Cellular and molecular pharmacology of auranofin and related gold complexes.
Explore the source record for details and available documents.
Biomedical subjects
Publications and source records attributed to S T Crooke.
Explore the source record for details and available documents.
Metallothioneins are low molecular weight, cysteine-rich proteins believed to participate in metal detoxification. Turnover of Cd-, Zn-, and Au-induced metallothionein was studied in a Chinese hamster ovary cell line which was resistant to Cd and the Au-containing drug auranofin. Cd, Zn, and Au were potent inducers of metallothionein mRNA and resulted in accumulation of approximately equal amounts of mRNA. Pulse-chase studies with [35S]cysteine revealed that the half-life of Au-, Zn-, and Cd-induced metallothionein was 0.75, 10, and 24 h, respectively. The differences in the half-life of metallothionein may be related to the tertiary structure of metal-metallothionein complexes. These results have implications in the mechanism of resistance to gold compounds.
Leukotriene D4 stimulates prostanoid synthesis in smooth muscle and endothelial cells. Because phospholipases A2 and C have been proposed to regulate prostanoid synthesis, we examined the effect of leukotriene D4 on these activities. Leukotriene D4 treatment resulted in a dose-dependent, stereospecific increase in phospholipase A2 activity with phosphatidylcholine as a substrate. The induction of phospholipase A2 activity occurred just prior to the appearance of prostanoids in the media. Protein and RNA synthesis were required for the increase in phospholipase A2 activity, and the increase in activity resulted from an increase in the apparent Vmax of the phospholipase A2 enzyme. Phospholipase C activity using various substrates was unchanged. We conclude that the increase in prostanoid synthesis observed after leukotriene D4 treatment is a result of an increase in a phospholipase A2 activity.
Vasopressin-induced phosphatidylinositol turnover and mobilization of intracellular Ca2+ was studied using an established smooth muscle cell line (A-10). The cells were subcloned to ensure a monoclonal cell population. The accumulation of inositol mono-, di-, and tris-phosphates (IP1, IP2, and IP3, respectively), and the mobilization of intracellular Ca2+ were dependent on the time of incubation and the concentration of arginine vasopressin (AVP). IP1, IP2, and IP3 were significantly elevated after 15 sec and remained elevated for up to 2 hr. The concentrations of AVP required for half-maximal stimulation of IP1, IP2, and IP3 formation were 2, 12, and 4 nM, respectively. LiCl was required to observe the accumulation of inositol phosphates in response to AVP. Significant 45Ca2+ efflux was observed within 15 sec after exposure to AVP. By employing the vasopressin receptor subtype selective antagonists [d(CH2)5Tyr(Me)AVP, V1; d(CH2)5D-Tyr(Et)VAVP,V1/V2; d(CH2) 5D-IleVAVP,V2] and agonists [AVP, V1/V2; dDAVP, V2; dVDAVP, V2], we found that the vasopressin-induced stimulation of phosphatidylinositol turnover and 45Ca2+ efflux were mediated by receptors of the vascular V1 subtype. Pertussis toxin pretreatment partially inhibited vasopressin-induced phosphatidylinositol turnover. These data demonstrate that activation of V1 receptors of vascular smooth muscle cells resulted in enhanced phosphatidylinositol turnover and mobilization of intracellular Ca2+.
Isoproterenol-induced desensitization of turkey erythrocyte adenylate cyclase is accompanied (1) by a decrease in the mobility of beta-adrenergic receptor proteins, specifically photoaffinity labeled with 125I-(p-azidobenzyl)carazolol (125I-PABC), on sodium dodecyl sulfate (SDS)-polyacrylamide gels and (2) by a 2-3-fold increase in phosphate incorporation into the beta receptor [Stadel, J.M., Nambi, P., Shorr, R. G. L., Sawyer, D. F., Caron, M. G., & Lefkowitz, R. J. (1983) Proc. Natl. Acad. Sci. U.S.A. 80, 3173]. Analysis of 32P-labeled beta receptors partially purified by affinity chromatography and subsequently hydrolyzed in 6 N HCl revealed that the beta receptor from control erythrocytes contained only phosphoserine and that agonist-promoted phosphorylation of the receptor in desensitized cells occurred on serine residues. Comparison of limited-digest peptide maps of 125I-PABC-labeled beta receptors from control and desensitized erythrocytes reveals distinctly different sensitivities of the two beta receptors to cleavage by chymotrypsin and Staphylococcus aureus protease. The altered mobility of the 125I-PABC-labeled beta receptor from desensitized erythrocytes was eliminated when 5 M urea was included in the SDS-polyacrylamide gels. Limited-digest peptide mapping of 32P-labeled beta receptors from control and desensitized cells with the protease papain identified a unique phosphorylated peptide in desensitized preparations. Our results are consistent with the hypothesis that the altered mobility of beta-receptor proteins on SDS gels following desensitization is due to changes in conformation promoted by prolonged exposure to agonists.
The potent and D-1 versus D-2 selective dopamine receptor antagonist, SK&F R-83566, was radiolabelled with tritium and was used as a radioligand for examination of D-1 receptors in rat striatum. Binding of the radioligand was stereoselective, saturable and reversible. In homogenates of rat striatum, nonspecific binding of the radioligand was less than 5% of total binding, the KD was 1.1 +/- 0.2 nM and the Bmax was 1130 +/- 70 fmoles/mg protein. Results of competition binding analyses yielded a pharmacological profile that was characteristic of dopamine D-1 receptor interaction. Competition studies of dopamine agonists against the potent antagonist radioligand indicated multiple affinities of agonist binding to the D-1 receptor. Displacement was best fit to a two-site model of ligand binding and high and low affinities were subject to regulation by guanine, but not adenine, nucleotides. Antagonist binding was not complex and was unaffected by guanine nucleotides. The role of monovalent cations in regulating D-1 receptor binding was evaluated by comparing effects of Na+, Li+, and K+ on binding of the antagonist [3H]SK&F R-83566 and the agonist [3H]fenoldopam (SK&F 82526). Whereas agonist binding was reduced in a concentration dependent fashion by monovalent cations with a ranking of potency Li+ greater than Na+ greater than K+, antagonist binding was enhanced by the cation Na+ but little affected by Li+ or K+. This effect of relatively low concentrations of Na+ to decrease agonist binding and increase antagonist binding suggests similarities between the D-1 receptor which is positively-coupled to adenylate cyclase and other receptors, e.g. alpha 2 adrenergic receptors, which are negatively-coupled to adenylate cyclase.
We investigated the interactions of 4'-(9-acridinylamino)methanesulfon-m-anisidide (mAMSA) with thiol-containing compounds and the potential binding of the thiolytic adducts to DNA. All thiols tested (glutathione, cysteine, coenzyme A, 2-mercaptoethanol and lactate dehydrogenase) formed adducts with mAMSA as evidenced by changes in the absorption spectrum of mAMSA and induction of fluorescence. Spectral changes induced by the thiols were different, suggesting that each thiol induced specific changes in the electronic structure of the acridine nucleus. Treatment of glutathione with p-chloromercuribenzoate eliminated the absorption spectral changes and induction of fluorescence, indicating that the reduced-thiol group is involved. In high ionic strength buffer, addition of calf thymus DNA induced fluorescence-quenching of both the mAMSA-glutathione and mAMSA-cysteine adducts without spectral shift. Viscometric studies showed that mAMSA and mAMSA-glutathione intercalated into DNA and produced similar increases in the length of linear DNA.
The interactions of certain gold(I) and gold(III) complexes with isolated plasmid pBR322 DNA were defined and compared to those of cis-diamminedichloroplatinum(II), CDDP, using an agarose gel electrophoresis assay. Trichloro(pyridine)gold(III) appeared to bind to DNA as evidenced by its ability to produce dose-dependent changes in the electrophoretic mobilities of closed circular, supercoiled, closed circular, relaxed, and open circular plasmid DNAs. These effects suggest that the gold containing complex induces conformational changes in the plasmid as a result of the compound binding to the DNA and the subsequent unwinding of the double helix and shorting of the DNA. Auranofin [(2,3,4,6-tetra-O-acetyl-1-thio-beta-D-glucopyranosato-S)-triethyl phosphine gold(I)] did not appear to interact with DNA under any conditions. However, its analog chloro(triethylphosphine) gold(I) interacted with DNA at pH 9.5 in borate buffer and produced electrophoretic mobility changes in pBR322 DNA which were different from those produced by the gold(III) complexes that were evaluated. Binding of chloro(triethylphosphine) gold(I) was inhibited by the co-addition of the thiosugar portion of auranofin suggesting preferential binding of the gold moiety to thiosugar, which results in the production of auranofin (or a sugar containing) gold complex and inhibition of gold binding to DNA. The interactions of a number of gold compounds with DNA were also evidenced by their abilities to inhibit the binding of ethidium bromide to DNA. The results from these studies indicate that: gold containing complexes can bind to, and produce conformational changes in, DNA; gold(I) and gold(III) complexes may interact with DNA via different chemical mechanisms to produce different conformational changes in DNA; and certain coordinating ligands in gold complexes (e.g. Cl, Br and SCN) can be exchanged for binding sites on DNA by gold.
The ability of gold coordination complexes to bind to DNA and produce inter-strand cross-links in DNA was assessed in an assay system based on the fluorescence properties of the DNA intercalative dye, ethidium bromide. Results from these studies using a variety of gold(I) and gold(III) complexes suggest that the ability of gold complexes to bind to and produce inter-strand cross-links in DNA is not dependent on the oxidation state of gold in the complex but is influenced by the nature of the coordinating ligands. Those complexes in which the gold was ligated through one or more weakly coordinating ligands showed evidence for DNA binding. However, only those complexes with two or more of these relatively weak coordinating ligands produced inter-strand cross-links. Both the amount of binding to and cross-linking of DNA by these compounds were decreased by treatment of the gold-DNA complex with 2-mercaptoethanol and other thiol containing agents. As shown by agarose gel electrophoresis, 2-mercaptoethanol caused a dissociation of the gold-DNA complexes and a regeneration of closed circular superhelical pBR322 DNA. DNA strand breakage also resulted from treatment of a number of gold-DNA complexes with 2-mercaptoethanol; this was observed with the gold compounds which were shown to produce inter-strand cross-links in DNA. The amount of DNA strand breakage produced by treatment of gold-DNA complexes with 2-mercaptoethanol was influenced by the initial conformation of the DNA; gold-DNA complexes which resulted from the binding of gold compounds to covalently closed superhelical DNA were more sensitive to the breakage induced by 2-mercaptoethanol treatment than those complexes in which closed circular, relaxed DNA was used as substrate. The DNA breakage was not reduced in partially anaerobic conditions or by free-radical scavengers, suggesting that it is not mediated by oxygen. The results are discussed with respect to the potential for the interaction of gold complexes with intracellular DNA and chromatin and their biological implications.
Cyclic voltammetry and UV/VIS spectrometry studies show that 4'-(9-acridinylamino)methanesulfon-m-anisidide (mAMSA) can be oxidized electrochemically to N1-methylsulfonyl-N4-(9-acridinyl)-3-methoxy-2,5-cyclohexadiene-1,4-d iimine (mAQDI) in Tris buffer, pH 7.5. The formal potential of this 2-electron process, as determined by spectroelectrochemical techniques, was 0.141 V versus saturated calomel electrode. Voltammetric data also indicate that an electron transfer reaction between mAMSA and Cu(II) was thermodynamically favored. Two lines of evidence suggest that mAQDI and Cu(I) are the active species in DNA breakage: (1) mAQDI, in the presence of Cu(I), induced both single- and double-strand DNA breakage of the superhelical pDPT275 form I DNA. mAQDI or Cu(I), when used alone, was less effective. (2) The DNA-breaking activity of an mAMSA-Cu(II) mixture was kinetically correlated with the production of both Cu(I) and mAQDI. Thin-layer chromatographic studies showed that mAMSA was oxidized to mAQDI which, in turn, was hydrolyzed. The end product was identified as 9-aminoacridine. When DNA breakage activity was measured as a function of reaction time, a biphasic response was observed. Maximal DNA-breaking activity was obtained upon mixing mAMSA and Cu(II) for 2-4 hr, depending on the concentrations of mAMSA and Cu(II), and was followed by a subsequent decrease in breakage. The decrease appears to be due to the decrease in Cu(I) production and the hydrolysis of mAQDI. These results substantiate the proposed mechanism that DNA breakage induced by mAMSA-Cu(II) involves a rate-limiting electron transfer step to form mAQDI and Cu(I), which are the active species for DNA breakages.
Auranofin (AF), an orally active, antiarthritic agent, modulates the functional activities of macrophages in vivo and in vitro. To better understand the molecular mechanism of action of auranofin with macrophages we have investigated its cellular association, intracellular distribution, and efflux with RAW 264.7 cells using auranofin radiolabeled within the triethylphosphine (Et3P) [3H], the gold [195Au] or the tetraacetylthioglucose (TATG)[14C] moieties of the molecule. Evaluation of the effects of auranofin on RAW 264.7 cells demonstrates that (1) cellular association of this compound was concentration, time and temperature dependent; (2) cellular association of AF was inhibited by N-ethylmaleimide but not by 2,4-dinitrophenol and NaF; (3) cellular association and uptake of Au and Et3P into cells was reduced when the drug was preincubated with increasing concentrations of fetal calf serum and albumin; (4) no tetraacetylthioglucose from the auranofin molecule became cell associated whereas the Au and Et3P moieties were internalized and distributed between the nuclear, cytosolic and membrane fractions of cells; and (5) efflux of Au and Et3P from RAW 264.7 cells was time and temperature dependent. Based on these data we propose a model, a sequential ligand exchange process, that describes the molecular interactions of auranofin and possibly other gold compounds with these cells.
The CPAE bovine endothelial cell line may be stimulated to produce eicosanoids. Leukotriene D4 increased the release of arachidonic acid primarily by activating phospholipase A2 while bradykinin activated the phospholipase C pathway. Cells pretreated with dexamethasone, a phospholipase A2 inhibitor, no longer responded to stimulation by LTD4 but did release arachidonic acid when treated with bradykinin. Aspirin blocked bradykinin-stimulated production of arachidonic acid but left the response to LTD4 unaffected. We conclude that these cells produce eicosanoids by activation of both PLA2 and PLC, and that the two different methods of arachidonic acid release can be distinguished by using the common anti-inflammatory drugs aspirin and dexamethasone.
Incubation of the bovine endothelial cell line, CPAE, with the calcium ionophore (A23187), bradykinin (BK), leukotriene D4 (LTD4) or leukotriene C4 (LTC4) resulted in concentration dependent increases in prostacyclin release measured as 6-keto-prostaglandin F1 alpha. The kinetics of induction of prostacyclin synthesis differed among the agents studied. Statistically significant increases in prostacyclin were observed one minute after treatment, with A23187, at slightly longer times with bradykinin and after approximately three minutes with the leukotrienes. Two other leukotrienes were tested. Both leukotriene B4 and leukotriene E4 (LTE4) were inactive at concentrations up to 10 microM. The induction of prostacyclin synthesis by LTC4 and LTD4 was inhibited by cycloheximide and actinomycin-D. The effect of BK was inhibited by cycloheximide but not by actinomycin-D. Induction by A23187 was not inhibited by either actinomycin-D or cycloheximide. The results suggest that these agents induced the increases in prostacyclin synthesis by different mechanisms.
A series of gold(I) coordination complexes including analogues of the antiarthritic agent auranofin 1 were evaluated for in vitro cytotoxic potency against both B16 melanoma cells and P388 leukemia cells and in vivo antitumor activity against P388 leukemia in mice. A number of the complexes showed potent cytotoxic activity in vitro and antitumor activity in vivo, with the phosphine-coordinated gold(I) thiosugar complexes demonstrating the greatest in vitro and in vivo activity. The data compiled for 63 complexes of the general structural formula LAuX provide the basis for the following observations: potent in vitro cytotoxic activity is observed for substituted (phosphine) gold complexes, lack of potency in vitro correlates well with lack of antitumor activity, potent cytotoxicity in vitro is not necessarily predictive of activity in vivo, in vivo antitumor activity is generally optimized by ligation of Au(I) with a substituted phosphine and a thiosugar.
Incubation of the bovine endothelial cell line, CPAE, with leukotriene D4, leukotriene C4, bradykinin, or the calcium ionophore A23187 results in the release of arachidonic acid metabolites including 6-keto-prostaglandin F1 alpha, the stable metabolite of prostacyclin. Pretreatment of these cells with the pertussis toxin islet-activating protein (IAP) results in a dose-dependent inhibition of the release of arachidonic acid metabolites and prostacyclin in response to leukotriene D4 and leukotriene C4. In contrast, similar responses evoked by bradykinin or ionophore were not significantly altered by the IAP pretreatment of the cells. IAP in the presence of [32P]NAD specifically [32P]ADP-ribosylates a 41-kDa protein in membranes prepared from CPAE cells. Pretreatment of the intact cells with IAP resulted in a dose-dependent inhibition of subsequent 32P labeling of the toxin substrate in the membranes and correlates with the uncoupling of the leukotriene responses. These results suggest that the 41-kDa IAP substrate, presumably a guanine nucleotide regulatory protein, mediates the response of CPAE cells to leukotriene D4 and leukotriene C4, but not to bradykinin or the calcium ionophore.
Rat thoracic aortic smooth muscle cells (line A10, ATCC CRL 1476) display a high density of atrial natriuretic factor (ANF) receptors. ANF stimulated the accumulation of cGMP in these cells in a time- and dose-dependent fashion. These cells are known to display a high density of vasopressin receptors of the vascular V1 subtype. These vasopressin receptors mediate inhibition of isoproterenol-stimulated cAMP accumulation and stimulation of inositol phosphate accumulation and calcium fluxes. Addition of [8-arginine]vasopressin ([Arg8]VP) to these cells inhibited ANF-stimulated cGMP accumulation. Inhibition of cGMP accumulation was dependent on the concentration of [Arg8]VP, with half-maximal and maximal effects occurring at 0.4 and 10 nM, respectively. [Arg8]VP did not have significant effects on basal cGMP levels. The inhibition by [Arg8]VP appears to be mediated by V1 receptors, since the V2 renal receptor agonist [1-desaminocysteine,8-D-arginine]vasopressin was ineffective. Also, the selective V1 antagonist [1-(beta-mercapto-beta,beta-cyclopentamethylenepropionic acid),2-(O-methyltyrosine),8-arginine]vasopressin and the mixed V1/V2 antagonist [1-(beta-mercapto-beta,beta-cyclopentamethylenepropionic acid),2-(O-ethyl-D-tyrosine),4-valine,8-arginine]vasopressin blocked the [Arg8]VP-mediated effect, whereas the selective V2 antagonist [1-(beta-mercapto-beta,beta-cyclopentamethylenepropionic acid), 2-D-isoleucine,4-valine,8-arginine]vasopressin was minimally effective. These data show that in rat aortic smooth muscle cells, V1 receptors are negatively coupled to guanylate cyclase. These data also suggest that the vasoconstrictor activity of [Arg8]VP might involve inhibition of ANF-receptor-mediated vascular relaxation through inhibition of cGMP accumulation in addition to its effects on isoproterenol-mediated cAMP accumulation and inositol phosphate accumulation and calcium fluxes.
Luzopeptin analogues, A, B and C (in decreasing degree of acetylation) were active, less active and inactive, respectively, against several animal tumour models. All bound to isolated DNA avidly. The present studies with cultured Novikoff hepatoma cells showed that luzopeptins A, B and C were inhibitory, less inhibitory and ineffective, respectively, on cell colony formation ability and whole-cell DNA and RNA synthesis. RNA synthesis was more sensitive than DNA synthesis to luzopeptins A and B. All had no direct effect on protein synthesis. All inhibited the DNA and RNA syntheses of isolated nuclei and luzopeptins B and C were slightly more active than luzopeptin A inhibition of RNA synthesis in isolated nuclei. All similarly inhibited DNA polymerase activity in vitro, but luzopeptin C was more active against RNA polymerase activity in vitro. It is concluded that luzopeptin cytotoxicity probably resulted from inhibition of DNA and RNA biosynthesis. Because all analogues bound avidly to isolated DNA and inhibited its functions, the differential anti-tumour activity may be attributed to the differential ability of these luzopeptins to traverse the cell membrane and, to some extent, other intracellular barriers. This probably is a result of difference in acetylation, resulting in differences in hydrophobicity.
Guinea pig lung membrane leukotriene D4 (LTD4) receptors were prelabeled with [3H]LTD4 and solubilized using digitonin, 3-[(3-cholamidopropyl)- dimethylammonio]-1-propane sulfonate, and other non-ionic, zwitterionic, and ionic detergents. [3H]LTD4 remains tightly associated with the receptor complex in the digitonin solubilized state. The dissociation rate of [3]LTD4 from the soluble receptor complex was increased in the presence of guanine nucleotides and sodium ions in a manner similar to that observed for the receptors in the membrane-bound state. The soluble [3H]LTD4 receptor complex was retained on wheat germ lectin affinity columns and destabilized by heat (40 +/- 4 degrees), trypsin, and chymotrypsin treatment, suggesting that the receptor is a glycoprotein. Size exclusion high pressure liquid chromatography of the soluble receptor complex showed that an apparent molecular weight of the soluble receptor complex, in the presence of digitonin, is in the range of 240,000-500,000. An approximately 20-fold enrichment of receptor-radioligand complex was achieved by passing the solubilized LTD4 receptor preparation successively through size exclusion and wheat germ lectin chromatography columns. These data provide the first step toward the purification and chemical characterization of LTD4 receptors.