereby directly regulates a large number of reported genes with functions in stress responses, hormone signaling, and developmental processes (Zong et al., 2016). OsbZIP46, another ABF subfamily member, also positively regulates ABA signaling and drought resistance (Tang et al., 2012). However, the transcriptional activity of native OsbZIP46 is blocked by a negative regulatory domain in OsbZIP46, but is significantly enhanced after being phosphorylated by SAPK-type protein kinases (Tang et al., 2016). In addition to the phosphorylation modification, the activity of transcription factors can be modulated by other post-translational modifications such as dephosphorylation and ubiquitination (DeLong, 2006; Wu et al., 2016; Zhu, 2016).

reported to be involved in regulating growth and development. H2Bub1 mediated by OsHUB1 and OsHUB2, in concert with H3K4me2, modulates the transcriptional regulation of genes involved in anther development (Cao et al., 2015). In addition, OsHUB2 regulates flowering time and yield potential in rice by affecting histone H2B monoubiquitination and gene expression (Du et al., 2016).

Chromatin modifications have also been demonstrated to play critical roles in transcriptional regulation. The covalent modifications of histones include methylation, acetylation, phosphorylation, ADP ribosylation, sumoylation, and monoubiquitination (Turner, 2002; Kouzarides, 2007). Histone monoubiquitination mainly involves histone H2A and H2B (Weake and Workman, 2008). Monoubiquitination of histones has been found to be associated with transcriptional regulation of gene expression and DNA-damage response (Ramanathan and Ye, 2012). Monoubiquitination of H2B (H2Bub1) was first detected in mouse cells and then discovered widely throughout eukaryotes spanning from yeast (Saccharomyces cerevisiae) to humans and plants. In yeast and humans, the E3 ubiquitin ligase Brefeldin A-sensitivity protein1 (BRE1) catalyzes the monoubiquitination of histone H2B, and a yeast bre1 mutant exhibits a developmental defect with increased cell size (Hwang et al., 2003). The BRE1 homologs in human, namely RNF20/hBRE1A and RNF40/ hBRE1B, act as tumor suppressors through regulation of H2Bub1 (Zhu et al., 2005). In Arabidopsis, there are two BRE1 homologs, HISTONE MONOUBIQUITINATION1 (AtHUB1) and AtHUB2, which function non-redundantly as a heterotetramer in the monoubiquitination of histone H2B and regulate a wide range of biological processes. Compared with the wild-type, the athub mutants exhibited diverse phenotypic changes including pale leaves, modified leaf shape, reduced rosette biomass and root growth, and early flowering (Fleury et al., 2007). Moreover, the cutin and wax composition, seed dormancy, and circadian clock were also affected in the athub mutants (Liu et al., 2007; Himanen et al., 2012; Menard et al., 2014). H2Bub1 also plays important roles in plant defense responses to pathogens. AtHUB1 and AtHUB2 have been reported to regulate disease resistance via histone monoubiquitination at disease resistance gene loci (Zou et al., 2014). Specifically, athub1 mutant plants showed susceptibility to necrotrophic fungal pathogens, but no obvious response to Pseudomonas syringae pv Tomato (Pst) DC3000 (Dhawan et al., 2009). The tomato histone H2B monoubiquitination enzymes SIHUB1 and SIHUB2 contribute to the resistance against Botrytis cinerea through balancing the salicylic acid and jasmonic acid/ethylene-mediated signaling pathways (Zhang et al., 2015). In rice, OsHUB1 and OsHUB2, the homologs of Arabidopsis AtHUB1 and AtHUB2, have been 2

Molecular Plant --, 1–15, -- 2019 ª The Author 2018.

Monoubiquitination of histone H2B is dynamically regulated and can also be reversed by histone deubiquitinases (Huang and Cochran, 2013). Among the 50 deubiquitination enzymes in Arabidopsis, ubiquitin-specific protease 26 (UBP26), UBP22, and OTLD1 (a putative histone deubiquitinase) are known to target histones H2B (Komander et al., 2009; Isono and Nagel, 2014; Nassrallah et al., 2018). UBP26 acts as a transcriptional repressor involved in seed development, and also acts as a transcriptional activator in flowering by activating FLOWERING LOCUS C (FLC) (Luo et al., 2008; Schmitz et al., 2009). UBP22, regulated by a light-signaling factor DE-ETIOLATED1, may contribute to the light-dependent control on H2Bub homeostasis and photomorphogenesis (Nassrallah et al., 2018). OTLD1 has been reported to be involved in plant organ growth and development by repressing the expression of GA20OX, WUS, OSR2, ARL, and ABI5, and may associate directly and function together with the KDM1-type lysine demethylases to regulate gene expression (Krichevsky et al., 2011; Keren and Citovsky, 2016). H2Bub1 is involved in the regulation of plant growth, development, light signaling, and immune responses. However, whether H2Bub1 is implicated in plant responses to drought stress, and if so the underlying molecular mechanisms, remain unknown. Previously, we reported that the transcription factor OsbZIP46 positively regulates ABA signaling and drought resistance in rice (Tang et al., 2012). The activity of native OsbZIP46 is suppressed by the negative regulatory region domain D, which is required for the interaction of OsbZIP46 with the repressor Mediator of OsbZIP46 Deactivation and Degradation (MODD) (Tang et al., 2016). MODD, a homolog of Arabidopsis thaliana ABSCISIC ACID-INSENSITIVE5 (ABI5)-binding protein (AFP), negatively regulates ABA signaling and drought resistance via deactivation and degradation of OsbZIP46. MODD represses OsbZIP46 activity by interacting with the OsTPR3-HDA702 corepression complex and downregulating histone acetylation level at the target genes of OsbZIP46, and MODD can also promote OsbZIP46 degradation via interaction with the E3 ligase OsPUB70 (Tang et al., 2016). Here, we report that OsHUB2, an H2B monoubiquitinase in rice, interacts with OsbZIP46 and plays a positive role in ABA signaling and drought resistance by increasing the H2Bub1 modification on OsbZIP46 target genes. Furthermore, we found that the H2Bub1 levels at the OsbZIP46 target genes can be repressed by MODD, which interacts with OsOTLD1, a homolog of Arabidopsis deubiquitinase OTLD1. Our study reveals a novel epigenetic regulatory mechanism on ABA and drought responses through reversible H2B monoubiquitination.

RESULTS OsbZIP46 Physically Interacts with OsHUB2 Our previous studies suggest that OsbZIP46 is a key regulator in ABA signaling and drought resistance (Tang et al., 2012, 2016).

Please cite this article in press as: Ma et al., Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice, Molecular Plant (2018), https://doi.org/10.1016/j.molp.2018.12.005

Histone Monoubiquitination in Rice Drought Response A

B

C

D

To further unveil the regulatory mechanism of OsbZIP46, we performed a yeast two-hybrid (Y2H) screening to probe interacting proteins of OsbZIP46. Full-length OsbZIP46 fused with the GAL4 DNA-binding domain was used as a bait to screen approximately 106 yeast transformants. Two positive clones were identified to contain the same cDNA with its full-length sequence encoding a protein identical to OsHUB2 (LOC_Os10g41590) (Figure 1A), which has been reported as an E3 ligase mediating H2B monoubiquitination (H2Bub1) (Cao et al., 2015). The interaction between OsHUB2 and OsbZIP46 was further confirmed by bimolecular fluorescence complementation (BiFC) and co-immunoprecipitation (coIP) in rice protoplasts. In BiFC assays, OsbZIP46-pVYCE(R) and OsHUB2-pVYNE(R) were cotransformed into rice protoplasts. A strong fluorescence signal was observed in cells co-expressing OsbZIP46-cYFP and OsHUB2-nYFP, but no signal was detected when each construct was co-expressed with an empty vector (Figure 1B and Supplemental Figure 1A). This result indicated an interaction between OsbZIP46 and OsHUB2 in rice protoplasts. In the coIP assays, 35S:OsHUB2-HA and 35S:OsbZIP46-Myc were transiently expressed in rice protoplasts. The transformation of 35S:HA and 35S:OsbZIP46-Myc was used as a negative control. Anti-hemagglutinin (HA) beads could successfully precipitate OsbZIP46-Myc only when OsHUB2 was present, suggesting that OsbZIP46 physically interacted with OsHUB2 in vivo (Figure 1C). Furthermore, OsbZIP46 constitutive active forms (CA1, DD, and PA), with deletions of domain D or phosphomimetic mutations (Tang et al., 2012, 2016), were investigated for their potential interactions with OsHUB2 by Y2H assays. The results revealed that OsHUB2 could interact with OsbZIP46 not only in the native form but also in the constitutive active forms (Supplemental Figure 1B), implying that OsHUB2 might be substantially associated with the functional regulation of OsbZIP46.

Molecular Plant Figure 1. OsHUB2 Interacts with OsbZIP46. (A) Interaction between OsHUB2 and OsbZIP46 detected by yeast two-hybrid assay. SC-LTH, synthetic complete Leu-Trp-His medium; 3-AT, 3-amino-1,2,4-triazole; CK+ is the positive control. AD, GAL4 activation domain. BD, GAL4 DNAbinding domain. (B) Confirmation of the interaction of OsHUB2 and OsbZIP46 by BiFC assay in rice protoplasts. Ghd7CFP as a marker for nuclear localization was cotransformed. Scale bars, 20 mm. (C) Confirmation of the OsHUB2-OsbZIP46 interaction by coIP assays in rice protoplasts. Total proteins from the protoplasts expressed with OsbZIP46-Myc and OsHUB2-HA or a vector control (hemagglutinin [HA]) were used. Proteins before (input) and after immunoprecipitation (IP) were detected with the anti-HA and anti-Myc antibodies. (D) The expression of OsHUB2 is upregulated by ABA and drought stress treatment. The x axis indicates the time course of the sampling (h, hours; d, days). Error bars represent the standard deviations (SD) based on three replicates. REL, relative expression level.

Since OsbZIP46 is a major regulator in ABA signaling and drought resistance, we checked the expression levels of OsHUB2 under ABA or drought stress conditions using qRT–PCR. The ABA (100 mM) treatment was applied on 14-day-old seedlings of rice Zhonghua 11 and leaf samples were collected at 0, 1, 3, and 6 h after the treatment. The drought stress treatment was applied to 4-leaf-stage seedlings, and the leaf samples were collected at 0 (when the fluidic water in soil was removed), 1, 3, and 5 days (when all the leaves were completely rolled). The results showed that the transcript abundance of OsHUB2 was upregulated by exposure to ABA treatment and drought stress treatment, and the regulation appeared at the early stages of ABA treatment and drought stress (Figure 1D). This result, together with the interaction of OsHUB2 with OsbZIP46, implies that OsHUB2 may be involved in the regulation of ABA signaling and drought responses.

OsHUB2 Positively Regulates ABA Signaling and Drought Resistance OsbZIP46 has been reported as a positive regulator of ABA signaling and drought resistance in rice. To investigate whether OsHUB2 also regulates ABA signaling and drought resistance, we overexpressed OsHUB2 in rice Zhonghua 11. We randomly selected two independent overexpression lines that contained a single copy of the transgene (Supplemental Figure 2) for further phenotypic and molecular analyses. These overexpression lines showed significantly shorter shoots and roots than the negative transgenic lines under ABA treatment, while there were no significant differences under normal growth conditions (Figure 2A and 2B). Furthermore, the OsHUB2 overexpression lines exhibited greater survival rates than the negative transgenic lines after drought stress treatment (Figure 2C and 2D). These results indicated that overexpressing OsHUB2 could increase ABA sensitivity and drought resistance. Molecular Plant --, 1–15, -- 2019 ª The Author 2018.

3

Please cite this article in press as: Ma et al., Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice, Molecular Plant (2018), https://doi.org/10.1016/j.molp.2018.12.005

Molecular Plant

Histone Monoubiquitination in Rice Drought Response

A

B

E

F

C

G

D

H

Figure 2. Phenotypes of OsHUB2-Overexpression and Knockdown Mutant Plants under ABA and Drought Stress Treatment.

(A) Growth performance of two independent OsHUB2-overexpression transgenic lines (OE-12 and OE-43) and two negative transgenic lines (WT0 -12 and WT0 -41) on 1/2 MS medium containing 0 mM ABA (normal) or 3 mM ABA (ABA treatment). Scale bars, 5 cm. (B) The shoot and root lengths were measured at 10 days after germination shown in (A). Error bars indicate SD (n = 24). The statistical significance between the OE and wild-type plants were determined by Student’s t-test (**P < 0.01, two-sided). (C) Performance of two OsHUB2-overexpression lines and the wild-type under normal and drought stress conditions. Scale bars, 15 cm. (D) Survival rates of the overexpression lines and the wild-type after drought stress treatment. Error bars indicate SD based on three replicates. The statistical significance was determined by Student’s t-test (**P < 0.01, two-sided). (E) Performance of the oshub2-1 and the wild-type seedlings (WT0 -1 and WT0 -9) grown on 1/2 MS medium containing 0 mM or 3 mM ABA. Scale bars, 5 cm. (F) The length of the shoots and roots were measured at 12 days after germination shown in (A). Error bars indicate SD (n = 30). The statistical significance was determined by Student’s t-test (**P < 0.01, *P < 0.05, two-sided). (G) Performance of the oshub2-1 and the wild-type under normal conditions and drought stress treatment. Scale bars, 15 cm. (H) Survival rates of the mutant lines and the wild-type after drought stress treatment. Error bars indicate SD based on three replicates. The statistical significance was determined by Student’s t-test (**P < 0.01, *P < 0.05, two-sided).

To further confirm the role of OsHUB2 in ABA signaling and drought response, we collected two allelic rice T-DNA insertion mutants, a knockdown line oshub2-1 (03Z11AY86-3 with a background of rice Zhonghua 11), and a knockout line oshub2-2 (PFG_2A-00426.R with a background of rice Dongjin), with the insertions in the 50 untranslated region and the 17th exon, respectively (Supplemental Figure 2). Contrary to the results of the OsHUB2 overexpression lines, oshub2-1 and oshub2-2 both showed decreased sensitivity to ABA (Figure 2E and 2F; Supplemental Figure 3). Furthermore, the knockdown line oshub2-1 showed increased sensitivity to drought stress treatment (Figure 2G and 2H), while the knockout line oshub2-2 produced not enough seeds for the evaluation of the drought resistance, which may be partially due to the indispensable 4

Molecular Plant --, 1–15, -- 2019 ª The Author 2018.

function of OsHUB2 in reproductive development such as anther development, as previously reported (Cao et al., 2015). These results together suggest that OsHUB2 positively affects ABA sensitivity and drought resistance.

Positive Correlation between H2Bub1 and Gene Expression in Response to Drought In light of the fact that OsHUB2 acts as an E3 ligase for H2Bub1 and plays a pivotal role in drought resistance, it is tempting to explore the general role of H2Bub1 in drought response. We first examined the overall H2Bub1 levels in wild-type rice Zhonghua 11 under moderate drought stress (relative soil water content at about 20%), severe drought stress (relative soil water content

Please cite this article in press as: Ma et al., Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice, Molecular Plant (2018), https://doi.org/10.1016/j.molp.2018.12.005

Histone Monoubiquitination in Rice Drought Response A

B

Molecular Plant Figure 3. Alteration of H2Bub1 Modification in Drought Response.

(A) The H2Bub1 level detected by western blot under normal (Nor), moderate drought (M-Dr) and severe drought (S-Dr) stress conditions. (B) The H2Bub1 level detected by western blot under ABA treatment. H2B level was detected as a loading control in (A) and (B). Values of band inE tensity relative to the first lanes from the left are C D indicated below the bands. (C) Patterns of drought-responsive genes with H2Bub1 modifications under moderate drought stress conditions. Group 1 are drought-upregulated genes with the H2Bub1 level increased; group 2 are drought-downregulated genes with the H2Bub1 level decreased; group 3 are droughtdownregulated genes with the H2Bub1 level increased; group 4 are genes with the H2Bub1 level changed but not drought responsive. (D) Patterns of drought-responsive genes with H2Bub1 modifications under severe drought stress. Group 1 are drought-upregulated genes with H2Bub1 level increased; group 2 are drought-downregulated genes with H2Bub1 level decreased; group 3 are drought-downregulated genes with the H2Bub1 level increased; group 4 are genes with H2Bub1 level changed but are not drought responsive. (E) Correlation of the fold change of the drought-induced H2Bub1 level and the fold change of the gene expression level based on the data in (C).

at about 10%), and ABA treatments. The results showed that the H2Bub1 level was significantly increased under drought stress and ABA treatments (Figure 3A and 3B). To answer whether the change of the H2Bub1 level is associated with the regulation of drought response, we applied chromatin immunoprecipitation sequencing (ChIP-seq) to obtain the genome-wide H2Bub1 profiles by using anti-ubiquitinated-histone H2B antibody. The DNA fragments were immunoprecipitated from leaf tissues of rice Zhonghua 11 under normal growth conditions, moderate drought stress treatment, and severe drought stress treatment (each with three biological replicates). Meanwhile, the transcriptomes in those plants were investigated using RNA sequencing (RNAseq). The ChIP-seq generated 25–30 million unique mapped reads and more than 10 000 peaks for each sample. Genomic distribution analysis revealed that more than 88% of the peaks were located within the transcribed regions (Supplemental Figure 4A–4D), in line with a previously characterized predominant distribution for H2Bub1 over gene bodies in Arabidopsis (Roudier et al., 2011). By comparing the peaks between normal and moderate drought stress conditions, we found that 555 genes (fold change R2, false discovery rate <0.05) exhibited a significantly difference in H2Bub1 level. Taking into account the RNA-seq results, we further classified these genes into four groups based on their H2Bub1 level changes under moderate drought stress treatment and their responsiveness to the drought stress (Figure 3C, Supplemental Figure 4E and Supplemental Data 1). Interestingly, more than half of the differentially H2Bub1 modified genes under moderate drought stress conditions were drought-responsive genes. Among the genes with H2Bub1 levels significantly increased, almost all (218 genes, group 1) were significantly upregulated by drought stress, and only a few of them were downregulated (six genes, group 3). All of the 80 genes with H2Bub1 levels significantly decreased were significantly downregulated (group 2) by drought stress treatment. Among the other differen-

tially H2Bub1 modified genes (group 4), most of the genes with H2Bub1 levels significantly increased (171 genes) or decreased (80 genes) were also slightly upregulated or downregulated by drought stress treatment. Similar patterns were observed for the differential H2Bub1 modification and differential gene expression between normal growth and severe drought stress conditions (Figure 3D and Supplemental Data 2). Moreover, our data showed that there was a significant positive correlation between the differential H2Bub1 modification and gene expression levels (Figure 3E; Supplemental Figure 4F and 4G), in agreement with a generally recognized positive role of H2Bub1 on gene expression (Roudier et al., 2011; Cao et al., 2015). Among these differentially H2Bub1-modified genes, at least 60 genes were ABA responsive according to the transcriptome data (Garg et al., 2012) under ABA treatment (Supplemental Data 3). These results implied that H2Bub1 is involved in the regulation of both ABA and drought responses. In addition, we noticed that there were more genes (394 versus 220) with significantly increased H2Bub1 levels but fewer genes (160 versus 1114) with significantly decreased H2Bub1 levels under moderate drought stress conditions compared with severe drought stress conditions, which suggested that the H2Bub1 modification under drought stress conditions may be dynamically regulated.

The Role of H2Bub1 in Drought Responsiveness Largely Depends on OsbZIP46 Since H2Bub1 is associated with the expression of droughtresponsive genes and OsHUB2 can interact with OsbZIP46, it would be intriguing to further examine the expression levels of OsbZIP46 target genes in the OsHUB2-overexpression lines. We found that the expression levels of OsbZIP46 and its target genes such as RAB21 were upregulated in the OsHUB2-overexpression lines, especially after 3 h of ABA treatment or under Molecular Plant --, 1–15, -- 2019 ª The Author 2018.

5

Please cite this article in press as: Ma et al., Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice, Molecular Plant (2018), https://doi.org/10.1016/j.molp.2018.12.005

Molecular Plant

Histone Monoubiquitination in Rice Drought Response

A

C

F

B

D

G

E

J

K

H

I

L

Figure 4. The Role of H2Bub1 in Drought Responsiveness Largely Depends on OsbZIP46. (A and B) The relative expression level (REL) of OsbZIP46 and RAB21 in the OsHUB2-OE and wild-type plants under ABA treatment. The x axis indicates the time course of the sampling (h, hours). Error bars represent SD from three replicates. The statistical significance was determined by Student’s t-test (*P < 0.05, two-sided). (legend continued on next page)

6

Molecular Plant --, 1–15, -- 2019 ª The Author 2018.

Please cite this article in press as: Ma et al., Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice, Molecular Plant (2018), https://doi.org/10.1016/j.molp.2018.12.005

Histone Monoubiquitination in Rice Drought Response

Molecular Plant

moderate drought stress treatment (Figure 4A–4D). Meanwhile, the H2Bub1 levels at OsbZIP46 and its target gene RAB21 were also significantly increased under drought stress treatment (Figure 4E). On the other hand, the H2Bub1 levels at several target genes of OsbZIP46 (randomly selected based on the results by Tang et al., 2012 and our unpublished ChIPseq data of OsbZIP46) were significantly decreased in the oshub2-1 and oshub2-2 mutants (Figure 4F and 4G) and increased in the OsHUB2 overexpression line (Figure 4H). These results indicate that both the H2Bub1 and expression levels of the OsbZIP46 target genes may be affected by OsHUB2, probably due in part to its interaction with OsbZIP46 and, apparently, its general effects on H2Bub1 regulation (Cao et al., 2015 and this study).

To further elucidate the association between OsbZIP46 and H2Bub1 regulation, we investigated the potential functional dependency of OsHUB2 on OsbZIP46 using a transient expression system in rice protoplasts. The activation of RAB21 by OsHUB2 was evaluated in protoplasts isolated from wild-type or osbzip46 knockout mutant plants. The results showed that OsHUB2 significantly activated the expression of RAB21 in wild-type, whereas the activation was greatly attenuated in osbzip46 mutant (Figure 4L), implying that the function of OsHUB2 on gene expression regulation could be, at least partially, dependent on OsbZIP46. Taken together, the results in this section suggest that the H2Bub1 modification at the OsbZIP46 target genes in response to drought stress is largely dependent on OsbZIP46 and its interaction with OsHUB2.

Next, we wondered whether the changes of H2Bub1 modification in response to drought stress are associated with OsbZIP46. We compared the genes with significantly changed H2Bub1 modification levels and the genes directly targeted by OsbZIP46 in drought response. We found that 140 of 555 H2Bub1 levelchanged genes were directly target genes of OsbZIP46 under drought stress treatment (Supplemental Data 4; Supplemental Figure 4E and 4H). Furthermore, we checked the H2Bub1 levels of several OsbZIP46 target genes in the osbzip46 mutant and wild-type lines by ChIP–qPCR. The results showed that the H2Bub1 levels of these genes were decreased in the osbzip46 mutant (Figure 4I). To establish this point further, we examined the global H2Bub1 profile in the osbzip46 mutant using ChIPseq (Supplemental Figure 5). In comparison with the wild-type, 1287 genes exhibited differential H2Bub1 levels in the osbzip46 mutant under drought stress condition (Supplemental Data 5). Notably, all of these genes were downregulated in the mutant, and 223 of them were directly target genes of OsbZIP46, among which 83 genes were drought responsive (Figure 4J and Supplemental Figure 5). Consistent with this, gene ontology (GO) analysis of the 223 target genes showed significant enrichments in the functional category of ‘‘response to stress’’ (Figure 4K). These results support that OsbZIP46 significantly regulates the H2Bub1 levels at its target genes in response to drought stress. Nevertheless, the target-binding capability of OsbZIP46 in oshub2-1 plants showed no difference when compared with the wild-type plants (Supplemental Figure 6), suggesting that OsHUB2 may not affect the binding of OsbZIP46 to its target genes.

MODD Negatively Regulates the H2Bub1 Level Although the H2Bub1 level was significantly upregulated under drought stress, it was decreased under severe drought treatment compared with moderate drought stress (Figure 3A), and was also decreased at 9 h after ABA treatment compared with 3 h after ABA treatment (Figure 3B). Meanwhile, the H2Bub1 levels of about 1000 genes were decreased under severe drought stress conditions (Figure 3D and Supplemental Data 2). Therefore, we speculated that there may be some negative regulators involved in the H2Bub1 modification. Since our previous study reported that MODD negatively regulates ABA signaling and drought response via deactivation and degradation of OsbZIP46, we wondered whether MODD is also involved in the H2Bub1 changes in drought response. First, we examined the expression of RAB21, a typical target gene of OsbZIP46, in rice protoplast with OsHUB2 or MODD transformed, or with OsHUB2 and MODD co-transformed under normal and ABA treatment. The results showed that the RAB21 was upregulated in the protoplast with OsHUB2 transformed under ABA treatment, but was downregulated when MODD and OsHUB2 were co-transformed (Figure 5A). To answer whether MODD affected the H2Bub1 modification, we further checked the H2Bub1 levels in the modd mutant and the wild-type using the anti-ubiquitinated histone H2B specific antibody. The results showed that the H2Bub1 level was significantly higher in the mutant than in wild-type (Figure 5B), suggesting that MODD could also affect the H2Bub1 level. We then quantified the H2Bub1 levels of several OsbZIP46 target genes in the modd

(C and D) The REL of OsbZIP46 and RBA21 in the OsHUB2-OE and wild-type plants under normal, moderate drought stress, and severe drought stress conditions. Error bars represent SD (n = 3). The statistical significance was determined by Student’s t-test (**P < 0.01, *P < 0.05, two-sided). (E) The H2Bub1 level (relative enrichment, RE) at RAB21 and OsbZIP46 detected by ChIP–qPCR under normal, moderate drought stress, and severe drought stress conditions. The data indicate the mean values ± standard errors (SE) of three biological replicates. The statistical significance was determined by Student’s t-test (**P < 0.01, *P < 0.05, two-sided). (F–I) H2Bub1 levels (relative enrichment, RE) at OsbZIP46 target genes in oshub2-1, oshub2-2, OsHUB2-OE, osbzip46, and wild-type (WT) plants. The data indicate mean values ± SE of three biological replicates. The statistical significance between mutant and wild-type plants was determined by Student’s t-test (**P < 0.01, *P < 0.05, two-sided). (J) Venn diagram showing 1287 genes with decreased H2Bub1 level in osbzip46 versus WT under moderate drought stress, and its overlapping with OsbZIP46 target and drought-responsive genes. (K) Gene ontology (GO) analysis for the 223 OsbZIP46 target genes shown in (B). The number of genes for each enriched functional category (P < 0.005) is shown. Note that the ‘‘other categories’’ consists of the following GO terms (each with 2–4 genes): starch biosynthetic process, protein heterodimerization activity, regulation of flower development, phosphorelay sensor kinase activity, protein import into peroxisome matrix, glucosinolate biosynthetic process, and mitochondrial transport. (L) The expression level of RAB21 was detected in the osbzip46 mutant and wild-type protoplasts co-transformed with OsHUB2 or the empty vector (CK). Error bars represent SD from three replicates. The statistical significance was determined by Student’s t-test (**P < 0.01, two-sided; n.s., not significant).

Molecular Plant --, 1–15, -- 2019 ª The Author 2018.

7

Please cite this article in press as: Ma et al., Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice, Molecular Plant (2018), https://doi.org/10.1016/j.molp.2018.12.005

Molecular Plant

Histone Monoubiquitination in Rice Drought Response Figure 5. MODD Can Repress the H2Bub1 Modification.

A

B

C

mutant and MODD-overexpression lines by ChIP–qPCR. The H2Bub1 levels at several random chosen target genes were significantly decreased in the MODD-overexpression lines compared with the wild-type rice Zhonghua 11 (Figure 5C). On the contrary, H2Bub1 levels at these genes were significantly increased in the modd mutant lines compared with the corresponding wild-type rice Dongjin (Figure 5C). These results implied that MODD could downregulate the H2Bub1 levels at the OsbZIP46 target genes.

MODD Interacts and Functions Together with Histone Deubiquitinase OsOTLD1 to Regulate H2Bub1 To explain how MODD attenuates the H2Bub1 levels of OsbZIP46 target genes, we explored the potential link between MODD and the removal of ubiquitin on H2B. OTLD1 is a transcriptional repressor of ABI5, a close homolog of OsbZIP46 in Arabidopsis, via deubiquitination of H2B (Keren and Citovsky, 2016). We cloned the only homologous gene of OTLD1 in rice for further study. It was exciting to discover that MODD interacted with OsOTLD1 (LOC_Os04g33780) in yeast (Figure 6A), and the interaction was confirmed by BiFC and coIP assays in rice protoplasts (Figure 6B and 6C; Supplemental Figure 7). We then examined the function of OsOTLD1 in H2Bub1 modification along with MODD in rice. We checked the transcript levels of OsOTLD1 under ABA and drought stress treatments. The results showed that OsOTLD1 transcript abundance was upregulated by ABA and drought treatments, but the regulation appeared at the late stages of the treatments (Figure 6D), which is quite similar to the regulation pattern of MODD at the late stages (Tang et al., 2016), implying that MODD and OsOTLD1 may work together to turn down the H2Bub1 levels during the late stages of the ABA and drought stress treatments. Next, we investigated 8

Molecular Plant --, 1–15, -- 2019 ª The Author 2018.

(A) MODD repressed the transcriptional activity of OsbZIP46 in rice protoplasts (in ZH11 background) whether OsHUB2 was transformed or not, as indicated by the transcript level of RAB21, a direct target gene of OsbZIP46. The data indicate mean values ± SE of three biological replicates. The statistical significance was determined by Student’s ttest (**P < 0.01, two-sided; n.s., not significant). (B) The H2Bub1 level was increased in the modd-2 mutant. The H2B protein level was detected as a loading control. Values of band intensity relative to the first lanes from the left are indicated below the bands. Only one set of three repeated data is shown. The statistical significance between modd2 and wild-type (DJ) plants based on overall data from three repeats (n = 6 for each genotype) was determined by Student’s t-test (**P < 0.01, twosided; n.s., not significant). (C) H2Bub1 levels at four target genes of OsbZIP46 were measured by ChIP–qPCR in MODD-overexpression lines (OE-20), modd mutant lines (modd-2), and the corresponding wild-types (ZH11 and DJ). The data indicate mean values ± SE of three biological replicates. The statistical significance was determined by Student’s t-test (**P < 0.01, *P < 0.05, two-sided).

whether OsOTLD1 possesses the histone deubiquitination function. Purified OsOTLD1 was incubated with bovine histones, and the H2B ubiquitination levels were checked by western blot. The results showed that the monoubiquitinated histone H2B levels were gradually decreased with the presence of increasing concentrations of OsOTLD1 (Figure 6E). These results confirmed that OsOTLD1 could deubiquitinate histone H2B in vitro just like the homologous protein OTLD1. The function of OsOTLD1 was further checked in rice protoplasts. When the OsOTLD1 was co-transformed with OsbZIP46 into the protoplasts, the expression of RAB21 was slightly suppressed under both normal and ABA treatment conditions, but the expression of RAB21 was dramatically suppressed when MODD was co-transformed with OsOTLD1 and OsbZIP46 (Figure 6F). In contrast, the suppression was abolished in modd-2 mutant protoplasts (Figure 6G), suggesting that OsOTLD1 negatively regulates the transactivation activity of OsbZIP46 in a MODD-dependent manner. In addition, we attempted to suppress the expression level of OsOTLD1 in rice protoplasts using the CRISPR/Cas9 technique to delete a fragment from the coding region of OsOTLD1. The deletion of the fragment was detected in part of the protoplasts by PCR (Supplemental Figure 8A and 8B). qRT–PCR suggested that the OsOTLD1 expression level, with the amplification region covering the deleted region, was suppressed, and the expression levels of OsbZIP46 and its target genes were upregulated (Figure 6H). Meanwhile, the H2Bub1 levels at these genes were increased in the OsOTLD1-suppressed protoplasts (Figure 6I). Furthermore, we identified a T-DNA insertion line osotld1, which appeared to be a loss-of-function mutant (Supplemental Figure 8C and 8D). osotld1 exhibited increased ABA sensitivity in comparison with wild-type (Figure 6J), suggesting that OsOTLD1, in contrast to OsbZIP46, negatively regulates ABA signaling. All of these results demonstrate that

Please cite this article in press as: Ma et al., Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice, Molecular Plant (2018), https://doi.org/10.1016/j.molp.2018.12.005

Histone Monoubiquitination in Rice Drought Response

Molecular Plant

the OsOTLD1 functions as a deubiquitinase of histone H2B to repress the expression of OsbZIP46 and its target genes by decreasing the H2Bub1 level. The results in this section taken together suggest that MODD negatively regulates the H2Bub1 level by interacting with the deubiquitinase OsOTLD1.

newly published work pointed to substantial roles for AtHUB2 and GhHUB2 (a homolog of AtHUB2 in cotton) in drought stress response (Chen et al., 2018). The overall data suggest that H2Bub1 may function as a general epigenetic regulatory mechanism in response to drought stress.

DISCUSSION

Dynamic and Reversible H2Bub1 Modification in Response to Drought Stress

H2Bub1 Participates in the Regulation of ABA Signaling and Drought Response

The overall H2Bub1 level was induced by drought stress treatments, but showed a dynamic change during the drought stress treatment (Figure 3A). For example, the H2Bub1 levels at the OsbZIP46 and RAB21 genes were induced by drought stress treatment, but the H2Bub1 levels were significantly lower under severe drought stress than moderate drought stress conditions (Figure 4E). Although the overall H2Bub1 level was increased compared with normal conditions, the proportion of drought-responsive genes with increased H2Bub1 levels was obviously lower under severe drought stress than moderate drought stress conditions (Figure 3C and 3D). For many genes, the H2Bub1 levels were decreased under severe drought conditions compared with the levels under moderate drought stress. These results indicate that the H2Bub1 levels at different genes may be differentially and dynamically modulated during the drought stress treatment. In agreement with our observations, a similar dynamical regulation of H2Bub1 was described for Arabidopsis in response to light treatments (Bourbousse et al., 2012), indicating a potential general regulatory mechanism in plant adaptive response. Presumably, the global low H2Bub1 level under severe drought stress may downregulate transcription through interplay, to some extent, with H3K4me2 (Cao et al., 2015), in a complex context of transcriptional regulatory network and chromatin remodeling (Kouzarides, 2007). Interestingly, the expression levels of most genes with decreased H2Bub1 levels under severe drought stress treatment were also downregulated by the stress. GO analysis suggests that these genes belong to drought stress-related categories including ‘‘response to hormone,’’ ‘‘response to stimulus,’’ and ‘‘response to abiotic stimulus’’ (Supplemental Figure 11), implying that the dynamic change of the H2Bub1 level is consistent with the dynamic change of the expression levels for many droughtresponsive genes. Although it is difficult to completely explain how H2Bub1 is dynamically regulated during the drought stress treatment, OsOTLD1, encoding an H2B deubiquitinase, may crucially account for the decreased H2Bub1 levels under severe drought stress conditions.

The modification of histone H2Bub1 is well conserved from yeasts to humans and plants, and is involved in various biological processes. E3 ligases OsHUB1 and OsHUB2 constitute major determinants of H2Bub1 in rice, and have been importantly recognized for functions on growth and developmental regulation (Cao et al., 2015; Du et al., 2016). In the present study, we found that OsHUB2 interacts with the key regulator OsbZIP46 and plays a positive role in ABA signaling and drought stress resistance, thereby substantiating the link between H2Bub1 modification and plant stress response. The previous study revealed that OsHUB1 can interact with OsHUB2 and each of them can interact with itself (Cao et al., 2015). Accordingly, OsHUB1 may also function in ABA signaling and drought stress response via the formation of a heterotetrameric complex with OsHUB2 or the interaction with transcriptional regulators. In fact, we found that overexpression of OsHUB1 also enhanced the ABA sensitivity and drought stress resistance just like OsHUB2 (Supplemental Figure 9), suggesting that both OsHUB2 and OsHUB1 positively regulates the ABA signaling and drought stress resistance. H2Bub1 was induced under drought stress conditions, especially under moderate drought stress (Figure 3A). The ChIP-seq profiles showed that a large proportion of genes with increased H2Bub1 levels under moderate drought stress treatment were drought responsive and/or ABA responsive (Figure 3C and Supplemental Data 3), indicating that H2Bub1 plays an important role in the regulation of ABA signaling and drought stress response. Notably, a significant proportion of the genes with H2Bub1 levels changed under moderate drought stress treatment are OsbZIP46 target genes. We further confirmed that the H2Bub1 modifications on target genes may be largely dependent on the presence of OsbZIP46 (Figure 4, Supplemental Figure 5, and Supplemental Data 5). Although a definitive analysis for the genetic interaction remains to be conducted, our results based on the transient expression system suggest that the function of OsHUB2 may be partially dependent on OsbZIP46. These data imply that the role of H2Bub1 in ABA and drought stress responses may be fulfilled mainly through the interaction with OsbZIP46, although it cannot be excluded that the H2Bub1 may be involved in the regulation of ABA and drought stress responses through the coordination with other regulators. Moreover, the transcript abundance of AtHUB1, AtHUB2, and AtOTLD1 in Arabidopsis were also upregulated by drought stress treatments (Supplemental Figure 10), with patterns similar to those for their homologs in rice. Interestingly, the H2Bub1 level was also induced in Arabidopsis under drought stress treatments (Supplemental Figure 10). Consistent with this, a

Reversible histone modifications, such as acetylation and methylation, have been reported to be involved in the regulation of stress-responsive genes (Chinnusamy et al., 2008; Kim et al., 2015). These modifications are established or erased by specific enzymes. For histone ubiquitination, histone H2B is monoubiquitinated by E3 ligases (Liu et al., 2007; Cao et al., 2015) and the modification can be removed by histone deubiquitinases (Luo et al., 2008; Krichevsky et al., 2011). Here, we found that the H2Bub1 level in rice was increased at the early stages of drought stress treatment, in which the OsHUB2 transcript abundance was strongly upregulated, and declined at the late stages of drought stress treatment when the OsOTLD1 gene was strongly upregulated. The elaborate Molecular Plant --, 1–15, -- 2019 ª The Author 2018.

9

Please cite this article in press as: Ma et al., Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice, Molecular Plant (2018), https://doi.org/10.1016/j.molp.2018.12.005

Molecular Plant A

Histone Monoubiquitination in Rice Drought Response B

C

D

F

E

G

H

J

I

Figure 6. OsOTLD1 Interacts with MODD and Acts as a Deubiquitinase of H2B to Repress the Expression of OsbZIP46 Target Genes. (A) Interaction between MODD and OsOTLD1 detected by Y2H assay. SC-LTH, synthetic complete Leu-Trp-His medium; 3-AT, 3-amino-1,2,4-triazloe; CK+ is the positive control. AD, GAL4 activation domain. BD, GAL4 DNA-binding domain. (B) MODD interacted with OsOTLD1 in rice protoplasts by the BiFC assays. Ghd7-CFP as a marker for nuclear localization was co-transformed. Scale bars, 20 mm. (legend continued on next page)

10

Molecular Plant --, 1–15, -- 2019 ª The Author 2018.

Please cite this article in press as: Ma et al., Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice, Molecular Plant (2018), https://doi.org/10.1016/j.molp.2018.12.005

Histone Monoubiquitination in Rice Drought Response

Molecular Plant Figure 7. Working Model for the H2Bub1 Modification in the Regulation of StressResponsive Genes. H2Bub1 mediated by the OsHUB2 complex plays pivotal roles in stress responses. OsHUB2 can interact with OsbZIP46 when plants are exposed to drought stress conditions, and activate the stressresponsive genes by increasing the H2Bub1 levels. The dotted MODD represents a low expression level of MODD in the early stage of drought stress condition (Tang et al., 2016). At the late drought stress or drought recovery stage, when the expression level of MODD increases, MODD may compete with OsHUB2 to interact with OsbZIP46 and recruit OsOTLD1 to attenuate the stress signaling by decreasing the H2Bub1 level.

adjustments of transcript abundance for OsHUB2 and OsOTLD1 could be attributed to regulation of mRNA synthesis and/or decay, while the underlying mechanisms remain to be further explored. These dynamic changes suggest that a reversible histone H2B monoubiquitination is involved in the fine-tuning of drought response. OsOTLD1 possessed the enzymatic activity of H2B deubiquitination in vitro and the H2Bub1 levels were increased at Osb ZIP46 target genes in the rice protoplast when OsOTLD1 was suppressed, which further supported the role of OsOTLD1 in the removal of H2B monoubiquitination. Interestingly, we found that OsOTLD1 interacts with MODD, a key suppressor protein mediating the deactivation and degradation of OsbZIP46. This implies that the reversal of H2Bub1 under severe drought stress conditions by OsOTLD1 may be associated with MODD. A few additional pieces of evidence support this hypothesis. First, the H2Bub1 levels of the OsbZIP46 target genes are lower in the MODD-overexpression plant but higher in the modd mutant (Figure 5C). Second, MODD and OsOTLD1 showed similar late responsiveness to the drought stress (Tang et al., 2016;

Figure 6D). Third, the expression of the target gene RAB21 was dramatically reduced in the rice protoplasts cotransformed with MODD and OsOTLD1 compared with the transformation of OsOTLD1 alone (Figure 6F). Lastly, OsOTLD1 deactivates OsbZIP46 in a MODD-dependent manner (Figure 6G). Presumably, the OsbZIP46-interacting protein MODD may recruit OsOTLD1 to erase the H2Bub1 modification on OsbZIP46 target genes, and thereby repress the transactivation activity of OsbZIP46 (Figure 7). In particular, the late responsiveness for OsOTLD1 to drought stress hinted at a potential role during late stress or recovery stage. OsOTLD1 and MODD may cooperatively contribute to the timely erasure of H2Bub1 enrichment when the stress is alleviated or removed, so as to coordinate plant stress responses with development and growth. Plants exhibit arrays of dynamic changes at different levels in responses to drought stress. At the transcriptional level, the dynamic changes of stress-responsive genes are regulated by a line of molecular mechanisms. Our results together suggest that the reversible and dynamic H2Bub1 modification mediated by the OsbZIP46–OsHUB2 complex (deposition) and MODD– OsOTLD1 complex (removal) provides a novel and elaborate mechanism to regulate the expression of drought-responsive genes (Figure 7).

(C) Confirmation of the MODD-OsOTLD1 interaction by coIP in rice protoplasts. Total proteins from the protoplasts expressed with OsOTLD-Myc and MODD-HA or a vector control (HA) were used. Proteins before (input) and after IP were detected by the anti-Myc and anti-HA antibodies. (D) The expression of OsOTLD1 is upregulated by ABA and drought stress treatment. The x axis indicates the time course of sampling (h, hours; d, days). Error bars represent SD from three replicates. (E) Deubiquitination of monoubiquitinated H2B by OsOTLD1. Purified GST-OsOTLD1 was incubated with 20 mg of total purified bovine histones and then the H2Bub1 level was detected by western blot. H2B level was detected as a loading control. Values of band intensity relative to the first lanes from the left are indicated below the bands. (F) The expression level of RAB21 was repressed in rice protoplasts (in ZH11 background) co-transformed with OsOTLD1. Error bars represent SD from three replicates. The statistical significance was determined by Student’s t-test (**P < 0.01, two-sided). (G) The expression level of RAB21 was detected in the modd-2 mutant and wild-type protoplasts (in DJ background) co-transformed with OsOTLD1. Error bars represent SD from three replicates. The statistical significance was determined by Student’s t-test (**P < 0.01, two-sided; n.s., not significant). (H) The expression of OsbZIP46 and its target genes were increased in rice protoplasts with OsOTLD1 silenced by the CRISPR/Cas9 technique. Error bars represent SD from three replicates. The statistical significance was determined by Student’s t-test (**P < 0.01, *P < 0.05, two-sided). (I) The H2Bub1 level of OsbZIP46 and its target genes were increased in rice protoplasts with OsOTLD1 expression suppressed. (J) Performance of the osotld1 mutant and the wild-type seedlings grown on 1/2 MS medium containing 0 mM or 3 mM ABA (Scale bars, 2 cm). The shoot and root lengths were measured at 10 days after germination shown in (I). Error bars indicate SD (n = 24). The statistical significance between the mutant and wild-type plants were determined by Student’s t-test (**P < 0.01, two-sided).

Molecular Plant --, 1–15, -- 2019 ª The Author 2018.

11

Please cite this article in press as: Ma et al., Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice, Molecular Plant (2018), https://doi.org/10.1016/j.molp.2018.12.005

Molecular Plant

Histone Monoubiquitination in Rice Drought Response

OsbZIP46 May Integrate Different Epigenetic Modifications to Fine-Tune ABA Signaling and Drought Response

drought stress response. In addition, as discussed earlier, there is also evidence to link AtHUB1, AtHUB2, and OTLD1 to drought response in Arabidopsis. Collectively, the findings indicate that the model (Figure 7) we proposed may be a general regulation mechanism in plants.

OsbZIP46 is a key transcription factor ‘‘translating’’ the upstream ABA signaling into the activation of downstream drought resistance-related genes (Tang et al., 2016). In the previous study, we proposed a model pertaining to the coordination of OsbZIP46 and MODD in ABA signaling and drought response: OsbZIP46 is induced and activated by SnRK2-type protein kinases at early drought stress stages, but MODD is not induced at the early stages, so that the function of activated OsbZIP46 is not significantly affected by the low levels of the MODD protein; When the stress strength continues, MODD is induced and interacts with OsbZIP46 (via its D domain) to attenuate the activity and stability of OsbZIP46 (Tang et al., 2016). MODD represses OsbZIP46 activity via interaction with the OsTPR3-HDA702 corepressor complex to turn down the histone acetylation level at OsbZIP46 target genes (Tang et al., 2016). Histone acetylation is a well-known epigenetic modification affecting the expression of stress-responsive genes. In this study, we found that OsbZIP46–OsHUB2 and MODD–OsOTLD1 complexes antagonistically regulate the H2Bub1 and expression levels of OsbZIP46 target genes in response to drought stress. These findings allow us to update our previous model, centered on OsbZIP46 and its interacting proteins in the regulation of ABA signaling and drought response, by adding the reversible H2Bub1 modification (Figure 7). In this updated model, the key transcription factor, OsbZIP46, fine-tunes ABA signaling and drought response by integrating at least two epigenetic regulation mechanisms: histone monoubiquitination and acetylation. At the early stages of drought stress, OsbZIP46 phosphorylation, activated by ABA signaling, and an increase in H2Bub1, mediated by OsbZIP46OsHUB2, jointly upregulate the expression of droughtresponsive genes that are mainly targeted by OsbZIP46. At the late drought stress or recovery stage, increased MODD and OsOTLD1 levels lead to downregulation of the genes through deacetylation and H2B deubiquitination, mediated by the MODD–OsTPR3–HDA702 complex and the MODD–OsOTLD1 complex, respectively. It should be noted that the stability of OsbZIP46 can also be downregulated via degradation mediated by MODD–OsPUB70 (Tang et al., 2016), which is not included in the updated model. The joint regulation by histone deacetylation and H2Bub1 in stress responses may be conserved in plants. In Arabidopsis, AFP (a homolog of MODD) interacts with the co-repressors TOPLESS-HDAC, similar to OsTPR3-HDA702 in rice, and inhibits ABA-regulated gene expression, and AFP can also modulate ABA response via TOPLESS-independent chromatin modification (Lynch et al., 2017). In view of the homologous relationship between AFP and MODD, H2Bub1 is likely one of the TOPLESS-independent chromatin modifications in stress responses. A previous study showed that the OTLD1 in Arabidopsis most likely functions as a part of a repressor complex, and the complex may contain some common chromatin modification members, such as histone deacetylases, histone methyltransferases, and histone lysine demethylases, based on the effects of OTLD1 on its target chromatin (Keren and Citovsky, 2016). This knowledge further supports our model on the integration of histone deacetylation and H2Bub1 in the regulation of ABA and 12

Molecular Plant --, 1–15, -- 2019 ª The Author 2018.

METHODS Plant Materials and Growth Conditions The full-length coding regions of OsHUB1 and OsHUB2 were cloned into the pCAMBIA1301U (driven by a maize ubiquitin promoter) vector using the KpnI and BamHI sites, and introduced into Zhonghua 11 (Oryza sativa ssp. japonica) by Agrobacterium tumefaciens-mediated transformation. T-DNA insertion mutants oshub2-1 (03Z11AY86-3, in Zhonghua 11 background), oshub2-2 (PFG_2A-00426.R, in Dongjin background), osbzip46 (PFG_1B-09829.L, in Dongjin background), and osotld1 (PFG_1C11130.R, in Hwayoung background) were obtained from the RMD or POSTECH mutant database (Jeon et al., 2000; Jeong et al., 2006; Zhang et al., 2006). The plants were grown in the field from April to October under natural conditions in Wuhan, China, or in the greenhouse with a 12-h light/12-h dark cycle at 25 C–30 C in winter.

Yeast Two-Hybrid Assay Y2H assays were performed using the ProQuest two-hybrid system (Invitrogen) as described previously (Tang et al., 2016). For the Y2H screening, the full-length coding region of OsbZIP46 was cloned into pDEST32 to generate a bait construct, which was further co-transformed into the yeast strain Mav203 with a prey cDNA library of rice. For the other Y2H assays, the full-length coding region of OsHUB2 was fused to the GAL4 activation domain by constructing it in the pDEST22 vector, while OsbZIP46 and MODD were cloned into the pDEST32 vector. For the interaction test, pDEST32 and pDEST22 constructs were co-transformed into the Mav203 yeast strain, and selected on the SD/-Leu-Trp-His medium. The b-galactosidase activity assay was further conducted according to the manufacturer’s manual.

Rice Protoplast Preparation and Transformation Rice protoplasts were prepared from 13-day-old seedlings of rice Zhonghua 11 as described previously (Xie and Yang, 2013; Shen et al., 2017). The protoplasts were isolated by digesting rice sheath strips in Digestion Solution for 4 h. The protoplasts were collected and incubated in W5 solution at room temperature for 60–90 min. Finally, the protoplasts were collected by centrifugation at 100 g for 8 min and resuspended in MMG solution to a final concentration of 0.5–1.0 3 107 ml1. For transformation, 5–20 mg of plasmids was gently mixed with 100 ml of protoplasts and 110 ml of polyethylene glycol (PEG)–CaCl2 solution (0.6 M mannitol, 100 mM CaCl2, and 40% PEG4000), and incubated at room temperature for 15 min. Two volumes of W5 solution were then added to stop the transformation, and the protoplasts were again collected by centrifugation followed by resuspension in WI solution and incubation in 24-well culture plates for 20 h in the dark. The protoplasts were then collected by centrifugation at 200 g for 5 min for the next assay.

BiFC Assay For the BiFC assays, the coding regions of OsbZIP46 and MODD were cloned into pVYCE(R) and fused with the C terminus of YFP, and OsHUB2 was cloned into pVYNE(R) and fused with the N terminus of YFP (Waadt et al., 2008). The recombinant constructs were co-transformed in pairs into rice protoplasts, and Ghd7-CFP as a marker for nuclear localization (Tang et al., 2016) was also co-transformed. The fluorescence signal was observed using an Olympus FV1200 confocal microscope.

Please cite this article in press as: Ma et al., Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice, Molecular Plant (2018), https://doi.org/10.1016/j.molp.2018.12.005

Molecular Plant

Histone Monoubiquitination in Rice Drought Response CoIP Assay The total proteins were extracted from the rice protoplasts, in which two proteins with HA and Myc tags, respectively, were co-expressed. Total proteins were then incubated with anti-HA agarose beads (Roche) in coIP buffer (50 mM Tris–HCl [pH 8.0], 150 mM KCl, 1 mM EDTA, 0.5% Triton X-100, 1 mM DTT, 1 mM PMSF, and protease inhibitor cocktail tablets [Roche]). The coIP samples were detected with the anti-Myc and antiHA antibodies (ABclonal).

Stress Treatments The ABA sensitivity and drought resistance of the materials were tested as described previously (Tang et al., 2016). In brief, the seeds were germinated on half-strength Murashige and Skoog (1/2 MS) medium for 3 days and then transplanted to 1/2 MS medium containing 0 mM or 3 mM ABA. After 7–10 days, the shoot and root length were measured and analyzed. For the drought resistance testing, half transgenic and half control plants (10–12 plants each line and three repeats) were grown in the pots. At the 4-leaf stage, watering was stopped for 5–7 days until the leaves completely rolled, after which the plants were recovered for 5–7 days and the survival rates were calculated. To define the moderate and severe drought stresses treatment, we subjected rice plants at 4-leaf stage growing in the paddy soil from natural rice field to progressive (slowly developed) drought stress with the soil water content monitored. This was regarded as moderate drought stress when the relative soil water content was about 20% and severe drought stress when the relative soil water content was about 10%. The relative soil water content was calculated as follows: 100% 3 (the weight of soil after stress – the weight of air-dried soil)/the weight of soil after stress (Gonzalez et al., 2012).

ChIP–qPCR and ChIP-Seq The ChIP assays were conducted as described previously (Huang et al., 2007). In brief, 2 g of mutant and wild-type seedlings (each with three biological replicates) were harvested and crosslinked in 1% formaldehyde under vacuum, and chromatin was extracted and fragmented by sonication. For the immunoprecipitations, specific antibodies (anti-H2Bub1, Millipore, 05-1312; anti-OsbZIP46, produced by ABclonal company in China), together with Dynabeads protein A or G (Invitrogen, 10002D/10004D), were added to the sonicated chromatin followed by incubation overnight. The precipitated and input DNAs were further analyzed using qPCR with specific primers (Supplemental Data 6) or sequencing. OsACTIN (Os11g06390) was used as a negative control locus in the ChIP–qPCR assays. For ChIP-seq, chromatin-immunoprecipitated DNAs were used to construct sequencing libraries following the protocol of the TruSeq ChIP Library Preparation Kit (Illumina), and sequenced by Novogene Bioinformatics Technology (China) with Hiseq-PE150 (Illumina). Raw sequencing reads were filtered by FASTP and mapped to the rice genome (MSU RGAP, release 7) using Bowtie 2 (version 2.3.4) with default parameters, and only uniquely mapped reads were used for peak calling. Peaks that were significantly enriched in ChIP-seq tags were identified by Modelbased Analysis of ChIP-Seq (MACS, version 2.1.1) with default parameters, and visualized using the Integrated Genomics Viewer (v.2.3.26). Genes (including gene body and the 3-kb upstream and 3-kb downstream regions) overlapping with the peaks were considered to have the H2Bub1 occupancy, and the differential modification of H2Bub1 was determined with a standard of fold change R2 and P value <0.05, based on analyses with the R package DiffBind (http://bioconductor.org). GO analysis was performed on an online platform named BMKClould (http://www. biocloud.net), and the significance threshold was set at a P value of 0.005.

RNA-Seq Analyses Total RNA samples were extracted from wild-type rice Zhonghua 11 at the 4-leaf stage under normal growth conditions, and moderate and severe drought stress treatments (each with three biological replicates) using TRIzol reagent (Invitrogen) according to the manufacturer’s instructions.

For each sample, mRNA purified from 10 mg of total RNA was used for constructing sequencing libraries with the TruSeq RNA Sample Preparation Kit (Illumina). The libraries were sequenced by Novogene Bioinformatics Technology (China) with Hiseq-PE150 (Illumina). Raw sequencing reads were filtered by FASTP and mapped to the rice genome (MSU RGAP, release 7) using HISAT2 (version 2.0.5) with default parameters. Differentially expressed genes, which were defined with an expression change fold R2 with P < 0.05, were analyzed using DESeq2 (version 1.20.0). The comparative analyses between RNA-seq and ChIP-seq data were conducted using Excel.

Transcriptional Regulation Activity Assay in Protoplast Transcriptional regulation activity assays were performed as previously described (Tang et al., 2016). For determination of their regulatory effects on the expression of OsbZIP46 target genes, OsbZIP46, OsHUB2, MODD, and OsOTLD1 were cloned into effector vectors to generate constructs driven by the CaMV 35S promoter. The effector vectors, together with the construct containing the Renilla luciferase (R-LUC) gene as an internal control, were co-transformed into rice protoplasts followed by culturing for more than 12 h. The protoplasts were collected to isolate total RNA. Total cDNA was obtained using the EasyScript One-Step gDNA Removal and cDNA Synthesis SuperMix (Transgen Biotech) kits. The expression levels of RAB21 and R-LUC were determined by qPCR (Hao et al., 2011).

In Vitro Histone Deubiquitination Assay Purified GST-OsOTLD1 protein was incubated with 20 mg of total bovine histones (Roche) in the buffer with 50 mM Tris–HCl (pH 8.0) and 50 mM NaCl for 3–4 h at room temperature as previously described (Krichevsky et al., 2011) with minor modifications. For western blot, the samples were resolved in 10% SDS–PAGE, blocked with 1% BSA and probed with the anti-ubiquitin H2B antibody (Medimabs, 1:500 dilution) and the anti-H2B antibody (Abcam, 1:5000 dilution).

Micro-ChIP Assay The rapid mChIP assay was performed as described previously (Dahl and Collas, 2009). OsOTLD1-pRGE or pRGE empty vector were transformed into rice protoplasts. The protoplasts were cultured for 12 h and then harvested and resuspended in PBS buffer at room temperature. The DNA and protein were crosslinked in 1% formaldehyde and stopped by 125 mM glycine. The formaldehyde crosslinked cells were centrifuged and washed three times in PBS buffer. The lysis buffer was then added and the cells were sonicated on ice for 3 3 30 s. The products were centrifuged at 12 000 g for 10 min at 4 C, then the supernatant was transferred into a clean 1.5-ml tube chilled on ice. Chromatin extracts were mixed with anti-H2Bub1 beads, incubated in a rotator at 40 rpm for 2 h at 4 C, and washed with ice-cold RIPA buffer (10 mM Tris–HCl [pH 7.5], 140 mM NaCl, 1 mM EDTA, 0.5 mM EGTA, 1% Triton X-100, 0.1% SDS) and TE buffer (10 mM Tris–HCl [pH 8.0], 10 mM EDTA). The input and precipitated DNAs were further analyzed using qPCR with specific primers (Supplemental Data 6) for OsbZIP46 and its target genes.

Data Availability Sequence data from this article can be found on the Rice Genome Annotation Project website (http://rice.plantbiology.msu.edu/) under the following accession numbers: OsHUB1, Os04g46450; OsHUB2, Os10g41590; OsOTLD1, Os04g33780; OsbZIP46, Os06g10880; MODD, Os03g11550; LEA3, Os05g46480; LEA14, Os01g50910; RAB16b, Os11g26780; RAB21, Os11g26790; PP2C51, Os05g49730; and OsACTIN, Os11g06390. The raw and processed ChIP-seq and RNA-seq data have been deposited in the Gene Expression Omnibus database under the series accession number GEO: GSE117047.

SUPPLEMENTAL INFORMATION Supplemental Information is available at Molecular Plant Online.

Molecular Plant --, 1–15, -- 2019 ª The Author 2018.

13

Please cite this article in press as: Ma et al., Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice, Molecular Plant (2018), https://doi.org/10.1016/j.molp.2018.12.005

Molecular Plant

Histone Monoubiquitination in Rice Drought Response

FUNDING

Fleury, D., Himanen, K., Cnops, G., Nelissen, H., Boccardi, T.M., Maere, S., Beemster, G.T., Neyt, P., Anami, S., Robles, P., et al. (2007). The Arabidopsis thaliana homolog of yeast BRE1 has a function in cell cycle regulation during early leaf and root growth. Plant Cell 19:417–432.

This work was supported by grants from the National Natural Science Foundation of China (31771359, 31821005), the National Key Research and Development Program of China (2016YFD0100600), and the Fundamental Research Funds for the Central Universities.

AUTHOR CONTRIBUTIONS S.M., N.T., and L.X. conceived and designed the research. S.M., N.T., Xu Li, Y.X., D.X., J.F., J.S., J.Y., H.T., and Xianghua Li performed the experiments. S.M., N.T., and L.X. analyzed the data. S.M., N.T., and L.X. wrote the article with input from H.H.

ACKNOWLEDGMENTS We are grateful to Dr. Shouyi Chen (Institute of Genetics and Developmental Biology, Chinese Academy of Sciences) for the gift of vectors for the transcriptional regulation activity assay in protoplasts. We are grateful to Dr. Yongxiu Liu (Institute of Botany, Chinese Academy of Sciences) for the gift of oshub2-2 mutant. No conflict of interest declared. Received: July 3, 2018 Revised: November 9, 2018 Accepted: December 9, 2018 Published: December 18, 2018

REFERENCES Banerjee, A., and Roychoudhury, A. (2017). Abscisic-acid-dependent basic leucine zipper (bZIP) transcription factors in plant abiotic stress. Protoplasma 254:3–16. Bourbousse, C., Ahmed, I., Roudier, F., Zabulon, G., Blondet, E., Balzergue, S., Colot, V., Bowler, C., and Barneche, F. (2012). Histone H2B monoubiquitination facilitates the rapid modulation of gene expression during Arabidopsis photomorphogenesis. PLoS Genet. 8:e1002825. Cao, H., Li, X., Wang, Z., Ding, M., Sun, Y., Dong, F., Chen, F., Liu, L., Doughty, J., Li, Y., et al. (2015). Histone H2B monoubiquitination mediated by HISTONE MONOUBIQUITINATION1 and HISTONE MONOUBIQUITINATION2 is involved in anther development by regulating tapetum degradation-related genes in rice. Plant Physiol. 168:1389–1405. Chen, H., Feng, H., Zhang, X., Zhang, C., Wang, T., and Dong, J. (2018). An Arabidopsis E3 ligase HUB2 increases histone H2B monoubiquitination and enhances drought tolerance in transgenic cotton. Plant Biotechnol. J. https://doi.org/10.1111/pbi.12998. Chinnusamy, V., Gong, Z., and Zhu, J.K. (2008). Abscisic acid-mediated epigenetic processes in plant development and stress responses. J. Integr. Plant Biol. 50:1187–1195. Cutler, S.R., Rodriguez, P.L., Finkelstein, R.R., and Abrams, S.R. (2010). Abscisic acid: emergence of a core signaling network. Annu. Rev. Plant Biol. 61:651–679. Dahl, J.A., and Collas, P. (2009). MicroChIP: chromatin immunoprecipitation for small cell numbers. Methods Mol. Biol. 567: 59–74. DeLong, A. (2006). Switching the flip: protein phosphatase roles in signaling pathways. Curr. Opin. Plant Biol. 9:470–477. Dhawan, R., Luo, H., Foerster, A.M., Abuqamar, S., Du, H.N., Briggs, S.D., Mittelsten Scheid, O., and Mengiste, T. (2009). HISTONE MONOUBIQUITINATION1 interacts with a subunit of the mediator complex and regulates defense against necrotrophic fungal pathogens in Arabidopsis. Plant Cell 21:1000–1019. Du, Y., He, W., Deng, C., Chen, X., Gou, L., Zhu, F., Guo, W., Zhang, J., and Wang, T. (2016). Flowering-related RING protein 1 (FRRP1) regulates flowering time and yield potential by affecting histone H2B monoubiquitination in rice (Oryza Sativa). PLoS One 11:e0150458.

14

Molecular Plant --, 1–15, -- 2019 ª The Author 2018.

Garg, R., Tyagi, A.K., and Jain, M. (2012). Microarray analysis reveals overlapping and specific transcriptional responses to different plant hormones in rice. Plant Signal. Behav. 7:951–956. Gonzalez, C.V., Ibarra, S.E., Piccoli, P.N., Botto, J.F., and Boccalandro, H.E. (2012). Phytochrome B increases drought tolerance by enhancing ABA sensitivity in Arabidopsis thaliana. Plant Cell Environ. 35:1958–1968. Hao, Y.J., Wei, W., Song, Q.X., Chen, H.W., Zhang, Y.Q., Wang, F., Zou, H.F., Lei, G., Tian, A.G., Zhang, W.K., et al. (2011). Soybean NAC transcription factors promote abiotic stress tolerance and lateral root formation in transgenic plants. Plant J. 68:302–313. Himanen, K., Woloszynska, M., Boccardi, T.M., De Groeve, S., Nelissen, H., Bruno, L., Vuylsteke, M., and Van Lijsebettens, M. (2012). Histone H2B monoubiquitination is required to reach maximal transcript levels of circadian clock genes in Arabidopsis. Plant J. 72:249–260. Hu, H., and Xiong, L. (2014). Genetic engineering and breeding of drought-resistant crops. Annu. Rev. Plant Biol. 65:715–741. Huang, L., Sun, Q., Qin, F., Li, C., Zhao, Y., and Zhou, D.X. (2007). Down-regulation of a SILENT INFORMATION REGULATOR2-related histone deacetylase gene, OsSRT1, induces DNA fragmentation and cell death in rice. Plant Physiol. 144:1508–1519. Huang, O.W., and Cochran, A.G. (2013). Regulation of deubiquitinase proteolytic activity. Curr. Opin. Struct. Biol. 23:806–811. Hwang, W.W., Venkatasubrahmanyam, S., Ianculescu, A.G., Tong, A., Boone, C., and Madhani, H.D. (2003). A conserved RING finger protein required for histone H2B monoubiquitination and cell size control. Mol. Cell 11:261–266. Isono, E., and Nagel, M.K. (2014). Deubiquitylating enzymes and their emerging role in plant biology. Front. Plant Sci. 5:56. Jeon, J.S., Lee, S., Jung, K.H., Jun, S.H., Jeong, D.H., Lee, J., Kim, C., Jang, S., Yang, K., Nam, J., et al. (2000). T-DNA insertional mutagenesis for functional genomics in rice. Plant J. 22:561–570. Jeong, D.H., An, S., Park, S., Kang, H.G., Park, G.G., Kim, S.R., Sim, J., Kim, Y.O., Kim, M.K., Kim, S.R., et al. (2006). Generation of a flanking sequence-tag database for activation-tagging lines in japonica rice. Plant J. 45:123–132. Keren, I., and Citovsky, V. (2016). The histone deubiquitinase OTLD1 targets euchromatin to regulate plant growth. Sci. Signal. 9:ra125. Kim, J.M., Sasaki, T., Ueda, M., Sako, K., and Seki, M. (2015). Chromatin changes in response to drought, salinity, heat, and cold stresses in plants. Front. Plant Sci. 6:114. Komander, D., Clague, M.J., and Urbe, S. (2009). Breaking the chains: structure and function of the deubiquitinases. Nat. Rev. Mol. Cell Biol. 10:550–563. Kouzarides, T. (2007). Chromatin modifications and their function. Cell 128:693–705. Krichevsky, A., Zaltsman, A., Lacroix, B., and Citovsky, V. (2011). Involvement of KDM1C histone demethylase-OTLD1 otubain-like histone deubiquitinase complexes in plant gene repression. Proc. Natl. Acad. Sci. U S A 108:11157–11162. Liu, Y., Koornneef, M., and Soppe, W.J. (2007). The absence of histone H2B monoubiquitination in the Arabidopsis hub1 (rdo4) mutant reveals a role for chromatin remodeling in seed dormancy. Plant Cell 19:433–444.

Please cite this article in press as: Ma et al., Reversible Histone H2B Monoubiquitination Fine-Tunes Abscisic Acid Signaling and Drought Response in Rice, Molecular Plant (2018), https://doi.org/10.1016/j.molp.2018.12.005

Histone Monoubiquitination in Rice Drought Response

Molecular Plant

Luo, M., Luo, M.Z., Buzas, D., Finnegan, J., Helliwell, C., Dennis, E.S., Peacock, W.J., and Chaudhury, A. (2008). UBIQUITIN-SPECIFIC PROTEASE 26 is required for seed development and the repression of PHERES1 in Arabidopsis. Genetics 180:229–236.

Tang, N., Zhang, H., Li, X., Xiao, J., and Xiong, L. (2012). Constitutive activation of transcription factor OsbZIP46 improves drought tolerance in rice. Plant Physiol. 158:1755–1768.

Lynch, T., Erickson, B.J., and Finkelstein, R.R. (2012). Direct interactions of ABA-insensitive(ABI)-clade protein phosphatase(PP) 2Cs with calcium-dependent protein kinases and ABA response element-binding bZIPs may contribute to turning off ABA response. Plant Mol. Biol. 80:647–658. Lynch, T.J., Erickson, B.J., Miller, D.R., and Finkelstein, R.R. (2017). ABI5-binding proteins (AFPs) alter transcription of ABA-induced genes via a variety of interactions with chromatin modifiers. Plant Mol. Biol. 93:403–418. Ma, Y., Szostkiewicz, I., Korte, A., Moes, D., Yang, Y., Christmann, A., and Grill, E. (2009). Regulators of PP2C phosphatase activity function as abscisic acid sensors. Science 324:1064–1068. Menard, R., Verdier, G., Ors, M., Erhardt, M., Beisson, F., and Shen, W.H. (2014). Histone H2B monoubiquitination is involved in the regulation of cutin and wax composition in Arabidopsis thaliana. Plant Cell Physiol. 55:455–466. Nakashima, K., Fujita, Y., Kanamori, N., Katagiri, T., Umezawa, T., Kidokoro, S., Maruyama, K., Yoshida, T., Ishiyama, K., Kobayashi, M., et al. (2009). Three Arabidopsis SnRK2 protein kinases, SRK2D/SnRK2.2, SRK2E/SnRK2.6/OST1 and SRK2I/ SnRK2.3, involved in ABA signaling are essential for the control of seed development and dormancy. Plant Cell Physiol. 50:1345–1363. Nassrallah, A., Rougee, M., Bourbousse, C., Drevensek, S., Fonseca, S., Iniesto, E., Ait-Mohamed, O., Deton-Cabanillas, A.F., Zabulon, G., Ahmed, I., et al. (2018). DET1-mediated degradation of a SAGAlike deubiquitination module controls H2Bub homeostasis. Elife 7. https://doi.org/10.7554/eLife.37892. Park, S.Y., Fung, P., Nishimura, N., Jensen, D.R., Fujii, H., Zhao, Y., Lumba, S., Santiago, J., Rodrigues, A., Chow, T.F., et al. (2009). Abscisic acid inhibits type 2C protein phosphatases via the PYR/PYL family of START proteins. Science 324:1068–1071. Ramanathan, H.N., and Ye, Y. (2012). Cellular strategies for making monoubiquitin signals. Crit. Rev. Biochem. Mol. Biol. 47:17–28.

Turner, B.M. (2002). Cellular memory and the histone code. Cell 111:285–291. Umezawa, T., Sugiyama, N., Mizoguchi, M., Hayashi, S., Myouga, F., Yamaguchi-Shinozaki, K., Ishihama, Y., Hirayama, T., and Shinozaki, K. (2009). Type 2C protein phosphatases directly regulate abscisic acid-activated protein kinases in Arabidopsis. Proc. Natl. Acad. Sci. U S A 106:17588–17593. Vishwakarma, K., Upadhyay, N., Kumar, N., Yadav, G., Singh, J., Mishra, R.K., Kumar, V., Verma, R., Upadhyay, R.G., Pandey, M., et al. (2017). Abscisic acid signaling and abiotic stress tolerance in plants: a review on current knowledge and future prospects. Front. Plant Sci. 8:161. Waadt, R., Schmidt, L.K., Lohse, M., Hashimoto, K., Bock, R., and Kudla, J. (2008). Multicolor bimolecular fluorescence complementation reveals simultaneous formation of alternative CBL/CIPK complexes in planta. Plant J. 56:505–516. Weake, V.M., and Workman, J.L. (2008). Histone ubiquitination: triggering gene activity. Mol. Cell 29:653–663. Wu, X., Gong, F., Cao, D., Hu, X., and Wang, W. (2016). Advances in crop proteomics: PTMs of proteins under abiotic stress. Proteomics 16:847–865. Xiang, Y., Tang, N., Du, H., Ye, H., and Xiong, L. (2008). Characterization of OsbZIP23 as a key player of the basic leucine zipper transcription factor family for conferring abscisic acid sensitivity and salinity and drought tolerance in rice. Plant Physiol. 148:1938–1952. Xie, K., and Yang, Y. (2013). RNA-guided genome editing in plants using a CRISPR-Cas system. Mol. Plant 6:1975–1983. Xiong, L., Schumaker, K.S., and Zhu, J.K. (2002). Cell signaling during cold, drought, and salt stress. Plant Cell 14:S165–S183. Zhang, J., Li, C., Wu, C., Xiong, L., Chen, G., Zhang, Q., and Wang, S. (2006). RMD: a rice mutant database for functional analysis of the rice genome. Nucleic Acids Res. 34:D745–D748.

Rodrigues, A., Adamo, M., Crozet, P., Margalha, L., Confraria, A., Martinho, C., Elias, A., Rabissi, A., Lumbreras, V., GonzalezGuzman, M., et al. (2013). ABI1 and PP2CA phosphatases are negative regulators of Snf1-related protein kinase1 signaling in Arabidopsis. Plant Cell 25:3871–3884.

Zhang, J., Li, X., He, Z., Zhao, X., Wang, Q., Zhou, B., Yu, D., Huang, X., Tang, D., Guo, X., et al. (2013). Molecular character of a phosphatase 2C (PP2C) gene relation to stress tolerance in Arabidopsis thaliana. Mol. Biol. Rep. 40:2633–2644.

Roudier, F., Ahmed, I., Berard, C., Sarazin, A., Mary-Huard, T., Cortijo, S., Bouyer, D., Caillieux, E., Duvernois-Berthet, E., Al-Shikhley, L., et al. (2011). Integrative epigenomic mapping defines four main chromatin states in Arabidopsis. EMBO J. 30:1928–1938.

Zhang, Y., Li, D., Zhang, H., Hong, Y., Huang, L., Liu, S., Li, X., Ouyang, Z., and Song, F. (2015). Tomato histone H2B monoubiquitination enzymes SlHUB1 and SlHUB2 contribute to disease resistance against Botrytis cinerea through modulating the balance between SA- and JA/ET-mediated signaling pathways. BMC Plant Biol. 15:252.

Schmitz, R.J., Tamada, Y., Doyle, M.R., Zhang, X., and Amasino, R.M. (2009). Histone H2B deubiquitination is required for transcriptional activation of FLOWERING LOCUS C and for proper control of flowering in Arabidopsis. Plant Physiol. 149:1196–1204.

Zhu, B., Zheng, Y., Pham, A.D., Mandal, S.S., Erdjument-Bromage, H., Tempst, P., and Reinberg, D. (2005). Monoubiquitination of human histone H2B: the factors involved and their roles in HOX gene regulation. Mol. Cell 20:601–611.

Schweighofer, A., Hirt, H., and Meskiene, I. (2004). Plant PP2C phosphatases: emerging functions in stress signaling. Trends Plant Sci. 9:236–243.

Zhu, J.K. (2016). Abiotic stress signaling and responses in plants. Cell 167:313–324.

Shen, J., Liu, J., Xie, K., Xing, F., Xiong, F., Xiao, J., Li, X., and Xiong, L. (2017). Translational repression by a miniature inverted-repeat transposable element in the 3’ untranslated region. Nat. Commun. 8:14651.

Zong, W., Tang, N., Yang, J., Peng, L., Ma, S., Xu, Y., Li, G., and Xiong, L. (2016). Feedback regulation of ABA signaling and biosynthesis by a bZIP transcription factor targets drought-resistance-related genes. Plant Physiol. 171:2810–2825.

Tang, N., Ma, S., Zong, W., Yang, N., Lv, Y., Yan, C., Guo, Z., Li, J., Li, X., Xiang, Y., et al. (2016). MODD mediates deactivation and degradation of OsbZIP46 to negatively regulate ABA signaling and drought resistance in rice. Plant Cell 28:2161–2177.

Zou, B., Yang, D.L., Shi, Z., Dong, H., and Hua, J. (2014). Monoubiquitination of histone 2B at the disease resistance gene locus regulates its expression and impacts immune responses in Arabidopsis. Plant Physiol. 165:309–318.

Molecular Plant --, 1–15, -- 2019 ª The Author 2018.

15