PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Med Chem. Author manuscript; available in PMC 2013 September 13.
Published in final edited form as:
PMCID: PMC3478695
NIHMSID: NIHMS400179

Synthesis, Aqueous Reactivity, and Biological Evaluation of Carboxylic Acid Ester-Functionalized Platinum–Acridine Hybrid Anticancer Agents

Abstract

The synthesis of platinum–acridine hybrid agents containing carboxylic acid ester groups is described. The most active derivatives and the unmodified parent compounds showed up to 6-fold higher activity in ovarian cancer (OVCAR-3) and breast cancer (MCF-7, MDA-MB-23) cell lines than cisplatin. Inhibition of cell proliferation at nanomolar concentrations was observed in pancreatic (PANC-1) and non-small cell lung cancer cells (NSCLC, NCI-H460) of 80- and 150-fold, respectively. Introduction of the ester groups did not affect the cytotoxic properties of the hybrids, which form the same monofunctional–intercalative DNA adducts as the parent compounds, as demonstrated in a plasmid unwinding assay. In-line high-performance liquid chromatography and electrospray mass spectrometry (LC-ESMS) shows that the ester moieties undergo platinum-mediated hydrolysis in a chloride concentration-dependent manner to form carboxylate chelates. Potential applications of the chloride-sensitive ester hydrolysis as a self-immolative release mechanism for tumor-selective delivery of platinum–acridines are discussed.

INTRODUCTION

The clinical utility of the DNA cross-linking agent cisplatin (1) (Chart 1) as a life-prolonging chemotherapeutic agent is limited by various mechanisms of acquired tumor resistance and the drug’s narrow spectrum of activity.1, 2 Most cancers are inherently resistant to platinum-based therapies, and curative effects are observed only in germ cell tumors.3 For instance, in non-small-cell lung cancer (NSCLC), a major cause of cancer mortality in both men and women, platinum is used as a palliative treatment without a major survival advantage.4 Recent insight into the genetics of NSCLC resistance to cisplatin suggests that this drug is largely ineffective because the DNA adducts it forms are removed by nucleotide excision repair (NER), which is upregulated in this type of aggressive cancer.57 Among the novel cytotoxics developed to combat this and other forms of chemoresistant cancers, non-classical platinum drugs mechanistically distinct from cisplatin have demonstrated unique clinical potential because of their ability to overcome resistance at the nuclear level and induce cancer cell death by alternative mechanisms.8, 9

Platinum–acridine agents derived from the prototypical agent [PtCl(en)(1-[2-(acridin-9-ylamino)ethyl]-1,3-dimethylthiourea)](NO3)2 (PT-ACRAMTU) (Chart 1), which form monofunctional–intercalative DNA adducts but do not cross-link DNA, have shown excellent activity in solid tumors resistant to cisplatin, in particular NSCLC cell lines.10 Specifically, compound 3 (Chart 1), generated by replacement of the thiourea in 2 (PT-ACRAMTU) with an amidine donor, and related derivatives have demonstrated a cytotoxic enhancement relative to cisplatin of up to 500-fold in NSCLC cell lines.11, 12 Compound 3 produces significantly higher intracellular concentrations and DNA adduct levels in NCI-H460 cells than cisplatin,13 and the hybrid adduct itself proves to be a significantly more severe form of DNA damage than the common bifunctional cross-link.14 The cumulative effect of these pharmacological parameters renders the hybrid agent an efficient inhibitor of DNA replication and inducer of cell death.15

While the second-generation platinum–acridines show potent cytotoxic properties that translate into promising tumor inhibition in vivo, the dose-limiting toxicity observed in test animals suggests that modifications of these agents are necessary to improve their pharmacological properties.12 Specifically, we are interested in introducing platinum–acridines as cytotoxic ‘warheads’ in receptor-targeted therapies or as payloads attached to tumor-directed carriers using chemically or enzymatically reversible linkers. To achieve this goal, modification of the original structures with conjugatable functional groups is necessary. This is not a trivial task as many modifications made to cytotoxics, for instance to deliver them as a payload reversibly attached to monoclonal antibodies, have been shown to diminish the agent’s biological activity.16 Here, we have investigated the chemical and biological effects of attaching a carboxylic acid ester moiety to the nonleaving acridine moiety in compound 3. We demonstrate that the ester-modified hybrid agents, which maintain high activity in various solid tumor cell lines, are excellent models of caged carboxylato-functionalized hybrids. Based on the aqueous reactivity of these molecules, a novel self-immolative mechanism of drug activation/release via platinum-catalyzed chloride concentration-sensitive ester hydrolysis to free carboxylate is proposed.

RESULTS

Design and Synthesis

The design of the ester-modified platinum–acridines reported here was inspired by previous attempts to reduce the reactivity of these agents in circulation by exchanging the chloro leaving group in compound 3 with sterically hindered monocarboxylato ligands.11 These ligands proved to be extremely labile, which resulted in undesired intramolecular substitution of the oxygen donor by the 9-amino group of the acridine moiety to produce a substitution-inert, marginally cytotoxic nitrogen chelate. We reasoned that incorporation of the carboxylato ligand as a pendant group into a chelate, similar to the dicarboxylato leaving group in the clinical drug carboplatin,1 should produce a more stable, but yet DNA reactive entity. Thus, initially the synthetic goal of this study was to generate a suitable precursor molecule by installing a carboxylic acid ester group at the amidine moiety of the acridine carrier ligand that can be converted into a carboxylate chelate (Scheme 1). Precedent for this reaction was found in analogous Pt(II) and Pd(II) complexes containing ester-substituted pyridine ligands.17

Scheme 1
Proposed Conversion of Ester Precursor to Chelated Carboxylato Ligand

The desired target compounds in this study were synthesized using methodology developed for compound 3 and related analogues, which involves highly efficient addition of a suitably modified 9-aminoacridine derivative across the activated CN triple bond in the appropriate platinum–nitrile complexes to produce an amidine linkage.11, 12 Three ester-modified hybrids (compounds 1012) and, for comparison, two simple propionamidine derivatives ( and 9) were synthesized and isolated as mononitrate salts (Scheme 2) (Note: compound is the +1 charged form of compound 3). While compounds , 9, 11, and 12 were isolated in >85% yield and >95% analytical purity, the synthesis of derivative 10 suffered from an inherent instability of precursor complex 5 (decarboxylation, dimerization) as a consequence of the reactivity of the methylcyanoacetate ligand (see Supporting Information). From the reaction mixtures, we were able to isolate milligram quantities of compound 10 using semipreparative reverse-phase HPLC (see Supporting Information) sufficient for cytotoxicity assays in one of the more sensitive cell lines.

Scheme 2
Synthesis of Target Compounds

Biological Activity

The newly synthesized platinum–acridines were studied along with cisplatin in NSCLC (NCI-H460), ovarian (OVCAR-3), breast (MDA-MB-231, MCF-7), and pancreatic (PANC-1) cancer cell lines using a colorimetric cell proliferation assay. The results are summarized in Table 1. The majority of the platinum–acridines screened performed significantly better than cisplatin in all cancer cell lines in this assay, based on the inhibitory concentrations calculated from the drug–response curves. While the most active analogues in each series showed a 3–6-fold enhanced cytotoxicity compared to the clinical drug in the ovarian and breast cancer cell lines, a dramatic increase in cytotoxic potency of approximately 80-fold and 150-fold was observed in the pancreatic and NSCLC cell line, respectively. Cisplatin inhibits cell proliferation at micromolar concentrations, as typically observed for this drug in solid tumor cell lines.18 By contrast, several of the simple ( and 9) and ester-modified (1012) platinum–acridines show effective cell kill in the nanomolar concentration range.

Table 1
Summary of Cytotoxicity Data for Platinum–Acridines and Cisplatin in Human Solid Tumor Cell Lines

In addition to these global trends in activity and chemosensitivity, two important structure–activity relationship features emerge in the set of compounds tested. First, extension of the propionamidine moieties in compounds and 9 with a methyl ester group, giving derivatives 11 and 12, did not have a negative effect on the cytotoxicity of the hybrids, except in the case of the pair 9/12 in OVCAR-3, where the activity of the ester-modified compound decreases by approximately 5-fold. In most cases, compounds 11 and 12 exhibited cytotoxicities similar to those of their unmodified counterparts. This was unexpected since it has previously been observed that addition of steric bulk in this part of the molecule results in a decrease in cytotoxicity.11 Another critical trend observed across the entire data set is that the hybrids in which the platinum moiety is separated from the intercalator by an extended propylene linker (m = 3 in 9 and 12) show greatly reduced cell kill potential. For instance, in NCI-H460, derivatives and 11 (with m = 2) are approximately 6 times more cytotoxic than compounds 9 and 12. This effect is even more pronounced in the cell lines OVCAR-3 and PANC-1. Compound 10 (n = 1, m = 2), which was tested only in NCI-H460 cells, was 3–4-times less cytotoxic than compounds and 11. In other words, extension of the amidine side chain by one methylene group (n = 2 in 11 vs. n = 1 in 10) significantly improves the cell kill activity in this particular cell line.

Aqueous Reactivity

Compounds 1012 were designed as potential prodrugs of the respective carboxylato complexes and as model compounds to test the utility of carboxylic acid esters as cleavable groups in cancer-cell or tumor-tissue targeted conjugates. Attempts to synthesize the corresponding closed carboxylato chelate complexes (Scheme 1) from compounds 1012 were unsuccessful and were abandoned after it was established that these form spontaneously on a physiologically relevant time scale in pH neutral solution, as described below.

To study the aqueous chemical reactivity of the pendant ester functional groups under physiologically relevant conditions, compounds 11 and 12 were incubated at 37 °C in phosphate buffer (pH 7.4) and in phosphate buffer supplemented with 150 mM NaCl. The latter media was chosen to simulate chloride concentrations in circulation. The reaction mixtures were then analyzed at various time points by in-line high-performance liquid chromatography–electrospray mass spectrometry (LC–ESMS). Complexes 11 and 12 were found to show the same reactivity, which was expected since the only difference between the two analogues is an extended acridine linkage (m = 2 in 11 vs. m = 3 in 12). Because the reaction mixtures produced by compound 12 resulted in superior chromatographic (baseline) separations, results were analyzed in detail and will be discussed for this derivative. Selected analytical data (for mass spectra of all species, see Supporting Information) and structures of the species identified in the mixtures are summarized in Figure 1.

Figure 1
Reverse-phase HPLC analysis of a 1 mM solution of complex 12 in (A) 10 mM phosphate buffer (pH 7.4) prior to incubation, (B) 10 mM phosphate buffer after 18 h of incubation at 37 °C, and (C) 10 mM phosphate buffer containing 150 mM NaCl after ...

When a freshly prepared sample of compound 12 in 10 mM phosphate buffer was analyzed, the HPLC trace showed 90% chemically unchanged 12 (Figure 1A), based on the molecular ion and fragment ions observed in mass spectra recorded in positive-ion mode (Figure 1D). Additionally, two minor products were identified consistent with substitution of the chloro ligand in 12 by phosphate (peak b, Figure 1A), as well as an aquated form that has undergone ester hydrolysis (peak c, Figure 1A). After 18 h of incubation at 37 °C, complete conversion of compound 12 is observed. The major species detected in the corresponding HPLC trace (peak d, Figure 1B), which accounts for 70% of the product mixture, is easily identified as the proposed carboxylate chelate in which the chloro ligand has been replaced with carboxylate oxygen (Figure 1E). In addition, small amounts (< 10%) of non-chelated species and acridine–amine 8 are observed (peak e), the latter product suggesting minor degradation of the hybrid (~10%), which was unexpected. When incubations were performed under the same conditions in a buffer containing 150 mM NaCl, approximately 55% of compound 12 remained in the mixture chemically unchanged (Figure 1C), and the carboxylate chelate shows an abundance of only 17%. Under these high-chloride conditions, 22% of the hydrolyzed ester product exists in the form of the chloro complex containing a dangling carboxylate (peak f, Figure 1C). This observation is consistent with facile opening of the seven-membered N,O chelate by this relatively weak donor.

The incubations in buffer mimicking biological media show that the efficiency of ester hydrolysis in compound 12 is reduced by high chloride concentrations, which suggests that cleavage of the ester group is mediated to a significant extent by platinum. To gain insight into the kinetics of ester cleavage in the presence and absence of chloride ion, the time course of ester conversion was monitored by LC–ESMS. This was possible by integrating the peaks in the chromatograms recorded for samples withdrawn from the reaction mixtures at various time points during the first 6 h of incubation. The percentage of the total platinum species present in solution in which the ester moiety was converted to chelated or dangling carboxylate (fractions c, d, and f, see Figure 1) was then plotted against reaction time (Figure 2). From curve fits of the data assuming first-order reaction kinetics, half-lives of 2 h and 8 h were estimated for ester hydrolysis in chloride-free and chloride-containing media, respectively. Incubations of compound 12 in chloride-free acidic solution in the pH range 1–5 (adjusted with HNO3) did not lead to significant cleavage of the ester group. By contrast, rapid ester hydrolysis is observed at pH > 9. Under these conditions, relative larger amounts of compound 12 decompose to produce free acridine–amine 8 (data not shown).

Figure 2
Time course of ester hydrolysis in compound 12 at 37 °C monitored by quantitative HPLC in phosphate-buffered (pH 7.4) solution (filled circles) and in 10 mM phosphate buffer containing 150 mM NaCl (open circles). Species in solution were quantified ...

Based on the pH and chloride concentration dependence of ester hydrolysis observed in complex 12, an intramolecular, metal-mediated mechanism is proposed in which platinum acts as a catalytic Lewis acid (Scheme 3). Ester cleavage may proceed either via internal attack of an activated aqua (hydroxo) ligand or via external attack of the metal activated ester group by water or hydroxide ion. Both pathways of the proposed mechanism, which are in complete analogy to the platinum-catalyzed intramolecular hydrolysis observed for peptide bonds,19 would explain why suppression of (reversible) aquation of the complex by high concentrations of external chloride inhibits ester hydrolysis catalyzed by platinum.

Scheme 3
Proposed Mechanism of Platinum-Catalyzed Ester Hydrolysis.

Reactivity with DNA Nitrogen and Amino Acid Sulfur

To study the effect of the pendant ester groups and the corresponding carboxylato chelates in the platinum–acridines in reactions with DNA nitrogen, compound 11 was incubated with excess 5´-guanosine monophosphate (5´-GMP, disodium salt). The reactions were performed with complex 11 in phosphate-buffered solution prior to and after complete hydrolysis of the ester and monitored by quantitative LC-ESMS analysis. For comparison, incubations were also performed with compound under the same conditions. Quantitative HPLC analysis shows that compound converts nearly quantitatively to monofunctional adduct with a half-life of 80 min (Figure 3). By contrast, the analogous reactions with intact compound 11 and its hydrolyzed form do not go to completion but show approximately 85% conversion, which may be a result of the above-noted partial decomposition of the compound. However, both species react with the mononucleotide significantly faster than compound with a half-life of less than 30 min. Formation of monofunctional adducts in all three cases was confirmed by electrospray mass spectrometry (Figure 4A). While the high reactivity of the seven-membered carboxylato chelate can be expected since oxygen is a better leaving group than chloride, the fact that compound 11 reacts faster than compound with guanine nitrogen despite the added steric bulk of the ester functional group is surprising. One possible explanation for this rate enhancement might be that the ester group acts as an internal auxiliary nucleophile that enhances nucleotide binding by catalyzing the rate-limiting loss of chloride (see external pathway in Scheme 3). In reactions of 11 with 5´-GMP, nucleotide binding outcompetes ester hydrolysis, resulting in modification of the nucleotide mainly with the intact ester derivative (adduct II in Figure 4A) rather than the hydrolyzed form, of which only traces are observed (adduct III in Figure 4A).

Figure 3
Time course of adduct formation with 5´-GMP at 37 °C (1 mM Pt, 10 mM nucleotide) for compound 11 (triangles), the carboxylate chelate formed from compound 11 after 18 h of pre-incubation in phosphate buffer (open circles), and compound ...
Figure 4
Electrospray mass spectra recorded in positive-ion mode in the 200–800 m/z range of adducts formed in reactions of platinum–acridines , 11, and hydrolyzed 11 with (A) 5´-GMP and (B) GSH in phosphate-buffered solution ...

The reactivity of the three platinum–acridines with glutathione (GSH) was also studied using the same experimental setup. Incubations of the platinum complexes with one equivalent of GSH at 25 °C resulted in stoichiometric replacement of the leaving group with the tripeptide to produce adducts of a 1:1 stoichiometry (Figure 4B). Reactions with the sulfur nucleophile proceeded rapidly in all cases making it impossible to determine half-lives chromatographically (and by proton NMR spectroscopy). An important outcome of this assay is that the carboxylate chelate in the hydrolysis product does not seem to protect platinum from nucleophilic attack by GSH sulfur, a reaction that has been implicated as a cause of tumor resistance and nephrotoxicity of cisplatin.20 The high reactivity with sulfur was expected based on the facile substitution of coordinated carboxylate by chloride observed in the high-salt phosphate buffer.

DNA Interactions

Introduction of an ester functionality in compounds (giving 11) and 9 (giving 12) did not have a major impact on the cytotoxic potency of the agents (Table 1). On the other hand, extension of the alkylene linker connecting platinum to the acridine intercalator in this set of compounds by only one carbon atom (e. g., m = 2 in vs. m = 3 in 9) resulted in a pronounced decrease in cytotoxicity. To assess if the effects observed in cells may be mediated at the DNA adduct level, a plasmid unwinding assay was performed. This assay has been used routinely to determine the local untwisting of closed-circular supercoiled DNA induced by different types of platinum–DNA adducts,21 including those produced by platinum–acridines.22, 23 The unwinding produced per adduct extracted from these experiments provides insight into the binding mode of a DNA-modifying agent. In this experiment, negatively supercoiled pUC19 plasmid was modified with compounds , 9, and 11 at varying platinum-to-DNA nucleotide ratios (rb) and analyzed by gel electrophoresis (Figure 5). The results show that the hybrid adduct formed by the prototypical agent unwinds the duplex by 17±1°. This value is in excellent agreement with the unwinding produced by compound 3 in site-specifically modified model duplexes.14 Furthermore, the data demonstrate that compounds 9 and 11 produce essentially the same degree of unwinding as the parent compound (16±1°), suggesting that extension of the linker by one methylene group or addition of a dangling ester group have no effect on the intercalation geometry in the hybrid adducts. Quantitation of platinum bound to the DNA samples after 24 h of incubation and subsequent removal of unbound agent by microdialysis shows that only ~64% of compound 11 forms permanent adducts with the plasmid, whereas binding of the parent compound is quantitative under the same conditions (see Supporting Information). The reduced binding levels observed for compound 11 may be caused by partial conversion of the ester into a carboxylate chelate and the electrostatically unfavorable negative charge introduced into the DNA adducts by the carboxylate. However, despite being a less efficient inducer of hybrid adducts in this assay, compound 11 appears to cause the same distortions in DNA per adduct as compound . Because of the inherent hydrolytic reactivity of compound 11, the plasmid reacted with it likely contains adducts featuring dangling ester and carboxylate groups. In that case, the observed unwinding angle would suggest that both adducts produce similar structural effects in DNA. To study the effect of a dangling anionic carboxylate group on the DNA binding mode, unwinding experiments were also performed with prehydrolyzed compound 11 (not shown). Because ester hydrolysis leads to product mixtures of varying composition depending on the buffer system used, unwinding experiments performed with the reaction mixtures gave inconsistent results and were not pursued further.

Figure 5
DNA unwinding study of pUC19 plasmid modified with varying amounts (0.02 < rb < 0.07) of compounds (A), 9 (B), and 11 (C). Ethidium stained agarose gels are shown along with quadratic polynomial fits of integrated band intensities ...

The differences in cancer cell kill observed for compounds and 9 are unlikely a consequence of differences in the DNA adducts they form, as both analogues produce the same DNA helical perturbations. However, while both compounds show the same mode of intercalation, extension of the linker from m = 2 (in ) to m = 3 (in 9) may have an effect on the rate with which platinum binds with nucleobase nitrogen. This is a possibility because increasing the spacing between acridine and platinum may affect the nucleophilic attack of nucleobase nitrogen on the metal at the site of intercalation. To investigate the time course of platination of DNA by derivatives and 9, we took advantage of the compounds’ ability to efficiently convert negatively supercoiled DNA (Form 1) into the fully relaxed form (Form 2). Because only irreversibly bound platinum–acridine contributes to the unwinding of plasmid DNA in this assay,22 the rate at which the relaxed (open-circular) form is generated is directly proportional to the rate of platination.

In this experiment, pUC19 plasmid was incubated at the same concentration of platinum that produces adduct levels high enough to completely convert Form 1 of the DNA into Form 2 (ri ≈ 0.07, see Figure 5). Samples were withdrawn from the incubation mixtures at various time points and quenched with excess thiourea under conditions where the sulfur nucleophile binds to platinum to prevent further reaction with DNA but does not reverse existing DNA adducts.24 Samples were then analyzed by agarose gel electrophoresis to determine the relative rates of formation of Form 2 of the plasmid. A comparison of the two electrophoretic separations (Figure 6A,B) clearly shows that derivative 9 produces Form 2 at a significantly slower rate than derivative . This qualitative assessment is based on the relatively lower intensity of the corresponding band on the gel observed for the former derivative and the fact that partially relaxed topoisomers (diffuse bands of intermediate mobility) persist longer for 9 than for . A plot of integrated band intensities vs reaction time for Form 2 of the plasmid generated by and 9 confirms this observation (Figure 6C). From curve fits of the data to first-order reaction kinetics, half-lives of 43 min and 77 min were estimated for and 9, respectively (Figure 6C). In conclusion, while the acridine chromophores in the two derivatives appear to produce similar intercalation geometries in the final DNA adducts they form, extending the linker by one methylene group leads to less efficient platination of nucleobase nitrogen. This decrease in DNA reactivity may contribute to the loss in cytotoxic potency observed for compound 9.

Figure 6
Kinetics of unwinding of pUC19 plasmid by platinum–acridines studied by electrophoretic analysis for compounds (A) and 9 (B). Electrophoretic separations in (A) and (B) were run on the same agarose gel to minimize artifacts due to inconsistent ...

DISCUSSION

In a search for targeted prodrug technologies for the delivery of highly cytotoxic platinum–acridine hybrid agents, we have evaluated the utility of carboxylic acid ester-modified derivatives. We demonstrated that addition of suitable 9-aminoacridine precursors to platinum-bound (functionalized) nitrile is a suitable synthetic strategy for incorporating an ester group into the carrier side chain of these hybrid agents. Platinum–nitrile complexes, such as compound 6, prove to be versatile synthons for generating structurally and functionally diverse hybrid agents. The clean and high-yield conversion to hybrid agent in the absence of by-products, which are the hallmarks of organic click chemistry,25 has been used here to generate compounds 11 and 12. The synthetic scheme, however, proved to be impractical for generating compound 10 due to the inherent instability of precursor complex 5. This limitation complicates the synthesis of carboxylate-modified hybrid agents of sufficient purity and stability able to form potentially less reactive six-membered carboxylate chelates (such as compound 10).

The biological consequences of introducing an ester group into the hybrid agents and structure–activity relationships in this set of compounds were studied in five solid tumor cell lines. NCI-H460, which represents an aggressive form of lung cancer, has been shown to be most responsive to platinum–acridine treatment and was included in this assay as a benchmark. The fact that compound 11 maintains nanomolar activity in this cell line correlates well with the DNA interactions established for the compound and supports previous observations implying that nanomolar activity in this type of cancer requires rapid formation of hybrid adducts in cellular DNA.13, 24 The approximately 6 times lower cytotoxicity observed for compound 9 compared to (the same trend is observed for the pair 11/12) suggests that DNA-binding efficiency is directly correlated with cytotoxicity in this rapidly proliferating cancer. However, it cannot be ruled out that subtle differences in compound uptake or in the structures and recognition of their DNA adducts may also contribute to the differences in biological activity.

OVCAR-3 is a cell line derived from an ovarian tumor refractory to cisplatin. The level of resistance observed in these cells is closely correlated with high levels of GSH,26 and we reasoned that the proposed intracellular formation of the carboxylate chelates from the ester precursors might protect platinum from detoxification by the sulfur-containing tripeptide. Overall, compounds and 11 show slightly better activity in this cell line than cisplatin. The fact that the ester derivatives had no advantage over the unmodified parent compounds, however, suggests that the conversion of compounds 11 and 12 into carboxylate chelates plays no role as a protective mechanism. While intracellular conversion to the chelated form can be expected during the 72-h incubations of the cancer cells with platinum (based on the t1/2 of 2 h of chelate formation in buffered neutral solution, see Figure 2), we also demonstrated that this form of the agent would readily react with GSH.

Significantly higher cytotoxicity is observed for the most potent platinum–acridines than for cisplatin in the breast cancer lines MCF-7 and MDA-MB-231, which were included in this panel because they are generally insensitive to chemotherapy.27 Especially in hormone-dependent breast cancers, such as estrogen receptor-positive MCF-7, platinum–acridines may have applications as a cytotoxic component in combination therapies or multifunctional targeted therapies (see discussion below). Finally, PANC-1 was included in this panel. Screening in this pancreatic cancer cell line was inspired by the compelling mechanistic similarities between the drug gemcitabine (Gemzar), one of the most effective therapies in pancreatic cancer,28 and platinum–acridine 3. Unlike cisplatin, both agents are very efficient inhibitors of DNA synthesis and induce cell-cycle arrest in late G1/early S phase.15, 29 Compounds and 11 are highly cytotoxic at submicromolar concentrations in PANC-1, whereas micromolar concentrations of cisplatin (Table 1) and gemcitabine28 are required to produce the same cell kill effect in this cell line. Given these promising results, platinum–acridines will be studied as systemic and targetable cytotoxics in additional pancreatic cancer models.

The high reactivity of the seven-membered carboxylate chelate precludes the use of compound 11 as a simple non-targeted prodrug to help overcome GSH-mediated tumor resistance and/or reduce systemic toxicity. The ester technology however provides a unique opportunity for tuning the biodistribution of the platinum–acridines and for delivering them to target tissues. This should be possible by attaching the hybrid agents to suitable carrier systems using cleavable ester linkages. Potential delivery vehicles include, for instance, polymeric and nanosized materials and small molecules that bind to cancer cell-specific receptors. Simple lipophilic groups, such as fatty acid esters and glycerides in place of the simple methyl ester in compound 11 would be an interesting option to tune the hydrophilic/lipophilic balance and enhance the passive permeability of the hybrids. The ideal conjugates would be sufficiently stable in circulation where the high chloride concentration should suppress complex aquation and metal-promoted ester cleavage. Once internalized into the cancer cell, hydrolysis of the ester linkage is triggered by the low chloride concentration in the cytosol resulting in the self-immolative release30 of hybrid agent (Figure 7). To achieve this design goal, additional structural modifications may be necessary that extend the half-life of the ester group, which may also be prone to spontaneous and enzymatic cleavage in plasma. In essence, this approach will exploit the classical mechanism of activation of platinum drugs by intracellular complex aquation31 for the targeted release of platinum–acridines.

Figure 7
Proposed mechanism of action of tumor-targeted platinum–acridines based on ester cleavage triggered in a chloride concentration gradient. (I) Passive accumulation in tumor tissue or active targeting of cancer cell by an ester-linked tumor-selective ...

In conclusion, in this study an ester-functionalized form of the platinum–acridine pharmacophore was generated and studied as a cytotoxic agent in various cancer models. The modified agent shows the same high cell-kill potential as the unmodified parent compound, which translates into promising submicromolar activity in NSCLC NCI-H460 and previously untested pancreatic PANC-1. The high sensitivity of the latter cell line to the hybrid agents is exciting given the urgent need for novel treatment options to tackle pancreatic cancer. Efficient targeting of this and other types of cancer investigated in this study using appropriate carrier molecules in conjunction with novel self-immolative release mechanisms may improve the toxicity profile and in vivo efficacy of the platinum–acridines.

EXPERIMENTAL SECTION

Materials and Procedures

Cisplatin (1), 5´-guanosine monophosphate (5´-GMP, disodium salt, 99%), and glutathione (reduced form, 98–100%) were purchased from Sigma. The platinum precursor [PtCl2(en)] (en = ethane-1,2-diamine) was synthesized by the method of Dhara.32 Compounds , 4, 7 and 8 were synthesized previously.11, 12, 33, 34 For the preparation of biological buffers, biochemical grade chemicals (Fisher/Acros) were used. HPLC-grade solvents were used for all HPLC and mass spectrometry experiments. All reagents were used as obtained without further purification.

1H NMR spectra of the target compounds and intermediates were recorded on Bruker Advance 300 and DRX-500 instruments operating at 300 and 500 MHz, respectively. 13C{1H} NMR spectra were recorded on a Bruker DRX-500 instrument operating 125.8 MHz. Chemical shifts (δ) are given in parts per million (ppm) relative to tetramethylsilane (TMS). Electrospray mass spectra were recorded on an Agilent 1100LC/MSD ion trap mass spectrometer. Ion evaporation was assisted by a flow of N2 drying gas (300–350 °C) at a pressure of 40–50 psi and a flow rate of 11 L/min. Mass spectra were typically recorded with a capillary voltage of 2800 V and a mass-to-charge (m/z) scan range of 200 to 2200. The purity of the target compounds and the composition of reactions mixtures was analyzed by reverse-phase high-performance liquid chromatography (HPLC) using the LC module of an Agilent Technologies 1100 LC/MSD trap system equipped with a multi-wavelength diode-array detector. Separations were performed on a 4.6 mm × 150 mm reverse-phase Agilent ZORBAX SB-C18 (5 µm) analytical column at 25 °C. The following solvent system was used: solvent A, optima water, and solvent B, methanol/0.1% formic acid, at a flow rate of 0.5 mL/min and a gradient of 95% A to 5% A over 25 min. HPLC traces were recorded over a wavelength range of 363–463 nm. Peak integration was accomplished with the LC/MSD Trap Control 4.0 data analysis software. For target compounds studied in cancer cells and DNA binding experiments an analytical purity of ≥ 95% was confirmed using this setup.

[PtCl(en)(C4H5NO2)]Cl (5)

The complex [PtCl2(en)] (300 mg, 1.23 mmol) was heated under reflux in DI H2O (5 mL) with methyl cyanoacetate (5 mL, excess) until the yellow suspension turned into a colorless solution (~1 h). Solvent was removed by rotary evaporation. The pale-yellow residue was dried in a vacuum at 40 °C and dissolved in 5 mL of dry methanol. The solution was passed through a syringe filter, and the colorless filtrate was added directly into 150 mL of vigorously stirred dry diethyl ether. The resulting white precipitate was isolated by membrane filtration and dried in a vacuum, yielding complex 5 as an off-white microcrystalline solid, which was filtered off and dried in a vacuum. Yield: 207 mg of a moisture-sensitive crude product, which could not be further purified but used directly for the synthesis of compound 10 (see Supporting Information for details of the product characterization by LC-MS).

[PtCl(en)(C5H7NO2)]Cl (6)

This analogue was prepared as described for 5 starting from 200 mg (0.613 mmol) and methyl cyanopropionate (2 mL, excess) in 3 mL DI H2O. Yield: 183 mg (68%). 1H NMR (MeOH-d4) δ 5.93 (4H, m), 3.75 (3H, s), 3.18 (2H, t, J = 6.6 Hz), 2.82 (2H, t, J = 6.6 Hz), 2.59 (4H, m). ESMS (+ve mode): m/z 403.0 ([M]+).

[PtCl(en)(C20H24N4)]NO3 (9)

Platinum–nitrile complex 4 (200 mg, 0.526 mmol) was converted to its nitrate salt by reaction with AgNO3 (85 mg, 0.500 mmol) in 4 mL of anhydrous DMF. AgCl was removed by syringe filtration, and the filtrate was cooled to −10 °C. N-(Acridin-9-yl)-N′-methylpropane-1,3-diamine (8) (140 mg, 0.524 mmol) was added, and the suspension was stirred at −10 °C until it turned into a deep orange solution (~2 h). After treatment with activated carbon, the reaction mixture was filtered through a Celite pad and added dropwise into 200 mL of vigorously stirred diethyl ether. The yellow slurry was stirred for ~30 min. The bright yellow precipitate was recovered by membrane filtration, washed exhaustively with dry diethyl ether to remove excess acridine–amine 8 and dried in a vacuum at 60 °C overnight. Yield 286 mg (85%). 1H NMR (MeOD-d4) δ 8.31 (2H, d, J = 8.1 Hz), 7.82 (2H, d, J = 8.7 Hz), 7.69 (2H, t, J = 7.7 Hz), 7.39 (2H, t, J = 7.7 Hz), 3.95 (2H, t, J = 6.8 Hz), 3.40 (2H, t, J = 7.5 Hz), 2.85 (5H, overl. m), 2.55 (4H, m), 2.03 (2H, p, J = 7.4 Hz), 1.17 (3H, t, J = 7.1 Hz). 13C{1H} NMR (MeOH-d4) δ 171.1, 155.2, 147.7, 132.6, 132.4, 129.2, 126.7, 125.6, 123.9, 117.2, 66.9, 28.9, 15.5, 11.7. ESMS (+ve mode): m/z 611.2 ([M]+), 305.7 ([M+H]2+), 575.2 (M−Cl−H]+, 287.2 (M−Cl]2+).

[PtCl(en)(C20H22N4O2)]NO3 (10)

This derivative was prepared analogously to compound 9. Starting from complex 5 (450 mg, 1.06 mmol) and acridine–amine 7 (258 mg, 1.03 mmol), 489 mg of crude product were obtained, which contained major amounts of impurities that could not be removed by recrystallization (see text and Supporting Information). An analytically pure sample of complex 10 was generated by HPLC purification on a Hitachi D-7000 HPLC system equipped with a L-7420 UV-Vis detector using a 9.4 mm × 250 mm reverse-phase Agilent ZORBAX SB-C18 (5 µm) semi-preparative column. The following solvent system was used: solvent A, optima water, and solvent B, methanol/0.1% formic acid, with a flow rate of 0.75 mL/min and a gradient of 95% A to 5% A over 45 min. A monitoring wavelength of 413 nm was used to identify acridine-containing products. Approximately 10 mg of compound 10 of 95% purity were generated. ESMS (+ve mode): m/z 641.2 ([M]+), 321.1 ([M+H]2+), 302.7 (M− Cl]2+).

[PtCl(en)(C21H24N4O2)]NO3 (11)

This derivative was prepared analogously to compound 9. Starting from complex 6 (300 mg, 0.684 mmol) and acridine–amine 7 (170 mg, 0.676 mmol), 419 mg (86%) of compound 11 were obtained. 1H NMR (MeOH-d4) δ 8.28 (2H, d, J = 8.5 Hz), 7.70 (4H, overl. m), 7.25 (2H, t, J = 7.5 Hz), 4.11 (2H, t, J = 6.3), 3.79 (2H, t, J = 6.3 Hz), 3.68 (3H, s), 3.29 (2H, t, J = 8.3 Hz), 2.60 (3H, br. s), 2.93 (2H, t, J = 6.3 Hz), 2.59 (4H, overl. m). 13C{1H} NMR (MeOH-d4) δ 172.7, 168.1, 154.2, 131.3, 125.2, 124.7, 122.4, 65.5, 30.3, 29.3, 14.1. ESMS (+ve mode): m/z 655.1 ([M]+), 328.0 ([M+H]2+), 309.2 (M−Cl]2+).

PtCl(en)(C22H26N4O2)]NO3 (12)

This derivative was prepared analogously to compound 9. Starting from 310 mg (0.708 mmol) of complex 6 and 187 mg (0.705 mmol) of acridine–amine 8, 459 mg (89%) of compound 12 were obtained. 1H NMR (MeOD-d4) δ 8.31 (2H, d, J = 8.5 Hz), 7.80 (2H, d, J = 8.5 Hz), 7.69 (2H, t, J = 7.3 Hz), 7.38 (2H, t, J = 7.3 Hz), 3.94 (2H, t, J = 7.0 Hz), 3.65 (3H, s), 3.46 (2H, t, J = 7.3 Hz), 3.21 (2H, t, J = 8.0 Hz), 2.96 (3H, s), 2.90 (2H, t, J = 8.3 Hz), 2.53 (4H, m), 2.06 (2H, p, J = 7.3 Hz). 13C{1H} NMR (MeOD-d4) δ 174.1, 168.8, 155.1, 147.6, 132.4, 125.6, 123.8, 117.3, 52.4, 49.931.7, 30.5, 29.4. ESMS (+ve mode): m/z 668.2 ([M]+), 335.1 ([M+H]2+), 316.2 (M−Cl]2+).

Incubations

Ester hydrolysis in compounds 11 and 12 was studied by incubating the respective agent (1 mM) in a 10 mM phosphate buffer (pH 7.4, with and without 150 mM NaCl) at 37 °C for 18 h. The reaction mixtures were analyzed by in-line LC-MS using the experimental setup described under Materials and Procedures. All reaction mixtures were diluted with methanol/formic acid to a final concentration of platinum of 20 µM immediately prior to injection into the LC unit. To study the kinetics of chelate formation, the same reactions were carried out, however, LC-MS spectra were collected of samples withdrawn from the incubation mixtures and quenched on ice approximately every 45 min over a period of 6 h. Reactions with 5´-guanosine monophosphate (5´-GMP) and glutathione (GSH) were performed at a 1:10 drug-to-nucleophile ratio in the same (NaCl-free) phosphate buffer. (The same conditions were applied for samples of 11 and 12 that were pre-incubated in chloride-free phosphate buffer at 37 °C for 18 h). Incubations with nucleophiles were performed at 37 °C for 24 h. The reaction mixtures were analyzed by LC-MS as described in the previous section. Time-dependent data were fitted to (pseudo) first-order reaction kinetics using SigmaPlot (version 12.0, Systat Software, San Jose, CA).

DNA Unwinding Assay

Plasmid DNA (pUC19, 2686 bp) was generated by transforming a chemically competent E. coli (NovaBlue, Novagen) with pUC19 and extracted from the cells using Qiagen’s MegaPrep Plasmid kit (Qiagen, Valencia, CA). Plasmid concentrations were determined spectrophotometrically at 260 nm with ε = 6500 M−1 · cm−1 (nucleotides, n.t.). To modify the DNA with known amounts of platinum agent, pUC19 plasmid (5 µM n.t.) was incubated in TE buffer (10 mM Tris/1 mM EDTA, pH 7.5) overnight at 37 °C with , 9, and 11 at various platinum-to-nucleotide ratios (ri) (ri = 0.01–0.2; for accurate determination of the coalescence point, rb(c), the range of modification was narrowed down to ri = 0.02–0.07). Incubation concentrations of the platinum hybrid agents were determined from their absorbances at 413 nm in water using an extinction coefficient of 105 M−1 · cm−1.12, 22 The progress of DNA adduct formation for and 9 was monitored at various time points (5–360 min) by quenching samples with 7 µM thiourea and chilling at 0 °C. Bound platinum-to-nucleotide ratios (rb) around the coalescent points were determined by inductively-coupled plasma mass spectrometry (ICP-MS) using a previously reported protocol.13 Samples were dialyzed against 8 L of MilliQ water in the dark at 4 °C (Fisher 6000–8000 MW cut-off dialysis membrane) prior to ICP-MS analysis. Platinum-modified plasmid was analyzed by electrophoresis on 1% agarose gels (60 V, 4 °C for 4 h, TAE buffer, pH 7.5). Gels were stained with ethidium bromide (1 µg/mL) for 30 min and washed with water (3×100 mL for 15 min). The bands on the gels were visualized and quantified on a Chemidoc XRS imaging system running Quantity One Software (Biorad, Hercules, CA). The rb(c) for each compound was determined by plotting integrated band densities of the relaxed DNA (Form 2) vs rb, and the data was fitted using a quadratic polynomial in SigmaPlot 12.0. Unwinding angles (ϕ were calculated using the following equation:

equation M1

where σ is the superhelical density of the plasmid under the specific conditions of this assay (−0.050, determined by electrophoresis using ethidium bromide with a known unwinding angle of 26°).35 The rb(c) values determined for each agent and the corresponding unwinding angles are means ± standard deviations of three independent experiments.

Cytotoxicity Studies

Cell proliferation assays were carried out using the Celltiter 96 aqueous non-radioactive cell proliferation assay (Promega, Madison, WI). For this assay, all cells were kept in a humidified atmosphere of 5% CO2 at 37 °C. Cells were seeded into 96-well plates with a total of 740 cells/well for NCI-H460, 10000 cells/well for OCVAR-3, MCF-7, and MDA-MB-231, and 2000 cells/well for PANC-1. 10 mM stock solutions of the tested agents were prepared in DMF and stored at −20 °C. Cells growing in log phase were incubated for 72 h with appropriate serial dilutions of the drugs (starting at a drug concentration of 100 µM). To each well were added 20 µL of MTS/PMS solution, and the mixtures were allowed to equilibrate for 2–4 h. The absorbance of formazan that is produced by bioreduction of MTS was measured at 490 nm (none of the derivatives tested absorbs at this wavelength), and the IC50 data was calculated from non-linear curve fits using a sigmoidal dose-response equation in GraphPad Prism (La Jolla, CA). The calculated IC50 values are averages of three individual experiments performed in triplicate.

Supplementary Material

1_si_001

ACKNOWLEDGMENTS

This research was sponsored by the National Institutes of Health/National Cancer Institute (grant CA101880). We thank Drs. K. E. Levine and A. S. Essader (RTI International, Research Triangle Park, North Carolina, USA) for the ICP-MS measurements.

ABBREVIATIONS USED

ACRAMTU
1-[2-(acridin-9-ylamino)ethyl]-1,3-dimethylthiourea
ESMS
electrospray mass spectrometry
GMP
guanosine monophosphate
GSH
glutathione
NER
nucleotide excision repair
NSCLC
non-small cell lung cancer
rb
bound platinum-to-nucleotide ratio
ri
platinum-to-nucleotide incubation ratio

Footnotes

ASSOCIATED CONTENT

Supporting Information

Spectroscopic and analytical data for target compounds, details of the LC-ESMS analysis of reaction mixtures. This material is available free of charge via the Internet at http://pubs.acs.org.

The authors declare no competing financial interest.

REFERENCES

1. Kelland L. The resurgence of platinum-based cancer chemotherapy. Nat. Rev. Cancer. 2007;7:573–784. [PubMed]
2. Koberle B, Tomicic MT, Usanova S, Kaina B. Cisplatin resistance: preclinical findings and clinical implications. Biochim. Biophys. Acta. 2010;1806:172–182. [PubMed]
3. Einhorn LH. Testicular cancer: An oncological success story. Clin. Cancer Res. 1997;3:2630–2632. [PubMed]
4. Archer VR, Billingham LJ, Cullen MH. Palliative chemotherapy: no longer a contradiction in terms. Oncologist. 1999;4:470–477. [PubMed]
5. Fujii T, Toyooka S, Ichimura K, Fujiwara Y, Hotta K, Soh J, Suehisa H, Kobayashi N, Aoe M, Yoshino T, Kiura K, Date H. ERCC1 protein expression predicts the response of cisplatin-based neoadjuvant chemotherapy in non-small-cell lung cancer. Lung Cancer. 2008;59:377–384. [PubMed]
6. Cobo M, Isla D, Massuti B, Montes A, Sanchez JM, Provencio M, Vinolas N, Paz-Ares L, Lopez-Vivanco G, Munoz MA, Felip E, Alberola V, Camps C, Domine M, Sanchez JJ, Sanchez-Ronco M, Danenberg K, Taron M, Gandara D, Rosell R. Customizing cisplatin based on quantitative excision repair cross-complementing 1 mRNA expression: a phase III trial in non-small-cell lung cancer. J. Clin. Oncol. 2007;25:2747–2754. [PubMed]
7. Oliver TG, Mercer KL, Sayles LC, Burke JR, Mendus D, Lovejoy KS, Cheng MH, Subramanian A, Mu D, Powers S, Crowley D, Bronson RT, Whittaker CA, Bhutkar A, Lippard SJ, Golub T, Thomale J, Jacks T, Sweet-Cordero EA. Chronic cisplatin treatment promotes enhanced damage repair and tumor progression in a mouse model of lung cancer. Genes Dev. 2010;24:837–852. [PubMed]
8. Momekov G, Bakalova A, Karaivanova M. Novel approaches towards development of non-classical platinum-based antineoplastic agents: Design of platinum complexes characterized by an alternative DNA-binding pattern and/or tumor-targeted cytotoxicity. Curr. Med. Chem. 2005;12:2177–2191. [PubMed]
9. Komeda S. Unique platinum-DNA interactions may lead to more effective platinum-based antitumor drugs. Metallomics. 2011;3:650–655. [PubMed]
10. Guddneppanavar R, Bierbach U. Adenine-N3 in the DNA minor groove - an emerging target for platinum containing anticancer pharmacophores. Anticancer Agents Med. Chem. 2007;7:125–138. [PubMed]
11. Graham LA, Wilson GM, West TK, Day CS, Kucera GL, Bierbach U. Unusual Reactivity of a Potent Platinum-Acridine Hybrid Antitumor Agent. ACS Med. Chem. Lett. 2011;2:687–691. [PMC free article] [PubMed]
12. Ma Z, Choudhury JR, Wright MW, Day CS, Saluta G, Kucera GL, Bierbach U. A non-cross-linking platinum-acridine agent with potent activity in non-small-cell lung cancer. J. Med. Chem. 2008;51:7574–7580. [PMC free article] [PubMed]
13. Qiao X, Zeitany AE, Wright MW, Essader AS, Levine KE, Kucera GL, Bierbach U. Analysis of the DNA damage produced by a platinum-acridine antitumor agent and its effects in NCI-H460 lung cancer cells. Metallomics. 2012;4:645–652. [PMC free article] [PubMed]
14. Kostrhunova H, Malina J, Pickard AJ, Stepankova J, Vojtiskova M, Kasparkova J, Muchova T, Rohlfing ML, Bierbach U, Brabec V. Replacement of a thiourea with an amidine group in a monofunctional platinum-acridine antitumor agent. Effect on DNA interactions, DNA adduct recognition and repair. Mol. Pharmaceutics. 2011;8:1941–1954. [PMC free article] [PubMed]
15. Smyre CL, Saluta G, Kute TE, Kucera GL, Bierbach U. Inhibition of DNA Synthesis by a Platinum-Acridine Hybrid Agent Leads to Potent Cell Kill in Non-Small Cell Lung Cancer. ACS Med. Chem. Lett. 2011;2:870–874. [PMC free article] [PubMed]
16. Alley SC, Okeley NM, Senter PD. Antibody-drug conjugates: targeted drug delivery for cancer. Curr. Opin. Chem. Biol. 2010;14:529–537. [PubMed]
17. Ma Y, Day CS, Bierbach U. Synthesis, structure, and reactivity of monofunctional platinum(II) and palladium(II) complexes containing the sterically hindered ligand 6-(methylpyridin-2-yl)acetate. J. Inorg. Biochem. 2005;99:2013–2023. [PubMed]
18. Manzotti C, Pratesi G, Menta E, Di Domenico R, Cavalletti E, Fiebig HH, Kelland LR, Farrell N, Polizzi D, Supino R, Pezzoni G, Zunino F. BBR 3464: A novel triplatinum complex, exhibiting a preclinical profile of antitumor efficacy different from cisplatin. Clin. Cancer Res. 2000;6:2626–2634. [PubMed]
19. Zhu L, Kostic NM. Toward artificial metalloproteases: Mechanisms by which platinum(II) and palladium(II) complexes promote selective, fast hydrolysis of unactivated amide bonds in peptides. Inorg. Chem. 1992;31:3994–4001.
20. Wang X, Guo Z. The role of sulfur in platinum anticancer chemotherapy. Anticancer Agents Med. Chem. 2007;7:19–34. [PubMed]
21. Brabec V, Kleinwachter V, Butour J-L, Johnson NP. Biophysical studies of the modification of DNA by antitumor platinum coordination-complexes. Biophys. Chem. 1990;35:129–141. [PubMed]
22. Baruah H, Rector CL, Monnier SM, Bierbach U. Mechanism of action of non-cisplatin type DNA-targeted platinum anticancer agents: DNA interactions of novel acridinylthioureas and their platinum conjugates. Biochem. Pharmacol. 2002;64:191–200. [PubMed]
23. Choudhury JR, Bierbach U. Characterization of the bisintercalative DNA binding mode of a bifunctional platinum-acridine agent. Nucleic Acids Res. 2005;33:5622–5632. [PMC free article] [PubMed]
24. Choudhury JR, Rao L, Bierbach U. Rates of intercalator-driven platination of DNA determined by a restriction enzyme cleavage inhibition assay. J. Biol. Inorg. Chem. 2011;16:373–380. [PMC free article] [PubMed]
25. Baskin JM, Prescher JA, Laughlin ST, Agard NJ, Chang PV, Miller IA, Lo A, Codelli JA, Bertozzi CR. Copper-free click chemistry for dynamic in vivo imaging. Proc. Natl. Acad. Sci. USA. 2007;104:16793–16797. [PubMed]
26. Godwin AK, Meister A, O'Dwyer PJ, Huang CS, Hamilton TC, Anderson ME. High resistance to cisplatin in human ovarian cancer cell lines is associated with marked increase of glutathione synthesis. Proc. Natl. Acad. Sci. USA. 1992;89:3070–3074. [PubMed]
27. Yde CW, Issinger OG. Enhancing cisplatin sensitivity in MCF-7 human breast cancer cells by down-regulation of Bcl-2 and cyclin D1. Int. J. Oncol. 2006;29:1397–1404. [PubMed]
28. Rejiba S, Bigand C, Parmentier C, Hajri A. Gemcitabine-based chemogene therapy for pancreatic cancer using Ad-dCK::UMK GDEPT and TS/RR siRNA strategies. Neoplasia. 2009;11:637–650. [PMC free article] [PubMed]
29. Morgan MA, Parsels LA, Parsels JD, Mesiwala AK, Maybaum J, Lawrence TS. Role of checkpoint kinase 1 in preventing premature mitosis in response to gemcitabine. Cancer Res. 2005;65:6835–6842. [PubMed]
30. Wang RE, Costanza F, Niu Y, Wu H, Hu Y, Hang W, Sun Y, Cai J. Development of self-immolative dendrimers for drug delivery and sensing. J. Control. Rel. 2012;159:154–163. [PubMed]
31. Gelasco A, Lippard SJ. Anticancer activity of cisplatin and related complexes. In: Clarke MJ, Sadler PJ, editors. Topics in Biological Inorganic Chemistry. New York: Springer; 1999. pp. 1–43.
32. Dhara SC. A rapid method for the synthesis of cis-[Pt(NH3)2Cl2] Indian J. Chem. 1970;8:193–194.
33. Augustus TM, Anderson J, Hess SM, Bierbach U. Bis(acridinylthiourea)platinum(II) complexes: synthesis, DNA affinity, and biological activity in glioblastoma cells. Bioorg. Med. Chem. Lett. 2003;13:855–858. [PubMed]
34. Martins ET, Baruah H, Kramarczyk J, Saluta G, Day CS, Kucera GL, Bierbach U. Design, synthesis, and biological activity of a novel non-cisplatin-type platinum-acridine pharmacophore. J. Med. Chem. 2001;44:4492–4496. [PubMed]
35. DeLeys RJ, Jackson DA. Electrophoretic analysis of covalently closed SV40 DNA: Boltzmann distributions of DNA species. Nucleic Acids Res. 1976;3:641–652. [PMC free article] [PubMed]