ABC transporters Cdr1p, Cdr2p and Cdr3p of a human pathogen Candida albicans are general phospholipid translocators.

We have used fluorescent 7-nitrobenz-2-oxa-1,3-diazol-4-yl (NBD)-tagged phospholipid analogues, NBD-PE (phosphatidylethanolamine), NBD-PC (phosphatidylcholine) and NBD-PS (phosphatidylserine), to demonstrate that Cdr1p and its other homologues, Cdr2p and Cdr3p, belonging to the ATP-binding cassette (ABC) superfamily behave as general phospholipid translocators. Interestingly, CDR1 and CDR2, whose overexpression leads to azole resistance in C. albicans, elicit in-to-out transbilayer phospholipid movement, while CDR3, which is not involved in drug resistance, carries out-to-in translocation of phospholipids between the two monolayers of plasma membrane. Cdr1p, Cdr2p and Cdr3p could be further distinguished on the basis of their sensitivities to different inhibitors. For example, the in-to-out activity associated with Cdr1p and Cdr2p is energy-dependent and sensitive to sulphydryl blocking agents such as N-ethylmaleimide (NEM) and cytoskeleton disrupting agent cytochalasin E, while Cdr3p-associated out-to-in activity is energy-dependent but insensitive to NEM and cytochalasin E. We found that certain drugs, such as fluconazole, cycloheximide and miconazole, to which Cdr1p confers resistance could also affect in-to-out transbilayer movement of NBD-PE, while the same drugs had no effect on Cdr3p-mediated out-to-in translocation of NBD-PE. The ineffectiveness of these drugs to affect Cdr3p mediated out-to-in phospholipid translocation further confirms the inherent difference in the directionality of phospholipid translocation between these pumps. Notwithstanding the role of some of the Cdrps in drug resistance, this study clearly demonstrates that these ABC transporters of C. albicans are phospholipid translocators and this function could represent one of the physiological functions of such large family of proteins.

Structural and kinetic modifications of aldose reductase by S-nitrosothiols.

Modification of aldose reductase (AR) by the nitrosothiols S-nitroso-N-acetyl penicillamine (SNAP) and N-(beta-glucopyranosyl)-N(2)-acetyl-S-nitrosopenicillamide (glyco-SNAP) resulted in a 3-7-fold increase in its k(cat) and a 25-40-fold increase in its K(m) for glyceraldehyde. In comparison with the native protein, the modified enzyme was less sensitive to inhibition by sorbinil and was not activated by SO(2-)(4) anions. The active-site residue, Cys-298, was identified as the main site of modification, because the site-directed mutant in which Cys-298 was replaced by serine was insensitive to glyco-SNAP. The extent of modification was not affected by P(i) or O(2), indicating that it was not due to spontaneous release of nitric oxide (NO) by the nitrosothiols. Electrospray ionization MS revealed that the modification reaction proceeds via the formation of an N-hydroxysulphenamide-like adduct between glyco-SNAP and AR. In time, the adduct dissociates into either nitrosated AR (AR-NO) or a mixed disulphide between AR and glyco-N-acetylpenicillamine (AR-S-S-X). Removal of the mixed-disulphide form of the protein by lectin-column chromatography enriched the preparation in the high-K(m)-high-k(cat) form of the enzyme, suggesting that the kinetic changes are due to the formation of AR-NO, and that the AR-S-S-X form of the enzyme is catalytically inactive. Modification of AR by the non-thiol NO donor diethylamine NONOate (DEANO) increased enzyme activity and resulted in the formation of AR-NO. However, no adducts between AR and DEANO were formed. These results show that nitrosothiols cause multiple structural and functional changes in AR. Our observations also suggest the general possibility that transnitrosation reactions can generate both nitrosated and thiolated products, leading to non-unique changes in protein structure and function.

Characterization of the glutathione binding site of aldose reductase.

Despite extensive investigations, the physiological role of the polyol pathway enzyme-aldose reductase (AR) remains obscure. While the enzyme reduces glucose in vivo and in vitro, kinetic and structural studies indicate inefficient carbohydrate binding to the active site of the enzyme. The active site is lined by hydrophobic residues and appears more compatible with the binding of medium- to long-chain aliphatic aldehydes or hydrophobic aromatic aldehydes. In addition, our recent studies show that glutathione (GS) conjugates are also reduced efficiently by the enzyme. For instance, the GS conjugate of acrolein is reduced with a catalytic efficiency 1000-fold higher than the parent aldehyde, indicating specific recognition of glutathione by the active site residues of AR. An increase in the catalytic efficiency upon glutathiolation was also observed with trans-2-nonenal, trans-2-hexenal and trans, trans-2,4-decadienal, establishing that enhancement of catalytic efficiency was specifically due to the glutathione backbone and not specific to the aldehyde. Structure-activity relationships with substitution or deletion of amino acids of GSH indicated specific interactions of the active site with gamma-Glu1 and Cys of GSH. Molecular modeling revealed that the glutathione-propanal conjugate could bind in two distinct orientations. In orientation 1, gamma-Glu1 of the conjugate interacts with Trp20, Lys21 and Val47, and Gly3 interacts with Ser302 and Leu301, whereas in orientation 2, the molecule is inverted with gamma-Glu1 interacting with Ser302, and Leu301. Taken together, these data suggest that glutathiolation of aldehydes enhances their compatibility with the AR active site, which may be of physiological significance in detoxification of endogenous and xenobiotic aldehydes.

Metabolic regulation of aldose reductase activity by nitric oxide donors.

Regulation of aldose reductase (AR), a member of the aldo-keto reductase superfamily, by nitric oxide (NO) donors was examined. Incubation of human recombinant AR with S-nitrosoglutathione (GSNO) led to inactivation of the enzyme and the formation of an AR-glutathione adduct. In contrast, incubation with S-nitroso-N-acetyl penicillamine (SNAP) or N-(beta-D-glucopyranosyl)-SNAP (GlycoSNAP) led to an increase in enzyme activity which was accompanied by the direct nitrosation of the enzyme and the formation of a mixed disulfide with the NO-donor. To examine in vivo modification, red blood cells (RBC) and rat aortic vascular smooth muscle cells (VSMC) were incubated with 1 mM GSNO or SNAP. Exposure of VSMC to SNAP and GSNO for 2 h at 37 degrees C led to approximately 71% decrease in the enzyme activity with DL-glyceraldehyde as the substrate. Similarly, exposure of RBC in 5 mM glucose to NO-donors for 30 min at room temperature, followed by increasing the glucose concentration to 40 mM, resulted in >75% decrease in the formation of sorbitol. These investigations indicate that NO and/or its bioactive metabolites can regulate cellular AR, leading to either activation (by nitrosation) or inactivation (by S-thiolation).

Metabolism of lipid peroxidation product, 4-hydroxynonenal (HNE) in rat erythrocytes: role of aldose reductase.

Lipid peroxidation represents a significant source of erythrocyte dysfunction and aging. Because the toxicity of lipid peroxidation appears to be in part due to aldehydic end products, we examined, in rat erythrocytes, the metabolism of 4-hydroxy-trans-2-nonenal (HNE), one of the most abundant and toxic lipid-derived aldehydes. Packed erythrocytes, 0.1 ml, completely metabolized 20 nmoles of HNE in 20 min. The glutathione conjugate of HNE and 4-hydroxynonanoic acid (HNA) represented 70 and 25% of the total metabolism, respectively. Approximately 70% of the metabolites were extruded to the medium. Upon electrospray ionization mass spectrometry, the glutathione conjugate resolved into two distinct species corresponding to glutathionyl HNE (GS-HNE) and glutathionyl 1,4-dihydroxynonene (GS-DHN). The concentration of GS-DHN formed was twice that of GS-HNE. Inhibition of aldose reductase by sorbinil and tolrestat led to a selective decrease in the formation of GS-DHN, although the extent of HNE glutathiolation was unaffected. Inhibitors of aldehyde or alcohol dehydrogenase, i.e., cyanamide and 4-methyl pyrazole, had no effect on the formation of HNA and GS-DHN, indicating that these enzymes are not significant participants in the erythrocyte HNE metabolism. Thus, oxidation to HNA, conjugation with glutathione, and further reduction of the conjugate by aldose reductase appear to be the major pathways of HNE metabolism in erythrocytes. These pathways may be critical determinants of erythrocyte toxicity due to lipid peroxidation-derived aldehydes.

Selective recognition of glutathiolated aldehydes by aldose reductase.

In this study, the selectivity and specificity of aldose reductase (AR) for glutathionyl aldehydes was examined. Relative to free aldehydes, AR was a more efficient catalyst for the reduction of glutathiolated aldehydes. Reduction of glutathionyl propanal [gammaGlu-Cys(propanal)-Gly] was more efficient than that of Gly-Cys(propanal)-Gly and gamma-aminobutyric acid-Cys(propanal)-Gly suggesting a possible interaction between alpha-carboxyl of the conjugate and AR. Two active site residues, Trp20 or Ser302, were identified by molecular modeling as potential sites of this interaction. Mutations containing tryptophan-to-phenylalanine (W20F) and serine-to-alanine (S302A) substitutions did not significantly affect reduction of free aldehydes but decreased the catalytic efficiency of AR for glutathiolated aldehydes. Combined mutations indicate that both Trp20 and Ser302 are required for efficient catalysis of the conjugates. The decrease in efficiency due to W20F mutation with glutathionyl propanal was not observed with gamma-aminobutyric-Cys(propanal)-Gly or Gly-Cys-(propanal)-Gly, indicating that Trp20 is involved in binding the alpha-carboxyl of the conjugate. The effect of the S302A mutation was less severe when gammaGlu-Cys(propanal)-Glu rather than glutathionyl propanal was used as the substrate, consistent with an interaction between Ser302 and Gly-3 of the conjugate. These observations suggest that glutathiolation facilitates aldehyde reduction by AR and enhances the range of aldehydes available to the enzyme. Because the N-terminal carboxylate is unique to glutathione, binding of the conjugate with the alpha-carboxyl facing the bottom of the alpha/beta-barrel may assist in the exclusion of unrelated peptides and proteins.

Kinetic and structural characterization of the glutathione-binding site of aldose reductase.

Aldose reductase (AR), a member of the aldo-keto reductase superfamily, has been implicated in the etiology of secondary diabetic complications. However, the physiological functions of AR under euglycemic conditions remain unclear. We have recently demonstrated that, in intact heart, AR catalyzes the reduction of the glutathione conjugate of the lipid peroxidation product 4-hydroxy-trans-2-nonenal (Srivastava, S., Chandra, A., Wang, L., Seifert, W. E., Jr., DaGue, B. B., Ansari, N. H., Srivastava, S. K., and Bhatnagar, A. (1998) J. Biol. Chem. 273, 10893-10900), consistent with a possible role of AR in the metabolism of glutathione conjugates of aldehydes. Herein, we present several lines of evidence suggesting that the active site of AR forms a specific glutathione-binding domain. The catalytic efficiency of AR in the reduction of the glutathione conjugates of acrolein, trans-2-hexenal, trans-2-nonenal, and trans,trans-2,4-decadienal was 4-1000-fold higher than for the corresponding free alkanal. Alterations in the structure of glutathione diminished the catalytic efficiency in the reduction of the acrolein adduct, consistent with the presence of specific interactions between the amino acid residues of glutathione and the AR active site. In addition, non-aldehydic conjugates of glutathione or glutathione analogs displayed active-site inhibition. Molecular dynamics calculations suggest that the conjugate adopts a specific low energy configuration at the active site, indicating selective binding. These observations support an important role of AR in the metabolism of glutathione conjugates of endogenous and xenobiotic aldehydes and demonstrate, for the first time, efficient binding of glutathione conjugates to an aldo-keto reductase.

Asymmetric distribution of phosphatidylethanolamine in C. albicans: possible mediation by CDR1, a multidrug transporter belonging to ATP binding cassette (ABC) superfamily.

By using two molecular probes, we demonstrate that only 4% of total phosphatidylethanolamine (PtdEtn) in the plasma membrane (PM) of a human pathogenic yeast, Candida albicans, is present in its external half. Evidence is presented to show that the availability of PtdEtn could be related to the expression of a multidrug transporter CDR1 of C. albicans, and the process is energy-dependent. A homozygous CDR1 disruptant strain of C. albicans shows almost 23% reduction in the external labelling of PtdEtn. This report shows that, similar to human MDRs, yeast multidrug transporter could also be involved in aminophospholipid translocation.

Role of the actin cytoskeleton in regulating the outer phosphatidylethanolamine levels in yeast plasma membrane.

Transbilayer phosphatidylethanolamine (PtdEtn) movements in the plasma membrane of Saccharomyces cerevisiae are regulated by an ATP-dependent, protein-mediated process(es). To examine whether this process is influenced by the actin cytoskeleton, we have studied the PtdEtn translocation in S. cerevisiae cells after treatment with microfilament disrupting and microtubule-disrupting agents. PtdEtn translocation was studied by measuring the external PtdEtn levels, using fluorescamine as the external membrane probe, in the ATP-depleted, ATP-depleted and repleted, and N-ethylmaleimide-treated cells. The microfilaments and microtubules were disrupted by treatment with various cytochalasins and colchicine (or benomyl) respectively PtdEtn translocation became abnormal in the cytochalasin-treated cells but not in cells that were treated with microtubule-disrupting agents, such as colchicine or benomyl. These results have been interpreted to suggest that the actin cytoskeleton is involved in regulating the PtdEtn translocase activity in the yeast cell plasma membrane.