Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

The conjugation of electrophilic substrates to glutathione (GSH) and subsequent disposition of the GSH S-conjugates formed are mediated by a large number of different enzymes and transport systems present in various tissues (Commandeur et al., 1995). One of the pathways involved in the catabolism of GSH-conjugates is the -elimination reaction of the corresponding cysteine S-conjugates by cysteine conjugate -lyases (-lyases), resulting in the formation of ammonia, pyruvic acid, and thiol compounds. In the case of glutathione S-conjugates of halogenated alkenes, the thiols formed by -lyase may rearrange rapidly to chemically highly reactive intermediates, such as thionoacyl halides, thiiranes, and/or thioketenes (Dekant et al., 1988; Commandeur et al., 1996). The combination of active uptake mechanisms and a relatively high activity of -lyase in the kidney may explain the relatively selective nephrotoxicity of many halogenated alkenes in rodents. A number of studies by the group of Elfarra have demonstrated that the biochemical basis of this kidney selectivity may also be applied to target pharmacologically active thiol-containing antitumor agents, such as mercaptopurine and thioguanine, to the kidney for the treatment of renal cell carcinoma (Hwang and Elfarra, 1989, 1991; Elfarra and Hwang, 1993; Elfarra et al., 1995). More recently, the structurally strongly related selenocysteine Se-conjugates were proposed as alternative prodrugs to target pharmacologically selenol compounds to the kidney by local -lyase enzyme systems (Andreadou et al., 1996).

Although the above-mentioned prodrugs appear to be activated by renal subcellular fractions, the enzymes that are actually involved in these -elimination reactions have not yet been identified. To elucidate the possibilities and limitations of this prodrug concept, however, the identity and tissue distribution of the enzymes involved in the activation of cysteine S-conjugates and selenocysteine Se-conjugates remain to be characterized.

Three major cysteine conjugate -lyase enzymes have been identified in rat kidney cytosol (Cooper, 1998). All are pyridoxal 5'-phosphate-dependent enzymes. One cytosolic -lyase in the kidney appeared to be identical with glutamine transaminase K (GTK), based on its composition and enzyme kinetic properties (Stevens et al., 1986). Originally denoted as GTK, this enzyme was isolated more than 25 years ago (Cooper and Meister, 1974). -Lyase/GTK is a dimeric protein consisting of two identical subunits of M_r 47,470. Recent cloning studies revealed that -lyase/GTK is also identical with kynurenine aminotransferase (Perry et al., 1993; Mosca et al., 1994). More recently, a closely related cysteine conjugate -lyase, with subunits of a slightly higher³ molecular weight, 48,500, was characterized with considerable overlap in substrate selectivity (Abraham and Cooper, 1996). Finally, a -lyase enzyme with high molecular weight (330,000) was demonstrated to be present in both cytosolic and mitochondrial fractions. This third -lyase has different enzyme characteristics compared with -lyase/GTK as demonstrated by its ability to convert leukotriene E₄ and 5'-S-cysteinyldopamine and by its lower specific activity toward cysteine conjugates of halogenated alkenes (Abraham et al., 1995).

Recently, we demonstrated that replacing the sulfur of cysteine S-conjugates by a selenium atom resulted in a dramatic increase in -elimination activity in rat renal cytosol (Andreadou et al., 1996). Therefore, selenocysteine Se-conjugates were proposed as alternative prodrugs to target pharmacologically active selenol compounds to the kidney (Fig. 1). The aim of the present investigation was to study whether selenocysteine Se-conjugates are substrates for purified rat renal -lyase/GTK. Next to the selenocysteine Se-conjugates, a number of corresponding cysteine S-conjugates were tested as substrates for purified rat renal -lyase/GTK as well. Because the enzyme kinetics obtained with the purified enzyme showed a relatively poor correlation with results previously obtained with rat renal cytosol, the possible involvement of multiple enzymes was studied by determining the activity profiles after fractionation of rat renal cytosol by two different chromatographic methods: anion exchange chromatography and gel permeation chromatography.

View larger version (7K): [in this window] [in a new window]   Fig. 1.   Concept of activation of cysteine S-conjugates and selenocysteine Se-conjugates to pharmacologically active thiols and selenols, respectively.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. -Keto--methiolbutyric acid (KMB), -chloro-L-alanine, S-methyl-L-cysteine, S-ethyl-L-cysteine, S-benzyl-L-cysteine, and a diagnostic kit for aspartate aminotransferase (DG158-K) were purchased from Sigma Chemical Co. (St. Louis, MO). Amino-oxyacetic acid and phenylmethylsulfonyl fluoride (PMSF) were obtained from Aldrich Chemie (Brussels, Belgium). 4-Methoxythiophenol was obtained from Fluka (Buchs, Switzerland). S-(4-Methylbenzyl)-L-cysteine was purchased from Advanced Chemtech (Louisville, KY). S-(4-Methoxybenzyl)-L-cysteine was purchased from Bachem Feinchemikalien AG (Bubendorf, Switzerland).

Synthesis of Se-Allyl-L-selenocysteine. L-Selenocystine (1.5 mmol, 500 mg) was dissolved in 8 ml of 0.5 N NaOH and 2 ml of ethanol. At 0°C, 0.4 g (15 mmol) of NaBH₄ was added while the reaction mixture was stirred. The mixture was allowed to reach room temperature, during which the color of the solution changed from yellow to colorless, After cooling again to 0°C, 4 ml of 2 N NaOH and 6 mmol of allylbromide were added, and the mixture was stirred for 3 h at room temperature. Concentrated HCl was added until pH 5 to 6 and cooled at 4°C. Se-Allyl-L-cysteine precipitated as a yellowish solid. The ¹H NMR spectrum obtained was almost identical with that of S-allyl-L-cysteine (17): ¹H NMR (D₂O, Na₂CO₃):  (ppm) 2.95 to 3.22 (2H, m, CH₂-CH-NH₂), 3.30 (2H, d, CH₂==CH-CH₂), 4.25 to 4.40 (1H, d of d, CH₂-CH-NH₂), 5.05 to 5.25 (2H, m, CH₂==CH-CH₂), 5.80 to 6.08 (1H, m, CH₂==CH-CH₂).

Purification of Rat Renal Cysteine Conjugate -Lyase/GTK. -Lyase/GTK was purified from rat renal cytosol obtained from male Wistar rats (160-250 g) supplied by Harlan (Zeist, the Netherlands). The procedure used is described in detail by Yamauchi et al. (1993) and yielded a highly purified enzyme that was 1000-fold enriched according to its increased specific activity with S-(1,2-dichlorovinyl)-L-cysteine (1,2-DCV-Cys) as a substrate. Matrix-assisted laser desorption time-of-flight (MALDI-TOF) mass spectrometry analysis of this protein revealed a single MH⁺ mass of M_r 47,700 ± 130,⁴ which is consistent with the mass as predicted from the sequence of -lyase/GTK as determined by Perry et al. (1993).

Enzymatic Incubations. Selenocysteine Se-conjugates and cysteine S-conjugates were incubated with the purified -lyase in 50 mM Tris-HCl buffer, pH 8.6, and at a temperature of 37°C. Unless stated otherwise, the final concentration of enzyme was 1 µg/ml. The cofactor KMB was added at a final concentration of 0.2 mM, which allows assessment of transamination reaction by measuring KMB consumption (see later). After 2 min of preincubation at 37°C, incubations were started by the addition of a 10-fold concentrated enzyme solution. -Elimination reactions were assessed by determining the formation of pyruvic acid. To correct for background production of pyruvic acid, parallel incubations were always performed in absence of enzyme. All incubations were always performed in duplicate.

Time Course of Enzyme Reactions. Before determining enzyme kinetics, the time course of product formation was determined to assess the linearity of the -elimination reactions. To this end, incubations were performed at an incubation volume of 2 ml. After starting the reaction by the addition of 200 µl of 10 µg/ml enzyme solution, samples of 100 µl were taken from the incubation at several time points and mixed with 500 µl of OPD solution (12 mM OPD in 3 M HCl) in 1-ml Eppendorf cups. The Eppendorf cups were closed and heated for 60 min at 60°C to complete the derivatization reaction (Stijntjes et al., 1992). To the resulting solutions, 600 µl of HPLC eluent (45% methanol, 54% water, 1% acetic acid) was added and transferred to HPLC vials for automated HPLC analysis (see later).

Enzyme Kinetic Parameters. Enzyme kinetic parameters (K_m and k_cat) of conjugates were determined by incubating substrates at six to eight concentrations ranging from 0.05 to 4 mM. Due to their poor solubility, the conjugates with phenyl and benzyl substituents were incubated at concentrations up to 2 mM. Corrections for nonenzymatic degradation were made by performing parallel incubations in absence of enzyme. All incubations were performed at 37°C and at a total volume of 300 µl and were started by adding 30 µl of 10 µg/ml purified -lyase/GTK. The final concentration of KMB in all incubations was 0.2 mM. Reactions were stopped after 5 min of incubation by the addition of 1 ml of 12 mM OPD in 3 M HCl. The resulting mixtures were subsequently heated for 60 min at 60°C. After this derivatization reaction, a volume of 600 µl was mixed with 600 µl of HPLC eluent (45% methanol, 54% water, 1% acetic acid) and transferred to HPLC vials for automated HPLC analysis (see later).

HPLC Analysis of Pyruvic Acid and KMB. HPLC vials were placed in a Waters 707 Autoinjector cooled at 4°C. Then, 100-µl samples were injected at intervals of 20 min. The analytes were chromatographed on two ChromSpher C18 columns (5-µm particles, 100 × 4.6 mm; Chrompack, Bergen op Zoom, The Netherlands), which was eluted isocratically with the above-mentioned HPLC eluent at a flow rate of 0.4 ml/min. Detection of derivatized pyruvic acid and KMB was accomplished with a Jasco fluorescence detector (model 821-FP) set at an excitation wavelength of 336 nm and an emission wavelength of 420 nm. Chromatograms were analyzed using the Class VP 4.1 software package of Shimadzu (Columbia, MD). Under these conditions, retention times of derivatized pyruvic acid and KMB were 5.8 and 14.9 min, respectively.

Fractionation of Rat Kidney Cytosol by Anion Exchange Fast Protein Liquid Chromatography (FPLC). Rat kidney cytosol was fractionated by FPLC using a Mono-Q anion exchange column (Pharmacia Biotech, Uppsala, Sweden). Before application to the column, rat kidney cytosol was dialysed for 20 h against 20 mM triethanol-HCl buffer, pH 7.45, containing 0.1 mM EDTA and 40 µM PMSF (buffer A). After dialysis, this enzyme fraction was filtered using a 0.2-µm Schlauer filter, and 0.5 ml (containing 19 mg of cytosolic protein) was applied to the Mono-Q column, which was equilibrated with the same buffer as used for dialysis.

Fractionation of Rat Kidney Cytosol by Gel Permeation Chromatography. Rat renal cytosol (5 ml) was applied to a HiLoad 16/60 Superdex 200 column (Pharmacia Biotech) and eluted at a flow rate of 0.5 ml/min with 50 mM sodium phosphate buffer, pH 7.4, containing 150 mM sodium chloride and 40 µM PMSF. After 65 min, representing the time necessary to elute the dead volume, fractions of 0.5 ml were collected and placed on ice. Elution of proteins was monitored continuously by UV detection at 280 nm. Activities were determined using Se-(4-methylbenzyl)-L-selenocysteine and CTFE-Cys as substrates. Then, 15 µl of the fractions was added to 135 µl of 50 mM Tris-HCl buffer, pH 8.6, containing substrate and 0.5 mM KMB. Incubations were performed for 20 min at 37°C, after which the reaction was terminated by adding 500 µl of 12 mM OPD in 3 M HCl. After 60 min of derivatization at 60°C, the samples were analyzed by HPLC, as described earlier.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Time Course of Biotransformation of Cysteine S-Conjugates and Selenocysteine Conjugates. A significant time-dependent formation of pyruvic acid was observed on incubation of purified -lyase/GTK with cysteine S-conjugates of halogenated alkenes and with various selenocysteine Se-conjugates (Fig. 2). Because pyruvate formation significantly deviated from linearity after approximately 5 min, the enzyme kinetic parameters K_m and k_cat for the conjugates were based on the specific activities obtained using incubation times of 5 min.

View larger version (31K): [in this window] [in a new window]   Fig. 2.   Time course of pyruvic acid formation and KMB consumption in incubation of purified cysteine conjugate -lyase/GTK (1 µg/ml) with 1 mM cysteine S-conjugates (top) or 1 mM selenocysteine Se-conjugates (middle and bottom). , pyruvic acid; , KMB. Error bars represent difference between two data points.

Substrate Selectivity and Kinetics of Purified -Lyase/GTK. The specific activities and enzyme kinetic parameters of purified -lyase/GTK toward 11 cysteine S-conjugates and 15 selenocysteine Se-conjugates as substrates are shown in Table 1. From these results, it appears that most cysteine S-conjugates tested showed only a very low -elimination activity. Significant consumption of KMB was observed, indicative for a preference for the transamination reaction. The only L-cysteine S-conjugates showing a sufficiently high -elimination activity enabling assessment of enzyme kinetic parameters were the four cysteine S-conjugates carrying halogenated alkyl and alkenyl substituents. A decrease in specific activity and k_cat/K_m was observed in the order TFE-Cys > CTFE-Cys > DCDFE-Cys  1,2-DCV-Cys. These nephrotoxic cysteine S-conjugates also appear to be transaminated to a significant extent as indicated by a significant consumption of KMB (Table 1). The relative importance of transamination appears to slightly increase in the order TFE-Cys < CTFE-Cys < DCDFE-Cys  1,2-DCV-Cys.

Fractionation of Rat Kidney Cytosol by Anion Exchange Chromatography. Rat renal cytosol was fractionated using Mono-Q anion exchange FPLC, and the 2-ml fractions collected were subsequently screened for -lyase activity using five different substrates. Using CTFE-Cys as a substrate, the highest activity was found in fraction 10, whereas lower activities were found in fractions 11, 12, and 13 (Fig. 3A). In incubations of fractions with CTFE-Cys as a substrate, only a relatively small consumption of KMB was observed (<20% of the initial 0.2 mM concentration). When using Se-phenyl-L-selenocysteine and Se-(4-methylbenzyl)-L-selenocysteine as substrate, however, a significantly different activity profile was observed, which rules out involvement of only a single enzyme in the -elimination reactions. First, in contrast to CTFE-Cys, maximal activity was observed in fraction 11 (Fig. 3, B and C). Second, using these selenocysteine conjugates, very significant pyruvic acid formation was also observed in fractions 14 to 18. Furthermore, with both substrates, a much stronger, up to 60%, consumption of KMB was observed, which again was significant from fractions 10 to 18. To test whether KMB might have become limiting in these experiments, the activity of these fractions were also determined in presence of 0.5 mM KMB. Interestingly, at this higher KMB concentration, the highest pyruvate formation was observed in fraction 10, as is the case with CTFE-Cys as substrate (data not shown).

View larger version (36K): [in this window] [in a new window]   Fig. 3.   Activity profiles of -elimination reactions (indicated by pyruvate formation) and transamination reactions (as indicated by KMB consumption) with CTFE-Cys and three selenocysteine Se-conjugates in FPLC fractions from rat renal cytosol applied to a Mono-Q anion exchange column. Activities were expressed as percentage of the highest peak area of pyruvate and KMB, respectively.

Fractionation of Rat Kidney Cytosol by Gel Filtration Chromatography. Rat renal cytosol was also fractionated by high-resolution gel filtration chromatography using a HiLoad 16/60 Superdex 200 column. By using Se-(4-methylbenzyl)-L-selenocysteine and CTFE-Cys as substrates, activity profiles were determined. As shown in Fig. 4, at least two broad peaks with -elimination activity were shown. For both substrates, maximal activity was present in fraction 20, which eluted 150 min after the application of renal cytosol to the column. According to the calibration by the marker protein mixture, the mass of this protein will be around M_r 300,000. The second peak had its highest activity in fraction 37, which eluted after 167 min. According to the calibration by the marker protein mixture, the mass of this protein will between M_r 66,000 and 132,000, which may be consistent with the 90-kDa -lyase protein or proteins.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Activity profiles of -elimination reactions (indicated by pyruvate formation) with CTFE-Cys () and Se-(4-methylbenzyl)-L-selenocysteine () in FPLC fractions from rat renal cytosol applied to a HiLoad 16/60 Superdex 200 gel filtration column.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Local activation of cysteine S-conjugates by renal -eliminating enzymes has been proposed as a novel approach to target antitumor compounds 6-mercaptopurine and 6-thioguanine to the kidney (Hwang and Elfarra, 1989, 1991). More recently, selenocysteine Se-conjugates were proposed as alternative kidney-selective prodrugs showing much higher -elimination activities in rat renal cytosol than their corresponding sulfur analogs (Andreadou et al., 1996). However, as yet little is known regarding which proteins are actually involved in the -elimination reactions of these S- and Se-conjugates because activities have been measured only in crude enzyme fractions such as kidney homogenates, mitochondria, and cytosols.

One of the enzymes capable of catalyzing transamination and -elimination reactions is -lyase/GTK. It has been suggested that large noncharged amino acids are transaminated by -lyase/GTK (Cooper and Meister, 1974). Using purified -lyase/GTK from rat kidney, next to transamination, a high -elimination activity has been demonstrated with 1,2-DCV-Cys and TFE-Cys as substrates. No -elimination activity was observed using S-(benzothiazolyl)-L-cysteine (BTC) (Yamauchi et al., 1993), 5'-S-cysteinyldopamine and leukotriene E₄ as substrates (Abraham et al., 1995). In the present study, nine additional cysteine S-conjugates were evaluated as substrates for purified renal -lyase/GTK. As shown in Table 1, high -elimination activities were also observed with CTFE-Cys and DCDFE-Cys as substrate. In accordance with the observations by Stevens et al. (1986) using 1,2-DCV-Cys, significant transaminase activity was also observed as implicated by the relatively high rates of KMB consumption that were observed.

The present study confirms that cysteine S-conjugates carrying nonhalogenated substituents are poor substrates in comparison with their selenium analogs (Table 1). Enzymatic pyruvic acid production was only 25 to 50% of the low nonenzymatic degradation of the conjugates (data not shown). Only at a higher enzyme concentration and with a longer incubation time was significant consumption of KMB observed for these S-conjugates, indicative of a preference for transamination reactions. Although the present nonhalogenated cysteine S-conjugates have not yet been tested with the purified rat -lyase/GTK, some of them have been tested previously as substrates for purified -lyases from bovine and turkey kidney (Bhattacharya and Schultze, 1967). Consistent with the present study, no -elimination was previously observed with S-methyl-L-cysteine, S-ethyl-L-cysteine, S-propyl-L-cysteine, S-benzyl-L-cysteine, and S-allyl-L-cysteine. Lash et al. (1990) reported that BTC and S-(benzothiazolyl)-L-homocysteine were actively -eliminated by a purified -lyase from human kidney. However, Yamauchi et al. (1993) did not find any activity of purified rat renal -lyase/GTK toward BTC, suggesting a significant species difference in substrate selectivity between the human and rat enzymes. This species difference is further supported by the overall 18% dissimilarity between the amino acid sequences of the rat and human enzyme (Perry et al., 1993).

Selenocysteine Se-conjugates previously were shown to be -eliminated at a high activity by renal cytosol (Andreadou et al., 1996). The results of the present study suggest that -lyase/GTK plays a major role in this reaction, because most Se-conjugates were -eliminated at a very high activity. Several selenocysteine Se-conjugates displayed -elimination activities as high as that observed with TFE-Cys, which was the best substrate known. As was observed previously in rat renal cytosol, the activity of -elimination of Se-conjugates was much higher than -elimination of corresponding S-conjugates (Table 1). Possible explanations for the higher -elimination reactions of selenocysteine Se-conjugates may be the weaker bond strength of the C-Se-bond (234 kJ/mol) versus C-S-bonds (272 kJ/mol) (Guziec, 1987) and/or a more facilitated -proton abstraction of the selenocysteine moiety (Miles, 1986). As indicated by the significant KMB consumption, transamination reaction is a prominent pathway of biotransformation of both selenocysteine Se-conjugates and cysteine S-conjugates. When comparing specific activities of transamination, selenocysteine Se-conjugates as substrates showed 5- to 10-fold higher activities compared with their sulfur analogs (Table 1). Because transamination reactions do not involve C-Se-scission, facilitation of -proton abstraction by electronic effects of the Se atom may be the most likely explanation.

Because only very few substrates have been identified as yet, little was known regarding the structure-activity relationship of -lyase/GTK-catalyzed -elimination reactions. Because of their surprisingly high activities, the class of selenocysteine Se-conjugates may be interesting probe substrates to characterize the substrate-binding site. For the alkyl-substituted Se-conjugates, an increase in the alkyl-chain appears to increase the affinity for the enzyme, as indicated by the decrease in their K_m value (Table 1). The benzyl-substituted Se-conjugates appeared to be extremely good substrates as well, especially when substituted at the para-position of the benzyl group. No clear electronic effects were observed because both electron-withdrawing and electron-donating para-substituents appeared to increase -elimination activity compared with Se-benzyl-L-selenocysteine. Therefore, apparently steric properties or lipophilicity plays a more important role. Surprisingly, the introduction of para-substituents in Se-phenyl-L-selenocysteine led to a very strong decrease in enzyme activities (Table 1).

When comparing enzyme kinetic parameters of the purified -lyase/GTK with those obtained previously with rat renal cytosol, a very poor correlation is found. The most striking difference is found with the para-substituted Se-phenyl-L-selenocysteine compounds, which showed high activity with rat renal cytosol (Andreadou et al., 1996) but demonstrated only a minor activity with purified -lyase/GTK. By determining the activity profile in fractionated rat renal cytosol, the present study reveals that the poor correlation between cytosolic and purified enzyme activities may be explained by the involvement of multiple enzymes in the cytosolic fractions (Figs. 3 and 4). Aspartate aminotransferase, which was proposed by Kato et al. (1996) as an alternative rat renal -lyase enzyme, does not appear to be involved in -elimination of the conjugates tested. According to the results of the gel filtration chromatography, an important role may be played by the high-molecular-weight -lyase previously characterized by Abraham et al. (1995).

Recently, it was shown that selenium-enriched garlic was more effective in cancer prevention than normally grown garlic (Lu et al., 1996). Among the selenium species identified in selenium-enriched garlic were two of the selenocysteine Se-conjugates, Se-methyl-L-selenocysteine and Se-allyl-L-selenocysteine, tested in this study. Dietary administration of these selenocysteine Se-conjugates to rats indeed provided strong protection against methylnitrosourea-induced carcinogenesis in rats (Ip et al., 1999). Chemopreventive activities of these selenocysteine Se-conjugates have been attributed to methyl selenol (Ip and Ganther, 1992) and diallyl selenide (Lu et al., 1996), respectively, both of which are formed via -elimination reactions. The relatively high activity of purified -lyase/GTK toward these selenocysteine Se-conjugates (Table 1) may implicate a role of -lyase/GTK in the chemopreventive activity of selenium-enriched garlic. Diallyl selenide was reported to have a 100-fold higher chemopreventive activity than diallyl sulfide, which is an important chemopreventive agent in normally grown garlic (el-Bayoumy et al., 1996). Therefore, the combination of higher bioactivation activity with formation of a 100-fold more potent product may explain the higher chemopreventive activity of Se-allyl-L-selenocysteine compared with S-allyl-L-cysteine.

In conclusion, the results presented here indicate that rat renal -lyase/GTK plays an important role in the -elimination of selenocysteine Se-conjugates by rat renal cytosol. The high -elimination activity in combination with the potent antitumor activities of the formed selenol compounds makes selenocysteine Se-conjugates promising prodrugs to treat renal cell carcinoma. Fractionation of renal cytosol by two different types of chromatography, however, revealed that next to -lyase/GTK, additional enzymes are active in the -elimination of selenocysteine Se-conjugates. The identity and tissue distribution of these additional enzymes remain to be established to predict the kidney selectivity of selenocysteine Se-conjugates as prodrugs.