- Email: [email protected]
Gamma-Glutamyl Transpeptidase-Dependent Lipid Peroxidation in Isolated Hepatocytes and HepG2 Hepatoma Cells
Free Radical Biology & Medicine, Vol. 22, No. 5, pp. 853–860, 1997 Copyright q 1997 Elsevier Science Inc. Printed in the USA. All rights reserved 0891-5849/97 $17.00 / .00
PII S0891-5849(96)00422-4
Original Contribution GAMMA-GLUTAMYL TRANSPEPTIDASE-DEPENDENT LIPID PEROXIDATION IN ISOLATED HEPATOCYTES AND HepG2 HEPATOMA CELLS Aldo Paolicchi,* Roberto Tongiani,* Patrizia Tonarelli,* Mario Comporti,† and Alfonso Pompella† *Dipartimento di Biomedicina Sperimentale, Universita` di Pisa, Italy; and †Istituto di Patologia Generale, Universita` di Siena, Italy (Received 23 May 1996; Revised 18 July 1996; Accepted 9 August 1996)
Abstract—Gamma-glutamyltranspeptidase (GGT), a plasma membrane-bound enzyme, provides the only activity capable to effect the hydrolysis of extracellular glutathione (GSH), thus favoring the cellular utilization of its constituent amino acids. Recent studies have shown however that in the presence of chelated iron prooxidant species can be originated during GGT-mediated metabolism of GSH, and that a process of lipid peroxidation can be started eventually in suitable lipid substrates. The present study was undertaken to verify if a GGT-dependent lipid peroxidation process can be induced in the lipids of biological membranes, including living cells, and if this effect can be sustained by the GGT highly expressed at the surface of HepG2 human hepatoma cells. In rat liver microsomes (chosen as model membrane lipid substrate) exposed to GSH and ADP-chelated iron, the addition of GGT caused a marked stimulation of lipid peroxidation, which was further enhanced by the addition of the GGT co-substrate glycyl-glycine. The same was observed in primary cultures of isolated rat hepatocytes, where the lipid peroxidation process did not induce acute toxic effects. GGT-stimulation of lipid peroxidation was dependent both on the concentration of GSH and of ADP-chelated iron. In GGT-rich HepG2 human hepatoma cells, the exposure to GSH, glycyl-glycine, and ADP-chelated iron resulted in a nontoxic lipid peroxidation process, which could be prevented by means of GGT inhibitors such as acivicin and the serine–boric acid complex. In addition, by co-incubation of HepG2 cells with rat liver microsomes, it was observed that the GGT owned by HepG2 cells can act extracellularly, as a stimulant on the GSH- and iron-dependent lipid peroxidation of microsomes. The data reported indicate that the lipid peroxidation of liver microsomes and of living cells can be stimulated by the GGT-mediated metabolism of GSH. Due to the well established interactions of lipid peroxidation products with cell proliferation, the phenomenon may bear particular significance in the carcinogenic process, where a relationship between the expression of GGT and tumor progression has been envisaged. Copyright q 1997 Elsevier Science Inc. Keywords—Gamma-glutamyl transpeptidase, Glutathione, Iron, Liver microsomes, Hepatocytes, HepG2 hepatoma cells, Lipid peroxidation, Free radicals
synthesis, and the latter in turn depends on an adequate supply of precursor amino acids. In several cell types, this is accomplished in part through gammaglutamyl transpeptidase (GGT) activity. Being a membrane-bound enzyme, with its active site oriented toward the outer surface of the cell, GGT in fact participates in the salvage pathway of extracellular glutathione by catalyzing the first step in its degradation, i.e., the hydrolytic release of the cysteinyl-glycine and the glutamyl moieties; these become available for subsequent transport across the
INTRODUCTION
The maintenance of intracellular levels of reduced glutathione (GSH) is critical for the functioning of important antioxidant defense mechanisms of the cell.1 Because intact GSH is poorly transported across cell membranes, its intracellular levels depend on a balance between its consumption and its de novo Address correspondence to: Alfonso Pompella, Istituto di Patologia Generale dell’Universita` di Siena—Via Aldo Moro, I-53100 Siena, Italy. 853
/ 2b25 2380 Mp 853 Wednesday Dec 18 01:17 AM EL–FRB 2380
854
A. PAOLICCHI et al.
membrane to the cytoplasm, where they enter the GSH-synthetic pathway.2 GGT is expressed constitutively in several organs and tissues.3 In the liver, its expression is detectable in biliary epithelium and in occasional periportal hepatocytes.4 GGT is also reexpressed in high levels in putative preneoplastic nodules induced by chemical carcinogens,5 as well as in several hepatoma cell lines6 and in human spontaneous neoplasia and metastasis, such as in colon,7 and ovary.8 Because it favors the reconstitution of intracellular GSH, GGT can be regarded as a member of the cellular antioxidant systems; indeed, the enrichment of cells in GGT has been shown to result in an increased resistance against toxic and prooxidant injury.9 – 11 On this basis, the reexpression of high GGT levels in preneoplastic lesions and in frank neoplasia has been interpreted as one of the factors involved in the resistance of transformed cells against chemical and radiant damage, which is often encountered in tumor tissue and can represent a major hindrance to efficient therapy.8,12,13 Recent studies have shown, however, that GGT is itself able to play a potentially prooxidant role in selected conditions. In fact, cysteinyl-glycine resulting from the GGT-mediated cleavage of GSH can reduce ferric iron, if present, and the redox-cycling of iron thus started has been shown to result in the initiation of lipid peroxidation.14 The potential role of the lipid peroxidation process in the initiation and promotion of carcinogenesis has been long debated (reviewed in 15); it is therefore noteworthy that a GGT-stimulation of lipid peroxidation has been indeed documented histochemically in fresh rat liver sections, in correspondence of GGT-positive hepatocytes of chemically induced putative preneoplastic lesions;16,17 in this experimental model, evidence was provided suggesting that the phenomenon can be operative in vivo, and its potential role in the modulation of the carcinogenic process therefore has been discussed.16 In order to expand the observations obtained histochemically in tissue sections, the present study was designed to ascertain (1) whether a GGT-dependent lipid peroxidation process can be induced in biological membranes of living cells, i.e., cells retaining their antioxidant defense systems; and (2) whether this effect can be sustained by the endogenous GGT, highly reexpressed at the surface of living HepG2 human hepatoma cells. The results indicate that in the presence of GGT substrates and chelated iron a GGTdependent, nonlethal lipid peroxidation process can easily be ignited, involving the membranes of both normal and neoplastic liver cells, and liable to be propagated to lipid substrates in the cellular outer environment.
EXPERIMENTAL PROCEDURES
Chemicals All reagents, including cell culture media, were from Sigma (St. Louis, MO). Isolation of rat liver microsomes Adult Sprague Dawley rats (200–250 g; Charles River, Italy) were sacrificed by exsanguination under diethylether anaesthesia; livers were quickly excised and homogenized (33%) in 5 mM EDTA, 100 mM Tris-HCl buffer (pH 7.4). Liver homogenates were centrifuged at 9,000 1 g for 20 min. The supernatants were removed and centrifuged at 100,000 1 g for 45 min. Pellets (Åmicrosomes) were resuspended in the same buffer and stored at 0807C until use. Hepatocyte and HepG2 hepatoma cell cultures Isolated rat hepatocytes were obtained by collagenase perfusion from the same animals as above. Isolated cells were resuspended (1 1 106 cells/ml) in Waimouth MB 752/1 medium containing 10 mg/ml insulin, 10% fetal calf serum, 100 U/ml penicillin, and 100 mg/ml streptomycin (Sigma), plated in 5-cm collagencoated Falcon petri dishes (2.5 ml/dish) and allowed to adhere for 4 h at 377C; medium (including nonadherent, dead cells) was then changed once with the same medium (without serum), and cells were used for the experiments. Human HepG2 hepatoma cells, maintained in DMEM/F12 medium containing penicillin and streptomicin (see above) plus 10% fetal calf serum, were plated in 25-cm2 Falcon culture flasks, and used before reaching confluence. Prooxidant incubations and lipid peroxidation All incubations were performed in Hank’s buffer, pH 7.2. Reagents were added with suitable aliquots of concentrated stock solutions, in order to obtain the final concentrations indicated. Rat liver microsomes were incubated at the concentration of 1 mg protein/ml. Hepatocytes or HepG2 hepatoma cells were exposed to fresh medium, in which the reagents had been dissolved at the indicated concentrations. At the indicated times, lipid peroxidation was determined as the content of malonaldehyde (MDA), or, for selected experiments, as thiobarbituric acid reacting substances (TBARS). For the determination of MDA, medium was discarded and cells were yielded in 5% TCA; after centrifugation, TCA extracts were then reacted with 1 vol 0.67% thiobarbituric acid
/ 2b25 2380 Mp 854 Wednesday Dec 18 01:17 AM EL–FRB 2380
Gamma-glutamyl transpeptidase and lipid peroxidation
Fig. 1. GGT-stimulation of iron-dependent lipid peroxidation (production of malonaldehyde, MDA) in rat liver microsomes. Incubations were carried out at 377C in Hank’s medium, and had the following composition: (h) 2 mM GSH, 1.5–0.15 mM ADP-Fe3/, 200 mU/ml GGT; (l) as above, plus 20 mM glycyl-glycine; (s) 2 mM GSH, 1.5–0.15 mM ADP-Fe3/, 20 mM glycyl-glycine (no added GGT); (m) as for l, plus 200 mM BHT. One typical experiment of five is shown.
(TBA) at 1007C for 10 min.18 For the determination of TBARS, 2 vol TBA reagent were added to each culture flask, and the mixture was reacted at 957C for 15 min, essentially as described.19
855
as expected.22 However, the addition of GGT to the incubation mixture greatly stimulated the process, and even higher lipid peroxidation levels were attained when the incubation included glycyl-glycine, acting as g-glutamyl acceptor and therefore as a stimulant for GGT activity.20 The addition of the lipid-soluble, chainbreaking antioxidant BHT resulted in the suppression of GGT-stimulated lipid peroxidation. The intensity of the stimulation showed a linear dose-dependence on the GSH concentration, as reported in Fig. 2. The possibility was then verified that the described GGT-stimulated lipid peroxidation might involve the membranes of intact, living cells (Fig. 3). In 24 h-cultured rat hepatocytes, a lipid peroxidation developed in the presence of GSH, glycyl-glycine, and ADP-iron; again, as in the case of liver microsomes, lipid peroxidation was markedly stimulated by the addition of GGT. Up to the latest time of incubation studied (90 min), hepatocyte viability as assessed by LDH leakage was not affected by lipid peroxidation (not shown). As shown in Table 1, the stimulation of lipid peroxidation was dependent on the addition of glycyl-glycine, which again points to a crucial role played by GGT activity in this effect. The extent of GGT-stimulated, GSH-dependent lipid peroxidation in cultured hepatocytes was proportionally decreased by decreasing the ADP-iron concentration; in all cases, however, ADP-iron plus GGT caused higher levels of lipid peroxidation than ADPiron alone (Fig. 4).
Other determinations GGT activity was determined according to Huseby and Stro¨mme20 as modified by Edwards.21 Protein content was determined by the method of Lowry. Statistical significance of data was assessed by the Student’s t test. RESULTS
Rat liver microsomes were chosen as a membrane lipid substrate for the envisaged prooxidant effects of GGT activity. For these and most subsequent experiments, the concentrations of GSH and glycyl-glycine (2 and 20 mM, respectively) were chosen, taking into account the corresponding Km values of GGT for both, in order to obtain the vmax of GGT activity.20 The stimulation by GGT on the iron-dependent lipid peroxidation of rat liver microsomes is shown in Fig. 1; microsomes exposed only to GSH and ADP-chelated ferric iron exhibited increasing levels of lipid peroxidation,
Fig. 2. Glutathione-dependence of the GGT-mediated lipid peroxidation (production of malonaldehyde, MDA) in rat liver microsomes. Incubations were carried out (60 min, 377C) in Hank’s medium containing 2 mM GSH, 20 mM glycyl-glycine, 200 mU/ml GGT, 1.5–0.15 mM ADP-Fe3/. Data shown are means { S.E.M. of three experiments.
/ 2b25 2380 Mp 855 Wednesday Dec 18 01:17 AM EL–FRB 2380
856
A. PAOLICCHI et al.
Fig. 3. Prooxidant effect of GGT on cultured rat hepatocytes. Incubations were carried out in Hank’s medium containing 2 mM GSH, 1.5–0.15 mM ADP-Fe3/ and 20 mM glycyl-glycine, in the presence (l) or absence (s) of GGT (200 mU/ml). Data are means { S.E.M. of triplicate samples; one typical experiment of three is shown. Repeated experiments gave fully comparable results. * Significantly different from the 0-time value, p õ 0.01. ** Significantly different from the value obtained in the absence of GGT, p õ 0.01.
Further experiments were performed with human hepatoma HepG2 cells, in which a high intrinsic GGT activity (67 mU/mg protein) was detected. Because the levels of lipid peroxidation inducible in hepatoma cells (as in the case of other tumor cells) are generally much lower than in normal hepatocytes,23 for the determination of lipid peroxidation in HepG2 cells, the less specific but more sensitive determination of TBARs was employed.19 As reported in Table 2, the exposure of HepG2 cells to GSH, in the presence of ADP-Fe3/, led to the development of lipid peroxidation; the latter
Fig. 4. Influence of different concentrations of ADP-chelated iron on GGT-dependent lipid peroxidation in cultured rat hepatocytes. Cells were incubated for 60 min at 377C in Hank’s buffer containing 2 mM GSH, 20 mM glycyl-glycine, and the indicated concentrations of ADP-Fe3/, in the presence (l) or absence (s) of GGT (200 mU/ ml). Data are means { S.E.M. of triplicate samples; one typical experiment of three is shown. * Significantly different from the respective values obtained in the absence of GGT, p õ 0.01. (ns) Not significantly different from the 15 mM Fe3/ value.
was further stimulated by the addition of the GGT-cosubstrate glycyl-glycine, and was inhibited by the addition of the GGT inhibitor acivicin, indicating the dependence of the process on the cell-bound GGT activity. The extent of GGT-dependent lipid peroxidation was compared with that obtainable with other well-established prooxidant mixtures (Table 3), and appeared somewhat higher. No alteration of cell viability as assessed by LDH leakage was observed during GGT-me-
Table 1. GGT-Mediated Lipid Peroxidation (Production of Malonaldehyde, MDA) in Primary Hepatocyte Culturesa Incubation System Hepatocytes (control) Hepatocytes / GSH / GGT / ADP-Fe / ADP-Fe / gly2 / ADP-Fe / gly2 / BHT Hepatocytes / GSH / gly2 / ADP-Fe (no GGT)
MDA Released, nmol/mg proteinb 0.23 0.33 3.56 6.63 0.31 2.39
{ { { { { {
0.04 0.09 (ns) 0.04** 1.23 0.12 (ns) 0.48*
a Incubations were carried out at 377C for 60 min. Concentration of reagents was as follows: GSH, 2 mM; glycyl-glycine, 20 mM; ADP-Fe, 1.5 mM–150 mM; GGT, 200 mU/ml; BHT, 200 mM. b Data are means { S.E.M. of three experiments. * Significantly different from control, p õ 0.02. ** Significantly different from the value obtained without ADP-Fe, p õ 0.001. (ns) Not significantly different from control.
/ 2b25 2380 Mp 856 Wednesday Dec 18 01:17 AM EL–FRB 2380
Gamma-glutamyl transpeptidase and lipid peroxidation Table 2. GSH-Dependent Lipid Peroxidation (Production of TBA-Reactive Substances, TBARS) in GGT-Positive Hep-G2 Hepatoma Cellsa Incubation Systemb HepG2 HepG2 HepG2 HepG2 HepG2
(control) / ADP-Fe / ADP-Fe / GSH / ADP-Fe / GSH / gly2 / ADP-Fe / GSH / gly2 / acivicin
857
Table 4. Accumulation in the Incubation Medium of the Products of GSH-Dependent Lipid Peroxidation in GGT-Positive Hep-G2 Hepatoma Cellsa
TBARS, nmol/mg proteinc 0.71 3.30 4.29 5.11 2.54
{ { { { {
0.05 0.02 0.05** 0.18* 0.11***
Incubations were carried out at 377C for 60 min. Concentration of reagents was as follows: GSH, 2 mM; glycylglycine, 20 mM; ADP-Fe, 1.5 mM–150 mM; acivicin, 1 mM. c Data are means { S.E.M. of triplicate samples; one typical experiment of four is shown. Repeated experiments gave fully comparable results. * Significantly different from the value obtained without gly2, p õ 0.02. ** Significantly different from the value obtained without GSH, p õ 0.001. *** Significantly different from the value obtained without acivicin, p õ 0.001; not significantly different from ‘‘HepG2 / ADPFe’’. a
b
TBARS Incubation Time (min)
nmol/mg cell protein
nmol/ml supernatant
0 30 60
0.40 { 0.03 1.27 { 0.09 0.89 { 0.02*
0 1.25 { 0.72 3.38 { 0.12*
At the indicated times, culture supernatants were separated from cells and centrifuged (800 g, 10 min); cells were washed once with fresh medium; TBARS was then determined separately in cells and supernatants. a Incubations were carried out at 377C. Concentration of reagents was as follows: cells, 1 mg protein/ml; GSH, 2 mM; glycyl-glycine, 20 mM; ADP-Fe, 1.5 mM–150 mM. b Data are means { S.E.M. of triplicate samples; one typical experiment of three is shown. * Significantly different from the respective 30 min value, p õ 0.05.
DISCUSSION
diated lipid peroxidation, up to the latest time of incubation studied (90 min) (not shown). When lipid peroxidation was determined separately, in cell monolayers as compared to cell supernatants, the extracellular, but not intracellular, accumulation of the produced TBARS was observed (Table 4). It was then investigated whether the lipid peroxidation process stimulated by the GGT activity present at the cell surface might be propagated to lipid substrates present in the extracellular environment. As shown in Table 5, the addition of GGT-rich HepG2 cells to a suspension of rat liver microsomes, in the presence of chelated iron and GGT substrates, greatly stimulated microsomal lipid peroxidation; the effect was GGT-dependent, as indicated again by the fact that the addition of the GGT inhibitor serine–boric acid complex caused its suppression. Table 3. GSH-Dependent Lipid Peroxidation in GGT-Positive HepG2 Hepatoma Cells, as Compared to Lipid Peroxidation Inducible With Other Prooxidant Mixturesa Incubation System HepG2 HepG2 HepG2 HepG2 HepG2
/ / / / /
GSH / gly2 / ADP-Feb Fe-NTA 100 mM CuOOH 500 mM ADP-Fe 1.5 mM-150 mM buffer (control)
TBARS, nmol/mg proteinc 5.70 2.71 1.88 3.16 0.77
{ { { { {
0.21 0.28 0.15 0.80 0.06
Incubations were carried out at 377C for 60 min. Concentration of reagents was: GSH, 2 mM; glycyl-glycine, 20 mM; ADP-Fe, 1.5 mM–150 mM. c Data are means { S.E.M. of triplicate samples; one typical experiment of three is shown. a
b
The data reported in the present study indicate that the GGT-mediated metabolism of GSH is liable to expose cellular membranes to a prooxidant insult, leading to the onset of lipid peroxidation. As previously suggested,14 the reason for this effect is allegedly due to the fact that GSH metabolites prior to their absorption by the cell can participate in metal-catalyzed redox reactions, leading to the production of radical oxygen species. In particular, the thiol group of the dipeptide cysteinyl-glycine, resulting from the GGT-mediated cleavage of GSH, being more dissociable at neutral pH than the thiol of GSH itself,24 is more prone to interact with transition metal ions; in the case of ferric iron, cysteinyl-glycine can accomplish its reduction to ferrous ions, whereas the oxidation of the thiol can lead Table 5. Lipid Peroxidation in Rat Liver Microsomes Exposed to GGT-Rich HepG2 Hepatoma Cells, in the Presence of GGT Substratesa Incubation System
TBARS, nmol/mg proteinb
Microsomes / GSH / gly2 / ADP-Fe / HepG2 cells / HepG2 cells / serine borate HepG2 / GSH / gly2 / ADP-Fe
14.8 26.0 12.6 2.9
{ { { {
1.38 2.24* 1.59 (ns) 0.24
a Incubations were carried out at 377C for 60 min. Concentration of reagents was as follows: GSH, 2 mM; glycyl-glycine, 20 mM; ADP-Fe, 1.5 mM–150 mM; microsomes, 1 mg protein/ml; serine 10 mM-boric acid 20 mM. b Data are means { S.E.M. of triplicate samples; one typical experiment of three is shown. * Significantly different from the value obtained without HepG2 cells, p õ 0.02. (ns) Not significantly different from the value obtained without HepG2 cells.
/ 2b25 2380 Mp 857 Wednesday Dec 18 01:17 AM EL–FRB 2380
858
A. PAOLICCHI et al.
to the formation of a cysteinyl-glycine thiyl radical.14 As in other model systems of thiol-driven iron-dependent lipid peroxidation,22 further interactions of reduced iron with dioxygen can then cause the appearance of species capable to initiate lipid peroxidation in membrane lipids.25 The ability of GGT to stimulate lipid peroxidation recently has been reported biochemically, in systems including the purified enzyme and a soluble lipid substrate.14 Also the development of GGT-dependent lipid peroxidation has been demonstrated by histochemical means in chemically induced GGT-positive preneoplastic lesions of rat livers, following the exposure in vitro of tissue cryostat sections to GSH, glycyl-glycine, and chelated iron.16,17 The data reported in the present study add to the biological significance of the phenomenon, in that they indicate that the prooxidant effects of GGT activity can involve as well the lipids of cellular membranes, including those of intact living cells. The addition of GGT plus its substrates GSH and glycyl-glycine, in the presence of ADP-chelated ferric iron, results, in fact, in a marked stimulation of lipid peroxidation both in rat liver microsomes and in isolated hepatocytes. Moreover, the plasma membrane-bound GGT activity intrinsically present at the outer surface of HepG2 human hepatoma cells is equally able to stimulate a lipid peroxidation process, which involves the membranes of the cells themselves, and also can be propagated to membranes present in the surrounding environment. Interestingly, both in the case of isolated rat hepatocytes and of HepG2 hepatoma cells, GGT-stimulated lipid peroxidation was not accompanied by acute alterations of cell viability. In this respect, it is tempting to speculate that this also could be a result of GGT activity, whose role in potentiating cellular resistance to oxidative stress has been documented.9 – 11 Indeed, data shown in Table 4 suggest that the products of the lipid peroxidation process do not accumulate within cells, where they allegedly undergo efficient metabolism. The experimental conditions employed in the present study included relatively high concentrations of reagents (GSH, glycyl-glycine, ADP-iron). These were chosen based on the described Km values of GGT for its substrates, in order to optimize the conditions for GGT activity and thus facilitate the detection of its prooxidant effects. However, GGT-stimulated lipid peroxidation exhibited a linear dependence on GSH concentration (Fig. 2), down to GSH concentrations (50–10 mM) in the range of those detectable in rat or human plasma;26 on the other hand, in the liver, where hepatocytes are net exporters of GSH, GSH can reach remarkably higher local concentrations. It seems, therefore, likely that although at levels allegedly lower that
those observed in the present study the described GGTstimulation of lipid peroxidation can take place in in vivo conditions as well. With respect to the requirement of the reaction for chelated iron, the linear dose-dependence shown in Figure 4 again suggests that GGT stimulation of lipid peroxidation is still possible down to very low concentrations of iron. On the other hand, the participation of iron in vivo to such reactions would require its prior release from its storage protein, transferrin, in the first place. In this respect, a number of studies have shown, however, that the release of iron from transferrin is facilitated in several conditions, such as mild acidic pH, exposure to reducing agents, and binding to cellular receptors.27,28 Previous biochemical and histochemical studies14,16 have indeed shown that GGT-stimulation of lipid peroxidation can be sustained by transferrin iron alone. The biological significance of GGT-dependent lipid peroxidation in vivo might be multifold. Varying levels of GGT activity can be detected in most cell types, ranging from very low, as in hepatocytes, to rather high, as in tubular epithelium. It is conceivable that the prooxidant effects of GGT activity are normally balanced by its established role in favoring the cellular uptake of precursors for GSH resynthesis, thus allowing the reconstitution of cellular antioxidant defenses. On the other hand, in cells overexpressing the enzyme, conditions could be in favor of a higher extracellular production of cysteinyl-glycine, liable to ignite lipid peroxidation through interactions with iron. One major aspect involved in the described phenomena can be tumor progression. Increasing evidence suggests, in fact, that the expression of GGT may play a role in tumor progression: transfection with GGT resulted in a growth advantage for mouse hepatoma cells;29 GGT is an established marker of progression in mouse skin carcinogenesis;29 in metastasis from human colon carcinomas, GGT is higher than in primitive tumors;7 transfection of epithelial cells with the oncogene ras while resulting in the appearance of a metastatic behavior is also accompanied by GGT expression.30 In addition, a recent study has shown in monoblastic leukemia cells that the inhibition of GGT can inhibit cell proliferation and favor differentiation.31 The enzyme catalytic site of cell-bound GGT is located at the cell surface, and its substrates and the potential targets of the prooxidant reactions can be found in the outer cell environment, i.e., out of reach and control for most of the cellular antioxidant systems. In this way, despite the fact that in tumor cells antioxidants may often be increased,32 cells reexpressing GGT activity at their surface might anyway promote prooxidant conditions in their vicinity, exposing themselves and the surround-
/ 2b25 2380 Mp 858 Wednesday Dec 18 01:17 AM EL–FRB 2380
Gamma-glutamyl transpeptidase and lipid peroxidation
ing cells to the effects of prooxidant species and lipid peroxidation products. The latter have been shown indeed to possess a wide range of antiproliferative,33,34 as well as mutagenic,35 – 37 tumor-initiating38 and tumorpromoting effects;39 – 41 the possibility therefore exists that some of these effects are involved in the described association of GGT expression with tumor progression. In conclusion, the data reported in the present study show that the metabolism of GSH mediated by (soluble and cell-bound) GGT activity can result in conditions favoring the ignition of lipid peroxidation in cell membranes. Basing on a number of experimental observations, GGT activity is believed to play a role in tumor progression. Future studies in our laboratory will be dedicated to investigating the possibility that aspects of the envisaged role of GGT in tumor progression may be mediated at least partly by the phenomenon of GGTdependent lipid peroxidation.
12. 13. 14.
15. 16.
17.
18.
Acknowledgements — The support of Consiglio Nazionale delle Ricerche (Italy), Progetto A.C.R.O., is gratefully acknowledged. Additional financial support was derived from the Association for International Cancer Research (U.K.).
19.
REFERENCES
21.
1. Sies, H. Strategies of antioxidant defense. Eur. J. Biochem. 215:213–219; 1993. 2. Tate, S. S.; Meister, A. g-Glutamyl transpeptidase from kidney. Methods Enzymol 113:400–419; 1985. 3. Rutenburg, A. M.; Kim, H.; Fischbein, J. W.; Hanker, J. S.; Wasserkrug, H. L.; Seligman, A. M. Histochemical and ultrastructural demonstration of g-glutamyl transpeptidase activity. J. Histochem. Cytochem. 17:517–526; 1969. 4. Lindros, K. O.; Penttila, K. E.; Janzen, J.-W. G.; Moorman, A. F. M.; Spelsky, H.; Israel, Y. The gamma-glutamyltransferase/glutamine synthetase activity ratio: A powerful marker of the acinar origin of hepatocytes. J. Hepatol. 8:338–343; 1989. 5. Cameron, R.; Kellen, J.; Kolin, A.; Malkin, A.; Farber, E. gGlutamyltransferase in putative premalignant liver cell populations during hepatocarcinogenesis. Cancer Res. 38:823–829; 1978. 6. Cheng, S.; Nassar, K.; Levy, D. g-Glutamyl transpeptidase activity in normal, regenerating and malignant hepatocytes. FEBS Lett. 85:310–312; 1978. 7. Munjal, D. D. Concurrent measurements of carcinoembryonic antigen, glucose phosphate isomerase, g-glutamyltransferase, and lactate dehydrogenase in malignant, normal adult, and fetal colon tissues. Clin. Chem. 26:1809–1812; 1980. 8. Hanigan, M. H.; Frierson, H. F., Jr.; Brown, J. E.; Lovell, M. A.; Taylor, P. T. Human ovarian tumors express g-glutamyl transpeptidase. Cancer Res. 54:286–290; 1994. 9. Rajpert-De Meyts, E.; Shi, M.; Chang, M.; Robison, T. W.; Groffen, J.; Heisterkamp, N.; Forman, H. J. Transfection with g-glutamyl transpeptidase enhances recovery from glutathione depletion using extracellular glutathione. Toxicol. Appl. Pharmacol. 114:56–62; 1992. 10. Shi, M.; Gozal, E.; Choy, H. A.; Forman, H. J. Extracellular glutathione and g-glutamyl transpeptidase prevent H2O2-induced injury by 2,3-dimethoxy-1,4-naphthoquinone. Free Rad. Biol. Med. 15:57–67; 1993. 11. Chang, M.; Shi, M.; Forman, H. J. Exogenous glutathione pro-
20.
22. 23.
24. 25. 26.
27. 28. 29.
30.
31.
859
tects endothelial cells from menadione toxicity. Am. J. Physiol. 6:L634–L637; 1992. Hanigan, M. H.; Pitot, H. C. Gamma-glutamyl transpeptidase— Its role in carcinogenesis. Carcinogenesis 6:165–172; 1985. Farber, E. Clonal adaptation during carcinogenesis. Biochem. Pharmacol. 39:1837–1846; 1990. Stark, A.-A.; Zeiger, E.; Pagano, D. A. Glutathione metabolism by g-glutamyltranspeptidase leads to lipid peroxidation: Characterization of the system and relevance to hepatocarcinogenesis. Carcinogenesis 14:183–189; 1993. Dianzani, M. U. Lipid peroxidation and cancer. Crit. Rev. Oncol. Hematol. 15:125–147; 1993. Pompella, A.; Paolicchi, A.; Dominici, S.; Comporti, M.; Tongiani, R. Selective colocalization of lipid peroxidation and protein thiol loss in chemically induced hepatic preneoplastic lesions: The role of g-glutamyl transpeptidase activity. Histochem Cell Biol. 106:275–282; 1996. Stark, A.-A.; Russell, J. J.; Langenbach, R.; Pagano, D. A.; Zeiger, E.; Huberman, E. Localization of oxidative damage by a glutathione-g-glutamyl transpeptidase system in preneoplastic lesions in sections of livers from carcinogen-treated rats. Carcinogenesis 15:343–348; 1994. Pompella, A.; Maellaro, E.; Casini, A. F.; Ferrali, M.; Ciccoli, L.; Comporti, M. Measurement of lipid peroxidation in vivo: A comparison of different procedures. Lipids 22:206–211; 1987. Buege, J.; Aust, S. D. Microsomal lipid peroxidation. Methods Enzymol. 52:302–310; 1978. Huseby, N. E.; Stro¨mme, J. H. Practical points regarding routine determination of g-glutamyl transferase (g -GT) in serum with a kinetic method at 377C. Scand. J. Clin. Lab. Invest. 34:357–361; 1974. Edwards, A. M. Regulation of g-glutamyltranspeptidase in rat hepatocyte monolayer cultures. Cancer Res. 42:1107–1115; 1982. Tien, M.; Bucher, J. R.; Aust, S. D. Thiol-dependent lipid peroxidation. Biochem. Biophys. Res. Commun. 107:279–285; 1982. Canuto, R. A.; Ferro, M.; Muzio, G.; Bassi, A. M.; Leonarduzzi, G.; Maggiora, M.; Adamo, D.; Poli, G.; Lindahl, R. Role of aldehyde metabolizing enzymes in mediating effects of aldehyde products of lipid peroxidation in liver cells. Carcinogenesis 15:1359–1364; 1994. Stark, A.-A.; Arad, A.; Siskindovich, S.; Pagano, D. A.; Zeiger, E. Effect of pH on mutagenesis by thiols in Salmonella typhimurium TA102. Mutat. Res. 224:89–94; 1989. Minotti, G.; Aust, S. D. An investigation into the mechanisms of citrate-Fe2/-dependent lipid peroxidation. Free Rad. Biol. Med. 3:379–387; 1987. Michelet, F.; Gueguen, R.; Leroy, P.; Wellman, M.; Nicolas, A.; Siest, G. Blood and plasma glutathione measured in healthy subjects by HPLC: Relation to sex, aging, biological variables and life habits. Clin. Chem. 41:1509–1517; 1995. Brieland, J. K.; Fantone, J. C. Ferrous iron release from transferrin by human neutrophil-derived superoxide anion: Effect of pH and iron saturation. Arch. Bioch. Biophys. 248:78–83; 1991. Sipe, D. M.; Murphy, R. F. Binding to cellular receptors results in increased iron release from transferrin at mildly acidic pH. J. Biol. Chem. 266:8002–8007; 1991. Warren, B. S.; Naylor, M. F.; Winberg, L. D.; Yoshimi, N.; Volpe, J. P. G.; Gimenez-Conti, I.; Slaga, T. J. Induction and inhibition of tumor progression. Proc. Soc. Exp. Biol. Med. 202:9–15; 1993. Braun, L.; Goyette, M.; Yaswen, P.; Thompson, N. L.; Fausto, N. Growth in culture and tumorigenicity after transfection with the ras oncogene of liver epithelial cells from carcinogen-treated rats. Cancer Res. 47:4116–4124; 1987. Bauvois, B.; Laouar, A.; Rouillard, D.; Wietzerbin, J. Inhibition of g-glutamyl transpeptidase activity at the surface of human myeloid cells is correlated with macrophage maturation and transforming growth factor b production. Cell Growth Differ. 6:1163–1170; 1995.
/ 2b25 2380 Mp 859 Wednesday Dec 18 01:17 AM EL–FRB 2380
860
A. PAOLICCHI et al.
32. Sun, Y. Free radicals, antioxidant enzymes and carcinogenesis. Free Rad. Biol. Med. 8:583–599; 1990. 33. Esterbauer, H.; Schaur, R. J.; Zollner, H. Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Rad. Biol. Med. 11:81–128; 1991. 34. Gonzalez, M. J.; Schemmel, R. A.; Gray, J. I.; Dugan, L., Jr.; Sheffield, L. G.; Welsch, C. W. Effect of dietary fat on growth of MCF-7 and MDA-MB231 human breast carcinomas in athymic nude mice: Relationship between carcinoma growth and lipid peroxidation product levels. Carcinogenesis 12:1231– 1235; 1991. 35. Esterbauer, H.; Eckl, P.; Ortner, A. Possible mutagens derived from lipids and lipid precursors. Mutat. Res. 238:223–233; 1990. 36. Canonero, R.; Martelli, A.; Marinari, U. M.; Brambilla, G. Mutation induction in Chinese hamster lung V79 cells by five alk2-enals produced by lipid peroxidation. Mutat. Res. 244:153– 156; 1990. 37. Zhang, J.-R.; Sevanian, A. The genotoxic effects of arachidonic acid in V79 cells are mediated by peroxidation products. Toxicol. Appl. Pharmacol. 121:193–202; 1993. 38. Chung, F.-L.; Chen, H.-J. C.; Guttenplan, J. B.; Nishikawa, A.; Hard, G. C. 2,3-Epoxy-4-hydroxynonanal as a potential tumorinitiating agent of lipid peroxidation. Carcinogenesis 14:2073– 2077; 1993. 39. Bartoli, G. M.; Bartoli, S.; Galeotti, T.; Bertoli, E. Superoxide
dismutase content and microsomal lipid composition of tumours with different growth rates. Biochim. Biophys. Acta 620:205– 211; 1980. 40. O’Brian, C. A.; Ward, N. E.; Weinstein, I. B.; Bull, A. W.; Marnett, L. J. Activation of rat brain protein kinase C by lipid oxidation products. Biochem. Biophys. Res. Commun. 155:1374– 1380; 1988. 41. Greenley, T. L.; Davies, M. J. Direct detection of radical generation in rat liver nuclei on treatment with tumour-promoting hydroperoxides and related compounds. Biochim. Biophys. Acta 1226:56–64; 1994.
ABBREVIATIONS
BHT—butylated hydroxytoluene CuOOH—cumene hydroperoxide Fe-NTA—ferric nitrilo-triacetate GGT—gamma-glutamyl transpeptidase gly2—glycyl-glycine GSH—reduced glutathione MDA—malonaldehyde TBARS—thiobarbituric acid-reacting substances
/ 2b25 2380 Mp 860 Wednesday Dec 18 01:17 AM EL–FRB 2380
PII S0891-5849(96)00422-4
Original Contribution GAMMA-GLUTAMYL TRANSPEPTIDASE-DEPENDENT LIPID PEROXIDATION IN ISOLATED HEPATOCYTES AND HepG2 HEPATOMA CELLS Aldo Paolicchi,* Roberto Tongiani,* Patrizia Tonarelli,* Mario Comporti,† and Alfonso Pompella† *Dipartimento di Biomedicina Sperimentale, Universita` di Pisa, Italy; and †Istituto di Patologia Generale, Universita` di Siena, Italy (Received 23 May 1996; Revised 18 July 1996; Accepted 9 August 1996)
Abstract—Gamma-glutamyltranspeptidase (GGT), a plasma membrane-bound enzyme, provides the only activity capable to effect the hydrolysis of extracellular glutathione (GSH), thus favoring the cellular utilization of its constituent amino acids. Recent studies have shown however that in the presence of chelated iron prooxidant species can be originated during GGT-mediated metabolism of GSH, and that a process of lipid peroxidation can be started eventually in suitable lipid substrates. The present study was undertaken to verify if a GGT-dependent lipid peroxidation process can be induced in the lipids of biological membranes, including living cells, and if this effect can be sustained by the GGT highly expressed at the surface of HepG2 human hepatoma cells. In rat liver microsomes (chosen as model membrane lipid substrate) exposed to GSH and ADP-chelated iron, the addition of GGT caused a marked stimulation of lipid peroxidation, which was further enhanced by the addition of the GGT co-substrate glycyl-glycine. The same was observed in primary cultures of isolated rat hepatocytes, where the lipid peroxidation process did not induce acute toxic effects. GGT-stimulation of lipid peroxidation was dependent both on the concentration of GSH and of ADP-chelated iron. In GGT-rich HepG2 human hepatoma cells, the exposure to GSH, glycyl-glycine, and ADP-chelated iron resulted in a nontoxic lipid peroxidation process, which could be prevented by means of GGT inhibitors such as acivicin and the serine–boric acid complex. In addition, by co-incubation of HepG2 cells with rat liver microsomes, it was observed that the GGT owned by HepG2 cells can act extracellularly, as a stimulant on the GSH- and iron-dependent lipid peroxidation of microsomes. The data reported indicate that the lipid peroxidation of liver microsomes and of living cells can be stimulated by the GGT-mediated metabolism of GSH. Due to the well established interactions of lipid peroxidation products with cell proliferation, the phenomenon may bear particular significance in the carcinogenic process, where a relationship between the expression of GGT and tumor progression has been envisaged. Copyright q 1997 Elsevier Science Inc. Keywords—Gamma-glutamyl transpeptidase, Glutathione, Iron, Liver microsomes, Hepatocytes, HepG2 hepatoma cells, Lipid peroxidation, Free radicals
synthesis, and the latter in turn depends on an adequate supply of precursor amino acids. In several cell types, this is accomplished in part through gammaglutamyl transpeptidase (GGT) activity. Being a membrane-bound enzyme, with its active site oriented toward the outer surface of the cell, GGT in fact participates in the salvage pathway of extracellular glutathione by catalyzing the first step in its degradation, i.e., the hydrolytic release of the cysteinyl-glycine and the glutamyl moieties; these become available for subsequent transport across the
INTRODUCTION
The maintenance of intracellular levels of reduced glutathione (GSH) is critical for the functioning of important antioxidant defense mechanisms of the cell.1 Because intact GSH is poorly transported across cell membranes, its intracellular levels depend on a balance between its consumption and its de novo Address correspondence to: Alfonso Pompella, Istituto di Patologia Generale dell’Universita` di Siena—Via Aldo Moro, I-53100 Siena, Italy. 853
/ 2b25 2380 Mp 853 Wednesday Dec 18 01:17 AM EL–FRB 2380
854
A. PAOLICCHI et al.
membrane to the cytoplasm, where they enter the GSH-synthetic pathway.2 GGT is expressed constitutively in several organs and tissues.3 In the liver, its expression is detectable in biliary epithelium and in occasional periportal hepatocytes.4 GGT is also reexpressed in high levels in putative preneoplastic nodules induced by chemical carcinogens,5 as well as in several hepatoma cell lines6 and in human spontaneous neoplasia and metastasis, such as in colon,7 and ovary.8 Because it favors the reconstitution of intracellular GSH, GGT can be regarded as a member of the cellular antioxidant systems; indeed, the enrichment of cells in GGT has been shown to result in an increased resistance against toxic and prooxidant injury.9 – 11 On this basis, the reexpression of high GGT levels in preneoplastic lesions and in frank neoplasia has been interpreted as one of the factors involved in the resistance of transformed cells against chemical and radiant damage, which is often encountered in tumor tissue and can represent a major hindrance to efficient therapy.8,12,13 Recent studies have shown, however, that GGT is itself able to play a potentially prooxidant role in selected conditions. In fact, cysteinyl-glycine resulting from the GGT-mediated cleavage of GSH can reduce ferric iron, if present, and the redox-cycling of iron thus started has been shown to result in the initiation of lipid peroxidation.14 The potential role of the lipid peroxidation process in the initiation and promotion of carcinogenesis has been long debated (reviewed in 15); it is therefore noteworthy that a GGT-stimulation of lipid peroxidation has been indeed documented histochemically in fresh rat liver sections, in correspondence of GGT-positive hepatocytes of chemically induced putative preneoplastic lesions;16,17 in this experimental model, evidence was provided suggesting that the phenomenon can be operative in vivo, and its potential role in the modulation of the carcinogenic process therefore has been discussed.16 In order to expand the observations obtained histochemically in tissue sections, the present study was designed to ascertain (1) whether a GGT-dependent lipid peroxidation process can be induced in biological membranes of living cells, i.e., cells retaining their antioxidant defense systems; and (2) whether this effect can be sustained by the endogenous GGT, highly reexpressed at the surface of living HepG2 human hepatoma cells. The results indicate that in the presence of GGT substrates and chelated iron a GGTdependent, nonlethal lipid peroxidation process can easily be ignited, involving the membranes of both normal and neoplastic liver cells, and liable to be propagated to lipid substrates in the cellular outer environment.
EXPERIMENTAL PROCEDURES
Chemicals All reagents, including cell culture media, were from Sigma (St. Louis, MO). Isolation of rat liver microsomes Adult Sprague Dawley rats (200–250 g; Charles River, Italy) were sacrificed by exsanguination under diethylether anaesthesia; livers were quickly excised and homogenized (33%) in 5 mM EDTA, 100 mM Tris-HCl buffer (pH 7.4). Liver homogenates were centrifuged at 9,000 1 g for 20 min. The supernatants were removed and centrifuged at 100,000 1 g for 45 min. Pellets (Åmicrosomes) were resuspended in the same buffer and stored at 0807C until use. Hepatocyte and HepG2 hepatoma cell cultures Isolated rat hepatocytes were obtained by collagenase perfusion from the same animals as above. Isolated cells were resuspended (1 1 106 cells/ml) in Waimouth MB 752/1 medium containing 10 mg/ml insulin, 10% fetal calf serum, 100 U/ml penicillin, and 100 mg/ml streptomycin (Sigma), plated in 5-cm collagencoated Falcon petri dishes (2.5 ml/dish) and allowed to adhere for 4 h at 377C; medium (including nonadherent, dead cells) was then changed once with the same medium (without serum), and cells were used for the experiments. Human HepG2 hepatoma cells, maintained in DMEM/F12 medium containing penicillin and streptomicin (see above) plus 10% fetal calf serum, were plated in 25-cm2 Falcon culture flasks, and used before reaching confluence. Prooxidant incubations and lipid peroxidation All incubations were performed in Hank’s buffer, pH 7.2. Reagents were added with suitable aliquots of concentrated stock solutions, in order to obtain the final concentrations indicated. Rat liver microsomes were incubated at the concentration of 1 mg protein/ml. Hepatocytes or HepG2 hepatoma cells were exposed to fresh medium, in which the reagents had been dissolved at the indicated concentrations. At the indicated times, lipid peroxidation was determined as the content of malonaldehyde (MDA), or, for selected experiments, as thiobarbituric acid reacting substances (TBARS). For the determination of MDA, medium was discarded and cells were yielded in 5% TCA; after centrifugation, TCA extracts were then reacted with 1 vol 0.67% thiobarbituric acid
/ 2b25 2380 Mp 854 Wednesday Dec 18 01:17 AM EL–FRB 2380
Gamma-glutamyl transpeptidase and lipid peroxidation
Fig. 1. GGT-stimulation of iron-dependent lipid peroxidation (production of malonaldehyde, MDA) in rat liver microsomes. Incubations were carried out at 377C in Hank’s medium, and had the following composition: (h) 2 mM GSH, 1.5–0.15 mM ADP-Fe3/, 200 mU/ml GGT; (l) as above, plus 20 mM glycyl-glycine; (s) 2 mM GSH, 1.5–0.15 mM ADP-Fe3/, 20 mM glycyl-glycine (no added GGT); (m) as for l, plus 200 mM BHT. One typical experiment of five is shown.
(TBA) at 1007C for 10 min.18 For the determination of TBARS, 2 vol TBA reagent were added to each culture flask, and the mixture was reacted at 957C for 15 min, essentially as described.19
855
as expected.22 However, the addition of GGT to the incubation mixture greatly stimulated the process, and even higher lipid peroxidation levels were attained when the incubation included glycyl-glycine, acting as g-glutamyl acceptor and therefore as a stimulant for GGT activity.20 The addition of the lipid-soluble, chainbreaking antioxidant BHT resulted in the suppression of GGT-stimulated lipid peroxidation. The intensity of the stimulation showed a linear dose-dependence on the GSH concentration, as reported in Fig. 2. The possibility was then verified that the described GGT-stimulated lipid peroxidation might involve the membranes of intact, living cells (Fig. 3). In 24 h-cultured rat hepatocytes, a lipid peroxidation developed in the presence of GSH, glycyl-glycine, and ADP-iron; again, as in the case of liver microsomes, lipid peroxidation was markedly stimulated by the addition of GGT. Up to the latest time of incubation studied (90 min), hepatocyte viability as assessed by LDH leakage was not affected by lipid peroxidation (not shown). As shown in Table 1, the stimulation of lipid peroxidation was dependent on the addition of glycyl-glycine, which again points to a crucial role played by GGT activity in this effect. The extent of GGT-stimulated, GSH-dependent lipid peroxidation in cultured hepatocytes was proportionally decreased by decreasing the ADP-iron concentration; in all cases, however, ADP-iron plus GGT caused higher levels of lipid peroxidation than ADPiron alone (Fig. 4).
Other determinations GGT activity was determined according to Huseby and Stro¨mme20 as modified by Edwards.21 Protein content was determined by the method of Lowry. Statistical significance of data was assessed by the Student’s t test. RESULTS
Rat liver microsomes were chosen as a membrane lipid substrate for the envisaged prooxidant effects of GGT activity. For these and most subsequent experiments, the concentrations of GSH and glycyl-glycine (2 and 20 mM, respectively) were chosen, taking into account the corresponding Km values of GGT for both, in order to obtain the vmax of GGT activity.20 The stimulation by GGT on the iron-dependent lipid peroxidation of rat liver microsomes is shown in Fig. 1; microsomes exposed only to GSH and ADP-chelated ferric iron exhibited increasing levels of lipid peroxidation,
Fig. 2. Glutathione-dependence of the GGT-mediated lipid peroxidation (production of malonaldehyde, MDA) in rat liver microsomes. Incubations were carried out (60 min, 377C) in Hank’s medium containing 2 mM GSH, 20 mM glycyl-glycine, 200 mU/ml GGT, 1.5–0.15 mM ADP-Fe3/. Data shown are means { S.E.M. of three experiments.
/ 2b25 2380 Mp 855 Wednesday Dec 18 01:17 AM EL–FRB 2380
856
A. PAOLICCHI et al.
Fig. 3. Prooxidant effect of GGT on cultured rat hepatocytes. Incubations were carried out in Hank’s medium containing 2 mM GSH, 1.5–0.15 mM ADP-Fe3/ and 20 mM glycyl-glycine, in the presence (l) or absence (s) of GGT (200 mU/ml). Data are means { S.E.M. of triplicate samples; one typical experiment of three is shown. Repeated experiments gave fully comparable results. * Significantly different from the 0-time value, p õ 0.01. ** Significantly different from the value obtained in the absence of GGT, p õ 0.01.
Further experiments were performed with human hepatoma HepG2 cells, in which a high intrinsic GGT activity (67 mU/mg protein) was detected. Because the levels of lipid peroxidation inducible in hepatoma cells (as in the case of other tumor cells) are generally much lower than in normal hepatocytes,23 for the determination of lipid peroxidation in HepG2 cells, the less specific but more sensitive determination of TBARs was employed.19 As reported in Table 2, the exposure of HepG2 cells to GSH, in the presence of ADP-Fe3/, led to the development of lipid peroxidation; the latter
Fig. 4. Influence of different concentrations of ADP-chelated iron on GGT-dependent lipid peroxidation in cultured rat hepatocytes. Cells were incubated for 60 min at 377C in Hank’s buffer containing 2 mM GSH, 20 mM glycyl-glycine, and the indicated concentrations of ADP-Fe3/, in the presence (l) or absence (s) of GGT (200 mU/ ml). Data are means { S.E.M. of triplicate samples; one typical experiment of three is shown. * Significantly different from the respective values obtained in the absence of GGT, p õ 0.01. (ns) Not significantly different from the 15 mM Fe3/ value.
was further stimulated by the addition of the GGT-cosubstrate glycyl-glycine, and was inhibited by the addition of the GGT inhibitor acivicin, indicating the dependence of the process on the cell-bound GGT activity. The extent of GGT-dependent lipid peroxidation was compared with that obtainable with other well-established prooxidant mixtures (Table 3), and appeared somewhat higher. No alteration of cell viability as assessed by LDH leakage was observed during GGT-me-
Table 1. GGT-Mediated Lipid Peroxidation (Production of Malonaldehyde, MDA) in Primary Hepatocyte Culturesa Incubation System Hepatocytes (control) Hepatocytes / GSH / GGT / ADP-Fe / ADP-Fe / gly2 / ADP-Fe / gly2 / BHT Hepatocytes / GSH / gly2 / ADP-Fe (no GGT)
MDA Released, nmol/mg proteinb 0.23 0.33 3.56 6.63 0.31 2.39
{ { { { { {
0.04 0.09 (ns) 0.04** 1.23 0.12 (ns) 0.48*
a Incubations were carried out at 377C for 60 min. Concentration of reagents was as follows: GSH, 2 mM; glycyl-glycine, 20 mM; ADP-Fe, 1.5 mM–150 mM; GGT, 200 mU/ml; BHT, 200 mM. b Data are means { S.E.M. of three experiments. * Significantly different from control, p õ 0.02. ** Significantly different from the value obtained without ADP-Fe, p õ 0.001. (ns) Not significantly different from control.
/ 2b25 2380 Mp 856 Wednesday Dec 18 01:17 AM EL–FRB 2380
Gamma-glutamyl transpeptidase and lipid peroxidation Table 2. GSH-Dependent Lipid Peroxidation (Production of TBA-Reactive Substances, TBARS) in GGT-Positive Hep-G2 Hepatoma Cellsa Incubation Systemb HepG2 HepG2 HepG2 HepG2 HepG2
(control) / ADP-Fe / ADP-Fe / GSH / ADP-Fe / GSH / gly2 / ADP-Fe / GSH / gly2 / acivicin
857
Table 4. Accumulation in the Incubation Medium of the Products of GSH-Dependent Lipid Peroxidation in GGT-Positive Hep-G2 Hepatoma Cellsa
TBARS, nmol/mg proteinc 0.71 3.30 4.29 5.11 2.54
{ { { { {
0.05 0.02 0.05** 0.18* 0.11***
Incubations were carried out at 377C for 60 min. Concentration of reagents was as follows: GSH, 2 mM; glycylglycine, 20 mM; ADP-Fe, 1.5 mM–150 mM; acivicin, 1 mM. c Data are means { S.E.M. of triplicate samples; one typical experiment of four is shown. Repeated experiments gave fully comparable results. * Significantly different from the value obtained without gly2, p õ 0.02. ** Significantly different from the value obtained without GSH, p õ 0.001. *** Significantly different from the value obtained without acivicin, p õ 0.001; not significantly different from ‘‘HepG2 / ADPFe’’. a
b
TBARS Incubation Time (min)
nmol/mg cell protein
nmol/ml supernatant
0 30 60
0.40 { 0.03 1.27 { 0.09 0.89 { 0.02*
0 1.25 { 0.72 3.38 { 0.12*
At the indicated times, culture supernatants were separated from cells and centrifuged (800 g, 10 min); cells were washed once with fresh medium; TBARS was then determined separately in cells and supernatants. a Incubations were carried out at 377C. Concentration of reagents was as follows: cells, 1 mg protein/ml; GSH, 2 mM; glycyl-glycine, 20 mM; ADP-Fe, 1.5 mM–150 mM. b Data are means { S.E.M. of triplicate samples; one typical experiment of three is shown. * Significantly different from the respective 30 min value, p õ 0.05.
DISCUSSION
diated lipid peroxidation, up to the latest time of incubation studied (90 min) (not shown). When lipid peroxidation was determined separately, in cell monolayers as compared to cell supernatants, the extracellular, but not intracellular, accumulation of the produced TBARS was observed (Table 4). It was then investigated whether the lipid peroxidation process stimulated by the GGT activity present at the cell surface might be propagated to lipid substrates present in the extracellular environment. As shown in Table 5, the addition of GGT-rich HepG2 cells to a suspension of rat liver microsomes, in the presence of chelated iron and GGT substrates, greatly stimulated microsomal lipid peroxidation; the effect was GGT-dependent, as indicated again by the fact that the addition of the GGT inhibitor serine–boric acid complex caused its suppression. Table 3. GSH-Dependent Lipid Peroxidation in GGT-Positive HepG2 Hepatoma Cells, as Compared to Lipid Peroxidation Inducible With Other Prooxidant Mixturesa Incubation System HepG2 HepG2 HepG2 HepG2 HepG2
/ / / / /
GSH / gly2 / ADP-Feb Fe-NTA 100 mM CuOOH 500 mM ADP-Fe 1.5 mM-150 mM buffer (control)
TBARS, nmol/mg proteinc 5.70 2.71 1.88 3.16 0.77
{ { { { {
0.21 0.28 0.15 0.80 0.06
Incubations were carried out at 377C for 60 min. Concentration of reagents was: GSH, 2 mM; glycyl-glycine, 20 mM; ADP-Fe, 1.5 mM–150 mM. c Data are means { S.E.M. of triplicate samples; one typical experiment of three is shown. a
b
The data reported in the present study indicate that the GGT-mediated metabolism of GSH is liable to expose cellular membranes to a prooxidant insult, leading to the onset of lipid peroxidation. As previously suggested,14 the reason for this effect is allegedly due to the fact that GSH metabolites prior to their absorption by the cell can participate in metal-catalyzed redox reactions, leading to the production of radical oxygen species. In particular, the thiol group of the dipeptide cysteinyl-glycine, resulting from the GGT-mediated cleavage of GSH, being more dissociable at neutral pH than the thiol of GSH itself,24 is more prone to interact with transition metal ions; in the case of ferric iron, cysteinyl-glycine can accomplish its reduction to ferrous ions, whereas the oxidation of the thiol can lead Table 5. Lipid Peroxidation in Rat Liver Microsomes Exposed to GGT-Rich HepG2 Hepatoma Cells, in the Presence of GGT Substratesa Incubation System
TBARS, nmol/mg proteinb
Microsomes / GSH / gly2 / ADP-Fe / HepG2 cells / HepG2 cells / serine borate HepG2 / GSH / gly2 / ADP-Fe
14.8 26.0 12.6 2.9
{ { { {
1.38 2.24* 1.59 (ns) 0.24
a Incubations were carried out at 377C for 60 min. Concentration of reagents was as follows: GSH, 2 mM; glycyl-glycine, 20 mM; ADP-Fe, 1.5 mM–150 mM; microsomes, 1 mg protein/ml; serine 10 mM-boric acid 20 mM. b Data are means { S.E.M. of triplicate samples; one typical experiment of three is shown. * Significantly different from the value obtained without HepG2 cells, p õ 0.02. (ns) Not significantly different from the value obtained without HepG2 cells.
/ 2b25 2380 Mp 857 Wednesday Dec 18 01:17 AM EL–FRB 2380
858
A. PAOLICCHI et al.
to the formation of a cysteinyl-glycine thiyl radical.14 As in other model systems of thiol-driven iron-dependent lipid peroxidation,22 further interactions of reduced iron with dioxygen can then cause the appearance of species capable to initiate lipid peroxidation in membrane lipids.25 The ability of GGT to stimulate lipid peroxidation recently has been reported biochemically, in systems including the purified enzyme and a soluble lipid substrate.14 Also the development of GGT-dependent lipid peroxidation has been demonstrated by histochemical means in chemically induced GGT-positive preneoplastic lesions of rat livers, following the exposure in vitro of tissue cryostat sections to GSH, glycyl-glycine, and chelated iron.16,17 The data reported in the present study add to the biological significance of the phenomenon, in that they indicate that the prooxidant effects of GGT activity can involve as well the lipids of cellular membranes, including those of intact living cells. The addition of GGT plus its substrates GSH and glycyl-glycine, in the presence of ADP-chelated ferric iron, results, in fact, in a marked stimulation of lipid peroxidation both in rat liver microsomes and in isolated hepatocytes. Moreover, the plasma membrane-bound GGT activity intrinsically present at the outer surface of HepG2 human hepatoma cells is equally able to stimulate a lipid peroxidation process, which involves the membranes of the cells themselves, and also can be propagated to membranes present in the surrounding environment. Interestingly, both in the case of isolated rat hepatocytes and of HepG2 hepatoma cells, GGT-stimulated lipid peroxidation was not accompanied by acute alterations of cell viability. In this respect, it is tempting to speculate that this also could be a result of GGT activity, whose role in potentiating cellular resistance to oxidative stress has been documented.9 – 11 Indeed, data shown in Table 4 suggest that the products of the lipid peroxidation process do not accumulate within cells, where they allegedly undergo efficient metabolism. The experimental conditions employed in the present study included relatively high concentrations of reagents (GSH, glycyl-glycine, ADP-iron). These were chosen based on the described Km values of GGT for its substrates, in order to optimize the conditions for GGT activity and thus facilitate the detection of its prooxidant effects. However, GGT-stimulated lipid peroxidation exhibited a linear dependence on GSH concentration (Fig. 2), down to GSH concentrations (50–10 mM) in the range of those detectable in rat or human plasma;26 on the other hand, in the liver, where hepatocytes are net exporters of GSH, GSH can reach remarkably higher local concentrations. It seems, therefore, likely that although at levels allegedly lower that
those observed in the present study the described GGTstimulation of lipid peroxidation can take place in in vivo conditions as well. With respect to the requirement of the reaction for chelated iron, the linear dose-dependence shown in Figure 4 again suggests that GGT stimulation of lipid peroxidation is still possible down to very low concentrations of iron. On the other hand, the participation of iron in vivo to such reactions would require its prior release from its storage protein, transferrin, in the first place. In this respect, a number of studies have shown, however, that the release of iron from transferrin is facilitated in several conditions, such as mild acidic pH, exposure to reducing agents, and binding to cellular receptors.27,28 Previous biochemical and histochemical studies14,16 have indeed shown that GGT-stimulation of lipid peroxidation can be sustained by transferrin iron alone. The biological significance of GGT-dependent lipid peroxidation in vivo might be multifold. Varying levels of GGT activity can be detected in most cell types, ranging from very low, as in hepatocytes, to rather high, as in tubular epithelium. It is conceivable that the prooxidant effects of GGT activity are normally balanced by its established role in favoring the cellular uptake of precursors for GSH resynthesis, thus allowing the reconstitution of cellular antioxidant defenses. On the other hand, in cells overexpressing the enzyme, conditions could be in favor of a higher extracellular production of cysteinyl-glycine, liable to ignite lipid peroxidation through interactions with iron. One major aspect involved in the described phenomena can be tumor progression. Increasing evidence suggests, in fact, that the expression of GGT may play a role in tumor progression: transfection with GGT resulted in a growth advantage for mouse hepatoma cells;29 GGT is an established marker of progression in mouse skin carcinogenesis;29 in metastasis from human colon carcinomas, GGT is higher than in primitive tumors;7 transfection of epithelial cells with the oncogene ras while resulting in the appearance of a metastatic behavior is also accompanied by GGT expression.30 In addition, a recent study has shown in monoblastic leukemia cells that the inhibition of GGT can inhibit cell proliferation and favor differentiation.31 The enzyme catalytic site of cell-bound GGT is located at the cell surface, and its substrates and the potential targets of the prooxidant reactions can be found in the outer cell environment, i.e., out of reach and control for most of the cellular antioxidant systems. In this way, despite the fact that in tumor cells antioxidants may often be increased,32 cells reexpressing GGT activity at their surface might anyway promote prooxidant conditions in their vicinity, exposing themselves and the surround-
/ 2b25 2380 Mp 858 Wednesday Dec 18 01:17 AM EL–FRB 2380
Gamma-glutamyl transpeptidase and lipid peroxidation
ing cells to the effects of prooxidant species and lipid peroxidation products. The latter have been shown indeed to possess a wide range of antiproliferative,33,34 as well as mutagenic,35 – 37 tumor-initiating38 and tumorpromoting effects;39 – 41 the possibility therefore exists that some of these effects are involved in the described association of GGT expression with tumor progression. In conclusion, the data reported in the present study show that the metabolism of GSH mediated by (soluble and cell-bound) GGT activity can result in conditions favoring the ignition of lipid peroxidation in cell membranes. Basing on a number of experimental observations, GGT activity is believed to play a role in tumor progression. Future studies in our laboratory will be dedicated to investigating the possibility that aspects of the envisaged role of GGT in tumor progression may be mediated at least partly by the phenomenon of GGTdependent lipid peroxidation.
12. 13. 14.
15. 16.
17.
18.
Acknowledgements — The support of Consiglio Nazionale delle Ricerche (Italy), Progetto A.C.R.O., is gratefully acknowledged. Additional financial support was derived from the Association for International Cancer Research (U.K.).
19.
REFERENCES
21.
1. Sies, H. Strategies of antioxidant defense. Eur. J. Biochem. 215:213–219; 1993. 2. Tate, S. S.; Meister, A. g-Glutamyl transpeptidase from kidney. Methods Enzymol 113:400–419; 1985. 3. Rutenburg, A. M.; Kim, H.; Fischbein, J. W.; Hanker, J. S.; Wasserkrug, H. L.; Seligman, A. M. Histochemical and ultrastructural demonstration of g-glutamyl transpeptidase activity. J. Histochem. Cytochem. 17:517–526; 1969. 4. Lindros, K. O.; Penttila, K. E.; Janzen, J.-W. G.; Moorman, A. F. M.; Spelsky, H.; Israel, Y. The gamma-glutamyltransferase/glutamine synthetase activity ratio: A powerful marker of the acinar origin of hepatocytes. J. Hepatol. 8:338–343; 1989. 5. Cameron, R.; Kellen, J.; Kolin, A.; Malkin, A.; Farber, E. gGlutamyltransferase in putative premalignant liver cell populations during hepatocarcinogenesis. Cancer Res. 38:823–829; 1978. 6. Cheng, S.; Nassar, K.; Levy, D. g-Glutamyl transpeptidase activity in normal, regenerating and malignant hepatocytes. FEBS Lett. 85:310–312; 1978. 7. Munjal, D. D. Concurrent measurements of carcinoembryonic antigen, glucose phosphate isomerase, g-glutamyltransferase, and lactate dehydrogenase in malignant, normal adult, and fetal colon tissues. Clin. Chem. 26:1809–1812; 1980. 8. Hanigan, M. H.; Frierson, H. F., Jr.; Brown, J. E.; Lovell, M. A.; Taylor, P. T. Human ovarian tumors express g-glutamyl transpeptidase. Cancer Res. 54:286–290; 1994. 9. Rajpert-De Meyts, E.; Shi, M.; Chang, M.; Robison, T. W.; Groffen, J.; Heisterkamp, N.; Forman, H. J. Transfection with g-glutamyl transpeptidase enhances recovery from glutathione depletion using extracellular glutathione. Toxicol. Appl. Pharmacol. 114:56–62; 1992. 10. Shi, M.; Gozal, E.; Choy, H. A.; Forman, H. J. Extracellular glutathione and g-glutamyl transpeptidase prevent H2O2-induced injury by 2,3-dimethoxy-1,4-naphthoquinone. Free Rad. Biol. Med. 15:57–67; 1993. 11. Chang, M.; Shi, M.; Forman, H. J. Exogenous glutathione pro-
20.
22. 23.
24. 25. 26.
27. 28. 29.
30.
31.
859
tects endothelial cells from menadione toxicity. Am. J. Physiol. 6:L634–L637; 1992. Hanigan, M. H.; Pitot, H. C. Gamma-glutamyl transpeptidase— Its role in carcinogenesis. Carcinogenesis 6:165–172; 1985. Farber, E. Clonal adaptation during carcinogenesis. Biochem. Pharmacol. 39:1837–1846; 1990. Stark, A.-A.; Zeiger, E.; Pagano, D. A. Glutathione metabolism by g-glutamyltranspeptidase leads to lipid peroxidation: Characterization of the system and relevance to hepatocarcinogenesis. Carcinogenesis 14:183–189; 1993. Dianzani, M. U. Lipid peroxidation and cancer. Crit. Rev. Oncol. Hematol. 15:125–147; 1993. Pompella, A.; Paolicchi, A.; Dominici, S.; Comporti, M.; Tongiani, R. Selective colocalization of lipid peroxidation and protein thiol loss in chemically induced hepatic preneoplastic lesions: The role of g-glutamyl transpeptidase activity. Histochem Cell Biol. 106:275–282; 1996. Stark, A.-A.; Russell, J. J.; Langenbach, R.; Pagano, D. A.; Zeiger, E.; Huberman, E. Localization of oxidative damage by a glutathione-g-glutamyl transpeptidase system in preneoplastic lesions in sections of livers from carcinogen-treated rats. Carcinogenesis 15:343–348; 1994. Pompella, A.; Maellaro, E.; Casini, A. F.; Ferrali, M.; Ciccoli, L.; Comporti, M. Measurement of lipid peroxidation in vivo: A comparison of different procedures. Lipids 22:206–211; 1987. Buege, J.; Aust, S. D. Microsomal lipid peroxidation. Methods Enzymol. 52:302–310; 1978. Huseby, N. E.; Stro¨mme, J. H. Practical points regarding routine determination of g-glutamyl transferase (g -GT) in serum with a kinetic method at 377C. Scand. J. Clin. Lab. Invest. 34:357–361; 1974. Edwards, A. M. Regulation of g-glutamyltranspeptidase in rat hepatocyte monolayer cultures. Cancer Res. 42:1107–1115; 1982. Tien, M.; Bucher, J. R.; Aust, S. D. Thiol-dependent lipid peroxidation. Biochem. Biophys. Res. Commun. 107:279–285; 1982. Canuto, R. A.; Ferro, M.; Muzio, G.; Bassi, A. M.; Leonarduzzi, G.; Maggiora, M.; Adamo, D.; Poli, G.; Lindahl, R. Role of aldehyde metabolizing enzymes in mediating effects of aldehyde products of lipid peroxidation in liver cells. Carcinogenesis 15:1359–1364; 1994. Stark, A.-A.; Arad, A.; Siskindovich, S.; Pagano, D. A.; Zeiger, E. Effect of pH on mutagenesis by thiols in Salmonella typhimurium TA102. Mutat. Res. 224:89–94; 1989. Minotti, G.; Aust, S. D. An investigation into the mechanisms of citrate-Fe2/-dependent lipid peroxidation. Free Rad. Biol. Med. 3:379–387; 1987. Michelet, F.; Gueguen, R.; Leroy, P.; Wellman, M.; Nicolas, A.; Siest, G. Blood and plasma glutathione measured in healthy subjects by HPLC: Relation to sex, aging, biological variables and life habits. Clin. Chem. 41:1509–1517; 1995. Brieland, J. K.; Fantone, J. C. Ferrous iron release from transferrin by human neutrophil-derived superoxide anion: Effect of pH and iron saturation. Arch. Bioch. Biophys. 248:78–83; 1991. Sipe, D. M.; Murphy, R. F. Binding to cellular receptors results in increased iron release from transferrin at mildly acidic pH. J. Biol. Chem. 266:8002–8007; 1991. Warren, B. S.; Naylor, M. F.; Winberg, L. D.; Yoshimi, N.; Volpe, J. P. G.; Gimenez-Conti, I.; Slaga, T. J. Induction and inhibition of tumor progression. Proc. Soc. Exp. Biol. Med. 202:9–15; 1993. Braun, L.; Goyette, M.; Yaswen, P.; Thompson, N. L.; Fausto, N. Growth in culture and tumorigenicity after transfection with the ras oncogene of liver epithelial cells from carcinogen-treated rats. Cancer Res. 47:4116–4124; 1987. Bauvois, B.; Laouar, A.; Rouillard, D.; Wietzerbin, J. Inhibition of g-glutamyl transpeptidase activity at the surface of human myeloid cells is correlated with macrophage maturation and transforming growth factor b production. Cell Growth Differ. 6:1163–1170; 1995.
/ 2b25 2380 Mp 859 Wednesday Dec 18 01:17 AM EL–FRB 2380
860
A. PAOLICCHI et al.
32. Sun, Y. Free radicals, antioxidant enzymes and carcinogenesis. Free Rad. Biol. Med. 8:583–599; 1990. 33. Esterbauer, H.; Schaur, R. J.; Zollner, H. Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Rad. Biol. Med. 11:81–128; 1991. 34. Gonzalez, M. J.; Schemmel, R. A.; Gray, J. I.; Dugan, L., Jr.; Sheffield, L. G.; Welsch, C. W. Effect of dietary fat on growth of MCF-7 and MDA-MB231 human breast carcinomas in athymic nude mice: Relationship between carcinoma growth and lipid peroxidation product levels. Carcinogenesis 12:1231– 1235; 1991. 35. Esterbauer, H.; Eckl, P.; Ortner, A. Possible mutagens derived from lipids and lipid precursors. Mutat. Res. 238:223–233; 1990. 36. Canonero, R.; Martelli, A.; Marinari, U. M.; Brambilla, G. Mutation induction in Chinese hamster lung V79 cells by five alk2-enals produced by lipid peroxidation. Mutat. Res. 244:153– 156; 1990. 37. Zhang, J.-R.; Sevanian, A. The genotoxic effects of arachidonic acid in V79 cells are mediated by peroxidation products. Toxicol. Appl. Pharmacol. 121:193–202; 1993. 38. Chung, F.-L.; Chen, H.-J. C.; Guttenplan, J. B.; Nishikawa, A.; Hard, G. C. 2,3-Epoxy-4-hydroxynonanal as a potential tumorinitiating agent of lipid peroxidation. Carcinogenesis 14:2073– 2077; 1993. 39. Bartoli, G. M.; Bartoli, S.; Galeotti, T.; Bertoli, E. Superoxide
dismutase content and microsomal lipid composition of tumours with different growth rates. Biochim. Biophys. Acta 620:205– 211; 1980. 40. O’Brian, C. A.; Ward, N. E.; Weinstein, I. B.; Bull, A. W.; Marnett, L. J. Activation of rat brain protein kinase C by lipid oxidation products. Biochem. Biophys. Res. Commun. 155:1374– 1380; 1988. 41. Greenley, T. L.; Davies, M. J. Direct detection of radical generation in rat liver nuclei on treatment with tumour-promoting hydroperoxides and related compounds. Biochim. Biophys. Acta 1226:56–64; 1994.
ABBREVIATIONS
BHT—butylated hydroxytoluene CuOOH—cumene hydroperoxide Fe-NTA—ferric nitrilo-triacetate GGT—gamma-glutamyl transpeptidase gly2—glycyl-glycine GSH—reduced glutathione MDA—malonaldehyde TBARS—thiobarbituric acid-reacting substances
/ 2b25 2380 Mp 860 Wednesday Dec 18 01:17 AM EL–FRB 2380