The Emerging Roles of γ-Glutamyl Peptides Produced by γ-Glutamyltransferase and the Glutathione Synthesis System

Proteins and peptides are produced via the translation of genetic codes found in mRNA by ribosomes; some small peptides, however, are produced solely through sequential enzymatic reactions that depend on substrate specificity, but are independent of the genetic information encoded in the nucleotide sequences of DNA. These small peptides may contain non-protein amino acids or promote non-canonical peptide bonds between amino acids. Glutathione is the most abundant peptide of this category and exhibits pleiotropic functions [ 1 , 2 ]. Glutathione, in its reduced form (GSH), is produced via the coordinated action of γ-glutamyl-cysteine synthetase (γ-GCS) and glutathione synthetase (GS) without referencing genetic information. γ-GCS first ligates the γ-carboxyl group of L-glutamate (Glu) to the amino group of L-cysteine (Cys), which results in the production of L-γ-glutamyl-L-cysteine (γ-Glu-Cys) [ 3 ]. GS then adds a glycine (Gly) unit to γ-Glu-Cys, which leads to the formation of the tripeptide γ-Glu-Cys-Gly (i.e., GSH). Due to its broad substrate specificity, however, γ-GCS could be the producer of many other γ-glutamyl peptides, which are largely dependent on the availability of free amino acids. The liver is the central organ that predominantly synthesizes GSH and secretes it into the bloodstream ( Figure 1 ). It is generally accepted that glutathione, whether in its intact form, as conjugates, or as an oxidized dimer designated as glutathione disulfide (GSSG), is secreted from cells via the multidrug resistance-associated protein (MRP), which is a protein in the ABCC subclass of the ABC transporter family [ 4 ]. In circulation, the γ-glutamyl group appears to act as a protector against the peptidases that are abundant in blood plasma [ 5 ]. Some cells, however, express γ-glutamyltransferase (GGT) on the plasma membrane, which is involved in the initiation of glutathione degradation outside cells and helps recruit constituent amino acids [ 6 ]. GGT either hydrolytically removes the γ-glutamyl moiety of glutathione or produces a variety of γ-glutamyl peptides by transferring the γ-glutamyl moiety to an amino acid or dipeptide; these processes depend on the abundance of acceptor molecules [ 7 ]. Although both γ-GCS and GGT are responsible for producing γ-glutamyl peptides, since their discovery, the roles of the resultant γ-glutamyl peptides in vivo have remained unclear. The role of γ-glutamyl peptides, such as glutathione, as systemic allosteric modulators of calcium-sensing receptors (CaSRs) expressed on the plasma membrane, has attracted much attention, particularly in the fields of food chemistry [ 8 , 9 ] and neuroscience [ 10 ]. Since the production of γ-glutamyl peptides is elevated under certain pathological conditions, it is conceivable that they transmit signals from damaged cells to surrounding cells via the modulation of CaSR action. Here, we briefly overview glutathione function and its metabolism, with a particular focus on GGT. We then discuss the potential roles of the γ-glutamyl peptides that are extracellularly produced by GGT, which are compounds that have received scant attention so far.

The essential roles that glutathione plays in maintaining cellular redox homeostasis are postulated to occur in most types of cells. The reactivity of GSH towards reactive oxygen species (ROS) is marginal and, in fact, the antioxidant action is largely accomplished by donating electrons to glutathione peroxidase (GPX) [ 25 ]. GPX uses two electrons from two GSH molecules to achieve the reductive detoxification of peroxides, which are oxidized into GSSG. While mammals carry eight genes encoding canonical GPX, proteins such as GST [ 26 ] and peroxiredoxin [ 27 ] also show GSH-dependent peroxidase activity to some extent. Until the discovery of ferroptosis, it was unclear whether a GSH deficiency could be closely associated with a specific type of cell death, although the association of decreased GSH levels with oxidative stress-related cell death is implied in many diseases [ 28 ]. Ferroptosis is an iron-dependent form of regulated necrosis that was first discovered in cultured cells, wherein the function of xCT, a transporter of oxidized Cys referred to as cystine, is inhibited by erastin [ 29 ]. Glutathione levels are decreased via xCT inhibition due to the deprivation of cellular Cys, and enzymes that require GSH are disabled. Among the GSH-requiring enzymes, GPX4 reduces phospholipid hydroperoxide to corresponding alcohol and is one of the most potent enzymatic suppressors of ferroptosis [ 30 , 31 ]. Iron plays the primary role in the lipid peroxidation reaction, so that iron chelation also effectively suppresses ferroptosis. Under conditions with disabled GPX4, lipid peroxidation products accumulate, which leads to membrane rupture [ 32 ]. Because ferroptosis is considered to be involved in a variety of diseases, such as neurodegenerative disease, ischemic disease, and cancer [ 33 ], the primary role of the Cys-GSH-GPX4 axis in combating ferroptosis has begun to attract much attention. Although this issue is extremely important from the aspect of glutathione function, an in-depth discussion of ferroptosis is beyond the scope of this article. Therefore, interested readers should refer to the review articles dedicated to this issue [ 33 , 34 , 35 , 36 , 37 , 38 ]. Glutaredoxin, also called thioltransferase, is encoded by the gene GLRX and protects the SH groups in the Cys residue in proteins from oxidative modification via a reduction in the reducing equivalent of GSH [ 14 , 39 ]. Cysteine sulfhydryl (Cys-SH) forms four oxidation states: cysteine sulfenic acid (Cys-SOH), cysteine sulfinic acid (Cys-SO 2 H), cysteine sulfonic acid (Cys-SO 3 H), and a disulfide. While Cys-SOH and the disulfide can be reversibly reduced to Cys-SH via the use of either physiological reductants or enzymatic reactions, Cys-SO 2 H and Cys-SO 3 H generally cannot be reduced back to Cys-SH via biological systems [ 40 ]. Under oxidative conditions, glutathione forms a conjugation with Cys-SH and more preferentially with Cys-SOH in proteins, which is called S-glutathionylation [ 41 ]. While Cys-SOH is prone to oxidation to either Cys-SO 2 H or Cys-SO 3 H, such oxidation is avoided by competitive mixed disulfide formation with glutathione. When cells are recovered from oxidative stress, S-glutathionylated Cys residue can be reduced back to Cys-SH residue by means of another GSH, which is accelerated via the catalytic action of GLRX. Thus, S-glutathionylation could prevent the irreversible oxidation of proteins under oxidative conditions. GLRX is encoded by two genes in mammals: GLRX1 and GLRX2. While GLRX1 resides mainly in the cytoplasm, GLRX2 is localized either in the mitochondria or nuclei [ 42 , 43 ].

γ-GCS is responsible for the production of γ-Glu-Cys, and then, GS completes glutathione formation via the ligation of a Gly unit. The enzymatic and protein chemical properties of γ-GCS have been precisely reviewed by Griffith [ 3 ], so the subject matter is only briefly described here. γ-GCS is a heterodimeric enzyme composed of a catalytic subunit that is encoded by the GCLC gene and a modifier subunit that is encoded by the GCLM gene [ 15 ]. Physiological levels of GSH suppress γ-GCS activity in an allosteric manner. The GCLC protein is responsible for catalysis and is regulated by a specific inhibitor, buthionine sulfoximine, which decreases glutathione to negligible levels after one day of treatment in most types of cells during cultivation [ 54 ]. This indicates that γ-GCS activity is the main component for maintaining glutathione levels inside cells. The generation of γ-Glu-Cys by the γ-GCS-catalyzed ligation reaction proceeds essentially in two partial reaction steps [ 55 ]. In the first step, Glu is phosphorylated by ATP to form a carboxylic phosphoric anhydride, γ-glutamyl phosphate. Then, this activated γ-carboxyl group is attacked by the amino group of Cys to form γ-Glu-Cys. Cys is the preferable amino acid substrate that is γ-glutamylated in the γ-GCS reaction (the Michaelis constant (Km) = 0.1–0.3 mM), but Cys-mimetic 2-aminobutyric acid (2AB) is also a good substrate [ 3 ]. Moreover, other amino acids also take the place of Cys, albeit to a lesser extent, as described in the following section. This broad specificity to the substrate allows for the production of a variety of γ-glutamyl peptides inside cells. Mice with a genetic ablation of GCLC exhibit embryonic lethality [ 56 ]. While at least six children with GCLC deficiency are reported to have shown anemia [ 57 ], significant levels of glutathione remained in their red blood cells, which suggests the preservation of some γ-GCS activity. GS is a homodimeric enzyme that adds a Gly unit to γ-Glu-Cys and also to γ-Glu-2-aminobutyrate (2AB), which results in GSH and γ-Glu-2AB-Gly, which is referred to as ophthalmic acid (OPT), respectively. A genetic deficiency of GSS encoding GS is quite rare and shows a relatively mild phenotype [ 58 , 59 ]. GSS-deficient patients experience 5-oxoprolinuria, hemolytic anemia, and neurological dysfunction. Whereas glutathione levels are decreased in the fibroblasts of GSS-deficient patients, Cys and γ-Glu-Cys levels increase [ 60 ]. Hemolytic anemia and neurological dysfunction can be explained as a decrease in glutathione levels. The excessive production of L-5-oxoproline (5-OP) causes 5-oxoprolinuria, and Cys sometimes accumulates inside the cells of patients with a GSS deficiency. In the case of the γ-GCS reaction, the generation of 5-OP and the accumulation of Cys are contradictory phenomena because 5-OP may be produced when γ-glutamyl phosphate does not meet Cys [ 55 ]. Another hypothetical mechanism involves a certain isozyme of γ-glutamylcyclotransferase, which produces 5-OP from glutathione and could also actively cleave γ-Glu-Cys to 5-OP and Cys. Currently, however, there is no evidence to support either mechanism.

There is a group of cell-penetrating (permeable) peptides that traverse the phospholipid bilayer barrier without the involvement of a transporter protein [ 66 ]. However, due to its hydrophilic nature, GSH does not follow this process. While S. cerevisiae import GSH into the cells via Hgt1p in a proton-coupled manner [ 67 , 68 ], no orthologous genes are known to exist in mammals. Although mitochondrial GSH import is mediated by a probable transporter protein, SLC25A39 [ 69 ], neither GSH nor related γ-glutamyl peptides appear to enter mammalian cells without modification, e.g., via esterification [ 70 ]. Glutathione is actively degraded on the surface of some types of cells under physiological and pathophysiological situations ( Figure 3 ). Extracellular GSH and GSSG experience removal of the γ-glutamyl moiety by GGT, which is a plasma membrane-anchored glycoprotein, and are further hydrolyzed into amino acids by dipeptidase [ 7 ]. A similar degradation process is also involved in the metabolism of glutathione S-conjugate with an aromatic compound, which is finally metabolized to mercapturic acid via the N-acetylation of the cysteine S-conjugate, and is then excreted from the body through the urine. The human GGT gene family, which includes GGT-related genes along with pseudogenes, has been classified by Heisterkamp et al. [ 71 ] in collaboration with the HUGO Gene Nomenclature Committee (HGNC). We adopted the proposed nomenclature in this article. 4.1. γ-Glutamyltransferase The reaction of glutathione with amino acids to yield peptides containing glutamic acid was first described by Hanes et al. in 1950 [ 72 ], and this led to the observation that an enzyme from the kidney catalyzes removal of the γ-glutamyl group from glutathione and the formation of a new γ-glutamyl bond with various acceptor amino acids, and thereby produces several types of γ-glutamyl amino acids. Thus, the enzyme was referred to as γ-glutamyl transpeptidase (γ-GTP) [ 73 ] and thereafter as γ-glutamyltransferase (GGT). The hydrolytic reaction was also found to be catalyzed by the same enzyme, which transfers a γ-glutamyl group to a water molecule. Mammalian GGT also exhibits glutaminase activity, by which glutamine is hydrolyzed to glutamic acid and ammonia, as indicated by the identification of GGT as maleate-stimulated glutaminase [ 74 , 75 ]. Mammalian GGT is a heterodimeric type-II transmembrane glycoprotein that is translated as a single precursor and post-translationally processed into a dimeric form by autocatalytic limited proteolysis [ 76 ]. The enzyme protein is anchored to a cell surface, and can be released into various body fluids via cleavage of the membrane-anchor domain and/or other non-hydrolytic processes [ 77 , 78 , 79 , 80 ]. GGT has been a traditional biomarker of hepatobiliary diseases, and recently has also been considered a potential biomarker for various other diseases [ 81 ]. Measurements of the blood levels of enzyme activity are of particular clinical importance in the diagnoses of liver diseases such as alcoholic damage and carcinogenesis. Many studies have accumulated evidence that this enzyme is induced by the administration of rugs and xenobiotics, including ethanol, and during carcinogenesis in various tissues. Elevation of the enzyme activity in the blood seems to be caused either by the increased expression of enzyme protein in lesions or by enhanced liberation from cells—and possibly by both of these factors. In clinical chemistry testing, various “GGT iso(en)zymes” have been observed in electrophoretic separation and have been investigated to improve diagnostic accuracy [ 79 ]. The appearance of different forms of GGT in clinical chemistry testing appears to be attributed to variation in the N-terminal structure, which is due to non-specific limited proteolysis as well as to an association with exosomes [ 80 ] and/or simply to the structures of glycans [ 82 ].

In addition to the transpeptidation reaction, GGT also catalyzes the hydrolysis of the γ-glutamyl bonding of γ-glutamyl substrates. Furthermore, when glutamine is used as the substrate, the enzyme hydrolyzes the side chain amide group, yielding glutamic acid and ammonia. All these reactions certainly appear to involve an acyl enzyme intermediate, which is the γ-glutamyl enzyme. The intermediate is subsequently attacked by amino groups of the acceptors or by water, and, thus, produces either new γ-glutamyl compounds or glutamic acid, respectively. The transfer reaction of the γ-glutamyl moiety could be considered GGT-mediated re-distribution and propagation of the γ-glutamyl group among various biological molecules in organisms. When γ-glutamyl-p-nitroanilide is used as the substrate for hydrolysis, the formation step of the acyl enzyme is faster than the step for the water-conducted breakdown of the intermediate, as shown by the faster reaction rate for the transpeptidation reaction compared with that of hydrolysis [ 86 ]. The nucleophilic attack of an amino group toward the enzyme intermediate is more efficient than that of water. However, since the simultaneous hydrolysis is not necessarily marginal compared with that of transpeptidation, researchers have long questioned whether the biochemical significance of the enzyme is attributed to transpeptidation or to the hydrolysis of γ-glutamyl compounds that include glutathione [ 87 ]. In addition to the usual transpeptidation and hydrolysis of the γ-glutamyl group, GGT also catalyzes autotranspeptidation, which is an unusual type of transpeptidation [ 88 ]. When the enzyme is allowed to react with a γ-glutamyl donor in the absence of an acceptor substrate, the γ-glutamyl group is transferred to another molecule of the γ-glutamyl donor, which results in the formation of a γ-glutamyl-γ-glutamyl compound and/or a γ-glutamyl-γ-glutamyl-γ-glutamyl compound [ 89 ]. Autotranspeptidation occurs only when using the L-stereoisomer of the γ-glutamyl group rather than the D-isomer, which is consistent with evidence that the D-isomers of amino acids do not serve as an acceptor for the reaction [ 90 ]. When the D-γ-glutamyl donor is used as a substrate, the enzyme exhibits only hydrolysis and not autotranspeptidation [ 86 ]. When the hydrolytic activity is assessed using a L-γ-glutamyl donor without any acceptor substrate, it should be noted that the appropriate concentration of the donor must be chosen. The concentration of the donor should be sufficiently low for the transpeptidation activity to be negligible. In the absence of an acceptor, a double reciprocal plot for the donor substrate shows a “substrate activation” profile rather than a straight curve, as indicated by the downward curvature with a decreasing 1/[substrate] [ 86 , 91 ]. The autotranspeptidation becomes noticeable as the substrate concentration increases, because the apparent Km value for the single substrate as an acceptor is much higher than its value as a donor.

Although GGT had long been considered a unique enzyme with γ-glutamyl bond-cleaving activity, particularly with respect to glutathione metabolism, another closely related but clearly different enzyme, referred to as the “GGT-related enzyme (GGT-rel)”, is known to catalyze the hydrolysis of glutathione [ 96 ]. Further studies have identified several genes with nucleotide or amino acid sequences that are similar to the most “traditional” γ-glutamyl transferase, the gene symbol of which has been systematically designated as GGT1. Thus, these related genes constitute a gene family, which includes at least 13 members that have been concisely summarized by Heisterkamp et al. [ 71 ]. Of the 13 genes, at least 6 appear to be active in terms of expression. Nevertheless, the protein products of only two genes, GGT1 and GGT5, the latter of which is the approved gene symbol for GGT-rel, were found to function as enzymes. GGT5 is capable of hydrolyzing glutathione but not γ-glutamyl-p-nitroanilide, which is an ordinary substrate for enzyme-activity assay. It seems most likely that one of the most important roles of GGT5 is the conversion of leukotriene C4 into D4 by hydrolyzing the γ-glutamyl bond, and therefore, the enzyme is also known as γ-glutamyl leukotrienase [ 97 ]. In fact, GGT5 is primarily expressed in the spleen, and GGT5-deficient mice indicate a defect in the LTD4 formation and the attenuation of acute inflammatory responses [ 98 , 99 ]. By contrast, GGT1-null mice show a substantial conversion of LTC4 to LTD4, which excludes GGT1 from the responsible genes [ 97 , 100 ]. The enzymatic properties of GGT5, as well as those of GGT 7, have not been well characterized in terms of the formation of γ-glutamyl peptides. Thirteen genes are systematically designated as GGT1 through GGT8 for the genes that comprise full-length whole proteins with both large and small subunits, and as GGTLC1 through GGTLC5 light-chain-only genes, wherein only the small subunit (light chain) is encoded [ 71 ]. Of the full-length protein genes, GGT3, GGT4, and GGT8 are not functional and are considered to be pseudogenes, thus being more exactly noted as GGT3P, GGT4P, and GGT8P, respectively. As described above, although GGT1 and GGT5 products have been well characterized in terms of enzymology, biochemistry, and structural biology, other full-length genes have not yet been investigated in sufficient detail. With respect to the light-chain-only genes, although GGTLC1 through GGTLC3 are functional genes, the roles of their protein products are unknown. Enzymatic activities such as the abilities of hydrolysis and the transpeptidation of γ-glutamyl compounds involve amino acid residues located in the large subunit, and therefore, it is very unlikely that these proteins exhibit the same or similar activities compared to those of γ-glutamyltransferase. On the other hand, GGTLC4 and GGTLC5 are pseudogenes, and should be noted as GGTLC4P and GGTLC5P, respectively.

The glutathione-degrading system is not very active inside cells, but the inhibition of glutathione synthesis by treatment with a γ-GCS inhibitor reduces the content to quite low levels. This is thought to be due mainly to the excretion of GSH, glutathione conjugates, and GSSG by MRP. The degradation of glutathione, however, may be accelerated under conditions such as starvation [ 104 ] and endoplasmic reticulum (ER) stress [ 105 ]. 5.1. GSH-Specific γ-Glutamylcyclotransferase Activity Inside Cells Glutathione degradation inside cells is initiated by isozymes of γ-glutamylcyclotransferase ( Figure 3 ). Initial studies have shown that γ-glutamylcyclotransferase liberates the γ-glutamyl moiety from γ-glutamyl amino acids, excluding GSH, as 5-OP [ 1 ]. While such activity has been recognized for a long time, the protein that exhibits glutathione-specific γ-glutamylcyclotransferase activity was only recently identified. The first identified enzyme was cation transport regulator protein 1 (CHAC1), which was originally reported as a proapoptotic component [ 105 ]. The CHAC family proteins consist of CHAC1 and CHAC2, and their genes are located on human chromosomes 15 and 2, respectively. CHAC specifically acts on GSH with a Km of 0.1–2 mM, which is the appropriate value for the degradation of GSH inside cells [ 106 , 107 ]. However, CHAC 1 shows activity that is approximately 20 times higher than that of CHAC 2. Studies on CHAC 1 and CHAC 2 have shown that the two isoenzymes both promote the intracellular degradation of GSH, but they appear to act in different contexts. ChaC1 is upregulated under stress conditions such as ER stress and amino acid starvation [ 105 ], whereas ChaC2 is constitutively expressed and may act in a housekeeping fashion [ 5 ]. An ATF4-CHOP cascade is involved in the induction of ChaC1 under ER stress conditions, which results in the accelerated degradation of intracellular glutathione and may be responsible for cell death. ChaC1 is also induced by xCT inhibition and therefore is considered a potential marker for ferroptosis [ 108 ]. Since the GPX4-catalyzed reduction of lipid peroxides is disabled by glutathione deprivation, elevated ChaC1 may play a role in ferroptosis by degrading glutathione [ 109 ]. While the participation of both ATF3 and ATF4 has been proposed in the induction of ChaC1 [ 110 ], the cooperation of both FOXO1 and AFT4 reportedly also plays a role in the induction [ 111 ]. The administration of paracetamol, another name for acetaminophen (APAP), induces ChaC1, and the eIF2α-ATF4 pathway that is activated under ER stress appears to be involved in this induction [ 112 ]. In any case, ATF4 appears to have a primary role in the induction of the ChaC1 gene. ChaC1, which is referred to as Botch in the report [ 113 ], is induced and promotes neurogenesis by antagonizing Notch signaling in mouse models. The enzymatic reaction that has been proposed for ChaC1/Botch by Chi et al. [ 113 ] is the formation of 5-OP by deglycinase. This enzymatic reaction appears to be unlikely, however, because ChaC1/Botch exhibits γ-glutamylcyclotransferase activity but not deglycinase activity [ 5 ]. DJ-1 is the protein that protects against the development of Parkinson’s disease (PD), and its mutation causes early-onset PD in an autosomal recessive manner [ 114 ]. Because ChaC1 is upregulated in DJ-1 knockout mice, DJ-1 is considered to maintain glutathione by suppressing ChaC1 expression, which thereby prevents PD development [ 115 ]. Mice with a knock-in of inactive mutant ChaC1 tend to sustain glutathione levels in their muscles during starvation, but otherwise show no significant changes [ 116 ]. Since mutant ChaC1-knock-in mice have been examined only in limited situations, it remains ambiguous whether a deficiency of CHAC 1 activity affects phenotypic properties under pathological conditions such as ER stress, oxidative stress, and ferroptosis. Unlike CHAC 1, CHAC 2 is enriched in undifferentiated human embryonic stem cells [ 117 ]. The downregulation of ChaC2 tends to decrease the levels of glutathione and blocks the self-renewal of these cells. Simultaneous knock-downs of ChaC1 and ChaC2, however, restore the self-renewability of the cells, which indicates that glutathione homeostasis is maintained by the balance between ChaC2 and ChaC1. ChaC2 levels are associated with some malignant diseases. For instance, ChaC2 expression is frequently downregulated in gastric and colorectal cancers, which suggests it is a tumor-suppressor gene [ 118 ]. However, the elevated expression of ChaC2 in breast cancer [ 119 ] and in hepatocellular carcinoma [ 120 ] is associated with a poor prognosis for these malignant diseases. While these observations are discrepant at a glance, a similar situation is sometimes observed in other antioxidant systems associated with tumors [ 121 ]. A possible explanation for such observations may be that antioxidant systems are originally protective against tumorigenicity t

Glutathione is a dominant reductant in most cells and is also considered to be a vehicle for transporting Cys in a safe form to cells in vivo. GGT-catalyzed removal of the γ-glutamyl moiety from glutathione releases a Cys-Gly dipeptide, which undergoes either direct uptake by peptide transporter PEPT2 in some cells or further degradation into free amino acids, Cys and Gly, via extracellular dipeptidases [ 127 , 128 ]. Several dipeptidases appear to perform this reaction in the extracellular space, and carnosine dipeptidase (CNDP) 2 has been identified as a protein that preferentially reacts on the Cys-Gly dipeptide in yeast [ 129 ]. While CNDP1 specifically degrades anti-inflammatory dipeptide carnosine (β-alanyl-L-histidine) in blood plasma, CNDP2 is a cytosolic isoform and is more specific to Cys-Gly dipeptide in animals [ 130 ]. We found that the CNDP2 level was elevated in primary macrophages isolated from xCT-knockout mice [ 131 ]. Although the genetic ablation of CNDP2 does not cause a vast change in Cys-Gly dipeptidase activity, APAP overdose induces renal damage more severely than that encountered in wild-type mice. Therefore, CNDP2 along with GGT may facilitate the recruitment of Cys from extracellular GSH when demand increases during emergencies. The risk of nephropathy increases in cases of type 2 diabetes that carry common variants of CNDP1 and CNDP2 genes [ 132 ]. Tumor suppressor action by CNDP2 has also been observed in hepatocellular carcinoma [ 133 ] and in gastric cancer [ 134 ]. Although the mechanism was not clarified in these studies, the resultant increase in GSH could enhance antioxidant capacity, thereby suppressing nephropathy development and tumorigenic mutagenesis. In contrast, the upregulation of CNDP2 has been reported in several tumors such as hepatocellular carcinoma [ 133 ]. CNDP2 may also stimulate the growth of colon cancer [ 135 ] and ovarian cancer cells [ 136 ] by recruiting Cys for the synthesis of GSH. Because ubenimex, also called bestatin, is an inhibitor of CNDP2, the anti-tumorigenic action of this drug could be partially attributed to the inhibition of CNDP2 [ 137 ]. It is also noteworthy that CNDP2 has a novel function in that it catalyzes N-lactoyl amino acid formation [ 138 ]. Exercise increases N-lactosyl phenylalanine in a CNDP2-dependent fashion, which may help control food intake and regulate systemic energy balance [ 139 ]. Thus, it is possible that elevated N-lactosyl amino acids produced by CNDP2 also have additional functions in tumorigenesis.

While beneficial actions of γ-Glu-Cys have been demonstrated in many studies [ 146 , 147 , 148 ], these are largely attributable to the reactivity of the Cys moiety. Typically, γ-Glu-Cys can also support GPX activity by directly donating electrons [ 149 ]. Cys is released from γ-Glu-Cys, and then, recruited to synthesize glutathione. However, other γ-glutamyl peptides do not have redox activity, so their production may have different underlying benefits. A list of the possible mechanisms follows, which could help elucidate this issue. One hypothetical explanation for producing these γ-glutamyl peptides could be associated with the unique enzymatic properties of γ-GCS that cause a futile cycle in cases of Cys deficiencies. γ-Glutamyl phosphate is the intermediary compound produced from Glu and ATP by the action of γ-GCS; this compound is autocyclized to form 5-OP when there is an insufficient amount of Cys as a substrate, as discussed above in the development of 5-oxoprolinuria. The resultant 5-OP is converted back to Glu by OPLAH, which is a reaction that also requires ATP. Since these collective reactions merely consume ATP without producing substantial products, they are considered to form a futile cycle, which may consequently impair cells and cause APAP hepatotoxicity [ 150 , 151 ]. Consistent with this mechanism, the development of anion gap metabolic acidosis is likely associated with 5-oxoprolinuria, which is a symptom sometimes observed in patients taking APAP [ 152 ]. On the other hand, due to the broad substrate specificity in the γ-GCS reaction, instead of Cys, any amino acid could serve as the substrate for γ-glutamylation with varying efficiency. As a result, the auto-cyclization of γ-glutamyl phosphate is prevented by the presence of such an amino acid, which prevents progression of the futile cycle and consequential metabolic acidosis. This preventive mechanism is particularly effective when amino acids with high affinity to γ-GCS are present, and, hence, 2AB is considered to play such a role. OPT is detectable frequently in non-survivors of APAP-induced liver failure. Mean OPT levels are not associated with survival [ 153 ], however, which suggests that the production of OPT may not sufficiently suppress a futile cycle. To end this debate, careful consideration based on appropriate experiments performed on animal models is required. A benefit has also been proposed in terms of the inhibition of ferroptosis by forming a γ-glutamyl peptide; in other words, the γ-glutamyl peptide-producing reaction is exploited to reduce cellular Glu content, which attenuates the production of ROS so as to suppress the lipid peroxidation that is responsible for cell death [ 154 ]. Cys starvation commonly causes ferroptosis in many types of cells through glutathione deprivation, whereas cells that abundantly produce γ-glutamyl peptides are resistant to nutritional deficiency. It is well established that a Cys deficiency decreases glutathione production, which precludes the GPX4-mediated reductive detoxification of lipid peroxides [ 30 ]. Also, the formation of γ-glutamyl peptides by γ-GCS consumes cellular Glu, which is converted to 2-ketoglutarate and becomes an intermediary compound to the tricarboxylic acid (TCA) cycle. As a result of the formation of γ-glutamyl peptides, the carbohydrate catabolism in the TCA cycle is attenuated. While the production of ATP is decreased by this, and the production of oxygen radicals that cause lipid peroxidation is also simultaneously decreased. Consequently, ferroptosis that is executed by lipid peroxidation is suppressed. This mechanism is supported by observations wherein inhibition of the electron transfer chain suppresses the ferroptosis caused by Cys deprivation [ 155 , 156 ]. A brief explanation is that a decline in the lipid peroxidation reaction due to attenuated Glu catabolism is a likely mechanism for the inhibition of ferroptosis in cells with a high capacity to produce γ-glutamyl peptides. Because each type of cell depends on a different type of metabolism, it is likely that the functions of γ-glutamyl peptides may also differ from cell to cell.

Glutathione is a pivotal molecule that protects cells from the cytotoxic effects of xenobiotic compounds via conjugation reactions and oxidative insults through the reductive detoxification of peroxides by donating electrons to the redox system. The degradation of glutathione by GGT recruits amino acids, notably Cys, to cells, but also produces γ-glutamyl peptides extracellularly under certain circumstances. γ-Glutamyl peptides are also produced intracellularly by γ-GCS-mediated reactions. These reactions appear to proceed under pathological conditions and act as a cellular defense. Although the significance of extracellularly produced γ-glutamyl peptides by GGT have long been obscured, the discovery that γ-glutamyl peptides modulate CaSR function has unveiled their potential roles in extracellular signaling. At present, the signaling role of γ-glutamyl peptides is known to be primarily manifested in the gastrointestinal tract and in the nervous system, but the systemic expression of CaSR may extend this phenomenon to other physiological functions, such as those in the cardiovascular and immune systems.