AtEAU1 and AtEAU2, Two EAR Motif-Containing ABA Up-Regulated Novel Transcription Repressors Regulate ABA Response in Arabidopsis

EAR (Ethylene-responsive element binding factor-associated Amphiphilic Repression) motif-containing transcription repressors have been shown to regulate plant growth and development, and plant responses to plant hormones and environmental stresses including biotic and abiotic stresses. However, the functions of most EAR-motif-containing proteins remain largely uncharacterized. The plant hormone abscisic acid (ABA) also plays important roles in regulating plant responses to abiotic stresses via activation/repression of ABA-responsive genes. We report here the identification and functional characterization of two ABA-responsive EAR motif-containing protein genes, AtEAU1 (Arabidopsis thaliana EAR motif-containing ABA Up-regulated 1) and AtEAU2. Quantitative RT-PCR results show that the expressions of AtEAU1 and AtEAU2 were increased by ABA treatment, and were decreased in the ABA biosynthesis mutant aba1-5. Assays in transfected Arabidopsis protoplasts show that both AtEAU1 and AtEAU2 were specifically localized in the nucleus, and when recruited to the promoter region of the reporter gene by a fused DNA binding domain, repressed reporter gene expression. By using T-DNA insertion mutants and a gene-edited transgene-free mutant generated by CRISPR/Cas9 gene editing, we performed ABA sensitivity assays, and found that ABA sensitivity in the both ateau1 and ateau2 single mutants was increased in seedling greening assays. ABA sensitivity in the ateau1 ateau2 double mutants was also increased, but was largely similar to the ateau1 single mutants. On the other hand, all the mutants showed a wild type response to ABA in root elongation assays. Quantitative RT-PCR results show that the expression level of PYL4, an ABA receptor gene was increased, whereas that of ABI2, a PP2C gene was decreased in the ateau1 and ateau1 single, and the ateau1 ateau2 double mutants. In summary, our results suggest that AtEAU1 and AtEAU2 are ABA-response genes, and AtEAU1 and AtEAU2 are novel EAR motif-containing transcription repressors that negatively regulate ABA responses in Arabidopsis, likely by regulating the expression of some ABA signaling key regulator genes.

Here, by using an opposite strategy, i.e., identifying ABA response genes from unknown function EAR motif-containing protein genes [31], and searching for protein homologs, we identified AtEAU1 (Arabidopsis thaliana EAR motif-containing ABA Upregulated 1) and AtEAU2 as novel regulators of ABA response in Arabidopsis.

AtEAU1 and AtEAU2 Are Closely Related EAR Motif-Containing Proteins
A protein encoded by gene At1g78170 is among the identified EAR motif-containing proteins [31]. As a protein encoded by gene At1g22250 is closely related to the protein encoded by gene At1g78170, and the expression of these two genes is induced by ABA treatment (see next section for details), we named them AtEAU1 (Arabidopsis thaliana EAR motif-containing ABA Up-regulated 1) and AtEAU2, respectively.
Protein homolog identification on Phytozome (https://phytozome-next.jgi.doe.gov, accessed on 1 September 2016) show that the next closely related protein to AtEAU1 and AtEAU2 is the protein encoded by gene At4g08910, which was also identified as an EAR motif-containing protein [31]. However, phylogenetic analysis shows that the most closely related protein identified from other plant species including soybean, tomato, rice and poplar are more closely related to At4g08910, and they formed a clade, whereas AtEAU1 and AtEAU2 formed another clade ( Figure 1A). some of the AITRs have a fully conserved LxLxL EAR motif [7], but none of them are identified in the genome-wide search of EAR motif-containing proteins [31]. Therefore we first tried to identify novel EAR motif-containing proteins that may involve in the regulation of ABA response by searching EAR motifs in proteins encoded by unknown function ABA-responsive genes from available transcriptome datasets, and we successfully identified SlEAD1 (Solanum lycopersicum EAR motif-containing ABA down-regulated 1) as a negative regulator of ABA response in tomato [33].
Here, by using an opposite strategy, i.e., identifying ABA response genes from unknown function EAR motif-containing protein genes [31], and searching for protein homologs, we identified AtEAU1 (Arabidopsis thaliana EAR motif-containing ABA Up-regulated 1) and AtEAU2 as novel regulators of ABA response in Arabidopsis.

AtEAU1 and AtEAU2 Are Closely Related EAR Motif-Containing Proteins
A protein encoded by gene At1g78170 is among the identified EAR motif-containing proteins [31]. As a protein encoded by gene At1g22250 is closely related to the protein encoded by gene At1g78170, and the expression of these two genes is induced by ABA treatment (see next section for details), we named them AtEAU1 (Arabidopsis thaliana EAR motif-containing ABA Up-regulated 1) and AtEAU2, respectively.
Protein homolog identification on Phytozome (https://phytozome-next.jgi.doe.gov, accessed on 1 September 2016) show that the next closely related protein to AtEAU1 and AtEAU2 is the protein encoded by gene At4g08910, which was also identified as an EAR motif-containing protein [31]. However, phylogenetic analysis shows that the most closely related protein identified from other plant species including soybean, tomato, rice and poplar are more closely related to At4g08910, and they formed a clade, whereas AtEAU1 and AtEAU2 formed another clade ( Figure 1A).  The full-length amino acid sequences of the AtEAU1 and AtEAU2 and their closely related proteins from Arabidopsis and other several different plants were obtained from phytozome (https://phytozome-next.jgi.doe.gov, accessed on 1 September 2016), and subjected to phylogenetic analysis by using the "One Click" mode with default settings on phylogeny (http: //www.phylogeny.fr/simple_phylogeny.cgi, accessed on 1 September 2016). Numbers above the branches indicate the branch support values. (B) Amino acid identity and similarity of AtEAU1, AtEAU2 and At4g08910. The full-length amino acid sequences of AtEAU1, AtEAU2, and At4g08910 were used for SIAS (http://imed.med.ucm.es/Tools/sias.html, accessed on 1 September 2016) assay. The amino acid identity percentage was shaded in light blue, and the similarity percentage in blue. (C) Sequence alignment of AtEAUs and At4g08910. Black shades indicate the identical amino acids and gray shades indicate the similar amino acids. Underline indicates the LxLxL motif, one of the two known EAR motifs. Consistent with the phylogenetic tree, amino acid identity and similarity assays show that AtEAU1 and AtEAU2 shared~57% amino acid identity and~62% similarity, respectively, but sharing between At4g08910 and AtEAU1 or AtEAU2 was only about 30% and 40%, respectively ( Figure 1B). On the other hand, even though At4g08910 shared relatively low amino acid identity and similarity with AtEAU1 or AtEAU2, it has a fully conserved LxLxL EAR motif presented in AtEAU1 or AtEAU2 ( Figure 1C).

The Expression of AtEAU1 and AtEAU2
Are Up-Regulated by ABA and They Share Similar Expression Patern Since AtEAU1 and AtEAU2 are closely related, but showed low amino acid identity and similarity with At4g08910, and are in a different clade than At4g08910 in the phylogenetic analysis (Figure 1), we characterized only AtEAU1 and AtEAU2 in this study.
Expression of AtEAU1 and AtEAU2 in response to exogenous ABA treatment was examined using the Col wild-type Arabidopsis seedlings. Seedlings of the Col wild type were treated with ABA, and RNA isolated was used for qRT-PCR analysis. As shown in Figure 2A, the expression levels of AtEAU1 and AtEAU2 increased about 2.5 and 1.5 fold, respectively, in seedlings treated with ABA compared with mock-treated seedlings. We also examined the expression of AtEAU1 and AtEAU2 in seedlings of the aba1-5, an ABA biosynthesis mutant in the Ler wild type background [34]. We found that the expression level of AtEAU1 and AtEAU2 decreased to about 1/3 and 1/2 of that in the Ler wild type seedlings, respectively ( Figure 2B). These results suggest that AtEAU1 and AtEAU2 are ABA response genes.
At4g08910 were used for SIAS (http://imed.med.ucm.es/Tools/sias.html, accessed on 1 September 2016) assay. The amino acid identity percentage was shaded in light blue, and the similarity percentage in blue. (C) Sequence alignment of AtEAUs and At4g08910. Black shades indicate the identical amino acids and gray shades indicate the similar amino acids. Underline indicates the LxLxL motif, one of the two known EAR motifs. Consistent with the phylogenetic tree, amino acid identity and similarity assays show that AtEAU1 and AtEAU2 shared ~57% amino acid identity and ~62% similarity, respectively, but sharing between At4g08910 and AtEAU1 or AtEAU2 was only about 30% and 40%, respectively ( Figure 1B). On the other hand, even though At4g08910 shared relatively low amino acid identity and similarity with AtEAU1 or AtEAU2, it has a fully conserved LxLxL EAR motif presented in AtEAU1 or AtEAU2 ( Figure 1C).

The Expression of AtEAU1 and AtEAU2 Are Up-Regulated by ABA and They Share Similar Expression Patern
Since AtEAU1 and AtEAU2 are closely related, but showed low amino acid identity and similarity with At4g08910, and are in a different clade than At4g08910 in the phylogenetic analysis (Figure 1), we characterized only AtEAU1 and AtEAU2 in this study.
Expression of AtEAU1 and AtEAU2 in response to exogenous ABA treatment was examined using the Col wild-type Arabidopsis seedlings. Seedlings of the Col wild type were treated with ABA, and RNA isolated was used for qRT-PCR analysis. As shown in Figure 2A, the expression levels of AtEAU1 and AtEAU2 increased about 2.5 and 1.5 fold, respectively, in seedlings treated with ABA compared with mock-treated seedlings. We also examined the expression of AtEAU1 and AtEAU2 in seedlings of the aba1-5, an ABA biosynthesis mutant in the Ler wild type background [34]. We found that the expression level of AtEAU1 and AtEAU2 decreased to about 1/3 and 1/2 of that in the Ler wild type seedlings, respectively ( Figure 2B). These results suggest that AtEAU1 and AtEAU2 are ABA response genes. Twelve-day-old seedlings of the Col wild type Arabidopsis were treated with 50 μM ABA for 4 h, RNA was then isolated and subjected to quantitative RT-PCR (qRT-PCR) to examine the expression of AtEAU1 and AtEAU2. ACT2 was used as an inner reference gene for qRT-PCR, and the expression levels of AtEAU1 and AtEAU2 in mock-treated control seedlings were set as 1. Data represent the mean ± SD of three replicates. (B) Expression of AtEAU1 and AtEAU2 in the aba1-5 ABA biosynthesis mutant seedlings. RNA was isolated from 12-day-old seedlings of the Ler wild type and the aba1-5 mutant, and qRT-PCR was used to examine the expression of AtEAU1 and AtEAU2. ACT2 was used as an inner Twelve-day-old seedlings of the Col wild type Arabidopsis were treated with 50 µM ABA for 4 h, RNA was then isolated and subjected to quantitative RT-PCR (qRT-PCR) to examine the expression of AtEAU1 and AtEAU2. ACT2 was used as an inner reference gene for qRT-PCR, and the expression levels of AtEAU1 and AtEAU2 in mock-treated control seedlings were set as 1. Data represent the mean ± SD of three replicates. (B) Expression of AtEAU1 and AtEAU2 in the aba1-5 ABA biosynthesis mutant seedlings. RNA was isolated from 12-day-old seedlings of the Ler wild type and the aba1-5 mutant, and qRT-PCR was used to examine the expression of AtEAU1 and AtEAU2. ACT2 was used as an inner reference gene for qRT-PCR, and expression of AtEAU1 and AtEAU2 in the Ler wild-type seedlings was set as 1. Data represent the mean ± SD of three replicates. (C) Expression pattern of AtEAU1 and AtEAU2. Roots, radicle, rosette leaves, stems including 1st, 2nd, and 3rd stem from bottom to top of the inflorescence, cauline leaves, inflorescences, and siliques were collected from 5-week-old plants, RNA was then isolated and subjected to RT-PCR analysis to examine the expression of AtEAU1 and AtEAU2. ACT2 was used as a control for RT-PCR analysis.
We further examined the expression pattern of AtEAU1 and AtEAU2. Different tissues and organs were collected from adult Arabidopsis plants, and RNA was isolated and used for RT-PCR analysis. As shown in Figure 2C, AtEAU1 and AtEAU2 shared a largely similar expression pattern, i.e., both have relative higher express levels in inflorescence and siliques. However, the expression of AtEAU1 was also detectable in rosette leaves and stems, whereas AtEAU2 was not, possibly due to its relatively low expression level ( Figure 2C).

AtEAU1 and AtEAU2 Function as Transcription Repressors
As AtEAU1 has been identified as EAR motif-containing protein [31], and AtEAU2 also has a fully conserved LxLxL EAR motif, we thus examined if they may function as transcription repressors by using transfection assays in Arabidopsis protoplasts.
We first examined the protein subcellular localization of AtEAU1 and AtEAU2. Plasmids of the AtEAU1-GFP and AtEAU2-GFP constructs were transfected into Arabidopsis protoplasts isolated from leaves of 3-4 weeks old Col wild type plants, and GFP fluorescence in the protoplasts was observed under a confocal microscope. As shown in Figure 3A, GFP fluorescence for both AEAU1-GFP and AtEAU2-GFP was observed only in the nucleus, suggesting that AEAU1 and AtEAU2 are nucleus proteins.
lings was set as 1. Data represent the mean ± SD of three replicates. (C) Expression pattern o AtEAU1 and AtEAU2. Roots, radicle, rosette leaves, stems including 1st, 2nd, and 3rd stem from bottom to top of the inflorescence, cauline leaves, inflorescences, and siliques were collected from 5-week-old plants, RNA was then isolated and subjected to RT-PCR analysis to examine the ex pression of AtEAU1 and AtEAU2. ACT2 was used as a control for RT-PCR analysis.
We further examined the expression pattern of AtEAU1 and AtEAU2. Different tissues and organs were collected from adult Arabidopsis plants, and RNA was isolated and used for RT-PCR analysis. As shown in Figure 2C, AtEAU1 and AtEAU2 shared a largely similar expression pattern, i.e., both have relative higher express levels in inflorescence and siliques. However, the expression of AtEAU1 was also detectable in rosette leaves and stems whereas AtEAU2 was not, possibly due to its relatively low expression level ( Figure 2C).

AtEAU1 and AtEAU2 Function as Transcription Repressors
As AtEAU1 has been identified as EAR motif-containing protein [31], and AtEAU2 also has a fully conserved LxLxL EAR motif, we thus examined if they may function as transcription repressors by using transfection assays in Arabidopsis protoplasts.
We first examined the protein subcellular localization of AtEAU1 and AtEAU2 Plasmids of the AtEAU1-GFP and AtEAU2-GFP constructs were transfected into Arabidopsis protoplasts isolated from leaves of 3-4 weeks old Col wild type plants, and GFP fluorescence in the protoplasts was observed under a confocal microscope. As shown in Figure 3A, GFP fluorescence for both AEAU1-GFP and AtEAU2-GFP was observed only in the nucleus, suggesting that AEAU1 and AtEAU2 are nucleus proteins. . Plasmid DNA of the AtEAU1-GD and AtEAU2-GD was cotransfected with a plasmid of the LD-VP activator gene, and the LexA-Gal4:GUS reporter gene into protoplasts isolated from the Col wild-type Ara bidopsis, and the transfected protoplasts were incubated for 20-22 h in darkness before GUS activ ity was assayed. Cotransfection of the GD plasmid DNA was used as a control. Data represent the mean ± SD of three replicates.
We then examined if AtEAU1 and AtEAU2 may able to repress reporter gene expression in transfected protoplasts when recruited to the promoter of the reporter gene by a fused DNA binding domain. Plasmids of the reporter construct LexA-Gal4:GUS and the transcriptional activator construct LD-VP were co-transfected with the effect construct AtEAU1-GD or AtEAU2-GD into Arabidopsis protoplasts. Co-transfection of effector construct GD was used as a control. As the transcriptional activator LD-VP is able to bind to the LexA promoter via the fused LD domain and activate the expression of the LexA-Gal4:GUS reporter gene, whereas the AtEAU1 or AtEAU2 proteins are able to bind to the Gal4 promoter via the fused GD domain, expression levels of the reporter gene will be decreased if AtEAU1 and AtEAU2 function as transcription repressors. As We then examined if AtEAU1 and AtEAU2 may able to repress reporter gene expression in transfected protoplasts when recruited to the promoter of the reporter gene by a fused DNA binding domain. Plasmids of the reporter construct LexA-Gal4:GUS and the transcriptional activator construct LD-VP were co-transfected with the effect construct AtEAU1-GD or AtEAU2-GD into Arabidopsis protoplasts. Co-transfection of effector construct GD was used as a control. As the transcriptional activator LD-VP is able to bind to the LexA promoter via the fused LD domain and activate the expression of the LexA-Gal4:GUS reporter gene, whereas the AtEAU1 or AtEAU2 proteins are able to bind to the Gal4 promoter via the fused GD domain, expression levels of the reporter gene will be decreased if AtEAU1 and AtEAU2 function as transcription repressors. As shown in Figure 3B, compared with cotransfection of GD, GUS activity was dramatically reduced when AtEAU1-GD or AtEAU2-GD was cotransfected, suggesting that AtEAU1 and AtEAU2 are transcription repressors.

Generation of Single and Double Mutants for Gene AtEAU1 and AtEAU2
In order to examine if AtEAU1 and AtEAU2 may involve in the regulation of ABA response in Arabidopsis, we isolated/generated single and double mutants for AtEAU1 and AtEAU2. Since there is 1 T-DNA insertion line available for gene AtEAU1, and 2 are available for AtEAU2 from ABRC, we obtained the seeds and isolated single homozygous mutants ateau1-1 (SAIL_197_H05), ateau2-1 (SAIL_642_C12), and ateau2-2 (SAIL_1242_E02) ( Figure 4A). By crossing ateau1-1 and ateau2-1 mutants and examining F2 plants, we obtained ateau1 ateau2 double mutants.
shown in Figure 3B, compared with cotransfection of GD, GUS activity was dramatically reduced when AtEAU1-GD or AtEAU2-GD was cotransfected, suggesting that AtEAU1 and AtEAU2 are transcription repressors.

Generation of Single and Double Mutants for Gene AtEAU1 and AtEAU2
In order to examine if AtEAU1 and AtEAU2 may involve in the regulation of ABA response in Arabidopsis, we isolated/generated single and double mutants for AtEAU1 and AtEAU2. Since there is 1 T-DNA insertion line available for gene AtEAU1, and 2 are available for AtEAU2 from ABRC, we obtained the seeds and isolated single homozy gous mutants ateau1-1 (SAIL_197_H05), ateau2-1 (SAIL_642_C12), and ateau2-2 (SAIL_1242_E02) ( Figure 4A). By crossing ateau1-1 and ateau2-1 mutants and examining F2 plants, we obtained ateau1 ateau2 double mutants. Considering that only one T-DNA insertion mutant was obtained for AtEAU1, we decided to generate an additional mutant by using CRISPR/Cas9 gene-editing o AtEAU1. Two different target sequences were selected and used to generate CRISPR/Cas9 construct. By transforming the Col wild-type Arabidopsis, examining gene editing status in T1 transgenic plants, and isolating Cas9-free homozygous mutants from T2 plants, we successfully obtained a transgene-free homozygous fragment deletion mutant, namely ateau1-c1. Sequencing results show that there is a 440bp fragment dele tion in the ateau1-c1 mutant ( Figure 4B). The deletion led to substitutions of a few amino acids and a premature stop in AtEUA1, resulting in a short amino acid sequence o AtEAU1 with only 50 amino acids in the ateau1-c1 mutant ( Figure 4C).

ABA Sensitivity Is Increased in the Single and Double Mutants of AtEAU1 and AtEAU2
By using the mutants obtained, we examined the role of AtEAU1 and AtEAU2 in regulating of ABA response in Arabidopsis using seedling greening assay. Seeds of the In the nucleotide sequence alignment, numbers beside sequences indicate the relative position of the nucleotides to the start codon, and underlines indicate the PAM sites. In the amino acid alignment, AtEAU1 sequences in the mutant were subjected to ORFfinder (https://www.ncbi.nlm.nih.gov/orffinder/, accessed on 1 September 2016) for ORF analysis, and the predicted amino acid sequences were aligned with AtEAU1 amino acid sequence. Numbers beside the sequences indicate the total amino acids in the corresponding proteins.
Considering that only one T-DNA insertion mutant was obtained for AtEAU1, we decided to generate an additional mutant by using CRISPR/Cas9 gene-editing of AtEAU1. Two different target sequences were selected and used to generate CRISPR/Cas9 construct. By transforming the Col wild-type Arabidopsis, examining gene editing status in T1 transgenic plants, and isolating Cas9-free homozygous mutants from T2 plants, we successfully obtained a transgene-free homozygous fragment deletion mutant, namely ateau1-c1. Sequencing results show that there is a 440bp fragment deletion in the ateau1-c1 mutant ( Figure 4B). The deletion led to substitutions of a few amino acids and a premature stop in AtEUA1, resulting in a short amino acid sequence of AtEAU1 with only 50 amino acids in the ateau1-c1 mutant ( Figure 4C).

ABA Sensitivity Is Increased in the Single and Double Mutants of AtEAU1 and AtEAU2
By using the mutants obtained, we examined the role of AtEAU1 and AtEAU2 in regulating of ABA response in Arabidopsis using seedling greening assay. Seeds of the Col wild type, the ateau1 and ateau2 single and the ateau1 ateau2 double mutants were plated on 1/2 MS plates at the presence of 1 µM ABA or solvent alone as a control, and seedlings with green cotyledons were counted 14 days after the transfer of the plates into the growth room. The percentage of green cotyledons was then calculated.
We found that, on the control plates, seedlings for all the plants turned to green, however, on the ABA-containing plates, the green seeding rate for the Col wild type was nearly 95%, whereas that for both the ateau1-1 and ateau1-c1 single mutants was about 60%, and that for the ateau2 single mutants was less than 20% ( Figure 5A). To our surprise, we found that the green seeding rate for the ateau1 ateau2 double mutants was about 60%, similar to the ateau1-1 and ateau1-c1 single mutants ( Figure 5A). Col wild type, the ateau1 and ateau2 single and the ateau1 ateau2 double mutants were plated on 1/2 MS plates at the presence of 1 μM ABA or solvent alone as a control, and seedlings with green cotyledons were counted 14 days after the transfer of the plates into the growth room. The percentage of green cotyledons was then calculated. We found that, on the control plates, seedlings for all the plants turned to green, however, on the ABA-containing plates, the green seeding rate for the Col wild type was nearly 95%, whereas that for both the ateau1-1 and ateau1-c1 single mutants was about 60%, and that for the ateau2 single mutants was less than 20% ( Figure 5A). To our surprise, we found that the green seeding rate for the ateau1 ateau2 double mutants was about 60%, similar to the ateau1-1 and ateau1-c1 single mutants ( Figure 5A). Cotyledon greening of the Col wild type Arabidopsis, the ateau1 and ateau2 single, and the ateau1 ateau2 double mutants in response to ABA treatment. Sterilized seeds were sown on 1/2 MS plates containing 1 μM ABA or solvent control, respectively. The plates were kept for 2 days at 4 °C in darkness, and then transferred to a growth room. Fourteen days after the transfer, seedlings with green cotyledons were counted and the percentage of green cotyledons was then calculated. Data represent the mean ± SD of three replicates. (B) Root elongation of the Col wild type, the ateau1 and ateau2 single, and the ateau1 ateau2 double mutants in response to ABA treatment. Sterilized seeds were sown on 1/2 MS plates, kept for 2 days at 4 °C in darkness, then transferred to a growth room and grown vertically for 2 days. Seedlings with similar root length were selected and transferred to plates containing 5 μM ABA or solvent control, respectively. New elongated root length was measured, and percentage of inhibition was then calculated. Data represent the mean ± SD of 10 seedlings.
We also examined response of the mutants to ABA by using root elongation inhibition assays. Sterilized seeds of the Col wild type, the ateau1 and ateau2 single and the ateau1 ateau2 double mutants were plated on 1/2 MS plates and grown vertically for 2 days, then seedlings with root at the same length were transferred to 1/2 MS plates at the presence or absence of 5 μM ABA and grown for 9 additional days, then new elongated root length was measured and the percentage of inhibition was calculated. We found that root elongation of all the seedlings, including the Col wild type, the ateau1 and ateau2 single, and the ateau1 ateau2 double mutants was inhibited by about 50% ( Figure  5B).

Expression Levels of Some ABA Signaling Genes Are Altered in the Single and Double Mutants of AtEAU1 and AtEAU2
Having shown that AtEAU1 and AtEAU2 are ABA response genes, and ABA sensitivity was increased in the ateau1 and ateau2 single and the ateau1 ateau2 double mutants, we wanted to further examine how AtEAU1 and AtEAU2 may regulate plant response to ABA. We examined the expression of ABA single key regulator genes in seedlings of the Col wild type, the ateau1-1 and ateau2-1 single, and the ateau1 ateau2 double mutants, and found that the expression level of the ABA receptor gene PYL4 was increased, but the PP2C gene ABI2 was decreased in the seedlings of the ateau1-1 and ateau2-1 single Cotyledon greening of the Col wild type Arabidopsis, the ateau1 and ateau2 single, and the ateau1 ateau2 double mutants in response to ABA treatment. Sterilized seeds were sown on 1/2 MS plates containing 1 µM ABA or solvent control, respectively. The plates were kept for 2 days at 4 • C in darkness, and then transferred to a growth room. Fourteen days after the transfer, seedlings with green cotyledons were counted and the percentage of green cotyledons was then calculated. Data represent the mean ± SD of three replicates. (B) Root elongation of the Col wild type, the ateau1 and ateau2 single, and the ateau1 ateau2 double mutants in response to ABA treatment. Sterilized seeds were sown on 1/2 MS plates, kept for 2 days at 4 • C in darkness, then transferred to a growth room and grown vertically for 2 days. Seedlings with similar root length were selected and transferred to plates containing 5 µM ABA or solvent control, respectively. New elongated root length was measured, and percentage of inhibition was then calculated. Data represent the mean ± SD of 10 seedlings.
We also examined response of the mutants to ABA by using root elongation inhibition assays. Sterilized seeds of the Col wild type, the ateau1 and ateau2 single and the ateau1 ateau2 double mutants were plated on 1/2 MS plates and grown vertically for 2 days, then seedlings with root at the same length were transferred to 1/2 MS plates at the presence or absence of 5 µM ABA and grown for 9 additional days, then new elongated root length was measured and the percentage of inhibition was calculated. We found that root elongation of all the seedlings, including the Col wild type, the ateau1 and ateau2 single, and the ateau1 ateau2 double mutants was inhibited by about 50% (Figure 5B).

Expression Levels of Some ABA Signaling Genes Are Altered in the Single and Double Mutants of AtEAU1 and AtEAU2
Having shown that AtEAU1 and AtEAU2 are ABA response genes, and ABA sensitivity was increased in the ateau1 and ateau2 single and the ateau1 ateau2 double mutants, we wanted to further examine how AtEAU1 and AtEAU2 may regulate plant response to ABA. We examined the expression of ABA single key regulator genes in seedlings of the Col wild type, the ateau1-1 and ateau2-1 single, and the ateau1 ateau2 double mutants, and found that the expression level of the ABA receptor gene PYL4 was increased, but the PP2C gene ABI2 was decreased in the seedlings of the ateau1-1 and ateau2-1 single and the ateau1 ateau2 double mutants when compared with that in the seedlings of the Col wild type plants ( Figure 6). and the ateau1 ateau2 double mutants when compared with that in the seedlings of the Col wild type plants ( Figure 6). Figure 6. Expression of PYL4 and ABI2 in mutants of AtEAUs. RNA was isolated from 12-day-old seedlings of the Col wild type, the ateau1-1, ateau2-1 single, and the ateau1 ateau2 double mutants, and used for qRT-PCR analysis. Expression of ACT2 was used as an inner reference gene, and expression of PYL4 and ABI2 in the seedlings of the Col wild-type plants was set as 1. Data represent the mean ± SD of three replicates.

Discussion
EAR motif-containing-protein mediated transcriptional repression has been considered to be the main transcriptional repression form in plants [23], consistent with this, EAR motif-containing proteins are involved in the regulation of plant growth and development, as well as plant response to hormones and environmental stresses [22][23][24][25][26][28][29][30][31]. So far it has also been shown that EAR motif-containing proteins are involved in the regulation of hormone responses including ethylene, jasmonate, auxin, and strigolactone [20,23,[26][27][28]. Some of the EAR motif-containing proteins, including ERF7, AITRs, and SlEAD1 have been shown to regulate ABA response in plants [6,25,33]. We provide evidence here that AtEAU1 and AtEAU2 are ABA responsive novel EAR motif-containing transcription repressors that regulating ABA response in Arabidopsis.
First, AtEAU1 has previously reported to be an EAR motif-containing protein [31], and our bioinformatics analysis identified AtEAU2 as its closely related EAR motif-containing protein ( Figure 1). Second, we found that the expression levels of AtEAU1 and AtEAU2 were increased in response to exogenously ABA treatment, but decreased in the ABA biosynthesis mutant aba1-5 ( Figure 2). Third, consistent with the fact that AtEAU1 and AtEAU2 are closely related and both have a fully conserved LxLxL EAR motif (Figure 1), both AtEAU1 and AtEAU2 were specifically localized in the nucleus in Arabidopsis protoplast assays, and functioned as transcription repressors (Figure 3). Last but not least, ABA sensitivity was increased in the ateau1 and ateau2 single mutants in green seedling assays ( Figure 5). These results show that AtEAU1 and AtEAU2 are ABA responsive EAR motif-containing protein genes, and AtEAU1 and AtEAU2 function as transcription repressors to regulate ABA response in Arabidopsis. However, in root elongation assays, no difference was observed between the seedlings of the Col wild type and the mutants, suggesting the AtEAU1 and AtEAU2 may not regulate ABA sensitivity at root elongation stage.
Considering that both AtEAU1 and AtEAU2 are ABA response genes (Figure 2), and AtEAU1 and AtEAU2 shared higher amino acid identity and similarity (Figure 1), and repressed reporter gene expression in transfected protoplasts (Figure 3), we expect that they may have redundant functions, therefore we generated ateau1 ateau2 double mutant, and tested whether that was the case. However, we found that the ateau2 single mutants display greater sensitivity to ABA than the ateau1 single mutants, but sensitivity in the ateau1 ateau2 double mutant is largely similar to the ateau1 single mutants ( Figure  5), and expression level changes of PYL4 and ABI2 in the ateau1 ateau2 double mutant are Figure 6. Expression of PYL4 and ABI2 in mutants of AtEAUs. RNA was isolated from 12-day-old seedlings of the Col wild type, the ateau1-1, ateau2-1 single, and the ateau1 ateau2 double mutants, and used for qRT-PCR analysis. Expression of ACT2 was used as an inner reference gene, and expression of PYL4 and ABI2 in the seedlings of the Col wild-type plants was set as 1. Data represent the mean ± SD of three replicates.

Discussion
EAR motif-containing-protein mediated transcriptional repression has been considered to be the main transcriptional repression form in plants [23], consistent with this, EAR motif-containing proteins are involved in the regulation of plant growth and development, as well as plant response to hormones and environmental stresses [22][23][24][25][26][28][29][30][31]. So far it has also been shown that EAR motif-containing proteins are involved in the regulation of hormone responses including ethylene, jasmonate, auxin, and strigolactone [20,23,[26][27][28]. Some of the EAR motif-containing proteins, including ERF7, AITRs, and SlEAD1 have been shown to regulate ABA response in plants [6,25,33]. We provide evidence here that AtEAU1 and AtEAU2 are ABA responsive novel EAR motif-containing transcription repressors that regulating ABA response in Arabidopsis.
First, AtEAU1 has previously reported to be an EAR motif-containing protein [31], and our bioinformatics analysis identified AtEAU2 as its closely related EAR motif-containing protein ( Figure 1). Second, we found that the expression levels of AtEAU1 and AtEAU2 were increased in response to exogenously ABA treatment, but decreased in the ABA biosynthesis mutant aba1-5 ( Figure 2). Third, consistent with the fact that AtEAU1 and AtEAU2 are closely related and both have a fully conserved LxLxL EAR motif (Figure 1), both AtEAU1 and AtEAU2 were specifically localized in the nucleus in Arabidopsis protoplast assays, and functioned as transcription repressors (Figure 3). Last but not least, ABA sensitivity was increased in the ateau1 and ateau2 single mutants in green seedling assays ( Figure 5). These results show that AtEAU1 and AtEAU2 are ABA responsive EAR motifcontaining protein genes, and AtEAU1 and AtEAU2 function as transcription repressors to regulate ABA response in Arabidopsis. However, in root elongation assays, no difference was observed between the seedlings of the Col wild type and the mutants, suggesting the AtEAU1 and AtEAU2 may not regulate ABA sensitivity at root elongation stage.
Considering that both AtEAU1 and AtEAU2 are ABA response genes (Figure 2), and AtEAU1 and AtEAU2 shared higher amino acid identity and similarity (Figure 1), and repressed reporter gene expression in transfected protoplasts (Figure 3), we expect that they may have redundant functions, therefore we generated ateau1 ateau2 double mutant, and tested whether that was the case. However, we found that the ateau2 single mutants display greater sensitivity to ABA than the ateau1 single mutants, but sensitivity in the ateau1 ateau2 double mutant is largely similar to the ateau1 single mutants (Figure 5), and expression level changes of PYL4 and ABI2 in the ateau1 ateau2 double mutant are also largely similar to those in the ateau1 single mutants ( Figure 6). These results suggest that AtEAU1 and AtEAU2 may function sequentially to regulate ABA response in Arabidopsis, and AtEAU1 may function down stream of AtEAU2. However, more experiments are required to examine if that is the case.
Consistent with the key roles played by ABA in regulating plant abiotic stress responses via regulating the expression of downstream ABA response genes, changes in the expression of ABA-signaling key regulator genes, including the PYR1/PYL/RCAR receptor genes, the PP2Cs phosphatase genes, the SnRKs kinase genes, and ABF/AREB/ABI5-type bZIP transcription factor genes all affected plant responses to abiotic stresses [1][2][3][4][5][6][7]. On the other hand, changes in the expression of some downstream ABA response genes also affected plant abiotic stress response, for example, the R2R3 MYB transcription gene MYB44, the heat shock factor gene HSFA6b, and the bHLH transcription factor gene bHLH112 are all ABA responsive genes involved in the regulation of plant responses to abiotic stresses [35][36][37]. Consistent with the importance of EAR motif-containing proteins in regulating transcriptional repression in plants [23], some EAR motif-containing proteins such as ERF7, AITRs, and SlEAD1 have been shown to be involved in regulating ABA responses [6,25,33]. So far we have used two different strategies to identify EAR motifcontaining proteins; i.e., searching EAR motifs in proteins encoded by ABA-responsive genes with unknown functions, and identifying ABA response genes from unknown function EAR motif-containing protein genes [31]. By using the first strategy, we have successfully identified SlEAD1 as a novel negative regulator of ABA response in tomato [33], and here by using the second strategy, we identified AtEAU1 and AtEAU2 as novel regulators of ABA response in Arabidopsis, indicating that both strategies are practical. Since most of the EAR motif-containing proteins that have been identified are functionally uncharacterized, and some of the EAR motif-containing proteins may have not been identified due to their sequence diversity [31,33], more EAR motif-containing ABA response regulator should be identified by using the two strategies mentioned above, and some of the genes may be used for plant breeding to improve plant abiotic stress tolerance.
Nevertheless, we identified the EAR motif-containing protein genes AtEAU1 and AtEAU2 as ABA responsive genes, we showed that AtEAU1 and AtEAU2 function as transcription repressors, and we found that AtEAU1 and AtEAU2 negatively regulate ABA responses in Arabidopsis, possibly by regulating the expression of the ABA signaling key regulator genes PYL4 and ABI2.

Bioinformatics Analysis
Homologs of AtEAU1 (At1g78170) in Arabidopsis, soybean, rice, tomato, and poplar were identified on Phytozome (https://phytozome-next.jgi.doe.gov, accessed on 1 September 2016) under the term Protein Homologs, and homologs were then used to search on Phytozome again to make sure their homolog protein is AtEUA1. The full-length amino acid sequences of AtEAU1 and its protein homologs were used for phylogenetic analysis on Phylogeny (http://www.phylogeny.fr/simple_phylogeny.cgi, accessed on 1 September 2016) by using the "One Click" mode with default settings, and for sequence alignment using Bioedit 7.2. Amino acid sequence identity and similarity assays of AtEAU1, AtEAU2 and At4g08910 were analyzed on SIAS (http://imed.med.ucm.es/Tools/sias.html, accessed on 1 September 2016).

Plant Materials and Growth Conditions
The Columbia-0 (Col) ecotype Arabidopsis (Arabidopsis thaliana) was used as a wild-type for plant transformation, protoplast isolation, and as a control for ABA response analysis. Seeds of SAIL_197_H05, a T-DNA insertion line for AtEAU1, and SAIL_642_C12 and SAIL_1242_E02, T-DNA insertion lines for AtEAU2 (At1g22250) obtained from the ABRC (Arabidopsis Biological Resource Center) are all in Col background, and were used to isolate the ateau1-1 and ateau2-1 and ateau2-2 mutants, respectively. The ateau1 ateau2-1 double mutant was generated by crossing ateau1-1 and ateau2-1 single mutants, and identifying homozygous double mutants in the F2 generation. The Ler wild type and the aba1-5 mutant in Ler background [34] were used to examine the expression of AtEAUs.
To generate plants for protoplast isolation and plant transformation, seeds of the Col wild type were sown directly into soil pots, germinated and grown in a growth room. To generate seedlings for RNA isolation and ABA response assays, seeds of the Col wild-type, the ateau1 and ateau2 single and ateau1 ateau2 double mutants, the Ler wild type, and the aba1-5 mutant were sterilized for 10 min with 25% (v/v) bleach, washed with sterilized water four times, and plated on 0.6% (w/v) phytoagar (PlantMedia) solidified 1/2 MS (Murashige and Skoog) plates with vitamins (Plant Media) and 1% (w/v) sucrose. The plates were kept in darkness at 4 • C for 2 days, and then transferred to a growth room. The growth conditions in the growth room were set at 22 • C, and a 16 h light/8 h dark long-day condition with photon density at~120 µmol m −2 s −1 .

RNA Isolation and RT-PCR
To examine the expression changes of AtEAU1 and AtEAU2 in response to ABA treatment, 12-day-old seedlings of the Col wild-type were treated for 4 h with 50 µM ABA or solvent methanol, and then RNA was isolated. To examine the expression levels of AtEAU1 and AtEAU2 in the aba1-5 mutant seedlings, RNA was isolated from 12-day-old seedlings of the Ler wild type and the aba1-5 mutant. To examine the expression of ABA signaling genes in the ateau1-1 and ateau2-1 single and ateau1 ateau2 double mutants, RNA was isolated from seedlings of 12-day-old Col wild-type, ateau1-1 and ateau2-1 single and ateau1 ateau2 double mutants. To examine the expression pattern of AtEAUs, roots, radicle, rosette leaves, first stems, second stems, third stems, cauline leaves, inflorescences, and siliques were collected from 5-week-old plants, and used for RNA isolation.
Total RNA isolated was used for cDNA synthesis as described previously [38], and then used to examine the expression of AtEAUs and ABA signaling genes by using RT-PCR/qRT-PCR. The primers used for AtEAU1 were 5 -TTCATGCACCCACACGATCA-3 and 5 -AGGTTCTTCTCTACAAAAAGCCTAA-3 ; for AtEAU2 they were 5 -TCAAAAG GAGCAAACAAGGAGA-3 and 5 -GGCTTTATCAACATGGCGCT-3 . ACT2 (ACTIN2) was used as a control gene for RT-PCR and an inner reference gene for qRT-PCR. The primers for PYL4, ABI2, and ACT2 were as reported previously [39][40][41].

Plant Transformation and Transgenic Plants Selection
About 5-week-old Col wild-type plants with several mature flowers were transformed with the CRISPR/Cas9 constructs to generate the ateau1-c1 single mutant by using the floral dip method [45]. Transgenic plants generated were selected by plating the T1 seeds collected on 1/2 MS plates with 30 µg/mL Hygromycin and 100 µg/mL Carbenicillin. Gene editing status in the T1 plants was examined by PCR amplification and sequencing the genomic sequence of AtEAU1. T2 seeds were collected from gene-edited T1 plants, germinated directly in soil pots, and used to identify Cas9-free homozygous mutant as described in the next section.

DNA Isolation and PCR
To examine gene editing status of AtEAU1, DNA was isolated from leaves of the T1 transgenic plants or the Cas9-free T2 plants, and used for PCR amplification of the genomic sequences of AtEAU1. To isolate transgene-free mutants, DNA was isolated from leaves of the T2 offspring of the gene-edited T1 plants, and subjected to PCR amplification of the Cas9 gene fragment by using the primers described previously [44].

Plasmid DNA Isolation, Protoplast Isolation and Transfection
Plasmid DNA of the reporter and effector genes was isolated by using the Gold Hi Endo Free Plasmid Maxi Kit (CWBIO) according to the manufacturers' instructions. Protoplasts were isolated from rosette leaves collected from 3-4-week-old Col wild-type plants, and the isolated protoplasts were transfected as described previously [6,38,46,47]. For subcellular location assays, plasmid DNA of AtEAU1-GFP and AtEAU2-GFP were transfected into protoplasts. For transcriptional activity assays, plasmid DNA of the reporter gene LexA-Gal4:GUS, the activator gene LD-VP, and the effector genes GD, AtEAU1-GD, or AtEAU2-GD were cotransformed into protoplasts. The transfected protoplasts were incubated in darkness at room temperature for 20-22 h, then GFP fluorescence was observed under an Olympus FV1000 confocal microscope, and GUS activities were measured using a Synergy TM HT microplate reader.

ABA Sensitivity Assays
Assays of ABA-inhibited cotyledon greening and root elongation were performed as described previously [6,[48][49][50]. For cotyledon greening assays, green seedlings on plates containing 1 µm ABA or solvent alone were counted 14 days after the plates were transferred to the growth room. For root elongation assays, 2-day-old seedlings grown on vertically placed plates were transferred to plates containing 5 µm ABA or solvent as a control; the length of new elongated roots was measured 9 days after the transfer. All the experiments were repeated at least three times.  Data Availability Statement: All data were obtained were presented in this article.