MOS1 Negatively Regulates Sugar Responses and Anthocyanin Biosynthesis in Arabidopsis

Sugars, which are important signaling molecules, regulate diverse biological processes in plants. However, the convergent regulatory mechanisms governing these physiological activities have not been fully elucidated. MODIFIER OF snc1-1 (MOS1), a modulator of plant immunity, also regulates floral transition, cell cycle control, and other biological processes. However, there was no evidence of whether this protein was involved in sugar responses. In this study, we found that the loss-of-function mutant mos1-6 (mos1) was hypersensitive to sugar and was characterized by defective germination and shortened roots when grown on high-sugar medium. The expression of MOS1 was enhanced by sucrose. Hexokinase 1, an important gene involved in sugar signaling, was upregulated in the mos1 mutant compared to wild-type Col-0 in response to sugar. Furthermore, the mos1 mutant accumulated more anthocyanin than did wild-type Col-0 when grown on high-sugar concentration medium or under high light. MOS1 was found to regulate the expression of flavonoid and anthocyanin biosynthetic genes in response to exogenous sucrose and high-light stress but with different underlying mechanisms, showing multiple functions in addition to immunity regulation in plant development. Our results suggest that the immune regulator MOS1 serves as a coordinator in the regulatory network, governing immunity and other physiological processes.


Introduction
Sugars not only serve as energy sources in plants but also as hormone-like molecules in regulating many important physiological processes, including metabolism [1,2], seed germination [3], and biotic and abiotic stress responses [4,5]. Many sucrose-insensitive or -hypersensitive mutants have been screened to identify genes involved in sugar signaling [6][7][8][9]. By studying these mutants, it has been recognized that sugars have crosstalk with other signals, such as light [10], hormones [11,12], stresses [12], and nutrients [13,14]. Sugar signaling is usually triggered by glucose [15], although sucrose is the main type of sugar for systemic transport in plants [16].
Sugar signaling pathways are conserved in eukaryotes [17]. Hexokinases (HXKs), a group of identified glucose sensors, also govern glucose phosphorylation and regulate sugar responses [10]. In Arabidopsis, HXK1 mutants are insensitive to glucose, and HXK1 has been reported to coordinate sugar, light, and hormones to control plant growth [10]. TREHALOSE-6-PHOSPHATE SYNTHASE (TPS), which is involved in the HXK-dependent glucose signaling pathway, catalyzes the biosynthesis of Mutant mos1 has been reported to have delayed flowering, increased ploidy level, and changed rosette size and other mutant phenotypes [39,40]. A previous study revealed that MOS1 antagonized MAD1 activity by interacting with MAD2 in endoreduplication regulation [40]. MAD2 loss-of-function mutants have defects in early seedling development, and these defects can be rescued by exogenous sugars [42]. Therefore, it is possible that exogenous sugar treatment also affects mos1 seedling development. To characterize the roles of MOS1 in plant development, in this study, we investigated the responses of mos1 mutants, wild-type Col-1, and mos1 complementation lines #1 and #2-9 under different sugar concentrations, including the germination rate, the expression of MOS1, a sugar-responsive gene Subunit 3 of ADP-Glucose Pyrophosphorylase (APL3) [43], and other genes involved in the sugar response pathway, i.e., HXK1 and TPS1 in the HXK1-dependent pathway [10,18], RGS1 and GPA1 in the RGS pathway [20,44], AKIN10 and AKIN11 in the SNF1-RELATED KINASE1 pathway [22,23,45,46], and SUGAR-INSENSITIVE 3 encoding an E3 ligase in an independent sugar-response pathway [47]. We found that the MOS1 knockout mutant mos1 exhibited hypersensitive responses to sugar. We hypothesized that MOS1 might participate in sugar signaling pathways and other processes related to sugar signaling. The HXK1 gene involved in sugar signaling pathways showed enhanced expression in the mos1 mutant in response to sugar. As sucrose is known as a positive factor in the accumulation of anthocyanin pigments [2] and high light stress can also boost the biosynthesis of anthocyanin [27], we speculated that MOS1 might also be involved in sugar-and light-induced anthocyanin biosynthesis. Then, we analyzed anthocyanin accumulation in wild-type Col-0 and mos1, #1, and #2-9 grown on medium supplemented with different concentrations of sugar and grown in the same soil medium but under different light intensities and found that MOS1 is involved in the regulation of anthocyanin biosynthesis triggered by sugar and light by affecting the expression of ABGs and FLS. This finding suggests that MOS1 has multiple roles in organizing sugar signaling and immune responses, thereby functioning as a coordinator in developmental, biotic, and abiotic stress responses.

mos1 Mutant Was Hypersensitive to Sugar during Early Seedling Development
When mos1 mutant and wild-type Col-0 seeds were sown on medium supplemented with exogenous sugars, there was no difference in the germination rates of the mos1 mutant from that of wild-type Col-0 when sown on half-strength Murashige and Skoog (MS) medium supplemented with 0.8% sucrose (w/v) (normal medium; Figure S1). However, when seeds were sown on half-strength MS medium supplemented with 4% glucose, the germination rate of the mos1 mutant was reduced to 40% ( Figure 1A,B), while the germination rate of wild-type Col-0 was 94% ( Figure 1A,B). A similar germination phenotype was observed when seeds were sown on medium supplemented with 6% sucrose ( Figure S1). However, when mos1 and wild-type Col-0 seeds were sown on half-strength MS medium supplemented with 4% mannitol (equimolar concentrations of glucose), no significant difference in germination rates was observed between them ( Figure 1A,B), indicating that the defective germination of mos1 sown on medium supplemented with 4% glucose or 6% sucrose was not caused by osmotic stress but by sugars.
Furthermore, we investigated the expressions of APL3 under different sugar treatments. The expression level of APL3 in mos1 was similar to that of wild-type Col-0 when seedlings were grown on half-strength MS supplemented with 6% mannitol ( Figure 1C), while it was 160% higher in mos1 when grown on half-strength MS supplemented with 6% sucrose ( Figure 1C). These results were consistent with a previous study [48] and indicated that the mos1 mutant was sensitive to sugar. The germination rates of Col-0 and mos1, #1, and #2-9 grown on half-strength MS medium with 4% M or 4% G. Different letters above the bars indicate significant differences (one-way ANOVA/Bonferroni p < 0.001). (C) qRT-PCR analysis of APL3 expression in Col-0 and mos1 seedlings grown on half-strength MS medium with 6% M or 6% S. Quantification was normalized to ACTIN2. Error bars indicate standard error (SE) of two independent biological replicates. The asterisk indicates a significant difference compared with the corresponding Col-0 (one-way ANOVA/Bonferroni p < 0.001). (D) Relative root lengths of 7-d-old Col-0 and mos1, #1, and #2-9 seedlings grown on half-strength MS medium with 0.8% sucrose (S), 2% M or 2% G. Furthermore, we investigated the expressions of APL3 under different sugar treatments. The expression level of APL3 in mos1 was similar to that of wild-type Col-0 when seedlings were grown on half-strength MS supplemented with 6% mannitol ( Figure 1C), while it was 160% higher in mos1 when grown on half-strength MS supplemented with 6% sucrose ( Figure 1C). These results were consistent with a previous study [48] and indicated that the mos1 mutant was sensitive to sugar.
To confirm whether the absence of MOS1 is responsible for sugar hypersensitivity in the mos1 mutant, we generated the MOS1 rescue construct pMOS1::MOS1:GFP and obtained two independent complementation lines, pMOS1::MOS1:GFP mos1-6 #1 (#1) and pMOS1::MOS1:GFP mos1-6 #2-9 (#2-9) ( Figure S2). These two lines partially rescued the defect germination rate of mos1 grown on medium supplemented with sugar ( Figures 1A,B and S1), confirming that the knockdown of MOS1 is responsible for the defective germination rate and contributes to the sugar hypersensitivity of mos1.
As high-sugar treatment also affected root elongation [25], we analyzed the root lengths of the wild-type Col-0 and mos1, #1, and #2-9 seedlings grown on medium containing different concentrations of sugar for 7 d. The root lengths of mos1 were comparable to the wild-type when seedlings were grown on normal medium or half-strength MS with 2% mannitol ( Figure 1D). However, when seedlings were grown on half-strength MS with 2% glucose, the roots of mos1 were The germination rates of Col-0 and mos1, #1, and #2-9 grown on half-strength MS medium with 4% M or 4% G. Different letters above the bars indicate significant differences (one-way ANOVA/Bonferroni p < 0.001). (C) qRT-PCR analysis of APL3 expression in Col-0 and mos1 seedlings grown on half-strength MS medium with 6% M or 6% S. Quantification was normalized to ACTIN2. Error bars indicate standard error (SE) of two independent biological replicates. The asterisk indicates a significant difference compared with the corresponding Col-0 (one-way ANOVA/Bonferroni p < 0.001). (D) Relative root lengths of 7-d-old Col-0 and mos1, #1, and #2-9 seedlings grown on half-strength MS medium with 0.8% sucrose (S), 2% M or 2% G.
To confirm whether the absence of MOS1 is responsible for sugar hypersensitivity in the mos1 mutant, we generated the MOS1 rescue construct pMOS1::MOS1:GFP and obtained two independent complementation lines, pMOS1::MOS1:GFP mos1-6 #1 (#1) and pMOS1::MOS1:GFP mos1-6 #2-9 (#2-9) ( Figure S2). These two lines partially rescued the defect germination rate of mos1 grown on medium supplemented with sugar ( Figure 1A,B and Figure S1), confirming that the knockdown of MOS1 is responsible for the defective germination rate and contributes to the sugar hypersensitivity of mos1.
As high-sugar treatment also affected root elongation [25], we analyzed the root lengths of the wild-type Col-0 and mos1, #1, and #2-9 seedlings grown on medium containing different concentrations of sugar for 7 d. The root lengths of mos1 were comparable to the wild-type when seedlings were grown on normal medium or half-strength MS with 2% mannitol ( Figure 1D). However, when seedlings were grown on half-strength MS with 2% glucose, the roots of mos1 were shorter than those of wild-type Col-0 ( Figure 1D). Complementation lines #1 and #2-9 both showed normal root elongation, as with the wild-type ( Figure 1D). This finding confirmed that the shortened root in mos1 was due to the loss of function of MOS1.

Expression of MOS1 Is Induced by Sucrose
As mos1 showed hypersensitivity to sugars, to explore the role of MOS1 in sugar signaling pathways, the expression of the MOS1 gene in wild-type Col-0 under different concentrations of sucrose was investigated. Compared to the seedlings grown on normal medium, the expression level of MOS1 in seedlings grown on medium supplemented with 6% sucrose increased by 270% (Figure 2A). To elucidate the expression pattern of MOS1 in response to exogenous sucrose, the transgenic plants pMOS1:GUS harboring the β-glucuronidase (GUS) reporter gene under the promoter and the first exon of MOS1 [39] were used. Histochemical analysis showed that the expression of GUS driven by the MOS1 promoter was related to the developmental stages of leaves and sucrose concentrations. When seedlings were grown on normal medium, strong GUS signals were detected in the emerging tissues, but only notably weak GUS signals were detected in the mature tissues. However, when seedlings were transferred to medium containing 6% sucrose, the intensity of the GUS signal was stronger, with obvious GUS signals being detected in the mature tissues ( Figure 2B). These results showed that the expression of MOS1 is promoted by exogenous sucrose. shorter than those of wild-type Col-0 ( Figure 1D). Complementation lines #1 and #2-9 both showed normal root elongation, as with the wild-type ( Figure 1D). This finding confirmed that the shortened root in mos1 was due to the loss of function of MOS1.

Expression of MOS1 is Induced by Sucrose
As mos1 showed hypersensitivity to sugars, to explore the role of MOS1 in sugar signaling pathways, the expression of the MOS1 gene in wild-type Col-0 under different concentrations of sucrose was investigated. Compared to the seedlings grown on normal medium, the expression level of MOS1 in seedlings grown on medium supplemented with 6% sucrose increased by 270% ( Figure  2A). To elucidate the expression pattern of MOS1 in response to exogenous sucrose, the transgenic plants pMOS1:GUS harboring the β-glucuronidase (GUS) reporter gene under the promoter and the first exon of MOS1 [39] were used. Histochemical analysis showed that the expression of GUS driven by the MOS1 promoter was related to the developmental stages of leaves and sucrose concentrations. When seedlings were grown on normal medium, strong GUS signals were detected in the emerging tissues, but only notably weak GUS signals were detected in the mature tissues. However, when seedlings were transferred to medium containing 6% sucrose, the intensity of the GUS signal was stronger, with obvious GUS signals being detected in the mature tissues ( Figure 2B). These results showed that the expression of MOS1 is promoted by exogenous sucrose.

MOS1 Affects the Expression of HKX1 in Response to Sugar
According to the qRT-PCR results, the expression of seven genes in several well-established sugar-response pathways was unchanged in both wild-type Col-0 and the mos1 mutant in response to sucrose, except for HKX1 and AKIN11 (Figure 3). The transcription of AKIN11 was downregulated by sucrose in both wild-type Col-0 and mos1 mutants, which made it difficult to determine whether the transcription change was associated with MOS1. However, the transcription of HKX1 was significantly upregulated by sucrose in the mos1 mutant and unchanged in wild-type Col-1 ( Figure  3). This finding suggested that the mos1 mutation may influence the HKX1-dependent sugar response pathway.

MOS1 Affects the Expression of HKX1 in Response to Sugar
According to the qRT-PCR results, the expression of seven genes in several well-established sugar-response pathways was unchanged in both wild-type Col-0 and the mos1 mutant in response to sucrose, except for HKX1 and AKIN11 (Figure 3). The transcription of AKIN11 was downregulated by sucrose in both wild-type Col-0 and mos1 mutants, which made it difficult to determine whether the transcription change was associated with MOS1. However, the transcription of HKX1 was significantly upregulated by sucrose in the mos1 mutant and unchanged in wild-type Col-1 ( Figure 3). This finding suggested that the mos1 mutation may influence the HKX1-dependent sugar response pathway. qRT-PCR analysis of the expression of genes involved in sugar-response pathways in Col-0 and mos1 grown on half MS at 3 hours after 6% sucrose (Suc+) or 6% mannitol (Suc-) treatment. Quantification was normalized to ACTIN2. Error bars indicate SE of two independent biological replicates. The asterisks indicate a significant difference between Suc+ and Suc-in each genotype (Student's t-test p < 0.001).

MOS1 Represses Anthocyanin Biosynthesis Induced by Sugar and High-Light Stress
When seedlings were grown on normal medium, mutant mos1 had comparable anthocyanin content to wild-type Col-0. However, when seedlings were grown on medium supplemented with 3% glucose or 6% sucrose, mos1 accumulated 2.5-and 2-fold more anthocyanin than wild-type Col-0, respectively ( Figure 4). In addition, complementation lines #1 and #2-9 accumulated similar amounts of anthocyanin pigments to wild-type Col-0 under all conditions (Figure 4). qRT-PCR analysis of the expression of genes involved in sugar-response pathways in Col-0 and mos1 grown on half MS at 3 hours after 6% sucrose (Suc+) or 6% mannitol (Suc-) treatment. Quantification was normalized to ACTIN2. Error bars indicate SE of two independent biological replicates. The asterisks indicate a significant difference between Suc+ and Suc-in each genotype (Student's t-test p < 0.001).

MOS1 Represses Anthocyanin Biosynthesis Induced by Sugar and High-Light Stress
When seedlings were grown on normal medium, mutant mos1 had comparable anthocyanin content to wild-type Col-0. However, when seedlings were grown on medium supplemented with 3% glucose or 6% sucrose, mos1 accumulated 2.5-and 2-fold more anthocyanin than wild-type Col-0, respectively ( Figure 4). In addition, complementation lines #1 and #2-9 accumulated similar amounts of anthocyanin pigments to wild-type Col-0 under all conditions ( Figure 4).
Additionally, the accumulation of anthocyanin in response to high light in wild-type Col-0 and mos1, #1, and #2-9 were analyzed concurrently. As shown in Figure 5, the anthocyanin content in mos1 was similar to that in wild-type Col-0 under normal conditions. After high-light treatment, mos1 accumulated anthocyanin pigments three times those in wild-type Col-0, while the contents of anthocyanin pigments in complementation lines #1 and #2-9 were similar to those in wild-type Col-0 ( Figure 5). This finding indicated that MOS1 could repress anthocyanin biosynthesis induced by sucrose and high-light stress, although the mechanisms governing the effect warrant further analysis.

MOS1 Affects the Expression of Genes Related to Anthocyanin Biosynthesis in Response to Sugar and High Light
To discover the molecular regulatory mechanisms of MOS1 on anthocyanin accumulation, the transcription of six early ABGs (PAL, C4H, CHS, CHI, F3H, F3 H), three late ABGs (DFR, ANS, LDOX), FLS, and two components of MBW complex, PAP and TT8, were analyzed. After treatment with 6% sucrose, the expression levels of PAL and F3H and F3 H in mos1 were 50%, 96%, and 109% higher than those in wild-type Col-0, respectively, and the expression levels of DFR, LDOX, and UF3GT in mos1 were 180%, 207%, and 124% higher than those in wild-type Col-0, respectively ( Figure 6A,B). However, the expression levels of DFR, LDOX, and UF3GT in mos1 treated with 6% mannitol were similar to those in wild-type Col-0 ( Figure 6A,B). Moreover, after sucrose treatment, the expression level of FLS was 60% lower in mos1 than in wild-type Col-0, although its expression was also lower upon treatment with 6% mannitol ( Figure 6C). Correspondingly, after treatment with 6% sucrose, the transcript levels of PAP1 and TT8 were 400% and 62% higher than those in wild-type Col-0, respectively, but there was no difference after treatment with 6% mannitol ( Figure 6D). mos1, #1, and #2-9 were analyzed concurrently. As shown in Figure 5, the anthocyanin content in mos1 was similar to that in wild-type Col-0 under normal conditions. After high-light treatment, mos1 accumulated anthocyanin pigments three times those in wild-type Col-0, while the contents of anthocyanin pigments in complementation lines #1 and #2-9 were similar to those in wild-type Col-0 ( Figure 5). This finding indicated that MOS1 could repress anthocyanin biosynthesis induced by sucrose and high-light stress, although the mechanisms governing the effect warrant further analysis.

MOS1 Affects the Expression of Genes Related to Anthocyanin Biosynthesis in Response to Sugar and High Light
To discover the molecular regulatory mechanisms of MOS1 on anthocyanin accumulation, the transcription of six early ABGs (PAL, C4H, CHS, CHI, F3H, F3′H), three late ABGs (DFR, ANS, LDOX), FLS, and two components of MBW complex, PAP and TT8, were analyzed. After treatment with 6% sucrose, the expression levels of PAL and F3H and F3′H in mos1 were 50%, 96%, and 109% higher than those in wild-type Col-0, respectively, and the expression levels of DFR, LDOX, and UF3GT in mos1 were 180%, 207%, and 124% higher than those in wild-type Col-0, respectively ( Figure 6A,B). However, the expression levels of DFR, LDOX, and UF3GT in mos1 treated with 6% mannitol were similar to those in wild-type Col-0 ( Figure 6A,B). Moreover, after sucrose treatment, the expression level of FLS was 60% lower in mos1 than in wild-type Col-0, although its expression was also lower upon treatment with 6% mannitol ( Figure 6C). Correspondingly, after treatment with 6% sucrose, the The expression of most ABGs in mos1 was equivalent to that in wild-type Col-0 under both normal and high-light conditions, except for CHS. The expression abundance of CHS after high-light treatment in mos1 was 54% higher than that in wild-type Col-0 ( Figure 7A,B). In addition, the expression of FLS in mos1 after high-light treatment was similar to that under normal light, while it was increased in wild-type Col-0 after high-light treatment ( Figure 7C). This finding indicates that MOS1 affects the expression of FLS. Moreover, the expression of PAP1 and TT8 was similar in the wild-type Col-0 and mos1 under both normal and high-light conditions ( Figure 7D). Although the responses of these genes to high light and sucrose were different in mos1, we can still conclude that MOS1 might regulate the accumulation of anthocyanin under sugar and light treatment by influencing the expression of some ABGs but through different regulatory mechanisms.

Discussion
Sugar signaling plays important roles in plant development and abiotic and biotic stress responses [7]. In this study, we found that the absence of the MOS1 gene function caused intense responses to sugars, as characterized by a reduced germination rate and shortened roots. Correspondingly, the expression of the glucose-responsive marker gene APL3 was increased (Figure 1), and MOS1 could respond to exogenous sucrose ( Figure 2). This finding indicated that MOS1 was a negative regulator of sugar responses and that there might be transcriptional feedback to control the responses within a certain range. The higher expression of HXK1 in response to sucrose in mos1 than in wild-type Col-0 ( Figure 3) suggested that MOS1 may influence sugar responses by regulating the transcriptional level of HXK1. HXKs have been identified as glucose sensors in many plant species, and recently, HXK1 was discovered to have multiple functions [9], i.e., promoting anthocyanin biosynthesis in apple by stabilizing a bHLH TF [49]. In agreement with this finding, mos1 accumulated more anthocyanin than wild-type Col-0 when exposed to exogenous sugars (Figure 4). transcript levels of PAP1 and TT8 were 400% and 62% higher than those in wild-type Col-0, respectively, but there was no difference after treatment with 6% mannitol ( Figure 6D). The expression of most ABGs in mos1 was equivalent to that in wild-type Col-0 under both normal and high-light conditions, except for CHS. The expression abundance of CHS after high-light Figure 6. Expression analysis of anthocyanin biosynthesis genes in response to sucrose. qRT-PCR analysis of the expression of early ABGs (A), late ABGs (B), FLS (C), and TFs (D) in 10-d-old Col-0 and mos1 after 6% mannitol (CTRL) or 6% sucrose (6%S) treatment. Quantification was normalized to ACTIN2. Error bars indicate SE of two independent biological replicates. The asterisks indicate significant differences compared with the corresponding Col-0 (one-way ANOVA/Bonferroni p < 0.001).
was increased in wild-type Col-0 after high-light treatment ( Figure 7C). This finding indicates that MOS1 affects the expression of FLS. Moreover, the expression of PAP1 and TT8 was similar in the wild-type Col-0 and mos1 under both normal and high-light conditions ( Figure 7D). Although the responses of these genes to high light and sucrose were different in mos1, we can still conclude that MOS1 might regulate the accumulation of anthocyanin under sugar and light treatment by influencing the expression of some ABGs but through different regulatory mechanisms.  In addition to sugars, anthocyanin biosynthesis is triggered by multiple stresses [2,5,27,50]. The overaccumulation of anthocyanin pigments in the mos1 mutant under sugar and high-light stress compared to wild-type Col-0 (Figures 4 and 5) indicated that MOS1 negatively regulates anthocyanin biosynthesis. Sugar activated the expression of several ABGs (Figure 6) [43], which was more pronounced in mos1 ( Figure 6). As mos1 exhibited increased sensitivity to sugars, the overaccumulation of anthocyanin pigments could be a consequence of the enhanced sugar response. Sugars also enhanced the expression of FLS [2,33] (Figure 6C); as a hypersensitive mutant, mos1 should have a higher expression of FLS. However, mos1 had a significantly lower expression level of FLS than did wild-type Col-0 ( Figure 6C), and the lower expression level of FLS led to dihydroflavonol accumulation as substrates for subsequent anthocyanin biosynthesis. This finding indicates that MOS1 has other mechanisms independent of sugar signals in regulating anthocyanin biosynthesis.
Moreover, the regulatory mechanisms of MOS1 in anthocyanin biosynthesis under sugar and light stresses might be different. Upon high-light treatment, only the expression of CHS in the mos1 mutant was higher than that in the wild-type, while under 6% sucrose, several ABGs but no CHS had different expression levels between the mos1 mutant and wild-type Col-0 (Figures 6 and 7). Moreover, the expression of FLS in CTRL under sugar treatment was different from that under light treatment (Figures 6 and 7), which might be due to the pretreatment in the dark before transport to medium containing different sugars compared to no pretreatment before transfer to chambers with different light intensities. Additionally, the expression of FLS was not induced by high-light treatment in mos1 ( Figure 7C). All these pieces of evidence indicate the specific function of MOS1 in the transcriptional regulation of FLS. A MOS1-interacting protein [40], TCP15, represses anthocyanin biosynthesis under high light [51], suggesting that TCP15 and MOS1 might also be involved in anthocyanin biosynthesis as well as immune responses. TCP15 affects the expression level of PAP1, TT8, and DFR under high light [51], while MOS1 showed no influences on PAP1, TT8, and DFR under high light (Figure 7). That might be because the regulation mechanisms of TCP15 and MOS1 on anthocyanin synthesis do not overlap completely, just like in immune responses [40]. Additionally, we used different sampling time points from Vialo et al. [51], while the effects of TCP15 had been found to be related to the irradiation time [51]. However, MOS1 showed negative regulations on the expression of PAP1, TT8, and DFR under 6% sucrose ( Figure 6). The expressions in the early part of the high-light treatment and the global gene expression changes with RNA-seq will be conducted in further studies, which will be beneficial for obtaining a better understanding of the regulation of MOS1 and the interactions of MOS1 and TCP15 on anthocyanin biosynthesis combined with genetic analysis.
Similar to anthocyanin biosynthesis, plant defense responses are affected by many factors, such as hormones, sugars, and light [52][53][54][55]. Recently, there has been increasing evidence supporting the contribution of sugar signals to plant immune responses. HXK1 plays positive roles in immune regulation, and the glucose phosphorylation capacity of HXK1 has been found to be essential for cell death and defense responses in the MIPS (myo-inositol 1-phosphate synthase) mutant [37]. MOS1 also plays positive roles in immunity, but the mos1 mutant has normal defense responses [37,39,40]. Thus, there is a possibility that the enhanced HXK1 expression in mos1 may be a compensation mechanism to maintain proper immune responses, and MOS1 might be the convergent regulator involved in the sugar-immunity regulation network.
Some studies also suggested that anthocyanin could take part in immunity in plants, but the precise underlying mechanism remains uncharacterized [56,57].
As MOS1 showed functions in anthocyanin accumulation (Figures 4 and 5), it may be worthwhile to identify convergent regulators in anthocyanin biosynthesis and immune response crosstalk, which may provide new insights into the coordinated network between immunity and other physiological processes.
To ensure that the plants grew normally, half-strength MS with 0.8% (w/v) sucrose was used as normal medium. For sugar treatment, 10-d-old seedlings grown on normal medium were transferred to the dark for 24 h to reduce intercellular sugar. After that step, the medium was replaced by half-strength MS medium with 3% (w/v) glucose, 6% (w/v) sucrose, 3% (w/v), and 6% (w/v) mannitol for an additional 3 h under light.
For the high-light treatment, plants were grown in soil under 150 µmol m −2 s −1 light (normal) for 14 d. Then, some plants were transferred to chambers with a light intensity of 450 mol m −2 s −1 (high light). Seedlings treated for 1 d were used for RNA isolation, and seedlings treated for 3 d were used to analyze the anthocyanin content.

Plasmid Construction and Generation of Transgenic Plants
A genomic fragment of the entire MOS1 coding region (without stop codon) and the 2680-bp sequence upstream of the ATG start codon were amplified by PCR from genomic DNA isolated from Col-0. The PCR product was cloned into the pDONR222 vector by BP reactions (Invitrogen, 11789020) and then cloned into the binary vector pGWB550 [58] to create pMOS1:MOS1:GFPCOM. The constructed vector was introduced into mos1 using Agrobacterium tumefaciens GV3101. Transgenic plants were selected on plates with hygromycin.

Germination Assay and Root Length Measurement
All seeds, harvested and stored identically, were sown on normal medium and medium containing 4% glucose or 4% mannitol. All plates were incubated at 4 • C for 2 d and then placed in a growth chamber for 7 d. The germination rate was scored by cotyledon greening. At least 50 seeds for each genotype were used for each independent biological repeat, and two repeats were conducted.
For root length measurement, seedlings were grown vertically on normal medium and medium supplemented with 2% glucose or 2% mannitol for 7 d. Images were captured by a digital camera, and the root lengths were calculated by ImageJ.

Measurement of Anthocyanin Content
Fresh seedlings grown on the indicated medium or after light treatment were used for measuring anthocyanin content. Leaf tissues of 20 mg were homogenized in 0.6 mL of methanol-HCl (1%, v/v) and then incubated at 4 • C for 1 d. After centrifugation, 0.4 mL chloroform and 0.4 mL ddH 2 O were added to the supernatant and vortexed vigorously. Then, the samples were centrifuged, and the absorbance of the supernatant was measured at 530 and 657 nm. Relative anthocyanin concentrations were calculated with the equation anthocyanin content = (A530-A657)/fresh weight (g).

RNA Extraction and Quantitative Real-Time PCR Analysis
Total RNA was isolated from plants with RNAiso Plus (Takara, Shiga, Japan, 9108), according to the manufacturer's instructions. cDNA was synthesized from 2 µg RNA by a PrimeSpcript TM RT reagent Kit with a gDNA Eraser Kit (Takara, Shiga, Japan, RR047). Quantitative RT-PCR was performed with a Bio-Rad CFX96™ Real-Time System (Bio-Rad, Hercules, USA) using TB Green Premix Ex Taq TM II (Tli RNaseH Plus; Takara, Shiga, Japan, RR820). Primers for RT-PCR are listed in Table S1. Two independent biological replicates were performed.

Conclusions
We provide evidence that the immune regulator MOS1 represses sugar responses and anthocyanin biosynthesis in Arabidopsis, possibly at the transcriptional level. Our findings highlight the involvement of MOS1 in sugar signaling. In the future, identifying MOS1 genetic interacting regulators and studying the regulation of MOS1 in sugar and hormone signaling may not only help to characterize the roles of MOS1 in specific biological processes but also elucidate the mechanism governing the balance of growth and defense.