Adriamycin: A proposal on the specificity of drug action

Adriamycin has been found to decrease the gel to liquid crystal transition temperature of liposomal membranes of varying compositions. However, when a low level of cardiolipin was inserted into a lecithin-containing membrane matrix, drug interaction caused the opposite effect on the thermal transition. It is suggested that this phenomenon may be indicative of specificity in the cytotoxic action of adriamycin on tumors, because evidence exists which indicates that certain neoplastic cells may contain cardiolipin in their plasma membrane and thus present a different surface to the drug than a non-malignant cell.     

Inhibition of plasma membrane NADH dehydrogenase by adriamycin and related anthracycline antibiotics

Doxorubicin (adriamycin) is cytotoxic to cells, but the biochemical basis for this effect is unknown, although intercalation with DNA has been proposed. This study suggests that the cytotoxicity of this drug may be due to inhibition of the plasma membrane redox system, which is involved in the control of cellular growth. Concentrations between 10^–6–10^–7 M adriamycin inhibit plasma membrane redox reactions >50%. AD32, a form of adriamycin which does not intercalate with DNA, but is cytotoxic, also inhibits the plasma membrane redox system. Thus, the cytotoxic effects of adriamycin, which limit its use as a drug, may be based on the inhibition of a transplasma membrane dehydrogenase involved in a plasma membrane redox system.     

Sensitization to the cytotoxicity of adriamycin by verapamil and heat in multidrug-resistant Chinese hamster ovary cells.

Development of multidrug resistance to anticancer agents is a major limitation for the success of cancer chemotherapy. The chemosensitizer verapamil increases intracellular accumulation of drugs such as adriamycin in certain multidrug-resistant cell lines. When combined with verapamil, hyperthermia should be able to alter membrane permeability to adriamycin and to enhance the cytotoxicity of the drug. Verapamil increased the cytotoxicity of adriamycin in multidrug-resistant Chinese hamster ovary cells (CH(R)C5) but not in drug-sensitive cells (AuxB1). Hyperthermia (42 degrees C) alone clearly increased the cytotoxicity of adriamycin in AuxB1 cells. There was also a small increase in CH(R)C5 cells at 42 and 43 degrees C. In drug-resistant cells, the cytotoxicity of adriamycin increased considerably when verapamil was combined with heat. This effect was dependent on temperature and increased with time of incubation. At 37 degrees C, verapamil increased the uptake of adriamycin in CH(R)C5 cells, while drug efflux decreased. When verapamil was combined with hyperthermia, drug efflux decreased even further. These results led to an overall increase in intracellular accumulation of the drug. In drug-sensitive cells, hyperthermia increased both the uptake and efflux of adriamycin, but verapamil had no effect. Verapamil plus heat increased the cytotoxicity of adriamycin in drug-resistant cells, and this was accompanied by altered permeability of the membrane to the drug. Hyperthermia combined with verapamil could be beneficial by increasing the effectiveness of adriamycin in the elimination of multidrug-resistant cells in a localized target region.     

The interaction of adriamycin with small unilamellar vesicle liposomes. A fluorescence study

The interaction of the antineoplastic agent adriamycin with sonicated liposomes composed of phosphatidylcholine alone and with small amounts (1-6%) of cardiolipin has been studied by fluorescence techniques. Equilibrium binding data show that the presence of cardiolipin increases the amount of drug bound to liposomes when the bilayer is below its phase transition temperature and when the ionic strength is relatively low (0.01 M). At higher ionic strength (0.15 M) and above the Tm (i.e. conditions which are closer to the physiological state) the binding of the drug to the two liposome types is nearly the same. Thus the differences in the interactions of adriamycin with cardiolipin-containing membranes, as opposed to those composed of phosphatidylcholine alone, are not due simply to increased binding but rather to an altered membrane structure when this lipid is present. Quenching of adriamycin fluorescence by iodide shows that bound drug is partially, but not completely, buried in the liposomal membrane. Both in the presence and absence of cardiolipin the bulk of the adriamycin is more accessible to the quencher below the Tm than above it; that is, a solid membrane tends to exclude the drug from deep penetration. Above the Tm, the presence of cardiolipin alters the nature of liposome-adriamycin interaction. Here the fluorescence quenching data suggest that the presence of small amounts of cardiolipin (3%) in a phosphatidylcholine matrix creates two types of binding environments for drug, one relatively exposed and the other more deeply buried in the membrane. The temperature dependence of the adriamycin fluorescence and the liposome light scattering reveal that cardiolipin alters the thermal properties of the bilayer as well as its interaction with adriamycin. At low ionic strength lateral phase separations may occur with both pure phosphatidylcholine and when 3% cardiolipin is present; under these conditions the bound adriamycin exists in two kinds of environment. It is notable that only adriamycin fluorescence reveals this phenomenon; thebulk property of liposome light scattering reports only on the overall membrane phase change. These data suggest that under certain conditions the drug binding sites in the membranes are decoupled from the bulk of the lipid bilayer.     

Effects of adriamycin on respiratory chain activities in mitochondria from rat liver, rat heart and bovine heart. Evidence for a preferential inhibition of complex III and IV

The inhibition of respiratory chain activities in rat liver, rat heart and bovine heart mitochondria by the anthracycline antibiotic adriamycin was measured in order to determine the adriamycin-sensitive sites. It appeared that complex III and IV are efficiently affected such that their activities were reduced to 50% of control values at 175 +/- 25 microM adriamycin. Complex I displayed a minor sensitivity to the drug. Of the complex-I-related activities tested, only duroquinone oxidation appeared sensitive (50% inhibition at approx. 450 microM adriamycin). Electron-transfer activities catalyzed by complex II remained essentially unaltered up to high drug concentrations. Of the activities measured for this complex, only duroquinone oxidation was significantly affected. However, the adriamycin concentration required to reduce this activity to 50% exceeded 1 mM. Mitochondria isolated from rat liver, rat heart and bovine heart behaved essentially identical in their response to adriamycin. These data support the conclusion that, in these three mitochondrial systems, the major drug-sensitive sites lie in complex III and IV. Cytochrome c oxidase and succinate oxidase activity in whole mitochondria exhibited a similar sensitivity towards adriamycin, as inner membrane ghosts, suggesting that the drug has direct access to its inner membrane target sites irrespective of the presence of the outer membrane. By measuring NADH and succinate oxidase activities in the presence of exogenously added cytochrome c, it appeared that adriamycin was less inhibitory under these conditions. This suggests that adriamycin competes with cytochrome c for binding to the same site on the inner membrane, presumably cardiolipin.     

Effect of adriamycin entrapped by sulfatide-containing liposomes on ovarian tumor-bearing nude mice

Sulfatide-containing liposomes showed the highest degree of adriamycin entrapment of all the liposomes tested. Adriamycin was bound to the sulfatide anions on the liposomal membrane, inserted into the membrane, and incorporated into the aqueous compartment of the vesicle. Liposome-entrapped adriamycin was maintained at a much higher blood level than free adriamycin, and reached a lower concentration in the heart than did the free drug, which might lead to lower cardiotoxicity of the drug. Incorporation of adriamycin into ovarian tumor transplanted into nude mice was increased when entrapped by the sulfatide-containing liposomes. Liposome-entrapped adriamycin did not induce the drastic loss of body weight which occurred with the free drug. The growth of ovarian tumor was inhibited by liposome-entrapped adriamycin to the same degree as free adriamycin. Having these advantages, sulfatide-containing liposomes could be useful carriers of adriamycin for cancer chemotherapy.     

The effect of bispecific monoclonal antibody recognizing both hepatoma-specific membrane glycoprotein and anthracycline drugs on the metastatic growth of hepatoma AH66

A monoclonal mouse antibody (MoHG) was produced using in vitro cultured AH66R tumor cells treated with cholesteryl hemisuccinate as an immunogen. The antibody identified a 90 kd membrane glycoprotein (HG-90) which is expressed on in vitro cultured hepatoma cell lines AH66 and AH66R. A monoclonal antibody was prepared to the anthracycline drug daunomycin, and it also reacted with adriamycin. A fusion was made of the hybridoma HG-90 with the hybridoma which recognized daunomycin/adriamycin. This bispecific hybridoma A8C recognized both determinants. We studied the therapeutic effect of the A8C bispecific antibody with adriamycin treatment and compared it to the effect of the bispecific antibody to which adriamycin had been conjugated via an albumin (Alb) bridge. The therapy model used was the tumor AH66R in Donryu rats. Tumor bearing rats had their subcutaneous tumors resected on day 10, a time when distant metastases were present. After the surgical resection of the tumor the rats were injected intravenously for two cycles with the bispecific antibodies, followed by the administration of adriamycin (ADR) or MoHG.Alb.ADR conjugates. A slight therapeutic effect occurred with either MoHG or ADR alone but treatment with the bispecific antibody followed by the administration of ADR or with the MoHG.Alb.ADR conjugates significantly prolonged survival, with 60% of the treated animals being "tumor free" when sacrificed on day 80. Lower serum concentrations of alphafetoprotein were observed with the bispecific antibody and drug treatment. This suggests that the bispecific antibody/drug treatment is potentially more beneficial in the suppression of distant metastases than the MoHG.Alb.ADR conjugate. This may be due to an increase in the local drug concentration of unmodified adriamycin.     

The interaction of adriamycin with cardiolipin in model and rat liver mitochondrial membranes

The interaction of adriamycin with cardiolipin in model membranes and in various membrane preparations derived from rat liver mitochondria was studied and the results are analyzed in the light of a possible specific interaction between adriamycin and cardiolipin. It was found that adriamycin binds to cardiolipin-containing model membranes with a fixed stoichiometry of two drug molecules per cardiolipin. Furthermore, the extent of drug complexation by mitochondria and mitoplasts (inner membrane plus matrix) is in reasonable agreement with their cardiolipin content. In contrast, adriamycin-binding curves of inner membrane ghosts and submitochondrial particles reveal considerable association to an additional site, presumably RNA. The evidence for the potential importance of RNA as a target comes from experiments on outer membranes and microsomes which both appear to bind substantial amounts of adriamycin. Removal of the major part of the RNA associated with these fractions by EDTA treatment is accompanied by a dramatic reduction of binding capacity. We propose that endogenous RNA present in mitochondria and mitoplasts is not accessible for adriamycin at low concentrations of the drug due to the presence of an intact lipid barrier. This potential site comes to expression in ghosts and submitochondrial particles, due to the absence of an intact lipid bilayer and due to the inside-out orientation of the limiting membrane, respectively. Electron microscopical studies show that adriamycin induces dramatic changes in mitochondrial morphology, similar to the uncoupler-induced effects described by Knoll and Brdiczka (Biochim. Biophys. Acta 733, 102-110 (1983). Adriamycin has an uncoupling effect on mitochondrial respiration and oxidative phosphorylation. The concentration dependence of this effect correlates with the adriamycin-binding curve for mitochondria which implies that only bound adriamycin actively inhibits respiration.     

Interaction of carnitine with mitochondrial cardiolipin.

The physiological role of L-carnitine is to determine the transport of acyl-CoA through the mitochondrial membrane. However, some observations may also suggest a direct effect of the molecule per se on the physical properties of the membrane, most probably at the level of the binding site. This possibility has been investigated by studying the influence of adriamycin, a drug that binds to cardiolipin, on the effect of carnitine on isolated rat liver mitochondria. It has been found that adriamycin almost abolishes the activating effect of carnitine on state 2 respiration. The effect and its inhibition is seen by using either the L-form of carnitine or the D-form or both. Cardiolipin removes the effect of adriamycin and restores the activation by carnitine. It is proposed that some effects of carnitine on mitochondrial properties may be the result of interaction of carnitine with cardiolipin at the membrane level.     

Localization of adriamycin in model and natural membranes. Influence of lipid molecular packing

The interaction of adriamycin with lipids was studied in model (monolayers, small unilamellar vesicles, large multilamellar vesicles) and natural (chinese hamster ovary cell) membranes by measurement of fluorescence energy transfer and fluorescence quenching. 2-APam, 7-ASte, 12-ASte and anthracene-phosphatidylcholine were used as fluorescent probes in which the anthracene group is well located at graded depths in the membrane. Egg-yolk phosphatidylcholine and a 1/1 mixture of it with bovine brain phosphatidylserine were used in model membrane systems. Large fluorescence energy transfer was observed between these molecules as donors and the drug as acceptor. With liposomes, at pH 7.4 and over an adriamycin concentration range of 0-100 microM, the efficiency of energy transfer was 12-ASte greater than 7-ASte greater than 2-APam, with 100% energy transfer for 12-ASte above a drug concentration of 30 microM. At pH 5, where the fatty acids are buried deeper (0.45 nm) in the lipid bilayer due to protonation of the carboxyl group, the order of energy transfer 7-ASTe greater than 12-ASte = 2-APam was observed. Measurements of fluorescence quenching using the non-permeant Cu2+ ion as quencher and spectrophotometric assays indicated that around 40% of the adriamycin molecules were deeply embedded in the lipid bilayer. Adriamycin molecules thus appear to penetrate the lipid bilayer, with the aminoglycosyl group interacting with the lipid phosphate groups and the dihydroanthraquinone residue in contact with the lipid fatty acid chains. In contrast, fluorescence energy transfer and quenching studies on CHO cells showed that adriamycin penetrated the plasma membrane of these cells to a much more limited extent than in the model membrane systems. This can be related to the squeezing out of the drug from a film of phosphatidylcholine which was observed in monolayers by means of surface pressure, potential and fluorescence experiments. These observations indicated that the penetration of adriamycin into lipid bilayers strongly depends on the molecular packing of the lipid.     

Uptake of antineoplastic agents into large unilamellar vesicles in response to a membrane potential

Many drugs exhibit lipophilic and cationic (basic) characteristics. Previous studies have shown that lipophilic cations can be accumulated into model membrane 'liposomal' (vesicular) systems in response to establishing a membrane potential (inside negative) across the vesicle membrane. We demonstrate here that the anticancer drugs, adriamycin and vinblastine, can be rapidly accumulated into egg phosphatidylcholine large unilamellar vesicles in response to a valinomycin-dependent K+ diffusion potential (delta psi) to achieve high effective interior concentrations. Further, trapping efficiencies approaching 100% can be easily achieved. The influence of lipid composition and the requirement for valinomycin have been examined for adriamycin. Equimolar cholesterol levels inhibit the uptake process at 20 degrees C. However, incubation at higher temperature results in enhanced uptake. Similarly, the presence of egg phosphatidylserine or incubation at elevated temperatures results in significant adriamycin uptake in the absence of valinomycin. It is shown that the adriamycin retention time in the vesicles is enhanced by an order of magnitude or more when actively trapped by the presence of a membrane potential in comparison to passive trapping procedures. It is suggested that such active trapping procedures may be of use for loading liposomal systems for drug delivery applications, and may provide avenues for controlled release of encapsulated material.     

Adriamycin induced changes in translocation of sodium ions in transporting epithelial cells.

The antitumor, antibiotic, adriamycin stimulates the net transport of sodium ions across frog skin epithelium under short circuit conditions. This stimulation is largely independent of Ca⁺⁺ concentration in the media or of previous treatment of the epithelium with amiloride, ouabain, and vasopressin. We believe adriamycin induces changes in membrane permeability to sodium ions and that such changes may explain, in part, the cardiotoxicity of this drug.     

Is the Drug-Responsive NADH Oxidase of the Cancer Cell Plasma Membrane a Molecular Target for Adriamycin?

Enhanced growth inhibition and antitumor responses to adriamycin have been observed repeatedly from several laboratories using impermeant forms of adriamycin where entry into the cell was greatly reduced or prevented. Our laboratory has described an NADH oxidase activity at the external surface of plasma membrane vesicles from tumor cells where inhibition by an antitumor sulfonylurea, N-(4-methylphenylsulfonyl)-N-(4-chlorophenyl)urea (LY181984), and by the vanilloid, capsaicin (8-methyl-N-vanillyl-6-noneamide) correlated with inhibition of growth. Here we report that the oxidation of NADH by isolated plasma membrane vesicles was inhibited, as well, by adriamycin. An external site of inhibition was indicated from studies where impermeant adriamycin conjugates were used. The EC₅₀ for inhibition of the oxidase of rat hepatoma plasma membranes by adriamycin was several orders of magnitude less than that for rat liver. Adriamycin cross-linked to diferric transferrin and other impermeant supports also was effective in inhibition of NADH oxidation by isolated plasma membrane vesicles and in inhibition of growth of cultured cells. The findings suggest the NADH oxidase of the plasma membrane as a growth-related adriamycin target at the surface of cancer cells responsive to adriamycin. Whereas DNA intercalation remains clearly one of the principal bases for the cytotoxic action of free adriamycin, this second site, possibly related to a more specific antitumor action, may be helpful in understanding the enhanced efficacy reported previously for immobilized adriamycin forms compared to free adriamycin.     

Evidence that adriamycin resistance in Chinese hamster lung cells is regulated by phosphorylation of a plasma membrane glycoprotein

Incubation of adriamycin resistant Chinese hamster lung cells with low levels of N-ethylmaleimide (NEM) results in a major increase in the cellular accumulation of drug. When resistant cells are prelabeled with [³²Pi] and thereafter treated with NEM there also occurs a selective superphosphorylation of an 180K plasma membrane glycoprotein (P-180). This phosphorylation reaction occurs at both serine and threonine residues. In similar experiments with drug sensitive cells only minor levels of this protein can be detected. Detailed studies have established that in cells which have reverted to drug sensitivity there is a parallel loss in the presence of phosphorylated P-180. Also in cells which have undergone partial reversion to drug sensitivity there is a correlation between levels of superphosphorylated P-180 and adriamycin resistance. These results provide evidence that adriamycin resistance is dependent on the presence of P-180. The results also suggest that the biological activity of this protein is highly regulated by phosphorylation and that in the superphosphorylated state P-180 is inactive and under these conditions the resistant cell is converted to a drug sensitive phenotype.     

Purification of the 170- to 180-kilodalton membrane glycoprotein associated with multidrug resistance. 170- to 180-kilodalton membrane glycoprotein is an ATPase

170-180-kDa membrane glycoprotein (P-glycoprotein) associated with multidrug resistance is involved in drug transport mechanisms across the plasma membrane of resistant cells. From sequence analysis of cDNAs of the P-glycoprotein gene, it is postulated that the active drug-efflux pump function may be attributable to the protein. However, purification of the P-glycoprotein while preserving its enzymatic activity has not been reported. In this study, we have purified the P-glycoprotein from the human myelogenous leukemia K562 cell line resistant to adriamycin (K562/ADM) by means of one-step immunoaffinity chromatography using a monoclonal antibody against P-glycoprotein. The procedure was simple and efficiently yielded an electrophoretically homogeneous P-glycoprotein sample. By solubilization with 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate, the purified P-glycoprotein was found to have ATPase activity. This ATP hydrolysis may be coupled with the active efflux of anticancer drugs across the plasma membrane of multidrug-resistant cells.     

Transferrin-adriamycin conjugates which inhibit tumor cell proliferation without interaction with DNA inhibit plasma membrane oxidoreductase and proton release in K562 cells.

Conjugates of adriamycin crosslinked to transferrin with glutaraldehyde inhibit proliferation of transformed cells. Conjugates of this type inhibit oxidoreductase activity in the plasma membrane of K562 cells, and the inhibition of electron transport is found at concentrations ten times lower than concentrations of free adriamycin which inhibit electron transport and cell growth. The transferrin-adriamycin conjugate inhibits ferricyanide reduction, diferric transferrin reduction and plasma membrane NADH oxidase activity stimulated by transferrin. Activation of proton release from the K562 cells by diferric transferrin also is inhibited by the conjugate, and conjugate kills cells more effectively than free adriamycin. Since the conjugate does not transfer adriamycin to the nucleus, the growth control may be based on inhibition of the transferrin regulated redox system and Na+/H+ antiport activity at the plasma membrane.     

Diferric transferrin reductase in the plasma membrane is inhibited by adriamycin

Diferric transferrin which is often necessary for growth of cells is reduced by the transplasma membrane electron transport system of HeLa cells with release of ferrous iron outside the cell. Reduction of external diferric transferrin is reflected in oxidation of internal NADH. Adriamycin, an antitumor drug, inhibits diferric transferrin reduction by the HeLa cells and inhibits concomittant oxidation of cytosolic NADH at concentrations, 10(-8)-10(-6)M, which inhibit cell growth. Isolated liver plasma membranes have an NADH diferric transferrin reductase activity which is inhibited by similar adriamycin concentrations. We propose that inhibition of cell growth by adriamycin can be based on inhibition of transplasmalemma diferric transferrin reductase.     

Adriamycin transport and sensitivity in fatty acid-modified leukemia cells

The membrane phospholipids of L1210 murine leukemia cells were modified by supplementing the growth medium with micromolar concentrations of polyunsaturated or monounsaturated fatty acids. This procedure results in enrichment of cellular phospholipids by the supplemented fatty acid. Enrichment with polyunsaturated fatty acids resulted in a marked increase in sensitivity to adriamycin as compared to enrichment with monounsaturated fatty acids. The increased cytotoxicity was directly proportional to the extent of unsaturation of the inserted fatty acid, but there was no difference in cells enriched with n-3 compared with n-6 family fatty acids. To explore the mechanism of this observation, we examined whether augmented uptake of the drug might explain the increased cytotoxicity. The uptake of [14C]adriamycin, which was approximately linear at later time points, was only partially temperature dependent and never reached a steady state. Initial uptake at time points prior to 60 s could not be measured due to high and variable rapid membrane adsorption. Cellular accumulation of drug was greater in the docosahexaenoate 22:6-enriched L1210 cells as compared to oleate 18:1-enriched cells and was about 32% greater after 20 min. When L1210 cells were enriched with six fatty acids of variable degrees of unsaturation, the accumulation of adriamycin was directly correlated with the average number of double bonds in the fatty acids contained in cellular phospholipids. There was no difference in efflux of drug from cells pre-loaded with adriamycin. We conclude that the greater accumulation of adriamycin by the polyunsaturated fatty acid-enriched L1210 cells likely explains the increased sensitivity of these cells to adriamycin compared to 18:1-enriched cells.     

Free radicals and anticancer drug resistance: Oxygen free radicals in the mechanisms of drug cytotoxicity and resistance by certain tumors

Certain anticancer agents form free radical intermediates during enzymatic activation. Recent studies have indicated that free radicals generated from adriamycin and mitomycin C may play a critical role in their toxicity to human tumor cells. Furthermore, it is becoming increasingly apparent that reduced drug activation and or enhanced detoxification of reactive oxygen species may be related to the resistance to these anticancer agents by certain tumor cell lines. The purposes of this review are to summarize the evidence pointing toward the significance of free radicals formation in drug toxicity and to evaluate the role of decreased free radical formation and enhanced free radical scavenging and detoxification in the development of anticancer drug resistance by a spectrum of tumor cell types. Studies failing to support the participation of oxyradicals in the cytotoxicity and resistance of adriamycin are also discussed.