Relationship between activity and amino sugar stereochemistry of daunorubicin and adriamycin derivatives.

The effects of 4'-epi-daunorubicin, 4'-epi-adriamycin, and the corresponding beta anomers on the in vitro activity of Escherichia coli DNA polymerase I and RNA polymerase were determined and compared with the effects of the parent compounds. The observed effects parallel the cytotoxic activities, assayed by inhibition of mouse embryo fibroblast proliferation, and the inhibitory activities on DNA synthesis in cultured cells. The data indicate that the inverted configuration at position 1 of the amino sugar results in a markedly reduced biological activity. This conclusion is also substantiated by the data obtained with the beta anomer of adriamycin. A preliminary investigation on the binding properties of these derivatives suggests that the inverted configuration at C-1' produces a significant decrease in the binding to DNA. In contrast, epimerization at position 4' did not produce any significant change in activity. The relationship between biological and biochemical activity and DNA binding properties of the tested compounds are discussed with particularly reference to antitumor activity.

A comparison of the effects of daunomycin and adriamycin on various DNA polymerases.

The effects of the anthracycline antiboties, daunomycin and adriamycin, on the DNA-directed activities of DNA polymerases from murine sarcoma virus, rat liver (high-molecular-weight species), Escherichia coli, and Micrococcus luteus were determined. Under all conditions tested, these compounds had greater inhibitory effect against the viral polymerase than against cellular polymerase. The inhibition of murine sarcoma virus DNA polymerase by daunomycin was competitive with respect to DNA. For viral DNA polymerase it was concluded that the inhibition was predominatly caused by the interaction of duanomycin with the primer-template DNA. Also, an appreciable reversal of the daunomycin-induced inhibition of this polymerase by an increase in Mg-2+ concentration is consistent with the conclusion derived by competition experiments. In contrast, the inhibition of both rat liver and M. luteus DNA polymerases was essentially noncompetitive with DNA. Also, bacterial enzymes wer e less sensitive to inhibition by these drugs than the virion polymerase. The strong and preferential inhibiton of viral DNA polymerase is discussed in relation to a differential sensitivity of normal as compared to tumor cells observed in some cell lines.

Identification of the product of two oncogenic rearranged forms of the RET proto-oncogene in papillary thyroid carcinomas.

In papillary thyroid carcinomas, we have identified two tumor-specific rearrangements of the RET proto-oncogene leading to the formation of different transforming fusion products sharing the tyrosine kinase (tk) domain of the proto-oncogene and designated ptc-1 and ptc-2. We have analysed ptc-1 and ptc-2 products by immunoprecipitation with specific anti-RET antibodies followed by immunoblotting with the same reagent or with antibodies specific for phosphotyrosine (P-tyr) residues. The anti-RET antibodies were reactive with 64-kDa (p64ptc-1) and 81-kDa (p81ptc-2) proteins from lysates of ptc-1 and ptc-2 transformed cells, respectively, and identified two proteins of 140 kDa and 160 kDa from extracts of SK-N-SH, a neuroblastoma cell line previously shown to express two differently glycosylated forms of the normal RET product. The anti P-tyr antibodies, while detecting the same p64ptc-1 and p81ptc-2 proteins from ptc-1 and ptc-2 extracts, did not show any specific band in the neuroblastoma lysates. An additional set of experiments led us to conclude that, whereas the normal product of the RET proto-oncogene is a membrane-associated receptor-like molecule not intrinsically phosphorylated on tyrosine, both oncogenic forms of RET, ptc-1 and ptc-2, are constitutively phosphorylated on tyrosine, display an 'in vitro' autophosphorylation activity, are translocated from the membrane to the cytoplasm and are apparently unaffected by protein kinase C modulation.

The interaction of daunorubicin and doxorubicin with DNA and chromatin.

Isotherms that describe the binding of anthracycline antibiotics (including daunorubicin and doxorubicin (adriamycin)) to calf thymus DNA and chromatin have been obtained by means of fluorescence measurements. As expected for charged ligands, the association constants for the interaction of all drugs examined with DNA were found to be dependent on the ionic strength. However, in the case of the daunorubicin-DNA interaction, a marked decrease in the number of binding sites was also observed when the ionic strength was increased. It is suggested that the effect of salt concentration on the number of potential binding sites of daunorubicin molecules to DNA may be the result of some salt-induced alterations in the DNA conformation. This interpretation is also supported by binding data obtained with calf thymus chromatin; Whereas at low salt concentration the binding parameters for the doxorubicin-chromatin interaction are similar to those expected by neutralization of the phosphate groups by histones, modifications of the DNA structure in chromatin are invoked to account for the reduction and heterogeneity of daunorubicin binding sites. The side chain at C-9 could play an important role in determining the strength and specificity of the anthracycline-DNA interaction.

The interaction of adriamycin and its beta anomer with DNA.

The results of thermal denaturation, fluorescence, calorimetric and viscosimetric studies on the interaction of adriamycin and its beta anomer with DNA are reported. Whereas all equilibrium, hydrodynamic and thermodynamic measurements are consistent with the proposed intercalative binding model for the adriamycin-DNA complex, the binding mechanism for the reaction of the beta anomer with DNA remains uncertaian. All DNA binding properties of this stereoisomer are substantially different from those of the parent compound. The results suggest that the amino sugar residue of the natural antibiotic may interact stero-specifically with the DNA helix, thus dictating the orientation of the tetracvclic chromophore within the intercalation site. The alteration in the DNA binding capacity and the changes in interactions with DNA following in inversion of configuration at C-1', parallel a lack of biological activity observed for the beta anomer.

Synthesis of alkyl chain-modified ether lipids and evaluation of their in vitro cytotoxicity.

A series of alkyl lysophospholipid (ALP) analogs of ET-18-OCH3 (1-O-octadecyl-2-O-methyl-rac-glycero-3-phosphocholine) containing modifications in the long C-1 chain has been synthesized and evaluated in human tumor cell line cytotoxicity assays. The compounds have also been evaluated in platelet activating factor (PAF) receptor agonism and hemolysis tests. Two modifications have been studied, introduction of a carbonyl group at different positions of the C-1 chain and branching of this chain, in some compounds with incorporation of a phenyl group. Several compounds showed a cytotoxic potency comparable to that of the reference compound ET-18-OCH3, associated with reduced proaggregating and hemolytic effects. The two enantiomers of 1-O-(7-oxooctadecyl)-2-O-methyl-rac-glycero-3-phosphocholine (2) showed the same level of cytotoxicity or antiproliferative activity, with the PAF-agonistic effect confined to R-2. The very low stereoselectivity found in the in vitro cytotoxicity confirms earlier results and indicates a lack of stereospecific interactions with a macromolecular target.

Effect of modulation of protein kinase C activity on cisplatin cytotoxicity in cisplatin-resistant and cisplatin-sensitive human osteosarcoma cells.

The effect of modulation of protein kinase C (PKC) activity by 12-O-tetradecanoylphorbol-13-acetate (TPA) on cisplatin cytotoxicity was examined in a human osteosarcoma U2-OS cell line and in a U2-OS variant (U2-OS/Pt) selected after continuous exposure to increasing concentrations of cisplatin. U2-OS/Pt cells showed a 7.5-fold resistance to the drug. A 24 h exposure of cells to TPA caused a potentiation of cisplatin cytotoxicity in sensitive and in resistant cells; under these conditions, PKC activity was shown to be down-regulated. In contrast, a short-term exposure of cells to TPA did not affect cisplatin cytotoxicity in U2-OS or in U2-OS/Pt cells. These results support the involvement of PKC in cellular response to cisplatin. However, this enzyme is probably not directly implicated in the mechanisms of acquired resistance in this cell system.

Changes in the three-dimensional organization of LoVo cells associated with resistance to doxorubicin.

LoVo cells and a derived subline resistant to doxorubicin were compared in the spheroids system. The resistant line, unlike the parent one, was unable to grow as spheroids, but formed irregular loose aggregates. Moreover treatment of the resistant cells with membrane-active agents able to reverse pleiotropic drug resistance had no effect on the capability of these cells to grow as spheroids. The results indicate that the inability of resistant cells to form spheroids is not related to the resistance mechanism.

Diversity of effects of two antitumor anthracycline analogs on the pathway of activation of PKC in intact human platelets.

Two antitumor antibiotics doxorubicin and daunorubicin were tested for their ability to influence the activation of protein kinase C in human platelets. Daunorubicin was found to inhibit the phosphorylation of the 40 K PKC substrate induced by thrombin and 12-O-tetradecanoyl-phorbol-13-acetate as well as the phosphorylation of the 20 K protein induced by thrombin. The serotonin release associated to these phosphorylative events was also inhibited by daunorubicin. In contrast the effects of doxorubicin, though inhibitory on the release reaction, were always stimulatory of the phosphorylations. Doxorubicin alone was able to induce the phosphorylation of both 40 K and 20 K phosphoproteins in a concentration-dependent manner. Whereas the stimulation by doxorubicin was not influenced by pretreatment with dibutyryl-cyclic-AMP which inhibits the effects of thrombin, this effect was inhibited by daunorubicin, neomycin and stimulated by the diacylglycerol-kinase inhibitor R 59 022. It is proposed that doxorubicin activates the protein kinase C by causing the breakdown of phosphoinositides.

Lipid peroxidation, phosphoinositide turnover and protein kinase C activation in human platelets treated with anthracyclines and their complexes with Fe(III).

The effects of the antitumor drugs daunorubicin, doxorubicin and their complexes with Fe(III) on phosphoinositide hydrolysis, lipid peroxidation and protein kinase C (PKC) activation were measured in intact human platelets. Doxorubicin and the Fe(III) complexes of both doxorubicin and daunorubicin quickly induced lipid peroxidation [as measured by the thiobarbituric acid (TBA) assay], phosphorylation of the 40 K substance of PKC, and increased levels of phosphatidic acid and inositol phosphates. Fe(III) alone or complexed to acetohydroxamic acid induced high levels of TBA-reactive material but did not affect either PKC activation or phosphoinositide turnover. In contrast, daunorubicin, which was ineffective per se, inhibited all these doxorubicin- and anthracyclines/Fe(III)-induced biochemical events. We suggest that phosphoinositide hydrolysis determined by anthracyclines, and consequently PKC activation, could be due to lipid peroxidation, thus triggering the activity of phospholipase C.

Clavilactones, a novel class of tyrosine kinase inhibitors of fungal origin.

Targeting of deregulated protein tyrosine kinases has been proposed as a new approach in the therapeutic intervention against pathological processes including proliferative disorders and cancer. Using a screening approach based on a comparative evaluation of antiproliferative effects in a panel of tumor cells with differential expression of protein tyrosine kinases, three benzoquinoid macrolidic fungal metabolites produced by Clitocybe clavipes, clavilactones A, B, and D (CA, CB, and CD) and two semisynthetic derivatives of these products, diacetyl-CA and dimethyl-CA, were identified as inhibitors of protein tyrosine kinases. Naturally occurring CA, CB, and CD showed inhibitory activity in kinase assays against the Ret/ptc1 and epidermal growth factor receptor (EGF-R) tyrosine kinases, while being less effective against the v-Abl tyrosine kinase and p34(cdc2) serine/threonine kinase (IC(50) 2.8, 5.5, 81.3, and 128 microM respectively, for the most potent compound CD). CB was shown to be a non-competitive inhibitor of EGF-R with respect to ATP or poly(Glu(6)Ala(3)Tyr). CD also preferentially inhibited the growth of A431 cells, which overexpress a constitutively active EGF-R, as opposed to IGROV-1 and SKOV-3 cells, which express low levels of the receptor. Further, EGF-R was shown to be a target for clavilactones in A431 cells, since EGF-induced receptor autophosphorylation was inhibited in the presence of CB, CD, and diacetyl-CA. Both CD and diacetyl-CA displayed weak activity when administered daily (i.p.) to mice bearing ascitic A431 tumor. These findings indicate that clavilactones represent the prototypes of a new structural class of tyrosine kinase inhibitors deserving further investigation.

Decreased drug accumulation and increased tolerance to DNA damage in tumor cells with a low level of cisplatin resistance.

In an attempt to examine the cellular changes associated with cisplatin resistance, we selected a cisplatin-resistant (A43 1/Pt) human cervix squamous cell carcinoma cell line following continuous in vitro drug exposure. The resistant subline was characterized by a 2.5-fold degree of resistance. In particular, we investigated the expression of cellular defence systems and other cellular factors probably involved in dealing with cisplatin-induced DNA damage. Resistant cells exhibited decreased platinum accumulation and reduced levels of DNA-bound platinum and interstrand cross-link frequency after short-term drug exposure. Analysis of the effect of cisplatin on cell cycle progression revealed a cisplatin-induced G2M arrest in sensitive and resistant cells. Interestingly, a slowdown in S-phase transit was found in A431/Pt cells. A comparison of the ability of sensitive and resistant cells to repair drug-induced DNA damage suggested that resistant cells were able to tolerate higher levels of cisplatin-induced DNA damage than their parental counterparts. Analysis of the expression of proteins involved in DNA mismatch repair showed a decreased level of MSH2 in resistant cells. Since MSH2 seems to be involved in recognition of drug-induced DNA damage, this change may account for the increased tolerance to DNA damage observed in the resistant subline. In conclusion, the involvement of accumulation defects and the increased tolerance to cisplatin-induced DNA damage in these cisplatin-resistant cells support the notion that multiple changes contribute to confer a low level of cisplatin resistance.

Lack of effect of glutathione depletion on cytotoxicity, mutagenicity and DNA damage produced by doxorubicin in cultured cells.

Since endogenous glutathione (GSH), the main non-protein intracellular thiol compound, is known to provide protection against reactive radical species, its depletion by diethylmaleate (DEM) was used to assess the role of free radical formation mediated by doxorubicin in DNA damage, cytotoxicity and mutagenicity of the anthracycline. Subtoxic concentrations of DEM that produced up to 75% depletion of GSH did not increase doxorubicin cytotoxicity in a variety of cell lines, including Chinese hamster ovary (CHO) and lung (V-79) cells, LoVo human carcinoma cells and P388 murine leukemia cells. Similarly, the number of doxorubicin-induced DNA single strand breaks in CHO cells and the mutation frequency in V-79 cells were not affected by GSH depletion. The results obtained suggest that mechanisms other than free radical formation are responsible for DNA damage, cytotoxicity and mutagenicity of anthracyclines.

Protein kinase C activation and lipid peroxidation by doxorubicin analogues.

Several doxorubicin analogues have been tested for their capacity to activate protein kinase C (PKC) and to induce lipid peroxidation in intact human platelets. Only doxorubicin and 4'-iodo-doxorubicin were able to induce lipid peroxidation and PKC activation the first being the most effective. N-acetyldoxorubicin, N-trifluoroacetyl-doxorubicin-14-valerate (AD32) and doxorubicin-14-propionate were not effective on either event. This correlation supports that PKC activation in human platelets by doxorubicin is mediated by lipid peroxidation and suggests that the effect is specific for anthracyclines with a doxorubicin aglycone and a free charged amino group in the sugar moiety. The results stress the new action of anthracyclines, whose pharmacologic implications are presently under investigation on nucleated cells.

A differential interaction of doxorubicin and daunorubicin with human serum proteins.

The results of a comparative investigation on the interaction of doxorubicin (adriamycin) and daunorubicin with serum proteins are reported. Whereas a strong interaction occurs in vitro between doxorubicin and human serum proteins, no appreciable binding to proteins could be detected for daunorubicin under similar experimental conditions. Since the protein-bound drug is only partially dissociated by physical procedures including gel-electrophoresis, column-chromatography and solvent extraction, the formation of a covalent bond is suggested. The doxorubicin binding to serum proteins is apparently nonselective for a class of proteins; it is strongly reduced in acid conditions and slightly dependent on the ionic strength. Two tentative reaction mechanisms have been considered.

Biologic activity of daunorubicin linked to proteins via the methylketone side chain.

New daunorubicin-protein conjugates were prepared by covalently linking the antitumor drug to various test proteins via its methylketone side chain. Attachment of daunorubicin to proteins was achieved by nucleophylic substitution reaction of the 14-bromo derivative of the drug, under mild coupling conditions. In contrast to conventional methods, this procedure did not involve reaction or modification of the amino sugar. As expected, the covalent linkage of the drug was generally associated with an appreciable reduction in the drug cytotoxicity to HeLa cells in vitro. The possible advantages of this method for coupling to specific protein carriers are discussed.

Discovery of 342 putative new genes from the analysis of 5'-end-sequenced full-length-enriched cDNA human transcripts.

In this work we describe the process that, starting with the production of human full-length-enriched cDNA libraries using the CAP-Trapper method, led us to the discovery of 342 putative new human genes. Twenty-three thousand full-length-enriched clones, obtained from various cell lines and tissues in different developmental stages, were 5'-end sequenced, allowing the identification of a pool of 5300 unique cDNAs. By comparing these sequences to various human and vertebrate nucleotide databases we found that about 40% of our clones extended previously annotated 5' ends, 662 clones were likely to represent splice variants of known genes, and finally 342 clones remained unknown, with no or poor functional annotation. cDNA-microarray gene expression analysis showed that 260 of 342 unknown clones are expressed in at least one cell line and/or tissue. Further analysis of their sequences and the corresponding genomic locations allowed us to conclude that most of them represent potential novel genes, with only a small fraction having protein-coding potential.