Synthesis and Cytotoxicity Studies of Novel NHC*-Gold(I) Complexes Derived from Lepidiline A

Ten novel N-heterocyclic carbene gold(I) complexes derived from lepidiline A (1,3-dibenzyl-4,5-dimethylimidazolium chloride) are reported here with full characterisation and biological testing. (1,3-Dibenzyl-4,5-diphenylimidazol-2-ylidene)gold(I) chloride (NHC*-AuCl) (1) was modified by substituting the chloride for the following: cyanide (2), dithiocarbamates (3–5), p-mercaptobenzoate derivatives (12–14) and N-acetyl-l-cysteine derivatives (15–17). All complexes were synthesised in good yields of 57–78%. Complexes 2, 12, 13, and 14 were further characterised by X-ray crystallography. Initial evaluation of the biological activity was conducted on all ten complexes against the multidrug resistant MCF-7topo breast cancer, HCT-116wt, and p53 knockout mutant HCT-116−/− colon carcinoma cell lines. Across the three cell lines tested, mainly single-digit micromolar IC50 values were observed. Nanomolar activity was exhibited on the MCF-7topo cell line with 3 displaying an IC50 of 0.28 μM ± 0.03 μM. Complexes incorporating a Au–S bond resulted in higher cytotoxic activity when compared to complexes 1 and 2. Theoretical calculations, carried out at the MN15/6–311++G(2df,p) computational level, show that NHC* is the more favourable ligand for Au(I)-Cl when compared to PPh3.


Introduction
Metal-based drugs are an important tool in the development of new therapeutic drugs.Auranofin, the successful gold(I)-based drug, exhibits both high potency antiarthritic and antitumour properties [1,2].Auranofin analogues have since been investigated for their interesting coordination to both a phosphine and a thioglucoside.In many cases, N-heterocyclic carbenes (NHCs) have been utilized as an alternative to the phosphine ligand [3][4][5].NHCs have proved to be suitable ligands for stabilizing the highly active gold(I) species, due to their good electron donating ability and their highly stable carbene from π-backbonding [6,7].As a result, several metal NHC complexes have reported strong anticancer activity [8][9][10].
Lepidiline A (Figure 1), a naturally occurring imidazolium compound extracted from the root of Lepidium meyenii, has presented many biological properties, including cytotoxicity [11].Lepidiline A exhibits activity against the human ovarian cancer cell line FDIGROV, with an ED 50 of 7.39 µg/mL [11].Furthermore, this biologically active imidazolium compound acts as a promising structural motif for NHC derivatives [12], and more effective applications of lepidiline A may lie in the development of metal-based complexes with lepidiline A as the coordinating ligand.
Gold(I) complexes are an important class of anticancer drugs, due to their unique mechanism of action.It has been shown that gold(I) complexes can elicit tumour cell death through targeting members of the intracellular redox-homeostasis system, such as the mitochondria associated thioredoxin reductase (TrxR), whose inhibition leads to reactive oxygen species formation [13][14][15].A selenocysteine-cysteine bridge at the C-terminal of the TrxR enzyme acts as the target for gold(I) [9,16].Gold(I) has a high affinity for thiols, due to their soft nature, resulting in strong Au-S bonds.However, gold(I) also binds strongly to blood thiols such as serum albumin or glutathione, reducing the amount of drug arriving at cancer cells [17].Therefore, there is a desire to design a gold(I)-NHC complex that has a suitably strong Au-S bond incorporated to lessen the chance of blood thiol conjugation.
The effectiveness of these gold(I)-NHC complexes are still restricted by cell selectivity.Introducing targeting biomolecules to the complex could ensure the drug is delivered directly to the cancer cells, thus minimizing the death of normal cells and increasing the drug's efficacy [18].Modifying the coordinating ligand of the NHC-gold(I) complex to include a carboxylic acid would allow increased functionality, such as esters or amides.
motif for NHC derivatives [12], and more effective applications of lepidiline A may lie in the development of metal-based complexes with lepidiline A as the coordinating ligand.
Gold(I) complexes are an important class of anticancer drugs, due to their unique mechanism of action.It has been shown that gold(I) complexes can elicit tumour cell death through targeting members of the intracellular redox-homeostasis system, such as the mitochondria associated thioredoxin reductase (TrxR), whose inhibition leads to reactive oxygen species formation [13][14][15].A selenocysteine-cysteine bridge at the C-terminal of the TrxR enzyme acts as the target for gold(I) [9,16].Gold(I) has a high affinity for thiols, due to their soft nature, resulting in strong Au-S bonds.However, gold(I) also binds strongly to blood thiols such as serum albumin or glutathione, reducing the amount of drug arriving at cancer cells [17].Therefore, there is a desire to design a gold(I)-NHC complex that has a suitably strong Au-S bond incorporated to lessen the chance of blood thiol conjugation.
The effectiveness of these gold(I)-NHC complexes are still restricted by cell selectivity.Introducing targeting biomolecules to the complex could ensure the drug is delivered directly to the cancer cells, thus minimizing the death of normal cells and increasing the drug's efficacy [18].Modifying the coordinating ligand of the NHC-gold(I) complex to include a carboxylic acid would allow increased functionality, such as esters or amides.Herein we present a structural assessment of NHC-Au(I) complexes, based on (1,3-dibenzyl-4,5diphenylimidazol-2-ylidene)gold(I) chloride (NHC*-AuCl), (1) Figure 1.The synthesis, characterisation, and biological testing of ten new NHC*-gold(I) complexes is reported.The effect of altering the coordinating ligands of the NHC*-gold(I) on the cytotoxicity is investigated via MTTbased proliferation assays.The cytotoxicity studies of these novel compounds have been conducted in vitro against three different tumour cell lines: MCF-7 topo (multidrug-resistant breast cancer), HCT-116 wt , and the p53 knockout mutant HCT-116 −/− (colon cancer).These cytotoxicity studies, compared to that of 1, can provide information on the ideal structures of future gold(I) chemotherapeutic complexes.Additionally, a computational study of 1 can highlight the advantages of employing an NHC ligand, as opposed to a phosphine.

Synthesis and Characterisation
The synthetic route for the ten NHC*-gold(I) complexes described in this paper are shown in Schemes 1-4.NHC*-Au(I)-Cl (1) was synthesised according to a procedure previously published [ 3 ] .The preparation of 1, p-mercaptobenzoate derivatives 7 and 8, and N-acetyl-L-cysteine (NAC) derivatives 10 and 11 (Scheme 3), were confirmed with 1 H and 13 C-NMR spectra.Novel complexes 2-5 and 12-17 were characterised with elemental analysis, high resolution mass spectrometry, IR spectroscopy, and melting point.See supplementary material for 1 H and 13 C spectra of complexes 2-5 and 12-17.Herein we present a structural assessment of NHC-Au(I) complexes, based on (1,3-dibenzyl-4,5-diphenylimidazol-2-ylidene)gold(I) chloride (NHC*-AuCl), (1) Figure 1.The synthesis, characterisation, and biological testing of ten new NHC*-gold(I) complexes is reported.The effect of altering the coordinating ligands of the NHC*-gold(I) on the cytotoxicity is investigated via MTT-based proliferation assays.The cytotoxicity studies of these novel compounds have been conducted in vitro against three different tumour cell lines: MCF-7 topo (multidrug-resistant breast cancer), HCT-116 wt , and the p53 knockout mutant HCT-116 −/− (colon cancer).These cytotoxicity studies, compared to that of 1, can provide information on the ideal structures of future gold(I) chemotherapeutic complexes.Additionally, a computational study of 1 can highlight the advantages of employing an NHC ligand, as opposed to a phosphine.

Synthesis and Characterisation
The synthetic route for the ten NHC*-gold(I) complexes described in this paper are shown in Schemes 1-4.NHC*-Au(I)-Cl (1) was synthesised according to a procedure previously published [3].The preparation of 1, p-mercaptobenzoate derivatives 7 and 8, and N-acetyl-L-cysteine (NAC) derivatives 10 and 11 (Scheme 3), were confirmed with 1 H and 13 C-NMR spectra.Novel complexes 2-5 and 12-17 were characterised with elemental analysis, high resolution mass spectrometry, IR spectroscopy, and melting point.See Supplementary Material for 1 H and 13 C spectra of complexes 2-5 and 12-17.
Complex 2 was formed in a 66% yield from the anion exchange of chloride to cyanide (Scheme 1).The precursor 1 was reacted with potassium cyanide in dry dichloromethane at reflux for 48 h to produce complex 2. The reaction does not form the desired product when conducted in a biphasic solvent system with ethyl acetate and water.Upon reaction in the presence of water, the carbene is protonated to form the corresponding imidazolium dicyanoaurate(I), confirmed by a signal at δ = 8.66 ppm, representing the protonated carbene.Complex 2 was formed in a 66% yield from the anion exchange of chloride to cyanide (Scheme 1).The precursor 1 was reacted with potassium cyanide in dry dichloromethane at reflux for 48 h to produce complex 2. The reaction does not form the desired product when conducted in a biphasic solvent system with ethyl acetate and water.Upon reaction in the presence of water, the carbene is protonated to form the corresponding imidazolium dicyanoaurate(I), confirmed by a signal at δ = 8.66 ppm, representing the protonated carbene.The 1 H-NMR spectrum of 2 shows a slight shift of the CH2 protons of the benzyl groups, from δ = 5.44 ppm to 5.37 ppm, when compared to the 1 H-NMR of 1.The quaternary carbon of the cyanide ligand appears in the 13 C-NMR spectra at δ = 152.6 ppm.An absorption band at 2144 cm −1 in the IR spectra of 2 represents the C≡N stretch.
Complexes 3-5 were prepared by reacting complex 1 with the corresponding sodium carbamate salt (Scheme 2).This was performed under biphasic conditions by stirring at room temperature in ethyl acetate and water for 48 h, with relatively good yields of 61-69%.Complexes 3-5 were also synthesised in dichloromethane at reflux for 24 h, this, however, gave lower yields.Similar to 2, the CH2 signal in the 1 H-NMR of complexes 3-5 is shifted to δ = 5.57-5.55ppm upon coordination to the dithiocarbamates.The addition of a new 1 H-NMR singlet at δ = 3.51 ppm corresponding to the two methyl groups of the dimethyldithiocarbamate moiety (3) confirms its coordination to the NHC*-Au(I).Similarly, the CH2 and CH3 peaks of the diethyldithiocarbamate complex 4 appear at δ = 3.96 ppm and 1.31 ppm, respectively.The pyrrolidine CH2 peaks of 5 appear at δ = 3.85 and 1.97 ppm, with a triplet and pentet distinguishing these two peaks.
Previous metal-dialkyldithiocarbamate complexes reported the IR stretch of the carbon sulphur bond from 820-1050 cm −1 [19,20].The IR spectra of 3, 4 and 5 show a medium band at 971, 910, and 949 cm −1 , respectively, corresponding to the C=S stretch.A nickel(II) dimethyldithiocarbamate complex exhibited a carbon-sulphur bond stretch at 975 cm −1 [19,21], which correlates well with the dimethyldithiocarbamate complex 3. IR spectra of 3, 4, and 5 show bands at 1447, 1411, and 1406 cm −1 , respectively, which correspond to the carbon-nitrogen stretching of the carbamate.Interestingly, these IR values account for an intermediate bond in the 1450-1550 cm −1 range [20].This indicates a resonance structure is present where the carbon-nitrogen bond exhibits more double bond character The 1 H-NMR spectrum of 2 shows a slight shift of the CH 2 protons of the benzyl groups, from δ = 5.44 ppm to 5.37 ppm, when compared to the 1 H-NMR of 1.The quaternary carbon of the cyanide ligand appears in the 13 C-NMR spectra at δ = 152.6 ppm.An absorption band at 2144 cm −1 in the IR spectra of 2 represents the C≡N stretch.
Complexes 3-5 were prepared by reacting complex 1 with the corresponding sodium carbamate salt (Scheme 2).This was performed under biphasic conditions by stirring at room temperature in ethyl acetate and water for 48 h, with relatively good yields of 61-69%.Complexes 3-5 were also synthesised in dichloromethane at reflux for 24 h, this, however, gave lower yields.Complex 2 was formed in a 66% yield from the anion exchange of chloride to cyanide (Scheme 1).The precursor 1 was reacted with potassium cyanide in dry dichloromethane at reflux for 48 h to produce complex 2. The reaction does not form the desired product when conducted in a biphasic solvent system with ethyl acetate and water.Upon reaction in the presence of water, the carbene is protonated to form the corresponding imidazolium dicyanoaurate(I), confirmed by a signal at δ = 8.66 ppm, representing the protonated carbene.The 1 H-NMR spectrum of 2 shows a slight shift of the CH2 protons of the benzyl groups, from δ = 5.44 ppm to 5.37 ppm, when compared to the 1 H-NMR of 1.The quaternary carbon of the cyanide ligand appears in the 13 C-NMR spectra at δ = 152.6 ppm.An absorption band at 2144 cm −1 in the IR spectra of 2 represents the C≡N stretch.
Complexes 3-5 were prepared by reacting complex 1 with the corresponding sodium carbamate salt (Scheme 2).This was performed under biphasic conditions by stirring at room temperature in ethyl acetate and water for 48 h, with relatively good yields of 61-69%.Complexes 3-5 were also synthesised in dichloromethane at reflux for 24 h, this, however, gave lower yields.Similar to 2, the CH2 signal in the 1 H-NMR of complexes 3-5 is shifted to δ = 5.57-5.55ppm upon coordination to the dithiocarbamates.The addition of a new 1 H-NMR singlet at δ = 3.51 ppm corresponding to the two methyl groups of the dimethyldithiocarbamate moiety (3) confirms its coordination to the NHC*-Au(I).Similarly, the CH2 and CH3 peaks of the diethyldithiocarbamate complex 4 appear at δ = 3.96 ppm and 1.31 ppm, respectively.The pyrrolidine CH2 peaks of 5 appear at δ = 3.85 and 1.97 ppm, with a triplet and pentet distinguishing these two peaks.
Previous metal-dialkyldithiocarbamate complexes reported the IR stretch of the carbon sulphur bond from 820-1050 cm −1 [19,20].The IR spectra of 3, 4 and 5 show a medium band at 971, 910, and 949 cm −1 , respectively, corresponding to the C=S stretch.A nickel(II) dimethyldithiocarbamate complex exhibited a carbon-sulphur bond stretch at 975 cm −1 [19,21], which correlates well with the dimethyldithiocarbamate complex 3. IR spectra of 3, 4, and 5 show bands at 1447, 1411, and 1406 cm −1 , respectively, which correspond to the carbon-nitrogen stretching of the carbamate.Interestingly, these IR values account for an intermediate bond in the 1450-1550 cm −1 range [20].This indicates a resonance structure is present where the carbon-nitrogen bond exhibits more double bond character Similar to 2, the CH 2 signal in the 1 H-NMR of complexes 3-5 is shifted to δ = 5.57-5.55ppm upon coordination to the dithiocarbamates.The addition of a new 1 H-NMR singlet at δ = 3.51 ppm corresponding to the two methyl groups of the dimethyldithiocarbamate moiety (3) confirms its coordination to the NHC*-Au(I).Similarly, the CH 2 and CH 3 peaks of the diethyldithiocarbamate complex 4 appear at δ = 3.96 ppm and 1.31 ppm, respectively.The pyrrolidine CH 2 peaks of 5 appear at δ = 3.85 and 1.97 ppm, with a triplet and pentet distinguishing these two peaks.
Previous metal-dialkyldithiocarbamate complexes reported the IR stretch of the carbon sulphur bond from 820-1050 cm −1 [19,20].The IR spectra of 3, 4 and 5 show a medium band at 971, 910, and 949 cm −1 , respectively, corresponding to the C=S stretch.A nickel(II) dimethyldithiocarbamate complex exhibited a carbon-sulphur bond stretch at 975 cm −1 [19,21], which correlates well with the dimethyldithiocarbamate complex 3. IR spectra of 3, 4, and 5 show bands at 1447, 1411, and 1406 cm −1 , respectively, which correspond to the carbon-nitrogen stretching of the carbamate.Interestingly, these IR values account for an intermediate bond in the 1450-1550 cm −1 range [20].This indicates a resonance structure is present where the carbon-nitrogen bond exhibits more double bond character than the carbon-sulphur bonds.Furthermore, the presence of only one band for the C=S bond implies the molecule is symmetrical, and therefore, in the resonant structure shown in Figure 2   The synthetic route to ester formation is highlighted below in Scheme 3. The esters 7, 8, 10, and 11 were made with Fischer esterification, by refluxing 4-mercaptobenzoic acid ( 6) and N-acetyl-Lcysteine (NAC) (9) (both commercially available) in methanol and ethanol with a catalytic amount of sulphuric acid, to make their corresponding methyl and ethyl esters.Compounds 6-11 were conjugated with complex 1, under basic conditions, to obtain complexes 12-17, in relatively good yields of 57-78% (Scheme 4).Scheme 3. General reaction scheme for the synthesis of esters 7, 8, 10, and 11.
Compounds 6 and 9 were initially conjugated to 1 to make the corresponding NHC*-Au-S-linker molecules 12 and 15.Esterification of the acid ends of 12 and 15 was unsucessful.Attempts were made to synthesise complexes 13, 14, 16, 17 by reacting 12 and 15 with methanol or ethanol; however, this also proved to be unsucessful.Due to the lack of success via the linear synthesis, we moved to convergent synthesis, which was successful.
The most diagnostic feature in the 1 H-NMR spectra of complexes 12-17 is the disappearance of the SH signal of the thiols once coordinated to the gold.This appears in the δ = 3.64-2.48ppm range for the p-mercaptobenzoate compounds (12)(13)(14) ,and δ = 1.33-1.31ppm range for the NAC compounds (15)(16)(17).In the NAC series, the acetyl protons on the nitrogen atom of compounds 10 and 11 are observed at δ = 2.07 and 2.09 ppm, respectively.However, once linked to the NHC*-Au(I) centre, there is an observed upfield chemical shift of the acetyl protons to δ = 1.95 and 1.94 ppm in compounds 16 and 17, respectively.For complexes 13 and 16 there is a slight upfield shift of the CH3 singlet of the methyl compounds upon coordination to the gold; however, in the ethyl compounds, a downfield shift is noted.The synthetic route to ester formation is highlighted below in Scheme 3. The esters 7, 8, 10, and 11 were made with Fischer esterification, by refluxing 4-mercaptobenzoic acid ( 6) and N-acetyl-L-cysteine (NAC) (9) (both commercially available) in methanol and ethanol with a catalytic amount of sulphuric acid, to make their corresponding methyl and ethyl esters.Compounds 6-11 were conjugated with complex 1, under basic conditions, to obtain complexes 12-17, in relatively good yields of 57-78% (Scheme 4).The synthetic route to ester formation is highlighted below in Scheme 3. The esters 7, 8, 10, and 11 were made with Fischer esterification, by refluxing 4-mercaptobenzoic acid ( 6) and N-acetyl-Lcysteine (NAC) (9) (both commercially available) in methanol and ethanol with a catalytic amount of sulphuric acid, to make their corresponding methyl and ethyl esters.Compounds 6-11 were conjugated with complex 1, under basic conditions, to obtain complexes 12-17, in relatively good yields of 57-78% (Scheme 4).Compounds 6 and 9 were initially conjugated to 1 to make the corresponding NHC*-Au-S-linker molecules 12 and 15.Esterification of the acid ends of 12 and 15 was unsucessful.Attempts were made to synthesise complexes 13, 14, 16, 17 by reacting 12 and 15 with methanol or ethanol; however, this also proved to be unsucessful.Due to the lack of success via the linear synthesis, we moved to convergent synthesis, which was successful.
The most diagnostic feature in the 1 H-NMR spectra of complexes 12-17 is the disappearance of the SH signal of the thiols once coordinated to the gold.This appears in the δ = 3.64-2.48ppm range for the p-mercaptobenzoate compounds (12)(13)(14) ,and δ = 1.33-1.31ppm range for the NAC compounds (15)(16)(17).In the NAC series, the acetyl protons on the nitrogen atom of compounds 10 and 11 are observed at δ = 2.07 and 2.09 ppm, respectively.However, once linked to the NHC*-Au(I) centre, there is an observed upfield chemical shift of the acetyl protons to δ = 1.95 and 1.94 ppm in compounds 16 and 17, respectively.For complexes 13 and 16 there is a slight upfield shift of the CH3 singlet of the methyl compounds upon coordination to the gold; however, in the ethyl compounds, a downfield shift is noted.Compounds 6 and 9 were initially conjugated to 1 to make the corresponding NHC*-Au-S-linker molecules 12 and 15.Esterification of the acid ends of 12 and 15 was unsucessful.Attempts were made to synthesise complexes 13, 14, 16, 17 by reacting 12 and 15 with methanol or ethanol; however, this also proved to be unsucessful.Due to the lack of success via the linear synthesis, we moved to convergent synthesis, which was successful.
The most diagnostic feature in the 1 H-NMR spectra of complexes 12-17 is the disappearance of the SH signal of the thiols once coordinated to the gold.This appears in the δ = 3.64-2.48ppm range for the p-mercaptobenzoate compounds (12)(13)(14) ,and δ = 1.33-1.31ppm range for the NAC compounds (15)(16)(17).In the NAC series, the acetyl protons on the nitrogen atom of compounds 10 and 11 are observed at δ = 2.07 and 2.09 ppm, respectively.However, once linked to the NHC*-Au(I) centre, there is an observed upfield chemical shift of the acetyl protons to δ = 1.95 and 1.94 ppm in compounds 16 and 17, respectively.For complexes 13 and 16 there is a slight upfield shift of the CH 3 singlet of the methyl compounds upon coordination to the gold; however, in the ethyl compounds, a downfield shift is noted.

Structural Discussion
X-ray crystallography data was obtained for four of the complexes synthesised.The crystal of complex 2 was developed from the slow diffusion of pentane into a saturated dichloromethane solution at −18 °C.Complex 2 crystallised in the monoclinic space group P21/m (#11) (Figure 3).The crystals of 12 and 13 were formed in a saturated solution of ethyl acetate with the slow infusion of pentane (Figures 4 and 5).Both crystallised in the triclinic space group P1 (#2), in the absence of any solvent molecules.Crystal 14 was formed in a saturated solution of dichloromethane with slow infusion of diethyl ether (Figure 6).Complex 14 crystallized in the monoclinic space group C2/c (#15), also in the absence of any solvent molecules.The X-ray crystal data and structure refinement of complexes 2, 12, 13, and 14 are found in Table 1, with the selected bond lengths and bond angles compiled in Tables 2 and 3.

Structural Discussion
X-ray crystallography data was obtained for four of the complexes synthesised.The crystal of complex 2 was developed from the slow diffusion of pentane into a saturated dichloromethane solution at −18 • C. Complex 2 crystallised in the monoclinic space group P2 1 /m (#11) (Figure 3).The crystals of 12 and 13 were formed in a saturated solution of ethyl acetate with the slow infusion of pentane (Figures 4  and 5).Both crystallised in the triclinic space group P1 (#2), in the absence of any solvent molecules.Crystal 14 was formed in a saturated solution of dichloromethane with slow infusion of diethyl ether (Figure 6).Complex 14 crystallized in the monoclinic space group C2/c (#15), also in the absence of any solvent molecules.The X-ray crystal data and structure refinement of complexes 2, 12, 13, and 14 are found in Table 1, with the selected bond lengths and bond angles compiled in Tables 2 and 3.

Biological Evaluation
The in vitro anticancer activity of 2-5 and 12-17 was tested via MTT-based proliferation assays against the human colon carcinoma cell line HCT-116 wt , the p53 knockout mutant HCT-116 −/− , and the multidrug-resistant (mdr) human breast cancer cell line MCF-7 topo (Table 4).Bar 2 and 13, all complexes reached low single-digit micromolar IC 50 values against the tested cell lines after 72 h of treatment.These two complexes exhibit only moderate toxicities with IC 50 values up to 20 µM.While the IC 50 values of the dithiocarbamate complexes 3-5 and the p-mercaptobenzoate complexes 12-14 vary depending on the nitrogen substitution, and the respective esterification, the complexes carrying NAC, 15-17, show single-digit IC 50 values in the low micromolar range for all tested cell lines, with almost similar cytotoxic activities throughout.Esterification of NAC with methanol or ethanol slightly increased the antitumor activity against all three cell lines.Amongst the three types of thiolated complexes, the dithiocarbamate complexes 3-5 showed the highest activity against the mdr MCF-7 topo breast cancer cells, with complex 3 being the most active complex in total, with IC 50 values of 1.5 ± 0.1 µM against the HCT-116 wt or 0.28 ± 0.03 µM against the MCF-7 topo cells.To test the complexes for their dependency on fully functional p53, one activator of the apoptotic cascade, the complexes were tested for their toxicity against a HCT-116 p53 knockout mutant.Surprisingly, only a few of the tested complexes showed similar or higher IC 50 values against the knockout mutant than against the wildtype cells.Complexes 4, 5, 12, and 13 exert a higher toxicity against the HCT-116 −/− than against the wildtype HCT-116 wt .Overall, the herein presented complexes exhibit high to moderate antitumoral activity against colon carcinoma cells and a mdr breast cancer cell line.Dithiocarbamate complex 3 shows the overall highest activity in all tested cell lines.

General Conditions
All chemicals were purchased and used as received, unless otherwise stated.Solvents were dried according to the standard procedures, when necessary. 1 H and 13 C spectra were recorded on either a 300 or 400 MHz Varian spectrometer at room temperature (rt).Both chloroform (CDCl3) and dimethyl sulfoxide (DMSO) were used as deuterated solvents.The residual solvent peak or tetramethylsilane (TMS) were used as the internal standard.All chemical shifts are reported as δ values in parts per million (ppm).Infrared spectra were recorded on a Bruker ALPHA PLATINUM ATR spectrometer (Millerica, MA, USA).High resolution accurate mass data were obtained on a Waters/Micromass LCT TOF spectrometer (Milford, MA, USA).under electrospray ionisation technique.Melting points were measured on a Stuart™ (Stone, UK).melting point apparatus SMP10.Elemental analysis was conducted on an Exeter Analytical CE-440 elemental analyser (Coventry, UK).X-ray crystallography data was collected on a Rigaku Oxford Diffraction (Chalgrove, UK) SuperNova A diffractometer.Absorbance measurements were done with a TECAN (Männedorf, Switzerland) Infinite F200 plate Also, for NHC*-AuCl, two backbonding donations from the gold into the π* C-N antibonding orbitals are observed, E(2) = 15.9 and 16.1 kJ/mol; while in Ph 3 P-AuCl, three backbonding donations are observed from the Au atom into the π* P-C antibonding orbitals with E(2) = 16.2, 16.0, and 14.9 kJ/mol.The additional backbonding in the Ph 3 P-AuCl molecule reduces its bond strength, resulting in a weaker donating ligand.Conclusively, these results give credence to NHCs being the more favourable ligand than phosphines.

General Conditions
All chemicals were purchased and used as received, unless otherwise stated.Solvents were dried according to the standard procedures, when necessary. 1 H and 13 C spectra were recorded on either a 300 or 400 MHz Varian spectrometer at room temperature (rt).Both chloroform (CDCl 3 ) and dimethyl sulfoxide (DMSO) were used as deuterated solvents.The residual solvent peak or tetramethylsilane (TMS) were used as the internal standard.All chemical shifts are reported as δ values in parts per million (ppm).Infrared spectra were recorded on a Bruker ALPHA PLATINUM ATR spectrometer (Millerica, MA, USA).High resolution accurate mass data were obtained on a Waters/Micromass LCT TOF spectrometer (Milford, MA, USA).under electrospray ionisation technique.Melting points were measured on a Stuart™ (Stone, UK).melting point apparatus SMP10.Elemental analysis was conducted on an Exeter Analytical CE-440 elemental analyser (Coventry, UK).X-ray crystallography data was collected on a Rigaku Oxford Diffraction (Chalgrove, UK) SuperNova A diffractometer.Absorbance measurements were done with a TECAN (Männedorf, Switzerland) Infinite F200 plate reader.

Structure Determination
X-ray crystallography data was collected on a Rigaku Oxford Diffraction SuperNova A diffractometer.Complex 12 was measured with Mo-K α (0.71073 Å), while complexes 2, 13, and 14 were measured with Cu-K α (1.54184 Å).A complete dataset was collected, assuming that the Friedel pairs are not equivalent.An analytical absorption correction based on the shape of the crystal was performed [32].The structures were solved by direct methods using SHELXS [33] and refined by full matrix least-squares on F2 for all data using SHELXL [33].Hydrogen atoms were added at calculated positions and refined using a riding model.Their isotropic temperature factors were fixed to 1.2 times (1.5 times for methyl and OH groups) the equivalent isotropic displacement parameters of the parent atom.Anisotropic thermal displacement parameters were used for all non-hydrogen atoms.CCDC 1854008 (2), CCDC 1850909 (12), CCDC 1850910 (13), CCDC 1850908 ( 14) contain the supplementary crystallographic data for this paper, available free of charge from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/structures.

MTT-Based Proliferation Assay
The cytotoxic activity of al gold complexes was determined via MTT-based proliferation assays for the colon carcinoma cell line HCT-116 wt , its p53 knockout mutant HCT-116 −/− , and the multidrug-resistant MCF-7 topo breast cancer cell line.The cells, kept in Dulbecco's Modified Eagle Medium (1% anti-anti, 10% FBS), were seeded into the wells of a clear 96 well plate (5 × 10 4 cells/well) and incubated for 24 h at standard cell culture conditions (37 • C, 5% CO 2 , 95% humidity).Appropriate pre-dilutions of freshly made stock solutions (10 mM in DMSO) of 2-5, 12-15, and DMSO as negative control, were added into the wells of the pre-incubated cells.After 72 h, the medium was exchanged
[19].Molecules 2018, 23, x FOR PEER REVIEW 4 of 17 than the carbon-sulphur bonds.Furthermore, the presence of only one band for the C=S bond implies the molecule is symmetrical, and therefore, in the resonant structure shown in Figure 2 [19].

Figure 2 .
Figure 2. The dominant resonance form of a dithiocarbamate complex.

Figure 2 .
Figure 2. The dominant resonance form of a dithiocarbamate complex.

Molecules 2018 ,
23, x FOR PEER REVIEW 4 of 17than the carbon-sulphur bonds.Furthermore, the presence of only one band for the C=S bond implies the molecule is symmetrical, and therefore, in the resonant structure shown in Figure2[19].

Figure 2 .
Figure 2. The dominant resonance form of a dithiocarbamate complex.

Figure 4 .
Figure 4. X-ray diffraction structure of 12; thermal ellipsoids are drawn on the 50% probability level.

Figure 6 .
Figure 6.X-ray diffraction structure of 14; thermal ellipsoids are drawn on the 50% probability level.

Figure 6 .
Figure 6.X-ray diffraction structure of 14; thermal ellipsoids are drawn on the 50% probability level.

Table 1 .
Crystal data and structure refinement for complexes 2

Table 1 .
Crystal data and structure refinement for complexes 2