Roy, Sudeshna
Citation
Roy, S. (2008, November 25). Synergy of intercalation and coordination binding to design novel DNA-targeting antineoplastic metallodrugs. Retrieved from
https://hdl.handle.net/1887/13281
Version: Corrected Publisher’s Version
License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden
Downloaded from: https://hdl.handle.net/1887/13281
Note: To cite this publication please use the final published version (if applicable).
| 4
DNA and protein interaction of derivatised pyridine and pyrimidine compounds of platinum(II) †
Chapter
Four Pt(II) compounds (C3-C6) have been selected to study the DNA and protein binding. The compounds have been shown (Chapter 3) to display limited activity in cancer cell lines. In addition the free ligands (L6-L7) have also been studied for DNA binding. Calf-thymus DNA and an abundant plasma protein have been taken as models for the two major targets for metallodrug interactions. The modifications in the carrier ligands (chloride or ammine) or ancillary secondary amines have been considered to reveal the mode of interactions. The optimised structures of platinum compounds facilitate to establish simple structure-activity relationship. The average combined effects of coordinative-binding and partial intercalation to DNA are evident from the spectroscopical studies.
To evaluate the permeability of platinum compounds through cellular membrane, partition coefficient has been measured for platinum compounds. This data provides reasonable predictability of antitumour activity against cancer cells when compared with standard platinum reference compounds.
4.1. Introduction
In chapter 3, synthesis, characterisations and in vitro antiproliferative activity of four platinum coordination compounds (C3-C6) has been described; in addition model- base studies for C3 and C5 were performed. In this follow-up chapter, DNA and protein binding interactions are described in details.
Genomic DNA is widely-accepted as the main target for platinum antitumour compounds such as cisplatin, carboplatin and oxaliplatin. In addition, there are several non-DNA platinophiles in the cellular milieu and cellular proteins consist a significant part of these platinum-deactivator agents. Besides the cytotoxicity assay for the new platinum compounds, the interaction with biological targets plays a crucial role to elucidate the mode of action.
DNA is a chiral molecule which is polymorphic in secondary structure. The number of bases per turn, the rise per base pair, the handedness of the helix, the inclination of the bases with respect to the helix axis and even the distance of the nucleobase from the helix axis are affected by the solution condition, salt concentration, temperature or presence of external molecules (e.g., platinum compound or ligands).1, 2 DNA in normal physiological condition exists mainly in right-handed helical B-form with base pairs stacked perpendicular to the helix axis. Two other major polymorphs of DNA are known namely, Z-DNA (left-handed helix) and A-DNA (right-handed helix, but different in base turn angle and helix diameter).3 Additionally, a C-form, P-form and a so- called holiday junction (mobile junction between four strands of DNA) can be observed upon changing solution conditions.3 The structures of three major forms of DNA have been depicted in Fig. 4.1 for reference purposes.
The extensively used metallodrug cisplatin gets activated in physiological condition via hydrolysis (loss of two chlorides) and forms several adducts with DNA bases, specially guanine. The major products are the intrastrand bidentate crosslinks [5'- GpG (~ 70%) and 5'-ApG (~ 15%)], whereas other interstrand crosslinks belong to a minor adduct category.4, 5 These DNA lesions form a significant kink in the DNA strand towards the major groove and they are eventually recognised by the HMG proteins. The coordination binding of cisplatin with the DNA inhibits replication both in vitro and in vivo.6, 7 The antiproliferative property of cisplatin thus proposed to be originating from DNA binding and distortion in DNA structure. The other platinum drugs such as carboplatin and oxaliplatin are also assumed to exhibit antitumour activity, similar to
cisplatin, via DNA binding. A few platinated-DNA crystal structures have been reported.8-17 Other than the coordinative monodentate or bidentate guanosine adducts on the DNA strand, intercalation, groove-binding or electrostatic binding modes have been observed for platinum compounds18-21 Therefore, up to now DNA has been accepted to be the prime intracellular target for cisplatin-like metallodrugs. For other platinum coordination compounds the binding mode is not known a priori.
Figure 4.1. Space-filling models for three major forms of DNA.
To elucidate the mode of action of new platinum compounds, the use of several spectroscopic techniques may be beneficial. Circular dichroism (CD) and ultraviolet- visible (UV-vis) spectroscopy possess extensive potential to display subtle changes in conformation, or electronic transitions in solution, respectively. Therefore, these techniques are extremely useful to probe the interaction of compounds with DNA or proteins. Calf-thymus DNA (CT DNA) and bovine serum albumin (BSA) have been selected as two substrates for this study and their interactions with platinum compounds or ligands have been followed over time.
The typical B-form DNA gives rise to two conservative CD bands in the UV- region. The positive band at 278 nm, which originates from base stacking and the negative band at 246 nm, which arises from the right-handed helicity of DNA are the two characteristic bands. Upon addition of a metal compound or a ligand to the B-DNA
B-DNA Z-DNA A-DNA
solution the changes in these two bands can be recorded. These changes in the overall spectra can indicate the mode of interaction. The typical B-DNA spectrum and the changes in the CD spectra after interaction with a classical intercalator (ethidium bromide) and a coordinative binder (cisplatin) are shown in Fig. 4.2.
UV-vis spectroscopy, based on the absorption of photons of monochromatic light, is applicable in biophysical studies to elucidate DNA-binding properties. When a platinum compound binds to DNA by coordination, hyperchromicity (increase in absorption of the DNA band at 270 nm) is observed.22, 23 After intercalation of a compound with DNA base pairs, hypochromism and a red shift is observed due to the strong π-π stacking interaction between the DNA base pairs and the aromatic part of the compound.22, 24, 25
Figure 4.2. CD spectral change of B-DNA upon addition of (a) ethidium bromide and (b) cisplatin in 10 mM phosphate buffer. The small differences in the blank DNA originate from concentration differences.
Albumin is the most abundant protein in the plasma and as a carrier protein it is responsible for the transport of a variety of compounds, including vitamins, fatty acids and pharmaceuticals.26 Interactions between a metal compound and albumin determine the overall distribution, excretion and differences in activity, toxicity and efficacy.27 Cisplatin can efficiently bind to albumin, since it has a high affinity for the sulfur-rich domains. Consequently, this leads to Pt-methionine adducts, which may act as a platinum reservoir for subsequent DNA platination. Therefore, it is worthwhile to check protein- platinum compound interaction to explain the in vivo efficacy, or on the other hand the deactivation.28
CD and UV-vis spectroscopy have been used to observe interactions between native bovine serum albumin and samples. Bovine serum albumin (BSA) has a high percentage of α-helical structure at physiological pH (=7). The CD spectrum of BSA shows two absorption bands at 208 and 222 nm. The compounds or ligands investigated do not show any absorption in this region in CD. Therefore, this fact can be exploited to follow the metal-protein or ligand-protein interactions. Spectra of cisplatin have also been recorded as a reference compound under identical experimental conditions.
The four platinum compounds and the corresponding ligands have been investigated for their interaction with DNA and BSA. The cisplatin analogues (C3 and C5) have solvolysable chloride ligands and are expected to interact with DNA similarly as cisplatin does. Firstly, the activation of the compounds is assumed via hydrolysis in buffer solution. Subsequently, the aquated positively charged species are attracted to negatively charged DNA-phosphate backbone by electrostatic attraction. The aromatic ligands may facilitate the interactions by sliding partially in between base pairs. The H- bonding amine ligands (L6 and L7) can facilitate the formation of the DNA-compound adduct further. On the other hand, compounds C4 and C6 are positively charged and devoid of any easily hydrolysable ligands. Therefore, these cationic compounds have other possibilities to interact with DNA via (a) electrostatic attraction, (b) hydrophobic interactions with the major or minor groove, (c) H-bonding and (d) intercalation between base pairs. The schematic structures of the studied compounds and ligands are shown in Fig. 4.3.
N N
NH
Pt
Cl Cl
C3
N N
NH
Pt NH3 NH3
(NO3)2
C4
N
N
N
N NH
Pt
Cl Cl
C5
N
N
N
N NH
Pt NH3 NH3
(NO3)2
C6
N N
NH
L6
N
N
N
N NH
L7
Cl Cl
H3N NH3
Pt
Cisplatin
N+
NH2 N
H2 Br
Ethidium bromide
Figure 4.3. Schematic structures of four platinum compounds (C3-C6), two ligands (L6- L7), cisplatin and ethidium bromide.
To predict the probable activity and biodistribution of a new compound, an inherent property was decided to be measured. Initially, if a compound has to reach the DNA or other cellular targets, it has to be lipophilic to cross the lipid bilayer of the cell- membrane via passive diffusion. To calculate the lipophilicity the partition coefficient, P, has to be measured. P is the ratio of the equilibrium concentrations of a compound dissolved in a two-phase system consisting of two immiscible solvents. The partition coefficient was calculated using the following equation (Eqn. 4.1).
octanol water
( ) = log( ) ( ) Partition coefficient P Pt
Pt Equation 4.1
The most used solvents are n-octanol, which is a model of the lipid bilayer of a cell membrane, and water, which represents the plasma inside and outside cells. The lipophilicity of a compound, the partition coefficient, can be calculated by taking the logarithm value of PO/W.29 A compound with a high positive value of log PO/W is lipophilic and can easily cross the cell membrane. A negative partition coefficient indicates hydrophilicity, making the compound preferring to stay in the aqueous solution.
Therefore, blood serum is the most preferred loading compartment for the second type of compounds.
4.2. Experimental 4.2.1. Chemicals
The detailed description of syntheses and characterisations for four platinum compounds (C3-C6) has been given in Chapter 3 (vide sections 3.2 and 3.3, respectively).
L6 was obtained from Sigma Aldrich B.V. (The Netherlands) and L7 has been synthesised following the reported synthetic route with slight modifications.30 Cisplatin was synthesised according to the reported literature.31 Sodium chloride, sodium dihydrogenphosphate (NaH2PO4) and n-octanol were purchased from Merck, Germany.
Disodium hydrogenphosphate (Na2HPO4) was obtained from ACF Chemiefarma NV (The Netherlands) and the water was of MilliQ quality from Millipore, USA. Ethidium bromide (EthBr), Bovine Serum Albumin (BSA) and Calf Thymus DNA (CT DNA) were purchased from Sigma-Aldrich B.V. (The Netherlands). CT DNA was purified following a typical method of size-exclusion chromatography and lyophilisation. The buffer used was 10 mM phosphate buffered saline (PBS) with 50 mM NaCl. The stock solutions for the studied compounds were prepared by dissolving the compounds either in dmso (C3 and
C5) or in milliQ water (cisplatin, C4 and C6) of a concentration of 0.5 M and subsequent dilution in phosphate buffer prior to the experiments has been carried out.
4.2.2. UV-vis spectroscopy
Spectra were recorded at room temperature using a Varian Cary 50 Scan UV- Visible Spectrophotometer. Quartz cuvettes with 1 cm path length were used. The range scanned was 200-800 nm and 10 mM phosphate buffer (pH = 7.0) having an ionic strength (NaCl) of 50 mM was used as blank sample. The DNA sample was prepared by proper dilution with phosphate buffer (pH = 7.0) at room temperature. The concentration of the DNA stock was calculated from A260 (molar extinction coefficient, Є = 6600 M-
1cm-1) per nucleotide phosphate. The BSA stock solution was made in phosphate buffer (10 mM). For titration experiments the required aliquots of samples were added just prior to the experiment and mixed thoroughly. The stock solutions and samples were always protected from external light sources. For titration in the presence of CT DNA, the platinum concentration was kept constant at 50 µM and aliquots from the concentrated DNA solution were added gradually at different ratios. The R values were varied from 0- 10, where R represented the ratio of DNA concentration in nucleotide phosphate units per metal.
4.2.3. Circular Dichroism
Circular dichroism spectra were recorded at 37 °C using a Jasco J-815 Spectropolarimeter equipped with a Jasco PTC-423 S Peltier temperature controller. The used scanning rate was 100 nm min-1 with a response time of 1 second. Spectra were recorded at standard sensitivity (100 mdeg) with a data-pitch of 0.5 nm in the continuous mode. The scanning range was 320-220 nm (for DNA) or 300-180 nm (for BSA). All the spectra are the average of four consequent accumulations. The cuvettes used were Hellma precision cells made of Quartz Suprasil of 2 mm path length. The baseline was corrected using a reference of 10 mM PBS. Volume of sample in the cuvettes was fixed at 700 µL.
A typical sample containing DNA/protein and metal compound in phosphate buffer was incubated at 37 °C and the spectra were collected at the given time intervals.
In addition the free ligands were also investigated under similar conditions. As reference compounds, cisplatin and EthBr were also followed for similar time- and concentration- dependent experiments.
4.2.4. DNA-binding studies
Supercoiled φX174 phage DNA (20 μM base pairs end concentration) was incubated with varying concentrations of the relevant compounds in 10 mM phosphate buffer (pH = 7). All reactions contained a compound:CuSO4:ascorbic acid in a 4:1:16 ratio, unless otherwise noted. Ascorbic acid and CuSO4 together (without added compound) result in unfolding of the DNA and therefore this mixture was used as control for all experiments. The compounds alone do not cleave when incubated with DNA.
Samples were incubated in the dark at 37 °C for 30 min. After incubation, the reaction was quenched by the addition of DNA loading buffer containing 1 mM of the Cu(I) chelator BCDA (bathocuproine disulfonic acid). Samples were run on a 1% agarose gel in TAE buffer, stained with EthBr, and imaged with a BioRad ChemiDoc XRS apparatus.
4.2.5. Partition coefficient measurement
The partition coefficient was determined by a typical water/n-octanol biphasic partition method. The samples were prepared at a concentration of 50 µM in 0.15 M saline by weighing out the appropriate amount of the platinum compounds and adding the solvents. These conditions were found sufficient to prevent rapid hydrolysis of C3 and C5. In all cases 3 mL of saline solution was mixed with 3 mL of saline saturated n- octanol. This biphasic solution mixture was shaken overnight by an automatic shaker and later centrifuged at 13000 rpm for 15 minutes. Then 1 mL aliquots of both layers (top layer: n-octanol and bottom layer: aqueous) were collected separately. The aqueous sample was diluted 20-fold and the total platinum content was determined by ICP-OES.
The n-octanol samples were injected to a FAAS and the total platinum content was determined. Each experiment was duplicated and results have been averaged.
The platinum concentration of samples in octanol was determined using a graphite-oven flameless atomic absorption spectroscope (FAAS), Perkin-Elmer 3100 AAS apparatus, equipped with a platinum hollow cathode lamp and an AS-60 graphite oven autosampler. All the measurements were repeated in triplicate and for each measurement 20 µL of sample was injected directly. The furnace was purged with argon gas and before starting a new sample the equipment was washed thoroughly with 1% HNO3 solution.
The furnace was programmed as mentioned below: drying 200 °C for 100 s, ashing 1300
°C for 60 s (first cycle) and 20 °C for 15 s (second cycle), atomisation and measurement 2650 °C/5 s, purging 2600 °C/5 s. The wavelength of the lamp is 265.9 nm with a slit width of 0.70 nm.
The platinum concentration of samples in the aqueous phase was determined by inductively coupled plasma–optical emission spectroscopy (ICP-OES). Measurements were carried out using a VISTA-MPX CCD Simultaneous ICP-OES instrument equipped with a Varian SPS 3 auto sampler. The samples were diluted 20-fold using a 3% HNO3
solution and 3 mL of sample was injected. The wavelength used was 214.424 nm and each measurement was performed in duplicate.
4.3. Results and discussion 4.3.1. General comments
In the present study four platinum compounds (C3-C6) are presented which have several differences in structures and properties (both chemically and electronically), with the aim to investigate the bio-physical properties and in vitro cytostatic activities. The changes in the ancillary ligands induce modifications of the chemical properties of the Pt complexes which undoubtedly affect their interaction with different biological targets.
These results are useful to correlate the biophysical studies with in vitro cytotoxicity.
4.3.2. UV-vis titration with CT DNA
Electronic spectroscopy in the UV-vis range is a quite convenient method to study the interaction of metal compounds with bio-molecules, like DNA or proteins. The absorption spectra of the four platinum compounds in the presence of increasing amounts of DNA (platinum compound concentration kept constant at 50 µM) are shown in Fig.
4.4.
Compounds C3 and C4 show similar behaviour on absorption spectral titration as both of them are coordinated to a common ligand, L6. The chloride and ammine ligands induce small differences in the spectra at high wave-length absorption. There are four peaks observed at 252, 278, 306 and 336 nm for C3 and at 204, 242, 301 and 333 nm for C4, respectively. Both compounds show a gradual increase in intensity against a blank DNA, which agrees with coordination binding to DNA. A subtle difference in their mode of binding is possible as C3 can bind to DNA in either mono- or bidentate fashion and C4 binds to DNA only by electrostatic and partial intercalation.
Figure 4.4. UV-vis spectral changes of (a) C3, (b) C4, (c) C5 and (d) C6 on addition of CT DNA. Arrows denote the gradual changes on increasing amounts of DNA (10-250 µM) in presence of a constant concentration (50 µM) of platinum compound at room temperature.
The UV-vis spectra of C5 and C6 are also similar in appearance, as both of them contain the ligand L7 with only a very small difference in spectra at a wavelength around 300 nm. There is a prominent peak for C5 at 257 nm and a shoulder at 316 nm, whereas for C6 the prominent peak appears at 258 nm with a shoulder at 318 nm and a very weak peak at 365 nm (which is not present for C5). For both C5 and C6, there is no prominent change in intensity of the 316 nm peak, though a significant increase in intensity at 257 nm peak is noteworthy. For a specific R (R=1), there is an insignificant blue shift (2 nm) (from 257 nm to 255 nm for C5 and from 258 to 256 nm for C6). The hyperchromism exhibited is a strong evidence for strong binding to DNA, but this observation is contrasting when compared C5 with C6. Therefore, the possible binding for C5 might be coordinative binding and for C6 it could be electrostatic binding to the phosphate backbone with partial intercalation.
4.3.3. Circular dichroism with calf thymus DNA
The difference in absorption of polarised light gives rise to a CD spectrum, which can distinguish subtle changes in DNA conformational changes in solution. In biological systems the most abundant form of DNA present is the “B-form”, which gives two characteristic peaks in the scan range of 220-320 nm.32 Upon interaction of metallodrugs with DNA, there are several possibilities of non-coordinative binding modes.3 The platinum antitumour compounds are known to be DNA-targeting drugs, so the interaction with DNA will be helpful to predict their mode of action. Different platinum drugs react with DNA at different fashion33, so CD studies support the prediction of the mode of interaction.
The CD spectra were recorded at a constant DNA concentration of 100 µM and the metal compounds were added in different ratios. After the measurements at time zero, incubation at 37 °C over a 24 h time-period and measurements at several intervals assisted the investigation of slow activation via hydrolysis for C3 and C5.
All the four platinum compounds exhibit significant changes in CD bands, when compared to free DNA of same concentration. The spectral changes of the CT DNA in the CD spectra are shown in Figs. 4.5 and 4.6 with different experimental conditions. C3 and C5 show similar changing patterns whereas C4 and C6 induce similar changes in CD signals. For C3 and C5 the negative band shows more noticeable changes than the positive peak at 275 nm. For C5, two smaller new peaks arise at around at 250 and 290 nm, whereas the negative band shows 3 nm of a red shift (bathochromic shift) with a sharp signal. The increase in the negative band is gradual with increasing amounts of the platinum compound. Thus, clearly base stacking is little disturbed by interaction with C3 and C5, whereas the right-handed helicity of B-DNA has started to change. The difference between the quite similar compounds C3 and C5 might be related to different binding modes.
The CD spectra of CT DNA in the presence of compounds C4 and C6 show similar changes in the positive band, but different changes in the negative band. The subsequent decrease in the positive bands with increasing Pt concentration shows the instability of the base stacking and at the highest Pt concentration; the B-DNA canonical form is almost lost. For C6, the intensity of negative band starts decreasing and also the sharp peak at 215 nm starts disappearing. This compound shows no apparent changes in base stacking, whereas at the highest concentration of the compound (ten times higher
nm is significant. The negative band shows a gradual red shift up to 6 nm. For C4, though the B-DNA form is retained (except for highest concentration of compound), there is no increase in the positive or negative bands (induced by classical intercalation). C4 and C6 are cationic compounds with no labile leaving group, so the coordination core around platinum is expected to remain intact. The dipositive charge of the compound initially facilitates for electrostatic attraction between the compound and the negatively charged DNA helix. The ligands around the platinum centres can form effective H-bonds, so they either can stabilise the DNA conformation, or may self-aggregate parallel to DNA helix.
The highest concentration of the compound may easily aggregate to form a helical structure along the DNA helix. In addition, the bidentate aromatic ligands, dipm and dpa both can lead the compounds to partial intercalation. So, a combined effect of both electrostatic interaction and partial intercalation may be an origin of the observed changes.
Figure 4.5. CD spectral changes of CT DNA (100 µM) in phosphate buffered saline in the absence (―) and in the presence of (---, ····, - - -) increasing concentrations of (a) C3, (b) C4, (c) C5 and (d) C6.
Figure 4.6. CD spectral changes of CT DNA (100 µM) in phosphate buffered saline in absence (―) and in presence of (---, ····, - - -) decreasing concentration of (a) C3, (b) C4, (c) C5 and (d) C6.
When the concentration of the platinum compounds is gradually reduced, while keeping the DNA concentration constant at 100 µM, the trend in CD spectral changes are less dramatic. For C3 and C5 the positive band (275 nm) was found to increase parallel with the appearance of a new ellipticity at 230 nm (C5) or at 225 nm (C3). The negative band for C5 first increased and then gradually decreased and red-shifted for 2 nm. C5 enhances the π-stacking, but the helicity of native DNA is somewhat inconclusive, as the changes in the negative band do not follow a typical trend. Therefore, the changes in the spectra are most probably the combined result of coordinative binding and partial intercalation. On the other hand, for C3 the negative band keeps increasing, suggesting ongoing intercalation in DNA base pairs.
For C4 and C6, gradual increases in positive and negative bands are showing clearly the intercalative binding mode. For these two compounds, no new peak appears in the 225-230 nm range. Thus these two compounds show typical intercalation by enhanced
The time-dependent spectra have also been measured for all these compounds. A significant time-effect on the changes of spectral band is observed. The observed spectral patterns are shown in Figs. 4.7-4.10.
Figure 4.7. CD spectral changes with different incubation times for C3 (a) R = 0.1, (b) R
= 1 and (c) R = 6 after interaction with CT DNA. R denotes the DNA-base/Pt ratio in concentration.
Figure 4.8. CD spectral changes with different incubation times for C4 (a) R = 0.1, (b) R
= 1 and (c) R = 6 after interaction with CT DNA.
Figure 4.9. CD spectral changes with different incubation times for C5 (a) R = 0.1, (b) R
= 1 and (c) R = 6 after interaction with CT DNA.
Figure 4.10. CD spectral changes with different incubation times for C6 (a) R = 0.1, (b) R = 1 and (c) R = 6 after interaction with CT DNA.
Several binding modes are possible for all these four platinum compounds (C3- C6), which also exhibit time-dependent intensity changes with shifts in wavelength. For all compounds, the higher concentration of platinum in the reaction solution induces conformational changes. From the time of mixing to 24 h incubation, the change of B- form to Z-form is gradual and conformational changes progress with longer incubation time. The CD spectra of all four compounds maintained the conservative doublets with subtle or almost no changes for R = 1. The presence of excess DNA in the titration solution induces hyperchromicity and bathochromic shifts (2-3 nm) for all the compounds, which enhances with longer incubation time. Therefore, the mode of interaction is not solely intercalation, but rather a combined effect of partial intercalation with coordinative binding and external stacking interactions. The optimised structures of all four platinum compounds depict the puckered form as shown in chapter 3. Thus, the complete insertion through base pairs might not be facilitated. The dipositive charge on compound C4 and C6 assists the external electrostatic interaction, whereas for compounds C3 and C5, coordinative binding is the other possible mode of interaction.
4.3.4. Interaction of platinum compounds with bovine serum albumin (BSA)
Albumin transports several small molecules through the blood flow such as calcium ions, progesterone and drugs.27 So it appeared worthwhile to investigate whether the new compounds can be transported by albumin.26, 27 On the other hand, a major route for deactivation of platinum drugs is the stable adduct formation with serum proteins.
Platinum compounds have a high affinity towards S-donor ligands (such as glutathione or metallothionein), so these adducts take a major part in bio-distribution, transport and removal of the active drug. Therefore, the interaction of the present platinum compounds with albumin has been followed by UV-vis and CD spectroscopy.
In the absorption spectral titration, the concentration of BSA was kept constant at 10 µM and different amounts of platinum compounds were added as described above for the DNA binding. It was followed by thorough mixing before recording the spectrum.
Five different ratios are compared with the native BSA spectrum at room temperature at pH = 7.2 (no platinum compound present and BSA concentration is 10 µM). The spectral changes are dramatic with higher concentration of compound. For C3 and C4, the three set of peaks appear at 279, 305, 336 (shoulder) nm and 285, 303, 330 nm, respectively.
Hypochromism is observed when the Pt concentration is decreased gradually.
Three peaks are appearing at 261, 286 (shoulder), 314 nm for C5 and at 265, 285 (shoulder) and 307 (shoulder) nm for C6. These peaks gradually increase with increasing concentration of the platinum compounds. When R is 1 to 10 (R is concentration ratio of protein to metal compound), the spectra return to the shape of native BSA with hypochromism for the peak at 279 nm (Fig. 4.11).
As a reference compound, cisplatin has also been tested for BSA binding [Fig.
4.11(e)]. The known preferred binding site of cisplatin on BSA is methionine-298, but there are also several possibilities of multiple binding at cysteine and tyrosine residues.34,
35 After prolonged incubation times 20 mol of platinum binds to 1 mol of protein. The present new platinum compounds thus agree with the binding mode of cisplatin as reported in literature.27, 36 Therefore, a higher concentration of the platinum compound in the solution induces hyperchromicity indicating to the stronger binding. On the contrary, hypochromicity is evident with lower platinum concentrations.36
The CD studies are quite helpful to investigate the changes in the secondary structure of the protein upon platinum binding. To correlate with the interaction mode of cisplatin (a classical coordinative binder) under same experimental conditions, changes in the CD have been recorded after incubation with BSA (Fig. 4.12).
The most abundant serum protein, serum albumin, (52% with 40-45 g per liter content for healthy humans) at physiological pH remains predominantly in the α-helical form (67%). The typical characteristic of this α-helix is evident in the CD spectra for the 180-300 nm scanning range. The two strong negative bands in the CD spectrum of bovine serum albumin appearing at 208 and 222 nm originate from π-π* and n-π* exciton transition. The investigated compounds do not give rise to any signal in this range. Thus, following the change in the ellipticity of these two bands, the extent of secondary structure change in protein can be assessed easily.
Figure 4.11. Changes in the UV-Vis spectra after incubation of a constant concentration of protein (50 µM) with different concentration of (a) C3, (b) C4, (c) C5, (d) C6 and (e) cisplatin (500-5 µM) at pH=7.2 in phosphate buffered saline at room temperature.
Arrows show the changes in absorption upon reduction of platinum compounds and R is the concentration ratio of protein to platinum compound.
Figure 4.12. Changes in CD spectra after incubation of a constant concentration of protein (50 µM) with different concentration of (a) C3, (b) C4, (c) C5, (d) C6 and (e) cisplatin (500-5 µM) at pH=7.2 in phosphate buffered saline at room temperature. Arrow marks show the changes in ellipticity upon addition of platinum compounds.
Upon addition of different amounts to a solution to protein (50 µM) the cisplatin- analogues, C3 and C5, exhibit a similar behaviour in the CD spectra as cisplatin. The absorption maximum of 222 nm shows a red shift of 8 nm in presence of a 10 fold excess of the platinum compound (R= 0.1, where R stands for concentration ratio of protein per platinum compound). In addition, the ellipticity at 208 nm completely disappeared with increasing amount of compound. The most reported mode of interaction of cisplatin with BSA is the coordinative binding after hydrolysis of chloride ions. These two new platinum compounds react in a similar fashion as cisplatin. After hydrolysis of one chloride ligand, a positively charged intermediate is formed. This highly reactive intermediate binds to BSA and destabilises the native α-helical form. The destabilisation is clearly evident from the CD spectra.
On the other hand, C4 and C6 contain non-hydrolysable ammine ligands.
Therefore, upon addition of increasing amounts of each compound, very little changes are observed. The stable helical form is retained with only a little increase in ellipticity, which can be explained by external attachment of the platinum to the protein.
4.3.5. Partition coefficient measurement
In order to measure the selective preference of a solute, the distribution property was measured by the shake-flask method. The two immiscible solvents chosen are water and octanol (octanol mimics the lipid membrane or cell wall37).
The four platinum compounds (C3-C6) along with cisplatin are used to determine the partition coefficient. Only C3 exhibits a positive log Po/w value indicating a very high lipophilicity. The other log Po/w values are negative, showing high hydrophilicity. C5 shows a partition coefficient comparable with satraplatin. C4 and C6 show partition coefficients comparable to oxaliplatin. If these compounds are arranged according to the order of increasing lipophilicity the trend will be C3 > C5 > C4 > C6. The comparison of the distribution with few others reported platinum drugs are summarised in Table 4.1.
4.3.6. DNA binding studies by gel electrophoresis
DNA binding or cleaving by a compound can be easily studied following the relative movement of DNA bands after agarose gel electrophoresis. The native DNA remains in the supercoiled form, also known as Form I (closed circular form). After single strand cutting, this can result into Form II (representing nicked DNA, also called open circle). If the compound cleaves DNA double-stranded, the Form III (representing linear
DNA) is formed. Form I (supercoiled DNA) has a tightly packed conformation and therefore migrates faster through agarose gels than linear DNA (intermediate migration) or open circle DNA (slowest migration). These movements on gel depend on the size of the DNA fragments and the conformation of the DNA. In this case, the electrophorectic movements are originated from the different conformation of the DNA, which has been induced by the platinum compounds.
Table 4.1. Comparison of partition coefficient values for several reported platinum compounds using known methodology in water/n-octanol.38
Studied
compounds Structure Partition coefficient
Reported
compounds Structure Partition coefficient
C3 N N
NH
Pt Cl Cl
0.11 cisplatin Pt
Cl
H3N H3N
Cl
-2.53
C4 N N
NH
Pt NH3 NH3
(NO3)2 -1.33 carboplatin Pt O
O H3N
H3N
O
O
-2.3
C5
N
N N
N NH
Pt
Cl Cl
-0.47 oxaliplatin Pt
O O
O
O N
N
-1.65
C6
N
N N
N NH
Pt NH3 NH3
(NO3)2 -1.74 satraplatin N Pt
O
O Cl
H3N Cl O
O
-0.16
Two platinum compounds (C5 and C6) have been tested for their DNA-cleavage properties, using supercoiled φX174 phage DNA in phosphate buffer at 37 °C. The samples were incubated for 0.5 h at dark with increasing amount of compound in presence of DNA (20 µM).
For C5, form II appears at 25 µM and form III appears at 40 µM. The most prominent change observed is the retardation of the supercoiled DNA in presence of increasing amount of compounds. This phenomenon is easily explained by simultaneous strong binding to DNA of multiple molecules. The gradual change in the electrophoretic movement in presence of C5 is shown in Fig. 4.13.
On the other hand, C6 shows strong binding from very low concentration (5 µM) evident by shift in the gel. The appearance of form III clearly starts from 35 µM concentration (Fig. 4.14). In addition, the effect of reducing agent, ascorbic acid (AA), the cleavage is more facilitated, whereas in absence of AA, the shift of DNA undoubtedly proves the binding of C6 (Fig. 4.15). The appearance of form III or linear DNA starts from 35 µM, which is therefore, a better cleaving agent than C5.
Figure 4.13. Agarose gel electrophoresis of φX174 phage DNA after 30 min incubation with C5. The gel at the top shows the lower concentration (in μM) with the appearance of form II and the gel below shows the higher concentration with cleavage and retardation.
Figure 4.14. Agarose gel electrophoresis of φX174 phage DNA after 30 min incubation with C6. The gel at the top shows the lower concentration (in μM) with the appearance of form II and the gel below shows the higher concentration with cleavage and retardation.
The effect of reducing agent (Ascorbic Acid, AA), on the cleavage of C6 has been shown in Fig. 4.15. C6 exhibits 50% single stranded cleavage stoichiometrically after 30 min of incubation (Pt to DNA= 1.25:1). The cleavage increases with increasing amount of
C 5 10 15 20 25 30 35
C 40 45 50 55 60 65 70
Form II (open circle/nicked) Form III (linear)
Form I (supercoiled) Form II
Form I
Form II Form I C 40 45 50 55 60 65 70
C 5 10 15 20 25 30 35
Form II Form I
Form III
compound suggesting multiple site attacks by the platinum compound. In the absence of AA insignificant extent of cleavage is observed, even at 1:2.5 ratio of DNA: Pt. The shift of form I is increasing with the platinum compound, suggesting a stronger binding to DNA.
Figure 4.15. Agarose gel electrophoresis of φX174 phage DNA after 30 min incubation with C6, in the presence and absence of ascorbic acid. For experimental details see section 4.2.4 and the numbers denote the concentration in μM.
4.4. Conclusions
Several platinum compounds have been investigated for their mode of interaction with two major biological targets. The neutral cisplatin analogues show coordinative binding and partial intercalation. The interaction with BSA further indicates a similar activation pathway (via hydrolysis of chloride ligand) as cisplatin. On the other hand, the cationic platinum compounds opt for electrostatic interaction with the negative DNA- phosphate backbone followed by partial intercalation. The ammine ligands on these compounds further stabilise the DNA adduct by hydrogen bonding. The other possible modes of interaction with DNA are external association with or without self-aggregation.
The non-planar folded structures of all these compounds inhibit the complete intercalation. The gel mobility studies additionally clarify multiple binding on DNA strands by platinum compounds.
A high affinity towards serum albumin is exhibited by neutral cisplatin-analogues.
The stable α-helical conformation of the protein (BSA) is destabilised by these compounds whereas the cationic compounds show no affinity towards BSA. Therefore, the probability of retaining the intact formulation in cellular milieu is higher for cationic compounds. The neutral compounds may therefore be deactivated or removed by BSA- platinum drug adduct formation. This selective affinity towards platinophiles directs a synthetic design of cationic compounds with higher activity in vivo.
C 5 5 25 25 50 50
+ - + - + - AA
Form I Form II
The partition coefficients for platinum compounds are in a comparable range with clinically approved drugs, except [Pt(dpa)Cl2]. This compound is clearly lipophilic and thus will be facilitated significantly to cross the cell membrane. The other three compounds are hydrophilic and will prefer to accumulate in the blood plasma. Both cationic compounds show lipophilicity in the range of oxaliplatin. Thus, lipophilicity tests can provide a rough idea of the behaviour of a compound in vivo, but no straightforward correlation can be drawn between lipophilicity and in vivo anticancer activity.
4.5. References
1. Johnson, W. C., CD of Nucleic Acids. In Circular Dichroism; Principles and Applications, Nakanishi, K.; Berova, N.; Woody, R. W., Eds. VCH Publishers Inc.: New York, 1994.
2. Snatzke, G., Circular Dichroism: Principles and Applications. VCH: 1994.
3. Richards, A. D.; Rodger, A., Chem. Soc. Rev. 2007, 36, 471-483.
4. Rice, J. A.; Crothers, D. M.; Pinto, A. L.; Lippard, S. J., Proc. Natl. Acad. Sci. U.
S. A. 1988, 85, 4158-4161.
5. Teuben, J. M.; Bauer, C.; Wang, A. H. J.; Reedijk, J., Biochemistry 1999, 38, 12305-12312.
6. Pinto, A. L.; Lippard, S. J., Biochim. Biophys. Acta 1984, 780, 167-180.
7. Roberts, J. J.; Thomson, A. J., Prog. Nucleic Acid Res. Mol. Biol. 1979, 22, 71- 133.
8. Wu, Y. S.; Koch, K. R.; Abratt, V. R.; Klump, H. H., Arch. Biochem. Biophys.
2005, 440, 28-37.
9. Reedijk, J.; Fichtinger-Schepman, A. M. J.; den Hartog, J. H. J., Inorg. Chim. Acta 1983, 79, 252-253.
10. Scheeff, E. D.; Briggs, J. M.; Howell, S. B., Mol. Pharmacol. 1999, 56, 633-643.
11. Sherman, S. E.; Gibson, D.; Wang, A. H. J.; Lippard, S. J., Science 1985, 230, 412-417.
12. Sherman, S. E.; Gibson, D.; Wang, A. H. J.; Lippard, S. J., J. Am. Chem. Soc.
1988, 110, 7368-7381.
13. Sherman, S. E.; Lippard, S. J., Chem. Rev. 1987, 87, 1153-1181.
14. Talman, E. G.; Brüning, W.; Reedijk, J.; Spek, A. L.; Veldman, N., Inorg. Chem.
1997, 36, 854-861.
15. Teo, B. K.; Eisenberger, P.; Reed, J.; Barton, J. K.; Lippard, S. J., J. Am. Chem.
Soc. 1978, 100, 3225-3227.
16. van Garderen, C. J.; Bloemink, M. J.; Richardson, E.; Reedijk, J., J. Inorg.
Biochem. 1991, 42, 199-205.
17. Yang, D. Z.; van Boom, S.; Reedijk, J.; van Boom, J. H.; Farrell, N.; Wang, A. H.
J., Nat. Struct. Biol. 1995, 2, 577-586.
18. Blackburn, G. M.; Gait, M. J., Nucleic acids in chemistry and biology. Oxford University Press: New York, 1996.
19. Cusumano, M.; di Pietro, M. L.; Giannetto, A.; Nicolo, F.; Rotondo, E., Inorg.
Chem. 1998, 37, 563-568.
20. Gibson, D.; Gean, K. F.; Benshoshan, R.; Ramu, A.; Ringel, I.; Katzhendler, J., J.
Med. Chem. 1991, 34, 414-420.
21. Jennette, K. W.; Gill, J. T.; Sadownick, J. A.; Lippard, S. J., J. Am. Chem. Soc.
1976, 98, 6159-6168.
22. Chen, L. M.; Liu, J.; Chen, J. C.; Tan, C. P.; Shi, S.; Zheng, K. C.; Ji, L. N., J.
Inorg. Biochem. 2008, 102, 330-341.
23. Kashanian, S.; Gholivand, M. B.; Ahmadi, F.; Taravati, A.; Colagar, A. H., Spectrochim. Acta A 2007, 67, 472-478.
24. Kemp, S.; Wheate, N. J.; Buck, D. P.; Nikac, M.; Collins, J. G.; Aldrich-Wright, J.
R., J. Inorg. Biochem. 2007, 101, 1049-1058.
25. Cusumano, M.; di Pietro, M. L.; Giannetto, A.; Vainiglia, P. A., Eur. J. Inorg.
Chem. 2005, 278-284.
26. Quiming, N. S.; Vergel, R. B.; Nicolas, M. G.; Villanueva, J. A., J. Health Sci.
2005, 51, 8-15.
27. Timerbaev, A. R.; Hartinger, C. G.; Aleksenko, S. S.; Keppler, B. K., Chem. Rev.
2006, 106, 2224-2248.
28. Bloemink, M. J.; Perez, J. M. J.; Heetebrij, R. J.; Reedijk, J., J. Biol. Inorg. Chem.
1999, 4, 554-567.
29. Platts, J. A.; Oldfield, S. P.; Reif, M. M.; Palmucci, A.; Gabano, E.; Osella, D., J.
Inorg. Biochem. 2006, 100, 1199-1207.
30. Yao, W. B.; Kavallieratos, K.; de Gala, S.; Crabtree, R. H., Inorg. Chim. Acta 2000, 311, 45-49.
31. Kukushkin, V. Y.; Oskarsson, A.; Elding, L. I.; Farrell, N., Inorg. Synth. 1998, 32, 141-144.
32. Ivanov, V. I.; Minchenk, L. E.; Schyolki, A. K.; Poletaye, A. I., Biopolymers 1973, 12, 89-110.
33. Macquet, J. P.; Butour, J. L., Eur. J. Biochem. 1978, 83, 375-387.
34. Ivanov, A. I.; Christodoulou, J.; Parkinson, J. A.; Barnham, K. J.; Tucker, A.;
Woodrow, J.; Sadler, P. J., J. Biol. Chem. 1998, 273, 14721-14730.
35. Neault, J. F.; Tajmir-Riahi, H. A., Biochim. Biophys. Acta, Protein Struct. Mol.
Enzymol. 1998, 1384, 153-159.
36. Peleg-Shulman, T.; Gibson, D., J. Am. Chem. Soc. 2001, 123, 3171-3172.
37. Ghezzi, A.; Aceto, M.; Cassino, C.; Gabano, E.; Osella, D., J. Inorg. Biochem.
2004, 98, 73-78.
38. Screnci, D.; McKeage, M. J.; Galettis, P.; Hambley, T. W.; Palmer, B. D.;
Baguley, B. C., Brit. J. Cancer 2000, 82, 966-972.
