α-Aminoadipic acid metabolism is controlled by the glutathione-dependent redox environment in Arabidopsis

α-Aminoadipic acid (Aaa) is an intermediate product in the Lys catabolism, on the saccharopine pathway. Since Lys content was increased by various abiotic stress factors affecting the redox environment, a glutathione-dependent redox regulation of the Aaa metabolism was assumed which was not tested earlier. The effect of the modified size and redox state of glutathione pool (reduced + oxidised, GSH + GSSG) on the Aaa content and related amino acid and transcript levels was compared in the wild-type Col-0 plants and the GSH-deficient pad2-1 Arabidopsis mutant with and without (control) treatment. Modification of the size and redox state of glutathione pool resulted in the alteration of Lys, Glu, Pro and Aaa contents and the transcription of the three investigated genes of the saccharopine pathway. Both the Aaa content and the expression of the two genes encoding enzymes of its synthesis (ketoglutarate reductase/saccharopine dehydrogenase, LKR/SDH; aminoadipic semialdehyde dehydrogenase, AASADH) was lower in the pad2-1 mutant, and these parameters, together with the GSSG content, were increased by GSH treatment both in Col-0 and mutant plants. The GSSG content and its ratio in the glutathione pool exhibited a positive correlation with the Aaa content and the transcription of the two genes of its synthesis and the Aaa catabolic gene (dihydrolipoamide-succinyltransferase), which indicates a GSH-dependent control of Aaa metabolism. This latter process in turn can affect the redox environment and stress response because of the NADPH use/formation in its several steps.


Introduction
Amino acids are known as the building blocks of proteins. However, the non-proteinogenic amino acids play other important biochemical and physiological functions. In plants they are involved in protection against oxidative stress, transformed into Aaa by aminoadipic semialdehyde dehydrogenase (AASADH) (Arruda et al. 2000). The dihydrolipoamide-succinyltransferase (DISU) plays central role in the catabolism of Aaa producing glutaryl-CoA. In several abiotic stress situations the level of Lys is increased in Arabidopsis (Obata and Fernie 2012). According to Navarova et al. (2012) strong accumulation of Aaa was observed at bacterial inoculation in Arabidopsis due to increased activation of LKR/SDH gene. In addition, its expression was upregulated by treatments with abscisic acid and jasmonic acid which hormones are regulators of stress response. This result can be explained by the presence of putative abscisic acid-and jasmonic acid-related elements in the promoter of LKR/SDH gene (Stepansky et al. 2005). The LKR/SDH gene was associated with the stress response in maize, Arabidopsis and rapeseed (Deleu et al. 1999;Less et al. 2011;Kiyota et al. 2015). AASADH enzyme is also involved in cellular defence against hyperosmotic stress (Brocker et al. 2010). In addition, the ectopically expressed soybean AASADH in Arabidopsis and tobacco resulted in enhanced stress tolerance against salinity, drought and oxidative stress (Rodrigues et al. 2006).
The relationship between the amino acid and glutathione (GSH) metabolism was shown by Noctor et al. (1998). The overexpression of the gene encoding γ-glutamylcysteine synthetase, the first enzyme of GSH synthesis increased not only the level of GSH, but also that of some free amino acids. This observation can be explained by the multifunctional role of GSH, which is an essential regulator of development, growth, flowering, biosynthetic pathways, detoxification, antioxidant biochemistry and redox homeostasis (Noctor et al. 2012). Both its pool size (reduced + oxidised forms, GSH + GSSG) and redox state (GSSG/GSH ratio) is important in the control of these processes, and changes in these parameters can be used as stress markers (Kranner et al. 2006). Treatment of Arabidopsis plants with GSH increased the level of Aaa extremely, while other redox treatments (ascorbate, dithiotreitol, hydrogen-peroxide) did not affect it in our previous experiment (Gulyás et al. 2017). In the present study, we wanted to clarify whether the changes in the size and redox state of the glutathione pool modify the amount of Aaa and the level of amino acids and transcripts related to its metabolism in the saccharopine pathway.

Plant material and growth conditions
Seedlings of the wild-type Arabidopsis (Col-0) and the glutathione-deficient (pad2-1) (Parisy et al. 2007) mutant were cultivated (8 h photoperiod, 100 µmol m − 2 s − 1 light intensity, 23/22°C temperature and 70/75% RH) as described earlier (Gulyás et al. 2017). After 5 weeks growth on soil blocks (Jiffy, Jiffy Products S.L. Ltd.), plants were transferred onto half-strength modified Hoagland-solution. After 5 days adaptation phase, 4 mM GSH was added to the solution for 3 days, which treatment was chosen based on our previous experiments (Gulyás et al. 2017). Samples were collected for biochemical analysis and gene expression studies after the treatment. There were 3 independent experiments with 3 parallel samples each.

Determination of glutathione
Plant samples were extracted for the determination of glutathione as described previously (Kocsy et al. 2000). Samples were analysed by reverse-phase HPLC using Waters Alliance 2695 Separations Module with a W474 scanning fluorescence detector (Waters, Milford, MA, USA).

Analysis of free amino acid content
The extraction of the samples and the analysis of amino acids was done according to an earlier used protocol (Gulyás et al. 2017). The measurements were carried out on an automatic amino acid analyser (Ingos Ltd., Czech Republic) equipped with an Ionex Ostion LCP5020 cation-exchange column (22 cm × 0.37 cm).

Gene expression studies
Total RNA extraction, the qRT-PCR (CFX96 thermocycler, Bio-Rad) and the data analysis was made based on our standard protocol (Gulyás et al. 2017). The primers (Table S1.) were designed by our group using Primer3 software (Koressaar and Remm 2007). The relative expression values were calculated with the ∆Ct method using the Actin2 (Wang et al. 2014) housekeeping gene (TAIR ID: AT3G18780.2) for normalization. Their log2 values are illustrated on a heatmap prepared by the MEV (Multiple Experiment Viewer, version 4.7.1.) programme.

Statistical analysis
The statistical analysis was done using one-way analysis of variance (SPSS 16.0.). Least significant difference method was used for post hoc analysis. Homogeneity of the variances was tested by Levene's test. The relationships between the various parameters were checked by correlation analysis (Microsoft Excel program) (Guilford 1950).

Changes in the amount and redox state of glutathione
As expected, the total glutathione content (GSH + GSSG) was much lower in pad2-1 mutants compared to the Col-0 plants under control conditions (Fig. 1). However, the GSH treatment recovered the total glutathione concentration in this genotype to level of the control (Col-0). In addition, the GSSG level and the GSSG/GSH ratio also greatly increased after GSH treatment in both genotypes.

Effect of glutathione treatment on free asp, lys, glu, pro and aaa contents
Among the free amino acids being associated with Aaa metabolism, the level of aspartic acid (Asp) was not influenced by the modification of GSH levels either by mutation or its exogenous application (Fig. 2a). The free Lys (Aaa precursor) content was lower in pad2-1 mutant compared to the Col-0 genotype. It was not influenced by the GSH treatment in either of the used genotypes (Fig. 2b). Although the amount of free Glu (side product of Aaa synthesis) was greater in pad2-1 mutant than in the Col-0 plants under control conditions, it was not affected by the GSH treatment (Fig. 2c). The prolin (Pro, its precursor is Glu) content was greater in pad2-1 mutant in which it was reduced by the addition of GSH (Fig. 2d). In pad2-1 mutant the Aaa level was much lower compared to the Col-0 plants. After the GSH treatment, its amount had a 5-fold increase in both genotypes. Thus, the amount of Aaa was recovered by the addition of GSH in the pad2-1 mutant to the level observed in the Col-0 plants before the treatment (Fig. 2e).

Modification of the gene expressions in the α-aminoadipic acid pathway by glutathione
The transcript levels of LKR/SDH and AASADH genes, involved in the Aaa synthesis, was lower in the GSH-deficient pad2-1 mutant under control conditions compared to the Col-0 plants (Fig. 3). They increased significantly after GSH treatment in both genotypes, but this change was smaller in pad2-1. The expression of DISU, related to the Aaa degradation, was also lower in the mutant than in the Col-0 plants before the GSH addition, while after the Fig. 1 Effect of glutathione treatment (4 mM GSH) on reduced (grey) and oxidised (white) glutathione content and their ratio in the investigated genotypes. Reduced glutathione: GSH; oxidised glutathione: GSSG. Values indicated by asterisks are significantly different from the control Col-0, treated with no GSH, at the P ≤ 0.05% level

Discussion
Although the reduced availability of GSH in pad2-1 mutant greatly decreased the amount of Aaa, and treatment of Col-0 and pad2-1 plants with exogenous GSH resulted in its 5-fold increase, there was no correlation between the GSH and Aaa levels. The lack of such relationship can be explained by treatment it increased in Col-0 and remained unchanged in pad2-1. Fig. 2 Changes in the level of free asparagine (Asp, A), lysine (Lys, B), glutamate (Glu, C), proline (Pro, D) and α-aminoadipic acid (Aaa, E) contents after the glutathione treatment (4mM GSH) in Col-0 (black) and pad2-1 (grey) genotypes. Values indicated by asterisks are significantly different from the Col-0, treated with no GSH, at the P ≤ 0.05% level increase was smaller or did not occur in the DISU transcript level related to Aaa degradation. The GSH-dependent, coordinated adjustment (very strong, positive correlation between their transcript levels, Table S2.) of these genes could be important in the stress response, in which these genes are involved according to previous studies (Deleu et al. 1999;Moulin et al. 2006;Rodrigues et al. 2006;Less et al. 2011;Kiyota et al. 2015).
Interestingly, treatment of the plants with GSH led to a great increase in the GSSG/GSH ratio, which could be the result of the inhibition of GSSG reduction by the GSH treatment. The increase in the GSSG content could induce the Aaa formation in order to increase the NADPH amount to ensure the GSSG reduction in the ascorbate-glutathione cycle. During the Aaa synthesis from Lys, there is one NADPH-consuming step and two NADPH-producing steps, thus the NADPH content will increase (Arruda and Neshich 2012). The excess of NADPH can ensure the subsequent reduction of GSSG in the ASA-GSH cycle. This hypothesis is supported by the positive correlation between GSSG and Aaa levels in the present experiments (Table S2.), and the earlier described protective role of Aaa during abiotic stresses (Arruda and Barreto, 2020). Thus, greater Lys, Pro, saccharopine and Aaa metabolite and related transcript levels were found in the tolerant sesame genotype compared to the sensitive one after drought stress (You et al. 2019). In addition, during the aminoadipate-semialdehyde formation from saccharopine, Glu, the precursor of Pro will be synthesised the unchanged amount of the endogenous GSH level after GSH addition in Col-0 plants, which could be a result of the inhibition of GSH formation by its great excess. Since the Aaa pathway is used for the catabolism of Lys, similar differences could be expected in its amount between the genotypes and after the treatment. Indeed, the Lys content was lower in pad2-1 mutant compared to the Col-0 plants after cultivation on the basic nutrient solution, but the addition of GSH did not influence the Lys levels. Thus, the GSHdependent alteration in Aaa level did not derive from the corresponding change in the Lys content and the low correlation between the amount of Lys and Aaa also did not indicate such relationship (Table S2.). Even the level of the Lys precursor, Asp was not modified by the GSH addition. Since Glu is the precursor of GSH and it is also a side-product of Lys catabolism to Aaa, an association between Glu and Aaa was also assumed, but it was not confirmed by the comparison of the amounts of Glu and Aaa by correlation analysis.
In contrast to GSH, the amount of GSSG and the GSSG/ GSH ratio exhibited a moderate (r: 0.43-0.64) and a strong (r: 0.73-0.88), positive correlation with Aaa content and the transcript levels of the genes encoding three enzymes of the Aaa metabolism, respectively (Table S2.). The redox state of the glutathione pool may influence the Aaa metabolism at transcriptional level, since both Aaa content and the expression of the genes encoding the enzymes related to its synthesis was lower in pad2-1 mutant and greater after the addition of exogenous GSH in both genotypes. This latter Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons. org/licenses/by/4.0/.