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).
| 3
Synthesis, cytotoxicity assay and model- base studies of derivatised pyridine and
pyrimidine compounds of platinum(II) †
Chapter
Four Pt(II) compounds have been synthesised with modification in the carrier ligands. Two secondary amines have been studied namely, bis(pyridine-2-yl)amine, L6 (dpa) and bis(pyrimidine-2-yl)amine, L7 (dipm) to design the basic chemical motif and to generate subtle differences in electronic properties and hydrogen bonding ability.
Different carrier ligands (chloride and ammine) have been introduced in search for structure-activity relationships. Two of the four compounds are cisplatin analogues. The compounds with L6 are Dichlorido-2,2'- dipyridinylamineplatinum(II), [Pt(dpa)Cl2], and Diammine-2,2'- dipyridinylamineplatinum(II) nitrate, [Pt(dpa)(NH3)2](NO3)2. The compounds Dichlorido-2,2'-dipyrimidinylamineplatinum(II), [Pt(dipm)Cl2], and Diammine-2,2'-dipyrimidinylamineplatinum(II) nitrate, [Pt(dipm)(NH3)2](NO3)2, bear L7 as the bidentate ligand. Two of these compounds are neutral with limited aqueous solubility, whereas the other two are positively charged and highly water-soluble. The structures of all these compounds have been optimised using DFT calculations. The energy-minimisation showed an open-book non-planar structure for all four compounds. In addition, the reaction of the new compounds with the model base 9-Ethylguanine has been followed by proton and platinum NMR spectroscopy. The cytotoxicity of these compounds has been investigated in seven human tumour cell lines using the SRB (Sulforhodamine B) assay. The most promising antitumour active compound appears to be [Pt(dipm)Cl2]. The effect of shorter incubation time and solvents has been tested and the two cationic platinum compounds show selective cytostatic activity against breast cancer cells only after relatively short exposure.
3.1. Introduction
The serendipitous discovery of cisplatin is generally considered to be a break- through in cancer treatment. In the years following the discovery of cisplatin it has been demonstrated that the combination of transition metals with appropriate ligands can lead to clinically approved anticancer metallodrugs (carboplatin, oxaliplatin, satraplatin), others like picoplatin, NAMI-A are in clinical trials. The versatile chemical properties of the central metal atom play a crucial role in explaining the activity against tumour cells;
however, the electronic and kinetic effects of the ligands need to be considered as well.1, 2 In this fertile field three main approaches to design new drugs are known: (1) cisplatin analogues or modifications, (2) use of other metals than platinum and (3) combination of active organic drugs with metal compounds.
Platinum coordination compounds have been the main focus for metal-antitumour drugs for the last decades. Though cisplatin is the most widely used drug in the treatment of solid tumours (head, neck, ovarian, testicular, bladder, stomach, lung and oesophageal cancer3), severe toxic side effects and resistance limit its prolonged application. The second-generation platinum drugs (carboplatin and oxaliplatin) are less toxic and are used effectively in combination therapy.4 However, the urge to obtain more efficient orally administrable drugs with minimum side effects continues. To reach that target, five key characteristics of the metallodrugs are required, namely:2
(a) saline solubility and stability;
(b) facile transport in blood and through cellular membranes;
(c) relatively stable DNA-binding ability, with slow or weak interaction to proteins, although non-DNA binding drugs cannot be excluded;
(d) selectivity and specificity towards cancer cells over healthy cells;
(e) activity in cisplatin-resistant cell lines.
The well-established cause of antitumour activity of cisplatin is the intrastrand DNA lesion. Therefore, genomic DNA is considered the primary intracellular target of platinum drugs. Activated cisplatin after hydrolysis of the chloride ligands coordinates to DNA (purine bases, specially guanine) thereby generating a kink in the helix.5, 6 On the other hand, two other antitumour drugs, doxorubicin and daunomycin, are known to act by intercalating between the base pairs of DNA.7 These compounds inhibit cell growth by mainly two pathways, (a) poisoning Topoisomerase I and II, which inhibits replication and (b) poisoning RNA polymerase which inhibits transcription. Therefore, bringing the
best of these two facets together i.e., combining coordination binding and intercalation in a single molecule might perhaps generate a new class of antitumour drugs.
The role of hydrogen bonding in DNA binding is quite important. Especially, when the metallodrugs are coordinated to DNA, additional hydrogen bonding renders better stability to the DNA-drug adduct. For metallointercalators the H-bonding from the metal compound might strengthen the DNA-binding affinity. Therefore, although not mentioned very often in the literature, the impact of hydrogen bonding on biomolecules (DNA or protein) is worth being explored. This non-coordinative interaction may also induce self-assembly of metal compounds, so that aggregated forms may arrange parallel to DNA strand. Therefore, a few heterocyclic ligands (known from literature) were studied in their binding to Pt(II), vide Fig. 3.1.
N NH
N M Donor
Metal binding site
L6
N N N
N NH
M Acceptor Donor
Acceptor
Metal binding site L7
Figure 3.1. Hydrogen-bond acceptor and donor sites in ligands L6 and L7.
The first ligand, L6, is a versatile ligand. It contains only a solitary H-bond donor moiety and the structure when attached to a metal has been schematically depicted in Fig.
3.1. L6 acts as a flexible chelating ligand by virtue of the central secondary amine moiety.
As a result the two pyridine rings can be either co-planar or tilted. Thus, the metal compound has a different structure, depending on the co-ligands bound to the metal and the packing. This ligand has been reported in some recent papers 8-10 when coordinated with platinum or palladium. Only two cell lines (B16-BL6 and human Jurkat T cells) have been used for the cytotoxicity tests of dichlorido-2,2'-dipyridinylamineplatinum(II), [C3, Pt(dpa)Cl2] reproduced following the literature procedure8, 9 for detailed biological experiments and another analogue with 2 monodentate ammine ligands. A newly prepared analogue, diammine-2,2'-dipyridinylamineplatinum(II) nitrate; C4 [Pt(dpa)(NH3)2](NO3)2, with two non-labile ammonia ligands has also been synthesised
for comparison. This highly water-soluble compound overcomes the poor solubility problem of C3. The molecular structures for C3 and C4 are shown in Fig. 3.2.
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
Figure 3.2. Molecular structures of four platinum compounds (C3-C6).
In the present study another unique ligand, L7 has been selected for coordination to platinum. Interestingly, L7 [bis(pyrimidine-2-yl)amine] (dipm), can form a linear acceptor-donor-acceptor array of hydrogen bonds (shown in Fig. 3.1) upon metal chelation. The previous report shows that, after coordination with Cu, L7 forms stable intermolecular Watson-Crick type hydrogen bonds.11, 12 It has also been reported that just a slight change in the anion can affect the hydrogen bonding of this compound significantly.11 These reports are the background to synthesise a new category of compounds via coordination to platinum. The structural variation has been incorporated by using different monodentate ligands (chloride or amine). The molecular structures of C5-C6 are depicted in Fig. 3.2. The expected hydrolysable chloride anions would make compound C5 a cisplatin analogue. The difference with cisplatin lies mainly in the chelating bidentate dipm ligand. The effect of ancillary ligands on the cytotoxicity and structural changes has been also investigated in compound C6 for comparison purposes.
Using DFT calculations all four compounds (C3-C6) have been optimised and their structures are reported. The two cisplatin-analogues, C3 and C5 have been studied for the DNA-model base reactions.
The 60-cell line panel is generally used in screening cytotoxic activities by the National Cancer Institute (NCI) Developmental Therapeutics Program, which provides the activity profile of new compounds. The web-based software analysis is efficiently and promptly performed by COMPARE (specially designed online programme).13 The validation of a smaller cell panel (19-cell line) compared to 60-cell line panel has been reported.14 In that specific article the effect of different solvent (DMF), different dye (crystal violet and MTT), different incubation time (96 h) and different concentration range are elucidated to allow a comparison with the typical NCI protocol.
These 60- or 19-cell line testing panels are quite exhaustive to evaluate cytotoxic activity of numerous new samples. So, it was decided to select a manageable further scaled-down 7-cell panel from different organ origins; this would provide the feasibility to test a wide range of samples in one cycle. The use of different solvents (dmso as used by NCI, or water) facilitates the activation of the compounds by different reaction solvolysis pathways, which might influence the activity in vitro. Short (48 h as used by NCI) or long (96 h) incubation times may have an effect on the uptake and accumulation of the various compounds. Three clinically-approved platinum drugs along with some typical reference compounds have been tested under the same conditions for comparison purposes, and the results are described in this chapter.
3.2. Experimental 3.2.1. Materials
2-chloropyrimidine was obtained from Acros Organics (The Netherlands).
Potassium carbonate was purchased from Merck (Germany). Anhydrous MgSO4 and N,N- diisopropylethylamine (dipea) were received from Fluka (The Netherlands). L6 (dpa), 9- Ethylguanine (9-EtG), 2-aminopyrimidine and deuterated solvents for NMR were obtained from Sigma Aldrich B.V. (The Netherlands) and used directly as received. The starting material, K2PtCl4 was received as a generous loan from Johnson Matthey (Reading, U.K.). Cisplatin was synthesised following a reported route.15 Another platinum precursor, cis-[Pt(dmso)2Cl2] was synthesised following the reported synthetic procedure.16 The ligand L7, bis-(pyrimidine-2-yl)amine (dipm), was synthesised following a reported synthetic route with some modifications and described in details in the following section.17 The solvents used for synthesis were purchased from Biosolve (AR grade, The Netherlands) and were used without further purifications.
3.2.2. Cell types and chemicals
The seven tumour cell lines used for this study are of the following origin: MCF7 (breast cancer (ER)+/(PgR)+), EVSA-T (breast cancer (ER)-/(PgR)-), WIDR (colon cancer), IGROV (ovarian cancer), M19MEL (melanoma), A498 (renal cancer), H226 (non-small cell lung cancer). The details of the cell lines have been tabulated in Chapter 2 (vide Table 2.1). The medium (RMPI-1640) and 10% fetal calf serum were purchased from Gibco, Invitrogen, Paisley Scotland. Penicillin, streptomycin, SRB and dmso were received from Sigma, St.Louis, MO, USA and PBS (phosphate buffered saline) was
obtained from NPMI B.V. Emmer-Compascum, The Netherlands. Cisplatin, Carboplatin, Oxaliplatin and R,R-[Pt(dach)Cl2] were purchased from Sigma-Aldrich (The Netherlands).
3.2.3. Nuclear magnetic resonance spectroscopy
One- and two-dimensional NMR spectra were recorded on a Bruker DPX 300 MHz spectrometer at room temperature (21 °C), or at physiological temperature (37 °C).
The solvents used were dmso-d6, dmf-d7, D2O or a mixture of these solvents. The 195Pt NMR spectra were referenced to an external standard, K2PtCl4 in D2O (δ = – 1614 ppm).
For the time-dependent one-dimensional serial studies spectra were collected with 48 scans and 15 minutes waiting time for a 14 h measurement. All data have been processed with XWIN-NMR and XWIN-PLOT.18
3.2.4. Electrospray ionisation mass spectroscopy
The ESI-MS measurements were carried out with a Thermo Finnigan Aqa mass detector. With the HPLC equipment 10 µL of the sample was introduced into the detector. Different solvents were used [methanol: water = 80:20 (v/v) or water] with a flow of 0.2 mL/min. Samples were measured in the positive mode using an ionisation voltage of 3 kV and a detector voltage of 20 mV with a probe temperature of 27 °C. The scanning range was m/z = 100 to 1400. For C3 and C5, the dmso stock solution was directly infused in the instrument.
3.2.5. Elemental analysis
Elemental analysis of the samples was performed in a Perkin Elmer Series II CHNS/O Analyser 2400. Typically 2 mg to 2.2 mg of sample was used for C, H, N and S percentage analysis. All measurements were performed in duplicate and the average value is reported.
3.2.6. Chemical syntheses
(a) L7: bis(pyrimidine-2-yl)amine (dipm):
The used synthetic route has been modified slightly from that reported in literature.17 dipea (37.1 g, 0.287 mol), 2-aminopyrimidine (10 g, 0.105 mol) and 2- chloropyrimidine (16.7 g, 0.105 mol) were mixed together and refluxed for 60 hours. The reaction mixture was cooled down to room temperature and the excess of DIPEA was decanted. The residual brown paste was redissolved in 200 mL of boiling water and the
aqueous layer was extracted several times with dichloromethane. The organic phase was collected, dried over anhydrous MgSO4 and finally evaporated under reduced pressure to yield a yellow solid. The crude ligand was recrystallised from water. Yield of the recrystallised ligand (light brown needles) was 60%.
(b) C3: Dichlorido-2,2'-dipyridinylamineplatinum(II), cis-[Pt(dpa)Cl2]
This compound had already been reported by others8, 9 and the synthesis has been reproduced with some additional characterisations. Two different platinum starting materials have been used.
Method 1: From cis-[Pt(dmso)2Cl2]
The starting material, cis-[Pt(dmso)2Cl2] (0.010 g, 2.40×10-4 mol) was dissolved in a minimum amount of dmso. A solution of dpa (0.040 g, 2.40×10-4 mol) in a minimum amount of dmso was added. The yellow solution was stirred for 3 days at room temperature. Removal of solvent under reduced pressure yielded a yellow coloured product (0.087 g, 83%).
Method 2: From K2PtCl4
The platinum precursor, K2PtCl4 (0.050 g, 1.2×10-4 mol) was dissolved in 6 mL of water and a solution of dpa (0.020 g, 1.2×10-4 mol) in 2 mL ethanol was added dropwise.
The mixture was stirred overnight at 50 °C. A yellow precipitate had formed, which was filtered off and washed with ice-cold water and acetone. The yield after purification was 0.018 g (34%).
(c) C4: Diammine-2,2'-dipyridinylamineplatinum(II) nitrate, cis-[Pt(NH3)2(dpa)](NO3)2
Cis-[Pt(NH3)2Cl2] (0.050 g, 1.67×10-4 mol) was dissolved in 2 mL of dmf. A solution of silver nitrate (0.056 g, 3.33×10-4 mol) in 1 mL of dmf was added and the mixture was stirred overnight at room temperature in the dark. The precipitate was filtered off and dpa (0.028 g, 1.67×10-4 mol) was added to the filtrate. The mixture was stirred for 4 days at room temperature in the dark. A small amount of a greenish precipitate (mostly polymer, or Magnus-salt like ionic aggregates19) was filtered off.
Diethyl ether was added slowly to the filtrate to precipitate the whitish product. The product was washed with a small amount of cold water and the product (0.041 g, 47%) was air dried.
(d) C5: Dichlorido-2,2'-dipyrimidinylamineplatinum(II), cis-[Pt(dipm)Cl2]
The title compound was synthesised using cis-[Pt(dmso)2Cl2] (prepared as reported in literature16) as the starting material. The cis-[Pt(dmso)2Cl2] (0.050 g, 1.18×10-
4 mol) was dissolved in a minimum amount (∼4 ml) of dmso. A solution of dipm (0.020 g, 1.18×10-4 mol) in a minimum amount of dmso (∼3 ml) was added dropwise to the platinum solution. The yellow solution was stirred for 3 days at 50 °C in the dark. Slow addition of diethyl ether to the reaction mixture initiated precipitation of the product. The yellow precipitate was dried in air to yield 0.032 g (63%) of product.
(e) C6: Diammine-2,2'-dipyrimidinylamineplatinum(II) nitrate, cis-[Pt(NH3)2(dipm)]
(NO3)2
The cis-[Pt(NH3)2Cl2] (0.050 g, 1.67×10-4 mol) was dissolved in a minimum amount of dmf in dark. A solution of silver nitrate (0.056 g, 3.33×10-4 mol) in 1 mL dmf was added and the mixture was stirred overnight at room temperature. The precipitate was filtered off and dipm (0.028 g, 1.67×10-4 mol) was added to the filtrate. The mixture was stirred for 2 days at room temperature. Diethyl ether was added to precipitate the pale yellow product with a yield of 0.043 g (64%).
3.2.7. Analysis (a) L7
The ligand was characterised by 1H, 13C and 2D COSY experiments (spectra not reproduced here). Elemental analysis (%) for C8H7N5·2H2O (water from recrystallisetion): expt. (calcd.) C 45.55 (45.93), N 33.58 (33.48), H 5.54 (5.30). ESI-MS measurements of a freshly prepared solution of recrystallised dipm in methanol show peaks at m/z = 196 and 228, which corresponds to Na + L7 and Na + L7 + CH3OH species, respectively.
(b) C3 (method 1)
The 1H and 195Pt NMR in dmso-d6 characterised the desired product. The 195Pt (dmso-d6) spectrum shows a peak at δ -2094 ppm, which corresponds to PtN2Cl2
environment. ESI-MS of a freshly prepared solution (in dmso and direct infusion to instrument without further dilution), shows the prominent peak at 479.6, which corresponds to the species, [Pt(dpa)Cl(dmso)]+ with typical platinum isotopic pattern.
Elemental analysis (%) for PtC10H9N3Cl2: expt. (calcd.) C 26.56 (27.47), N 9.30 (9.61), H 2.16 (2.07).
C3 (method 2)
The 1H and 195Pt NMR spectra in dmso-d6 show characteristic peaks of the product. The 195Pt NMR spectrum shows a peak at δ = -2093 ppm, which corresponds to a PtN2Cl2 environment. ESI-MS of a freshly prepared solution (in dmso), shows the prominent peak at m/z = 479.6, which corresponds to the species, [Pt(dpa)Cl(dmso)]+ with typical platinum isotopic pattern. Elemental analysis (%) for PtC10H9N3Cl2: expt.
(calcd.) C 26.96 (27.47), N 9.48 (9.61), H 2.14 (2.07).
(c) C4
The 1H and 195Pt NMR were recorded in D2O and show characteristic peaks from dpa coordination. The 195Pt shows peak at δ = -2584 ppm, corresponding to a PtN4
environment. A fresh solution in water (elution with water only) gives rise to the peak at m/z = 401, corresponding to the species, [Pt(dpa)(NH3)2]2+. Elemental analysis (%) for PtC10H15N7O6: expt. (calcd.) C 22.77 (22.91), N 18.41 (18.70), H 2.94 (2.88).
(d) C5
The 1H NMR spectrum in dmf-d7 shows small shifts compared to the free ligand.
195Pt spectrum (in dmf-d7) shows a peak at δ -2049 ppm, which corresponds to PtN2Cl2
environment. MS from a fresh solution (solution made in dmso and direct infusion the spectrometer) shows the major peak at m/z = 481, which corresponds to [Pt(dipm)Cl(dmso)]+ species with a typical platinum isotopic pattern. Elemental analysis (%) for PtC8H7N5Cl2: expt. (calcd.) C 21.33 (21.88), N 14.80 (15.95), H 1.84 (1.61).
(e) C6
The 1H NMR in dmso-d6 shows small shifts compared to free ligand and 195Pt NMR in the same solvent shows a single peak at δ -2504 ppm, which corresponded to PtN4 coordination motif. MS (elution with water) shows a major peak at m/z = 401, which corresponds to [Pt(dipm)(NH3)2]2+ species with a typical platinum isotopic pattern.
Elemental analysis (%) for PtC8H13N9O6: expt. (calcd.) C 18.26 (18.26), N 23.69 (23.95), H 2.42 (2.49).
3.2.8. Computational details
Density functional theory (DFT) calculations have been performed with the Amsterdam density functional (ADF) program.20-22 The BLYP23, 24 exchange-correlation functional and applied relativistic corrections within the Zero Order Regular Approximation (ZORA)25 for all the elements have been used. The relativistic corrections are particularly relevant for heavy elements, such as Pt. In the ADF code the electronic orbitals are written in terms of Slater-type functions (STO). The ZORA formalism requires specially adapted basis sets, and the relativistic basis set ZORA/TZ2P (triple-zeta including two sets of polarisation functions) was used. The ZORA formalism has been used also for the geometry optimisation of all the compounds. The effect of the relativistic corrections is significant and the agreement with experimental data for the C3, [Pt(dpa)Cl2] compound improves considerably.8, 9 In particular for the Pt-Cl (Pt-N) distance, an error of 5% (9%) observed without ZORA, decreases to 2% (3%) including ZORA.
3.2.9. Model-base studies
9-Ethylguanine was used as a model nucleobase to follow the interaction with platinum compounds, C3 and C5. One-dimensional NMR and ESI-MS spectra were recorded over time. Different molar equivalents (platinum compound: model base = 1:1or 1:4) of model base were added to the compound solution and allowed to react in deuterated solvent over 12 h. To investigate fast reactions, the ratio of platinum compound: model base was also used as 1:4. The volume of the solvent was kept constant at 700 µL and the Pt-sample concentration was 5 mM for all experiments. Reactions were followed by NMR spectroscopy over 7 days. To follow the changes around the platinum coordination sphere, 195Pt spectra were taken after 24 h of reaction time. ESI-MS spectra were recorded over specific time intervals.
3.2.10. In vitro cytotoxicity assay
The cytotoxicity of platinum compounds (C3-C6) and respective free ligands (L6- L7) was analysed using the microculture sulforhodamine B (SRB) test.26 This colorimetric assay was carried out generously by TEVA-Pharmachemie (Haarlem, The Netherlands). Briefly, the experiments were started on day 0. Flat-bottomed 96-welled microtiter plates (Cellstar, Greiner Bio-one) were used to plate 150 µL of trypsinised
tumour cells (1500-2000 cells/well). The plates were then preincubated for 48 h at 37 °C in 5% CO2 atmosphere to allow cells to adhere. On day 2, a three-fold dilution sequence of 10 steps was made in full medium starting from 0.25 mg/mL stock solution. Each dilution was used in quadruplicate by adding 50 µL to a column of four wells. This stepwise addition from low to high concentration of sample reached the highest concentration of 0.062 mg/mL in column 12. In column 1, only full medium was added to diminish interfering evaporation and column 2 was used for a blank in each cell plate.
The pictorial presentation of the cell plates is shown in Fig. 3.3. The further experimental details have been mentioned in chapter 2 (vide section 2.2.6). Data were used for concentration-response curves and ID50 values using Deltasoft 3 software (Biometallics Inc., Princeton, NJ, USA). The IC50 values were obtained after proper unit conversion of the ID50 values. The obtained IC50 values correspond to the concentration of the drug required for 50% inhibition of cell growth.
M E D I U M
C E L L + M E D.
Compound 2
low concentration high
Compound 1
B L A N K 1
B L A N K 2
Figure 3.3. Schematic representation of cell plates during incubation of the platinum compounds.
3.3. Results and discussion 3.3.1. General comments
In this study, four new quite different platinum compounds with differences in structure and property (both chemically and electronically) are reported. The aim has been to investigate the bio-physical properties and in vitro cytostatic activities. These compounds carry solvolysable chloride ligands (for C3 and C5), or non-leaving hydrogen-bond donor am(m)ine groups (C4 and C6). The changes in these ancillary ligands may help to modify the chemical properties, which undoubtedly will be reflected
in their interaction with different biological targets. These results are useful to correlate the biophysical studies with in vitro cytotoxicity. The flexibility of the aromatic ligands opens up the possibility of new ligand design, which is promising for new anticancer Pt- drugs.
3.3.2. Syntheses and characterisations
The synthesis of L7 and the platinum compounds (C3-C6) are straightforward.
For compounds C3 and C5, the metal precursor (K2PtCl4 or cis-[Pt(dmso)2Cl2]) has been allowed to react with 1 molar equivalent of the ligands in different solvents (see experimental part 3.2.6) and the products are obtained as yellow precipitates. For compounds C4 and C6, cis-[PtCl2(NH3)2] is used as precursor. Recrystallised cisplatin is activated by using 2 molar equivalent of AgNO3, followed by ligand addition. The desired products are obtained by thorough washing with ether (and acetone, for C4) and drying in air.
The one-dimensional (1H, 13C and 195Pt) and two-dimensional (COSY) NMR spectroscopic data provide unambiguous identification of the expected peaks. As a solvent, dmso-d6 or dmf-d7 is used for C3 and C5, whereas the high aqueous solubility of C4 and C6 allows recording the NMR spectra in D2O.
The ESI-MS analysis provides evidence for the major species present in solution.
For C3 and C5, direct infusion to the spectrometer shows the cationic species as a major peak, with one chloride substituted for dmso (solvent). For C4 and C6, the molecular ion species is the major peak along with some minor dissociation species. For elemental analysis the data are within the range when compared to the calculated weight percentage of the elements.
3.3.3. Optimised structure by DFT calculation
The structures of the compounds C3-C6 have been optimised with DFT in vacuum and are shown in Figs. 3.4-3.7. A direct comparison with experimental data is possible only for the compound C3, whose structure has been reported in the literature.8, 9 The crystals obtained for C5 and C6 are too small to diffract and no crystal structure is available for C4. Thus the structure of C3 has been used as a starting point to generate the initial models for the remaining compounds.9 The HOMO-LUMO pictures for the compounds are shown in Figs. 3.4-3.7. The comparison of some structure details for C3, calculated and observed, is given in Table 3.1.
Figure 3.4. Energy minimised structure of (a) C3, (b) HOMO and (c) LUMO.
Figure 3.5. Energy minimised structure of (a) C4, (b) HOMO and (c) LUMO.
(c) (a)
(b)
(c) (a)
(b)
Figure 3.6. Energy minimised structure of (a) C5, (b) HOMO and (c) LUMO.
Figure 3.7. Energy minimised structure of (a) C6, (b) HOMO and (c) LUMO.
(c) (a)
(b)
(c) (a)
(b)
Table 3.1. Comparison of geometric data for C3 with DFT calculation.
Experimental Calculated Bond lengths (Å) Bond lengths (Å)
Pt-Cl1 2.301 Pt-Cl1 2.331 Pt-Cl2 2.292 Pt-Cl2 2.332 Pt-N1 2.010 Pt-N1 2.084 Pt-N2 2.012 Pt-N2 2.086 Bond angles (°) Bond angles (°)
Cl-Pt-N 90.95 Cl-Pt-N 90.82 Cl-Pt-Cl 89.37 Cl-Pt-Cl 88.62 N-Pt-N 88.68 N-Pt-N 89.83
The calculated structure for C3 is in good agreement with the available crystal structure data (vide Table 3.1), both in terms of bond angles and bond lengths. DFT tends to slightly overestimate bond lengths by approximately 2% on average. The crystal packing effects are neglected in the model. The structures show that the aromatic dpa (L6) and dipm (L7) ligands are not planar. In particular for compound C3 the dihedral angle between the two pyridyl rings is predicted to be 148.5°, which compares well with the experimental value of 147.6°.9 This dihedral angle is calculated to be 142.8° in C4, 150.6° in C5 and 148.7° in C6. The substitution of chlorides with ammine ligands affects also the dihedral angle between the plane containing Pt and the monodentate ligands and the plane of the aromatic rings. This angle varies from ca. 152° in C3 and C5 to ca. 137°
in C4 and C6. These structural features may affect the interaction with the biological targets and ultimately the in vitro cytotoxicity, in addition to differences in chemical behaviour, like solvolysis.
Compounds C3 and C5 have the same coordination sphere around the platinum i.e., two N-atoms and two chlorine atoms. Similarly the other two compounds (C4 and C6) have a four N-atom environment, two of which originate from monodentate ammine ligands and the other two are from the bidentate ligands. The HOMO and LUMO pictures are provided along with the optimised structures. As might have been expected, the HOMO and LUMO character remains the same when changing the ligand from dpa to dipm, but it is strongly modified by substituting Cl- with NH3. From the HOMO-LUMO pictures it is clear that in C3 and C5 the nonbonding electrons reside in the p-orbital of
chlorides. And they can take part in pπ-dπ bonding, but there is no such situation for NH3
ligands in C4 and C6.
3.3.4. Model-base studies with 9-EtG
The model base 9-ethyl guanine is very widely used to follow the binding of metal compounds (as platinum, ruthenium or gold) by NMR, or ESI-MS analytical techniques.27-39 This simple molecule binds to metal compounds in a straightforward way to form kinetically stable adducts. Generally, the easily hydrolysable ligands are replaced by 9-EtG in a time scale which can be followed spectroscopically. The numbering scheme for the platinum compound and 9-EtG is shown in Fig. 3.8.
N N
NH
Pt Cl Cl
N N
N N NH
Pt
Cl Cl
N N
N NH O
NH2 1 H
2 3 4 6 5
7
1' 2' 3' 4'
5' 6'
1 2
3 4 6 5
7
1' 2' 3' 4'
5' 5 1 6'
2 3 4
6 7
8 9
(a) (b) (c)
Figure 3.8. Numbering scheme of atoms for (a) C3, (b) 9-EtG and (c) C5.
The model base is added in different molar equivalence to the platinum compound (C3 and C5) NMR spectra have been recorded within certain time intervals. In the presence of an excess of 9-EtG (1:4 molar ratio of Pt:9-EtG) prominent changes starts appearing for both C3 and C5 after 45 min at 37 °C and the changes continue up to 18 h.
The changes in the proton NMR spectra are shown in Figs. 3.9-3.12 and they do not show any further changes over days. The spectra of the adducts remain the same when different amounts of model base are used and the some of these selected stacked spectra are shown below. Different symbols have been used to differentiate between unreacted 9-EtG (♦), starting platinum compounds (C3 or C5; ●) and the adducts with 9-EtG (■) in the Figs.
3.9 and 3.11, respectively.
(a) C3 with 9-EtG
For C3, two different ratios of platinum compound to model base (1:1 and 1:4) have been studied at 37 °C (as physiological condition) in dmso-d6. The presence of
Figure 3.9. Time-dependent changes in the 1H NMR spectrum of C3 upon addition of 4 equivalent of 9-ethylguanine (in dmso-d6) with product marking, as unreacted 9-EtG (♦), C3 (●) and adduct with 9-EtG (■).
●
♦
■
●
■
■ ■
■ ■
●
● ●
H8E H1E
H2E -CH2E
H7c
H2,2'c H3,3'c
H4,4'c H5,5'c H7A
H2,2'A H5,5'A H4,4'A
-CH2A
Figure 3.10. Time-dependent changes in the 195Pt NMR spectrum of C3 upon addition of 4 equivalents of 9-EtG (in dmso-d6).
characteristic peaks from starting materials, C3 and 9-EtG has revealed that the reaction is not complete after days. 195Pt NMR also reveals the presence of three different species in the reaction mixture. Three peaks can be assigned as PtN2Cl2 (-2086 ppm),40 PtN3Cl (- 2216 ppm) and PtN2ClS (-2835 ppm) coordination sphere according to the database.41, 42 The gradual changes in the stacked spectra are depicted in Figs. 3.9-3.10.
PtN3Cl
PtN2Cl2
PtN2ClS
Figure 3.11. Time-dependent changes in the 1H NMR spectrum of C5 upon addition of 4 equivalents of 9-EtG (in dmso-d6) with product marking, as unreacted 9-EtG (♦), C5 (●) and adduct with 9-EtG (■) and solvated species (◘), respectively.
▪
♦
♦ ♦
▪ ▪ ▪
● ●
●
●
H1E H2E
H7c H2,2'c H7A
H3,3'c
H4,4'c ◘ ◘
● H2,2'A H3,3'A
Figure 3.12. Time-dependent change in the 195Pt NMR spectrum of C5 after reaction with 4 equivalents of 9-EtG.
After solvolysis of one of the chloride ligands the compound [Pt(dpa)(dmso)Cl]Cl with a PtN2ClS coordination sphere appears to be formed (the same peak is observed without addition of 9EtG). The peak did not increase or decrease in intensity after 36 h.
The third peak appearing at -2216 ppm, corresponds to Pt in a N3Cl environment. This indicates that one of the chloride ligands has been replaced by 9-EtG, giving rise to [Pt(dpa)(9-EtG)Cl]Cl, the mono substituted adduct. The appearance of the three peaks (relative peak heights 1:5:2) was found not dependent on the ratio of the model base to the platinum compound during the observed reaction period. The presence of the three species was further supported by ESI-MS analysis of the reaction mixture directly from NMR tubes.
(b) C5 with 9-EtG
The 1H NMR spectra of C5 with 4 equivalent of 9-EtG are shown in Fig. 3.11.
After 45 minutes new peaks are showing up, confirming that a reaction is taking place
PtN3Cl
PtN2Cl2 PtN2ClS
between the platinum compound and 9-EtG. Both the chloride ligands from the platinum compound may be solvolysed by dmso, or substituted by 9-EtG. From the 195Pt NMR spectra, shown in Fig. 3.12, changes can be observed with time. C5 gives rise to a peak at -2048 ppm, corresponding to platinum in a N2Cl2 envionment.40 After 6 h, a new peak appearing at -2835 ppm, corresponds to platinum in a N2ClS environment (from [Pt(dipm)(dmso)Cl]Cl species), via solvolysis.41, 42 The peak, surprisingly, appears not to increase or decrease in intensity over the observed time window (18 h). The third peak appears at -2177 ppm, corresponding to platinum in a N3Cl environment (from [Pt(dipm)(9-EtG)Cl]Cl species via substitution). This indicates that one of the chloride ligands has been replaced by 9-EtG, giving rise to the mono substituted adduct. The presence of the three species has been confirmed by ESI-MS (data not shown). Reactions of C4 and C6 were not investigated, because of the expected inertness under these conditions.
3.3.5. In vitro cytotoxicity
The results of the cytotoxicity assays are summarised in a tabular form. The typical 120 h incubation for non-metallic organic reference drugs are summarised in Table 3.2, whereas the results for platinum drugs for 120 h and 48 h incubation period have been depicted in Tables 3.3 and 3.4, respectively. Each sample was tested for ten different concentrations (from freshly prepared 5 mg/mL dmso stock solutions) and in quadruplicate. The 48 h incubation time was studied with freshly prepared samples from a stock in milliQ water (1 mg/mL) and investigated in the same cell lines. These two parameters have been changed in order to find out the effect of various activated species (via solvolysis) and duration of incubation period.
The SRB assay performed in slightly different conditions than common practice, did lead to a significant observation. Usual experimental methods in SRB assay involve incubation times of 120 h and a stock solution of 5 mg/mL in dmso (followed by subsequent dilution with medium). The reference compounds used generally are organic drugs (except cisplatin) with no structural similarity or mode of action. The cell lines used are seven human tumour cell lines with different origin and type. Among the organic compounds the 5-Flurouracil is the least active one, which is significant from the high IC50 values in all cell lines corresponding to higher amount of compound needed for 50%
growth inhibition. Taxol, methotrexate and doxorubicin are all equally hyper-active in all cell lines, which is prominent from the nanomolar values of IC .
Table 3.2. IC50 values (µM) for a selection of standard organic drugs (120 h incubation).
Cell lines
Samples A498 EVSA-T H226 IGROV M19 MEL WIDR MCF7
Doxorubicin 0.16 0.01 0.36 0.11 0.02 0.01 0.02 5-flurouracil 1.09 3.65 2.61 2.28 3.39 5.76 1.72
Methotrexate 0.08 0.01 5.03 0.01 0.05 0.04 0.01
Etoposide 2.23 0.53 6.68 0.98 0.85 4.40 0.25 Taxol < 0.004 < 0.004 < 0.004 < 0.004 < 0.004 < 0.004 < 0.004
Table 3.3. IC50 values (µM) for platinum drugs, cisplatin and C3-C6 (120 h incubation).
Cell lines
Samples A498 EVSA-T H226 IGROV M19 MEL WIDR MCF7
cisplatin 7.51 1.41 10.9 0.56 1.85 2.32 3.22
C3 74.2 16.7 62.6 62.6 47.7 60.3 52.8
C4 68.1 > 100 64.5 > 100 88.8 98.7 94.6
C5 17.1 6.33 17.8 19.4 18.6 12.1 19.2
C6 78.2 13.4 76.4 65.4 88.8 > 100 > 100
Table 3.4. IC50 values (µM) for several platinum compounds (48 h incubation).
Cell lines
Samples A498 EVSA-T H226 IGROV M19 MEL WIDR MCF7
C4 > 100 17.2 > 100 > 100 > 100 > 100 > 100 C6 > 100 7.08 > 100 71.2 > 100 > 100 > 100 Carboplatin 56.5 27.7 30.9 10.3 18.1 37.1 27.2 Oxaliplatin 0.054 0.44 0.64 1.49 0.90 0.54 0.65
[Pt(dach)Cl2] 0.44 0.51 0.56 1.29 0.79 0.49 0.62
In case of the platinum reference compounds the behaviour is different.
Oxaliplatin and [Pt(dach)Cl2] are very active even in 48 h of incubation time, in all cell lines. The low IC50 values in all cases are less then 2 µM for these two compounds. On
the other hand, carboplatin and cisplatin exhibit higher IC50 values compared to organic or the other two platinum compounds.
Under similar experimental condition the only active newly synthesised compound is C5. The activity is 1.6-fold (H226) to 10-fold (M19MEL) less, compared to cisplatin. In spite of similarity in the structure (there are two hydrolysable chloride groups on both of them, so a similar mode of action is expected) of C5 [Pt(dipm)Cl2] with cisplatin, the activity towards cancer cell lines are quite different. The other cisplatin analogue with N2Cl2 coordination sphere, C3 [Pt(dpa)Cl2] is active in only EVSA-T cell line. The other two coordinatively saturated compounds, C4 [Pt(dpa)(NH3)2]2+ and C6 [Pt(dipm)(NH3)2]2+ are showing almost no activity in the chosen cell lines. These two compounds having 2+ charges are expected to strongly interact with DNA at least by electrostatic attraction to phosphate backbone; the compounds in a way resemble those [Pt(en)(N-N)]2+ reported by Aldrich-Wright et al. which show activity (L1210 cell line).43, 44, 45
When the incubation time is changed from 120 h to 48 h, the two highly water soluble platinum compounds; C4 and C6 show enhanced activity in EVSA-T cell lines.
When compared with carboplatin the activity increases up to 4 and 1.5 fold, respectively.
In this experiment the stock solution is prepared in milliQ water, but absence of any easily hydrolysable ligand on platinum centre nullifies the activation of sample (unlike the cisplatin activation via hydrolysis). Therefore the proper explanation of the enhanced activity is yet to be unravelled.
3.4. Conclusions
In this study successful synthesis and characterisation of three new platinum compounds (C4, C5 and C6) with reproduction of fourth one (C3) are reported. The structures of all four coordination compounds have been energy-minimised using DFT calculations. This optimisation confirms that all the compounds are not absolutely planar, but rather more stabilised in a puckered position. The bidentate ligands (L6 and L7) induce a fold around the central–NH moiety of an open book structure.
The compounds with two chloride ligands, C3 and C5, were reacted with DNA- model base, 9-Ethylguanine in dmso solution. Both the compounds exhibit reaction to form a monoadduct with 9-EtG; despite excess of ligand, no bis-adduct is observed as formed by cisplatin.
The activity of C5 towards selected cell lines is promising to balance toxicity and selectivity. The cationic compounds (C4 and C6) are significantly active towards EVSA- T cell lines, whereas inactive towards MCF-7 cell lines (both cell lines are breast cancer cells with difference in their hormone level). These compounds might be potential candidates for use in leukemic cell lines as these cationic [Pt(en)(N-N)]2+ compounds exhibit significant activity in L1210 cell lines. All the compounds could bind DNA as primary target though the detailed cellular process is needed to be explored. Modification of the ligands, chemically and sterically, might give further clues to this hypothesis. The attachment of fluorescent tag or a receptor target with the amine N-atom of the ligand offers to follow the drug in vitro and to target specific organs respectively. In addition, the selective changes in the carrier ligand modify the activity profile, solubility and bio- availability. The following chapter (Chapter 4) will deal with detailed DNA and protein interactions of these four platinum compounds.
3.5. References
1. Reedijk, J., J. Inorg. Biochem. 2003, 96, 81-81.
2. Reedijk, J., Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 3611-3616.
3. Jamieson, E. R.; Lippard, S. J., Chem. Rev. 1999, 99, 2467-2498.
4. Kelland, L., Nat. Rev. Cancer 2007, 7, 573-584.
5. Brabec, V.; Kleinwachter, V.; Butour, J. L.; Johnson, N. P., Biophys. Chem. 1990, 35, 129-141.
6. Brabec, V., Prog. Nucleic Acids Res. 2002, 71, 1-68.
7. Brána, M. F.; Cacho, M.; Gradillas, A.; de Pascual-Teresa, B.; Ramos, A., Curr.
Pharm. Des. 2001, 7, 1745-1780.
8. Li, D. C.; Liu, D. J., Cryst. Res. Technol. 2004, 39, 359-362.
9. Tu, C.; Wu, X. F.; Liu, Q.; Wang, X. Y.; Xu, Q.; Guo, Z. J., Inorg. Chim. Acta 2004, 357, 95-102.
10. Rauterkus, M. J.; Fakih, S.; Mock, C.; Puscasu, I.; Krebs, B., Inorg. Chim. Acta 2003, 350, 355-365.
11. van Albada, G. A.; Mutikainen, I.; Turpeinen, U.; Reedijk, J., Inorg. Chem.
Commun. 2006, 9, 1067-1070.
12. van Albada, G. A.; Mutikainen, I.; Turpeinen, U.; Reedijk, J., J. Mol. Struct. 2007, 837, 43-47.
13. Paull, K. D.; Shoemaker, R. H.; Hodes, L.; Monks, A.; Scudiero, D. A.;
Rubinstein, L.; Plowman, J.; Boyd, M. R., J. Natl. Cancer Inst. 1989, 81, 1088- 1092.
14. Bracht, K.; Boubakari; Grunert, R.; Bednarski, P. J., Anti-Cancer Drugs 2006, 17, 41-51.
15. Dhara, S. C., Ind. J. Chem. 1970, 8, 193-6.
16. Price, J.; Schramm, R.; Wayland, B.; Williams, A., Inorg. Chem. 1972, 11, 1280- 1284.
17. Yao, W. B.; Kavallieratos, K.; de Gala, S.; Crabtree, R. H., Inorg. Chim. Acta 2000, 311, 45-49.
18. Brown, M. XWIN plot : simple and easy plotting, Version 01092006.meb 19. Interran, L. V.; Messmer, R. P., Inorg. Chem. 1971, 10, 1174-1178.
20. http://www.scm.com
21. Guerra, C. F.; Snijders, J. G.; te Velde, G.; Baerends, E. J., Theor. Chem. Acc.
1998, 99, 391-403.
22. te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Guerra, C. F.; van Gisbergen, S.
J. A.; Snijders, J. G.; Ziegler, T., J. Comput. Chem. 2001, 22, 931-967.
23. Becke, A. D., Phys. Rev. A 1988, 38, 3098-3100.
24. Lee, C. T.; Yang, W. T.; Parr, R. G., Phys. Rev. B 1988, 37, 785-789.
25. van Lenthe, E.; Ehlers, A.; Baerends, E. J., J. Chem. Phys. 1999, 110, 8943-8953.
26. Keepers, Y. P.; Pizao, P. E.; Peters, G. J.; van arkotte, J.; Winograd, B.; Pinedo, H. M., Eur. J. Cancer 1991, 27, 897-900.
28. Aoki, K.; Salam, M. A.; Munakata, C.; Fujisawa, I., Inorg. Chim. Acta 2007, 360, 3658-3670.
29. Bierbach, U.; Farrell, N., J. Biol. Inorg. Chem. 1998, 3, 570-580.
30. Chifotides, H. T.; Hess, J. S.; Angeles-Boza, A. M.; Galan-Mascaros, J. R.;
Sorasaenee, K.; Dunbar, K. R., Dalton Trans. 2003, 4426-4430.
31. Garza-Ortiz, A.; Maheswari, P. U.; Siegler, M.; Spek, A. L.; Reedijk, J., Inorg.
Chem. 2008, 47, 6964-6973.
32. Kampf, G.; Willermann, M.; Freisinger, E.; Lippert, B., Inorg. Chim. Acta 2002, 330, 179-188.
33. Komeda, S.; Lutz, M.; Spek, A. L.; Yamanaka, Y.; Sato, T.; Chikuma, M.;
Reedijk, J., J. Am. Chem. Soc. 2002, 124, 4738-4746.
34. Navarro, J. A. R.; Salas, J. M.; Romero, M. A.; Vilaplana, R.; Gonzalez-Vilchez, F.; Faure, R., J. Med. Chem. 1998, 41, 332-338.
35. Peacock, A. F. A.; Habtemariam, A.; Fernandez, R.; Walland, V.; Fabbiani, F. P.
A.; Parsons, S.; Aird, R. E.; Jodrell, D. I.; Sadler, P. J., J. Am. Chem. Soc. 2006, 128, 1739-1748.
36. Raudaschl-Sieber, G.; Schollhorn, H.; Thewalt, U.; Lippert, B., J. Am. Chem. Soc.
1985, 107, 3591-3595.
37. Sartori, D. A.; Miller, B.; Bierbach, U.; Farrell, N., J. Biol. Inorg. Chem. 2000, 5, 575-583.
38. Zamora, F.; Sabat, M., Inorg. Chem. 2002, 41, 4976-4977.
39. Corral, E.; Hotze, A. C. G.; Magistrato, A.; Reedijk, J., Inorg. Chem. 2007, 46, 6715-6722.
40. Lakomska, I.; Szlyk, E.; Sitkowski, J.; Kozerski, L.; Wietrzyk, J.; Pelczynska, M.;
Nasulewicz, A.; Opolski, A., J. Inorg. Biochem. 2004, 98, 167-172.
41. Berners-Price, S. J.; Ronconi, L.; Sadler, P. J., Prog. Nucl. Magn. Reson.
Spectrosc. 2006, 49, 65-98.
42. Still, B. M.; Kumar, P. G. A.; Aldrich-Wright, J. R.; Price, W. S., Chem. Soc. Rev.
2007, 36, 665-686.
43. Jaramillo, D.; Buck, D. P.; Collins, J. G.; Fenton, R. R.; Stootman, F. H.; Wheate, N. J.; Aldrich-Wright, J. R., Eur. J. Inorg. Chem. 2006, 839-849.
44. Jaramillo, D.; Collins, J. G.; Junk, P.; Aldrich-Wright, J. R., J. Inorg. Biochem.
2003, 96, 159-159.
45. Kemp, S.; Wheate, N. J.; Buck, D. P.; Nikac, M.; Collins, J. G.; Aldrich-Wright, J.
R., J. Inorg. Biochem. 2007, 101, 1049-1058.
