The present invention relates to novel metallo tetrapyridyl complexes for use as cytotoxic anti-cancer or antibacterial agents, and to their use in the preparation of a medication for the treatment of unlimited cell growth, such as in infections or in cancer and tumour cells, and their preparation.

Background of the Invention

Current therapies directed against cells with uncontrolled growth properties, such as highly proliferating cells, cancerous cells, and tumour cells, require that they include a set of unique criteria in order to avoid damaging normal growing and/or non-proliferating cells.

The leading anticancer drug cisplatin is one of the landmarks in modern inorganic chemistry. In the body, cisplatin enters cells, hydrolyses, and forms irreversible adducts with DNA, ultimately triggering cell death via apoptosis. However, because of their lack of selectivity, severe side effects are commonly associated with the use of cisplatin, including nephrotoxicity, neurotoxicity, ototoxicity, and nausea. Furthermore, cancer cells can acquire increasing levels of cisplatin-resistance. Accordingly, new anticancer metallodrugs that may offer less severe side effects and may retain activity against resistant cells would be highly desirable.

Summary of the Invention

Accordingly, in a first aspect, the present invention relates to metallo tetrapyridyl ligand complexes based on periodic system column 10 or 11 transition metals according to claim 1, which have chemotherapeutic activities, particularly anticancer properties. In more detail, the present invention relates to metallo tetrapyridyl ligand complexes of the formula(l):

[Z-L] ^M Qn (l)

wherein Z represents a metal cation selected from the group consisting of palladium, platinum or gold or mixtures thereof;

L is a tetradentate polypyridyl ligand of the general formula N-N'-N'-N (II) bonded to Z to form a complex Z-L:

wherein group X may independently represent a bridging group connecting two adjacent pyridyl groups, and may be an amino group, an optionally substituted methylene group, carboxyl or sulfoxyl group, a sulphur atom, or an oxygen atom; wherein group Y may be as group X, or may independently represent two hydrogen atoms, halogen atoms, optionally substituted alkyl groups, optionally substituted aryl groups, optionally substituted allyl groups, optionally substituted ester groups, optionally substituted ether groups;

M is an integer representing the positive charge of the Z-L complex;

Q represents the counter-anion(s) ionically bonded to the Z-L complex; and n is an integer representing the number of counter anions ionically bonded to the Z-L complex; and

R, R', R" and R'" independently represent one or more hydrogen atoms, halogen atoms, hydroxide, carboxylic acid, optionally substituted alkyl groups, , optionally substituted allyl groups, optionally substituted ester groups, optionally substituted ether groups, or R, R', R" and R'" represent a monocyclic or bicyclic aliphatic structure with one pyridyl groups, or two groups R, R', R" and R'" may together form a tricyclic structure with two adjacent pyridyl groups, and

R ^iv and R may represent a carbon-carbon bridge between two adjacent pyridyl groups, or in the case wherein X and Y are both amino groups linking two pyridyl groups, may each individually represent a hydrogen atom, a halogen atom, an optionally substituted alkyl group, and an optionally substituted ether group.

In a second aspect, the present invention relates to the synthesis of the metallo tetrapyridyl ligand complexes coordinated to periodic system column 10 or 11 transition metals, in particular gold (III), palladium(ll), and platinum(ll). It has been surprisingly found that the in vitro cytotoxic activity of these complexes ranges from EC50 values in micromolar concentrations to spectacular nanomolar EC50 values.

In a further aspect, the present invention relates to the preparation of the metallo tetrapyridyl ligand complexes, their biochemical activity in vitro and in vivo, and to the use of the complexes in the manufacture of medications for the treatment of unlimited cell growth, such as in infections, in cancer and/or tumour cells.

Influence of pH, interactions with DNA, cellular uptake, cellular fractionation, cell cytotoxicity on several commercially available human cancer cell lines and 1 non-cancerous cell line, in vitro mechanism of action, and in vivo studies of the metallo tetrapyridyl ligand complexes are described below.

Brief Description of the Figures

Fig. 1 shows the chemical structures of a number of suitable ligands for the metallo tetrapyridyl ligand complexes of this invention.

Figure 2 shows the chemical structures of two tetrapyridyl ligands, H ₂ bapbpy and Hbbpya, used in Examples 1 to 5. Hbbpya is a tetrapyridyl ligand similar to H ₂ bapbpy except for its single non-coordinated amine bridge (while H ₂ bapbpy has two amine bridges).

Figure 3 shows a schematic synthesis of Pd(Hbbpya)CI ₂ ("[1]CI ₂ ") and Pt(Hbbpya)CI ₂ C'[2]CI ₂ ") .

Figure 4 shows a schematic synthesis of [Pd(H ₂ bapbpy)] )(PF6) ₂ C'[3](PF6) ₂ ") and

[Pt(H ₂ bapbpy)(PF ₅ ) ₂ ("[4](PF ₅ ) ₂ ")

Figure 5 shows palladium(ll) and platinum(ll) complexes of H ₂ bapbpy (N-(6-(6- (pyridin-2-ylamino)pyridin-2-yl)pyridin-2-yl)pyridin-2-amine ) and Hbbpya (N,N-bis(2,2'- bipyrid-6-yl)amine). Figure 6 shows the displacement ellipsoid of the cationic part of [1]CI ₂ (50% probability level). Counter anions and lattice solvent molecules have been omitted for clarity.

Figure 7 illustrates the evolution of the UV-vis spectrum of [1]2+ (left) and [2]2+ (right) upon increasing the pH of an HCI aqueous solution (33 mM) of the metal complex using NaOH. [Pd] = [Pt] = 67 μΜ.

Figure 8: depicts the tumour volume of Balb/c mice treated with (PF6) ₂ and [4] (PFe) ₂ . CT-26 cells were injected subcutaneously into the right flank of Balb/c mice. Mice were treated on day 3, 5, 7, 10, and 12 (indicated by black triangles) with either compound [3](PF6 at concentrations of 2.5 mg/kg (i.p.) or [4](PF6 at concentrations of 2.8 mg/kg (i.p). ***, p <0.001, two-way ANOVA, Bonferroni post-test.

Figure 9 shows the chemical structures of three additional tetrapyridyl ligands, hhbiqbpyl, H2biqbpy2 and H2biqbpy3, used in Examples 15 to 17.

Figure 10 schematically shows the three-dimensional configuration of [4](PF6) ₂

(right) as compared to the substantially planar configuration of [2]CI ₂ (left), thereby showing the distortion of [Pt(Hbbpya)]+ vs [Pt(Hbapbpy)]+ d8 metal complexes. The view is given from the side of the metal complexes, along one of the Pt-N directions, where N is a non- coordinated NH bridge.

Figure 11 shows [Pt(Hbapbpy)]+ compound in a crystal structure with a 4-way DNA junction, whereby the complex forms a 4-way junction with four DNA strands. Note the non-planar Platinum aromatic compounds does not simply intercalate within the DNA base pair double helix, but stabilizes a more complex superstructure.

Detailed description of the invention

Inorganic anticancer compounds typically interact with biomolecules following two different modes. A first mode may be direct binding of biomolecules such as DNA or proteins to the metal center to form a metal-ligand coordination bond. The formation of these coordination bonds is commonly preceded by the release of a ligand present in the original or intermediate complex. For instance the anticancer drug cisplatin and most of its derivatives rely on this mechanism of action.

The second type of interaction may be an indirect binding mode of the complex with its biomolecular target. Typically, one of the ligands of a metal complex intercalates via re-re or hydrophobic interactions to DNA or proteins. Typical intercalators consist of planar polyaromatic and extended polypyridyl ligands. Whereas organic intercalators such as daunorubicin are widely used in the clinics as cancer-treatment, no inorganic intercalators have been employed clinically so far.

The present invention relates to the use of coordinatively saturated tetrapyridyl metal complexes as potentially intercalating anticancer drugs. The complexes have a metal centre Z selected from the group consisting of palladium, platinum and or gold, preferably palladium(ll), platinum(M) or gold(lll); to facilitate a prodrug (photo)reduction strategy, a palladium(IV) or platinum(IV) center is preferred. The complexes also have cyclometallated versions of ligands , such as the H ₂ bapbpy,Hbbpya, H ₂ biqbpyl, H2biqbpy2 and H2biqbpy3 ligands. The metal ion Z is centrally bonded to the ligand, thus forming a complex.

N-N'-N'-N represents a tetradentate ligand. M represents the charge of the complex, and typically M=+2, but upon deprotonation of the NH bridging amines it can become lower by one (M=+l) or two (M=0) units. The number of counter anions will depend on the charge M of the complex and its protonation state. Solvates of these compounds, for example with acetonitrile, water, methanol, ethanol, or other organic molecules, are also included.

In the compound, group X may independently represent a bridging group connecting two adjacent pyridyl groups, and may be an optionally substituted amino group, an optionally substituted methylene group, carboxyl or sulfoxyl group, a sulphur atom, or an oxygen atom. Preferably, X represents an amino group, more preferably a N(H) group. It was found that in particular -N(H)- groups were suitable to solubilise the complex, and therefore render the complexes particularly suitable for use according to the subject invention. It was found that in particular the complex disclosed in Lei He et al., Journal of inorganic Biochemistry, 166, (2017), 135-140, while exhibiting some cytotoxicity, was difficult to dissolve, which precludes efficient and direct use as a drug, or requires additional pharmacological work to formulate in a water-soluble form.

Group Y may be as X, thereby leading to an open circular ligand structure, or it may independently represent two hydrogen atoms, halogen atoms, optionally substituted alkyl groups, optionally substituted aryl groups, optionally substituted allyl groups, optionally substituted ester groups, optionally substituted ether groups, thereby leading to an embracing, non-cyclic ligand.

R, R', R" and R'" independently represent one or more hydrogen atoms, halogen atoms, preferably chloride, bromide or fluoride, optionally substituted alkyl groups, optionally substituted aryl groups, optionally substituted allyl groups, optionally substituted ester groups, optionally substituted ether groups, hydroxyl groups, carboxylic acid groups, amines, or R represents a monocyclic or bicyclic aliphatic or aromatic structure with one pyridyl groups, or two groups R, R', R" and R'" may together form a tricyclic structure with two adjacent pyridyl groups, e.g. forming quinolone or isoquinoline structures. Preferably, there are only the necessary number of substituents on the pyridyl rings, e.g. chlorides or ether groups, which may vary the electronic structure and hence binding strength of the ligand to the metal ion.

Generally, suitable tetradentate ligands N-N'-N'-N typically comprise a good metal chelating ligand, and include, but are not limited to, the compounds disclosed in Figure 1.

Q represents a counter anion, for example, chloride or hexafluorophosphate (PF6), but it could also be triflate, iodide, perchlorate, nitrate, sulfate, phosphate, or other as set out below. The counter anions of the complex may be any useful counter ion that allows formation of the complex, as well as use in the in vitro or in vivo processes. Preferably, Q. is selected from the group consisting of fluoride, chloride, bromide, iodide, sulphate, methyl sulphate, tetrafluoroborate, hexafluorophosphate, tetraphenylborate, triflate (CF3SO2 ^"" ), nonaflate (CF ₃ (CF ₂ )3S02 ^" ), bis(triflyl)amide ((CF ₃ S0 ₂ )2N ), trifluoroacetate (CF3CO2 ^~ ), heptafluorobutanoate (CF3(CF ₂ )3CC>2 ^~ ), acetate (CH ₃ C0 ₂ ^~ ), trifluoroacetate (CF3CCV) , and trifluoromethanesulphonate (CF3SO3 ^~ ), perchlorate (CIO4 ^~ ), a nitrate ion (NO3 ^~ ), a nitrite ion (N02~), phosphate, tetraphenylate, trisphate, tosylate and/or methylsufonate (CH3SO3 ), preferably chloride or hexafluorophosphate. More preferred as chloride and

hexafluorophosphate anions.

The present invention preferably relates to a compound of the general formula:

More preferably yet, the present invention relates to a compound selected from:

, or

L

More preferably, the compound is a complex based on a tetradentate ligand H2bapbpy (6,6'-bis[N-(pyridyl)-l-amino]-2,2'-bipyridine) coordinated to Pd (II) or Pt (II).

It has been discovered that while palladium complexes as analogues of platinum complexes are typically disregarded as anticancer drugs due to their unfavorably fast ligand exchange kinetics, which may be a relevant feature crucial for its success as a homogenous catalysts for many coupling reactions, [3](PFe) ₂ and [4](PF6) ₂ were found to exhibit strikingly different biological activities, whereas [2]CI ₂ harbours a moderate biologically activity,

[4](PFe) ₂ exhibits high cytotoxicity comparable to cisplatin, while [3](PF6) ₂ shows activities against a wide range of human cancer cell lines in the nanomolar range. Remarkably, the palladium compound [3](PFe) ₂ was not only found to be much more cytotoxic than its platinum analogue [4](PF6) ₂ ; it appeared to exert a different mechanism of action.

Accordingly, it represents the first palladium anticancer drug that causes a complete collapse of the cell redox balance and shows promising in vivo anticancer activity.

It has also been discovered that, as seen from Figure 10, that certain metallo tetrapyridyl ligand complexes of this invention have a 3-dimensional configuration, such as for instance Hbapbpy complexes. It was found that due to this skewed conformation, whereby the pyridyl rests closest to each other are pushed out of the plane, these complexes no longer appear to intercalate in the helical bends of DNA or RNA molecules. Instead, such complexes were surprisingly found to form so-called 4-way junctions connecting four separate strands of DNA, an entirely different mechanism than traditional planar intercalators.

Without wishing to be bound to any particular theory, it is believed that this positively enhances the effectiveness of the complexes., as exemplified by the high cell toxicity also against cisplatin-resistant cancer cell lines.

It is believed that the biochemical effects of the subject metallodrugs such as those disclosed herein are not limited to DNA-damage induced apoptosis but may include coordination of the biologically available thiol glutathione to e.g. the cisplatin centre may be associated with detoxification. Cisplatin can also disturb cellular homeostasis and cell redox balance, and increases the presence of reactive oxygen species (ROS). In this context, the present metallodrugs are specifically designed to disturb the intracellular redox balance have been developed as anticancer compounds. As cancer cells already exhibit high metabolic activity and low level of antioxidants they have an increased sensitivity toward such perturbations.

The present invention also relates to a composition comprising a compound as defined herein above. Preferably, the composition further comprises a suitable solvent.

Preferably, the solvent comprises acetonitrile, water, methanol, ethanol or combinations thereof.

The present invention also relates to a pharmaceutical composition comprising a compound, and a pharmaceutically acceptable excipient, adjuvant, diluent and/or carrier.

In a further aspect, the present invention also relates to a compound for use as a cytotoxic agent and/or for use as an anti-cancer agent, preferably for cisplatin-resistant cancer cells.

In a further aspect, the present invention also relates to a composition for use in therapy.

In a further aspect, the present invention also relates to a compound, or a composition according to the invention, for use in treating cancer or a high growth infection, preferably testicular cancer, bladder cancer, lung cancer, colon cancer, skin cancer or ovarian cancer in a subject in need thereof.

In a further aspect, the present invention also relates to a method of treating a disease, wherein the method comprises administering a therapeutically effective amount of a compound as defined herein above, or a composition as defined herein, to a subject in need thereof. Preferably, the disease comprises testicular cancer, bladder cancer or ovarian cancer.

In a further aspect, the present invention also relates to the use of a compound or a composition comprising the compound according to the invention, for the manufacture of a medicament for the treatment of a disease in a subject in need thereof. Preferably, the disease is selected from at least one of: testicular cancer, bladder cancer or ovarian cancer, in particular in a platinum resistant cancer.

The choice of the metal centre may have a strong influence on the biological activity of the complex in vitro and in vivo, on its cellular uptake, and on its pKa.

Accordingly, in a further aspect, the present invention also relates to the in vitro, or in vivo use of a use of a compound or a composition comprising the compound according to the invention, for determining if the compound or composition permits retarding or stopping the growth of uncontrolled growth cells, and for retarding or stopping the growth of uncontrolled growth cells.

In a further aspect, the present invention also relates to a method of making a compound according to the invention, comprising the step of combining:

i) a ligand of Formula II, or a derivative thereof; with

ii) a metal salt selected from palladium, platinum, and/or gold, wherein ligand i) and metal salt ii) form the compound; and optionally isolating, and optionally formulating the isolated or obtained compound into a medication. Preferably, the metal is selected from palladium, platinum, or gold.

The following, non-limiting examples will illustrate the invention further.

General procedures

Chemicals were bought from Alfa Aesar (foPtCU), Acros organics (foPdCU), and Sigma-

Aldrich. Reactions were performed under a nitrogen atmosphere using standard Schienk techniques. NMR spectra were recorded on a Bruker DPX-500 spectrometer. Chemical shifts are indicated in ppm relative to TMS. Mass spectra were recorded by using a Thermoquest Finnagen AQA Spectrometer. UV-vis experiments were performed on a Cary Varian spectrometer. pH measurements were performed using a PHM220 lab pH meter. Metal concentrations for logP determination were measured on a Vista MPX ICP-OES. Example 1: Synthesis of [llCb (formula (V)):

ffK¾

(V) In a 2-necked round bottom flask foPdCU (304 mg, 0.72 mmol) and Hbbpya (171 mg,

0.53 mmol) were added into a degassed ethanol-water mixture (7:3, 100 mL) preheated at 80 °C. The resulting yellow suspension was stirred overnight at 80 °C in a nitrogen atmosphere resulting in a dark green/black suspension. The solution was cooled down to room temperature, filtered, and from the filtrate all solvents were rotary evaporated. The solids were dissolved in MeOH (5 mL) and purified by size exclusion chromatography (Sephadex L- 20) using MeOH as eluent. The yellow band was collected and MeOH was removed by rotary evaporation, which resulted in [l]Cb as a yellow powder. Yield: 42% (111 mg). ¹ H NMR (500 MHz, D ₂ 0, 293 K, in ppm): δ = 8.55 (d, = 5.5 Hz, 2H, H ^5' ), 8.36 (m, 2H, H ^3' ,H ^6' ), 8.19 (t, J = 8.0 Hz, 2H, H ⁴ ), 8.00 (d, J = 7.5 Hz, 2H, H ³ ), 7.71 (m, 2H, H ^4' ), 7.44 (d, J = 8.5 Hz, 2H, H ⁵ ). ¹³ C NMR (126 MHz, D ₂ 0, 293 K, in ppm): δ = 155.5 (C ⁶ ), 152.0 (C ² ), 148.7 (C ^s ), 145.1 (C ² ), 143.0 (C ³ ), 142.4 (C ⁴ ), 128.4 (C ⁴ ), 124.6 (C ⁶ ), 118.3 (C ³ ), 118.2 (C ⁵ ). High resolution ES MS m/z (calc): 430.0286 (430.0284 = [M - H - 2xCI] ⁺ ). Bern. Anal. Calcd. For C2oCI ₂ H ₁₅ N ₅ Pd + MeOH + H ₂ 0: C, 45.63; H, 3.83; N, 12.67. Found: C, 45.11; H, 3.94; N, 12.86. Example 2: Synthesis of [2Kb (formula (VI)):

(VI)

In a 2-necked round bottom flask foPtCU (300 mg, 0.730 mmol) and Hbbpya (200 mg, 0.590 mmol) were added into a degassed ethanol-water mixture (3:2, 100 mL) preheated at 80 °C. The resulting yellow suspension was stirred overnight at 80 °C in a nitrogen atmosphere resulting in a dark red suspension. The suspension was cooled to room temperature, subsequently cooled on ice, and then filtered. From the filtrate all solvents were rotary evaporated. The solids were dissolved in MeOH (5 mL) and purified by size exclusion chromatography (Sephadex L-20) using MeOH as eluent. The yellow band was collected and MeOH was removed by rotary evaporation, which resulted in [2]CI ₂ as a a red/orange powder. Yield 42% (155 mg), ² H NMR (500 MHz, CD ₃ OD and DO, 293 K, in ppm): δ = 8.99 (d, J = 5.5 Hz, 2H, H ⁶ ), 8.58 (d, J = 7.5 Hz, 2H, H ^3' ), 8.50 (t, J = 7.5 Hz, 2H, H ^4' ), 8.39 (t, ; = 8.0 Hz, 2H, H ⁴ ), 8.23 (d, J = 6.5 Hz, 2H, H ^4' ), 7.99 (m, 2H, H ^5' ), 7.77 (d, J = 8.5 Hz, 2H, H ^s ). ¹³ C NMR (126 MHz, 293 K, CD3OD and DCI): δ = 156.4 (C ⁶ ), 152.4 (C ² ), 150.4 (C ^5' ), 144.4 (C ² ), 143.7 (C ^3' ), 142.4 (C ⁴ ), 129.6 (C'j, 125.6 (C ^5' ), 119.9 (C ³ ), 118.8 (C ⁵ ) ppm. High resolution ES MS (H ₂ 0) m/z (caic) 519.0893 (519.0897 = [M - H - 2xCI] ⁽ ). Elem. Anal. Calcd. For CioClHuNsPt + A ¹ H ₂ 0: C, 41.93; H, 2.82; N, 12.22. Found: C, 41.78; H, 3.12; N, 13.51.

Example 3: Synthesis of f31(PFe)2 (See Figure 4) :

ivm In a 2-necked round-bottom flask K ₂ [PdCI ₄ ] (80 mg, 0.24 mmol), H ₂ bapbpy (200 mg, 0.58 mmol,obtained as disclosed in S. Bonnet, M. A. Siegler, J. Sanchez Costa, G. Molnar, A. Bousseksou, A. L. Spek, P. Gamez and J. eedijk, Chem. Commun., 2008, 5619-5621) and degassed ethanol-water mixture (3 : 2; 30 mL) were added. The brown suspension was stirred for 2 hours at 80 ⁹ C under nitrogen, and an additional K ₂ [PdCU] (80 mg, 0.24 mmol) was added. Again, the brown suspension was stirred for 2 hours at 80 ⁹ C under nitrogen, and an additional K ₂ [PdCI ₄ ] (80 mg, 0.24 mmol) was added. After full conversion of hhbapbpy was confirmed by TLC (eluens = Acetone/H ₂ 0/KN0 ₃ (sat.), (100:10:1)) a precipitate was filtered off and by addition of 1 mL HCI (1.0 M), it was made sure that the solution was acidic. Then, a saturated aqueous solution of KPF6 (5 mL) was added to the filtrate inducing the precipitation of a yellow solid. The suspension was filtered over a membrane filter and the complex [Pd(H ₂ bapbpy)](PF6) ₂ was obtained as a yellow powder and dried in vacuo. Example 4: Synthesis of f31Cl ₂ :

To obtain the more water-soluble chloride salt [Pd(H ₂ bapbpy)](PF6) ₂ (50 mg) was dissolved in acetonitrile (250 mL) and heated a clear solution was obtained. Then, concentrated HCI (37% w/v, 1 mL) was added upon which a precipitate appeared. The suspension was filtered over a membrane filter and the complex [Pd(H ₂ bapbpy)]CI ₂ was obtained as a yellow powder and dried in vacuo.

Example 5: Synthesis of [4l(PFe) ₂ :

Mil

In a 2-necked round-bottom flask K ₂ [PtCI ₄ ] (101 mg, 0.24 mmol), H ₂ bapbpy (200 mg, 0.58 mmol) and degassed ethanol-water mixture (7 : 3; 30 mL) were added. The brown suspension was stirred for 2 hours at 80 ⁹ C under nitrogen, and an additional K ₂ [PtCU] (101 mg, 0.24 mmol) was added. Again, the brown suspension was stirred for 2 hours at 80 ⁹ C under nitrogen, and an additional K ₂ [PtCI ₄ ] (101 mg, 0.24 mmol) was added. After full conversion of H ₂ bapbpy was confirmed by TLC (eluens = Acetone/H ₂ 0/KN03 (sat.),

(100:10:1)) a precipitate was filtered off and an aliquot of HCI (1.0 M, 1 mL) was added to make the solution acidic. Then, a saturated aqueous solution of KPF6 (5 mL) was added to the filtrate inducing the precipitation of a yellow solid. The suspension was filtered over a membrane filter and the complex was obtained as a yellow powder and dried in vacuo. The product obtained here is [Pt(H ₂ bapbpy)](PF6) ₂

Example 6: Synthesis of [4l(CI)2: To obtain the chloride salt [Pt(H ₂ bapbpy)](PF6) ₂ was dissolved in acetonitrile and heated to ensure a better and full solubility - typically 50 mg of complex in 250 mL of acetonitrile. Then, to keep concentrations of water low, 1 mL of HCI (37% w/v) was added upon which the product precipitated. The suspension was filtered over a membrane filter and the complex [Pt(H ₂ bapbpy)]CI ₂ was obtained as a yellow powder and dried in vacuo.

Example 7: pK _a determination

pH titration: 6 mL of a 67 μΜ solution of [1]CI ₂ or [2]CI ₂ in hydrochloric acid (0.033 M) was put in a 15 mL vial. A pH measurement electrode was added and when stable the pH was logged, aliquots of aqueous NaOH (0.05 - 5M) were added to give a range of pH values while stirring. After each stable pH a UV-vis absorbance spectrum was recorded. The relative concentration of [M(Hbbpya] ^{2 t} was calculated using the Lambert-Beer law and then plotted vs. pH. The pKa was determined by modelling the curve using a simplified two-parameter Hill- slope equation: _{(1+lo((fogloP} - ). _Wi « 5 _ioPC)) - ^Lo g ^{p value} determination: 1.00 mM stock solutions of [l]C and [2]CI ₂ in octanol-saturated water were prepared. 0.2 mL aliquots were transferred to 15 mL corning centrifugation tubes and diluted with octanol-saturated water to 1.0 mL, to give 0.2 mM solutions (in threefold). Then, 1.0 mL water-saturated octanol was added to each solution (not for the control samples, which remained at a volume of 1 mL). The solutions were shaken for 60 minutes on a GFL 3016 shaker at maximum speed and centrifuged for 10 minutes at RCF = 2800 g at T = 293 K. Then, 100 μΐ of the aqueous phase (in threefold) was transferred into a vial, and 1.0 ml 65% nitric acid was added to degrade the compounds overnight. 0.8 mL of the resulting solution was then diluted with 11 mL MilliQ water in corning tubes. Then the metal concentration of each sample was measured using ICP-JOS, and the initial equilibrium metal concentration in the water and octanol phases was calculated from the dilution factors. Finally, the logP value was calculated using the following equation: lo _oct/ai? = log

aq

Example 8: Cell culturing and EC50 cytotoxicity assay

Cellular uptake of [1]CI ₂ , [2]CI ₂ , [3](PF ₆ ) ₂ , and [4](PF ₆ ) ₂

Uptake studies for complexes [1]CI ₂ , [2]CI ₂ , [Pd(H ₂ bapbpy)](PF ₆ ) ₂ ([3](PF ₆ ) ₂ ) and[Pt(H ₂ bapbpy)](PF ₅ ) ₂ ([4](PF ₆ ) ₂ ) were conducted on A375 and A549 lung cancer cells. 10 x 10 ³ A375 cells and 17 x 10 ³ A549 cells were seeded at t = 0 h in Opti-MEM complete in each well of a 6-well plate. At t = 48 h cells were treated with complexes to give a final concentration close to the EC50 values (1.9, 1.7, 0.05 or 0.2, and 1.0 μΜ for respectively [1]CI ₂ , [2]CI ₂ , [3](PF ₆ ) ₂ , and [4](PF ₆ ) ₂ ) in the dark after 24 h in a total volume of 4 mL. After 24 h of drug incubation at 37 °C, the medium was aspirated, the cells were washed with PBS-buffer, trypsinized, counted using BioRad Cell Counting Slides on a BioRad TC10 automated cell counter, and pelleted by centrifugation (700 ^χ g, 5 min). The supernatant was removed, and each sample was digested overnight in concentrated nitric acid at room temperature (> 65%). MilliQ water was added to each sample to obtain a final concentration of 5% H NO3. For ICP- MS measurements, the system was optimized with a palladium-platinum solution. The calibration range was from 0 to 25 μg/], and obtained detection limit for all isotopes was 0.01 μg/l. Silver and indium were used for internal standard, to correct for sample-dependent matrix effects. No reference sample was available; therefore several samples were spiked with a known concentration. The recoveries of the spiked concentrations were all within a 10% deviation.

Example 9: Synthesis and crystal structure

In a preferred embodiment of the subject invention, the square-planar complexes

[Pd(Hbbpya)]CI ₂ ([1]CI ₂ ) and [Pt(Hbbpya)]CI ₂ ((2]CI ₂ ) were synthesized by reacting ligand Hbbpya with K ₂ [PdCl4] and K ₂ [PtCI ₄ ], respectively, in an ethanol-water mixture (Figure 3).

After purification using size exclusion chromatography, analysis with NMR spectroscopy, elemental analysis, and high resolution mass spectrometry confirmed that the compounds were analytically pure. Vapour diffusion of acetone into a solution of [1]CI ₂ in methanol yielded yellow-colored crystals suitable for X-ray diffraction studies (Figure 3, schematic synthesis of [1]CI ₂ and [2]CI ₂ , and Figure 6, X-ray crystal structure of [1]CI ₂ )

The compound [1]CI ₂ crystallizes with one molecule of MeOH. Pdl-Nl and Pdl-N5 bond distances of the terminal pyridyl groups are 2.0600(18) and 2.0680(18) A, respectively whereas the Pdl-N2 and Pdl-N4 bond distances of the internal pyridines are 1.9887(10) and 1.9933(18) A, respectively, thus much shorter. These trends are similar to those in

[Pd(H ₂ bapbpy)](PF6) ₂ ([3](PF6) ₂ ), but are even more pronounced in the analogue

palladium(ll) complex without any amine bridge, i.e. [Pd(qtpy)](PFe) ₂ .

Table 4: Selected bond lengths (A) in the crystal structures of [1]CI ₂ , of the H ₂ bapbpy analogue [3](PF ₆ ) ₂ , and of [Pd(qtpy)](PF ₆ ) ₂ .

Bond [1]CI ₂ Bond [3](PF ₆ ) ₂ ^a Bond [Pd(qtpy)](PF ₆ ) ₂ ^b

Pdl-Nl 2.0600(18) Pdl-Nl 2.0251(16) Pdl-N2 2.056

Pdl-N2 1.9887(18) Pdl-N3 1.9846(16) Pdl-Nl 1.928

Pdl-N4 1.9933(18) Pdl-N4 1.9802(16) Pdl-NIB 1.928

Pdl-N5 2.0680(18) Pdl-N6 2.0312(16) Pdl-N2B 2.056

Pdl-Nave ^C 2.028 Pdl-Nave 2.005 Pdl-Nave 1.992 b: Values taken from Constable et al. ¹²

c: Average value of the four coordinating Pd-N bonds in each complex. The differences between the open and closed side of the tetrapyridyl ligands are also apparent when comparing the bond angles. For [l]C the Nl-Pdl-N5 angle on the open side is 108.84(7)°, while the angle N2-Pdl-N4 on the closed side is 93.22(7)°, thus more than 15° smaller. In [3](PF ₆ )2 and [Pd(qtpy)](PF ₅ )2 bond internal angles N3-Pdl-N4 and Nl-Pdl-N5 are 82.19(7) and 81.2°, respectively. These differences can be explained by the formation of one 6-membered ring and two 5-membered rings with Hbbpya, while coordination of h bapbpy generates one 5-membered ring and two 6-membered rings, and qtpy generates only 5- membered rings. Such differences increase the opening of the ligands as follows: H ₂ bapbpy < Hbbpya < qtpy: in [3](PF ₆ )2 N6-Pdl-Nl is 96.38(7)°, in [1]CI ₂ Nl-Pdl-N5 is 105.84(7)°, and in [Pd(qtpy)](PF ₆ ) ₂ N2-Pdl-N2B is 117.8°.

Table 5: Selected bond angles, torsion angles (°), and x4 values for [1]CI2, [3](PF6)2, and [Pd(qtpy)](PF6)2.

Angle [i]ci ₂ Angle [3](PF ₅ )2 ^a Angle [Pd(qtpy)](PF ₆ )2 ^b

Nl-Pdl-N2 80.38(7) Nl-Pdl-N3 91.43(7) N2-Pdl-Nl 80.5

N2-Pdl-N4 93.22(7) N3-Pdl-N4 82.19(7) Nl-Pdl-N1B 81.2

N4-Pdl-N5 80.48(7) N4-Pdl-N6 92.34(7) N1B-Pdl-N2B 80.8

N5-Pdl-Nl 105.84(7) N6-Pdl-Nl 96.38(7) N2B-Pdl-N2 117.8

Nl-Pdl-N4 173.35(7) Nl-Pdl-N4 165.35(6) N2-Pdl-N1B 161.7

N2-Pdl-N5 173.37(7) N3-Pdl-N6 167.38(6) Nl-Pdl-N2B 161.7

N1-N2-N4-N5 0.20 N1-N3-N4-N6 17.30 N2-N1-N1B-N2B 0.54 τ ₄ * 0.090 τ ₄ * 0.193 X4* 0.260

* τ ₄ is obtained using the equation ³⁶⁰ _i4 ^" ⁺ a value of 0 corresponds to a square- planar coordination geometry and a value of 1 to a tetrahedral coordination geometry. a Values taken from Chapter 6

b Values taken from Constable et al. ¹²

As a result of the different chelating rings generated by the three ligands H2bapbpy, Hbbpya, and qtpy, the distortions of the square-planar coordination sphere of the metal also greatly vary. A perfect square-planar coordination geometry can be defined as a four- coordinate complex with ligands that coordinate in a single plane, and it should contain bond angles of only 90 and 180°. The x ₄ value introduced by Houser and co-workers quantifies the square-planarity of tetracoordinated metal complexes by distinguishing complexes with a τ ₄ value of 0 that are perfectly square-planar, and complexes with a t ₄ value of 1 that have a perfect tetrahedral geometry.

Using the crystal structures of [1]CI ₂ , [3](PF6>2, and [Pd(qtpy)](PF6)2 τ ₄ values of respectively 0.090, 0.193, and 0.260 (Table 5) were calculated indicating that [1]CI ₂ is by far the most square-planar geometry, while [3](PF6) ₂ and [Pd(qtpy)](PF6) ₂ have both severely distorted geometries. The torsion angle of the coordinating nitrogen atoms is a measure of the degree of flatness of the coordination plane of square-planar complexes. For [1]CI ₂ , [3](PF ₅ ) ₂ , and [Pd(qtpy)](PF ₆ )2 these torsion angles are 0.20, 17.30, and 0.54°, respectively. Thus, whereas in [1] ²⁺ and [Pd(qtpy)] ²⁺ the tetrapyridyl ligands are coordinated almost in one single plane, in [3] ²⁺ the ligand assumes a more helical geometry. To conclude, [1]CI ₂ is highly square-planar and very flat. The distance between the hydrogen atoms of CI and C20 is 1.905 A, which is apparently long enough to prevent distortion of the coordinated Hbbpya ligand.

Previously, Hbbpya has already demonstrated a high degree of planarity in octahedral iron(M) complexes, and preliminary data shows that it is also highly planar in nickel(ll) and ruthenium(ll) complexes.

This planar geometry significantly differs from the H ₂ bapbpy ligand that coordinates to metal ions in a distorted fashion. Overall, the removal of an amine bridge in H ₂ bapbpy, results in the ligand Hbbpya that can coordinate to palladium(ll) or platinum(ll) with a highly square-planar geometry.

Compounds [1]CI ₂ and [2]CI ₂ were found to have mildly to strong cytotoxic properties. Without wishing to be bound to any particular theory, it was originally believed that the similar coordination mode of Hbbpya and H ₂ bapbpy to palladium(ll) and platinum(ll) ions may permit to predict similar activities for these metal complexes in vitro.

To check this hypothesis, also the cytotoxicity of the several Hbbpya and H ₂ bapbpy metal complexes were tested in vitro against three different cancer cell lines (A375 skin,

A549 lung, and MDA-MB231 breast cancer) and one non-malignant lung cell line (MRC-5).

The cell-testing protocol was adapted from S. L. Hopkins, B. Siewert, S. H. C. Askes, P.

Veldhuizen, R. Zwier, M. Heger and S. Bonnet, Photochem. Photobiol. Sci., 2016, 15, 644-

653. 24 h after seeding, the cells were incubated for 24 h with the drug, the media was refreshed, and the cells were further incubated with drug-free medium for 48 h. A sulfo- rhodamine B (SRB) cell-counting assay was performed at 96 h, and effective concentrations (EC50) were determined by comparing cell viability of treated wells with drug-free control wells. The EC50 values are summarized in Table 6.

ECso values for A375 and A549 cancer cell lines for [1]CI ₂ and [2]CI ₂ were similar and ranged between 1.6-1.9 μΜ. For MDA-MB231 and MRC-5 cancer cells both compounds showed lower cytotoxicity in the 4.7-6.8 μΜ range. Overall, the cytotoxicity of [1]CI ₂ and [2]CI ₂ was found comparable with that of cisplatin, i.e. in the low micromolar range. The cytotoxicity of [2]CI ₂ is comparable with that of [Pt(H ₂ bapbpy)](PF ₆ ) ₂ ([4](PF ₆ h, but the cytotoxicity of [1]CI ₂ was nowhere near the exceptional activity of [3](PF6> ₂ , that showed EC50 values below 0.2 μΜ for all the cell lines tested in this panel. Although the ligands are highly similar, once coordinated to palladium they have highly different biological properties as their palladium(ll) complexes.

Table 6: Cytotoxicity expressed as cell growing inhibition effective concentrations

(EC50 with 95% confidence intervals, in μΜ) for [1]CI2, [2]CI2, and cisplatin on skin (A375), lung (A549), and breast (MDA-MB231) cancer cell lines, and on noncancerous lung fetal cell line (MRC-5) Cell Line.

[1]CI ₂ [2]CI ₂ cisplatin

-0.26 -0.39 -0.52

A375

+0.30 +0.50 +0.77

-0.34 -0.47 -0.98

A549 3.1

+0.42 +0.67 +1.4

-1.43

MDA-MB231 6.6 4.7 ^1,2 > 25

+1.8 +1.5

-1.5 _ ₂ o -3.9

MRC-5 6.8 5.6 ^,u 12

+1.9 ₊₃ _ ₀ +5.9 The cytotoxicity of the two complexes [3](PF6) ₂ and [4](PFe)2 was tested against five different cancer cell lines (A375, A431, A549, MCF-7, DA- B231) and one non-malignant lung cell line (MRC-5). The cell testing protocol was adapted from Hopkins et al. as set out above. The cytotoxicity results are summarized in Table 7.

Table 7: Cytotoxicity of [3](PF6>2 and [4](PF6>2 expressed as cell growing inhibition effective concentrations (ECso with 95% confidence intervals, in μΜ)

Ceil line l3](P h [4]{PF ₆ ) ₂ cisplatin

-0.006 -0.140 -0.52

A375 0.058 0.95 1.60

+0.005 +0.122 +0.77

■0.014

A431 ^■ 0.142

0.087 1.15 3.92

+0.012 +0.126

A549 -0.020 -0.123 -0.9S

0.188 0.91 3.14

+0.018 +0.108 + 1.42

MCF-7 -0.027 -0.143 -1.49

0.095 0.80 4.02

+0.021 +0.121 +2.35

MDA-MB231 •0.033 • ^■ 0.752

0.123 1.33 >25

0.026 +0.480

MRC-5 -0.012 -0.633 -3.91

0.105 1.38 11.6

+0.011 +0.434 +5.89

For [4](PF6)2 EC50 values between 0.80 μΜ for MCF-7 cells and 1.4 μΜ for MRC-5 cells were found, which makes the platinum complex [4](PF6)2 slightly more cytotoxic than cisplatin. For [3](PF6)2 the EC50 values ranged between 0.058 μΜ for A375 cells and 0.19 μΜ for A549 cells, which is remarkably low. This makes the compound [3](PF6)2 one of the most cytotoxic palladium compounds known to date, as it is 520 times more toxic than its platinum analogue [4](PF6)2 or cisplatin over the full panel of tested cell lines in spite of the almost identical structural features.

The cytotoxicity of [Au(bapbpy)]CI was tested against four different cancer cell lines

(A549, A431, A375, and MCF-7) and one non-malignant lung cell line (MRC-5). The cell testing protocol was adapted from Hopkins et al. as set out above. The cytotoxicity results are summarized in Table 7A. Table 7A: Cytotoxicity of [Au(bapbpy)]CI expressed as cell growing inhibition effective concentrations ( EC50 with 95% confidence intervals, in μΜ)

Cell line (μΜ) ±CI

+3

A549 23

-3

+4.5

A431 9.5

-3.1

+0.5

A375 5.0

-0.1

+0.2

MCF7 2.1

-0.2

+4

MRC-5 18

-3

Example 10: Cellular Uptake

Considering the different in vitro activity of [1]CI ₂ and [4](PFe) ₂ uptake studies were performed to determine whether the differences in cytotoxicity are caused by different intracellular concentrations of both palladium compounds. A375 or A549 cells were seeded, and treated with [1]CI ₂ , [2]CI ₂ , [3](PF6) ₂ , [4](PFe) ₂ , or cisplatin at their EC50 concentrations. After 24 h drug incubation, cells were washed, trypsinized, counted, digested and the amount of metal content was measured using ICP-MS (see Experimental section for the full procedure). [1]CI ₂ and [2]CI ₂ showed similar uptake by A375 or A549 cells (7.2 and 9.4 nmol/10 ⁶ cells) with an uptake efficiency of 8 and 16%. [3](PFe) ₂ showed an uptake of 1.2 (54%) and 5.3 nmol Pd/10 ⁵ cells (39%) for A375 and A549 cells, respectively. For [4](PF _s ) ₂ a similar trend was observed with 19 and 22 nmol Pt/10 ⁶ cells for A375 and A549 cells, respectively, corresponding to an uptake efficiency of 58% in both cell lines. Cisplatin, by comparison, showed very low metal uptake (0.06 and 0.07 nmol Pt/10 ⁶ cells for A375 and A549 cells, respectively), which demonstrates that once in the cell cisplatin is much more harmful than the tetrapyridyl complexes. The lower cytotoxicity of the Hbbpya complexes compared to the H ₂ bapbpy complexes does not seem to be related to a lower metal uptake: treatment at the ECsothe uptake values were overall similar. Compound [4](PFe) ₂ may owe its higher cytotoxicity compared to [1]CI ₂ and [2]CI ₂ due to a higher uptake, but [3](PF6) ₂ clearly has a much lower metal uptake and much higher cytotoxicity. These uptake results emphasize the remarkable cytotoxicity of [3](PFe)2 as its cellular concentrations in this experiment are the lowest by far, while it exhibited the highest cytotoxicity.

Table 8: Cellular uptake for A375 and A549 for compounds [1]CI2, [2]CI2, [3](PF ₆ ) ₂ , [ ](PF6) ₂ , and cisplatin.

Cell Compound Treatment Concentration Metal Uptake Rel. Uptake Efficiency line (μΜ) (nmol/10 ⁶

cells)

[1]CI ₂ L9 9.4 ± 2.8 16

[2]CI ₂ 1.7 7.9 1 1.8 12

A375 (3](PF ₆ ) ₂ 0.05 1.2 1 0.2 54

[4](PF ₆ ) ₂ 1.0 19 + 4.4 58 cisplatin 1.6 < 0.01 < 1

Example 11: Lipophilicity and pKa

In medicinal chemistry higher uptake and improved cytotoxicity are often associated with higher lipophilicitiy. The octanol-water partition coefficient (logP value) is widely used as an indicator of lipophilicity (positive logP value) or hydrophilicity (negative logP value). The PFe salts of the H ₂ bapbpy complexes were first converted to [3]CI ₂ and [4]CI ₂ , respectively, as DMSO would be required for initial solubilization of the PF6 salts.

The logP values for complexes [1]CI ₂ to [4]CI ₂ are shown in

Table . The logP value for [4]CI ₂ (-0.45) was the highest of all the complexes, and also corresponded with the highest cellular uptake of [4] (PF6) ₂ . [3]CI ₂ with a logP value of -1.63 is the most hydrophilic of the compounds, and as the PF6 ^" salt was taken up second highest. The logP values of [l]C and [2]CI ₂ were found to be -0.96 and -1.59, respectively. In conclusion, all four tested complexes are rather hydrophilic, and no obvious relationship between metal uptake and logP could be observed for this series of complexes.

Each amine bridge in the tetrapyridyl ligands H ₂ bapbpy and Hbbpya bears an acidic proton. Consecutive deprotonation of the two amine bridges in H ₂ bapbpy can result in dicationic, monocationic, or neutral complexes, whereas the single amine bridge in Hbbpya complexes can only lead to dicationic or monocationic species. Of course, the acid-base equilibrium may influence both the logP values and the uptake of these complexes in vitro. The pKa for the first deprotonation of [3] ²⁺ and [4] ²⁺ is 7.8 and 8.3 in water, respectively.

For [1] ²⁺ and [2] ²⁺ the pKa in water was also determined by monitoring with UV-vis spectroscopy the titration of [1]CI ₂ and [2]CI ₂ in HCI (33 mM) with NaOH. Under acidic conditions the UV-vis spectra of [1] ²⁺ and [2] ²⁺ only show absorption in the visible region with a tail up to 423 nm and 425 nm, respectively (green curves in Figure 6). Upon base addition a significant increase in absorbance was observed, characterized by a new absorption maximum at 416 nm for [1] ²⁺ and 430 nm for [2] ²⁺ . Simultaneously, a major decrease in absorbance was found at 261 and 362 nm for [1] ²⁺ and at 239 and 372 nm for [2] ²⁺ . The evolution of the absorbance spectra for both complexes occurred with isosbestic points at 316 and 397 nm for [1] ²⁺ , and at 312 and 405 nm for [2] ²⁺ , indicating that only one reaction takes place. Further basification did not lead to any further change in the absorbance. The pH-dependence of the absorbance spectra of [1] ²⁺ and [2] ²⁺ in water can be explained by the deprotonation of [1] ²⁺ or [2] ²⁺ to form [Pd(bbpya)] ⁺ or [Pt(bbpya)] ⁺ , respectively. The corresponding pKa was determined to be 5.5 and 4.6 for [1] ²⁺ and [2] ²⁺ , respectively. Thus, [1] ²⁺ and [2] ²⁺ have much lower pKa's compared to [3] ²⁺ and [4] ²⁺ . Such higher acidity can be explained by a difference in steric repulsion: Hbbpya complexes have an almost perfect square-planar geometry (τ = 0.090), whereas complexes of H ₂ bapbpy are much more distorted (τ = 0.193, Figure 6). As a result the negative charge located on the deprotonated bbpya ^" ligand in [Pd(bbpya)] ⁺ and [Pt(bbpya)] ⁺ is much better delocalized than the negative charge located on Hbapbpy ^"" in [Pd(Hbapbpy)] ⁺ and [Pt(Hbapbpy)] ⁺ .

Another reason for the lower pKa of [1] ²⁺ and [2] ²⁺ compared to [3] ²⁺ and [4] ²⁺ may be that the negative charge of the [Pd(bbpya)] ⁺ and [Pt(bbpya)] ⁺ complexes may be delocalized over two bpy moieties, whereas that of the [Pd(Hbapbpy)] ⁺ and [Pt(Hbapbpy)] ⁺ complexes may be delocalized over only one bpy moiety and two pyridyl fragments.

Overall, at physiological pH (7.4) the Hbbpya complexes will be fully deprotonated, thus monocationic, whereas the H ₂ bapbpy complexes will be mixtures of dicationic and monocationic species (see Table 9).

Table 9: LogP and pKa values for [1]CI2, [2]CI2, [3]CI2, and [4]CI2.

Compound logP pKa

[1]CI ₂ -0.96 5.5

[2]CI ₂ -1.59 4.6

[3]CI ₂ -1.63 7.8

[4]CI ₂ -0.45 8.3

Although the tetrapyridyl ligands Hbbpya and H2bapbpy structurally are highly similar, the coordination to palladium(ll) or platinum(ll) centers results in complexes with significant differences in coordination geometry, acidity, cellular uptake, and cytotoxicity.

The crystal structures of the palladium(ll) complexes with qtpy, Hbbpya, or H ₂ bapbpy show that the palladium Hbbpya complex is the most square-planar as well as the flattest of the three complexes. In the H ₂ bapbpy complexes electronic conjugation in the distorted ligand is less possible, making deprotonation less favorable compared to Hbbpya complexes.

As a result [1] ²⁺ and [2] ²⁺ will not be present as dicationic species in vitro whereas [3] ²⁺ and [4] ²⁺ are present in both dicationic and monocationic forms.

Without wishing to be bound to any particular theory, it is hypothesized that the complexes described in this chapter are taken up via passive diffusion, both the charge state and the lipophilicity may play a key role.

Specifically in case of the H2bapbpy complexes, their pKa values close to

physiological conditions makes the complexes flexible regarding charge-dependent biochemical interactions with for instance negatively charged DNA. While observed differences in cytotoxicity may advantageously be explained by differences in uptake for [l]Cl2, [2]CI ₂ , and [4](PF6) ₂ , provided that these complexes have similar mechanism of actions, it was however found that the cytotoxicity and uptake of [3](PF6> ₂ demonstrates that it is orders of magnitude more cytotoxic, and thus efficient in killing the cell once taken up, and that it may have a distinct mechanism of actions compared to the other complexes.

[1]CI ₂ and [2]Cb, albeit less cytotoxic than [3](PFe) ₂ , also show that the class of non-cyclic tetrapyridyl complexes is promising as anticancer agents when the ligand is functionalized with a single non-coordinating amine bridge. These complexes also advantageously illustrate that non-coordinated chloride counter-anions, combined with the amine bridges, may result in much more water-soluble complexes, as compared to cisplatin.

Example 12: In vivo experiments

The in vivo anticancer activity of [3](PF6) ₂ and [4] (PF6) ₂ was evaluated in CT-26 colon cancer-bearing mice as disclosed in J. Mayr, P. Heffeter, D. Groza, L. Galvez, G.

Koellensperger, A. Roller, B. Alte, M. Haider, W. Berger, C. R. Kowol and B. K. Keppler, Chem. Sci., 2017, 8, 2241-2250.

Treatment existed of equimolar dosages [3](PF6) ₂ (2.5 mg/kg) or

[4](PF6) ₂ (2.8 mg/kg), which was found to be the maximum-tolerated dose for

[3](PF6) ₂ . The complexes were administered by intraperitoneal injections for 5 times (on day 3, 5, 7, 10, and 12) and mice were sacrificed 24 h after the last application.

The impact of the metal complexes on tumour growth was determined by measuring the tumour volume (mm ³ ) and the results are shown in Figure 6.4. Compared to untreated control BALB/c mice, [4] (PF6) ₂ treatment had no effect on tumour growth, indicating that in this treatment regime [4](PF6) ₂ is inactive at the doses used. In contrast, [3](PFe) ₂ induced from the first day a significant delay in tumour growth and on day 12 a 37% reduction compared to the control group. Thus, as previously witnessed in in vitro experiments, treatment with [3] (PFe) ₂ or [4](PF6) ₂ leads to different activity, despite structural similarities, also under in vivo conditions. H ₂ bapbpy complexes [3) ₂ , and [4](PFe) ₂ demonstrate very diverse cytotoxicity patterns. Exposure of various cancer cell lines to the drugs results in EC50 values for [3] (PF6) ₂ in the nanomolar range, and for [4](PF6) ₂ in the submicromolar to low micromolar range. Example 13: synthesis of fAu(bapbpy)lPF6

hhbapbpy (34 mg, 0.10 mmol) was first dissolved in hot ethanol (30 mL) and filtered to make a clear solution of H ₂ bapbpy. In a 2-necked round-bottom flask NaAuCU (61 mg, 0.15 mmol), excess KPF6 (2 mmol, 368 mg) and the ethanol solution of H ₂ bapbpy were added. The initially yellow solution was stirred for 4 hours at 60 ⁹ C under nitrogen, upon which it turned into a red suspension. The suspension was filtered over a membrane filter and the solid washed by water and ether, then dried in vacuum. The complex

[Au(bapbpy)](PF6) was obtained as a red powder in 38% yield (26 mg).

Example 14: synthesis of [Au(bbpya)]X ₂ (X ^~ =CI ^~ or AuCI ₂ ^~ )

Hbbpya (100 mg, 0.31 mmol) and NaAuCU ^χ 2H ₂ 0 (122 mg, 0.31 mmol) were dissolved in ethanol (30 mL) and heated to 60 °C under N ₂ . After 24 h, the reaction was cooled down to room temperature and the solid was collected by filtration over a micropore filter, washed with ether and dried. The product was then thoroughly washed with H ₂ 0 (~500 mL) until the intense yellow colour of the filtrate slowly faded to a colourless solution. The filtrate was concentrated under reduced pressure to obtain a dark green solid (24 mg) and the solid was washed with diethyl ether and dried in vacuum overnight obtaining a dark brown solid (92 mg).

Example 15: synthesis of [Au(biqbpyl)]CI

H ₂ biqbpyl (200 mg, 0.45 mmol) and NaAuCU x 2H ₂ 0 (181 mg, 0.45 mmol) were dissolved in methanol (80 mL) and heated at 72 ^C in an oil bath under N ₂ for 70 h. The reaction mixture was filtered over a micropore filter and the brown solid was washed with methanol and dried in air. H ₂ 0 (50 mL) was then added to the solid and the suspension was stirred at room temperature for 30 min. The solid was collected by filtration, washed diethyl ether and dried. Purification was done using two separate column chromatography on silica using the same eluent mixture of DCM:acetone as eluents starting with pure DCM and increasing the acetone ratio to 20:1 and 20:5. After purifying 69 mg of impure compound by chromatographic column, 44 mg of pure compound were obtained (63% yield).

Example 16: synthesis of [Au(biqbpy2)]CI

H ₂ biqbpy2 (150 mg, 0.34 mmol) and NaAuCU x 2H ₂ 0 (135 mg, 0.34 mmol) were dissolved in methanol (80 mL) at 75 °C. An excess of ΚΡΓ-6 was added after 1 h and the reaction mixture was stirred at 75 °C for 24 h. The solvents of the previously cooled down suspension were evaporated under reduced pressure obtaining an orange solid which was subsequently vigorously stirred with H ₂ 0 (30 mL) for 30 min. The emulsion was then filtered and the orange solid washed with ether and dried. Purification by column chromatography on silica was then carried out. Dichloromethane and methanol were used as eluents at an increasing ratio of 1:0, 10:0.5 and 10:1. The compound was obtained pure as an orange solid (80 mg, 44% yield). Example 17: synthesis of [Au(biqbpy3)]CI

H ₂ biqbpy3 (165 mg, 0.37 mmol) and NaAuCI ₄ x 2H ₂ 0 (151 mg, 0.38 mmol) were dissolved in methanol (75 mL) and heated to 72 under N ₂ in an oil bath. After 72 h, the precipitate was collected over a micropore filter, washed with few methanol and diethyl ether and dried. H ₂ 0 (50 mL) was then added to the solid and the suspension was vigorously stirred at room temperature for 30 min. After filtration, the solid was dried in vacuum overnight obtaining the pure product as a dark brown solid (90 mg, 32% yield). No further purification steps were needed.