Silencing of GSTP1, a prostate cancer prognostic gene, by the estrogen receptor-ß and endothelial nitric oxide synthase complex.

We recently identified in prostate tumors (PCa) a transcriptional prognostic signature comprising a significant number of genes differentially regulated in patients with worse clinical outcome. Induction of up-regulated genes was due to chromatin remodeling by a combinatorial complex between estrogen receptor (ER)-ß and endothelial nitric oxide synthase (eNOS). Here we show that this complex can also repress transcription of prognostic genes that are down-regulated in PCa, such as the glutathione transferase gene GSTP1. Silencing of GSTP1 is a common early event in prostate carcinogenesis, frequently caused by promoter hypermethylation. We validated loss of glutathione transferase (GST) P1-1 expression in vivo, in tissue microarrays from a retrospective cohort of patients, and correlated it with decreased disease-specific survival. Furthermore, we show that in PCa cultured cells ERß/eNOS causes GSTP1 repression by being recruited at estrogen responsive elements in the gene promoter with consequential remodeling of local chromatin. Treatment with ERß antagonist or its natural ligand 5α-androstane-3ß,17ß-diol, eNOS inhibitors or ERß small interference RNA abrogated the binding and reversed GSTP1 silencing, demonstrating the direct involvement of the complex. In vitro, GSTP1 silencing by ERß/eNOS was specific for cells from patients with worse clinical outcome where it appeared the sole mechanism regulating GSTP1 expression because no promoter hypermethylation was present. However, in vivo chromatin immunoprecipitation assays on fresh PCa tissues demonstrated that silencing by ERß/eNOS can coexist with promoter hypermethylation. Our findings reveal that the ERß/eNOS complex can exert transcriptional repression and suggest that this may represent an epigenetic event favoring inactivation of the GSTP1 locus by methylation. Moreover, abrogation of ERß/eNOS function by 3ß-adiol emphasizes the significance of circulating or locally produced sex steroid hormones or their metabolites in PCa biology with relevant clinical/therapeutic implications.

A histidine-rich metal binding domain at the N terminus of Cu,Zn-superoxide dismutases from pathogenic bacteria: a novel strategy for metal chaperoning.

A group of Cu,Zn-superoxide dismutases from pathogenic bacteria is characterized by histidine-rich N-terminal extensions that are in a highly exposed and mobile conformation. This feature allows these proteins to be readily purified in a single step by immobilized metal affinity chromatography. The Cu,Zn-superoxide dismutases from both Haemophilus ducreyi and Haemophilus parainfluenzae display anomalous absorption spectra in the visible region due to copper binding at the N-terminal region. Reconstitution experiments of copper-free enzymes demonstrate that, under conditions of limited copper availability, this metal ion is initially bound at the N-terminal region and subsequently transferred to an active site. Evidence is provided for intermolecular pathways of copper transfer from the N-terminal domain of an enzyme subunit to an active site located on a distinct dimeric molecule. Incubation with EDTA rapidly removes copper bound at the N terminus but is much less effective on the copper ion bound at the active site. This indicates that metal binding by the N-terminal histidines is kinetically favored, but the catalytic site binds copper with higher affinity. We suggest that the histidine-rich N-terminal region constitutes a metal binding domain involved in metal uptake under conditions of metal starvation in vivo. Particular biological importance for this domain is inferred by the observation that its presence enhances the protection offered by periplasmic Cu,Zn-superoxide dismutase toward phagocytic killing.

Valine 10 may act as a driver for product release from the active site of human glutathione transferase P1-1.

We have probed the electrophilic binding site (H-site) of human glutathione transferase P1-1 through mutagenesis of two valines, Val 10 and Val 35, into glycine and alanine, respectively. These two residues were previously shown to be the only conformationally variable residues in the H-site and hence may play important roles in cosubstrate recognition and/or product dissociation. Both of these mutant enzymes have been expressed in Escherichia coli and purified and their kinetic properties characterized. The results demonstrate that Val35Ala behaves similarly to wild-type, whereas Val10Gly exhibits a strong decrease of k(cat) and k(cat)/K(m) (cosub) toward two selected cosubstrates: ethacrynic acid and 1-chloro-2,4-dinitrobenzene. Pre-steady-state kinetic analysis of the GSH conjugation with ethacrynic acid shows that both wild-type and Val10Gly mutant enzymes exhibit the same rate-limiting step: the dissociation of product. However, in the Val10Gly mutant there is an increased energetic barrier which renders the dissociation of product more difficult. Similar results are found for the Val10Gly mutant with 1-chloro-2,4-dinitrobenzene as cosubstrate. With this latter cosubstrate, Val 10 also exerts a positive role in the conformational transitions of the ternary complex before the chemical event. Crystallographic analysis of the Val10Gly mutant in complex with the inhibitor S-hexyl-GSH suggests that Val 10 optimally orientates products, thus promoting their exit from the active site.

The ligandin (non-substrate) binding site of human Pi class glutathione transferase is located in the electrophile binding site (H-site).

Glutathione S -transferases (GSTs) play a pivotal role in the detoxification of foreign chemicals and toxic metabolites. They were originally termed ligandins because of their ability to bind large molecules (molecular masses >400 Da), possibly for storage and transport roles. The location of the ligandin site in mammalian GSTs is still uncertain despite numerous studies in recent years. Here we show by X-ray crystallography that the ligandin binding site in human pi class GST P1-1 occupies part of one of the substrate binding sites. This work has been extended to the determination of a number of enzyme complex crystal structures which show that very large ligands are readily accommodated into this substrate binding site and in all, but one case, causes no significant movement of protein side-chains. Some of these molecules make use of a hitherto undescribed binding site located in a surface pocket of the enzyme. This site is conserved in most, but not all, classes of GSTs suggesting it may play an important functional role.

Temperature adaptation of glutathione S-transferase P1-1. A case for homotropic regulation of substrate binding.

Human glutathione S-transferase P1-1 (GST P1-1) is a homodimeric enzyme expressed in several organs as well as in the upper layers of epidermis, playing a role against carcinogenic and toxic compounds. A sophisticated mechanism of temperature adaptation has been developed by this enzyme. In fact, above 35 degrees C, glutathione (GSH) binding to GST P1-1 displays positive cooperativity, whereas negative cooperativity occurs below 25 degrees C. This binding mechanism minimizes changes of GSH affinity for GST P1-1 because of temperature fluctuation. This is a likely advantage for epithelial skin cells, which are naturally exposed to temperature variation and, incidentally, to carcinogenic compounds, always needing efficient detoxifying systems. As a whole, GST P1-1 represents the first enzyme which displays a temperature-dependent homotropic regulation of substrate (e.g. GSH) binding.

In vivo formation of Cu,Zn superoxide dismutase disulfide bond in Escherichia coli.

We have found that the in vivo folding of periplasmic Escherichia coli Cu,Zn superoxide dismutase is assisted by DsbA, which catalyzes the efficient formation of its single disulfide bond, whose integrity is essential to ensure full catalytic activity to the enzyme. In line with these findings, we also report that the production of recombinant Xenopus laevis Cu,Zn superoxide dismutase is enhanced when the enzyme is exported in the periplasmic space or is expressed in thioredoxin reductase mutant strains. Our data show that inefficient disulfide bond oxidation in the bacterial cytoplasm inhibits Cu,Zn superoxide dismutase folding in this cellular compartment.

Shifting substrate specificity of human glutathione transferase (from class Pi to class alpha) by a single point mutation.

Substrate selectivity, among glutathione transferase (GST) isoenzymes, appears to be determined by a few residues. As part of study to determine which residues are class-specific determinants, Tyr 108 (an important residue of the class Pi) has been changed to a valine, the structural equivalent of a class Alpha enzyme. Using a panel of selected substrates, "diagnostic" for either class Pi or Alpha, it is shown here that this single mutation significantly alters the catalytic properties of the class Pi enzyme and shifts the substrate specificity of the enzyme toward that of the class Alpha enzyme.

Flexibility of helix 2 in the human glutathione transferase P1-1. time-resolved fluorescence spectroscopy.

Time-resolved fluorescence spectroscopy and site-directed mutagenesis have been used to probe the flexibility of alpha-helix 2 (residues 35-46) in the apo structure of the human glutathione transferase P1-1 (EC 2.5.1.18) as well as in the binary complex with the natural substrate glutathione. Trp-38, which resides on helix 2, has been exploited as an intrinsic fluorescent probe of the dynamics of this region. A Trp-28 mutant enzyme was studied in which the second tryptophan of glutathione transferase P1-1 is replaced by histidine. Time-resolved fluorescence data indicate that, in the absence of glutathione, the apoenzyme exists in at least two different families of conformational states. The first one (38% of the total population) corresponds to a number of slightly different conformations of helix 2, in which Trp-38 resides in a polar environment showing an average emission wavelength of 350 nm. The second one (62% of the total population) displays an emission centered at 320 nm, thus suggesting a quite apolar environment near Trp-38. The interconversion between these two conformations is much slower than 1 ns. In the presence of saturating glutathione concentrations, the equilibrium is shifted toward the apolar component, which is now 83% of the total population. The polar conformers, on the other hand, do not change their average decay lifetime, but the distribution becomes wider, indicating a slightly increased rigidity. These data suggest a central role of conformational transitions in the binding mechanism, and are consistent with NMR data (Nicotra, M., Paci, M., Sette, M., Oakley, A. J., Parker, M. W., Lo Bello, M., Caccuri, A. M., Federici, G., and Ricci, G. (1998) Biochemistry 37, 3020-3027) and pre-steady state kinetic experiments (Caccuri, A. M., Lo Bello, M., Nuccetelli, M., Nicotra, M., Rossi, P., Antonini, G., Federici, G., and Ricci, G. (1998) Biochemistry 37, 3028-3034) indicating the existence of a pre-complex in which GSH is not firmly bound to the active site.

Mutations of Gly to Ala in human glutathione transferase P1-1 affect helix 2 (G-site) and induce positive cooperativity in the binding of glutathione.

Previous kinetic studies on human glutathione transferase P1-1 have indicated that the motions of an irregular alpha-helix (helix 2) lining the glutathione (GSH) binding site are viscosity dependent and may modulate the affinity of GSH binding. The effect of single amino acid residue substitutions (Gly to Ala) in this region is investigated here by site-directed mutagenesis. Three mutants (Gly41Ala, Gly50Ala and Gly41Ala/Gly50Ala) were overexpressed in Escherichia coli, purified, and characterized by kinetic, structural, and spectroscopic studies. All these mutant enzymes show kcat values similar to that of the wild-type enzyme, while the [S]0.5 for GSH increases about eight-fold in the Gly41Ala mutant and more than 100-fold in the Gly41Ala/Gly50Ala double mutant. This change in affinity towards GSH is accompanied by an induced positive cooperativity as reflected by Hill coefficients of 1.4 (Gly41Ala) and 1.7 (Gly41Ala/Gly50Ala) upon substrate binding. Taken together, these data suggest that the region around helix 2 is markedly altered leading to the observed intersubunit communication. Molecular modeling of the Gly41Ala/Gly50Ala mutant and of the inactive oxidized form of the native enzyme provides a structural explanation of our results.

Multifunctional role of Tyr 108 in the catalytic mechanism of human glutathione transferase P1-1. Crystallographic and kinetic studies on the Y108F mutant enzyme.

The possible role of the hydroxyl group of Tyr 108 in the catalytic mechanism of human glutathione transferase P1-1 has been investigated by means of site-directed mutagenesis, steady-state kinetic analysis, and crystallographic studies. Three representative cosubstrates have been used, i.e. ethacrynic acid, 7-chloro-4-nitrobenz-2-oxa-1,3-diazole, and 1-chloro-2,4-dinitrobenzene. In the presence of ethacrynic acid, the enzyme follows a rapid equilibrium random bi-bi mechanism with a rate-limiting step which occurs after the addition of the substrates and before the release of products. The replacement of Tyr 108 with Phe yields a 14-fold decrease of k(cat), while it does not change appreciably the affinity of the H site for the substrate. In this case, it would appear that the role of the hydroxyl function is to stabilize the transition state for the chemical step, i.e. the Michael addition of GSH to the electrophilic substrate. Crystallographic data are compatible with this conclusion showing the hydroxyl group of Y108 in hydrogen bonding distance of the ketone moiety of ethacrynic acid [Oakley, A. J., Rossjohn, J., Lo Bello, M., Caccuri, A. M., Federici, G., & Parker, M. W. (1997) Biochemistry 36, 576-585]. Moreover, no structural differences are observed between the Y108F mutant and the wild type, suggesting that the removal of the hydroxyl group is solely responsible for the loss of activity. A different involvement of Tyr 108 appears in the catalyzed conjugation of 7-chloro-4-nitrobenz-2-oxa-1,3-diazole with GSH in which the rate-limiting step is of a physical nature, probably a structural transition of the ternary complex. The substitution of Tyr 108 yields an approximately 7-fold increase of k(cat) and a constant k(cat)/Km(NBD-Cl) value. Lack of a critical hydrogen bond between 7-chloro-4-nitrobenz-2-oxa-1,3-diazole and Tyr 108 appears to be the basis of the increased k(cat). In the 1-chloro-2,4-dinitrobenzene/GSH system, no appreciable changes of kinetics parameters are found in the Y108F mutant. We conclude that Y108 has a multifunctional role in glutathione transferase P1-1 catalysis, depending on the nature of the electrophilic cosubstrate.

The glutathione conjugate of ethacrynic acid can bind to human pi class glutathione transferase P1-1 in two different modes.

The diuretic drug ethacrynic acid, an inhibitor of pi class glutathione S-transferase, has been tested in clinical trials as an adjuvant in chemotherapy. We recently solved the crystal structure of this enzyme in complex with ethacrynic acid and its glutathione conjugate. Here we present a new structure of the ethacrynic-glutathione conjugate complex. In this structure the ethacrynic moiety of the complex is shown to bind in a completely different orientation to that previously observed. Thus there are at least two binding modes possible, an observation of great importance to the design of second generation inhibitors of the enzyme.

Structural flexibility modulates the activity of human glutathione transferase P1-1. Role of helix 2 flexibility in the catalytic mechanism.

Presteady-state and steady-state kinetic studies performed on human glutathione transferase P1-1 (EC 2.5.1.18) with 1-chloro-2, 4-dinitrobenzene as co-substrate indicate that the rate-determining step is a physical event that occurs after binding of the two substrates and before the final sigma-complex formation. It may be a structural transition involving the ternary complex. This event can be related to diffusion-controlled motions of protein portions as kcat degrees /kcat linearly increases by raising the relative viscosity of the solution. Similar viscosity dependence has been observed for Km GSH, while Km CDNB is independent. No change of the enzyme structure by viscosogen has been found by circular dichroism analysis. Thus, kcat and Km GSH seem to be related to the frequency and extent of enzyme structural motions modulated by viscosity. Interestingly, the reactivity of Cys-47 which can act as a probe for the flexibility of helix 2 is also modulated by viscosity. Its viscosity dependence parallels that observed for kcat and Km GSH, thereby suggesting a possible correlation between kcat, Km GSH, and diffusion-controlled motion of helix 2. The viscosity effect on the kinetic parameters of C47S and C47S/C101S mutants confirms the involvement of helix 2 motions in the modulation of Km GSH, whereas a similar role on kcat cannot be ascertained unequivocally. The flexibility of helix 2 modulates also the homotropic behavior of GSH in these mutants. Furthermore, fluorescence experiments support a structural motion of about 4 A occurring between helix 2 and helix 4 when GSH binds to the G-site.

Interaction of glutathione transferase P1-1 with captan and captafol.

Glutathione transferase (GST, EC 2.5.1.18) P1-1 was strongly inhibited by captan and captafol in a time- and concentration-dependent manner. The IC50 values for captan and captafol were 5.8 microM and 1.5 microM, respectively. Time-course inactivation of GSTP1-1 by two pesticides was prevented by 3 microM of hexyl-glutathione, but not by methylglutathione. The fact that the inactivated enzyme recovered all the 5,5'-dithiobis(2-nitrobenzoic acid) titrable thiol groups, with concomitant recovery of all its original activity after treatment with 100 microM dithiothreitol, suggested that captan and captafol were able to induce the formation of disulfide bonds. That the inactivation of GSTP1-1 by captan and captafol involves the formation of disulfide bonds between the four cysteinil groups of the enzymes was confirmed by the SDS-PAGE experiments on nondenaturant conditions. In fact, on SDS-PAGE, GSTP1-1 as well as the cys47ala, cys101ala, and cys47ala/cys101ala GSTP1-1 mutants treated with captan and captafol showed several extra bands, with apparent molecular masses higher and lower than the molecular mass of native GSTP1-1 (23.5 kDa), indicating that both intra- and inter-subunit disulfide bonds were formed. These extra bands returned to the native 23.5 kDa band with concomitant restoration of activity when treated with dithiothreitol.

Cytoplasmic and periplasmic production of human placental glutathione transferase in Escherichia coli.

An expression vector yielding large amounts of GST P1-1 in the cytoplasm of Escherichia coli was constructed. The recombinant enzyme, obtained after purification, was characterized in its physicochemical and kinetic properties and appeared to be indistinguishable from that purified from human placenta. However, N-terminal amino acid sequencing revealed that about 50% of the recombinant GST still contained methionine as the N-terminal amino acid. Such an incomplete processing was not simply due to overproduction of GST. In fact, under growth conditions that lead to a sharp decrease in the production of the protein the N-terminal methionine was not removed. GST was unable to translocate across the bacterial membrane when it was fused to the leader peptide of the pelB gene from Erwinia carotovora and accumulated in the cytoplasm in a soluble and active conformation. However, when this fusion protein was produced in a bacterial strain overexpressing the bacterial chaperonins GroEL and GroES, a fraction of GST was exported into the periplasmic space with the correct N-terminal sequence. The yield of correctly processed GST accounted for 12% of total GST present in the E. coli cells. Our results suggest that chaperonins are able to interact with nascent GST, thus maintaining the protein in an export-competent form and that E. coli strains with enhanced secretory characteristics may be obtained by genetic engineering technology.

Site-directed mutagenesis of human glutathione transferase P1-1. Spectral, kinetic, and structural properties of Cys-47 and Lys-54 mutants.

In the human placental glutathione transferase, Cys-47 possesses, at physiological pH values, a pK alpha value of 4.2 and may exist as an ion pair with the protonated epsilon-amino group of Lys-54. Using site-directed mutagenesis we investigate spectral, kinetic, and structural properties of Cys-47 and Lys-54 mutants. The results shown indicate that the thiolate ion detected at 229 nm should be assigned exclusively to Cys-47. The contribution of Lys-54 to the activation of Cys-47 is assessed by the spectral properties of the K54A mutant enzyme. The induced cooperativity toward glutathione, as a consequence of mutation of Lys-54 to alanine, clearly parallels that observed for the Cys-47 mutant enzymes (see the preceding paper (Ricci, G., Lo Bello, M., Caccuri, A. M., Pastore, A., Nuccetelli, M., Parker, M. W., and Federici, G. (1995) J. Biol. Chem. 270, 1243-1248) and points out the importance of this electrostatic interaction in shaping the correct spatial arrangement for the binding of glutathione and in anchoring the flexible helix alpha 2. When this ion pair is disrupted, by mutation of either residue, the flexibility of this region could be greatly increased, causing helix alpha 2 to come in contact with the other subunit and generating a structural communication, which is the basis of the observed cooperativity.

Temperature-dependent protein folding in vivo--lower growth temperature increases yield of two genetic variants of Xenopus laevis Cu,Zn superoxide dismutase in Escherichia coli.

Two genetic variants of Xenopus laevis Cu,Zn superoxide dismutase, XSODA and XSODB, have been expressed in Escherichia coli by recombinant DNA techniques. Production of both proteins was obtained, although with different yields, XSODB being more abundant than XSODA in all the conditions tested. Lowering the temperature of growth was found to be a specific factor, decisive in obtaining quantitatively abundant, active Xenopus enzymes. Impaired folding of these proteins in the E.coli cytoplasm was found to parallel their in vitro properties.