TY - JOUR
AU - Amir, Rachel
AB - Abstract S-Methylmethionine (SMM) was suggested previously to participate in the metabolism of methionine (Met) in seeds. To further reveal its roles, we had previously produced transgenic Arabidopsis (Arabidopsis thaliana) RNA interference (RNAi) seeds with lower transcript expression of CYSTATHIONINE γ-SYNTHASE (AtCGS), Met’s main regulatory enzyme. Unexpectedly, these seeds accumulated significantly higher levels of Met compared with control seeds through an as yet unknown mechanism. Here, transcript and metabolic analyses coupled with isotope-labeled [13C]SMM and [13C]Met feeding experiments enabled us to reveal that SMM that was synthesized in rosette leaves of RNAi plants significantly contributed to the accumulation of Met in their seeds at late stages of development. Seed-specific repression of AtCGS in RNAi seeds triggered the induction of genes operating in the SMM cycle of rosette leaves, leading to elevated transport of SMM toward the seeds, where higher reconversion rates of SMM to Met were detected. The metabolic rearrangements in RNAi seeds resulted in an altered sulfur-associated metabolism, such as lower amounts of Cys and glutathione, as well as a differential composition of glucosinolates. Together, the data propose a novel cross talk existing between seeds and rosette leaves along with mutual effects between the Asp family and SMM pathways operating in these tissues. They also shed light on the effects of higher Met levels on seed physiology and behavior. Met is a nutritionally essential sulfur-containing amino acid found at low levels in plants and in their seeds and, thus, limits the nutritional value of crop plants as a source of dietary protein for humans and animals (Hesse et al., 2004; Galili et al., 2005; Amir, 2008; Galili and Amir, 2013). In plants, Met plays key roles in protein synthesis and mRNA translation and regulates indirectly a variety of cellular processes through its main catabolic product S-adenosyl-MET (SAM; Fig. 1A). SAM serves as the precursor for the synthesis of the plant hormone ethylene, polyamines, and biotin and donates a primary methyl group that is essential for methylation reactions involved in a variety of developmental processes in plant cells (Fig. 1A; Roje, 2006; Sauter et al., 2013). Figure 1. Open in new tabDownload slide Met and S-methylmethionine (SMM) metabolism in plants. A, The metabolic pathways involved in the metabolism of Met and SMM in vegetative tissues and seeds. Key enzymes involving Asp family and SMM pathways are presented in italics. Solid arrows represent one metabolic step, while dashed arrows represent several metabolic steps. B, The SMM cycle. ATP, Adenosine triphosphate; CGS, cystathionine γ-synthase; HC, homocysteine; HMTs, homocysteine S-methyltransferases; MMT, met S-methyltransferase; SAHC, S-adenosylhomocysteine; SAMS, S-adenosyl-Met synthase. Figure 1. Open in new tabDownload slide Met and S-methylmethionine (SMM) metabolism in plants. A, The metabolic pathways involved in the metabolism of Met and SMM in vegetative tissues and seeds. Key enzymes involving Asp family and SMM pathways are presented in italics. Solid arrows represent one metabolic step, while dashed arrows represent several metabolic steps. B, The SMM cycle. ATP, Adenosine triphosphate; CGS, cystathionine γ-synthase; HC, homocysteine; HMTs, homocysteine S-methyltransferases; MMT, met S-methyltransferase; SAHC, S-adenosylhomocysteine; SAMS, S-adenosyl-Met synthase. Genetic and biochemical studies suggest that, in seeds, Met can be synthesized de novo as in vegetative tissues via the classical Asp family pathway by the activity of its main regulatory enzyme, CGS (Hanafy et al., 2013; Cohen and Amir, 2017). CGS combines the carbon/amino skeleton derived from Asp with the sulfur moiety donated by Cys (Kim and Leustek, 2000; Fig. 1A). Indeed, recent studies showed that seed-specific expression of a feedback-insensitive form of Arabidopsis (Arabidopsis thaliana) CGS (AtCGS) led to higher Met contents in seeds of four soybean (Glycine max) cultivars (Hanafy et al., 2013; Song et al., 2013; Cohen et al., 2016b), tobacco (Nicotiana tabacum; Matityahu et al., 2013), and Arabidopsis (Cohen et al., 2014). However, isotope-labeling experiments suggest that Met can be synthesized in seeds through an additional and/or alternative pathway by which Met produced in vegetative tissues is converted to SMM that is then transported via the phloem into the reproductive tissues (Fig. 1A). The first observation supporting this possibility was reported by Bourgis et al. (1999), who labeled wheat (Triticum aestivum) leaves with [13C]Met and demonstrated that [13C]SMM was the major sulfur compound transported through the phloem toward the reproductive tissues. Later, Lee et al. (2008) revealed that Arabidopsis seeds can reconvert transported SMM back to soluble Met. Two types of methyltransferases operate in the SMM cycle: MET S-METHYLTRANSFERASE (MMT), which uses Met and SAM to form SMM and S-adenosylhomocysteine; and HOMOCYSTEINE S-METHYLTRANSFERASEs (HMTs), which use SMM together with homocysteine to form two Met molecules (Fig. 1B; Giovanelli, 1987). Although it is clear how plants synthesize SMM and reconvert it back to Met, the physiological roles of SMM and its cycle remain mostly undefined, mainly since the activities of MMT and HMTs together have been proposed to constitute a futile cycle. Each turn of this cycle consumes and then regenerates two Met molecules while converting ATP to adenosine, PPi and Pi (Fig. 1B). Proposed general roles for SMM and its cycle include Met storage (Giovanelli, 1987; Mudd and Datko, 1990), short-term control of SAM levels (Ranocha et al., 2001), donation of methyl groups to other plant metabolites by SMM (Giovanelli, 1990), maintaining methylation ratios (Kocsis et al., 2003), and repairing SAM that suffers from chemical damage reactions under physiological conditions (Bradbury et al., 2014). Nevertheless, the contribution of transported SMM from nonseed tissues to Met accumulation in seeds is still indistinct, as the lack of SMM in Arabidopsis mmt mutant did not alter Met, thiol, and sulfur contents in seeds (Kocsis et al., 2003). Additionally, transgenic tobacco plants accumulating 176- and 39-fold higher Met and SMM contents in their leaves, respectively, exhibited only minor accumulation of Met in their seeds (Hacham et al., 2008). These lines of evidence shed light on the complexity of Met synthesis in seeds, emphasizing the prerequisite to further explore the relative contribution of Asp and SMM pathways to Met synthesis in seeds and their metabolic regulatory interactions during seed development. Thus, to gain more knowledge of the contribution of SMM to Met synthesis in seeds, we recently generated transgenic Arabidopsis seeds with lower transcript expression of AtCGS (Cohen and Amir, 2017). An RNA interference (RNAi) construct controlled by the seed-specific phaseolin promoter was used to regulate the gene during late stages of seed development (Fait et al., 2011). Unexpectedly, gas chromatography-mass spectrometry (GC-MS) measurements indicated that seeds produced by the transgenic RNAi lines accumulated up to 2.5-fold more soluble Met compared with wild-type seeds and seeds expressing an empty vector (EV; Cohen and Amir, 2017). This implied that other regulatory factors were triggered in these transgenic plants to support the synthesis and accumulation of Met. In this study, we further characterized RNAi lines by transcript and metabolic analyses coupled with isotope-labeled [13C]SMM and [13C]Met feeding experiments. These revealed new insights into the regulatory roles of SMM and the relative contribution of the Asp family and SMM pathways to Met synthesis in seeds. RESULTS Protein Degradation and/or Decreased Catabolism of Met Could Not Explain the Higher Soluble Met Found in RNAi Seeds Seeds produced by RNAi 1 and 2 lines were shown to accumulate 1.5- and 2.5-fold more soluble Met, respectively, and up to 1.8-fold total soluble amino acid contents compared with wild-type/EV seeds (Cohen and Amir, 2017). Overaccumulation of Met in RNAi seeds might result from increased activity of the SMM cycle to compensate for the lower expression level of AtCGS but also from released protein-bound Met and other amino acids due to protein degradation and/or difficulties in protein synthesis. To examine the latter two possibilities, protein extracts from RNAi seeds were hydrolyzed and the levels of protein-bound amino acids were evaluated by GC-MS. The results indicated insignificant 5% and significant 18% increases in protein-bound Met contents in RNAi 1 and 2 seeds, respectively, compared with wild-type/EV seeds (Fig. 2A). The higher protein-bound Met was associated with 1.3- and 1.5-fold increases in overall total amino acid contents in RNAi seeds, respectively (Fig. 2B), in accordance with previous reports (Matityahu et al., 2013; Cohen et al., 2014, 2016b; Cohen and Amir, 2017). The higher levels of Met and other amino acids in these seeds are apparently due to higher synthesis of several 12S-globulins and 2S-albumin subunits, which are the major seed-storage proteins, as we reported previously for Arabidopsis seeds expressing AtCGS (Cohen et al., 2016a). Despite these lines of evidence, we cannot rule out the possibility that proteins also were degraded, albeit at lower rates than their synthesis, and thus kinetic measurements should be made to examine this notion further. Figure 2. Open in new tabDownload slide Protein degradation and/or decreased catabolism of Met could not explain the higher soluble Met found in RNAi seeds. A and B, GC-MS analysis of protein-bound Met (A) and total protein-bound amino acid content (B) after protein hydrolysis. Levels were normalized according to the nor-Leu internal standard. Data shown are means ± se of seven replicates. FW, Fresh weight; WT, wild type. C and D, qRT-PCR analysis of AtSAMS3 (C) and AtMGL (D) relative expression levels. Expression levels were normalized according to the AtPP2A-3 constitutive gene. Data shown are means ± se of three replicates. Significance was calculated using the two-way ANOVA test of P < 0.05 and identified by different letters. Figure 2. Open in new tabDownload slide Protein degradation and/or decreased catabolism of Met could not explain the higher soluble Met found in RNAi seeds. A and B, GC-MS analysis of protein-bound Met (A) and total protein-bound amino acid content (B) after protein hydrolysis. Levels were normalized according to the nor-Leu internal standard. Data shown are means ± se of seven replicates. FW, Fresh weight; WT, wild type. C and D, qRT-PCR analysis of AtSAMS3 (C) and AtMGL (D) relative expression levels. Expression levels were normalized according to the AtPP2A-3 constitutive gene. Data shown are means ± se of three replicates. Significance was calculated using the two-way ANOVA test of P < 0.05 and identified by different letters. An additional option for gaining higher levels of Met in RNAi seeds might be lower catabolism. To this end, we performed quantitative real-time (qRT)-PCR and measured the expression level of the two predominant catabolic enzymes of Met, S-ADENOSYL-MET SYNTHASE3 (AtSAMS3) and MET γ-LYASE (AtMGL; Huang et al., 2014). Both genes exhibited up to 4-fold higher expression in the transgenic lines compared with wild-type/EV seeds (Fig. 2, C and D), further suggesting that the overaccumulation of Met in these seeds is not due to lower catabolism. RNAi Seeds Exhibit a Greater Ability to Reconvert Transported SMM Back to Met The above results reinforced a possible role of SMM in Met metabolism in RNAi seeds. To test this assumption, we employed qRT-PCR and revealed that RNAi 1 and 2 seeds raised the expression levels of AtHMT1 by 3.4- and 8.5-fold and those of AtHMT3 by 5.4- and 7.3-fold, respectively. Markedly, no changes were detected in the expression of AtHMT2 (Fig. 3A); likewise, a previous report demonstrated that AtHMT2 plays an insignificant role in seeds but an important role in nonseed tissues (Lee et al., 2008). Apart from AtHMTs, AtMMT also exhibited 2.2- and 4.9-fold increased expression in RNAi seeds (Fig. 3A), respectively, suggesting altered SMM levels and higher activity of the SMM cycle in RNAi seeds. Correspondingly, high-performance liquid chromatography (HPLC) analyses demonstrated significant 2- and 2.2-fold increased SMM levels in RNAi 1 and 2 seeds, respectively (Fig. 3B). Figure 3. Open in new tabDownload slide RNAi seeds have a greater ability to reconvert SMM to Met and higher SMM contents. A, qRT-PCR analysis of AtHMT and AtMMT relative expression levels in mature seeds. Expression levels were normalized according to the AtPP2A-3 constitutive gene. Data shown are means ± se of three replicates. B, HPLC analysis of SMM levels in mature seeds. Levels were normalized according to the nor-Leu internal standard. Data shown are means ± se of five replicates. FW, Fresh weight; WT, wild type. C, GC-MS analysis of [13C]Met in developing seeds fed with [13C]SMM. [13C]Met levels were normalized according to the nor-Leu internal standard. Data shown are means ± se of four replicates. Significance was calculated using the two-way ANOVA test of P < 0.05 and identified by different letters. Figure 3. Open in new tabDownload slide RNAi seeds have a greater ability to reconvert SMM to Met and higher SMM contents. A, qRT-PCR analysis of AtHMT and AtMMT relative expression levels in mature seeds. Expression levels were normalized according to the AtPP2A-3 constitutive gene. Data shown are means ± se of three replicates. B, HPLC analysis of SMM levels in mature seeds. Levels were normalized according to the nor-Leu internal standard. Data shown are means ± se of five replicates. FW, Fresh weight; WT, wild type. C, GC-MS analysis of [13C]Met in developing seeds fed with [13C]SMM. [13C]Met levels were normalized according to the nor-Leu internal standard. Data shown are means ± se of four replicates. Significance was calculated using the two-way ANOVA test of P < 0.05 and identified by different letters. To investigate further the likelihood that RNAi seeds facilitate higher import and conversion of SMM, we designed a feeding experiment where inflorescences from 44-d-old wild-type and RNAi plants were cut just above the rosette leaves and placed in tubes filled with 20 mL of feeding medium containing 2 mm isotope-labeled [13C]SMM in double-distilled water (DDW; Supplemental Fig. S1A). Following 36 h of feeding, silique hulls were harvested, and the developing seeds were isolated and profiled for [13C]Met levels by GC-MS. The [13C]Met could be differentiated from endogenous soluble Met due to the 13C labeling on its α-carbon, leading to shifts in two mass fragments, 219 mass-to-charge ratio (m/z) and 293 m/z in nonlabeled Met, to 220 m/z and 294 m/z in [13C]Met. Ratio calculations were based on the second shift, as it showed predominance over the first shift according to GC-MS validation (Supplemental Fig. S2). RNAi 1 seeds exhibited a 3.7% insignificant increase in 294 m/z to 293 m/z compared with wild-type developing seeds; however, a significant 12.3% increased ratio was detected in RNAi 2 seeds that have a lower AtCGS expression level and higher Met content (Cohen and Amir, 2017). Such differences were absent in seeds from both genotypes when they were fed with DDW (Fig. 3C). The 294 m/z fragment could be obtained solely from supplemented [13C]SMM; thus, the results implied that RNAi 2 seeds efflux more SMM from stems than RNAi 1 and wild-type seeds and reconvert it back to soluble Met at higher rates. This was further strengthened by the observation that a higher level of [13C]SMM in the medium of wild-type inflorescences did not significantly increase the level of [13C]Met in seeds (Fig. 3C). Vegetative and Reproductive Tissues of RNAi Plants Exhibit Differential Accumulation of SMM and Met and Altered Expression of Genes Involved in Their Metabolism Lee et al. (2008) suggested that the upper flower stalk contributes SMM to the seeds. To gain more knowledge about the source of SMM, we collected rosette leaves, stems, flower stalks, and silique hulls from 44-d-old wild-type/EV and RNAi plants and analyzed AtCGS, AtMMT, and AtHMT expression levels as well as the contents of SMM and Met. The expression level of AtCGS in RNAi 2 rosette leaves increased 1.56-fold (Table I), implying that the lower AtCGS expression in seeds triggered the biosynthesis pathway of Met in rosette leaves. Yet, Met levels in these leaves were similar to those measured in RNAi 1 and wild-type/EV rosette leaves (Table I), suggesting that Met was converted to SMM. This assumption was further supported by the significant 1.6- and 2-fold increased expression levels of AtMMT in rosette leaves from both RNAi 1 and 2 lines, respectively (Table I). We previously found that AtMMT and AtHMTs are both expressed in all plant tissues in Arabidopsis (Frank et al., 2015). Therefore, the expression levels of AtHMTs were measured to reveal increased expression of AtHMT2 and AtHMT3 in RNAi 2 by 2- and 1.8-fold, respectively, while no changes were detected in the expression of AtHMT1 (Table I). Despite the induced expression of AtCGS and two AtHMTs in RNAi 2 rosette leaves and the induced expression of AtMMT in rosette leaves of both RNAi plants, these tissues accumulated 24% and 29% less SMM than wild-type/EV leaves, respectively (Table I). We assume that the lower amounts of SMM are due to higher transportation to the reproductive tissues. Levels of SMM and Met and expression patterns of genes involved in the SMM cycle and the Asp family pathway in rosette leaves, stems, flower stalks, and silique hulls of RNAi compared with wild-type/EV plants Table I. Levels of SMM and Met and expression patterns of genes involved in the SMM cycle and the Asp family pathway in rosette leaves, stems, flower stalks, and silique hulls of RNAi compared with wild-type/EV plants Metabolitea/Geneb . RNAi 1 . RNAi 2 . Rosette Leaves . Stems . Flower Stalks . Silique Hulls . Rosette Leaves . Stems . Flower Stalks . Silique Hulls . SMM 0.76 c 1.34 1.43 1.35 0.71 2.10 3.40 1.92 Met 0.93 0.97 0.95 1.03 0.91 1.03 1.93 2.40 AtCGS 1.03 1.06 0.92 1.70 1.56 0.91 1.08 1.62 AtMMT 1.64 1.03 0.78 0.66 2.00 1.13 1.01 0.73 AtHMT1 0.96 1.02 0.45 0.97 0.99 1.23 0.41 1.02 AtHMT2 0.94 0.94 0.87 0.58 1.98 1.08 1.11 0.39 AtHMT3 0.88 0.88 0.95 0.74 1.84 1.14 0.87 0.64 Metabolitea/Geneb . RNAi 1 . RNAi 2 . Rosette Leaves . Stems . Flower Stalks . Silique Hulls . Rosette Leaves . Stems . Flower Stalks . Silique Hulls . SMM 0.76 c 1.34 1.43 1.35 0.71 2.10 3.40 1.92 Met 0.93 0.97 0.95 1.03 0.91 1.03 1.93 2.40 AtCGS 1.03 1.06 0.92 1.70 1.56 0.91 1.08 1.62 AtMMT 1.64 1.03 0.78 0.66 2.00 1.13 1.01 0.73 AtHMT1 0.96 1.02 0.45 0.97 0.99 1.23 0.41 1.02 AtHMT2 0.94 0.94 0.87 0.58 1.98 1.08 1.11 0.39 AtHMT3 0.88 0.88 0.95 0.74 1.84 1.14 0.87 0.64 a SMM and Met levels were measured by HPLC. Data shown are means ± se of five replicates.   b Relative expression of genes was measured by qRT-PCR and normalized according to the constitutive AtPP2A-3 gene. Data shown are means ± se of three replicates.   c The relative level/expression of a metabolite/gene compared with the wild type/EV. Boldface values represent significant changes compared with the wild type according to Student’s t test at P < 0.05. Open in new tab Table I. Levels of SMM and Met and expression patterns of genes involved in the SMM cycle and the Asp family pathway in rosette leaves, stems, flower stalks, and silique hulls of RNAi compared with wild-type/EV plants Metabolitea/Geneb . RNAi 1 . RNAi 2 . Rosette Leaves . Stems . Flower Stalks . Silique Hulls . Rosette Leaves . Stems . Flower Stalks . Silique Hulls . SMM 0.76 c 1.34 1.43 1.35 0.71 2.10 3.40 1.92 Met 0.93 0.97 0.95 1.03 0.91 1.03 1.93 2.40 AtCGS 1.03 1.06 0.92 1.70 1.56 0.91 1.08 1.62 AtMMT 1.64 1.03 0.78 0.66 2.00 1.13 1.01 0.73 AtHMT1 0.96 1.02 0.45 0.97 0.99 1.23 0.41 1.02 AtHMT2 0.94 0.94 0.87 0.58 1.98 1.08 1.11 0.39 AtHMT3 0.88 0.88 0.95 0.74 1.84 1.14 0.87 0.64 Metabolitea/Geneb . RNAi 1 . RNAi 2 . Rosette Leaves . Stems . Flower Stalks . Silique Hulls . Rosette Leaves . Stems . Flower Stalks . Silique Hulls . SMM 0.76 c 1.34 1.43 1.35 0.71 2.10 3.40 1.92 Met 0.93 0.97 0.95 1.03 0.91 1.03 1.93 2.40 AtCGS 1.03 1.06 0.92 1.70 1.56 0.91 1.08 1.62 AtMMT 1.64 1.03 0.78 0.66 2.00 1.13 1.01 0.73 AtHMT1 0.96 1.02 0.45 0.97 0.99 1.23 0.41 1.02 AtHMT2 0.94 0.94 0.87 0.58 1.98 1.08 1.11 0.39 AtHMT3 0.88 0.88 0.95 0.74 1.84 1.14 0.87 0.64 a SMM and Met levels were measured by HPLC. Data shown are means ± se of five replicates.   b Relative expression of genes was measured by qRT-PCR and normalized according to the constitutive AtPP2A-3 gene. Data shown are means ± se of three replicates.   c The relative level/expression of a metabolite/gene compared with the wild type/EV. Boldface values represent significant changes compared with the wild type according to Student’s t test at P < 0.05. Open in new tab Comparable measurements were performed on stems of both RNAi plants to reveal that the content of SMM increased 1.3- and 2.1-fold, respectively, but similar Met levels were detected as in wild-type/EV stems (Table I). Remarkably, the expression levels of all investigated genes were similar in the stems of all tested lines (Table I), suggesting that SMM, and much less Met, was transported through the stems of RNAi plants. SMM was most likely transported through the stems toward the upper flower stalks of RNAi 1 and 2, as these accumulated 1.4- and 3.4-fold more SMM than wild-type/EV plants, respectively (Table I). The level of Met increased 1.9-fold compared with wild-type/EV flower stalks only in RNAi 2 plants (Table I). Despite changes in the levels of these metabolites, AtHMT1 was the only gene exhibiting altered expression, being 55% and 59% lower compared with wild-type/EV flower stalks, respectively (Table I). Thus, as indicated for the stems, we proposed that the higher SMM levels found in the flower stalks of RNAi plants are the result of elevated transport of SMM and less de novo synthesis in these tissues. Lastly, silique hulls from RNAi 1 plants also accumulated 1.4-fold higher SMM levels but similar Met contents as wild-type/EV siliques, but silique hulls from RNAi 2 plants accumulated 1.9- and 2.4-fold more SMM and Met, respectively, compared with wild-type/EV siliques (Table I). Significantly lower expression of AtHMT2 and AtHMT3 was detected in silique hulls from both RNAi plants, providing a partial explanation for the higher SMM contents in these tissues (Table I). Notably, AtCGS was expressed 1.6-fold higher in both RNAi silique hulls compared with wild-type/EV silique hulls, implying that Met synthesis through the Asp family pathway increased significantly in this tissue (Table I). Exogenous Feeding of [13C]Met into RNAi Rosette Leaves Provides Physiological Evidence of Higher SMM Efflux toward Their Developing Seeds To reveal the physiological mechanism by which the SMM cycle supports the synthesis of Met in RNAi seeds, a second labeling experiment was designed where [13C]Met was applied to fully expanded mature rosette leaves of 44-d-old wild-type and RNAi flowering plants, and vegetative and reproductive tissues were harvested 4, 8, and 24 h after labeling (Supplemental Fig. S1B). Tissue extracts were analyzed by ultra-performance liquid chromatography equipped with electrospray ionization and mass spectrometry detection (UPLC-TQ-MS/MS) and specific multiple reaction monitoring (MRM) transitions. We calculated the ratios between the peak area of [13C]Met and the peak area of nonlabeled Met in the samples and subtracted the ratio calculated in control-treated samples fed with DDW. Similar manipulations were made to determine [13C]SMM ratios. Generally, the highest abundances of [13C]Met and [13C]SMM were measured in labeled rosette leaves from both wild-type and RNAi plants, as expected (Fig. 4). Both 13C-labeled compounds decreased constantly from 4 to 24 h after labeling, implying that [13C]Met successfully penetrated into the leaves and was converted to [13C]SMM. Yet, we cannot exclude the possibility that it was also converted to other Met-related metabolites. The constant decrease of [13C]SMM content in labeled rosette leaves further suggests that it was transported from these tissues to other vegetative and reproductive tissues. The observation that [13C]SMM was detected to some extent in these tissues 24 h after labeling may suggest that [13C]Met was still converted to [13C]SMM at this time point and/or that the flux toward other tissues remained, even after 24 h from labeling (Fig. 4). Figure 4. Open in new tabDownload slide Exogenous feeding of [13C]Met into RNAi rosette leaves provides physiological evidence of higher SMM efflux toward their developing seeds. A schematic representation of an isotope-labeling feeding experiment appears in Supplemental Figure S3B, while ultra-performance liquid chromatography (UPLC) conditions and MRM parameters are listed in Supplemental Tables S2 and S3. Data shown are means ± se of three replicates. Significance was calculated using the Student’s t test of P < 0.05 and identified by asterisks. WT, Wild type. Figure 4. Open in new tabDownload slide Exogenous feeding of [13C]Met into RNAi rosette leaves provides physiological evidence of higher SMM efflux toward their developing seeds. A schematic representation of an isotope-labeling feeding experiment appears in Supplemental Figure S3B, while ultra-performance liquid chromatography (UPLC) conditions and MRM parameters are listed in Supplemental Tables S2 and S3. Data shown are means ± se of three replicates. Significance was calculated using the Student’s t test of P < 0.05 and identified by asterisks. WT, Wild type. [13C]Met was significantly lower at all time points in RNAi 2-labeled rosette leaves, implying that it was converted at higher rates to [13C]SMM in correlation with the higher AtMMT expression observed in these tissues. Yet, significantly lower levels of [13C]SMM were measured in labeled rosette leaves of both RNAi plants, implying higher transport into other vegetative and reproductive tissues. Markedly, both 13C-labeled compounds were apparently able to transport from the feeding sites to the adjacent nonlabeled rosette leaves, mainly up to 4 h after labeling in RNAi 2 plants (Fig. 4). Similar [13C]Met ratios were detected in the stems among all three genotypes, further suggesting that [13C]SMM and not [13C]Met was transported from the feeding sites. Indeed, stems of RNAi 2 plants accumulated much higher levels of [13C]SMM compared with those of RNAi 1 and wild-type plants (Fig. 4). These differences, although significant, became much less distinct 24 h after labeling, suggesting that, at this time, the majority of [13C]Met was already converted to [13C]SMM that was transported into the reproductive tissues through the stems. As in the stems, we could not detect any differences in [13C]Met levels in the flower stalks of the three genotypes, but significantly higher levels of [13C]SMM in RNAi 2 flower stalks and silique hulls were detected 4 and 8 h after labeling (Fig. 4). These observations provide additional evidence that [13C]SMM is the transported compound. Interestingly, at all time points, we could detect higher [13C]Met levels in RNAi 2 silique hulls; thus, we could not rule out the possibility that some of the transported [13C]SMM was reconverted back to [13C]Met by the activity of the SMM cycle in the silique hulls (Fig. 4). Finally, RNAi 2 developing seeds exhibited much higher levels of [13C]Met compared with their RNAi 1 and wild-type counterparts (Fig. 4), correlating with the observations of our first isotope-labeling experiment showing the ability of these seeds to reconvert transported [13C]SMM back to [13C]Met at higher rates and with higher expression of AtHMT1 and AtHMT3 measured in these seeds. [13C]SMM was higher in RNAi 2 developing seeds 4 h after labeling but significantly lower 8 and 24 h after labeling (Fig. 4). This implies that more [13C]SMM reached these tissues after labeling that was then converted back to [13C]Met by AtHMTs operating in the SMM cycle in seeds. Met and SMM Overaccumulation in RNAi Seeds Affects the Metabolism of Other Sulfur-Associated Metabolites Cys is the sulfur moiety donor for Met biosynthesis (Fig. 1A) as well as the precursor for the important antioxidant reagent of glutathione (GSH; Kopriva, 2006). Thus, we were interested in assessing whether the overaccumulation of Met and SMM in RNAi seeds affected the levels of these two compounds. HPLC analyses indicated a significant 33% reduced level of Cys in RNAi 2 seeds together with a corresponding significant 20% reduction in GSH levels (Fig. 5A). Figure 5. Open in new tabDownload slide Met and SMM overaccumulation in RNAi seeds affects the metabolism of other sulfur-associated metabolites. A, HPLC analysis of Cys and GSH levels in mature seeds. Levels were normalized according to the nor-Leu internal standard. Data shown are means ± se of five replicates. B, UPLC-qTOF-MS/MS analysis of glucosinolate levels in mature seeds. The inset shows a representative chromatogram of the peaks of the nine corresponding glucosinolates positively identified in Table II. Data shown are means ± se of five replicates. Significance was calculated using the two-way ANOVA test of P < 0.05 and identified by different letters. FW, Fresh weight; WT, wild type. Figure 5. Open in new tabDownload slide Met and SMM overaccumulation in RNAi seeds affects the metabolism of other sulfur-associated metabolites. A, HPLC analysis of Cys and GSH levels in mature seeds. Levels were normalized according to the nor-Leu internal standard. Data shown are means ± se of five replicates. B, UPLC-qTOF-MS/MS analysis of glucosinolate levels in mature seeds. The inset shows a representative chromatogram of the peaks of the nine corresponding glucosinolates positively identified in Table II. Data shown are means ± se of five replicates. Significance was calculated using the two-way ANOVA test of P < 0.05 and identified by different letters. FW, Fresh weight; WT, wild type. Met is the precursor for the synthesis of various metabolites. Among them are the Met-derived glucosinolates, which are plant sulfonated thioglucosides originating principally in Arabidopsis from Phe, Trp, and chain-elongated Met derivatives (Reintanz et al., 2001). Using ultra-performance liquid chromatography equipped with quadrupole time of flight mass spectrometry detection (UPLC-qTOF-MS/MS), we were able to detect and analyze the levels of nine glucosinolates in the extracts of wild-type and RNAi mature seeds (Fig. 5B; Table II). These include four short-chain Met derivatives (4MTB, 4-methylthiobutyl glucosinolate; 5MTP, 5-methylthiopentyl glucosinolate; 3BZO, 3-benzoyloxypropyl glucosinolate; and 4BZO, 4-benzoyloxybutyl glucosinolate), four long-chain Met derivatives (7MSOH, 7-methylsulfinylheptyl glucosinolate; 8MSOO, 8-methylsulfinyloctyl glucosinolate; 7MTH, 7-methylthioheptyl glucosinolate; and 8MTO, 8-methylthiooctyl glucosinolate), and one Trp derivative (I3M, indol-3-yl-methyl glucosinolate). Glucosinolates detected in mature dry seeds following UPLC-qTOF-MS/MS analysis Table II. Glucosinolates detected in mature dry seeds following UPLC-qTOF-MS/MS analysis Metabolite No.a . Metabolite Name . Abbreviation . Glucosinolate Type . Retention Time, RT (min) . Molecular Formula . Molecular weight, Mw (m/z [M-H])b . 1 7-Methylsulfinylheptyl glucosinolate 7MSOH Long-chain Met derivative 3.47 C15H29NO10S3 478.0875 2 4-Methylthiobutyl glucosinolate 4MTB Short-chain Met derivative 3.69 C12H23NO9S4 420.0454 3 Indol-3-yl-methyl glucosinolate I3M Trp derivative 4.46 C16H20N2O9S2 447.0532 4 8-Methylsulfinyloctyl glucosinolate 8MSOO Long-chain Met derivative 5.33 C16H31NO10S3 492.1032 5 5-Methylthiopentyl glucosinolate 5MTP Short-chain Met derivative 5.82 C13H25NO9S3 434.0613 6 3-Benzoyloxypropyl glucosinolate 3BZO Short-chain Met derivative 8.87 C17H23NO11S2 480.0634 7 4-Benzoyloxybutyl glucosinolate 4BZO Short-chain Met derivative 11.25 C18H24NO11S2 494.0798 8 7-Methylthioheptyl glucosinolate 7MTH Long-chain Met derivative 12.34 C15H29NO9S3 462.0926 9 8-Methylthiooctyl glucosinolate 8MTO Long-chain Met derivative 15.92 C16H31NO9S3 476.1083 Metabolite No.a . Metabolite Name . Abbreviation . Glucosinolate Type . Retention Time, RT (min) . Molecular Formula . Molecular weight, Mw (m/z [M-H])b . 1 7-Methylsulfinylheptyl glucosinolate 7MSOH Long-chain Met derivative 3.47 C15H29NO10S3 478.0875 2 4-Methylthiobutyl glucosinolate 4MTB Short-chain Met derivative 3.69 C12H23NO9S4 420.0454 3 Indol-3-yl-methyl glucosinolate I3M Trp derivative 4.46 C16H20N2O9S2 447.0532 4 8-Methylsulfinyloctyl glucosinolate 8MSOO Long-chain Met derivative 5.33 C16H31NO10S3 492.1032 5 5-Methylthiopentyl glucosinolate 5MTP Short-chain Met derivative 5.82 C13H25NO9S3 434.0613 6 3-Benzoyloxypropyl glucosinolate 3BZO Short-chain Met derivative 8.87 C17H23NO11S2 480.0634 7 4-Benzoyloxybutyl glucosinolate 4BZO Short-chain Met derivative 11.25 C18H24NO11S2 494.0798 8 7-Methylthioheptyl glucosinolate 7MTH Long-chain Met derivative 12.34 C15H29NO9S3 462.0926 9 8-Methylthiooctyl glucosinolate 8MTO Long-chain Met derivative 15.92 C16H31NO9S3 476.1083 a Metabolite number corresponds to peaks appearing in Figure 5B.   b Molecular weight, Mw, detected following UPLC-qTOF-MS/MS analysis in the negative mode. Open in new tab Table II. Glucosinolates detected in mature dry seeds following UPLC-qTOF-MS/MS analysis Metabolite No.a . Metabolite Name . Abbreviation . Glucosinolate Type . Retention Time, RT (min) . Molecular Formula . Molecular weight, Mw (m/z [M-H])b . 1 7-Methylsulfinylheptyl glucosinolate 7MSOH Long-chain Met derivative 3.47 C15H29NO10S3 478.0875 2 4-Methylthiobutyl glucosinolate 4MTB Short-chain Met derivative 3.69 C12H23NO9S4 420.0454 3 Indol-3-yl-methyl glucosinolate I3M Trp derivative 4.46 C16H20N2O9S2 447.0532 4 8-Methylsulfinyloctyl glucosinolate 8MSOO Long-chain Met derivative 5.33 C16H31NO10S3 492.1032 5 5-Methylthiopentyl glucosinolate 5MTP Short-chain Met derivative 5.82 C13H25NO9S3 434.0613 6 3-Benzoyloxypropyl glucosinolate 3BZO Short-chain Met derivative 8.87 C17H23NO11S2 480.0634 7 4-Benzoyloxybutyl glucosinolate 4BZO Short-chain Met derivative 11.25 C18H24NO11S2 494.0798 8 7-Methylthioheptyl glucosinolate 7MTH Long-chain Met derivative 12.34 C15H29NO9S3 462.0926 9 8-Methylthiooctyl glucosinolate 8MTO Long-chain Met derivative 15.92 C16H31NO9S3 476.1083 Metabolite No.a . Metabolite Name . Abbreviation . Glucosinolate Type . Retention Time, RT (min) . Molecular Formula . Molecular weight, Mw (m/z [M-H])b . 1 7-Methylsulfinylheptyl glucosinolate 7MSOH Long-chain Met derivative 3.47 C15H29NO10S3 478.0875 2 4-Methylthiobutyl glucosinolate 4MTB Short-chain Met derivative 3.69 C12H23NO9S4 420.0454 3 Indol-3-yl-methyl glucosinolate I3M Trp derivative 4.46 C16H20N2O9S2 447.0532 4 8-Methylsulfinyloctyl glucosinolate 8MSOO Long-chain Met derivative 5.33 C16H31NO10S3 492.1032 5 5-Methylthiopentyl glucosinolate 5MTP Short-chain Met derivative 5.82 C13H25NO9S3 434.0613 6 3-Benzoyloxypropyl glucosinolate 3BZO Short-chain Met derivative 8.87 C17H23NO11S2 480.0634 7 4-Benzoyloxybutyl glucosinolate 4BZO Short-chain Met derivative 11.25 C18H24NO11S2 494.0798 8 7-Methylthioheptyl glucosinolate 7MTH Long-chain Met derivative 12.34 C15H29NO9S3 462.0926 9 8-Methylthiooctyl glucosinolate 8MTO Long-chain Met derivative 15.92 C16H31NO9S3 476.1083 a Metabolite number corresponds to peaks appearing in Figure 5B.   b Molecular weight, Mw, detected following UPLC-qTOF-MS/MS analysis in the negative mode. Open in new tab RNAi 2 seeds accumulated 7% more 4MTB, the most abundant glucosinolate, compared with RNAi 1 and wild-type seeds (Fig. 5B). These seeds also exhibited slightly higher levels of 3BZO and 7MTH but lower amounts of 8MSOO, 4BZO, and 8MTO. The most predominant reduction was detected in 4BZO, where RNAi 2 seeds accumulated 43% less of this metabolite compared with wild-type seeds (Fig. 5B). Apart from a 22% reduction in 4BZO levels, no significant changes were detected in the glucosinolate profile of RNAi 1 seeds compared with the wild-type seeds, further suggesting that the changes observed in sulfur-associated metabolites in RNAi seeds are affected directly by the levels of Met and SMM. The above changes encouraged us to search for altered expression of sulfur-associated genes in our microarray transcriptomic data set of RNAi::AtCGS seeds. Yet, we could not find any of these, despite changes in the metabolism of some sulfur-related metabolites mentioned above. Since we had previously found that the metabolic and transcriptomic effects of higher Met tightly depend on its levels within seeds (Cohen and Amir, 2017), we assumed that possible alterations in the expression of some sulfur-associated genes are relatively minor and presumably evaded the common P value and fold change cutoffs following the analysis. DISCUSSION SMM Produced by Rosette Leaves Supports Met Synthesis in Seeds Our study strongly suggests that rosette leaves of both wild-type and RNAi plants convert Met to SMM, which is then transported to the reproductive tissues to support the synthesis of Met in the developing seeds. We base this assumption on isotope-labeled feeding experiments demonstrating the conversion of [13C]Met to [13C]SMM in rosette leaves, which was then detected in stems and flower stalks. Similar expression levels of AtCGS, AtMMT, and AtHMTs were detected in these two latter tissues of RNAi plants as in the wild type, proposing that SMM is not produced in these tissues but rather transported from the rosette leaves to the seeds through them. Therefore, our observations reinforce the model of Bourgis et al. (1999) claiming that SMM originated from leaves and is used as a source of organic sulfur for reproductive organs. Yet, these are in partial contradiction to those reported by Lee et al. (2008), who performed flower stalk grafting experiments on Arabidopsis wild-type and hmt2 mutant plants. By using the wild type as a scion and hmt2 as a stock, the authors found significantly higher levels of SMM in the upper stalk, while plants derived from their corresponding reciprocal grafting accumulated similar SMM levels to wild-type plants. This suggested that the maternal genotype at the top of the flower stalk determines SMM and Met contents in seeds and not SMM production in rosette leaves. They also proposed that SMM is a nonessential intermediate in the movement of Met from vegetative tissues to the seeds (Lee et al., 2008), supporting the normal sulfur contents detected in seeds produced by the Arabidopsis mmt mutant that do not produce SMM (Kocsis et al., 2003). The differences between the experiments most likely lie in the fact that Lee et al. (2008) used grafted plants and mutants, while in our study, we used intact wild-type and transgenic plants. [13C]Met was transported from labeling sites to other tissues, since it was also detected in nonlabeled rosette leaves of both wild-type and RNAi plants. Only minor traces were measured in the stems and flower stalks, implying that [13C]Met was inadequately transported to these reproductive tissues but was mainly dispersed to tissues adjacent to labeling sites. Nevertheless, when SMM was absent, such as in the Arabidopsis mmt mutant, the transport of Met was reported to increase (Kocsis et al., 2003). These researchers reported that the exogenous application of [13C]Met resulted in its accumulation in mmt seeds, indicating that Met is possibly transported to the seeds either directly and/or via an intermediate other than SMM when its synthesis through the SMM cycle is limited (Kocsis et al., 2003). The Compensatory Role of the SMM Cycle in Maintaining Met Synthesis in RNAi::AtCGS Plants Where the Asp Family Pathway Is Restricted Our main finding is that the SMM cycle in RNAi plants plays crucial roles in sustaining the metabolism of Met in seeds when its synthesis through the classical Asp family pathway is limited. Several lines of evidence led to this conclusion: (1) the induction of AtCGS and AtMMT in the rosette leaves of RNAi plants, suggesting higher production of Met and conversion to SMM; (2) higher SMM levels in RNAi stems together with stable levels of Met, implying that SMM was transported to the upper stalk; (3) SMM accumulation in the flower stalks and silique hulls of RNAi plants along with lower expression levels of some AtHMTs, proposing that SMM was most likely transported from the rosette leaves via these tissues to the developing seeds (the higher levels of Met in silique hulls probably come from de novo synthesis, as these exhibited the induction of AtCGS); (4) higher SMM contents together with the elevated expression of AtHMT1 and AtHMT3 in RNAi seeds, providing a reliable explanation for Met overaccumulation; and (5) isotope-labeling experiments showing that [13C]Met is converted to [13C]SMM in rosette leaves and then transported through the stems to the flower stalks and developing seeds. Subsequently, significantly higher levels of [13C]SMM and [13C]Met were found in RNAi seeds. The results raise questions about the function of AtMMT and AtHMTs in seeds, particularly at late stages of Arabidopsis seed development. Recently, we showed that AtHMT1 and AtHMT3, but not AtHMT2 and/or AtMMT, are the predominant enzymes operating in Arabidopsis seeds, since the altered expression of these two genes affected the levels of Met and SMM in their corresponding transgenic seeds (Cohen et al., 2017). These observations are in agreement with those of this study showing the compensatory role of AtHMT1 and AtHMT3 in seeds when AtCGS expression is limited. Gallardo et al. (2007) initiated a comparative proteomic and transcriptomic study of different sections of Medicago truncatula seeds, suggesting that Met is synthesized through the Asp family pathway in the endosperm and seed coat tissues up to the midstages of seed filling. At late stages of seed filling, however, Met is then synthesized from the pool of SMM in the seed coat (Gallardo et al., 2007). The assumption was based on the expression levels of three genes/proteins: HMT, MET SYNTHASE, and SAMS (Fig. 1A). Despite the fact that these results are based on transcript and protein levels rather than on distinct metabolic changes, it suggests the contributions of both Asp family and SMM pathways to Met synthesis in seeds, but at different seed compartments and different stages of seed development. Our results propose that AtCGS is less vital for Met synthesis at late stages of Arabidopsis seed development compared with the contribution of transported SMM. Yet, the lower expression of AtCGS in RNAi seeds resulted in the overaccumulation of Met and the induction of genes operating in the SMM cycle in seeds and leaves. Evidence of Cross Talk between Seeds and Leaves and between the Asp Family Pathway and SMM Cycle-Related Enzymes Lower expression of AtCGS in RNAi seeds triggered changes in the expression of several genes involved in Met and SMM metabolism in rosette leaves, silique hulls, and seeds, leading to the differential accumulation of SMM and Met in some of these tissues. This type of interaction between sink and source organs is not trivial and apparently is far more complex than either efflux from higher to lower levels according to a gradient or one that results from the conversion of one metabolite to another in maintaining the osmotic gradient, as higher levels of both Met and SMM were found in RNAi seeds. Stems of RNAi plants were able to transport more [13C]SMM compared with wild-type stems, implying that the upper stalks and developing seeds triggered this higher transport. One of the options might be an as yet unknown signal that is transported through the plant or through air from the developing seeds to the rosette leaves, opposite to the transport direction from leaves to seeds of SMM and other metabolites. Such a signal, apparently among other factors, mediates the cross talk between seeds and leaves but also between Asp and Met biosynthesis pathways and those of the SMM cycle. Lower expression of AtCGS in RNAi seeds triggered the induced expression of AtMMT, AtHMT2, AtHMT3, and AtCGS in rosette leaves and of AtHMT1 and AtHMT3 in seeds. Yet, these changes did not affect all tissues, as none of these genes exhibited altered expression in stems and flower stalks. At least regarding the seeds, these results are in line with our previous observations. For example, transgenic Arabidopsis seeds overexpressing a feedback-insensitive form of AtCGS were shown to induce the expression of AtMMT, AtHMT1, and AtHMT3 (Cohen et al., 2014). Likewise, altered expression of AtCGS was detected in transgenic RNAi Arabidopsis seeds with reduced expression levels of AtHMT1 and AtHMT3 (Cohen et al., 2017). Together, these observations highlight the complex regulation of Met metabolism between and within tissues and between the SMM and Asp family pathways. This Study Implies Other Possible Roles of the SMM Cycle in Plant Organs Our results indicated that the SMM cycle converts Met to SMM and vice versa in leaves and seeds, since we found higher expression of AtHMT1, AtHMT3, and AtMMT in seeds and higher expression of AtMMT, AtHMT2, and AtHMT3 in leaves. Consequently, the higher level of [13C]Met and [13C]SMM found in seeds might result from the full conversion and reconversion of [13C]SMM transported from vegetative tissues. Indeed, a previous report showed that Arabidopsis leaves, roots, and developing seeds can metabolize [35S]Met to [35S]SMM and vice versa (Ranocha et al., 2001). All of these genes were shown previously to be expressed simultaneously in 17 different vegetative and reproductive tissues along the entire Arabidopsis life cycle (Frank et al., 2015). We observed that a higher expression level of AtMMT stimulated the induction of AtHMTs and vice versa, at least in seeds and rosette leaves. This stimulation is likely regulated by the contents of Met/SMM, as we found previously in Arabidopsis seeds expressing a feedback-insensitive form of AtCGS having significantly higher levels of Met but also higher expression of AtHMT1 and AtHMT3 (Cohen et al., 2014). We previously found a strong association between higher Met contents in seeds and the accumulation of stress-related transcript and metabolites (Cohen et al., 2014). Hence, the induction of AtMMT in RNAi seeds might produce more SMM, apparently trying to reduce the higher levels of Met and minimize the stress effects. A similar phenomenon also might occur in rosette leaves. Indeed, a metabolic profiling analysis of tobacco leaves overexpressing AtCGS shows that they have higher contents of stress-related metabolites, and they were more sensitive to short- and long-term oxidative stresses (Hacham et al., 2017). Since the SMM cycle is considered futile, it would be interesting to investigate whether AtHMTs and AtMMT are found in the same type of cells within seeds and leaves and whether SMM can be stored in other organelles, such as vacuoles, in order to discard its futility when HMTs and MMT cooperate at the same cell. In such a case, specific transporters, yet to be found (Gigolashvili and Kopriva, 2014), are expected to transport SMM and/or Met to and from this organelle. Higher Met Levels Alter Sulfur-Associated Metabolism in RNAi Seeds The levels of Cys and its product, GSH, were reduced only in RNAi 2 seeds but not in RNAi 1. We previously observed a similar phenomenon in transgenic seeds accumulating 4- and 6-fold more Met than wild-type seeds (Cohen et al., 2014), suggesting that the 50% more Met detected in RNAi 1 seeds was not sufficient to increase the flux toward Met biosynthesis at such levels to interrupt the synthesis of Cys and GSH. This also reinforces the aforementioned observations that a higher Met level in Arabidopsis seeds leads to dose-dependent metabolic and transcriptomic effects (Cohen and Amir, 2017). The metabolism of both short- and long-chain Met-derived glucosinolates also was altered in RNAi seeds, apparently due to metabolic rearrangements in Met and SMM biosynthetic pathways. For example, 3BZO and 4MTB, two products of homomethionine and dihomomethionine glucosinolates, respectively, exhibited higher levels in RNAi seeds compared with wild-type seeds, while lower levels of 4BZO were detected. 7MTH from pentahomomethionine increased in RNAi seeds, while both side chain modification products of hexahomomethionine, 8MTO and 8MSOO, decreased significantly in these seeds. Despite the fact that we could not point to a clear trend of change, these results suggest that changes in the levels of Met and/or SMM can change the biosynthetic pathways of glucosinolates. Additionally, they could partially explain why RNAi seeds exhibited induced expression of two myrosinases, AtTGG1 (At5g25980) and AtTGG2 (At5g26000), involved in the hydrolysis of glucosinolates, as identified previously by microarray analysis (Cohen and Amir, 2017). CONCLUSION All in all, this research highlights the complexity of the SMM cycle in Arabidopsis seeds, demonstrates mutual effects existing between the Asp family and SMM pathways throughout the plant, and proposes a novel cross talk between developing seeds and rosette leaves. According to our findings, wild-type plants produce Met de novo in rosette leaves by AtCGS through the Asp family pathway, where it is converted to SMM by AtMMT. SMM is then translocated through the phloem into the reproductive tissues (flower stalks, silique hulls, and seeds) and reconverted in the developing seeds back to Met by AtHMTs (Fig. 6, left). RNAi::AtCGS seeds, however, cannot produce Met by AtCGS and apparently trigger the formation of an as yet unknown signal reaching the rosette leaves, leading to the elevation of Met and SMM production by AtCGS and AtMMT, also inducing the expression of AtHMTs. The access of SMM that is transferred through the stems and flower stalks is converted back to Met in RNAi developing seeds at higher rates than in wild-type seeds, presumably by AtHMT1 and AtHMT3 but not by AtHMT2, to maintain stable Met levels. The elevation of soluble Met during the development of RNAi seeds reduces the synthesis of other sulfur-associated compounds such as Cys and glutathione and alters the metabolism of glucosinolates (Fig. 6, right). These also lead to the induction of desiccation stress and the accumulation of stress-related transcripts and metabolites (Cohen et al., 2014; Cohen and Amir, 2017). Notably, the global transcript and metabolic effects observed in the different tissues of RNAi 1 plants were of lower magnitude compared with those of RNAi 2 plants, further indicating that these changes depend on the expression levels of AtCGS and/or Met contents within seeds. Figure 6. Open in new tabDownload slide Proposed model for the metabolism of Met and SMM in wild-type (WT) and RNAi::AtCGS plants. Dashed lines represent multiple metabolic steps, while solid lines represent a single metabolic step. Boldface genes/metabolites represent higher expression/abundance found in RNAi::AtCGS tissues compared with their wild-type counterparts. Figure 6. Open in new tabDownload slide Proposed model for the metabolism of Met and SMM in wild-type (WT) and RNAi::AtCGS plants. Dashed lines represent multiple metabolic steps, while solid lines represent a single metabolic step. Boldface genes/metabolites represent higher expression/abundance found in RNAi::AtCGS tissues compared with their wild-type counterparts. From a nutritional quality prospective, these data indicate that manipulating the SMM cycle, such as increasing the levels of MMT and CGS in leaves together with increasing the levels of HMTs at the late developmental stage in seeds, may provide a new approach for improving Met contents in seeds. Altogether, the data provide new insights into the factors regulating Met and the mechanisms mediating its contents in seeds. MATERIALS AND METHODS Growth Conditions and Tissue Collection Throughout this work, Arabidopsis (Arabidopsis thaliana) wild-type, EV, and RNAi::AtCGS plants (called RNAi) were grown as described previously (Cohen and Amir, 2017). Rosette leaves, stems, silique hulls, and flower stalks were harvested from 44-d-old plants, immediately snap frozen under liquid nitrogen, cryolyophilized for 3 d until full desiccation, and stored at −80°C. Mature dry seeds were collected from plants at the end of the desiccation period, allowed to dry fully for 3 d in a vacuum desiccator, and stored under dry conditions at 4°C. Transcript Expression Analyses Using qRT-PCR qRT-PCR was used to determine the expression levels of AtCGS, AtHMT1, AtHMT2, AtHMT3, AtMMT, AtSAMS3, AtMGL, and the constitutive endogenous control gene AtPP2A-3 (Czechowski et al., 2005). RNA extraction, cDNA synthesis, and qRT-PCR procedures were according to Cohen et al. (2014). The values presented are means of three biological replicates, each with three technical replicates. Primers used for qRT-PCR analyses are listed in Supplemental Table S1. Determination of SMM and Met Contents Using HPLC and GC-MS SMM and Met were extracted from all harvested tissues and analyzed by HPLC as described previously by Cohen et al. (2017). Exogenous Feeding of [13C]SMM into 44-d-Old Inflorescences and Analysis of [13C]Met Levels by GC-MS Authentic [13C]Met was ordered from Cambridge Isotope Laboratories (http://www.isotope.com). [13C]SMM was synthesized by incubating [13C]Met with 250 mm methanol in 6 n HCl for 4 h at 110°C as described previously (Lavine et al., 1954; Gage et al., 1997). Successful conversion of [13C]SMM was verified against the retention time of authentic SMM standard using HPLC (Supplemental Fig. S3). Inflorescences from mature 44-d-old wild-type and RNAi plants were cut just above the rosette leaves, and the stalk base was inserted into 20 mL of labeling medium containing 2 mm [13C]SMM or DDW. Twelve individual inflorescences from each genotype were inserted into 50-mL tubes in five replicates and put in a growing chamber for 36 h under a controlled environment (22°C ± 1°C, 100 mmol m–2 s–1 constant light, and 70% relative humidity; Supplemental Fig. S1A). After labeling, silique hulls were immediately snap frozen under liquid nitrogen, cryolyophilized for 3 d until fully dried, and stored at −80°C. Siliques were then separated into developing seeds and the maternal tissue of the silique hulls. [13C]Met was extracted from developing seeds as described previously (Cohen et al., 2016b) and quantified by GC-MS, where [13C]Met could be differentiated from endogenous soluble Met due to the 1-13C label on its α-carbon, leading to the shift of the 293 m/z fragment to 294 m/z (Supplemental Fig. S2). Exogenous Feeding of [13C]Met into Whole Mature Plants, and Analysis of [13C]Met and [13C]SMM Levels in Vegetative and Reproductive Tissues by UPLC-TQ-MS/MS [13C]Met was applied to fully expanded rosette leaves of 44-d-old wild-type and RNAi flowering plants for 4, 8, and 24 h. Application sites on the leaves’ adaxial surface were gently abraded with fine silica powder to increase the uptake of [13C]Met. Then, the labeled/nonlabeled rosette leaves, stems, flower stalks, silique hulls, and developing seeds were collected and extracted for amino acid analysis. A control assay was performed by feeding wild-type and RNAi flowering plants for only 24 h with DDW (Supplemental Fig. S1B). Tissue extracts were then separated by UPLC-TQ-MS/MS using a SIELC Obelisc N column and MRM. A full description of UPLC conditions and MRM parameters is provided in Supplemental Tables S2 and S3. First, nonlabeled and 13C-labeled Met and SMM were measured using two MRM transitions for each compound; the first was used for quantification and the second for verification. The feasibility of feeding was validated by comparing the ratios between peak areas in MRMs of 13C-labeled and nonlabeled Met and SMM in wild-type labeled rosette leaves fed with [13C]Met or control rosette leaves fed with DDW. Indeed, 13C labeling in control rosette leaves was almost absent, showing similar ratios to those in Met/SMM external standard, while leaves fed with [13C]Met exhibited significantly higher 13C labeling ratios for both compounds (Supplemental Figs. S4–S6). The matrix effect in different tissues was determined by calculating the percentage recovery of Met and SMM in wild-type samples from all tissues spiked with Met/SMM standard or DDW. Calculated percentage recovery for all tissues ranged between 62% and 115% for Met and between 108% and 144% for SMM, indicating that there was no significant matrix effect in any tissue (Supplemental Fig. S7). Raw data were processed and analyzed using MassLynx software with TargetLynx (version 4.1; Waters). Determination of Cys and GSH Contents Using HPLC To determine Cys and GSH contents in control and transgenic seeds, 100 mg of mature dry seeds was collected, ground using a mortar and pestle, and then extracted and analyzed by HPLC as described previously (Matityahu et al., 2013). Glucosinolate Analysis by UPLC-qTOF-MS/MS To determine the composition of glucosinolates in control and transgenic seeds, 10 mg of mature dry seeds was extracted in 300 μL of 80% (v/v) methanol containing 0.1% (v/v) formic acid. The mixture was sonicated for 20 min at room temperature, centrifuged at 3,000g for 15 min, and filtered through a 0.22-μm polytetrafluoroethylene membrane (Acrodisc CR 13 mm; PALL) into running vials. Samples were injected into a UPLC-qTOF-MS/MS instrument (HDMS Synapt; Waters) connected online to a photodiode array and a mass spectrometry detector. Separation of metabolites was performed on a UPLC BEH C18 column (100 mm, 2.1 mm i.d., 1.7 μm; Waters) in negative ionization mode. The analysis of raw mass spectrometry data was carried out using XCMS software including chromatogram alignment, mass signal detection, and peak integration (Smith et al., 2006). Metabolites were identified using standard compounds by comparison of their retention times, UV spectra, mass spectrometry fragments, and dual-energy fragments. The identification of metabolites was carried out using an in-house compound library and previous reports (Malitsky et al., 2008; Tzin et al., 2009). Statistical Analyses Statistical analyses were performed using GraphPad Prism 5.01 scientific software (http://www.graphpad.com/). Significance was calculated using the two-way ANOVA and Student’s t tests of P < 0.05. Supplemental Data The following supplemental materials are available. Supplemental Figure S1. Schematic representation of isotope-labeled feeding experiments. Supplemental Figure S2. Validation of [13C]Met by GC-MS analysis. Supplemental Figure S3. Preparation of 13C-labeled compounds and verification by HPLC analysis. Supplemental Figure S4. MRMs for nonlabeled and 13C-labeled Met and SMM in Met/SMM external standard according to UPLC-TQ-MS/MS analysis. Supplemental Figure S5. MRMs for nonlabeled and 13C-labeled Met in labeled wild-type rosette leaves fed with DDW or [13C]Met according to UPLC-TQ-MS/MS analysis. Supplemental Figure S6. MRMs for nonlabeled and 13C-labeled SMM in labeled wild-type rosette leaves fed with DDW or [13C]Met according to UPLC-TQ-MS/MS analysis. Supplemental Figure S7. Determining matrix effects by calculating percentage recovery of Met and SMM in tissue samples of wild-type plants spiked with [13C]Met or DDW. Supplemental Table S1. List of primers used for qRT-PCR analyses. Supplemental Table S2. MRM parameters used for measuring nonlabeled and 13C-labeled Met and SMM by UPLC-TQ-MS/MS. Supplemental Table S3. UPLC conditions used for measuring nonlabeled and 13C-labeled Met and SMM. Glossary SAM S-adenosylmethionine SMM S-methylmethionine RNAi RNA interference GC-MS gas chromatography-mass spectrometry EV empty vector qRT quantitative real-time DDW double-distilled water UPLC-TQ-MS/MS ultra-performance liquid chromatography equipped with electrospray ionization and mass spectrometry detection GSH glutathione UPLC-qTOF-MS/MS ultra-performance liquid chromatography equipped with quadrupole time of flight mass spectrometry detection 4MTB 4-methylthiobutyl glucosinolate 5MTP 5-methylthiopentyl glucosinolate 3BZO 3-benzoyloxypropyl glucosinolate 4BZO 4-benzoyloxybutyl glucosinolate 7MSOH 7-methylsulfinylheptyl glucosinolate 8MSOO 8-methylsulfinyloctyl glucosinolate 7MTH 7-methylthioheptyl glucosinolate 8MTO 8-methylthiooctyl glucosinolate I3M indol-3-yl-methyl glucosinolate UPLC ultra-performance liquid chromatography LITERATURE CITED Amir R ( 2008 ) Towards improving methionine content in plants for enhanced nutritional quality . Funct Plant Sci Biotechnol 2 : 36 – 46 Google Scholar OpenURL Placeholder Text WorldCat Bourgis F , Roje S, Nuccio ML, Fisher DB, Tarczynski MC, Li C, Herschbach C, Rennenberg H, Pimenta MJ, Shen TL, et al. ( 1999 ) S-Methylmethionine plays a major role in phloem sulfur transport and is synthesized by a novel type of methyltransferase . Plant Cell 11 : 1485 – 1498 Google Scholar Crossref Search ADS PubMed WorldCat Bradbury LM , Ziemak MJ, El Badawi-Sidhu M, Fiehn O, Hanson AD ( 2014 ) Plant-driven repurposing of the ancient S-adenosylmethionine repair enzyme homocysteine S-methyltransferase . Biochem J 463 : 279 – 286 Google Scholar Crossref Search ADS PubMed WorldCat Cohen H , Amir R ( 2017 ) Dose-dependent effects of higher methionine levels on the transcriptome and metabolome of transgenic Arabidopsis seeds . Plant Cell Rep 36 : 719 – 730 Google Scholar Crossref Search ADS PubMed WorldCat Cohen H , Israeli H, Matityahu I, Amir R ( 2014 ) Seed-specific expression of a feedback-insensitive form of CYSTATHIONINE-γ-SYNTHASE in Arabidopsis stimulates metabolic and transcriptomic responses associated with desiccation stress . Plant Physiol 166 : 1575 – 1592 Google Scholar Crossref Search ADS PubMed WorldCat Cohen H , Pajak A, Pandurangan S, Amir R, Marsolais F ( 2016 a ) Higher endogenous methionine in transgenic Arabidopsis seeds affects the composition of storage proteins and lipids . Amino Acids 48 : 1413 – 1422 Google Scholar Crossref Search ADS PubMed WorldCat Cohen H , Salmon A, Tietel Z, Hacham Y, Amir R ( 2017 ) The relative contribution of genes operating in the S-methylmethionine cycle to methionine metabolism in Arabidopsis seeds . Plant Cell Rep 36 : 731 – 743 Google Scholar Crossref Search ADS PubMed WorldCat Cohen H , Shir OM, Yu Y, Hou W, Sun S, Han T, Amir R ( 2016 b ) Genetic background and environmental conditions drive metabolic variation in wild type and transgenic soybean (Glycine max) seeds . Plant Cell Environ 39 : 1805 – 1817 Google Scholar Crossref Search ADS PubMed WorldCat Czechowski T , Stitt M, Altmann T, Udvardi MK, Scheible WR ( 2005 ) Genome-wide identification and testing of superior reference genes for transcript normalization in Arabidopsis . Plant Physiol 139 : 5 – 17 Google Scholar Crossref Search ADS PubMed WorldCat Fait A , Nesi AN, Angelovici R, Lehmann M, Pham PA, Song L, Haslam RP, Napier JA, Galili G, Fernie AR ( 2011 ) Targeted enhancement of glutamate-to-γ-aminobutyrate conversion in Arabidopsis seeds affects carbon-nitrogen balance and storage reserves in a development-dependent manner . Plant Physiol 157 : 1026 – 1042 Google Scholar Crossref Search ADS PubMed WorldCat Frank A , Cohen H, Hoffman D, Amir R ( 2015 ) Methionine and S-methylmethionine exhibit temporal and spatial accumulation patterns during the Arabidopsis life cycle . Amino Acids 47 : 497 – 510 Google Scholar Crossref Search ADS PubMed WorldCat Gage DA , Rhodes D, Nolte KD, Hicks WA, Leustek T, Cooper AJ, Hanson AD ( 1997 ) A new route for synthesis of dimethylsulphoniopropionate in marine algae . Nature 387 : 891 – 894 Google Scholar Crossref Search ADS PubMed WorldCat Galili G , Amir R ( 2013 ) Fortifying plants with the essential amino acids lysine and methionine to improve nutritional quality . Plant Biotechnol J 11 : 211 – 222 Google Scholar Crossref Search ADS PubMed WorldCat Galili G , Amir R, Hoefgen R, Hesse H ( 2005 ) Improving the levels of essential amino acids and sulfur metabolites in plants . Biol Chem 386 : 817 – 831 Google Scholar Crossref Search ADS PubMed WorldCat Gallardo K , Firnhaber C, Zuber H, Héricher D, Belghazi M, Henry C, Küster H, Thompson R ( 2007 ) A combined proteome and transcriptome analysis of developing Medicago truncatula seeds: evidence for metabolic specialization of maternal and filial tissues . Mol Cell Proteomics 6 : 2165 – 2179 Google Scholar Crossref Search ADS PubMed WorldCat Gigolashvili T , Kopriva S ( 2014 ) Transporters in plant sulfur metabolism . Front Plant Sci 5 : 442 Google Scholar Crossref Search ADS PubMed WorldCat Giovanelli J ( 1990 ) Regulatory aspects of cysteine and methionine biosynthesis . In Rennenberg H , Brunold CH , de Kok LJ , Stulen I , eds, Sulfur Nutrition and Sulfur Assimilation in Higher Plants. SBP Academic Publishers , The Hague, The Netherlands , pp 33 – 48 Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Giovanelli JG ( 1987 ) Sulfur amino acids in plants: an overview . Methods Enzymol 143 : 419 – 428 Google Scholar Crossref Search ADS WorldCat Hacham Y , Matityahu I, Amir R ( 2017 ) Transgenic tobacco plants having a higher level of methionine are more sensitive to oxidative stress . Physiol Plant (in press) doi.org/10.1111/ppl.12557 Google Scholar OpenURL Placeholder Text WorldCat Hacham Y , Matityahu I, Schuster G, Amir R ( 2008 ) Overexpression of mutated forms of aspartate kinase and cystathionine gamma-synthase in tobacco leaves resulted in the high accumulation of methionine and threonine . Plant J 54 : 260 – 271 Google Scholar Crossref Search ADS PubMed WorldCat Hanafy MS , Rahman SM, Nakamoto Y, Fujiwara T, Naito S, Wakasa K, Ishimoto M ( 2013 ) Differential response of methionine metabolism in two grain legumes, soybean and azuki bean, expressing a mutated form of Arabidopsis cystathionine γ-synthase . J Plant Physiol 170 : 338 – 345 Google Scholar Crossref Search ADS PubMed WorldCat Hesse H , Kreft O, Maimann S, Zeh M, Hoefgen R ( 2004 ) Current understanding of the regulation of methionine biosynthesis in plants . J Exp Bot 55 : 1799 – 1808 Google Scholar Crossref Search ADS PubMed WorldCat Huang T , Joshi V, Jander G ( 2014 ) The catabolic enzyme methionine γ-lyase limits methionine accumulation in potato tubers . Plant Biotechnol J 12 : 883 – 893 Google Scholar Crossref Search ADS PubMed WorldCat Kim J , Leustek T ( 2000 ) Repression of cystathionine γ-synthase in Arabidopsis thaliana produces partial methionine auxotrophy and developmental abnormalities . Plant Sci 151 : 9 – 18 Google Scholar Crossref Search ADS WorldCat Kocsis MG , Ranocha P, Gage DA, Simon ES, Rhodes D, Peel GJ, Mellema S, Saito K, Awazuhara M, Li C, et al. ( 2003 ) Insertional inactivation of the methionine S-methyltransferase gene eliminates the S-methylmethionine cycle and increases the methylation ratio . Plant Physiol 131 : 1808 – 1815 Google Scholar Crossref Search ADS PubMed WorldCat Kopriva S ( 2006 ) Regulation of sulfate assimilation in Arabidopsis and beyond . Ann Bot (Lond) 97 : 479 – 495 Google Scholar Crossref Search ADS WorldCat Lavine TF , Floyd NF, Cammaroti MS ( 1954 ) The formation of sulfonium salts from alcohols and methionine in sulfuric acid . J Biol Chem 207 : 107 – 117 Google Scholar Crossref Search ADS PubMed WorldCat Lee M , Huang T, Toro-Ramos T, Fraga M, Last RL, Jander G ( 2008 ) Reduced activity of Arabidopsis thaliana HMT2, a methionine biosynthetic enzyme, increases seed methionine content . Plant J 54 : 310 – 320 Google Scholar Crossref Search ADS PubMed WorldCat Malitsky S , Blum E, Less H, Venger I, Elbaz M, Morin S, Eshed Y, Aharoni A ( 2008 ) The transcript and metabolite networks affected by the two clades of Arabidopsis glucosinolate biosynthesis regulators . Plant Physiol 148 : 2021 – 2049 Google Scholar Crossref Search ADS PubMed WorldCat Matityahu I , Godo I, Hacham Y, Amir R ( 2013 ) Tobacco seeds expressing feedback-insensitive cystathionine γ-synthase exhibit elevated content of methionine and altered primary metabolic profile . BMC Plant Biol 13 : 206 Google Scholar Crossref Search ADS PubMed WorldCat Mudd SH , Datko AH ( 1990 ) The S-methylmethionine cycle in Lemna paucicostata . Plant Physiol 93 : 623 – 630 Google Scholar Crossref Search ADS PubMed WorldCat Ranocha P , McNeil SD, Ziemak MJ, Li C, Tarczynski MC, Hanson AD ( 2001 ) The S-methylmethionine cycle in angiosperms: ubiquity, antiquity and activity . Plant J 25 : 575 – 584 Google Scholar Crossref Search ADS PubMed WorldCat Reintanz B , Lehnen M, Reichelt M, Gershenzon J, Kowalczyk M, Sandberg G, Godde M, Uhl R, Palme K ( 2001 ) Bus, a bushy Arabidopsis CYP79F1 knockout mutant with abolished synthesis of short-chain aliphatic glucosinolates . Plant Cell 13 : 351 – 367 Google Scholar Crossref Search ADS PubMed WorldCat Roje S ( 2006 ) S-Adenosyl-L-methionine: beyond the universal methyl group donor . Phytochemistry 67 : 1686 – 1698 Google Scholar Crossref Search ADS PubMed WorldCat Sauter M , Moffatt B, Saechao MC, Hell R, Wirtz M ( 2013 ) Methionine salvage and S-adenosylmethionine: essential links between sulfur, ethylene and polyamine biosynthesis . Biochem J 451 : 145 – 154 Google Scholar Crossref Search ADS PubMed WorldCat Smith CA , Want EJ, O’Maille G, Abagyan R, Siuzdak G ( 2006 ) XCMS: processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching, and identification . Anal Chem 78 : 779 – 787 Google Scholar Crossref Search ADS PubMed WorldCat Song S , Hou W, Godo I, Wu C, Yu Y, Matityahu I, Hacham Y, Sun S, Han T, Amir R ( 2013 ) Soybean seeds expressing feedback-insensitive cystathionine γ-synthase exhibit a higher content of methionine . J Exp Bot 64 : 1917 – 1926 Google Scholar Crossref Search ADS PubMed WorldCat Tzin V , Malitsky S, Aharoni A, Galili G ( 2009 ) Expression of a bacterial bi-functional chorismate mutase/prephenate dehydratase modulates primary and secondary metabolism associated with aromatic amino acids in Arabidopsis . Plant J 60 : 156 – 167 Google Scholar Crossref Search ADS PubMed WorldCat Author notes 1 This work was supported by the Israeli Science Foundation (grant no. 1004/15). 2 Address correspondence to rachel@migal.org.il. The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Rachel Amir (rachel@migal.org.il). H.C. and R.A. planned and designed the research; H.C. and Y.H. performed the experiments; H.C., Y.H., I.P., I.R., and A.A analyzed the data; H.C. and R.A. wrote the article. www.plantphysiol.org/cgi/doi/10.1104/pp.17.00579 © 2017 American Society of Plant Biologists. All Rights Reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Repression of CYSTATHIONINE γ-SYNTHASE in Seeds Recruits the S-Methylmethionine Cycle
JF - Plant Physiology
DO - 10.1104/pp.17.00579
DA - 2017-06-28
UR - https://www.deepdyve.com/lp/oxford-university-press/repression-of-cystathionine-synthase-in-seeds-recruits-the-s-wKLjGA0quk
SP - 1322
EP - 1333
VL - 174
IS - 3
DP - DeepDyve
ER -