Vol. 49 No. 3/2002 693–701 QUARTERLY

Earlier studies from our laboratories have shown that RLIP76, a previously described Ral-binding GTPase activating protein (Jullien-Flores et al., 1995, J. Biol. Chem. 270: 22473), is identical with the xenobiotic transporter DNP-SG ATPase, and can catalyze ATP-dependent transport of glutathione-conjugates as well as doxorubin (Awasthi et al., 2000, Biochemistry, 39: 9327). We have now reconstituted purified bacterially expressed RLIP76 in proteoliposomes, and have studied ATP-dependent uptake of the glutathione conjugate of 4-hydroxynonenal (GS-HNE) by these vesicles. Results of these studies show that RLIP76 reconstituted in proteoliposomes catalyzes ATP-dependent transport of GS-HNE against a concentration gradient. The transport of GS-HNE is saturable with respect to ATP as well as GS-HNE with K(m) values of 1.4mM and 2.5 microM, respectively. These studies demonstrate that RLIP76 mediates active transport of GS-HNE, and are consistent with our previous work showing that RLIP76-mediated efflux of GS-HNE regulates the intracellular concentration of 4-HNE and thereby affects 4-HNE mediated signaling.

active transport of GS-HNE, and are consistent with our previous work showing that RLIP76-mediated efflux of GS-HNE regulates the intracellular concentration of 4-HNE and thereby affects 4-HNE mediated signaling.
4-Hydroxynonenal (4-HNE), a relatively stable and toxic end product of lipid peroxidation, has been implicated in cell cycle signaling mechanisms.Cells in culture exposed to 4-HNE show activation of c-Jun N terminal kinase (JNK) and caspase 3 and eventually undergo apoptosis (Uchida et al., 1999;Cheng et al., 2001a;2001b;Yang et al., 2001).Exposure to 4-HNE has also been shown to induce erythroid differentiation in human erythroleukemia cells (Cheng et al., 1999).On the other hand, at low concentrations, 4-HNE is reported to cause cell proliferation in at least some cell types (Ruef et al., 1998;Cheng et al., 1999).These studies suggest that 4-HNE influences signaling pathways in a concentration dependent manner and that the regulation of its intracellular concentrations may be crucial for some of the signaling pathways.4-HNE is generated during lipid peroxidation, the extent of which depends on uncontrollable extracellular factors such as exposure to radiation, xenobiotics, oxidative stress, and heat.Thus, the intracellular concentrations of 4-HNE must be regulated by its metabolism and/or exclusion from the cell (Cheng et al., 2001b).
Cells metabolize 4-HNE through its conjugation to glutathione (GSH) which is catalyzed by glutathione S-transferases (GSTs) (Alin et al., 1985).4-HNE and its GSH-conjugate (GS-HNE) are also substrates for aldose reductase, which catalyzes their NADPH dependent reduction to corresponding alcohols (Srivastava et al., 1995;1998).GS-HNE is inhibitory to GSTs, and to sustain the GST-mediated cellular mechanisms of detoxification, it must be removed from cells through ATP-dependent transport (Ishikawa, 1989;Awasthi et al., 2000;Sharma et al., 2001).We have recently demonstrated that RLIP76 (Ral interacting protein or Ral-BP1), is the major transporter of GS-HNE in erythrocytes and plays a role in the regulation of intracellular concentration of 4-HNE (Awasthi et al., 2000;Cheng et al., 2001b;Sharma et al., 2001).RLIP76 is a GTPase activating protein (GAP) first described by Jullien-Flores and co-workers (Jullien-Flores et al., 1995).Our studies have shown that DNP-SG ATPase, a previously described transport protein (LaBelle et al., 1988;Awasthi et al., 1994;1998a;1998b) which catalyzed ATP-dependent primary active transport of various GSH-conjugates as well as cationic amphiphilic drugs such as doxorubicin (DOX), is identical with RLIP76 (Awasthi et al., 2000).Transfection of K562 cells with RLIP76 accelerates the efflux of GS-HNE (Awasthi et al., 2000;Cheng et al., 2001b).Likewise, induction of RLIP76 in cells subjected to oxidative stress is accompanied by increased transport of GS-HNE (Cheng et al., 2001b).RLIP76 and hGST5.8, a GST isoenzyme with substrate preference for 4-HNE, are induced in a variety of cells upon exposure to conditions causing enhanced lipid peroxidation and 4-HNE formation (Cheng et al., 2001b).Cells with induced RLIP76 show accelerated efflux of GS-HNE and acquire resistance to 4-HNE-or H 2 O 2 -induced apoptosis through this mechanism.The transport of GS-HNE and resistance to 4-HNE-mediated apoptosis can be abrogated by coating the cells with anti-RLIP76 IgG (Cheng et al., 2001b).These studies suggest that a coordinated action of GSTs and RLIP76 plays a major role in regulating the intracellular concentrations of 4-HNE.Studies with inside-out oriented vesicles (IOVs) prepared from human erythrocyte membranes also indicate that the majority of GS-HNE transport from erythrocytes is catalyzed by RLIP76, and that this transport can be inhibited by antibodies against RLIP76 (Sharma et al., 2001;Cheng et al., 2001b).It has been shown that about 70% of the transport of GS-HNE from erythrocytes is mediated by this transporter (Sharma et al., 2001).Together, these studies strongly suggest that RLIP76-mediated transport of GS-HNE is a major determinant of the intracellular concentrations of 4-HNE.
RLIP76-mediated, ATP-dependent uptake of GS-HNE has been studied primarily in the IOVs prepared from membrane vesicles (Ishikawa, 1989;Awasthi et al., 2000;Sharma et al., 2001).Since this transport is inhibited by anti-RLIP76 IgG (Cheng et al., 2001b;Sharma et al., 2001), it has been concluded that it is specifically mediated by RLIP76 but direct evidence for the ATP-dependent transport of GS-HNE by RLIP76 was lacking.To conclude unequivocally that RLIP76 indeed catalyzes transport of GS-HNE, a functional reconstitution of RLIP76 in proteoliposomes capable of catalyzing ATP-dependent transport of GS-HNE must be achieved.In the present studies, we demonstrate that proteoliposomes reconstituted with highly purified, bacterially expressed RLIP76 can catalyze ATP-dependent primary active transport of GS-HNE.In this communication we describe the kinetic characteristics of RLIP76-mediated transport of GS-HNE, and discuss the physiologic significance of RLIP76 in defense mechanisms against oxidative stress.

Bacterial expression and purification by metal affinity chromatography of RLIP76 carrying an N-terminal histidine tag.
Escherichia coli strain BL21(DE3) was transformed with a pET-30a vector containing the full length RLIP76 cDNA ligated in frame to the plasmid-supplied upstream sequence encoding a 6x histidine tag.The bacteria were grown at 37°C until A 600 reached 0.6, and were induced with 0.4 mM IPTG at 37°C overnight (Awasthi et al., 2001).The bacterial culture was centrifuged at 1400 ´g, and the pellet was lysed in 5 ml buffer containing 50 mM sodium phosphate, 300 mM NaCl, 20 mM imidazole, pH 8.0, 100 mM phenylmethylsulfonyl fluoride (PMSF), and 0.5% polidocanol by sonication on ice (3 ´15 s at 50 W).The lysate was incubated for 2 h at 4°C with gentle shaking followed by centrifugation at 13 000 r.p.m. for 20 min.The supernatant was mixed with Ni-NTA superflow resin (Qiagen) which was preequilibrated with the above lysis buffer.The resin was incubated for 2 h at 4°C with gentle shaking, and washed with buffer containing 50 mM sodium phosphate, 300 mM NaCl, 0.5% polidocanol and 10 mM imidazole, pH 8.0, until A 280 returned to zero.The protein bound to the resin was eluted with buffer containing 50 mM sodium phosphate, 300 mM NaCl, 0.025% polidocanol, 100 mM PMSF, 50 mM butylated hydroxytoluene (BHT) and 250 mM imidazole, pH 8.0.The eluted protein was concentrated using Amicon Centriprep concentrators followed by sequential dialysis against buffer containing 10 mM Tris/HCl, pH 7.4, 1.4 mM b-mercaptoethanol, 50 mM BHT, 100 mM EDTA, 100 mM PMSF and 0.025% polidocanol.

SDS/PAGE and Western blotting.
Purified RLIP76 was subjected to SDS/PAGE analysis on 12.5% polyacrylamide gels and stained with Coomassie Brilliant Blue R-250.Western blot analysis of the purified protein was performed by blotting the polyacrylamide gels to nitrocellulose using the tank blotting method of Towbin et al. (1979).The membranes were then incubated with polyclonal antibodies raised in rabbit against bacterially expressed RLIP76 (Sharma et al., 2001) followed by goat anti-rabbit secondary antibodies conjugated with horse radish peroxidase (HRP).The blots were developed using a HRP color developing reagent (Bio-Rad).The protein concentration was determined by the Bradford assay (Bradford, 1976).
Reconstitution of RLIP76 into proteoliposomes.Reconstitution of purified RLIP76 into proteoliposomes was performed by the method described by us previously (Awasthi et al., 2000).Briefly, the purified RLIP76 was dialyzed against reconstitution buffer (10 mM Tris/HCl, pH 7.4, 100 mM KCl, 1 mM EGTA, 2 mM MgCl 2 , 40 mM sucrose, 2.8 mM b-mercaptoethanol, and 0.025% (by vol.) polidocanol.A mixture of soybean lipids (asolectin; 40 mg/ml) and cholesterol (10 mg/ml) was prepared in the reconstitution buffer by sonication, and a 100 ml aliquot of this mixture was added to 0.9 ml of dialyzed purified RLIP76 containing 10 mg of protein.The mixture was sonicated for 15 s at 50 W after which 200 mg of SM-2 biobeads preequilibrated with reconstitution buffer (without polidocanol) were added to initiate vesiculation by removal of detergent.After incubation for 4 h at 4°C, SM-2 beads were removed by centrifugation at 3620 ´g.
Transport studies.ATP-dependent transport of [ 3 H]GS-HNE by RLIP76 proteoliposomes were measured by the rapid filtration technique described by Awasthi et al. (1994).Proteoliposomes were equilibrated at 37°C for 5 min in 120 ml of reconstitution buffer containing 10 mM [ 3 H]GS-HNE (100 c.p.m./pmol).The transport reaction was started by addition of ATP (final concentration 4 mM), and the reaction mixture was incubated for the indicated time at 37°C with gentle shaking.Control proteoliposomes were incubated with 4 mM 5¢AMP in place of ATP.The reactions were stopped by placing the tubes on ice.Aliquots (30 ml) of the reaction mixture were filtered through nitrocellulose membrane (0.45 mm) in a 96 well filtration plate under uniform suction.The plate was then air-dried, and the filters were cut by Millipore punch hole assembly.The filters were individually placed in glass scintillation vials containing 10 ml of scintillation fluid, and radioactivity was determined in a scintillation counter (Beckman LS6800).Measurements were performed in triplicate, and the mean ±S.D. is reported.Net ATP-dependent transport of GS-HNE was calculated by subtracting vesicle-associated radioactivity in the presence of AMP from that determined in the presence of ATP.The dependence of transport on RLIP76 protein was demonstrated by comparing protein-containing with protein-free liposomes with or without ATP.Kinetic constants for the ATP-dependent uptake of GS-HNE were determined by varying the concentration of [ 3 H]GS-HNE from 0.5 mM to 20 mM, and that of ATP from 0.5 to 8 mM.

Purification of RLIP76
Bacterially expressed RLIP76 was purified from transformed E. coli by chromatography on Ni-affinity columns as described by us previously (Awasthi et al., 2001).Purified preparations showed several bands on SDS polyacrylamide gel electrophoresis (Fig. 1A), and these bands were recognized by antibodies against RL1P76 (Fig. 1B).We and others have previously shown that even though the molecular mass of RLIP76 calculated from its amino-acid composition is 76 kDa, the protein shows several bands at higher as well lower molecular mass in SDS gels (Jullien-Flores et al., 1995;Awasthi et al., 2000;Cheng et al., 2001b;Sharma et al., 2001;Awasthi et al., 2001;Sighal et al., 2001).The N-terminal sequences of all these bands represent internal sequences of RLIP76, indicating that the peptides arose from proteolytic processing of RLIP76.The amino acid composition of purified RLIP76 also indicates that the preparations are more than 96% pure (Awasthi et al., 2000), and thus suitable for transport studies.

ATP-dependent uptake of GS-HNE by proteoliposomes reconstituted with RLIP76
Purified RLIP76 was reconstituted into proteoliposomes as described previously (Awasthi et al., 2000;2001) and detailed in the Materials and Methods Section.Electron microscopic evaluation indicated that the proteoliposomes prepared by this method were closed and mostly unilamellar.We have previously determined by inulin entrapment experiments that the internal volume of vesicles prepared under these conditions as is approximately 19 ml/ml of proteoliposomes (Awasthi et al., 1998b).In preliminary experiment it was shown that the ATP-dependent uptake of [ 3 H]GS-HNE by proteoliposomes was linear for up to 10 min.Thus, for transport experiments the reaction mixtures were incubated for 10 min.
As shown in Fig. 2, a significant amount of [ 3 H]GS-HNE was associated with liposomes reconstituted without RLIP76.However, this association was not dependent on the presence of ATP (Fig. 2) or the presence of equiosmolar NaCl, ADP, or Met-ATP, the non-hydrolyzable analog of ATP (data not presented).These results are consistent with our previous studies on transport of DOX and DNP-SG which show a significant amount of RLIP76 was expressed in E. coli BL21(DE3) transfected with full length RLIP76 cDNA, and purified by Ni-NTA metal affinity chromatography as described in the Materials and Methods.Panel A: Coomassie Brilliant Blue stain of SDS/PAGE, lane1, protein molecular mass markers; lane 2, 28 000 ´g supernatant (5 ml) of bacterial extract prior to addition of 0.4 mM IPTG, lane 3, 28 000 ´g supernatant (5 ml) of bacterial extract, 12 h after addition of 0.4 mM IPTG; lane 4, 10 mg of purified RLIP76.Panel B: Western blot probed with anti-RLIP76 IgG and visualized by HRP color developing reagent (Bio-Rad), lane 1, 28 000 ´g supernatant (5 ml) of bacterial lysate (induced); lane 2, purified RLIP76 (10 mg).
non-specific association of DOX and DNP-SG with liposomes (Awasthi et al., 2000), perhaps due to the amphiphilic nature of these compounds.
The association of [ 3 H]GS-HNE with RLIP-76-containing proteoliposomes in the absence of ATP was similar to that observed with protein-free liposomes, with or without ATP.However, in the presence of ATP, a significant increase was observed in the uptake of [ 3 H]GS-HNE by RLIP76 proteoliposomes (Fig. 2).Such an increase was not observed in the presence of Met-ATP or equiosmolar NaCl or ADP used as controls (not shown).These results indicated that [ 3 H]GS-HNE uptake was ATP-dependent, and that the hydrolysis of ATP was required for the uptake.The uptake was temperature-dependent, and maximal uptake was observed at 37°C (data not presented).Based on the specific radioactivity of [ 3 H]GS-HNE (100 c.p.m./pmol), the internal vesicle volume (19 ml/ml of reaction mixture), and the assumption that [ 3 H]GS-HNE is in the lumen of the vesicles, the intravesicular concentration of GS-HNE after 10 min incubation in absence of ATP was estimated to be about 13 mM which was similar to the extravesicular concentration of GS-HNE in the reaction mixture.In the presence of ATP, the estimated concentration of GS-HNE in RLIP76 reconstituted proteoliposomes was found to be about 20 mM, indicating that the transport of GS-HNE into the vesicles was against a concentration gradient.
Uptake of [ 3 H]GS-HNE by RLIP76 proteoliposomes was found to be saturable with respect to GS-HNE and ATP (Fig. 3A and 3B, respectively).The K m of RLIP76-mediated transport for ATP was found to be 1.4 mM, while the K m for GS-HNE was 2.5 mM, values similar to those observed for GS-HNE transport in crude membrane vesicles prepared from human erythrocytes.These results also demonstrate that RLIP76 has a significantly higher affinity for GS-HNE as compared to that of the non-physiological glutathione conjugate DNP-SG (K m of approx.60 mM) (Sharma et al., 2001).We have previously demonstrated that the efflux of GS-HNE from K562 cells can be blocked by anti-RLIP76 IgG (Cheng et al., 2001b;Awasthi et al., 2002).Likewise, the ATP-dependent uptake of GS-HNE by erythrocyte membrane IOVs can be blocked by anti-RLIP76 IgG (Sharma et al., 2001).Even though these studies strongly suggested that the transport of GS-HNE was mediated by RLIP76, direct evidence for its involvement in GS-HNE transport was lacking.The present studies clearly demonstrate that RLIP76 can be functionally reconstituted in proteoliposomes, and that it is capable of catalyzing ATP-dependent, saturable transport of GS-HNE against a concentration gradient.ATP-dependent transport of 4-HNE conjugate by membrane vesicles from cells overexpressing MRP1 has been previously reported (Renes et al., 2000).Results of our studies show that the K m of RLIP76 for GS-HNE is in a similar range to that of MRP1 (Renes et al., 2000).The V max of RLIP76-mediated transport of GS-HNE found in present studies is about 1.4 ± 0.1 nmol/min per mg protein.Considering that a substantial expression of RLIP76 is observed in all human tissues and cell lines examined so far, the present studies suggest that RLIP76 may be a major contributor to transport of GS-HNE in human tissues.Particularly in tissues such as erythrocytes where, in our hands, the expression of MRP1 is undetectable, RLIP76 contributes to a major portion of the transport of GS-HNE or other physiologic GSH-conjugates such as leukotriene C4 (LTC4) (Sharma et al., 2001).It is possible that RLIP76-mediated transport of GS-HNE and LTC4 is predominant in normal tissues, and MRP1-mediated transport may take over in cells overexpressing MRP1.This postulate must, however, be substantiated by further experimental evidence.
In summary, the present studies clearly demonstrate that RLIP76 catalyzes ATP-dependent transport of GS-HNE.These results are consistent with our previous studies suggesting that RLIP76-mediated transport of GS-HNE is an important component of defense mechanisms against oxidative stress and lipid peroxidation, and that RLIP76 can modulate cellular signaling mechanisms by regulating intracellular concentrations of 4-HNE (Cheng et al., 2001b).The latter compound has been implicated (Ruef et al., 1998;Cheng et al., 1999;Uchida et al., 1999;Cheng et al., 2001a;Yang et al., 2001;Cheng et al., 2001b ) in signaling for apoptosis, differentiation, and the cell cycle.Proteoliposomes with 10 mg of RLIP76 protein were diluted 6-fold in transport buffer containing 0.5-20 mM [ 3 H]GS-HNE (Panel A).After equilibration at 37°C for 5 min, transport was initiated by the adding 4 mM of either AMP or ATP.Aliquots of the reaction mixture (30 ml) were filtered in triplicate after 10 min.Radioactivity remaining on the nitrocellulose filter of each well in 96 well plate was quantified.ATP-dependent uptake of GS-HNE was calculated by subtracting uptake in presence of AMP from that observed in presence of ATP.Points represent averages of triplicate determinations.Panel B, the uptake of GS-HNE by proteoliposomes measured at varying concentrations (0.5-8 mM) of ATP at invariant 10 mM of [ 3 H]GS-HNE.

Figure 2 .
Figure 2. Uptake of GS-HNE in the presence of 5¢AMP and ATP by liposomes and proteoliposomes.Uptake by liposomes and proteoliposomes of [3 H]GS-HNE (10 mM, specific radioactivity 100 c.p.m./pmol) was compared in presence of AMP and ATP.Proteoliposomes used for these studies were reconstituted with 10 mg of purified RLIP76 protein and 5 mg of lipid/ml reconstitution buffer.Protein was omitted for reconstitution of liposomes.Liposomes and proteoliposomes were diluted 6-fold in transport buffer containing 10 mM [ 3 H]GS-HNE.After equilibrating at 37°C for 5 min, transport was initiated by addition of either 5¢AMP (4 mM) or ATP (4 mM).Aliquots of reaction mixture (30 ml) were filtered in triplicate after 10 min of incubation, and radioactivity remaining on the nitrocellulose filter of each well in the 96 well plate was quantified.Legends for different bars are shown on the figure.*Indicates a statistically significant difference between uptake of GS-HNE by proteoliposomes in presence of ATP as compared to AMP.

Figure 3 .
Figure 3. Double reciprocal plots of the kinetics of GS-HNE uptake by proteoliposomes with respect to varying GS-HNE and ATP concentrations.