Elucidation of the Molecular Interaction between Cisplatin and

Jan 29, 2013 - McGuire , W. P.; Hoskins , W. J.; Brady , M. F.; Kucera , P. R.; Partridge , E. E.; Look , K. Y.; Clarke-Pearson , D. L.; Davidson , M...
0 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/jmc

Elucidation of the Molecular Interaction between Cisplatin and Flavonol(s) and their Effect on DNA Binding Theodore J. Zwang, Kavisha Singh, Malkiat S. Johal, and Cynthia R. Selassie* Department of Chemistry, Pomona College, 645 North College Avenue, Claremont, California 91711, United States ABSTRACT: Combination therapy of cisplatin with flavonols is a promising treatment for increasing the efficacy of cisplatin when combating cancer. However, little is known about the molecular interactions between cisplatin and flavonols. The data herein helps to elucidate this interaction. Spectrophotometric data in the UV− visible range indicates that hydroxyl groups on the B-ring of flavonols are essential for reactivity with cisplatin. The use of a quartz crystal microbalance with dissipation monitoring approach clearly supports the critical role played by B-ring hydroxyls in their interactions with a cisplatin-bound double-stranded DNA surface; an increase in the number of hydroxyl groups on the B-ring of flavonols parallels the increase in their reaction rates with cisplatin and correlates well with their reported effects on leukemia cell apoptosis efficacy. This study underscores the importance of B-ring hydroxyls in cisplatin’s toxicity and may be used to better understand and improve combination therapies of flavonols with cisplatin.

■

INTRODUCTION Despite advances in the multimodal management of a wide spectrum of human cancers, the full clinical utility of cisdiaminedichloro-platinum(II) (cisplatin, cis-[PtCl2(NH3)2]) is limited because of acute and chronic nephrotoxicity, ototoxicity, and peripheral neuropathy.1−7 In addition to these side effects, one of cisplatin’s greatest setbacks is the development of resistance that significantly reduces the efficacy of continued treatment.8−12 Combination therapy with other drugs has been proposed to overcome early onset resistance to cisplatin. This has recently been explored through various studies combining cisplatin with palitaxel/SN-38,13 new classes of platinum drugs,14 multinuclear platinum compounds,15 and various flavonols.16 In some cases, combinations have shown increased efficacy over individual treatments17 and reduced toxicity in cultured renal cells.18 The combination of cisplatin with flavonols is particularly promising in light of recent studies which found an increased level of cell apoptosis upon treatment with cisplatin−flavone derivatives.16,19,20 The polyphenolic structure of flavonols enhances their ability to accept free radicals or complex with metal ions.21 Studies have also shown that flavonols can intercalate into deoxyribonucleic acid (DNA) as well as covalently bind to DNA and other proteins.22,23 As a result of these characteristics, flavonols have complex biological interactions, and combinations of them with cisplatin are poorly understood, although multiple theories currently exist that try to explain the interactions between cisplatin and flavonols.24−29 While some flavonols such as quercetin and luteolin have been found to improve the efficacy and cytotoxicity of cisplatin, other flavonols, such as apigenin, galangin, and chrysin, have been shown to have a negative effect on cisplatin’s cytotoxicity.24,25 The mechanistic underpinnings of this change in efficacy is not known but presents a potentially © 2013 American Chemical Society

interesting avenue for modulating cisplatin’s efficacy. The data shows a trend between the hydroxylation of the flavonol and increases in the efficacy of cisplatin. Unfortunately, the structures of cisplatin−flavonol complexes have not been well characterized, which makes it difficult to understand the structure−activity relationships. Some of the most studied flavonol−metal complexes are those of quercetin (Figure 1) and metal(II) centers. Quercetin forms complexes with copper(II), nickel(II), zinc(II), cobalt(II), manganese(II), and lead(II).30−32 One study suggests that the quercetin acts as a bidentate ligand for metal centers, tentatively forming an octahedral complex that involves its

Figure 1. (A) Structure of flavonols with changes in hydroxylation in the B-ring. (B) Possible cisplatin binding modes for quercetin.30−34 Received: July 23, 2012 Published: January 29, 2013 1491

dx.doi.org/10.1021/jm3016798 | J. Med. Chem. 2013, 56, 1491−1498

Journal of Medicinal Chemistry

Article

(MW = 25000 g/mol, mixture of linear and branched chains, water ≤1%), silver nitrate (AgNO 3: ACS reagent grade, ≥99%), dimethylsulfoxide (DMSO, ≥99%), and calf thymus DNA sodium salt were obtained from Sigma Aldrich and used without further treatment. Phosphate buffered saline (PBS) at pH 7.4 was obtained from Fischer. PEI solutions were prepared at a concentration of 10 mM (using the molecular weight of the polymer’s repeat unit) with ultrapure water (resistivity >18 MΩ cm, Milli-Q) at pH 7. DNA solutions were also prepared at a concentration of 0.1 mg/mL in a PBS solution with 5% DMSO by volume following the procedure of Rawle et al.36 The flavonol solutions were all prepared at a concentration of 0.1 mM PBS with 5% DMSO by volume because DMSO was required to ensure the solubility of the polyphenolic compounds. The cisplatin solutions were prepared at 2.56 mM in PBS with 5% DMSO by volume. Activated cisplatin ([Pt(NH3)2(H2O)2]+2) solutions were prepared by adding twice the molar amount of AgNO3 and allowing the precipitation reaction to complete over a time period of 4−7 days. The precipitate of AgCl was removed by multiple vacuum filtrations over this period of time, and finally a clear solution of activated cisplatin was obtained. Minimal HCl or NaOH was added as needed to flavonol or cisplatin solutions to maintain pH 7.4 before use. UV−vis Experiments. All UV−vis spectroscopic work was carried out on a Cary 300 UV−visible spectrophotometer (Varian). The cisplatin and flavonols were mixed in 50 mL centrifuge tubes at an equal molar ratio of cisplatin to flavonol and then incubated at 37 °C. The spectra were taken in quartz cuvettes and the solutions of flavonols, cisplatin, and cisplatin−flavonol mixtures were directly analyzed by pipetting 2 mL into the cuvettes at various time points. The reference solution used for these measurements was a solution of PBS with 5% DMSO by volume. QCM-D Experiments. A quartz crystal microbalance with dissipation monitoring (QCM-D) (E4, Q-Sense, Gothenburg, Sweden) was used to obtain all the frequency (F) and dissipation (D) data of the DNA, cisplatin, flavonols, and PEI surfaces. The QCMD sensor consisted of an AT-cut piezoelectric quartz crystal disk, coated with an Au electrode (100 nm thick) on the back and an active surface layer of SiO2 (∼50 nm thick). The QCM-D sensor crystal (14 mm × 0.3 mm, active area of 0.2 cm2) operated at a frequency of 4.95 MHz ± 50 kHz. The crystals were optically polished with a surface roughness of less than 3 nm (root-mean-square). The active side was in contact with the solutions of PEI, DNA, and the experimental solutions. The crystal was mounted in a flow cell with a total volume of 40 μL. Multiple overtone values were collected, but only the fifth overtone values for frequency and dissipation are reported herein. At this overtone, the frequency values were obtained with accuracy of ±0.2 Hz or less, and the dissipation factor was obtained with accuracy of ±0.2 × 10−6 or less. F and D values were measured relative to a baseline obtained in ultrapure water. Before usage, the SiO2-coated quartz crystals were exposed to a 2% Hellmanex solution (Hellma Co.) for 30 min, followed by an ethanol rinse and an ultrapure water rinse. The crystals were dried with N2 and treated by UV/ozone for 10 min and then placed in the QCM-D flow cells. A stable baseline for frequency and dissipation was obtained by flushing the QCM-D flow cells with ultrapure water. The sensor surface was then exposed to PEI in order to form a substrate onto which the DNA surface could form. The various other solutions were passed through the QCM-D flow cells as indicated. In between experimental solutions, the sensor surface was rinsed to remove any weakly attached adsorbents and to ensure that the observed changes in F and D were not due to changes in the solutions to which the surfaces were being exposed. The flow rate for each experiment was 100 μL/ min, and all QCM-D experiments were conducted at 23 °C.

CO group and the 3-OH oxygen atoms coordinating with the platinum.30 Lemanska et al., have shown that the low pKa (7.03) of the 4′-OH moiety in the B-ring results in easy deprotonation that is facilitated by charge delocalization of the anion to the C4 keto moiety. The 7-OH in the A-ring is also susceptible to deprotonation at physiological pH.33 This implies that these two sites are also capable of forming a complex with cisplatin. Another study on the absorption spectra of metal complexes of quercetin suggests that the metal(II) center complexes with the 3′-OH and 4′-OH sites of the B-ring of quercetin.34 To determine which sites of flavonols play a role in cisplatin complexation, we examined the UV−vis spectra of cisplatin with four flavonols that have different hydroxylation patterns on the B-ring, only. We also used a quartz crystal microbalance with dissipation monitoring (QCM-D) approach to assess the effects of hydroxylation patterns on the flavonols abilities to complex with cisplatin that is bound to double-stranded (ds) DNA. This study also examines how exposure of DNA to quercetin and cisplatin in series differs from using a cisplatin− quercetin complex concurrently. QCM-D is a gravimetric technique that involves the measurement of changes in resonance frequency and energy dissipation upon the adsorption of a material to a sensor surface. Essentially, the sensor vibrates with some constant amount of energy. Because the amount of energy in the system remains constant, the frequency of the oscillations changes any time mass is added to the surface. Small changes in the amount of adsorbed mass can be measured by relating changes in frequency to the amount of mass adsorbed by the Sauerbrey equation.35 However, a decrease in frequency does not always indicate an increase in mass on the surface. Energy can be lost in other ways that would also decrease the frequency at which the sensor oscillates. This is measured by changes in dissipation, which give a qualitative measure of how rigidly bound the adsorbent is. Increases in dissipation indicate more energy is being lost to the surrounding environment due to a relatively more “floppy” layer. Formation of a DNA surface on the QCM-D sensor allows for the measurement of relative amounts of flavonol−cisplatin combinations that bind to DNA as well as the rigidity of that binding. QCM-D has previously been used to study interactions of molecules with DNA surfaces that are similar to those explored herein, including quercetin−copper(II) induced DNA damage.36 The ability of flavonols to modulate cisplatin’s efficacy suggests that an understanding of the interactions of flavonols with cisplatin could greatly benefit the development of improved chemotherapy regimens. This study elucidates whether cisplatin and various flavonols undergo complexation, the types of interactions that are involved, and their effects on binding to dsDNA at the molecular level. UV−vis spectrophotometry in combination with a QCM-D approach allows us to examine the molecular interactions of quercetin and its analogues, shown in Figure 1, with cisplatin that is both free and bound to DNA in a noncellular environment.

■

■

MATERIALS AND METHODS

RESULTS Flavonol−Cisplatin Spectra. Upon the addition of cisplatin, the quercetin solution gradually changes color. As can be seen in Figure 2, this change in color coincides with a decrease in the characteristic absorbance of quercetin at 374 and 274 nm as well as the appearance of a new peak at 293 nm

Materials. All chemicals, unless otherwise stated, were purchased from commercial sources and used without further purification. cisDiaminedichloro-platinum(II) was purchased from Sigma-Aldrich and was 99.9+%, metals basis. The flavanols, kaempferol, quercetin, and myricetin, and galangin are ≥95% pure by HPLC, as certified by Sigma-Aldrich. High molecular weight poly(ethylenimine) (PEI) 1492

dx.doi.org/10.1021/jm3016798 | J. Med. Chem. 2013, 56, 1491−1498

Journal of Medicinal Chemistry

Article

Figure 3. UV−visible absorption spectra of kaempferol incubated with cisplatin over time. Each spectrum starting at black and progressing to purple represents an additional hour of incubation. The gray line is the spectrum after 20 h of incubation. The absorbance peaks at 376 and 271 nm decrease with time, and the absorbance peak at 333 nm increases with time. Two isosbestic points are observed at 316 and 345 nm.

Figure 2. UV−visible absorption spectra of quercetin incubated with cisplatin over time. Each spectrum starting at black and progressing to purple represents an additional hour of incubation. The gray line is the spectrum after 20 h of incubation. The absorbance peaks at 374 and 274 nm decrease with time, and the absorbance peak at 292 increases with time. Two isosbestic points are observed at 312 and 280 nm.

that increases in absorbance over time. Because the absorbance peak at 293 nm is not associated with the spectra of either cisplatin or quercetin, it suggests the formation of a new species. The spectra of quercetin alone had no noticeable change over the same time period, so the peak at 293 nm is unlikely to be due to a solvent interaction or quercetin’s degradation over time. An isosbestic point at 280 nm indicates that there are two distinct species in solution and that the newly formed species has the same molar absorptivity as quercetin at 280 nm. Therefore, the existence of an isosbestic point predicates the formation of a single new species that is driven by a specific interaction between cisplatin and quercetin because if a third species were formed the spectra would be expected to intersect at varying wavelengths as the concentrations change.37 Furthermore, because the molar absorptivity is shared between the two species, this suggests that much of the structure of quercetin remains unaffected by the addition of cisplatin. A second isosbestic point is also observed at 312 nm. The changes in spectra of the quercetin solution were found to be identical when using either activated cisplatin or its chlorinated form. Addition of cisplatin to analogues of quercetin with differently hydroxylated B-rings exhibited similar results. Kaempferol, with only a single hydroxyl at the 4′ position of the B-ring, has two absorbance maxima at 376 and 271 nm (Figure 3).38 After mixing with cisplatin, a new peak slowly emerged at 332 nm and two isosbestic points are observed at 316 and 345 nm. Myricetin has three hydroxyls on the B-ring at the 3′, 4′, and 5′ positions. In Figure 4, it has two absorbance maxima at 332 and 490 nm.38 On addition of cisplatin, a new peak forms quickly at 292 nm and an isosbestic point is observed at 330 nm. The spectra of both kaempferol and myricetin without cisplatin have no noticeable change over the same time period. These results indicate that cisplatin undergoes a site-specific interaction with both kaempferol and

Figure 4. UV−visible absorption spectra of myricetin incubated with cisplatin over time. Each spectrum starting at black and progressing to purple represents an additional hour of incubation. The gray line is the spectrum after 20 h of incubation. The absorbance peaks at 332 and 490 nm decrease with time, and the absorbance peak at 292 increases with time. An isosbestic point is observed at 330 nm.

myricetin that produces a single product, similar to its interaction with quercetin. Interestingly, galangin does not behave in the same manner as the other analogues. Galangin has no hydroxyl groups attached to the B-ring. In Figure 5, it has an absorbance maximum at 268 and 368 nm39 that shows no change relative to the rest of its spectra upon mixing with cisplatin. Additionally, no new peaks are formed and no isosbestic point is observed. Also unlike the other flavonols, galangin’s entire spectrum shows a decrease in absorbance over time in 1493

dx.doi.org/10.1021/jm3016798 | J. Med. Chem. 2013, 56, 1491−1498

Journal of Medicinal Chemistry

Article

Figure 6. QCM-D response to the binding of cisplatin to a dsDNA surface, with subsequent exposure to quercetin. Dotted lines represent PBS rinses, which remove weakly adhered molecules. The black line is the change in frequency, which corresponds to the left y-axis. The gray line is the change in the energy dissipation, which corresponds to the right y-axis. The asterisk indicates the end of the activated cisplatin− quercetin sequence, which resulted in a 13 Hz decrease. The subsequent rinse did not result in any significant change in frequency or dissipation.

Figure 5. UV−visible absorption spectra of galangin incubated with cisplatin over time. Each spectrum starting at black and progressing to purple represents an additional hour of incubation. The gray line is the spectrum after 20 h of incubation. The absorbance decreases over the 20 h incubation. No new peaks were formed, nor were any peaks’ absorbance diminished relative to the rest of the spectrum.

the absence or presence of cisplatin. This is likely due to its decreased solubility in water, which is consistent with its hydrophobicity; the other flavonols are more hydrophilic (see Table 1). The absence of a new absorbance peak suggests that the production of a second species/complex does not occur when galangin is mixed with cisplatin. Drug Interactions with dsDNA Surface. In Figure 6, a dsDNA surface was formed on the QCM-D sensor and first exposed to cisplatin, resulting in a frequency decrease of 5 Hz and a corresponding increase in dissipation. This indicates binding of cisplatin to the DNA. Subsequent addition of quercetin causes a further 10 Hz decrease in frequency and increase in dissipation. In Figure 7, the quercetin solution was flowed over a dsDNA surface prior to cisplatin binding; no mass deposition was observed upon exposure to quercetin and the same 5 Hz decrease in frequency and increase in dissipation was observed for the following addition of cisplatin. The nonbinding of quercetin with a dsDNA surface agrees with previous data.36 These results indicate that the degree of quercetin deposited in Figure 6 can be attributed entirely to its interactions with cisplatin that was already bound to the DNA surface. In Figure 8, activated ([Pt(NH3)2(H2O)2]+2) cisplatin was flowed over the dsDNA surface followed by quercetin. Exposure of the dsDNA to activated cisplatin results in a 30 Hz decrease in frequency, which indicates that approximately 6fold more activated cisplatin bound to the dsDNA surface than chlorinated ([Pt(NH3)2(Cl)2]) cisplatin. It should be noted

Figure 7. QCM-D response to the binding of quercetin to a dsDNA surface, with subsequent exposure to cisplatin. Dotted lines represent PBS rinses, which remove weakly adhered molecules. The black line is the change in frequency, which corresponds to the left y-axis. The gray line is the change in the energy dissipation, which corresponds to the right y-axis.

that subsequent addition of quercetin increased the frequency and decreased the dissipation; the complexation of quercetin in

Table 1. UV−Vis Spectroscopy Results from Incubation of Various Flavonols with an Equimolar Concentration of Cisplatin flavonoid

no. of hydroxyls on Bring

myrcetin quercetin kaempferol galangin

3 2 1 0

initial peaks (nm) 332, 374, 376, 368,

490 274 271 268

isosbestic point (nm)

new peaks (nm)

ΔAbs ((abs/mmol−1)/ (20 h−1))

relative initial rate

ClogP

330 280, 312 316, 345

292 293 332

0.569 0.192 0.114 0

3.31 2.02 1 0

0.84 1.50 2.10 2.76

1494

dx.doi.org/10.1021/jm3016798 | J. Med. Chem. 2013, 56, 1491−1498

Journal of Medicinal Chemistry

Article

Figure 9. QCM-D response to the binding of quercetin to a dsDNA surface, with subsequent exposure to activated cisplatin. Dotted lines represent PBS rinses, which remove weakly adhered molecules. The black line is the change in frequency, which corresponds to the left yaxis. The gray line is the change in the energy dissipation, which corresponds to the right y-axis.

Figure 8. QCM-D response to the binding of activated cisplatin to a dsDNA surface, with subsequent exposure to quercetin. Dotted lines represent PBS rinses, which remove weakly adhered molecules. The black line is the change in frequency, which corresponds to the left yaxis. The gray line is the change in the energy dissipation, which corresponds to the right y-axis. The asterisk indicates the end of the activated cisplatin−quercetin sequence, which resulted in a 12 Hz decrease. The subsequent rinse resulted in a 6 Hz increase in frequency and a significant decrease in dissipation, which suggests the loss of weakly adsorbed molecules from the surface.

quercetin is effectively tearing the activated cisplatin away from the surface by binding to it rather than to the DNA, by demonstrating that none of the activated cisplatin is rinsed off in the quercetin-activated cisplatin sequence. Incubated solutions of cisplatin−quercetin (Figure 10) and activated cisplatin−quercetin (Figure 11) were also exposed to

solution to bound cisplatin appears to be tearing some of the cisplatin away from the dsDNA. It is prudent to compare Figures 8 and 6 to see that the total decrease in frequency caused by the deposition of activated cisplatin followed by quercetin is the same as the decrease in frequency caused by the deposition of chlorinated cisplatin followed by quercetin. However, the subsequent rinse increases the frequency and decreases the dissipation for the dsDNA exposed to the activated cisplatin−quercetin sequence but does not affect the dsDNA exposed to the cisplatin−quercetin sequence. The increase in frequency and decrease in dissipation are characteristic of the loss of weakly adsorbed molecules from the surface. This suggests that activated cisplatin binds to dsDNA in both high- and low-affinity modes while chlorinated cisplatin only binds in high-affinity modes. This is reflected by the respective changes in dissipation during the initial deposition of the cisplatins: the larger increase in dissipation for activated cisplatin binding suggests floppiness or some incidence of weaker binding. The access to low-affinity binding sites of activated cisplatin but not the chlorinated form is likely caused by the activated form’s positive charge, whereas the chlorinated form does not have access to these sites because it is neutral. The electrostatic forces between the positively charged cisplatin and the negatively charged dsDNA surface could alter the affinity of activated cisplatin and allow it access to otherwise unfavorable binding sites. It appears that when cisplatin binds to a high-affinity binding site, quercetin binds to it and the complex remains bound to the dsDNA surface. However, when cisplatin is at a low-affinity binding site, quercetin binds to it and pulls it away from the dsDNA. On subsequent rinsing, the dissipation drops dramatically and the frequency increases but remains slightly depressed, indicating the removal of weakly bound cisplatin while only the tightly bound cisplatin remains attached to the dsDNA. Figure 9 supports this claim, that

Figure 10. QCM-D response to the binding of the quercetin−cisplatin complex to a dsDNA surface. Dotted lines represent PBS rinses, which remove weakly adhered molecules. The black line is the change in frequency, which corresponds to the left y-axis. The gray line is the change in the energy dissipation, which corresponds to the right y-axis.

a dsDNA surface because the UV−vis data indicates that these compounds might be undergoing complexation over time. Indeed, this resulted in a significant frequency decrease of 15 Hz with no change in dissipation for the chlorinated form and a decrease of 20 Hz with a significant drop in dissipation for the activated form. This difference in frequency indicates that the complexes formed between the two forms of cisplatin and quercetin both possess the ability to bind to dsDNA and that the binding causes a frequency change similar to that which occurred when cisplatin and quercetin were exposed to the surface, in series. However, the change in dissipation is 1495

dx.doi.org/10.1021/jm3016798 | J. Med. Chem. 2013, 56, 1491−1498

Journal of Medicinal Chemistry

Article

Figure 13. QCM-D response to the binding of cisplatin to a dsDNA surface, with subsequent exposure to kaempferol. Dotted lines represent PBS rinses, which remove weakly adhered molecules. The black line is the change in frequency, which corresponds to the left yaxis. The gray line is the change in the energy dissipation, which corresponds to the right y-axis.

Figure 11. QCM-D response to the binding of the activated cisplatin− quercetin complex to a dsDNA surface. Dotted lines represent PBS rinses, which remove weakly adhered molecules. The black line is the change in frequency, which corresponds to the left y-axis. The gray line is the change in the energy dissipation, which corresponds to the right y-axis.

hydroxyl group on the B-ring, respectively, resulted in a significant mass deposition and increase in dissipation comparable to those for quercetin. Galangin (Figure 14),

distinctly different when the incubated samples are used together compared to the change in dissipation when the compounds are not incubated and used in series. Usually, the deposition of mass on a crystal surface is accompanied by an increase in floppiness and dissipation, but this is not observed with the incubated solutions. This suggests that the mixtures bind very tightly to the dsDNA surface, and in the case of the activated cisplatin complex, its binding appears to be making the overall film significantly more rigid. This is consistent with the rigidification of DNA that would be expected from an intercalative binding mode. To further explore the effect of hydroxyl groups on the Bring of flavonols, quercetin analogues were exposed to a dsDNA surface with bound cisplatin. Myricetin (Figure 12) and kaempferol (Figure 13), with three hydroxyl groups and one

Figure 14. QCM-D response to the binding of cisplatin to a dsDNA surface with subsequent exposure to galangin. Dotted lines represent PBS rinses, which remove weakly adhered molecules. The black line is the change in frequency, which corresponds to the left y-axis. The gray line is the change in the energy dissipation, which corresponds to the right y-axis.

with no hydroxyl groups on the B-ring, resulted in no significant mass deposition. This further strengthens the argument that phenolic, hydroxyl groups on the B-ring are essential for the binding of flavonols to cisplatin and that includes cisplatin that is bound to dsDNA.

■

DISCUSSION DNA is regarded as the primary biological target of cisplatin, although alternative targets have been investigated.40−43 Once cisplatin enters a cell, the low chloride concentration results in the displacement of chlorine ligands by water to generate the diaqua active form of cisplatin, [Pt(NH3)2(H2O)2]+2, which

Figure 12. QCM-D response to the binding of cisplatin to a dsDNA surface, with subsequent exposure to myricetin. Dotted lines represent PBS rinses, which remove weakly adhered molecules. The black line is the change in frequency, which corresponds to the left y-axis. The gray line is the change in the energy dissipation, which corresponds to the right y-axis. 1496

dx.doi.org/10.1021/jm3016798 | J. Med. Chem. 2013, 56, 1491−1498

Journal of Medicinal Chemistry

Article

cytotoxicity-enhancing reaction between the flavonol and cisplatin is too slow. However, the inability of quercetin and the other flavonols to bind to the cisplatin-free dsDNA on the surface may be confounding this part of the analysis.

binds to DNA more easily than the chlorine-bound form.44 The platinum core binds to the N7 nitrogens of guanosine and adenosine, forming intrastrand adducts that distort the double DNA helix. This is further stabilized by hydrogen bonds between the amine ligands of cisplatin and the phosphate groups of DNA.41 The binding of cisplatin bends DNA and disrupts mitosis by inhibiting the function of replication and transcription factors. This induces early apoptosis and is advantageous for the destruction of tumor cells.45,46 Flavonols have the well-established ability to complex with metal ions.30−32,34 There are multiple possible complexing sites within a flavonol. Our data indicates that only flavonols with Bring hydroxyls complex with cisplatin. This is strong evidence that B-ring hydroxyls are the site of cisplatin complexation. By measuring the absorbance spectrum of a mixture of cisplatin with different flavonols, we have noted that the speed of this reaction increases with the number of hydroxyl groups on the B-ring. Over the first 4 h, the relative rates at which the new peak is formed are 3.31:2.02:1:0 for myricetin:quercetin:kaempferol:galangin. This ratio changes slightly to 3.31:2.35:1:0 if the rates are calculated over the first 8 h. The ratio of relative rates appears to mirror the ratio of hydroxyl groups on the flavonols’ B-ring. This, combined with the presence of only one complex in each reaction mixture, suggests that the complex being formed clearly involves the interaction of hydroxyls on the B-ring with the platinum metal center, which is a structure that shares similarities with a promising new DNA-binding platinum anticancer drug candidate.47 QCM-D results suggest that the enhanced cytotoxicity of the flavonol−cisplatin combination could result either from binding of the flavonols to cisplatin already bound to dsDNA or from the binding of a flavonol−cisplatin complex to the dsDNA. It is worth noting that differences in the rate at which the flavonols bind to cisplatin in dsDNA is much smaller than the differences in the rate at which the flavonols bind to cisplatin in solution. Our results allow us to elucidate the relationship between structure and activity for the phenomena reported by Cipak et al.24 In an elegant study, it was reported that flavonols can differentially modulate the efficacy of cisplatin in the induction of leukemia cell apoptosis. Quercetin and luteolin, both with 2 catecholic-hydroxyls on the B-ring, potentiate the cytotoxicity of cisplatin by increasing cisplatin-induced apoptotic cell death. Apigenin, with a single hydroxyl, and chrysin and galingin, with no hydroxyls, have either minimal effects or antagonistic effects on the cytotoxicity of cisplatin. The effects on apoptotic DNA fragmentation in leukemia cells caused by these flavonols, in decreasing order of magnitude, are quercetin, luteolin, and then apigenin, with chrysin and galingin having no effect.24 These flavonols are also listed in order of decreasing number of B-ring hydroxyls. Our data indicates that a decrease in the number of B-ring hydroxyls will reduce the rate at which flavonols can complex with cisplatin and that flavonol−cisplatin complexes can bind to dsDNA, albeit at different rates. This strongly suggests that the enhanced cytotoxicity may be attributed to the effects of the complex and that increasing the rate of cisplatin−flavonol complex formation by using flavonols such as myricetin with more B-ring hydroxyls may be able to further increase the synergistic effects of flavonols on leukemia cell apoptosis. The potentiation caused by the complex appears to be competing with a chemopreventative effect of flavonols, as observed by the antagonistic effects of apigenin, chrysin, and galingin on cisplatin cytotoxicity,24 which takes precedence when the

■

CONCLUSION Our data indicates that hydroxyl groups on the B-ring of flavanols are essential for interactions between cisplatin and certain flavonols. Increasing the rate of formation of this complex potentiates the efficacy of cisplatin in the induction of leukemia cell apoptosis. This study enhances understanding of the flavonol-modulated efficacy of cisplatin that will aid in the rational discovery of more effective chemotherapeutics. Furthermore, these results indicate that previously reported metal(II) complexes of flavonols are not adequate models for understanding the interaction of flavonols and cisplatin.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: (909) 621-8446. Fax: (909) 607-7726. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS This research was supported in part by the Pomona College Undergraduate Research Program. ABBREVIATIONS USED cisplatin, cis-diaminedichloro-platinum(II); DNA, deoxyribonucleic acid; QCMD, quartz crystal microbalance with dissipation monitoring; dsDNA, double-stranded DNA; PEI, poly(ethyleneimine); AgNO3; silver nitrate; DMSO, dimethylsulfoxide; PBS, phosphate buffered saline; F, frequency; D, dissipation

■

REFERENCES

(1) Rosenberg, B.; Vancamp, L.; Trosko, James, E.; Mansour, Virginia, H. Platinum Compounds: A New Class of Potent Antitumour Agents. Nature 1969, 222, 385−386. (2) Rose, P. G.; Bundy, B. N.; Watkins, E. B.; Thigpen, J. T.; Deppe, G.; Maiman, M. A.; Clarke-Pearson, D. L.; Insalaco, S. Concurrent cisplatin-based radiotherapy and chemotherapy for locally advanced cervical cancer. N. Engl. J. Med. 1999, 340, 1144−1153. (3) Crino, L.; Scagliotti, G.; Ricci, S.; De Marinis, F.; Rinaldi, M.; Gridelli, C.; Ceribelli, A.; Bianco, R.; Marangolo, M.; Di Costanzo, F. Gemcitabine and cisplatin versus mitomycin, ifosfamide, and cisplatin in advanced non-small-cell lung cancer: A randomized phase III study of the Italian Lung Cancer Project. J. Clin. Oncol. 1999, 17, 3522− 3530. (4) McGuire, W. P.; Hoskins, W. J.; Brady, M. F.; Kucera, P. R.; Partridge, E. E.; Look, K. Y.; Clarke-Pearson, D. L.; Davidson, M. Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer. N Engl. J. Med. 1996, 334, 1−6. (5) Burtness, B.; Goldwasser, M. A.; Flood, W.; Mattar, B.; Forastiere, A. A. Phase III randomized trial of cisplatin plus placebo compared with cisplatin plus cetuximab in metastatic/recurrent head and neck cancer: an Eastern Cooperative Oncology Group study. J. Clin. Oncol. 2005, 23, 8646−8654. (6) Treskes, M.; Vijgh, W. J. F. WR-2721 as a modulator of cisplatinand carboplatin-induced side effects in comparison with other chemoprotective agents: a molecular approach. Cancer Chemother. Pharmacol. 1993, 33, 93−106.

1497

dx.doi.org/10.1021/jm3016798 | J. Med. Chem. 2013, 56, 1491−1498

Journal of Medicinal Chemistry

Article

(7) Hamers, F.; Gispen, W.; Neijt, J. Neurotoxic side-effects of cisplatin. Eur. J. Cancer 1991, 27, 372−376. (8) Gately, D.; Howell, S. Cellular accumulation of the anticancer agent cisplatin. Br. J. Cancer 1993, 67, 1171−1176. (9) Aebi, S.; Kuri-Haidar, B.; Gordon, R.; Cenni, B.; Zheng, H.; Fink, D.; Christen, R. D.; Boland, C. R.; Koi, M.; Fishel, R. Loss of DNA mismatch repair in acquired resistance to cisplatin. Cancer Res. 1996, 56, 3087−3090. (10) Godwin, A.; Meister, A.; O’Dwyer, P.; Huang, C.; Hamilton, T.; Anderson, M. High resistance to cisplatin in human ovarian cancer cell lines is associated with marked increase in glutathione synthesis. Proc. Natl. Acad. Sci. U. S. A. 1992, 89, 3070−3074. (11) Kartalou, M.; Essigmann, J. M. Recognition of cisplatin adducts by cellular proteins. Mutat. Res., Fundam. Mol. Mech. Mutagen. 2001, 478, 23−43. (12) Gore, M.; Fryatt, I.; Wiltshaw, E.; Dawson, T.; Robinson, B.; Calvert, A. Cisplatin/carboplatin cross-resistance in ovarian cancer. Br. J. Cancer 1989, 60, 767−769. (13) Komuro, Y.; Udagawa, Y.; Susumu, N.; Aoki, D.; Kubota, T.; Nozawa, S. Paclitaxel and SN-38 Overcome Cisplatin Resistance of Ovarian Cancer Cell Lines by Down-Regulating the Influx and Efflux System of Cisplatin. Cancer Sci. 2001, 92, 1242−1250. (14) Kelland, L, R.; Sharp, S. Y.; O’Neill, C. F.; Raynaud, F. I.; Beale, P. J.; Judson, I. R. Discovery and development of platinum complexes designed to circumvent cisplatin resistance. J. Inorg. Biochem. 1999, 77, 111−115. (15) Perego, P.; Caserini, C.; Gatti, L.; Carenini, N.; Romanelli, S.; Supino, R.; Colangelo, D.; Viano, I.; Leone, R.; Spinelli, S. A novel trinuclear platinum complex overcomes cisplatin resistance in an osteosarcoma cell system. Mol. Pharmacol. 1999, 55, 528−534. (16) Kosmider, B.; Osiecka, R. Flavonoid compounds: a review of anticancer properties and interactions with cis-diamminedichloroplatinum. Drug Dev. Res. 2004, 63, 200−211. (17) Neelam, S. S.; Bernabei, A.; Freedland, C.; Thompson, R.; Coebett, T. H.; Luk, G. D. Combination of flavone acetic acid (FAA) with adriamycin, cis-platinum and difluoromethylornithine (DFMO) in vitro against human colon cancer cells. Invest. New Drugs 1990, 8, 263−268. (18) Kuhlman, M. K.; Horsch, E.; Burkhardt, G.; Wagner, M.; Kohler, H. Reduction of cisplatin toxicity in cultured renal tubular cells by the bioflavonoid quercetin. Arch. Toxicol. 1998, 72, 536−540. (19) Kosmider, B.; Wyszynska, K.; Janik, Spiechowicz, E.; Osiecka, R.; Zyner, E.; Ochoki, J.; Ciesielska, E.; Wasowicz, W. Evaluation of the genotoxicity of cis-bis(3-aminoflavone) dichloroplatinum(II) in comparison with cis-DDP. Mutat. Res., Genet. Toxicol. Environ. Mutagen. 2004, 558, 93−110. (20) Choi, J. A.; Kim, J. Y.; Lee, J. Y.; Kang, C. M.; Kwon, H. J.; Yoo, Y. D.; Kim, T. W.; Lee, Y. S.; Lee, S. J. Induction of cell cycle arrest and apoptosis in human breast cancer cells by quercetin. Int. J. Oncol. 2001, 19, 837−844. (21) Dehghan, G.; Dolatabadi, J. E. N.; Jouyban, A.; Zeynali, K. A.; Ahmadi, S. M.; Kashanian, S. Spectroscopic studies on the interaction of quercetin−terbium(III) complex with calf thymus DNA. DNA Cell Biol. 2011, 30, 195−201. (22) Kanakis, C. D.; Tarantilis, P. A.; Polissiou; Diamantoglou, S.; Tajmir-Riahi, H. A. An overview of DNA and RNA bindings to antioxidant flavonoids. Cell Biochem. Biophys. 2007, 49, 29−36. (23) Janjua, N. K.; Siddiqa, A.; Yaqub, A.; Sabahat, S.; Qureshi, R.; Haque, S. Spectrophotometric analysis of flavonoid−DNA binding interactions at physiological conditions. Spectrochim. Acta, Part A 2009, 74, 1135−1137. (24) Cipak, L.; Novotny, L.; Cipakova, I.; Rauko, P. Differential modulation of cisplatin and doxorubicin efficacies in leukemia cells by flavonoids. Nutr. Res. (N. Y., NY, U. S.) 2003, 23, 1045−1057. (25) Cipak, L.; Rauko, P.; Miadokova, E.; Cipakova, I.; Novotny, L. Effects of flavonoids on cisplatin-induced apoptosis of HL-60 and L1210 leukemia cells. Leuk. Res. 2003, 27, 65−72. (26) Ahmed, M. S.; Ramesh, V.; Nagaraja, V.; Parish, J.; Hadi, S. Mode of binding of quercetin to DNA. Mutagenesis 1994, 9, 193−197.

(27) Cross, H. J.; Tilby, M.; Chipman, J. K.; Ferry, D. R.; Gescher, A. Effect of quercetin on the genotoxic potential of cisplatin. Int. J. Cancer 1996, 66, 404−408. (28) Behling, E. B.; Sendao, M. C.; Francescato, H. D. C.; Antunes, L. M. G.; Costa, R. S.; de Lourdes, P.; Bianchi, M. Comparative study of multiple dosage of quercetin against cisplatin-induced nephrotoxicity and oxidative stress in rat kidneys. Pharmacol. Rep. 2006, 58, 526−532. (29) Devi, P.; Shyamala, D. Protective effect of quercetin in cisplatin induced cell injury in the rat kidney. Indian J. Pharmacol 2009, 31, 422−426. (30) Zhou, J.; Wang, L.; Wang, J.; Tang, N. Antioxidant and antitumour activities of solid quercetin zinc(II) complex. Transition Met. Chem. (Dordrecht, Neth.) 2001, 26, 57−63. (31) Tan, J.; Zhu, L.; Wang, B. DNA binding and cleavage activity of quercetin nickel(II) complex. Dalton Trans. 2009, 24, 4722−4728. (32) Tan, J.; Wang, B.; Zhu, L. DNA binding and oxidative DNA damage induced by a quercetin copper(II) complex: potential mechanism of its antitumor properties. J. Biol. Inorg. Chem. 2009, 14, 727−739. (33) Lemanska, K.; Szymudiak, H.; Tyrakowska, B.; Zielinski, R.; Soffers, A. E. M. F.; Rietjens, I. M. C. M. The influence of pH on antioxidant properties and the mechanism of antioxidant action of hydroxyflavones. Free Radical Biol. Med. 2001, 31, 869−881. (34) Jurd, L.; Geissman, T. Absorption spectra of metal complexes of flavonoid compound. J. Org. Chem. 1956, 21, 1395−1401. (35) Sauerbrey, G. Verwendung von Schwingquarzen zur Wägung dunner Schichten und zur Mikrowägung. Z. Phys. 1959, 155, 206−222. (36) Rawle, R. J.; Johal, M. S.; Selassie, C. R. D. A real-time QCM-D approach to monitoring mammalian DNA damage using DNA adsorbed to a polyelectrolyte surface. Biomacromolecules 2007, 9, 9− 12. (37) Moore, J. W.; Pearson, R. G. Kinetics and Mechanism, 3rd ed.; John Wiley & Sons: New York, 1981. (38) Szostek, B.; Orsaka-Gawrys, J.; Surowiec, I.; Trojanowicz, M. Investigation of natural dyes occurring in historical Coptic textiles by high-performance liquid chromatography with UV−vis and mass spectrometric detection. J. Chromatogr., A 2003, 1012, 179−192. (39) Ahmad, M. S.; Fazal, F.; Rahman, A.; Hadi, S.; Parish, J. H. Activities of flavonoids for the cleavage of DNA in the presence of Cu(II): correlation with generation of active oxygen species. J. . 1992, 13, 605−608. (40) Bose, R. Biomolecular targets for platinum antitumor drugs. Mini-Rev. Med. Chem. 2002, 2, 103−111. (41) Jamieson, E. R.; Lippard, S. J. Structure, recognition and processing of cisplatin DNA adducts. Chem. Rev. 1999, 99, 2467− 2498. (42) Reed, E.; Ozols, R. F.; Tarone, R.; Yuspa, S. H.; Poirier, M. C. The measurement of cisplatin−DNA adduct levels in testicular cancer patients. Carcinogenesis 1988, 9, 1909−1911. (43) Reed, E.; Yuspa, S.; Zwelling, L.; Ozols, R.; Poirier, M. Quantitation of cis-diamminedichloroplatinum II (cisplatin)−DNAintrastrand adducts in testicular and ovarian cancer patients receiving cisplatin chemotherapy. J. Clin. Invest. 1986, 77, 545−550. (44) Ziegler, C. J.; Silverman, A. P.; Lippard, S. J. High-throughput synthesis and screening of platinum drug candidates. J. Biol Inorg. Chem. 2000, 5, 774−783. (45) Trimmer, E. E.; Essigmann, J. M. Cisplatin. Essays Biochem. 1999, 34, 191−211. (46) Sorenson, C. M.; Barry, M. A.; Eastman, A. Analysis of Events Associated with Cell Cycle Arrest at G2 Phase and Cell Death Induced by Cisplatin. J. Natl. Cancer Inst. 1990, 82, 749−755. (47) Park, G. Y.; Wilson, J. J.; Song, Y.; Lippard, S. J. Phenanthriplatin, a monofunctional DNA-binding platinum anticancer drug candidate with unusual potency and cellular activity profile. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 11987−11992.

1498

dx.doi.org/10.1021/jm3016798 | J. Med. Chem. 2013, 56, 1491−1498