Synthesis and Anticancer Activity of Gold Porphyrin Linked to Malonate Diamine Platinum Complexes

HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.


Introduction
Cancer still remains one the major cause of mortality in the world in spite of the many progresses made in medical treatments for several decades. Conventional cancer treatment includes surgery, chemotherapy and radiotherapy. Platinum based compounds, such as Cisplatin, are widely used as anticancer therapeutic agents, but they have dose-limiting side effects and some cancer lines have become resistant to these drugs. 1 3 Photodynamic therapy (PDT) is an approved clinical treatment in oncology. It has many advantages such as low-cost, less invasiveness and minimal side-effects than conventional chemotherapy. 2 PDT requires a photosensitizer, such as a porphyrin derivatives, which are known to better concentrate within the tumor tissue. 3 Currently, many porphyrin based photosensitizers are developed used in the clinic. 2c, 34 The Photofrin was the first photosensitiser approved by the Food and Drug Administration (FDA) in 1993 for bladder cancer therapy. Now, it is used for several cancers. Then, a second generation of photoactive compounds was produced and developed for clinical use such as Foscan, Metvix, Laserphyrin, Redaporphin. In addition, several compounds such as Tookad, Radachlorin are under clinical trials for cancer treatment. Light excitation of photosensitizers produces reactive oxygen species (ROS) in tumor and ultimately leads to cancer cell death, a method that confers higher selectivity to cancer cells than Cisplatin-based chemotherapy treatments.
The combination of a porphyrin or a phthalocyanine with an appended platinum, 5 gold, 6 or ruthenium, 7 complex has been attempted to design innovative anticancer drugs. Interesting synergetic anticancer properties are obtained thanks to the dark chemostatic effect of the metal complex amplified by the PDT effect generated upon light excitation of the porphyrin or the phthalocyanine. 5a, b Recently, Che and co-workers reported the great pverotential of cationic gold(III) porphyrins as cytotoxic agent towards various cancer cell lines. 8 This discovery has stimulated many studies with gold porphyrins to develop anticancer drugs. 9 Interestingly, while Cisplatin anticancer drugs bind to DNA and induce cell death by binding to purine nucleic bases such as guanine and adenine, 1a gold porphyrins target mitochondria and particularly the heat shock protein. 10 As a consequence, the combination of Cisplatin derivative with gold porphyrin could lead to synergistic effect owing to different modes of action.
The most general mechanism of action of PDT is the generation of singlet oxygen by energy transfer from the photosensitizer excited triplet state to the triplet ground state of oxygen (classified as type II mechanism in PDT). It is well-accepted that photosensitizers with triplet excited states are more favorable than singlet ones to sensitize oxygen, because oxygen has a triplet ground state therefore energy transfer process from excited states of similar multiplicity are more efficient. 11 Gold porphyrins are known to undergo intersystem crossing with unity quantum yield owing to the heavy atom effect induced by the presence of gold. 12 As a result, itis reasonable to anticipate that gold porphyrins could be promising candidates for PDT, such as it was shown for palladium porphyrins. 11d, e However, to the best of our knowledge, there is so far no single study of the phototoxic activity of gold porphyrin derivatives.
In this work, we have investigated the combination of Cisplatin derivatives covalently connected to a gold porphyrin in order to take advantage, on one hand, of the intrinsic cytotoxicity of the platinum complex added to that of the gold porphyrin and, on the other hand, of the potentially 4 high phototoxicity of gold porphyrin upon light excitation (Chart 1). In addition, the affinity of porphyrin for tumor cells could also enhance selectivity and concentration of Cisplatin into tumor and thus increase the anticancer activity. However, this affinity is not sufficient legitimating that targeting strategies are currently explored to enhance the selectivity towards cancer cells. 13 Overall, the gold porphyrin platinum complex conjugates could have superior anticancer activity than individual compounds.
To test this idea, we report herein the preparation and the biological properties of three new dyads composed of gold porphyrin connected to a platinum diamine complex liganded via a malonate anchor (Chart 1). The diamine ligands around platinum(II) used in this work are ammoniac, cyclohexane diamine (CyDA) and pyridine methyl amine (Py).These three amine ligands where selected in this study, because related platinum complexes, such as Cisplatin, Picoplatin and oxaliplatin, have shown effective anticancer activities. 1a, 5f, 14 The three dyads were tested in human healthy and tumor cell lines to quantify their cytotoxicity without light excitation and their phototoxicity under light excitation.

Chart 1. Structures of the compounds investigated in this study.
It was demonstrated that gold(III)/Pt(II)conjugates are more potent by 2 to 5.6-fold than the corresponding platinum complexes. The cyclohexyldiamine (CyDA) derivative exhibited the most cytotoxic effect among the series. Singlet oxygen measurements indicated that gold(III) porphyrin derivatives are poor oxygen sensitizers and cell death occurred potentially due to generation of others 5 reactive oxygen species (ROS) upon reductive quenching of the gold porphyrin excited state. Overall this study indicates that the incorporation of two different cytotoxic metals in the same molecule represents a remarkable cytotoxic effect compared to traditional homometallic Pt(II) drugs and that an increased cancer cell death was obtained after light irradiation only in cancer cells.

Synthesis of the compounds
The synthesis of the reference platinum compounds (Pt-NH3, Pt-Py and PtCyDA) is illustrated in Scheme 1. First, p-cresol 1 was O-alkylated with bromoditertbutyl malonate 2 according to a Williamson substitution reaction affording 3 in 36% yield. We suspect that this lower yield stems from a second O-alkylation of compound 3, concomitant with its bromination and further nucleophilic substitution by p-cresol as already reported on a similar compound. 15 The tertbutyl esters of the malonate were subsequently hydrolyzed with trifluoroacetic acid (TFA) to give compound 4. Finally, the dicarboxylato platinum complexes were obtained in 92% yield according to Dhara's methodology, which consists in activating the carboxylic acid groups by deprotonation using sodium hydroxide before being reacted with the platinum amino precursor containing two weakly bound aqua ligands. 14a, 16 Scheme 1. Synthetic route for the preparation of the reference platinum complexes. Reagents and conditions: a) THF, NaOH, RT, 24h, 36% ; b) TFA, 70 o C, 1h, 100% ; c) EtOH, NaOH ; d) EtOH/Water (5/5), diamine(dinitro)platinum(II), RT, 48h, 92%.

6
The preparation of the dyads made of gold porphyrin/platinum complex required the porphyrin 8 as key intermediate (Scheme 2). Towards this goal, the ditertbutyl 4-(4-formylbenzyl)malonate 7 was first prepared by C-alkylation of the 4-(bromomethyl)-benzaldehyde6 in 64% yield. The tertbutyl ester was preferred over methyl or ethyl malonate owing its fast and quantitative hydrolysis in acidic conditions. Indeed, gold porphyrins are electron deficient molecules and in basic conditions they can undergo nucleophilic addition of hydroxide in meso position. 17 The key porphyrin 8 was synthesized by crosscondensation of pyrrole with benzaldehyde and ditertbutyl 4-(4-formylbenzyl)malonate 7. Using Adler conditions (reflux of the reagents in propionic acid), 18 the porphyrin was obtained in 12% yield, while Lindsey conditions (reaction conducted at RT in CH2Cl2) 19 with BF3-OEt2 as catalyst afforded 8 in lower yield (5%). However, Lindsey conditions with a mixture of two catalysts BF3-OEt2/TFA: 0.019/1 at 0.01 M improved the yield of 8 to 22%.
The insertion of Au(III) into the porphyrin was performed according to Sauvage methodology 20 using a Au(I) based complex surrounded with weakly binding ligand (THT=tetrahydrothiophene) and can be conducted under mild reaction conditions (Scheme 2). Classical conditions, using the salt KAuCl4 in refluxing acetic acid, were too harsh and caused hydrolysis of the ester groups and partial decarboxylation. Lastly, the introduction of the platinum complex was accomplished with Dhara's methodology similarly as for the reference complexes except that the solvent was adapted to these more hydrophilic compounds (mixture of ethanol and water). The complexes were characterized by proton NMR and IR spectroscopy, elemental analysis as well as by high resolution mass spectrometry.
The dyads AuP-PtNH3, AuP-PtPy and AuP-PtCyDA are soluble in ethanol, DMSO and acetone. 7 Scheme 2. Synthetic route for the preparation of the dyads gold porphyrin linked to platinum complex.

UV-vis electronic absorption spectra
The absorption spectra of the dyads gold porphyrin/platinum complex along with that of the reference AuTPP recorded in dichloromethane are shown in Figure 1 and the spectroscopic data are gathered in Table 1. The spectra of the dyads are essentially dominated by the transitions of the gold(III)porphyrin, since the Pt complexes does not exhibit any absorption band in the visible region. The presence of the malonate on one phenyl substituent does not modify the transition on the porphyrin, since there is no -conjugation with it. In addition, the meso aryl substituents of porphyrins are known to orient with circa 60° angle preventing electronic interaction between them. 21 8

Singlet oxygen measurements
Most of the porphyrins used in PDT operate according to a type II mechanism, that is the production of singlet oxygen by energy transfer from the photosensitizer excited state to oxygen present in the biological tissues. 2a, b The singlet oxygen quantum yield production was therefore measured to assess the potential of these gold porphyrin systems to act as sensitizers for PDT. The measurements were recorded in methanol and the production of singlet oxygen was assayed by recording the band at 1270 nm of singlet oxygen emission after gold porphyrin systems illumination. Unfortunately, none of the gold porphyrin and dyads prepared in this work produced detectable luminescence signal setting, thus a singlet oxygen quantum yield was below 1%. This unexpected result probably stems from the short lived triplet excited state of gold(III) porphyrins (≈ 1.5 ns), which decay to ground state before the energy transfer to surrounding oxygen could take place. 12,23 Biological activity In order to determine the therapeutic potential of these new compounds, we have first studied the cytotoxicity induced in human breast cancer cells (MCF-7) and in healthy human fibroblasts (FS-68).
The cells were incubated in darkness during 72 h with increasing concentrations of each compound (from 0.01 to 100 µM). The concentrations of the drugs which led to 50% cell mortality (IC50) were determined. As shown in Figure 2, obtained cytotoxicity data in both cell lines showed the classical sigmoidal dose-response curves when plotted as a logarithmic function of concentration (µM). Importantly, the cell death is very low in healthy fibroblasts, IC50 is too high to be calculate precisely (≥100 µM) (Figure 2c). This result highlights the targeting of cancer cells by these compounds. Then, the cytotoxic activity of gold porphyrins AuTPP, 9 and 10 was investigated in both human breast cancer cells ( Figure 3a) and in healthy fibroblasts cells (Figure 3b). The results first show that overall, the gold porphyrins are more potent than the reference platinum complexes. Although the difference of cytotoxicity in cancer cells between Pt-CyDA and 10 is small, but on healthy fibroblasts this difference become very high; close to 100 fold (Table 3c). In addition, the introduction of the malonate substituent on the porphyrin decreases its cytotoxic activity, since the simple AuTPP has the lowest IC50. This result is consistent with the report of Che and co-workers, showing that any substituent pattern on AuTPP results in a weakening of its cytotoxicity. 8a Important features for the anticancer activity of gold(III) porphyrins are the size of the macrocycle, which must fit in the cavity of the target receptor (certainly heat shock protein) 10 and the lipophilicity of the porphyrin because its biological properties might be related to the fact it is a planar aromatic organic cation. 26 Thus, complex 9 with bulky hydrophobic malonate substituent displayed the least activity, which might be attributed to traffic entrance into cells. A long floppy alkyl chain might have been more favorable to tether the platinum complex and to maintain the cytotoxic activity of gold(III) porphyrin. Finally, we note that for gold porphyrins AuTPP and 10, the cytotoxicity is lower on healthy cells than cancerous ones, a significant advantage to decrease side effects during the treatment. Similarly as for the reference gold porphyrins (AuTPP, 9 and 10), light excitation increases cell mortality, but to a low extent ( Figure 6). Similarly as for the reference gold porphyrins (AuTPP, 9 and 10), light excitation increases cell mortality, but to a low extent ( Figure 6). In addition, AuP-PtCyDA does not exhibit PDT efficiency in these conditions." shown that all these porphyrin derivatives are poor oxygen photosensitizers, therefore it is likely that the ROS production is induced by generating cytotoxic radicals via electron transfer from excited AuP + to surrounding organic molecules such as glutathione, a ubiquitous antioxidant in cells, upon reductive quenching of AuP + excited state, which is a strong oxidant. 12 Accordingly, PDT effect can arise by a type I mechanism, which passes via the formation of oxidized species, inducing thus an oxidative stress by oxidizing GSH to glutathione disulfide. Photosensitizers operating upon a type I mechanism is particularly interesting in hypoxic environments found in many solid tumours.

Conclusion
For the first time, we report three new dyads consisting of two appended cytotoxic moieties such as a gold(III) porphyrin and a platinum complex and their biological properties towards cancer cells were investigated in the dark (chemostatic effect) and upon light excitation (PDT effect) and compared to their corresponding individual platinum(II) complexes. To evaluate the decrease in side effects that could occur with such compounds, the effect on healthy cells was also studied. The main results of this work are: i) the diamine ligand around the Pt centre has a high impact on the cytotoxicity of the resulting complex and cyclohexane diamine is the best ligand; ii) the gold porphyrins are poor singlet 16 oxygen photosensitizers and thereby exhibit low phototoxicity with the present systems, but improvement can be obtained with modification of the porphyrin core, such as fluorination. The presence of fluorine atoms on meso phenyl substituents could enhance both the oxidizing power 29 and the lifetime of the triplet excited state. Indeed, it is postulated that gold porphyrins display ligand to metal charge transfer transitions, which are responsible of the fast decay of the triplet excited state. 23 However, the phototoxicity is not null and most probably result by a type I mechanism; iii) the introduction of a malonate group of one phenyl substituent of AuTPPs lightly decreases cyto-and photoxicity the gold porphyrin; iv) the dyad AuP-PtCyDA displays interesting cytotoxic activity towards cancer cells, for which the mechanism of action would be interesting to be investigated.The enhanced cytotoxicity indicates that porphyrins allow the delivery of the complex into cancer cells. The complex might have dual biological functioning between Au(III) and Pt(II) arms. The former as reported would interact with mitochondria and the latter would target DNA; both mechanisms cause cancer cell death.
Importantly, all compounds present a close or clearly higher cytotoxic activity in cancer cells than healthy cells and the PDT effect, even slight, induced around 15 or 20% more cancer cell death for AuTPP, AuP-PtNH3 and AuP-PtPy in conditions totally harmless for healthy cells (0% cell death).
Further biological analysis would clarify the anticancer activity if it is based on specific metal or both.
Additional analysis of the intracellular metal contents such as ICP-analysis to monitor the uptake of the whole stable bimetal conjugate by cancer cells would clarify the delivery mechanism. On the other hand, the incorporate mode of action exhibited by the two metals may be even amplified using an alternative photosensitizer. These studies are in due course in our laboratory.

Ditertbutylmalonate p-cresol (3).
A total of 1 g (9.24 mmol) of p-cresol 1 and 2.7 g (9.24 mmol) of bromoditertbutyl malonate2were dissolved under argon in 20 ml of distilled THF. 0.36 g (9 mmol) of pulverized NaOH was added, and the mixture was stirred for 24 h at room temperature. The organic solution was diluted with water and extracted with dichloromethane. The combined organic phases were washed with water and dried over Na2SO4.The solvent was evaporated and the crude product was purified by column chromatography (SiO2; CH2Cl2/petroleum ether 5/5). Yield: (1.08 g, 36% Then, they are sonicated during 30 seconds and diluted at the required concentrations in culture medium of each cell line".
Cytotoxicity study. MCF-7 and FS-68 cells were seeded into 96-well plates at a density of 10 3 cells/cm 2 .
Phototoxicity assay. Both cell lines, MCF-7 and FS-68 cells, were seeded into 384-well plates at a density of 10 3 cells/cm 2 and allowed to grow for 24 h. Cells were incubated or not (control) with 0.5 µM concentration of compounds solution for 24 h. After incubation, cells were exposed, or not, to light using mercury lamp of a fluorescence microscope at excitation ranges between 390-420 nm for 20 min (39 J.cm -2 ). Two days after, MTT assay was performed to evaluate the phototoxic effect of compounds.
ROS production. The detection of ROS was performed during the phototoxicity experiment. Forty five minutes before irradiation, cells were incubated with 20 µM concentration of DCFH2-DA(Cellular ROS Assay Kit, Abcam, USA). Cells were exposed to light using mercury lamp of a fluorescence microscope at excitation ranges between 390-420 nm for 20 min (39 J.cm -2 ). After irradiation, cells were washed and fluorescence emission of DCF was detected at 450 nm using fluorescent microscope.