Characterization of γ-glutamyltransferase- and phytochelatin synthase-mediated catabolism of glutathione and glutathione S-conjugates in Arabidopsis thaliana

Abstract

Glutathione (GSH, γ-L-glutamyl-L-cysteinyl-glycine) has been implicated in a multitude of cellular functions, such as protection of cells against oxidative stress, detoxification of xenobiotics via degradation of GSH S-conjugates, and disease resistance. Glutathione also serves as a precursor of phytochelatins, and thereby plays an essential role in heavy metal detoxification. The Arabidopsis genome encodes three functional γ-glutamyltransferase genes (AtGGT1, AtGGT2, AtGGT4) and two phytochelatin synthase genes (AtPCS1, AtPCS2). The function of plant GGT has not yet been clearly defined, although it is thought to be involved in GSH and GSH S-conjugate catabolism. On the other hand, besides its role in heavy metal detoxification, PCS has also been involved in GSH S-conjugate catabolism. Herein we describe the HPLC characterization of GSH and GSH S-conjugate catabolism in Arabidopsis mutants deficient in GSH biosynthesis (pad2-1/gsh1), atggt and atpcs1 T-DNA insertion mutants, atggt pad2-1, atggt atpcs1 double mutants, and the atggt1 atggt4 atpcs1 triple mutant. The results of our HPLC analysis confirm that AtGGT and AtPCS play important roles in two different pathways related with GSH and GSH S-conjugate (GS-bimane) catabolism in Arabidopsis.

Introduction

Glutathione (GSH, γ-L-glutamyl-L-cysteinyl-glycine), which is a thiol-containing tripeptide conserved in most organisms, plays essential roles in cellular functions such as storage and transport of reduced sulfur compounds, transport of amino acids, protection of cells against oxidative stress, disease resistance, detoxification of xenobiotics and heavy metals, redox regulation of enzymes, and gene expression (Noctor et al. 2012). GSH is synthesized from Glu, Cys, and Gly by two ATP-dependent reactions catalyzed by γ-glutamylcystein synthetase (γ-ECS, EC 6.3.2.2) and glutathione synthetase (GSHS, EC 6.3.2.3) (Noctor et al. 2012).

γ-Glutamyltransferase (GGT, EC 2.3.2.2), which is found in plants, mammals and microorganisms, catalyzes the hydrolysis of γ-glutamyl linkages of γ-glutamyl peptides, or the transfer of the γ-glutamyl moiety to a large number of amino acid and peptide acceptors, resulting in a new amide bond with an N-terminus. In mammals, yeast and Escherichia coli, GGTs are well characterized and have been shown to possess several physiologically important functions including glutathione GSH catabolism. The function of plant GGTs has not yet been clearly defined, although they are thought to be involved in extra-cytosolic (apoplastic and vacuolar) GSH catabolism and detoxification of xenobiotics via degradation of GSH S-conjugates (GS-X) (Ferretti et al. 2009; Grzam et al. 2007; Martin et al. 2007; Masi et al. 2015; Nakano et al. 2006; Ohkama-Ohtsu et al. 2007a, b; Prieto et al. 2009). The Arabidopsis genome contains three functional genes encoding GGT: AtGGT1 (At4g39640), AtGGT2 (At4g39650), and AtGGT4 (At4g29210). At4g29210 was called AtGGT3 in earlier studies (Ohkama-Ohtsu et al. 2007a, b; Prieto et al. 2009), but was later renamed as AtGGT4 to agree with the nomenclature of Martin et al. (2007) (Ohkama-Ohtsu et al. 2008). RT-PCR and AtGGT::GUS expression analyses indicated that AtGGT1 and AtGGT4 are probably constitutively expressed genes, whereas AtGGT2 is mainly expressed in siliques (Martin et al. 2007; Ohkama-Ohtsu et al. 2007a, b; Prieto et al. 2009). AtGGT1 and AtGGT2 have been proposed to be involved in apoplastic GSH catabolism (Destro et al. 2011; Martin et al. 2007; Ohkama-Ohtsu et al. 2007a; Prieto et al. 2009), and AtGGT4 in vacuolar degradation of GS-X (Grzam et al. 2007; Ohkama-Ohtsu et al. 2007b).

In both animals and plants, the γ-glutamyl cycle consists of intracellular GSH synthesis, extrusion to the extracellular space or apoplast, respectively, and degradation by GGTs and Cys-Gly dipeptidases into its constituent amino acids, which are transported into the cell for reutilization (Masi et al. 2015; Ohkama-Ohtsu et al. 2008).

Phytochelatin synthase (PCS, EC 2.3.2.15) is a specific γ-L-glutamyl-cysteine dipeptidyl transferase that catalyzes the synthesis of phytochelatins (PCs) from GSH in the presence of heavy metal stress (Grill et al. 1989; Vatamaniuk et al. 2000). GSH is polymerized to PCs, which detoxify heavy metals such as Cd and As by forming complexes that are then transported into the vacuole (Cobbett et al. 1998; Howden et al. 1995a, b; Li et al. 2004). The Arabidopsis genome encodes two PCS genes (AtPCS1=At5G44070, AtPCS2=At1G03980). AtPCS1 is the major player in PC biosynthesis, and its deficiency results in impaired heavy metal tolerance (Howden et al. 1995b). On the other hand, AtPCSs have been reported to be involved in the cytosolic degradation of GSH S-conjugates (Blum et al. 2010, 2007).

This work presents new evidence about the involvement of AtGGTs in GSH catabolism, and the mechanism of interaction between AtGGT and AtPCS mediated degradation of GSH S-conjugates (GS-bimane). Toward that end, Arabidopsis mutants deficient in GSH biosynthesis (pad2-1/gsh1), atggt, and atpcs1 single mutants, as well as atggt, atggt pad2-1, atggt atpcs1 double and triple mutants were obtained and characterized. Taken together, the results of our HPLC analysis confirm that AtGGTs are involved in the apoplastic and vacuolar catabolism of GSH and GSH S-conjugates of xenobiotics, whereas AtPCSs play an important role in their cytoplasmic degradation.

Materials and methods

Plant materials and growth conditions

The atggt1, atggt2, and atpcs1 T-DNA insertion mutants (SALK_080363, SALK_147881, and SAIL_650_C12, ecotype Columbia, Col) (Alonso et al. 2003; Prieto et al. 2009; Sessions et al. 2002) and the pad2-1 (gsh1) mutant (CS3804, Col) (Parisy et al. 2007) were obtained from the Arabidopsis Biological Resource Center (ABRC) at Ohio State University, Columbus, USA. The atggt4 line (GK-631A04.01) was provided by the Nottingham Arabidopsis Stock Centre (University of Nottingham, Loughborough, UK).

A. thaliana seeds were surface sterilized in a mixture of 17% sodium hypochlorite (v/v) and 4% triton-X (v/v), and allowed to germinate on plates containing Arabidopsis nutrient solution (Haughn and Somerville 1986), 1.5% sucrose and 1% agar. Plates were routinely kept in the dark for 2 days at 4°C to break seed dormancy, and then incubated in a near vertical position at 23°C with a 16 h light/8 h dark cycle. Day 0 of grow is defined as the time when plates were transferred to 23°C. Plants were grown to maturity on metromix 350 supplemented with 1/1,000 Hyponex (Hyponex, Japan).

Genotype analysis

The homozygous plants containing the atggt1-1 T-DNA insert were screened by genomic PCR analysis using the T-DNA left border primer LBa1 (5′-TGGTTCACGTAGTGGGCCATCG-3′) and the AtGGT1 specific primers At1-5.1 (5′-GCAGAGAGTCTGAACAATCGCT-3′) and At1.2-3.1 (5′-TTCCCCGGGAACGCCTACTGA-3′). The atggt2 allele was verified using LBa1, At2-5g.1 (5′-ACATTGACCAACAATAGATGGA-3′) and At1.2-3.1 primers. The genotype of the atggt4 line was confirmed using the T-DNA left border primer Gabi-Kat (5′-ATATTGACCATCATACTCATTGC-3′) and the AtGGT4 specific primers At3-5O (5′-TGACGCAATCATCGCCGATCCT-3′) and At3-3.3 (5′-GTCTCAAGACTCTGAGCTAGCT-3′). The atpcs1 allele was confirmed using the T-DNA left border primer Sail-LB1 (5′-GCCTTTTCAGAAATGGATAAATAGCCTTGCTTCC-3′), and the AtPCS1 specific primers PCS5.2 (5′-AGCGCAGAGGAGAAGTCTAGGCT-3′) and PCS3.1 (5′-GTCTCAAGACTCTGAGCTAGCT-3′). The pad2-1 mutation was analyzed by using cleaved amplified polymorphic sequence (CAPS) markers. There was a SphI site in the 202 bp PCR product amplified from wild type DNA with Pad2F (5′-GTTTCGTGCTGGTCTTGCTTTAC-3′) and Pad2R (5′-CTGAATCTAGATACCTTCGCATG-3′) that was absent in the PCR products amplified from pad2-1 genotypes (J. Glazebrook, personal communication).

Genetic crosses

Plants harboring the atggt, pad2-1 and atpcs1 mutations were selected in F2 populations by genotyping analysis, and the homozygous double and triple mutants in the F3 generation were used for phenotypic analysis.

RNA isolation and RT-PCR analysis

Total RNA was isolated using RNeasy mini kit (Qiagen, Germany) following manufacturer’s instructions. First strand cDNA was synthesized from about 1 µg total RNA from Arabidopsis siliques and seedlings, using PrimeScript™ first strand cDNA synthesis kit (Takara, Japan) as described by the manufacturer. PCR was carried out under standard conditions, using At3-5.2 (5′-CCTCAACCGTGAATTACCGT-3′) and At4g29210 3′-2 (5′-CTCCACCTGATAGCTCCT-3′), which leads to the amplification of a 456 bp AtGGT4 cDNA fragment. A 512 bp AtPCS1 cDNA fragment was amplified using the primer set AtPCS1 (PCS5.2 and PCS3.1). The primer set ACT2 (5′-GTTGGTGATGAAGCACAA-3′ and 5′-CAAGACTTCTGGGCATCT-3′) was used to amplify a 425 bp fragment of Actin2 cDNA as an internal standard of gene expression.

Analysis of GSH metabolism

Seedlings of wild type, atggt single mutants, and atggt double mutants were grown in solid medium for 8 days, and then transferred to control liquid medium, and liquid medium containing 100 µM GSH (reduced glutathione) or 50 µM GSSG (oxidized glutathione) under sterile conditions. After incubation of the seedling for different times, the amounts of GSH, Cys-Gly, Cys, and γ-Glu-Cys present in the different seedlings and liquid media were determined.

GSH S-conjugate degradation analysis

Seedlings of wild type, atggt, atggt double mutants, atpcs1, atggt atpcs1 double and triple mutant strains were grown in solid medium for 8 days, and then transferred to liquid medium containing 30 µM monochlorobimane (mBCl) with and without supplementation of 30 µM Cd₂⁺ to activate PCS. After incubation of the seedling for 18 h under sterile conditions, the amounts of GSH, Cys-Gly, Cys, and γ-Glu-Cys conjugates of bimane were determined.

Measurement of thiol compounds

Total thiol compounds (GSH, Cys-Gly, Cys, and γ-Glu-Cys) were extracted and analyzed by HPLC after derivatization with monobromobimane (mBBr) as described elsewhere (Prieto et al. 2009). To measure in vivo mBCl-labeled thiol compounds the samples were analyzed without mBBr derivatization.

Results

Identification of new atggt and atpcs1 mutants

In order to undertake the functional characterization of Arabidopsis GGT and PCS isoforms, we obtained new T-DNA insertion mutants for both AtGGT4 and AtPCS1 (Figure 1). The genotype of the mutants was confirmed by genomic PCR, and homozygous were recovered at the expected 1/4 ratio for single gene insertion. DNA sequencing of the corresponding PCR products indicated that atggt4 has a T-DNA inserted into the third exon, and atpcs1 has a T-DNA insertion in the eighth exon (Figure 1A). RT-PCR expression analysis indicated that atggt4 and atpcs1 showed no detectable AtGGT4 and AtPCS1 mRNA accumulation, respectively, compared to wild type (Figure 1B). Therefore, the atggt and atpcs1 strains used in the present study appear to be null mutant alleles (Figure 1B; Prieto et al. 2009).

Figure 1. atggt, pad2-1/gsh1, and atpcs1 mutants. (A) Genomic structure of AtGGTs and AtPCS1 genes. Rectangles and lines represent exons and introns, respectively. Triangles indicate the position of T-DNA insertions (atggt1 and atggt2, Prieto et al. 2009; atggt4 and atpcs1, this paper). (B) RT-PCR analysis. Seven-day-old wild type, atggt4, and atpcs1 seedlings were analyzed for expression of AtGGT4 and AtPCS1. (C) Partial nucleotide and amino acid sequences of PAD2/GSH1. Comparison of pad2-1 and wild type (Col) nucleotide sequences identified a single G to A transition at position 1,697 from the start codon of the gene, which resulted in replacement of a serine by an asparagine residue at position 298 in the 522 amino-acid protein (Parisy et al. 2007). (D) The pad2-1 and atpcs1 mutants are hypersensitive to cadmium. Wild type, pad2-1 and atpcs1 seedlings were grown for 12 days in solid medium supplemented with and without 1 µM Cd₂⁺. Bars=1 cm.

Next, we analyzed the level of sensitivity of wild type, pad2-1, atggt, and atpcs1 mutant to cadmium. As previously described for others GSH and PC biosynthesis mutants (Howden et al. 1995a, b), pad2-1 and atpcs1 mutants showed reduced growth in the presence of cadmium compared with wild type. On the other hand, no significant difference in the growth or appearance of atggt mutants compared with the wild type was observed under our experimental conditions (data not shown).

AtGGT1 is involved in GSH/GSSG catabolism in seedlings

In order to evaluate the role of AtGGT in GSH homeostasis, we analyzed the GSH metabolic rate in wild type, atggt1, atggt2, atggt4, atggt1 atggt4, and atggt2 atggt4 seedlings. After incubation of the seedlings in control (glutathione-free) liquid medium, and 100 µM GSH or 50 µM GSSG containing media, the amounts of GSH, Cys-Gly, Cys, and γ-Glu-Cys present in seedlings and liquid media were determined.

GSH and GSSG were lost in similar way from the solution of wild type and atggt mutants (Supplementary Figure S1). Also, it is noteworthy to indicate that we could not find significant differences between wild type and atggt seedlings in terms of GSH, Cys, and γ-Glu-Cys contents (Supplementary Figures S1, S2). In contrast, the amount of Cys-Gly present in the GSH or GSSG solution, and the amount of Cys-Gly in seedlings were significantly lower in strains harboring the atggt1 mutation compared to wild type and the rest of the mutants (Figure 2A, B). Moreover, the amount of Cys-Gly present in the medium of atggt1 mutants supplemented with GSSG was significantly lower compared to that of the medium supplemented with GSH (Figure 2A). Taken together, these results indicate that AtGGT1 is mainly involved in GSSG catabolism.

Figure 2. GSH degradation analysis in seedlings. Wild type, atggt1, atggt2, atggt4, atggt1 atggt4, and atggt2 atggt4 seedlings were incubated in control liquid medium, and liquid medium containing 100 µM GSH or 50 µM GSSG. (A) The liquid media was sampled at different times and used for Cys-Gly quantification. (B) After incubation for 18 h, the amount of Cys-Gly in seedlings was determined. Values are means±SD of three independent experiments. *=significant difference from wild type by Student’s t-test (p<0.05). △=significant difference from the liquid medium of atggt1 mutants supplemented with GSH by Student’s t-test (p<0.05).

Characterization of GSH metabolism in siliques

Since AtGGT2 is mainly expressed in siliques (Martin et al. 2007; Ohkama-Ohtsu et al. 2007a, b; Prieto et al. 2009), next we analyzed the GSH metabolic rate in siliques of wild type, atggt1, atggt2, atggt4, atggt1 atggt4, and atggt2 atggt4 mutants. However, no significant difference in the steady-state levels of GSH, Cys-Gly, Cys, and γ-Glu-Cys between atggt and wild type siliques were observed (Figure 3, data not shown). Since these results may be explained by a lower GSH biosynthesis rate in atggt mutants relative to wild type, we constructed and analyzed GSH biosynthesis mutants (pad2-1/gsh1) containing atggt mutations by genetic crosses. As illustrated in Figure 3, the pad2-1 mutant exhibited lower level of GSH and higher level of Cys compared to wild type. In contrast, pad2-1 atggt1, and pad2-1 atggt2 double mutants contained higher amount of GSH and lower amount of Cys compared to pad2-1. These results suggest the possibility that AtGGT1 and AtGGT2 are involved in GSH catabolism in siliques.

Figure 3. Quantification of thiol metabolites in siliques. The GSH (A) and Cys (B) contents in wild type, atggt1 atggt4, pad2-1, pad2-1 atggt1, and pad2-1 atggt2 siliques were determined. Values are means±SD of three independent experiments. An asterisk indicates a significant difference from pad2-1 by Student’s t-test (p<0.05).

AtGGT1, AtGGT4, and AtPCS1 are involved in GSH S-conjugate catabolism in seedlings

In order to evaluate the role of AtGGT and AtPCS in detoxification of xenobiotic compounds, we used monochlorobimane (mBCl) as a model xenobiotic to follow GSH S-conjugate turnover. Seedlings of wild type, atggt1, atggt2, atgg4, atggt1 atggt4, atggt2 atggt4, atpcs1, atpcs1 atggt1, atpcs1 atggt4, atpcs1 atggt1 atggt4 mutants were transferred to liquid medium containing 30 µM mBCl with and without supplementation of 30 µM Cd₂⁺ to activate AtPCS. After an incubation period of 18 h, the seedlings were analyzed for in vivo mBCl-labeled thiol compound levels (Figures 4–6).

Figure 4. AtGGT and AtPCS1 mediated GS-bimane catabolism. Wild type, atggt1, atggt2, atggt4, atggt1 atggt4, atggt2 atggt4, atpcs1, atpcs1 atggt1, atpcs1 atggt4, atpcs1 atggt1 atggt4 seedlings were exposed to 30 µM mBCl in the presence (A) and absence (B) of 30 µM Cd₂⁺. After an incubation period of 18 h, the seedlings were harvested and the levels (%) of GSH■, Cys-Gly■, Cys■, and γ-Glu-Cys□ conjugates of bimane were determined. Values are means of three independent experiments.

Figure 5. AtGGT1 is involved in GS-bimane catabolism. Wild type, atggt1, atggt2, atgg4, atggt1 atggt4, atggt2 atggt4, atpcs1, atpcs1 atggt1, atpcs1 atggt4, atpcs1 atggt1 atggt4 seedlings were exposed to 30 µM mBCl in the presence (A) and absence (B) of 30 µM Cd₂⁺. After an incubation period of 18 h, the seedlings were harvested and analyzed for Cys-Gly-bimane levels. Values are means±SD of three independent experiments. An asterisk indicates a significant difference from wild type by Student’s t-test (p<0.05).

Figure 6. AtGGT4 and AtPCS1 are involved in GS-bimane catabolism. Wild type, atggt1, atggt2, atggt4, atggt1 atggt4, atggt2 atggt4, atpcs1, atpcs1 atggt1, atpcs1 atggt4, atpcs1 atggt1 atggt4 seedlings were exposed to 30 µM mBCl in the presence (A) and absence (B) of 30 µM Cd₂⁺. After an incubation period of 18 h, the seedlings were harvested and the amount of GSH-bimane, Cys-bimane, and γ-Glu-Cys-bimane were determined. Values are means±SD of three independent experiments. *=significant difference from wild type by Student’s t-test (p<0.05). △=significant difference from atpcs1 by Student’s t-test (p<0.05).

As previously described by Ohkama-Ohtsu et al. (2007b) and Grzam et al. (2007), mutants harboring the atggt4 mutation contained higher levels of GS-bimane and lower levels of Cys-bimane compared to the rest of strains, which indicates that AtGGT4 is involved in GS-bimane catabolism. On the other hand, mutants harboring the atggt1 mutation exhibited lower levels of Cys-Gly-bimane compared to wild type (Figures 4, 5), which indicates that AtGGT1 is involved in the conversion of GS-bimane into Cys-Gly-bimane.

No significant difference in GS-bimane degradation was found between wild type and atpcs1 seedlings incubated in medium without Cd₂⁺ (Figures 4–6). In contrast, when incubated in medium containing Cd₂⁺, the strains harboring the atpcs1 mutation contained higher amounts of GS-bimane and lower amounts of γ-Glu-Cys-bimane compared to wild type (Figures 4–6). Moreover, the atpcs1 atggt4 and atpcs1 atggt1 atggt4 mutants showed higher GS-bimane levels than atpcs1 (Figures 4–6). Taken together, these results indicate that AtPCS1 plays an important role in the conversion of GS-bimane into γ-Glu-Cys-bimane in seedling exposed to heavy metals such as Cd₂⁺. On the other hand, we could not find significant differences between wild type and atggt seedlings, and between atpcs1 and atpcs1 atggt seedlings in terms of γ-Glu-Cys-bimane contents (Figures 4, 6), which indicates that AtGGTs do not play a significant role in the degradation of γ-Glu-Cys-bimane.

Discussion

The Arabidopsis genome contains three genes encoding functional GGTs, and two genes encoding PCSs. AtGGT1 and AtGGT2 have been proposed to be localized in the apoplastic space (Destro et al. 2011; Martin et al. 2007; Ohkama-Ohtsu et al. 2007a), whereas AtGGT4 is considered to be localized in the vacuole (Grzam et al. 2007; Ohkama-Ohtsu et al. 2007b). On the other hand, AtPCS1, which is the major player in PC biosynthesis, has been reported to be localized in the cytosol (Blum et al. 2010). In the present study, which is based on the analysis of pad2-1/gsh1, atggt, and atpcs1 single mutants, as well as atggt, atggt pad2-1, atggt atpcs1 double and triple mutants, we present new evidence to elucidate the role of AtGGTs in GSH catabolism, and the mechanism of interaction between AtGGT and AtPCS mediated degradation of GSH S-conjugates.

Giaretta et al. (2017) reported that apoplastic AtGGT1 and AtGGT2 silencing induces a decrease in the number of organs with a high GSH demand such as seeds and trichomes, which suggest that the degradation of GSH translocated in the phloem functions for delivering GSH and supplying Cys to sink tissues. On the other hand, it has been suggested that AtGGT1 is important in preventing oxidative stress by metabolizing extracellular GSSG in the apoplast (Ohkama-Ohtsu et al. 2007a). Besides, it has been shown that single atggt1 and atggt2 mutants were unable to abolish apoplastic GGT (Martin et al. 2007; Ohkama-Ohtsu et al. 2007a; Prieto et al. 2009), and one study showed evidence of a compensatory overexpression of AtGGT2 in atggt1 roots (Destro et al. 2011). The results of our HPLC analysis using atggt, and atggt pad2-1 mutants showed that siliques of pad2-1 atggt1, and pad2-1 atggt2 mutants contained higher levels of GSH and lower levels of Cys compared to the pad2-1 mutant (Figure 3), which indicates that glutathione homeostasis is disturbed in atggt mutants. Even though we cannot discard the possibility that atggt1 and atggt2 mutations lead to an increased oxidative stress resulting in induction of metabolism toward GSH synthesis in the pad2-1 genetic background, our results suggest that AtGGT1 and AtGGT2 could play an important role in GSH catabolism in siliques. On the other hand, we tested the hypothesis of AtGGT1 having a role in GSH catabolism in seedlings, and the salvage of exogenous and extracellular GSH. In barley seedlings, the uptake of GSH/GSSG from nutrient solution and the capacity to degrade GSH/GSSG to Cys-Gly in the external medium was impaired when serine/borate, a competitive inhibitor of GGT, was added to the solution (Ferretti et al. 2009). The results of our HPLC analysis indicate that the amount of Cys-Gly present in the GSH or GSSG containing liquid media, and the amount of Cys-Gly in seedlings were significantly lower in strains harboring the atggt1 mutation (Figure 2A, B). Moreover, the amount of Cys-Gly present in the GSSG containing medium of strains harboring the atggt1 mutation was significantly lower compared to that of the GSH containing medium (Figure 2A). On the other hand, we could not find significant differences between wild type and atggt mutants in terms of GSH uptake (Supplementary Figure S1), which may be explained by the compensatory expression of AtGGT2 in atggt1 mutant lines (Destro et al. 2011). Taken together, our results indicate that AtGGT1 is mainly involved in GSSG catabolism, and support the idea of AtGGT having an important role in adaptation to oxidative stress (Figure 7), GSH/GSSG degradation to its constituent aminoacids for uptake from the external medium, and delivery of GSH to sink tissues such as seeds.

Figure 7. GSH and GS-conjugate catabolic pathways. γ-ECS: γ-glutamylcystein synthetase, GSHS: glutathione synthetase, GST: glutathione S-transferase, GGT: γ-glutamyltransferase, DPase: dipeptidases, PCS: phytochelatin synthase, GS-X: GSH S-conjugate, X: xenobiotic. Exposure to heavy metals such as Cd₂⁺ induces AtPCS-mediated PC biosynthesis and GS-X degradation.

Blum et al. (2010, 2007) found that AtPCSs play an important role in the cytosolic conversion of GS-bimane into γ-Glu-Cys-bimane and Gly in Arabidopis seedlings and protoplasts. Under our experimental conditions, we couldn’t find significant differences in GS-bimane degradation between wild type and atpcs1 seedlings incubated in medium without Cd₂⁺ (Figures 4–6). In contrast, as previously reported by Blum et al. (2007), atpcs1 seedlings incubated in media containing Cd₂⁺ presented higher levels of GS-bimane and lower levels of γ-Glu-Cys-bimane compared to wild type (Figures 4, 6). On the other hand, the results of our HPLC analysis confirm that AtGGT1 and AtGGT4 are involved in the conversion of GS-bimane into Cys-Gly-bimane (Figures 4–6). Our results also indicated that AtGGTs do not play a significant role in γ-Glu-Cys-bimane degradation (Figures 4, 6), which strongly suggests that AtGGT and AtPCS1 participate in two independent pathways related with GS-bimane catabolism. Taking together, our results indicate that AtGGTs are involved in the apoplastic (AtGGT1) and vacuolar (AtGGT4) catabolism of GSH S-conjugates of xenobiotics (GS-bimane), whereas AtPCSs are involved in their cytoplasmic degradation (Figure 7).

While GGTs control GSH and GSH S-conjugate degradation in the apoplastic and vacuolar compartments, alternative cytosolic pathways of GSH catabolism by means of γ-glutamylcyclotransferase (GGCT, EC2.3.2.4) and γ-glutamylpeptidase (GGP, EC3.4.19.16) have been proposed to dominate these processes in the cytosol (Ito et al. 2022; Kumar et al. 2015; Ohkama-Ohtsu et al. 2008; Paulose et al. 2013). Arabidopsis contains three potential GGCT genes (GGCT1=At1g44790, GGCT2;1=At5g26220, GGCT2;2=At4g31290) encoding proteins that cleave GSH when overexpressed in yeast (Kumar et al. 2015; Paulose et al. 2013). Characterization of ggct2;1 mutants suggests that GGCT2;1 may protect Arabidopsis plants from heavy metal toxicity by recycling glutamate to maintain GSH homeostasis (Paulose et al. 2013). GGCT2;1 has also been proposed to play an important role in utilization of GSH as a source of Cys during S-starvation (Joshi et al. 2019). On the other hand, the Arabidopsis genome contains five potential genes encoding GGP (GGP1=At4g30530, GGP2=At4g30540, GGP3=At4g30550, GGP4=At2g23960, GGP5=At2g23970) (Geu-Flores et al. 2011). GGP1 and GGP3 were originally shown to be involved in the processing of GSH S-conjugates in the glucosinolate and camalexin pathways (Geu-Flores et al. 2011, 2009). However, Ito et al. (2022) have recently reported that recombinant GGP1 and GGP3 are capable of hydrolyzing the γ-glutamyl residue of GSH in vitro. Moreover, yeast complementation assays, and the characterization of ggp1 mutants suggest that GGP1 play a major role in cytoplasmic degradation of GSH in Arabidopsis (Ito et al. 2022). The complex biological significance of the different Arabidopsis GGT, GGCT, and GGP isoforms, and the mechanism of interaction between GGT, GGCT, and GGP mediated GSH catabolism should be determined in further studies.

Abbreviations

GGT

γ-glutamyltransferase

GSH

glutathione

GS-X

GSH S-conjugate

PCS

phytochelatin synthase