Interactions of [alpha,beta]-unsaturated carbonyl compounds with the glutathione-related biotransformation system

IntroductionModulation of glutathione-related biotransformation steps may play a role in important phenomena as anticarcinogenicity and multidrug resistance. Glutathione-related biotransformation comprises three main aspects i.e. glutathione, the glutathione S-transferases and the multidrug resistance associated protein pump. In Figure 6.1 is shown how the levels and relative activities of these three entities interact.The research presented in this thesis focused on the effects of the ubiquitous class ofα,β-unsaturated carbonyl compounds on these glutathione-related processes, especially glutathione S-transferase P1-1, while a secondary aim was to provide insight in the metabolism of these compounds.Figure 6.1 Interactions between the three aspects of the glutathione-related biotransformation system.Firstly, studies were conducted to expand understanding of the mechanisms of both GST inhibition and glutathione conjugation.Secondly, the effects of a series of exogenousα,β-unsaturated carbonyl compounds on the glutathione-related biotransformation were studied in a cellular system, as all three aspects are integrated in such a system and the relative importance of the various steps can be estimated.Finally, an important endogenousα,β-unsaturated ketone, prostaglandin A 2 , was selected and its metabolism and effects were studied to emphasise the significance of the glutathione-related metabolism for endogenous compounds and obtain insight into the possible role of GST inhibition in regulation of physiological processes.SummaryTo elucidate mechanistic features of the covalent interaction betweenα,β-unsaturated carbonyl compounds and GSTP1-1 and to study the involvement of the cysteine residues in this interaction, investigations were performed with mutants of GSTP1-1 ( chapter 2 ). In these mutants cysteine 47 and/or cysteine 101 were mutated into a serine. Theα,β-unsaturated carbonyl compounds, used in this study, inhibited GSTP1-1 activity, but when both cysteine residues were mutated, almost no inhibition of GSTP1-1 could be observed. Mutation of only the cysteine 47 residue already had the same effect. However, especially high concentrations of theα,β-unsaturated compounds still inhibited the double mutant GSTP1-1 to a certain extent, suggesting that lower reactivity sites in the enzyme can be modified as well. From the compounds studied, ethacrynic acid, acrolein, curcumin, and 4-hydroxy-2-nonenal were the most potent covalent inhibitors. As the compounds used are Michael acceptors, reversal of inhibition of activity by an excess of glutathione was investigated. Only for ethacrynic acid and crotonaldehyde, inhibition could be totally reversed. But inhibition by, for instance, acrolein could not be reversed; experiments using matrix-assisted laser desorption/ionization-mass spectrometry (MALDI-MS) showed that covalent cross linking between subunits occurred by acrolein (results not shown), as was demonstrated for eugenol previously (Rompelberg et al ., 1996). For the other compounds used, partial restoration of GSTP1-1 activity was observed, again indicating that reactions with other amino acids or binding places play a role as well.The results of this chapter indicate thatα,β-unsaturated carbonyl derivatives inhibit GSTP1-1 irreversibly mainly by binding to the cysteine residues, especially cysteine 47. This covalent inactivation might in particular be important, when glutathione concentrations are low.The potential relevance of stereoselective formation of glutathione conjugates ofα,β-unsaturated carbonyl compounds for the actual effects of these compounds led to the investigation of the relative formation of the two diastereoisomers of model compound ethacrynic acid. Ethacrynic acid has become a thoroughly and widely studied compound, lately especially with regard to multidrug resistance (Schultz et al ., 1997; Shen et al ., 1997). Glutathione conjugation of ethacrynic acid leads to the formation of two diastereoisomers, chemically in almost equal amounts (48:52). Although it has been shown that rat GST mu did not catalyse glutathione conjugation stereoselectively (Ploemen et al ., 1993b), human GSTP1-1 is stereospecific for the formation of one diastereoisomer, in this thesis designated as diastereoisomer A. GSTA1-1, but not GSTA1-2 and GSTA2-2 is stereoselective for the same diastereoisomer. No significant deconjugation of the diastereoisomeric mixture or of diastereoisomer A alone could be detected, chemically or upon addition of GSTP1-1; the latter probably due to inhibition of the enzyme. Clearly the equilibrium for glutathione conjugation of ethacrynic acid is strongly in favour of product formation. As a first step to study the role of conjugation in relation to the other glutathione-related aspects, IGR-39 human melanoma cells, containing high levels of GSTP1-1, were exposed to ethacrynic acid. Diastereoisomer A was preferentially produced in the medium. Although this has not yet been proven definitively, this was probably due to the stereoselective formation of diastereoisomer A by GSTP1-1 catalysis rather than stereoselective transport of the conjugates ( chapter 3 ).From this chapter it is clear that the chemical and enzyme catalysed equilibria for the reaction between ethacrynic acid and glutathione are strongly in favour of product formation. GSTP1-1 stereospecifically catalyses the glutathione conjugation of ethacrynic acid GSTA1-1 is stereoselective for the same diastereoisomer. Furthermore it was shown that stereoselectivity plays a role in cellular systems.As glutathione conjugation and inhibition of GST activity have mainly been studied using cytosol or purified enzymes, a method was developed to investigate modulation of glutathione conjugation in intact IGR-39 human melanoma cells by the quantification of the excretion of S-(2,4-dinitrophenyl)glutathione (DNPSG), the glutathione conjugate of the standard substrate 1-chloro-2,4-dinitrobenzene (CDNB). By investigating intracellular glutathione levels, GST activity and intra- and extracellular DNPSG concentration, some determinants involved in the mechanisms of inhibition of DNPSG excretion could be identified ( chapter 4 ). These mechanisms include depletion of glutathione levels, reversible and irreversible inhibition of glutathione S-transferase activity, and modulation of the efflux of glutathione conjugates by an effect on the multidrug resistance associated protein (MRP) pump.Using this assay, a series ofα,β-unsaturated carbonyl compounds were tested for their inhibiting properties of DNPSG excretion. Curcumin, an antioxidant and anticarcinogenic compound, was the most potent inhibitor of DNPSG excretion in these cells, followed by ethacrynic acid. Citral did not show any effect of DNPSG excretion up to 100 mM and acrolein was too toxic to get any effect ( chapter 4 ). The mechanisms of inhibition differed between the variousα,β-unsaturated carbonyl compounds. For curcumin and ethacrynic acid, glutathione depletion, reversible inhibition of GSTs and covalent modification of GSTP1-1 all three play a role in the inhibition of DNPSG excretion. However for trans -2-hexenal and cinnamaldehyde, reversible GST inhibition seems to be the major determinant for its effect. Crotonaldehyde mainly inhibits DNPSG excretion by depleting glutathione, but reversible inhibition does presumably plays a role as well. Curcumin and ethacrynic acid also inhibit the efflux of DNPSG by an effect on the transport of the glutathione conjugate out of the cells, probably by the multidrug resistance associated protein (MRP) pump. Indeed, it has been shown that the glutathione conjugate of EA has an equal efficiency for transport by the MRP pump compared to DNPSG (Zaman et al. , 1996).Some of theα,β-unsaturated carbonyl compounds, used in this study, thus inhibit GST avtivity in human melanoma cells. They modulate the glutathione related biotransformation system in these cells in different ways, i.e. glutathione depletion, reversible and irreversible inhibition of GST activity and/or influence on the efflux of glutathione conjugates.For the endogenousα,β-unsaturated ketone prostaglandin A 2 a more complete picture of its metabolism and the influence on glutathione-related biotransformation could be obtained, as it was also possible to analyse PGA 2 -glutathione conjugate excretion into the medium during exposure of the cells to PGA 2 . After loading IGR-39 human melanoma cells with [ 3H] glycine and subsequent exposure to PGA 2 , both diastereoisomers of the PGA 2 -glutathione conjugate are excreted into the medium, however with a clear excess of the S-form. Previous work indicates that this is the result of the preferential formation of the S-form by GSTP1-1 that is present in the cells and not from a stereoselectivity in transport (Bogaards et al ., 1997; Evers et al ., 1997). Incubation of IGR-39 human melanoma cells with PGA 2 during 1 or 4 hours clearly influenced the glutathione-related metabolism. After 1 hour exposure, DNPSG excretion was reduced mainly due to inhibition of the efflux of the conjugate. Indeed, it has recently been shown that the glutathione conjugates of PGA 2 have a higher affinity for MRP compared to DNPSG (Evers et al ., 1997). After 4 hours, total DNPSG formation was reduced markedly, resulting from depletion of glutathione and reversible and irreversible inhibition of GSTs; inhibition of efflux then only played a minor role. Although irreversible inhibition already accounted for about 25% of the GST inhibition by PGA 2 , depletion of intracellular GSH with BSO resulted in an even higher level of covalent inactivation. Experiments with purified GSTP1-1 and mutants missing one or two cysteine residues, revealed that this covalent inhibition of GSTP1-1 resulted from the binding of PGA 2 to mainly the cysteine 47 moiety of the enzyme. This inactivation could be totally reversed by an excess of glutathione, indicative of a retro-Michael reaction ( chapter 5 ).The results of this chapter again prove that GSTP1-1 might play a role in scavenging alkylating agents especially when GSH concentrations are low, or conversely might serve as a storing or transport protein for physiologically important compounds such as PGA 2 .PerspectivesBoth endogenous and exogenousα,β-unsaturated carbonyl compounds thus appear to influence several aspects of the glutathione-related biotransformation system. They are conjugated to glutathione, thereby depleting glutathione and thus influencing the redox status of the cells, which plays a role in regulation of Phase II enzymes (Talalay et al ., 1995; Primiano et al ., 1997). These conjugates usually are less toxic than their parent compounds, but it is also possible that they undergo retro-Michael cleavage, releasing the reactive compound under different circumstances, then also influencing the redox status. The glutathione conjugates themselves are inhibitors of GST activity (product binding) and probably by binding the enzyme, they are transported to efflux pumps such as MRP. Stereoselectivity in the formation of GSH conjugates might influence function and toxicity ofα,β-unsaturated carbonyl compounds. Future studies should be performed to further elucidate the conjugation mechanism and the relevance of stereoselectivity in vivo . One can only speculate about the physiological importance of this phenomenon. Studies with the recently developed GST pi knock-out mice (Henderson et al. , 1996) and MRP knock-out mice (Wijnholds et al ., 1997) could give some useful information.α,β-Unsaturated carbonyl compounds can inhibit GST activity both by competitive inhibition on the active site as well as covalent inactivation on the cysteine residues of the enzyme. This covalent binding is reversible due to retro-Michael reaction and most likely constitutes a functional role as well.Man is exposed to substantial amounts of theseα,β-unsaturated carbonyl compounds every day, dependent on life style factors, such as diet, smoking, contact with traffic exhaust. Therefore these findings are of considerable relevance. Especially when considering the total daily exposure to the variousα,β-unsaturated carbonyl compounds, concentrations reached in the body might equal the concentrations used in this study. For instance: acrolein is present in wine up to about 3.8 ppm (70 mM) (Feron et al .,1991); curcumin, the major component of the spice curry, is widely used and consumption for adult Indians is estimated on about 125 mg/day (Opdyke and Letizia, 1983); cinnamaldehyde is present in food up to 700ppm (4.7 mM) (Feron et al ., 1991); adding the additional endogenously producedα,β-unsaturated carbonyl compounds, the combined exposure very likely influences the glutathione-related biotransformation system.The fact that endogenousα,β-unsaturated carbonyl compounds as 4-hydroxy-2-nonenal, trans -2-hexenal and prostaglandin A 2 are good covalent inhibitors of GSTP1-1 and that covalent modification occurs intracellularly, supports the assumption, that GSTP1-1 might not only play a role in glutathione conjugation but also has other cellular functions. In this respect one can think of GSTP1-1 as a transport or storage protein for endogenous compounds and/or as a general intracellular scavenging protein for electrophilic agents.The suggestion that GSTP1-1 might function as a storage for endogenous compounds, is a commonly accepted function of GSTs in general (Listowsky, 1988). GSTP1-1 is known to have a hydrophobic pocket, which binds fatty acids (Nishihira et al ., 1992). Experiments with fatty acids and human GSTP1-1 revealed that linolenic acid is capable of inhibiting GSTP1-1 activity thereby not affecting covalent modification of GSTP1-1 by ethacrynic acid (unpublished results). This means that it should be possible to bind covalently modified GSTP1-1 on the fatty acid binding site. Interesting possibilities arise when this would be possible with regard to biomonitoring exposure to electrophilic compounds as GSTP1-1 is a major GST present in erythrocytes.Closely linked to this storage function, is the possible function of GSTP1-1 as a transport protein. As indicated for prostaglandin A 2 in chapter 5 , GSTP1-1 can transport this compound intracellularly to the nucleus by binding it. Localization of GSTP1-1 in human tissue, using immunohistochemical techniques, indeed show the presence of this isoenzyme in the nucleus (Terrier et al ., 1990). Compounds that are delivered in the nucleus can for instance change thiols, from GST to a transcription factor or other protein and accordingly trigger all sorts of events. It becomes now more and more accepted that genes, involved in protection against carcinogens are regulated by the redox status of cells (Talalay et al ., 1995; Primiano et al ., 1997; Itoh et al ., 1997). Another aspect in the role of GSTP1-1 as a transport protein is the capability of the enzyme to bind products. For instance the glutathione conjugate of ethacrynic acid is an even better inhibitor of GSTP1-1 than the parent compound. One could clearly think of a role of GSTP1-1 in transporting glutathione conjugates from the site of formation to efflux pumps in the plasma membrane, for instance MRP.The third notion, that GSTP1-1 might function as a general scavenging protein, might especially be apparent when glutathione levels are low. Lipid peroxidation products as HNE and reactive oxygen species can thus be neutralized, but also other electrophilic compounds. The inactivation of GSTP1-1 by 4-hydroxy-2-nonenal (HNE) and the only partial recovery of activity after incubation with a molar excess of glutathione ( chapter 2 ) are in line with previous findings with H 2 O 2 (Sluis-Cremer et al ., 1996). The ability of otherα,β-unsaturated carbonyl compounds to inactivate purified GSTP1-1 as well as GSTP1-1 in cells ( chapter 2, 4 and 5 ), together with previous results (Berhane and Mannervik, 1990; Terada et al ., 1995) also support a general scavenging role of GSTP1-1.The significance of MRP in the maintenance of intracellular concentrations of both functional and toxic or carcinogenic agents is under current investigation. However,α,β-unsaturated carbonyl compounds may influence its transport activity, for a start by depletion of glutathione, which seems to be essential for MRP. Furthermore, as the glutathione conjugates of both EA and PGA 2 are substrates, the glutathione conjugates of otherα,β-unsaturated carbonyl derivates might be substrates as well. Future research should focus on structure activity relationships for MRP substrates. The importance of stereoselectivity in the transport of these conjugates by MRP also merits further investigation (Evers et al ., 1997, Loe et al ., 1997).ConclusionIn conclusion, the results in this thesis demonstrate for the first time that GST activity is inhibited in cells exposed toα,β-unsaturated carbonyl compounds. It also became clear that GST activity should not be studied on its own, but, as it is a part of a glutathione-mediated biotransformation system, it should be investigated in conjunction with glutathione levels and the multidrug resistance associated protein (MRP). Moreover, the apparent involvement of GSTP1-1 in the metabolism of the endogenous compound prostaglandin A 2 , indicates a possible role of this isoenzyme in regulation of cell proliferation. Mostα,β-unsaturated carbonyl compounds, studied in this thesis, interact with the glutathione-related biotransformation system (i.e. glutathione conjugation, glutathione depletion, both reversible and irreversible inhibition of GST activity, modulation of MRP); some with three aspects, some only with one or two. In view of the multiple roles of this system in cellular physiology, cell proliferation, gene regulation, anticarcinogenicity and multidrug resistance,α,β-unsaturated carbonyl compounds indeed seem very important, especially as man is exposed to this class of compounds in everyday life. The results open further perspectives for the development of therapeutic agents regarding multidrug resistance and anticarcinogenicity. The potential effect of these compounds on vital processes emphasise the need for future research on the total exposure of people to these compounds, especially via diet and environment.