Cytotoxicity, cellular uptake and DNA damage by daunorubicin and its new analogues with modified daunosamine moiety.

Daunorubicin (DRB) and its two analogues containing a trisubstituted amidino group at the C-3' position of the daunosamine moiety have been compared regarding their cytotoxic activity, cellular uptake, subcellular localization and DNA damaging properties. An analogue containing in the amidino group a morpholine moiety (DRBM) as well as an analogue with a hexamethyleneimine moiety (DRBH), tested against cultured L1210 cells, exhibited lower cytotoxicity then DRB. The decrease of cytotoxic activity was not related to cellular uptake and subcellular localization of drugs. Although all tested drugs were active in the induction of DNA breaks and DNA-protein crosslinks, they differed in the mechanism of induction of DNA lesions. DRB produced DNA breaks mediated solely by topoisomerase II, whereas DRBM and DRBH induced two types of DNA breaks by two separate processes. The first is related to the inhibition of topoisomerase II and the second presumably reflects a covalent binding of drug metabolites to DNA. It is hypothesized that the replacement of the primary amino group (-NH(2)) at the C-3' position of the daunosamine moiety by a trisubstituted amidino group (-N=CH-NRR) may be a route to the synthesis of anthracycline derivatives with enhanced ability to form covalent adducts to DNA.

Effects of anticancer drugs on transcription in vitro.

The effects of DNA interacting drugs on: (1) total RNA synthesis catalyzed by E. coli and T7 RNA polymerase; (2) synthesis of the initiating dinucleotide (pppApU) by E. coli RNA polymerase ("abortive initiation"); (3) elongation of RNA chains synthesized by T7 RNA polymerase on pT7-7 plasmid DNA bearing T7 RNA polymerase promoter phi 10 with human Cu/Zn superoxide dismutase coding sequence, (4) interaction of transcription factor Sp1 and its binding site were studied. Intercalating ligands which form quickly dissociating complexes with DNA (anthracyclines, proflavine, ethidium bromide) are compared with the slowly dissociating drug of d(G x C) specificity (actinomycin D), the non-intercalating, d(A x T) specific pyrrole antibiotics (netropsin and distamycin A) and covalently binding to DNA 1-nitroacridine derivative (nitracrine). The obtained results indicate that rapidly dissociating ligands, proflavine and ethidium bromide, inhibit total RNA synthesis in vitro and the abortive initiation to a similar extent while they do not induce discrete elongation stops of RNA polymerase. Actinomycin D and nitracrine exhibit a high inhibitory effect on total RNA synthesis and induce stops of RNA polymerase while not affecting abortive initiation. Pyrrole antibiotics primarily inhibit the initiation, while no elongation stops are induced. Actinomycin D inhibits complex formation between nuclear proteins and the Sp1 binding site. Netropsin, ethidium bromide, proflavine and other intercalating acridines do not affect Sp1 binding. The results indicate that the effects primarily depend on sequence specificity and secondarily on the dissociation rate of ligands from their complexes with DNA.

Inhibition of RNA synthesis in vitro and cell growth by anthracycline antibiotics.

New derivatives of doxorubicin and daunorubicin with amidine group bonded to daunosamine at C-3' atom and bearing the morpholine ring attached to the amidine group have been recently synthesized. Their cytotoxic activities and effects on RNA synthesis in vitro were assayed. The drug concentrations inhibiting mouse leukaemia L1210 cell growth to 50% were about two- and three fold higher for the derivatives compared to doxorubicin and daunorubicin respectively. Inhibition of phage T7 RNA polymerase by the non-covalently interacting derivatives was also slightly lower than that by the parent compounds. As doxorubicin and daunorubicin, their amidine derivatives in the presence of dithiothreitol and Fe(III) ions are activated and covalently bind to DNA. The adducts formed affect RNA polymerase activity. Several bands corresponding to prematurely terminated RNA chains are observed by means of polyacrylamide gel electrophoresis. The patterns of bands are virtually identical for all the anthracyclines studied here and are similar to the terminations induced by actinomycin D. This observation is consistent with a notion that the adducts are formed at guanine in GpC sequences which are also binding sites of actinomycin D. A substantial difference between daunorubicin and its amidine derivative is shown by means of high performance liquid chromatography. The derivative undergoes rapid rearrangements in the presence of dithiothreitol and Fe(III) ions, while daunorubicin is stable for several hours under these conditions. The results presented here indicate that the amidine derivatives despite bulky morpholine substitution exhibit biological activity in the systems used here.

Crosslinking of cellular DNA by nitracrine and furocoumarin derivatives.

The anticancer drug, nitracrine, a 1-nitro-9-aminoalkyl derivative of acridine exhibits potent cytotoxic effects which are due to its metabolic activation, followed by covalent binding to macromolecules--DNA being the target for the drug. The renaturable fraction of DNA from L-1210 cells pretreated with nitracrine is assayed by means of ethidium bromide fluorescence assay and chromatography on hydroxyapatite column. The effect of the drug was compared with furocoumarins of different DNA crosslinking potencies. The existence of crosslinks in DNA upon incubation of cells with nitracrine (1-4 microM) have been confirmed with two different methods under the conditions where 8-methoxypsoralen, a classic crosslinking agent induced the renaturation. The DNA preparation isolated from the drug pretreated cells exhibited decreased transcriptional template activity with E. coli DNA-dependent RNA polymerase.

Effect of DNA-interacting drugs on phage T7 RNA polymerase.

9-Aminoacridine carboxamide derivatives studied here form with DNA intercalative complexes which differ in the kinetics of dissociation. Inhibition of total RNA synthesis catalyzed by phage T7 and Escherichia coli DNA-dependent RNA polymerases correlates with the formation of slowly dissociating acridine-DNA complex of time constant of 0.4-2.3 s. Their effect on RNA synthesis is compared with other ligands which form with DNA stable complexes of different steric properties. T7 RNA polymerase is more sensitive to distamycin A and netropsin than the E. coli enzyme while less sensitive to actinomycin D. Actinomycin induces terminations in the transcript synthesized by T7 RNA polymerase. Despite low dissociation rates of DNA complexes with acridines and pyrrole antibiotics no drug dependent terminations are observed with these ligands.

Inhibition of RNA synthesis in vitro by acridines--relation between structure and activity.

The effects of acridine derivatives (proflavine and 2,7-dialkyl derivatives, diacridines and triacridines, 9-aminoacridine carboxamides, and 9-anilinoacridine, amsacrine and its congeners) on overall RNA synthesis in vitro, on synthesis of initiating oligonucleotides and the binding of the enzyme to DNA were studied. The primary mechanism of action is related to inhibition of the enzyme binding to DNA. The acridines (intercalating or non-intercalating and bis-intercalating ligands) assayed here differ in the properties of their complexes with DNA. Correlation is generally observed between inhibition of RNA synthesis in vitro and cytotoxicity in cell cultures for di- and triacridines and 9-aminoacridine carboxamide derivatives. No relationship was found between the effect on RNA polymerase system and biological effects for amsacrine and its derivatives in contrast to the other series of acridines studied here. The aniline ring seems to decrease the inhibitory potency of a ligand. The discrepancy between the biological effect and RNA synthesis inhibition may be due to a different mechanism of cytotoxicity action of amsacrine which is a potent topoisomerase II poison.

Uptake of acridinecarboxamide derivatives by L1210 cells.

The uptake of six 9-aminoacridinecarboxamide derivatives by L1210 cells in relation to their lipophilicity and cytotoxic activity was studied. The amount of acridines taken up by cells was estimated by fluorimetric measurements. It was found that the uptake efficiency of this class of compounds by cells depends on the size of carboxamide residue as well as on position of the substituent. The increase of size of carboxamide chain resulted in the loss of capability of acridines to penetrate cell membrane. Cytotoxic effects of acridines were well correlated with the level of drugs accumulated by cells, whereas no clear correlation between uptake and lipophilicity was observed. It is concluded that uptake of 9-aminoacridinecarboxamides is the most important factor determining their antiproliferative activity.

Transcriptional template activity of covalently modified DNA.

The transcriptional template activity of covalent modified DNA is compared. 8-Methoxypsoralen (MOP), 3,4'dimethyl-8-methoxypsoralen (DMMOP) and benzopsoralen (BP) forming with DNA covalent complexes upon UV irradiation and exhibiting preference to pyrimidines, mostly thymines, differ in their cross-linking potency. MOP and DMMOP form both monoadducts and diadducts while no cross-links are formed by BP. Nitracrine (NC) forms covalent complexes with DNA upon reductive activation with dithiothreitol exhibiting a preference to purines and low cross-linking potency. Semilogarithmic plots of the relative template activity against the number of the drugs molecules covalently bound per 10(3) DNA nucleotides fit to regression lines corresponding to one-hit inactivation characteristics. The number of drug molecules decreasing RNA synthesis to 37% differ from 0.25 to 1.26 depending on the template used and the base preference but no dependence on the cross-linking potency was found.

A new complementation group of mitomycin C-hypersensitive Chinese hamster cell mutants that closely resembles the phenotype of fanconi anemia cells.

Three Mitomycin C (MMC)-hypersensitive mutants (CL-V1B, CL-V5B, and CL-V101B) were isolated from Chinese hamster V79B cells by the replica plating technique. In comparison to the parental cell line, CL-V1B, CL-V5B, and CL-V101B show about a 22-, 32-, and 13-fold increased sensitivity to MMC, respectively (judged by the D10). These mutants are also sensitive to other DNA cross-linking agents, such as 1,2,3,4-diepoxybutane (9-, 19-, and 12-fold, respectively) and cis-diamminedichloroplatinum(II) (17-, 12-, and 6-fold, respectively). CL-V5B and CL-V101B display an exclusive sensitivity to DNA cross-linking agents, whereas CL-V1B also shows an increased sensitivity to monofunctional alkylating agents, such as methyl methanesulfonate (3-fold) and ethyl methanesulfonate (2-fold), and UV254mm (2-fold). Approximately 2-3-fold higher levels of spontaneous chromosomal aberrations are found in these three mutants in comparison to wild-type V79B cells. At a MMC survival level of 80%, CL-V5B demonstrates a 16-fold higher level of MMC-induced chromosomal damage than V79B. Despite phenotypical heterogeneity within this group of mutants, hybrid clones derived after fusion remained MMC sensitive, indicating that these mutants belong to the same complementation group. To determine whether the mutants represent a new complementation group among other Chinese hamster cell mutants that also display hypersensitivity to MMC, CL-V1B cells were fused with mutants representing different complementation groups i.e., irs1, irs3, irs1SF, UV20, UV41, V-H4, and V-C8 cells. In all cases, the derived hybrids regained MMC sensitivity similar to wild-type cells, indicating that the CL-V1B mutant represents a new complementation group. The phenotype of CL-V1B, CL-V5B, and CL-V101B cells closely resembles the phenotype of Fanconi anemia cells, suggesting that these hamster mutants could be defective in a gene that is involved in this disorder.

Mutagenic response and repair of cis-DDP-induced DNA cross-links in the Chinese hamster V79 cell mutant V-H4 which is homologous to Fanconi anemia (group A).

Previously, it has been shown that the V-H4 mutant of Chinese hamster V79 cells is homologous to Fanconi anemia (FA) group A cells. This hamster cell mutant shows a specific sensitivity to DNA cross-linking agents; therefore, the induction and repair of DNA cross-links were studied in V-H4 and wild-type V79 cells after cis-DDP treatment by the DNA alkaline elution technique. A significant difference in repair of these lesions in V-H4 and wild-type cells was observed. After the cis-DDP treatment (24 h) about 3 times more cross-links remained in V-H4 cells in comparison to the parental V79 cells. These results indicate that the process of cross-link repair in V-H4 cells is hampered when compared to that of wild-type cells. To assess the effect of slower removal of DNA cross-links on the mutability of V-H4, the induction of mutants at the hypoxanthine-guanine phosphoribosyltransferase locus (HPRT) by cis-DDP was studied in V-H4 and V79 cells. Despite the increased cytotoxicity of cis-DDP to V-H4 cells, the mutation induction at the HPRT locus was not significantly different in both cell lines, but when the frequency of the hprt mutants was plotted against survival, hypomutability was observed in V-H4 cells after the cis-DDP treatment.

Effects of alkylating metabolites of ifosfamide and its bromo analogues on DNA of HeLa cells.

The in vitro cytotoxicity and mechanism of action of three alkylating compounds: an active metabolite of ifosfamide (1, isophosphoramide mustard, N,N'-bis(2-chloroethyl)phosphorodiamidic acid) and its bromo substituted analogues, 2 (one chlorine atom replaced by bromine atom) and 3 (two chlorine atoms replaced by bromine atoms), were studied in cultured HeLa cells. Alkaline elution analysis of cellular DNA demonstrated the presence of concentration- and time-dependent interstrand crosslinks, DNA-protein crosslinks and alkali-labile sites (ALSs) in HeLa cells following a 1 hr exposure to the compounds. The bromo analogues were more cytotoxic than 1 and exhibited higher crosslinking potency. The time-course of crosslink formation and removal for the three compounds was similar. ALSs in DNA were produced by all tested drugs but 3 exhibited exceptionally high activity and was able to induce two kinds of alkali-labile lesion (fast- and slow-appearing) whereas 1 and 2 generated only slow-appearing ones. The results suggest that 1 and 2 are more specific in their reaction with DNA in that they produced a lesser variety of lesions than 3. A potential advantage of 2 over 1 seems to be its higher DNA interstrand crosslinking activity.

DNA damage and cytotoxicity of nitracrine in cultured HeLa cells.

The effects of nitracrine (1-nitro-9-(3,3-N,N-dimethylaminopropylamino)acridine on DNA of cultured HeLa cells were studied. DNA strand breakage and interstrand cross-linking as well as DNA-protein cross-linking were measured by means of an alkaline elution technique and were compared with the cytotoxic effect of the drug. Interstrand cross-links were not detectable in the concentration range that inhibited cell growth up to 99%. DNA single-strand breaks were found when cells were treated with highly cytotoxic doses of the drug. DNA breakage was not reparable and exhibited a tendency to increase during incubation after drug removal. The only chromatin lesion induced by sublethal doses of nitracrine were DNA-protein cross-links which persisted for 24 h after drug treatment. It is concluded that DNA breaks represent degraded DNA from dying cells, whereas DNA-protein cross-links are specific cellular lesions, which may be responsible for the cell-killing effect of nitracrine.

DNA binding, cytotoxicity and inhibitory effect on RNA synthesis of two new 1-nitro-9-aminoacridine dimers.

Two 1-nitro-9-aminoacridine dimers were prepared: one bearing a spermine flexible linking chain, compound 4, the other a rigid dipiperidine-type linker, compound 7. Both dimers elicited a higher affinity constant for DNA than the parent monomeric drug nitracrine 2. This affinity was several orders lower than what was found for other dimeric compounds having the same linkers and no nitro group on the acridine ring (3, 5, 6 and 8). Bisintercalation was evidenced for compound 4 by viscosimetric measurements. In the absence of dithiothreitol, an inhibitory effect of RNA synthesis in vitro was observed for all the tested compounds except 2 and 7. In the presence of dithiothreitol, 4 and 7 formed irreversible complexes with DNA of decreased template properties. The level of the dimers binding was lower than that of the parent compound 2. Cross-links were detected by means of hydroxylapatite chromatography in a complex of the dimer bearing a flexible linking chain, compound 4 with DNA, while the compound 7-DNA complex eluted in the single-stranded DNA region. The extent of cytotoxicity of the two 1-nitro-9-aminoacridine dimers against L1210 cultured cells was different.

Inhibition of DNA-dependent RNA synthesis by 8-methoxypsoralen.

The effect of the photobinding of 8-methoxypsoralen to phage T7 DNA on different steps of RNA synthesis in vitro was assayed. Total RNA synthesis is reduced to a few percent and the transcript size is decreased, as shown by means of gel filtration on a Sepharose 4B column when DNA of the adduct content of six drug molecules per 10(3) nucleotides is used. The initiation of RNA chains seems to be less affected, as inferred from an abortive initiation assay. Synthesis of pppApU on DNA of the same adduct content is inhibited to 34% of the corresponding controls, while the overall RNA synthesis is inhibited to 6%. The amount of the enzyme needed for maximal retention of DNA, the kinetics of its binding and the decay of the polymerase-DNA complex at high ionic strength (or on decrease of the temperature) are similar with DNA either irradiated in the absence of the drug or DNA bearing six 8-methoxypsoralen molecules per 10(3) nucleotides. It is concluded from this study that 8-methoxypsoralen partially inhibits initiation and blocks movement of RNA polymerase along the template, inducing premature termination. It does not appear to influence the binding of the enzyme to DNA.

H2O2 as a DNA fragmenting agent in the alkaline elution interstrand crosslinking and DNA-protein crosslinking assays.

A method for DNA fragmentation by H2O2 in the DNA alkaline elution procedure is described. Treatment of cell suspensions for 1 h with 100 microM H2O2 or 5 mM H2O2 at 0-1 degree C resulted in DNA breakage equivalent to doses of 300 and 3000 rad of gamma-rays, respectively. The elution profiles were reproducible and H2O2 was used for measurements of interstrand crosslinks and DNA-protein crosslinks induced in HeLa cells by mitomycin C, cis-diamminedichloroplatinum(II), and trans-diamminedichloroplatinum(II). The comparison of data obtained with the use of H2O2 and gamma-rays has shown that both methods have similar sensitivity and reproducibility.