A Ubiquitously Expressed UDP-Glucosyltransferase, UGT74J1, Controls Basal Salicylic Acid Levels in Rice

Salicylic acid (SA) is a phytohormone that regulates a variety of physiological and developmental processes, including disease resistance. SA is a key signaling component in the immune response of many plant species. However, the mechanism underlying SA-mediated immunity is obscure in rice (Oryza sativa). Prior analysis revealed a correlation between basal SA level and blast resistance in a range of rice varieties. This suggested that resistance might be improved by increasing basal SA level. Here, we identified a novel UDP-glucosyltransferase gene, UGT74J1, which is expressed ubiquitously throughout plant development. Mutants of UGT74J1 generated by genome editing accumulated high levels of SA under non-stressed conditions, indicating that UGT74J1 is a key enzyme for SA homeostasis in rice. Microarray analysis revealed that the ugt74j1 mutants constitutively overexpressed a set of pathogenesis-related (PR) genes. An inoculation assay demonstrated that these mutants had increased resistance against rice blast, but they also exhibited stunted growth phenotypes. To our knowledge, this is the first report of a rice mutant displaying SA overaccumulation.


Introduction
Salicylic acid (SA), 2-hydroxy benzoic acid, is a small phenolic compound that was first isolated from willow bark. SA functions as a phytohormone and regulates many physiological processes, such as germination, cell growth, respiration, and senescence [1][2][3][4]. SA also regulates tolerance against environmental stresses, including heat, chilling, salinity, and oxidative damage [5][6][7][8]. However, the most important role of SA is the regulation of disease resistance. The hypersensitive response (HR), an aspect of the plant immune response that is characterized by localized necrosis and tissue lesions, is associated with SA accumulation [9]. The HR prevents the penetration and spread of pathogens from the site of infection [10]. Infected local tissues also transmit a signal to distant healthy tissues to induce systemic acquired resistance (SAR) [11,12]. SAR confers resistance to a wide range of pathogens and requires biosynthesis of mobile signals such as methyl salicylate [13].
SA is biosynthesized from chorismic acid via two independent pathways. In the phenylalanine ammonia-lyase (PAL) pathway, cinnamic acid produced by PAL is converted to SA by either β-oxidation or a non-β-oxidation route. In the isochorismic acid synthase (ICS) pathway, isochorismic acid produced by ICS is converted to SA. The intermediate conversion step of the ICS pathway was unknown until a recent study clarified that SA is synthesized from the intermediate isochorismoyl-glutamate in a two-step reaction catalyzed by PBS3 and EPS1 [14]. As an active phytohormone, SA undergoes a variety 2 of 12 of modifications, such as methylation, hydroxylation, and sugar conjugation [15]. SAglycosides are inactive storage forms of SA that accumulate in vacuoles [16][17][18]. SA 2-O-β-glucoside (SAG) and salicyloyl-glucose ester (SGE) have been isolated as major and minor SA glycosides, respectively [19,20]. SA glucosylation is catalyzed by UDP-glucosyltransferases (UGTs). UGT74F1 and UGT76B1 exhibit glucosyltransferase activity toward the hydroxyl group of SA, while UGT74F2 and UGT75B1 show the activity toward its carboxyl group in Arabidopsis thaliana [19,21,22]. Tobacco (Nicotiana tabacum) SAGT (UGT74G1-like) exhibits transferase activities toward both the hydroxy and carboxyl groups of SA [23], and rice (Oryza sativa) OsSGT1 (UGT74H3) catalyzes the glucosylation of the hydroxyl group of SA [24,25]. The UGT74 family proteins function as SA-glucosyltransferases in many plant species, and UGT74 mutant or overexpressing plants have altered SA and SA glucoside levels. The ugt74f1 and ugt74f2 Arabidopsis single mutants show reduced SAG and SGE levels, respectively, and the ugt74f1 mutant also shows an increased SA level [19]. Overexpression of SAGT in Nicotiana benthamiana results in decreased SA and increased SAG levels [26]. The rice OsSGT1 is a probenazole (PBZ)-inducible gene, and its knock-down mutants exhibit a reduced SAG level under PBZ treatment [25].
The changes in endogenous SA levels before and after pathogen infection differ depending on the plant species. In Arabidopsis, tobacco, and cucumber (Cucumis sativus), the basal level of SA is low (less than 0.3 µg/g fresh weight (FW)), but its abundance quickly increases in response to pathogen infection (0.5-2.0 µg/g FW) [11,12,27]. In contrast, rice accumulates a relatively large amount of SA under non-stressed conditions (5.0-30.0 µg/g FW), even more than infected Arabidopsis or tobacco tissues [28,29]. The SA level in rice does not significantly change following inoculation with either bacterial or fungal pathogens [29]. In addition, a comparative study of rice varieties revealed a positive correlation between basal SA level and general blast resistance [29]. These observations support the hypothesis that the SA level before infection is an important factor that determines rice blast resistance. SA-deficient transgenic NahG rice plants exhibit reduced rice blast resistance, further supporting this idea [30]. However, it is not known whether elevated SA levels increase disease resistance in rice, because no rice mutants that constantly overaccumulate SA had been identified.
In this study, we created rice mutants with high basal levels of SA by knocking out a novel UDP-glucosyltransferase gene, UGT74J1. We demonstrate that an increased basal SA level contributes to an increase in disease resistance in rice.

UGT74J1 Is a Ubiquitously Expressed UDP-Glucosyltransferase
It is unclear whether elevated basal levels of SA improve disease resistance in rice. We expected that disrupting SA glucosylation might increase endogenous levels of SA, so we searched the rice genome database, MSU Rice Genome Annotation Project (http: //rice.plantbiology.msu.edu (accessed on 8 September 2021)), for a highly and ubiquitously expressed SA-glucosyltransferase gene. Nine unknown UGT74 proteins with significant homology to a previously characterized SA-glucosyltransferase, OsSGT1, were identified in the database (Figure 1). These UGT genes are clustered on chromosomes 4 and 9, suggesting they originated through gene duplications (Supplementary Figure S1). Plants 2021, 10, x FOR PEER REVIEW 3 of 13 Figure 1. Phylogenetic analysis of UDP-glucosyltransferases in Arabidopsis and rice. The amino acid sequences of the UDP-glucosyltransferase family protein domain (accession: PLN02173) were aligned using ClustalW, and a phylogenetic tree was created using the neighbor-joining method. The bar represents the evolutionary distance between the proteins expressed as the number of substitutions per amino acid residue. Green and blue type indicate genes from rice and Arabidopsis UGTs, respectively. The clade including known SA-glucosyltransferase (UGT74F1, UGT74F2, and UGT74H3) is indicated by orange lines.
To identify UGT genes that control basal SA levels, we analyzed the tissue-specific expression of the UGT74 family genes under non-stressed conditions ( Figure 2). UGT74J13, UGT74J14, UGT74J2, UGT74J3, UGT74J5, and UGT74H12 showed similar patterns of expression, with most mRNA accumulating in the roots, palea, lemma, and ovary. OsSGT1 (UGT74H3) was specifically expressed in the inflorescence and anthers. Expression of UGT74H1 and UGT74H4 was observed primarily in leaf blade and in floral organs such as the palea, lemma, and ovary. On the other hand, UGT74J1 (LOC_Os04g12980, Os04g0206700) mRNA was detected in all tissues examined, although expression levels were tissue-dependent. Analysis of a dataset available at RiceXPro (https://ricexpro.dna.affrc.go.jp/ (accessed on 8 September 2021)) confirmed that UGT74J1 is a ubiquitously expressed UGT74 gene (Supplementary Figure S2). We, therefore, chose UGT74J1 as a target gene for modulating steady-state SA levels. Figure 1. Phylogenetic analysis of UDP-glucosyltransferases in Arabidopsis and rice. The amino acid sequences of the UDP-glucosyltransferase family protein domain (accession: PLN02173) were aligned using ClustalW, and a phylogenetic tree was created using the neighbor-joining method. The bar represents the evolutionary distance between the proteins expressed as the number of substitutions per amino acid residue. Green and blue type indicate genes from rice and Arabidopsis UGTs, respectively. The clade including known SA-glucosyltransferase (UGT74F1, UGT74F2, and UGT74H3) is indicated by orange lines.
To identify UGT genes that control basal SA levels, we analyzed the tissue-specific expression of the UGT74 family genes under non-stressed conditions ( Figure 2). UGT74J13, UGT74J14, UGT74J2, UGT74J3, UGT74J5, and UGT74H12 showed similar patterns of expression, with most mRNA accumulating in the roots, palea, lemma, and ovary. Os-SGT1 (UGT74H3) was specifically expressed in the inflorescence and anthers. Expression of UGT74H1 and UGT74H4 was observed primarily in leaf blade and in floral organs such as the palea, lemma, and ovary. On the other hand, UGT74J1 (LOC_Os04g12980, Os04g0206700) mRNA was detected in all tissues examined, although expression levels were tissue-dependent. Analysis of a dataset available at RiceXPro (https://ricexpro.dna. affrc.go.jp/ (accessed on 8 September 2021)) confirmed that UGT74J1 is a ubiquitously expressed UGT74 gene (Supplementary Figure S2). We, therefore, chose UGT74J1 as a target gene for modulating steady-state SA levels. UGT74J1 is a ubiquitously expressed UDP-glucosyltransferase gene in rice. Each tissue was sampled at the following growth stages: root, leaf sheath, and leaf blade were collected from 30-d-old plants; inflorescence and anther were collected from 7-and 9-wk-old plants, respectively; palea, lemma, and ovary were collected from florets on the fifth day after flowering. Semiquantitative RT-PCR was performed using gene-specific primers, and the cycle number is indicated at right. OsACT1 was used as the endogenous control.

Creation of ugt74j1 Knockout Mutants using CRISPR/Cas9
Mutants for UGT74J1 were created using clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9-mediated genome editing. Two guide RNA sequences were designed using the CRISPRdirect software ( Figure 3A). Transgenic plants expressing Cas9 and either of two targeted guide RNAs were created through Agrobacterium-mediated transformation. Genome-edited plants were selected by cleaved amplified polymorphic sequence (CAPS) and sequencing analysis. We selected two genome-edited T1 plants that had a heterozygous G-nucleotide deletion at target 2 and a heterozygous G-nucleotide insertion at target 3, respectively (Supplementary Figure S3A). These mutated UGT74J1 genes encode shorter peptides, 212 and 370 amino acid (aa), than intact protein, 491 aa. Homozygous mutant plants without the transgene integration were selected at the T2 generation (Supplementary Figure S3B,C). Two independent genome edited lines, 2-7 and 3-13, were utilized in the following study ( Figure 3B). UGT74J1 transcript levels were suppressed in both mutants, while most other UGT74 genes remained unaffected ( Figure 3C). This suppression may be caused by nonsense-mediated mRNA decay [31]. The transcript levels of UGT74J2 and UGT74J3 were slightly increased in the mutants. This may be associated with the SA accumulation in the mutants. Together, these data indicate that lines 2-7 and 3-13 are both knockout mutants of UGT74J1.
x 30 x 30 x 30 L e a f s h e a t h

OsACT1
x 28 x 34 L . UGT74J1 is a ubiquitously expressed UDP-glucosyltransferase gene in rice. Each tissue was sampled at the following growth stages: root, leaf sheath, and leaf blade were collected from 30-d-old plants; inflorescence and anther were collected from 7-and 9-wk-old plants, respectively; palea, lemma, and ovary were collected from florets on the fifth day after flowering. Semi-quantitative RT-PCR was performed using gene-specific primers, and the cycle number is indicated at right. OsACT1 was used as the endogenous control.

Creation of ugt74j1 Knockout Mutants Using CRISPR/Cas9
Mutants for UGT74J1 were created using clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9-mediated genome editing. Two guide RNA sequences were designed using the CRISPRdirect software ( Figure 3A). Transgenic plants expressing Cas9 and either of two targeted guide RNAs were created through Agrobacterium-mediated transformation. Genome-edited plants were selected by cleaved amplified polymorphic sequence (CAPS) and sequencing analysis. We selected two genome-edited T 1 plants that had a heterozygous G-nucleotide deletion at target 2 and a heterozygous G-nucleotide insertion at target 3, respectively (Supplementary Figure S3A). These mutated UGT74J1 genes encode shorter peptides, 212 and 370 amino acid (aa), than intact protein, 491 aa. Homozygous mutant plants without the transgene integration were selected at the T 2 generation (Supplementary Figure S3B,C). Two independent genome edited lines, 2-7 and 3-13, were utilized in the following study ( Figure 3B). UGT74J1 transcript levels were suppressed in both mutants, while most other UGT74 genes remained unaffected ( Figure 3C). This suppression may be caused by nonsense-mediated mRNA decay [31]. The transcript levels of UGT74J2 and UGT74J3 were slightly increased in the mutants. This may be associated with the SA accumulation in the mutants. Together, these data indicate that lines 2-7 and 3-13 are both knockout mutants of UGT74J1.  The genotypes of two independent mutants, 2-7 and 3-13. A partial fragment including the CRISPR/Cas9 target site was amplified from the mutants and aligned with the wildtype sequence. (C) The expression of UGT74 genes in the mutants. Total RNA was extracted from 14-d-old plants. Semi-quantitative RT-PCR was performed using gene-specific primers, and the cycle numbers applied are indicated at the right column. OsACT1 was used as an endogenous control.

Ugt74j1 Mutants show SA Overaccumulation
We suspected that the ugt74j1 mutants might have changes in SA level due to the blockage of glucosylation, so we determined SA levels for the mutants by ultraperformance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS). The ugt74j1 mutants 2-7 and 3-13 exhibited 1.7-and 1.3-fold increases in SA levels, respectively ( Figure 4A). We also determined SAG levels in the mutants. Although they showed a tendency towards lower SAG levels, there was no significant difference from that of the wild type ( Figure 4B). Similarly, there was no significant difference in SGE levels between the mutant and wild-type plants ( Figure 4C). These data suggest that UGT74J1 regulates basal levels of SA under non-stressed conditions. In addition, the ugt74j1 mutants showed reduced stature and delayed heading relative to the parental strain under non-stressed conditions ( Figure 5A). The leaf sheath and blade were 10-20% shorter than in the wild type ( Figure 5B). The heading date of the ugt74j1 mutants was delayed by 5-6 d ( Figure 5C), while no significant alterations were observed in fertility (Supplementary Figure S4).
x 36 x 36 x 32 x 32 x 32 x 30 x 36 x 28 x 36 x 36 The genotypes of two independent mutants, 2-7 and 3-13. A partial fragment including the CRISPR/Cas9 target site was amplified from the mutants and aligned with the wildtype sequence. (C) The expression of UGT74 genes in the mutants. Total RNA was extracted from 14-d-old plants. Semi-quantitative RT-PCR was performed using gene-specific primers, and the cycle numbers applied are indicated at the right column. OsACT1 was used as an endogenous control.

Ugt74j1 Mutants Show SA Overaccumulation
We suspected that the ugt74j1 mutants might have changes in SA level due to the blockage of glucosylation, so we determined SA levels for the mutants by ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS). The ugt74j1 mutants 2-7 and 3-13 exhibited 1.7-and 1.3-fold increases in SA levels, respectively ( Figure 4A). We also determined SAG levels in the mutants. Although they showed a tendency towards lower SAG levels, there was no significant difference from that of the wild type ( Figure 4B). Similarly, there was no significant difference in SGE levels between the mutant and wildtype plants ( Figure 4C). These data suggest that UGT74J1 regulates basal levels of SA under non-stressed conditions. In addition, the ugt74j1 mutants showed reduced stature and delayed heading relative to the parental strain under non-stressed conditions ( Figure 5A). The leaf sheath and blade were 10-20% shorter than in the wild type ( Figure 5B). The heading date of the ugt74j1 mutants was delayed by 5-6 d ( Figure 5C), while no significant alterations were observed in fertility (Supplementary Figure S4). Semi-quantitative RT-PCR was performed using gene-specific primers, and the cycle numbers applied are indicated at the right column. OsACT1 was used as an endogenous control.

Ugt74j1 Mutants show SA Overaccumulation
We suspected that the ugt74j1 mutants might have changes in SA level due to the blockage of glucosylation, so we determined SA levels for the mutants by ultraperformance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS). The ugt74j1 mutants 2-7 and 3-13 exhibited 1.7-and 1.3-fold increases in SA levels, respectively ( Figure 4A). We also determined SAG levels in the mutants. Although they showed a tendency towards lower SAG levels, there was no significant difference from that of the wild type ( Figure 4B). Similarly, there was no significant difference in SGE levels between the mutant and wild-type plants ( Figure 4C). These data suggest that UGT74J1 regulates basal levels of SA under non-stressed conditions. In addition, the ugt74j1 mutants showed reduced stature and delayed heading relative to the parental strain under non-stressed conditions ( Figure 5A). The leaf sheath and blade were 10-20% shorter than in the wild type ( Figure 5B). The heading date of the ugt74j1 mutants was delayed by 5-6 d ( Figure 5C), while no significant alterations were observed in fertility (Supplementary Figure S4). x 36 x 36 x 32 x 32 x 32 x 30 x 36 x 28 x 36 x 36 leaf of 25-d-old plants and quantified using UPLC-MS/MS. The data represent the means of four independent experiments ± SE. Significant differences were calculated using Student's t-tests: * p < 0.05; ** p < 0.01. . Asterisks indicate statistically significant differences compared with the wild type (t-test, * p < 0.05, ** p < 0.01). (C) The heading dates of the mutants. The data represent the mean ± SE (n = 15). Asterisks indicate statistically significant differences compared with the wild type (t-test, ** p < 0.01).

SA Overaccumulation Upregulates Defense-related Gene Expression
To determine whether the difference in basal SA levels affects defense gene expression, we compared the transcriptomes of the ugt74j1 mutant 2-7 and wild-type lines using a microarray. Several pathogenesis-related (PR) genes were identified as upregulated genes in the mutant (Supplementary Table S1), and the expression of these PR genes was validated by using real-time PCR in both ugt74j1 mutants, 2-7 and 3-13. In the ugt74j1 mutants, the expression of PR1 (OsPR1#011), PR2 (Gns10), PR4 (PR4b), PR8 (OsChi3a), and PR10 (JIOsPR10) were significantly increased compared with wild type ( Figure 6). Moreover, there seemed to be a correlation between the basal SA accumulation and the level of PR gene upregulation in these three lines (Figures 4 and 6). Together, these data suggested that overaccumulation of SA induces resistance to a broad spectrum of pathogens. . Asterisks indicate statistically significant differences compared with the wild type (t-test, * p < 0.05, ** p < 0.01). (C) The heading dates of the mutants. The data represent the mean ± SE (n = 15). Asterisks indicate statistically significant differences compared with the wild type (t-test, ** p < 0.01).

SA Overaccumulation Upregulates Defense-Related Gene Expression
To determine whether the difference in basal SA levels affects defense gene expression, we compared the transcriptomes of the ugt74j1 mutant 2-7 and wild-type lines using a microarray. Several pathogenesis-related (PR) genes were identified as upregulated genes in the mutant (Supplementary Table S1), and the expression of these PR genes was validated by using real-time PCR in both ugt74j1 mutants, 2-7 and 3-13. In the ugt74j1 mutants, the expression of PR1 (OsPR1#011), PR2 (Gns10), PR4 (PR4b), PR8 (OsChi3a), and PR10 (JIOsPR10) were significantly increased compared with wild type (Figure 6). Moreover, there seemed to be a correlation between the basal SA accumulation and the level of PR gene upregulation in these three lines (Figures 4 and 6). Together, these data suggested that overaccumulation of SA induces resistance to a broad spectrum of pathogens. Figure 6. PR gene expression is increased in the ugt74j1 mutants. qRT-PCR was performed using total RNA isolated from the fifth leaves of the wild type and mutants grown under non-stressed conditions. The expression level of each PR gene was normalized to the expression of OsACT1. These data represent the means of eight independent experiments ± SE. Significant differences compared with the wild type were calculated using Student's t-tests: * p < 0.05; ** p < 0.01.

Ugt74j1 Mutants Have Increased Rice Blast Resistance
We gauged disease resistance of the ugt74j1 mutants by a leaf inoculation assay using P. oryzae. The detached fifth leaves of the mutants and wild type were inoculated with a spore suspension of P. oryzae. After a 5-d incubation, we measured the lesion lengths on the inoculated leaves. The ugt74j1 mutant lines 2-7 and 3-13 showed a markedly decreased lesion length as compared to wild-type leaves ( Figure 7A). The average reduction was 50% and 62%, respectively, and these changes were statistically significant ( Figure 7B). Thus, activation of PR genes in response to increased endogenous SA levels likely contributes to enhanced resistance to P. oryzae.  Figure 6. PR gene expression is increased in the ugt74j1 mutants. qRT-PCR was performed using total RNA isolated from the fifth leaves of the wild type and mutants grown under non-stressed conditions. The expression level of each PR gene was normalized to the expression of OsACT1. These data represent the means of eight independent experiments ± SE. Significant differences compared with the wild type were calculated using Student's t-tests: * p < 0.05; ** p < 0.01.

Ugt74j1 Mutants Have Increased Rice Blast Resistance
We gauged disease resistance of the ugt74j1 mutants by a leaf inoculation assay using P. oryzae. The detached fifth leaves of the mutants and wild type were inoculated with a spore suspension of P. oryzae. After a 5-d incubation, we measured the lesion lengths on the inoculated leaves. The ugt74j1 mutant lines 2-7 and 3-13 showed a markedly decreased lesion length as compared to wild-type leaves ( Figure 7A). The average reduction was 50% and 62%, respectively, and these changes were statistically significant ( Figure 7B). Thus, activation of PR genes in response to increased endogenous SA levels likely contributes to enhanced resistance to P. oryzae. . Asterisks indicate statistically significant differences compared with the wild type (t-test, p < 0.01).

Discussion
Our expression analysis identified UGT74J1 as a novel UGT74 group gene that is expressed ubiquitously. In the ugt74j1 mutants, we observed overaccumulation of SA and P. oryzae. The fifth leaves of the wild type and the ugt74j1 mutants were inoculated with a spore suspension. After 5-d incubation, lesion length was imaged. (B) Quantification of the lengths of the lesions shown in (A). The data are represented as the mean ± SE (n = 24). Asterisks indicate statistically significant differences compared with the wild type (t-test, p < 0.01).

Discussion
Our expression analysis identified UGT74J1 as a novel UGT74 group gene that is expressed ubiquitously. In the ugt74j1 mutants, we observed overaccumulation of SA and a concomitant upregulation of PR genes, and the plants exhibited resistance against rice blast. This supported the previous observation of a positive correlation between basal SA levels and disease resistance in a wide range of rice varieties.
So far, OsSGT1 was the only rice UDP-glucosyltransferase known to have activity toward SA [24,25]. Our data now suggest that UGT74J1 also utilizes SA as a substrate. Our phylogenetic analysis identified ten members in the UDP-glucosyltransferase clade that includes OsSGT1 and UGT74J1 (Figure 1). The sub-clade including OsSGT1 has four members that cluster on chromosome 9, while the other sub-clade, represented by UGT74J1, has six members and clusters on chromosome 4. It is not known whether members of these two sub-clades have distinct enzymatic activities. Among the genes in the chromosome 9 group, OsSGT1 has been functionally characterized as an SA-glucosyltransferase that can transfer a glucose moiety to the hydroxy group of SA [24,25]. In contrast to our ugt74j1 mutant, OsSGT1 knockdown plants do not show altered SA levels under non-stressed conditions [25]. This could be explained by the fact that OsSGT1 is a PBZ-inducible gene [25], and basal expression of OsSGT1 is maintained at low levels in healthy leaves and roots ( Figure 2). Therefore, the contribution of OsSGT1 to the basal SA level appears to be small.
The UGT genes that form a clade with OsSGT1 exhibit tissue-specific expression in leaves (UGT74H1), in florets (UGT74H4), and in roots and florets (UGT74J13, UGT74J14, UGT74J2, UGT74J3, UGT74J5, UGT74H12) ( Figure 2). Therefore, these UGTs are expected to catalyze SA glucosylation in specific tissues. On the other hand, UGT74J1 was unique in its ubiquitous expression throughout the plant under non-stressed conditions (Figure 2). The ugt74j1 mutants showed significant increases in basal SA levels ( Figure 4A). These data suggested that UGT74J1 is involved in the glucosylation of SA. SAG is the major glucosylation form of SA. Measurement of SAG levels in wild type and the ugt74j1 mutants revealed a trend of lower SAG levels in the mutants, but the differences were not statistically significant ( Figure 4A,B). We also determined basal SGE levels in the mutants and wild type and no significant difference was found in the mutants and wild type. However, the levels of SGEs were less than one-thousandth of SA and SAG levels ( Figure 4C). Although more detailed verification is required, it is possible that SGE does not function as a storage form of SA under non-stress conditions. We believe that UGT74J1 is a constitutive SAglucosyltransferase that controls SA-SAG homeostasis under non-stressed conditions.
The endogenous level of SA in the absence of stress is relatively high in rice as compared with other plant species [28,32]. For example, non-stressed leaf tissues contaiñ 0.3 µg SA/g FW in Arabidopsis [27] and 0.04-0.07 µg SA/g FW in tobacco [33], while in rice (cultivar 'Nipponbare') leaves contain 10 µg SA/g FW [25,30]. It is hypothesized that rice has a different mode of action for SA in defense signaling, and a comparative study of basal SA levels in 28 rice varieties revealed that they ranged from 7 µg/g FW to 30 µg/g FW and were positively correlated with blast resistance [29]. This suggests that the blast resistance of a variety could be enhanced by increasing the endogenous SA level. The endogenous SA level of our ugt74j1 mutant line 3-13 was 6.30 µg/g FW, which was 1.3-fold higher than that of the wild type (cultivar 'Yukihikari') at 4.94 µg/g FW (Figure 4). This increase was enough to induce expression of a set of defense-related genes ( Figure 6) and confer increased resistance against rice blast (Figure 7). The basal SA level of wild-type 'Yukihikari' was lower than the range of SA levels observed in the comparative study [29]. This could be explained by differences in the methods utilized for quantifying SA level; we utilized UPLC-MS/MS, which generally allows more specific and accurate measurement than can be obtained using high-performance liquid chromatography with an ultraviolet detector, as in the Silverman study.
Our study of the ugt74j mutants strongly supports the hypothesis that basal SA levels determine disease resistance in rice. Based on this insight, it should be possible to improve the disease resistance of existing rice cultivars by modifying endogenous SA levels. On the other hand, we also showed that excessive SA accumulation has a negative effect on growth. Future studies will be needed to address how to uncouple these two traits.

Accession Number
The accession numbers of Arabidopsis and rice UGT genes in this study are shown below:

Phylogenetic Analysis
The amino acid sequences of UGTs from Arabidopsis and rice were obtained from TAIR (https://www.arabidopsis.org/ (accessed on 8 September 2021)) and the MSU Rice Genome Annotation Project (http://rice.plantbiology.msu.edu/ (accessed on 8 September 2021)). The accession numbers of these amino acid sequences are listed in Supplementary Table S2. The amino acid sequence alignment was performed using ClustalW (http://clustalw.ddbj.nig.ac.jp/ (accessed on 8 September 2021)). The phylogenetic tree was drawn by the neighbor-joining method using Geneious software (Biomatters, Auckland, New Zealand).

Plant Materials and Growth Conditions
Rice cultivar 'Yukihikari' was used in this study. The seeds were sterilized with 70% ethanol for 5 min and with 2% sodium hypochlorite for 30 min and subsequently incubated for 2 d in the dark at 30 • C for gemination. The germinated seeds were planted in soil and grown in a greenhouse maintained at a 14 h light/10 h dark cycle at 28 • C. The length of the leaf blade and leaf sheath was measured in 50-d-old plants. The heading day was recorded when the first spike emerged from the tip of the leaf sheath. Spikelet fertility was calculated as the number of set seeds per total spikelet number and was expressed as a percentile.

Semi-Quantitative RT-PCR
Plant tissues were sampled at several growth stages. Root, leaf sheath, and leaf blade were collected from 30-d-old plants. Inflorescence and anther were collected from 7-and 9-wk-old plants, respectively. Palea, lemma, and ovary were collected from florets at 5 d after flowering. Total RNA was extracted from each tissue using an RNeasy Plant Mini Kit (Qiagen, Germany), and first-strand cDNA was synthesized from 500 ng total RNA using the PrimeScript RT reagent Kit (Takara, Japan). Semi-quantitative RT-PCR was performed using PrimeSTAR GXL (Takara, Japan). Gene-specific primers are listed in Supplementary  Table S3. The following conditions were used for amplification of partial UGT fragments: 98 • C for 30 s, followed by 30-38 cycles of 98 • C for 10 s, 60 • C for 15 s, and 68 • C for 30 s. The housekeeping gene OsACT1 was amplified under the following conditions: 98 • C for 30 s, followed by 28 cycles of 98 • C for 10 s, 55 • C for 15 s, and 68 • C for 30 s. Semiquantitative RT-PCR was repeated at least two times for each gene. Representative results were shown in the figures.

Creation of the Ugt74j1 Mutants by Genome Editing
The CRISPR/Cas9 system was used to create the ugt74j1 mutants. The CRISPR/Cas9 target sites were located in the first or second exon of UGT74J1 and were designed using the CRISPRdirect software (https://crispr.dbcls.jp/ (accessed on 8 September 2021)). The expression vectors for the CRISPR/Cas9 system were constructed as in a previous study [34]. The paired oligonucleotides encoding a 20-nt sequence on the target site were annealed and introduced into pU6gRNA: GTTGGGTTGCTGAGCGTGGAGCTC and AAACGAGCTCCACGCTCAGCAACC for target site 2; GTTGGTGAGATAAGACCTCAAGCT and AAACAGCTTGAGGTCTTATCTCAC for target site 3, where the underline indicates the 20-nt target site. Subsequently, the gRNA expression cassettes on pU6gRNA were digested and subcloned into the pZH_gYSA_MMCas9 binary vector using Pac I and Asc I. Transgenic plants expressing the CRISPR/Cas9 system were generated using an Agrobacterium tumefaciens-mediated transformation method [35]. Partial UGT74J1 fragments containing target 2 or target 3 sites were amplified with each specific primer set. Mutations were detected by cleaved amplified polymorphic sequences (CAPS) assay using Sac I and Alu I, respectively [34]. The CAPS-positive bands were then sequenced using the BigDye Terminatorv3.1 Cycle Sequencing Kit and a 3130xl Genetic Analyzer (Thermo Fisher Scientific, Waltham, MA, USA). Homozygous mutant plants lacking the CRISPR/Cas9 expression cassette were isolated by self-crossing. Separation of the expression cassettes was confirmed by amplifying hygromycin phosphotransferase fragments. The primers used in the creation of the mutant are listed in Supplementary Table S3.

Quantification of Endogenous SA and SA Glucosides
Endogenous levels of SA, SAG, and SGE were determined using the fifth leaf of 25-dold plants. The leaf tissue was frozen in liquid nitrogen immediately after harvesting and then ground to a fine powder. Sample preparation and measurement using a deuteriumlabeled internal control were performed as described previously [36,37].

Microarray Analysis
Total RNA was extracted from the fifth leaf of wild-type and mutant plants using the RNeasy Plant Mini Kit (Qiagen, Hilden, Germany). First-strand cDNA was synthesized from 400 ng total RNA and labeled with cyanine 3 or cyanine 5 using a Quick Amp Labeling Kit (Agilent Technologies, Santa Clara, CA, USA). Two-color microarray using a rice 44k oligo microarray slide (Agilent Technologies, USA) was performed. Hybridization and scanning were performed according to the manufacturer's instructions.

Expression Analysis Using Real-Time PCR
Total RNA was extracted from the fifth leaf of wild-type and mutant plants using the RNeasy Plant Mini Kit (Qiagen, Hilden, Germany). First-strand cDNA was synthesized from 500 ng total RNA using the PrimeScript RT reagent kit (Takara, Otsu, Japan). Quantitative RT-PCR (qRT-PCR) was performed using an ABI7500 Real-Time PCR System (Thermo Fisher Scientific, Waltham, MA, USA) and TB Green Premix Ex Taq II (Takara, Otsu, Japan) according to the manufacturer's instructions. Gene-specific primers used for the qRT-PCR are listed in Supplementary Table S3.

Detached Leaf Inoculation Assay
The rice cultivar 'Yukihikari' carries the Pia resistance gene and is susceptible to Pyricularia oryzae (race: 007). A spore suspension of P. oryzae (race: 007) was prepared as described in a previous study [38]. The fifth leaves of 25-d-old plants were detached and placed on wet paper towels in a plastic box. Each leaf was inoculated with four 3-µL spots of the spore suspension (1.0 × 10 5 spores/mL) and incubated for 5 d at 25 • C. The lesion lengths were measured from digital images of the leaves using ImageJ. The differences in lesion size between the wild type and each mutant were statistically analyzed using a Student's t-test.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/plants10091875/s1, Figure S1: Schematic model of the UGT gene clusters on chromosomes 4 and 9, Figure S2: Tissue-specific expression of the UGT74 genes, Figure S3: Screening for the ugt74j1 mutants, Figure S4: Spikelet fertility of the ugt74j1 mutants, Table S1: Screening for defense genes using microarray, Table S2: Accession numbers of UDP-glucosyltransferases used in phylogenetic analysis, Table S3: Primers used in this study.