Versatile Platinum(IV) Prodrugs of Naproxen and Acemetacin as Chemo-Anti-Inflammatory Agents

Simple Summary Traditional intravenous platinum(II) chemotherapy drugs such as cisplatin, oxaliplatin and carboplatin are highly effective in treating multiple cancer types. Unfortunately, this treatment is most often beset with detrimental side effects that inevitably impact the patient’s willingness to comply with treatment programs. Platinum(II) drugs are scarcely selective, have poor bioavailability, and exhibit inherent and acquired resistance. A promising approach to address these impediments is the development of kinetically stable octahedral platinum(IV) complexes. This design strategy has been exploited for cisplatin and its derivatives and has been reported widely in the literature. This is a highly attractive approach for synthetic chemists due to the versatility it offers. Here, we contribute to this paradigm shift by using structurally distinct platinum(IV) scaffolds as effective prodrugs. The findings reported are expected to advance our understanding of cancer treatment. Abstract Developing new and versatile platinum(IV) complexes that incorporate bioactive moieties is a rapidly evolving research strategy for cancer drug discovery. In this study, six platinum(IV) complexes (1–6) that are mono-substituted in the axial position with a non-steroidal anti-inflammatory molecule, naproxen or acemetacin, were synthesised. A combination of spectroscopic and spectrometric techniques confirmed the composition and homogeneity of 1–6. The antitumour potential of the resultant complexes was assessed on multiple cell lines and proved to be significantly improved compared with cisplatin, oxaliplatin and carboplatin. The platinum(IV) derivatives conjugated with acemetacin (5 and 6) were determined to be the most biologically potent, demonstrating GI50 values ranging between 0.22 and 250 nM. Remarkably, in the Du145 prostate cell line, 6 elicited a GI50 value of 0.22 nM, which is 5450-fold more potent than cisplatin. A progressive decrease in reactive oxygen species and mitochondrial activity was observed for 1–6 in the HT29 colon cell line, up to 72 h. The inhibition of the cyclooxygenase-2 enzyme was also demonstrated by the complexes, confirming that these platinum(IV) complexes may reduce COX-2-dependent inflammation and cancer cell resistance to chemotherapy.


Introduction
With the rapid expansion of knowledge in coordination and organometallic chemistry, plenty of innovation towards exploiting metal elements to design and create valuable diagnostic and therapeutic drugs have been conveyed in the literature, especially for cancer treatment [1]. Cancer continues to be a major public health concern, and 1. Introduction With the rapid expansion of knowledge in coordination and organometallic chemistry, plenty of innovation towards exploiting metal elements to design and create valuable diagnostic and therapeutic drugs have been conveyed in the literature, especially for cancer treatment [1]. Cancer continues to be a major public health concern, and cancer treatment strategies have become more advanced in an attempt to provide effective treatment [2]. Traditional intravenous chemotherapy is still the established clinical regimen and is considered to be the most effective approach to kill cancer cells. This is predominantly due to the ground-breaking clinical success of platinum(II) drugs, cis-diamminedichloroplatinum(II) (cisplatin), trans-L-(1R,2R-diaminocyclohexane) oxalatoplatinum(II) (oxaliplatin) and cis-diammine (1,1-cyclobutanedicarboxylato) platinum(II) (carboplatin) (Figure 1) as mainstream anticancer agents [3][4][5][6]. The therapeutic potential of these drugs is attributed to their ability to covalently bind with deoxyribonucleic acid (DNA) and form crosslinks (i.e., DNA inter/intra-strand crosslinks and DNA-protein crosslinks) that prevent further DNA replication, and consequently, induce apoptosis [5,[7][8][9][10]. Although these drugs are effective against multiple tumour types, indiscriminate toxicity, poor bioavailability, and easily acquired drug resistance are challenging parameters that limit their use in the clinic [11][12][13][14]. We have previously reported a class of non-DNA coordinating platinum(II) complexes in the form of [Pt II (HL)(AL)] 2+ , where HL is the heterocyclic ligand (i.e., 1,10-phenanthroline (phen), 5-methyl-1,10-phenanthroline (5-Mephen) or 5,6-dimethyl-1,10-phenanthroline (5,6-Me2phen)) and AL is an ancillary ligand which is chiral, 1S,2S-diaminocyclohexane (SS-DACH or DACH) (Figure 1), which exhibits exceptional in vitro activity [15][16][17][18][19][20]. This unconventional class of platinum(II) complexes demonstrated potency significantly better than cisplatin, oxaliplatin or carboplatin in multiple cancer cell lines [20][21][22]. Our most potent complex, [Pt II (5,6-Me2phen)(SS-DACH)] 2+ (56MESS), demonstrates a mode of action that differs from traditional platinum(II) drugs, as this complex is proposed to modify the cytoskeletal architecture, induce bioenergetic stress that reduces the mitochondrial membrane potential (MtMP) and promotes epigenetic changes [21,23]. This class of platinum(II) complexes exhibits incomparable in vitro activity, but this was not initially seen during in vivo experiments. Treatment with 56MESS of BD-IX rats exhibiting peritoneal carcinomatosis through intravenous and intraperitoneal routes did not demonstrate tumour suppression response but instead induced nephrotoxicity [24]. We have previously reported a class of non-DNA coordinating platinum(II) complexes in the form of [Pt II (H L )(A L )] 2+ , where H L is the heterocyclic ligand (i.e., 1,10-phenanthroline (phen), 5-methyl-1,10-phenanthroline (5-Mephen) or 5,6-dimethyl-1,10-phenanthroline (5,6-Me 2 phen)) and A L is an ancillary ligand which is chiral, 1S,2S-diaminocyclohexane (SS-DACH or DACH) (Figure 1), which exhibits exceptional in vitro activity [15][16][17][18][19][20]. This unconventional class of platinum(II) complexes demonstrated potency significantly better than cisplatin, oxaliplatin or carboplatin in multiple cancer cell lines [20][21][22]. Our most potent complex, [Pt II (5,6-Me 2 phen)(SS-DACH)] 2+ (56MESS), demonstrates a mode of action that differs from traditional platinum(II) drugs, as this complex is proposed to modify the cytoskeletal architecture, induce bioenergetic stress that reduces the mitochondrial membrane potential (MtMP) and promotes epigenetic changes [21,23]. This class of platinum(II) complexes exhibits incomparable in vitro activity, but this was not initially seen during in vivo experiments. Treatment with 56MESS of BD-IX rats exhibiting peritoneal carcinomatosis through intravenous and intraperitoneal routes did not demonstrate tumour suppression response but instead induced nephrotoxicity [24]. Alternatively, when PHENSS or cisplatin were administered to mice transplanted with a PC3 (prostate carcinoma) tumour, both complexes exhibited a comparable reduction in mean tumour mass relative to the vehicle-treated control group [19]. Notably, treatment with PHENSS did not elicit any toxic side effects, while those mice treated with cisplatin were euthanised due to adverse side effects by Day 20. Due to the in vitro and in vivo discrepancies observed, The configurational arrangement of platinum(IV) complexes is an attribute of the design and a game changer for researchers in this area of research. Cisplatin, oxaliplatin and carboplatin are the most popular platinum(II) cores used to generate platinum(IV) scaffolds in many studies, and each have been coordinated to various enzymatic inhibitors that supress cancer development and progression [28,36]. In particular, the coordination of non-steroidal anti-inflammatory drugs (NSAIDs) to the cores of cisplatin and its derivatives have resulted in chemo-anti-inflammatory prodrugs that are capable of reversing chemoresistance because of enhanced bioavailability [37][38][39][40][41]. Since the 1970s, NSAIDs have been extensively administered to cancer patients, simply for cancer pain management, but there was also a growing curiosity as to whether the regular ingestion of NSAIDs would help combat cancer and decrease cancer risk [42][43][44][45]. NSAIDs suppress Figure 2. Exemplar platinum(IV) complex incorporating the platinum(II) scaffolds (PHENSS, 5MESS or 56MESS) with axial ligands (X). The separation of the axial ligands from the platinum(II) scaffolds is initiated by the two-electron reduction process. Counter-ions were omitted for clarity.
The configurational arrangement of platinum(IV) complexes is an attribute of the design and a game changer for researchers in this area of research. Cisplatin, oxaliplatin and carboplatin are the most popular platinum(II) cores used to generate platinum(IV) scaffolds in many studies, and each have been coordinated to various enzymatic inhibitors that supress cancer development and progression [28,36]. In particular, the coordination of non-steroidal anti-inflammatory drugs (NSAIDs) to the cores of cisplatin and its derivatives have resulted in chemo-anti-inflammatory prodrugs that are capable of reversing chemoresistance because of enhanced bioavailability [37][38][39][40][41]. Since the 1970s, NSAIDs have been extensively administered to cancer patients, simply for cancer pain management, but there was also a growing curiosity as to whether the regular ingestion of NSAIDs would help combat cancer and decrease cancer risk [42][43][44][45]. NSAIDs suppress the activity of cyclooxygenase (COX) enzymes, COX-1 and COX-2, which play a role in prostaglandin biosynthesis [46]. Prostaglandins are a group of physiologically active lipids that are involved in cellular signal transduction pathways and also, are key mediators in the inflammatory response [47]. Chronic inflammation is a key factor in cancer progression and occurs in 20% of all human cancers [48]. Because chronic inflammation is directly involved with the overexpression of COX enzymes [49,50], this makes COX enzymes as viable therapeutic targets, especially COX-2, as it is predominantly overexpressed in a few cancer types (i.e., lung, colon, prostate, and breast) [51].
We present here six mono-substituted platinum(IV) complexes that incorporate either naproxen (NPX) or acemetacin (ACE), which is the prodrug of indomethacin ( (Figure 3). The axial coordination of NPX or ACE into the platinum(IV) scaffolds is expected to produce prodrugs with enhanced biological activity. This approach will also provide a better understanding of the influence of NSAIDs in platinum(IV) prodrug chemotherapeutics in cancer treatment.
the activity of cyclooxygenase (COX) enzymes, COX-1 and COX-2, which play a role in prostaglandin biosynthesis [46]. Prostaglandins are a group of physiologically active lipids that are involved in cellular signal transduction pathways and also, are key mediators in the inflammatory response [47]. Chronic inflammation is a key factor in cancer progression and occurs in 20% of all human cancers [48]. Because chronic inflammation is directly involved with the overexpression of COX enzymes [49,50], this makes COX enzymes as viable therapeutic targets, especially COX-2, as it is predominantly overexpressed in a few cancer types (i.e., lung, colon, prostate, and breast) [51].
We present here six mono-substituted platinum(IV) complexes that incorporate either naproxen (NPX) or acemetacin (ACE), which is the prodrug of indomethacin ( Figure  3). The general structure of the synthesised platinum(IV) complexes are defined as  (Figure 3). The axial coordination of NPX or ACE into the platinum(IV) scaffolds is expected to produce prodrugs with enhanced biological activity. This approach will also provide a better understanding of the influence of NSAIDs in platinum(IV) prodrug chemotherapeutics in cancer treatment. The composition and homogeneity of the studied platinum(IV) complexes (1-6) was confirmed by high-performance liquid chromatography (HPLC), nuclear magnetic resonance ( 1 H-NMR; two-dimensional correlation spectroscopy (2D-COSY); heteronuclear The composition and homogeneity of the studied platinum(IV) complexes (1-6) was confirmed by high-performance liquid chromatography (HPLC), nuclear magnetic resonance ( 1 H-NMR; two-dimensional correlation spectroscopy (2D-COSY); heteronuclear multiple quantum correlation ( 1 H-195 Pt-HMQC)), ultraviolet-visible (UV), circular dichroism (CD), high-resolution electrospray ionisation mass spectrometry (ESI-MS), and infrared (IR) spectroscopy. The solubility of 1-6 was determined. The reduction behaviour of 1-6 in aqueous solution containing AsA at 37 • C was reported with the aid of 1 H-NMR and onedimensional 195 Pt-NMR (1D-195 Pt-NMR). The stability of 1-6 in aqueous solution without a reducing agent was monitored using HPLC, at room temperature and at 37 • C for 36 h. Lipophilicity measurements were also undertaken for 1-6, utilising HPLC. In vitro cytotox-  assessed in multiple cell lines, including: HT29 colon, U87 glioblastoma,  MCF7 breast, A2780 ovarian, H460 lung, A431 skin, Du145 prostate, BE2C neuroblastoma, SJG2 glioblastoma, MIA pancreas, ADDP ovarian variant (cisplatin-resistant A2780  clone), and the non-tumour derived MCF10A breast line. Finally, the reactive oxygen species (ROS) potential, changes in MtMP, and COX-2 inhibition capability of 1-6 were assessed to evaluate how impactful these parameters are to the overall in vitro cytotoxicity of the complexes.

Materials
All laboratory reagents used in the experiments were of analytical grade.

Cell Viability Assays
Cell viability assays were completed at the Calvary Mater Newcastle Hospital, NSW, Australia, as previously reported (see Supplementary Materials) [52,53,55,60]. The GI 50 values of platinum(II) precursors and platinum(IV) scaffolds, together with cisplatin, oxaliplatin and carboplatin, which were determined using the previously reported method, were also presented in this study for comparison [53,55]. The selectivity cytotoxicity index (SCI) for 1-6 and ligands (NPX and ACE) was calculated by dividing the GI 50 values of the complexes or ligands in the normal breast cell line MCF10A by their GI 50 in the cancerous prostate cell line Du145. Generally, a greater SCI denotes higher selectivity towards cancer cells [61,62].

Reactive Oxygen Species (ROS) Potential
To determine the presence of ROS in treated cells, a DCFDA/H 2 DCFDA-cellular ROS Assay Kit (Abcam, Cambridge, MA, USA) was utilised, as previously reported [53,55,63,64]. The ROS potential of platinum(II) precursors and platinum(IV) scaffolds, which were determined using the same method, were also presented in this study for comparison [53].

Synthesis and Characterisation
The NHS esters of NPX and ACE were synthesised using previously established methods [52,53]. The successful synthesis of the NHS esters was confirmed by HPLC (Figures S1 and S2), 1 H-NMR ( Figures S3 and S4), and ESI-MS ( Figures S5 and S6) experiments. The platinum(II) precursors and platinum(IV) scaffolds of types, [Pt II (HL)(AL)] 2+ and [Pt IV (HL)(AL)(OH)2] 2+ , were synthesised as previously described [52,53,55]. To prepare [Pt IV (HL)(AL)(X)(OH)] 2+ (1-6) (Figure 4), previously established protocols were applied [52,53]. All resulting complexes were purified through a flash chromatography system to obtain higher purity. Characterisation confirming the composition and homogeneity of 1-6 included HPLC ( Figures S7-S12 Figures S55-S60). All experimental yields, HPLC peak areas (%) and TR, and mass-to-charge ratios (m/z) of 1-6 are outlined in Table 1.   observed. With respect to the 1 H-NMR spectrum of 1 ( Figure 5), 17 peaks were recorded including the large signal at 4.70 ppm induced by the deuterated solvent used. A notable upfield shift movement was observed along the aromatic region (8-9 ppm) according to the resonances elicited by the phen protons. While this is mostly influenced by the axial coordination of NPX, a more viable explanation to this phenomenon is reflective of the structure of NPX. NPX contains an electron-rich naphthalene group, which may have interacted with the phen ring system of the complex through π-π stacking. Considering the high electron density of the aromatic rings (i.e., phen and naphthalene), this effect generates greater opposition to the applied magnetic field, which causes the hydrocarbons to shield and resonate at a lower frequency. The protons originating from phen, H2 and H9 resonated as a triplet at 9.12 ppm with a calculated J-coupling constant of 5.0 Hz. Due to the deshielding effect induced by the electronegativity of the nitrogen (N) atoms bound to the phen ring, the protons resonated furthest downfield. The two individual doublets at 8.74 and 8.70 ppm were assigned to H4 and H7 protons, respectively, and both have a calculated J-coupling constant of 8.3 Hz This was followed by a multiplet at 8.10 ppm, exhibited by H3 and H8 protons. The quartet at 7.67 ppm with a calculated J-coupling constant of 8.9 Hz was assigned to protons H5 and H6. Overall, these multiplicities observed for 1 in the aromatic region contrast what is typically reported for its corresponding platinum(II) and platinum(IV) complexes, PHENSS and PHENSS(IV)(OH) 2 , as they typically resonate as doublets and particularly a singlet for H5 and H6 ( Figure 6) [22,52,53,[55][56][57]65]. Table 3. Summary of the 1 H-NMR and 1 H-195 Pt-HMQC data of platinum(IV)-ACE derivatives (4-6), including chemical shifts (δ ppm), multiplicity, integration and J-coupling constants (Hz). As for the protons originating from the naphthalene group of NPX represented by a, b, c, d, e and f, varied multiplicity was also demonstrated ( Figure 5). The more deshielded protons near the alkoxy group, c and d, resonated as a multiplet at 7.12 ppm, while the less deshielded protons, a and b, resonated as separate singlets at 7.04 and 6.71 ppm, respectively. Moreover, e resonated as a doublet at 6.84 ppm with a calculated J-coupling constant of 8.5 Hz, while f, which is the least deshielded proton in the naphthalene group, resonated at 6.40 ppm as a doublet of doublets with calculated J-coupling constants of 1.4 and 8.5 Hz. The sharp singlet at 3.99 ppm was assigned to the methyl at the alkoxy group of NPX represented by γ. The methylene (α) and methyl (β) protons near the carbonyl group of NPX were assigned to the quartet and doublet at 3.50 and 1.15 ppm, respectively. Finally, the chemical multiplicity in the aliphatic region (1-3 ppm) exhibited by the ancillary ligand, SS-DACH agrees with the literature data [22,25,52,53,[55][56][57][58][59]66]. and 3, including chemical shifts (δ ppm), multiplicity, integration, and calculated J-coupling constants (Hz). Due to proton exchange with D2O, no amine proton resonances were observed. With respect to the 1 H-NMR spectrum of 1 ( Figure 5), 17 peaks were recorded including the large signal at 4.70 ppm induced by the deuterated solvent used. A notable upfield shift movement was observed along the aromatic region (8-9 ppm) according to the resonances elicited by the phen protons. While this is mostly influenced by the axial coordination of NPX, a more viable explanation to this phenomenon is reflective of the structure of NPX. NPX contains an electron-rich naphthalene group, which may have interacted with the phen ring system of the complex through π-π stacking. Considering the high electron density of the aromatic rings (i.e., phen and naphthalene), this effect generates greater opposition to the applied magnetic field, which causes the hydrocarbons to shield and resonate at a lower frequency.  To further confirm the successful coordination of NPX to the platinum core and only occupied one axial position, 1 H-195 Pt-HMQC experiments were undertaken at -2800 and 400 ppm. Normally, the platinum(II) scaffolds used in this study resonate at −2800 ppm while their corresponding platinum(IV) derivatives resonate at 400 ppm, as previously reported [22,25,52,53,[55][56][57][58][59]66]. According to the 1 H-195 Pt-HMQC spectrum of 1 (Figure 7), the correlation of the phen protons to the platinum was confirmed when two peaks resonated at 545 ppm. Specifically, the cross-coupling of H2 and H9 (9.12 ppm) and of H3 and H8 (8.10 ppm) protons with the platinum peaks at 545 ppm demonstrate correlation.

Proton
With respect to the 1 H-NMR spectrum of 2 ( Figure S14), a few differences in multiplicity were noted compared to the results observed for 1 ( Figure 5) and 3 ( Figure S15), specifically the multiplicity in the aromatic region (8-9 ppm). For example, the H2 and H9 protons of 2 appeared as two separate signals, a triplet at 9.12 ppm and a doublet 9.04 ppm, exhibiting the same J-coupling constant of 5.6 Hz, respectively. Additionally, the two separate quartets at 8.84 and 8.60 ppm were assigned to H4 and H7 protons, while the multiplet at 8.10 ppm was assigned to H3 and H8 protons. The H6 proton resonated as a doublet at 7.40 ppm, with a J-coupling constant of 19 Hz. This high J-coupling constant is a result of the long-range coupling between H6 and its neighbouring methyl group at the 5 position (consisting of three protons) in the heterocyclic ring system ( Figure S20). Overall, these differences in multiplicity of the heterocyclic protons are indicative of the structure of the complexes, considering that 2 and 3 contain methyl groups in their heterocyclic ring systems, while 1 does not. Nonetheless, the successful coordination of NPX to platinum was confirmed according to the 1 H-195 Pt-HMQC results observed for 2 (Figures S27 and S28) and 3 ( Figures S29 and S30). a calculated J-coupling constant of 5.0 Hz. Due to the deshielding effect induced by the electronegativity of the nitrogen (N) atoms bound to the phen ring, the protons resonated furthest downfield. The two individual doublets at 8.74 and 8.70 ppm were assigned to H4 and H7 protons, respectively, and both have a calculated J-coupling constant of 8.3 Hz This was followed by a multiplet at 8.10 ppm, exhibited by H3 and H8 protons. The quartet at 7.67 ppm with a calculated J-coupling constant of 8.9 Hz was assigned to protons H5 and H6. Overall, these multiplicities observed for 1 in the aromatic region contrast what is typically reported for its corresponding platinum(II) and platinum(IV) complexes, PHENSS and PHENSS(IV)(OH)2, as they typically resonate as doublets and particularly a singlet for H5 and H6 ( Figure 6) [22,52,53,[55][56][57]65]. As for the protons originating from the naphthalene group of NPX represented by a, b, c, d, e and f, varied multiplicity was also demonstrated ( Figure 5). The more deshielded protons near the alkoxy group, c and d, resonated as a multiplet at 7.12 ppm, while the less deshielded protons, a and b, resonated as separate singlets at 7.04 and 6.71 ppm, respectively. Moreover, e resonated as a doublet at 6.84 ppm with a calculated J-coupling constant of 8.5 Hz, while f, which is the least deshielded proton in the naphthalene group, resonated at 6.40 ppm as a doublet of doublets with calculated J-coupling constants of 1.4 and 8.5 Hz. The sharp singlet at 3.99 ppm was assigned to the methyl at the alkoxy group of NPX represented by γ. The methylene (α) and methyl (β) protons near the carbonyl group of NPX were assigned to the quartet and doublet at 3.50 and 1.15 ppm, respectively. Finally, the chemical multiplicity in the aliphatic region (1-3 ppm) exhibited by the ancillary ligand, SS-DACH agrees with the literature data [22,25,52,53,[55][56][57][58][59]66].
To further confirm the successful coordination of NPX to the platinum core and only occupied one axial position, 1 H-195 Pt-HMQC experiments were undertaken at -2800 and Furthermore, the results obtained for the remaining complexes incorporating the ACE ligand, 4-6, also showed distinct differences in their chemical shifts, which are attributed to the axial coordination of ACE and the number of methyl groups in the heterocyclic ligands. The assignment of resonances for the 1 H-NMR spectra of 4-6 ( Table 3 and Figures S16-S18) followed the same rationale to that described for 1-3. Lastly, the successful coordination of ACE to platinum at one axial position was also confirmed by 1 H-195 Pt-HMQC ( Figures S31-S36).
The UV spectra for 1-6 ( Figures S37-S42) were recorded. For the UV measurements, both metal-to-ligand charge transfer interactions and π-π* transitions were observed. These are primarily influenced by the heterocyclic ligands of the complexes (i.e., phen, 5-Mephen, and 5,6-Me 2 phen) that demonstrate ligand-centred π-π* transitions. The UV spectra of platinum(IV) derivatives incorporating NPX (1-3) exhibited similar absorption bands ( Figures S37-S39 and 8). Three prominent absorption bands were recorded at the wavelengths,~200,~230 and~279-289 nm. The differences in the methylation of the heterocyclic ligands of the complexes resulted in slight bathochromic shifts or red shifts, as shown in Figure 8. This pattern is also in agreement with the literature data [25,52,53,[55][56][57][58][59]66,67]. Of further note, a prominent absorption band at~235 nm was acquired for the uncoordinated NPX ligand, followed by three small peaks between 260 and 290 nm. Upon coordination of NPX to platinum(IV), the band at~235 nm was shifted to lower wavelengths ( Figure 8). This may be influenced by the π-π interactions between the aromatic system of NPX and the heterocyclic system of the platinum(IV), since larger conjugated systems tend to absorb at lower wavelengths.  [22,25,52,53,[55][56][57][58][59]66]. According to the 1 H-195 Pt-HMQC spectrum of 1 (Figure 7), the correlation of the phen protons to the platinum was confirmed when two peaks resonated at 545 ppm. Specifically, the cross-coupling of H2 and H9 (9.12 ppm) and of H3 and H8 (8.10 ppm) protons with the platinum peaks at 545 ppm demonstrate correlation. With respect to the 1 H-NMR spectrum of 2 ( Figure S14), a few differences in multiplicity were noted compared to the results observed for 1 ( Figure 5) and 3 ( Figure S15), specifically the multiplicity in the aromatic region (8-9 ppm). For example, the H2 and H9 protons of 2 appeared as two separate signals, a triplet at 9.12 ppm and a doublet 9.04 ppm, exhibiting the same J-coupling constant of 5.6 Hz, respectively. Additionally, the two separate quartets at 8.84 and 8.60 ppm were assigned to H4 and H7 protons, while the multiplet at 8.10 ppm was assigned to H3 and H8 protons. The H6 proton resonated as a doublet at 7.40 ppm, with a J-coupling constant of 19 Hz. This high J-coupling constant is a result of the long-range coupling between H6 and its neighbouring methyl group at the 5 position (consisting of three protons) in the heterocyclic ring system ( Figure S20). Overall, these differences in multiplicity of the heterocyclic protons are indicative of the structure of the complexes, considering that 2 and 3 contain methyl groups in their heterocyclic ring systems, while 1 does not. Nonetheless, the successful coordination of NPX to platinum was confirmed according to the 1 H-195 Pt-HMQC results observed for 2 ( Figures  S27 and S28) and 3 ( Figures S29 and S30).
Furthermore, the results obtained for the remaining complexes incorporating the ACE ligand, 4-6, also showed distinct differences in their chemical shifts, which are attributed to the axial coordination of ACE and the number of methyl groups in the heterocyclic ligands. The assignment of resonances for the 1 H-NMR spectra of 4-6 ( Table 3 and  -195 Pt-HMQC spectrum of 1, highlighting the correlation between H2, H9, H3 and H8 protons with the platinum core. Region: 400 ppm. Inset: structure of 1 with proton labelling system and arrows indicating correlation. Table 4. Characteristic peaks in the UV and CD spectra of 1-6.

Platinum (IV) Complexes
UV λ max nm (ε/M.cm −1 ± SD × 10 4 ) CD λ max nm (∆ε/M.cm −1 × 10 1 )  In comparison, the UV spectra of the platinum(IV) derivatives incorporating ACE (4-6) also followed a similar trend to 1-3, particularly the red shift recorded between 270 and 290 nm (Figures 9 and S40-S42). The absorption bands recorded for 4-6 were slightly blue shifted compared to the absorption bands measured for 1-3. This may be due to the structural differences of the axial ligands, NPX and ACE. NPX contains a naphthalene group that experiences hyperconjugation, which may be increased by the presence of the oxygen (O) in its alkoxy group. The O at the alkoxy group is electron-donating, thus this increases the delocalisation of electrons within the naphthalene ring. On the contrary, while ACE also contains aromatic rings, these are surrounded by electron-withdrawing groups, such as Cl and N atoms, and a carbonyl that reduces electron density. In comparison, the UV spectra of the platinum(IV) derivatives incorporating ACE (4-6) also followed a similar trend to 1-3, particularly the red shift recorded between 270 and 290 nm (Figures 9 and S40-S42). The absorption bands recorded for 4-6 were slightly blue shifted compared to the absorption bands measured for 1-3. This may be due to the structural differences of the axial ligands, NPX and ACE. NPX contains a naphthalene group that experiences hyperconjugation, which may be increased by the presence of the oxygen (O) in its alkoxy group. The O at the alkoxy group is electron-donating, thus this increases the delocalisation of electrons within the naphthalene ring. On the contrary, while ACE also contains aromatic rings, these are surrounded by electron-withdrawing groups, such as Cl and N atoms, and a carbonyl that reduces electron density.
Since 1-6 incorporate a chiral ancillary ligand, SS-DACH, it was only appropriate to confirm if the complexes' chirality has been retained during synthesis. Chirality is an essential parameter that influences the potency of the complexes, and it has been previously established that by substituting the SS-DACH to its enantiomer, RR-DACH, significant differences in overall cytotoxicity of the complexes were observed [17,18,20,65], where those complexes incorporating RR-DACH were significantly less potent than those complexes that incorporate the SS-DACH as the ancillary ligand.
For the CD measurements of 1-6, variations were exhibited ( Figures S43-S48). The absorption bands of 1-3 ( Figures S43-S45) were more defined and prominent compared to the absorption bands of 4-6 ( Figures S46-S48). Evidently, 1-3 displayed stronger positive and negative absorption bands, while 4-6 displayed weaker positive and negative absorption bands (Table 4). Prominent positive absorption bands at the lower wavelengths (239-244 nm) were exhibited by 1-3 ( Figures S43-S45), which were not recorded for 4-6 ( Figures S46-S48). These variations are likely more influenced by the axial ligands, NPX and ACE, rather than the methylation of the heterocyclic systems of the complexes. In summary, except for the strong positive absorption bands exhibited by 1-3 between 239-244 nm, the CD measurements obtained follow a comparable trend with published examples of platinum(IV) complexes [22,52,53,[55][56][57][58][59]66], confirming that chirality was retained during synthesis. Since 1-6 incorporate a chiral ancillary ligand, SS-DACH, it was only appropriate to confirm if the complexes' chirality has been retained during synthesis. Chirality is an essential parameter that influences the potency of the complexes, and it has been previously established that by substituting the SS-DACH to its enantiomer, RR-DACH, significant differences in overall cytotoxicity of the complexes were observed [17,18,20,65], where those complexes incorporating RR-DACH were significantly less potent than those complexes that incorporate the SS-DACH as the ancillary ligand.
IR measurements were also undertaken to further verify the functional groups present in the structures of 1-6 ( Figures S55-S60). Some of the prominent IR absorption peaks recorded for the complexes are summarised in Table 5. Because of the structural similarities of the complexes, the peaks for all complexes follow almost the same pattern ( Figures  S55-S60). As shown in Table 5  IR measurements were also undertaken to further verify the functional groups present in the structures of 1-6 ( Figures S55-S60). Some of the prominent IR absorption peaks recorded for the complexes are summarised in Table 5. Because of the structural similarities of the complexes, the peaks for all complexes follow almost the same pattern ( Figures S55-S60). As shown in Table 5, the types of bonds observed in the IR spectra of 1-6 included the following: O-H stretch (3410-3385 cm −1 ), C-H aromatic stretch (3062-3076 cm −1 ), C-H alkyl stretch (2936-2943 cm −1 ), C-C aromatic stretch (1603-1609 cm −1 ), C-O alkyl aryl ether stretch (1218-1265 cm −1 ) and a C-Cl halogen stretch (845-846 cm −1 ). These functional groups are also present in the complexes. The C-Cl stretch was only exhibited by the platinum(IV) complexes incorporating the ACE axial ligand, 4-6 (845-846 cm −1 ), considering that ACE is the only axial ligand that has a Cl atom. Because of the multiple functional groups present in the complexes, the frequencies observed cannot be assigned to a specific functional group. For example, the C-H and C-C vibrations from the heterocyclic ligands (phen, 5-Mephen and 5,6-Me 2 phen) of the platinum(II) cores may exhibit overlapping vibrations as those with the coordinated ACE and NPX ligands. Overall, the vibrations observed in the IR spectra of 1-6 are comparable to literature data [68][69][70][71][72].

Solubility Measurements
Solubility is an essential parameter that can predict drug effectiveness [73]. Generally, a drug with poor solubility will only promote pharmacokinetic challenges ( Figure 10). Platinum(II) drugs such as cisplatin, oxaliplatin and carboplatin are effective; however, they cannot be administered orally due to their limited solubility in aqueous conditions [36]. The poor solubility of these drugs is also a consequence of their poor bioavailability, and generally, high-dose administration is often required to achieve adequate pharmacological response. Although, high-dose administration can increase patient burden, especially the risks of patients developing adverse side effects considering the high toxicity of the drugs [74].

Solubility Measurements
Solubility is an essential parameter that can predict drug effectiveness [73]. Generally, a drug with poor solubility will only promote pharmacokinetic challenges ( Figure  10). Platinum(II) drugs such as cisplatin, oxaliplatin and carboplatin are effective; however, they cannot be administered orally due to their limited solubility in aqueous conditions [36]. The poor solubility of these drugs is also a consequence of their poor bioavailability, and generally, high-dose administration is often required to achieve adequate pharmacological response. Although, high-dose administration can increase patient burden, especially the risks of patients developing adverse side effects considering the high toxicity of the drugs [74]. The inherent characteristics of platinum(IV) complexes have inspired optimism that an effective orally available prodrug can be designed. Evidence has been demonstrated previously with bis-(acetate)-ammine dichloro-(cyclohexylamine) platinum(IV) (satraplatin), which was the first oral platinum agent that took part in phase III clinical trials but had to be discontinued due to low overall survival benefit [75]. The availability of orally available anticancer prodrugs will undoubtedly improve treatment experience and patient care, as hospitalisation would not be necessary.
In this study, the solubility for 1-6 in d.i.H2O at room temperature was determined. All values were expressed in mg/mL and mol/L, as shown in Table 6. Complexes 1-6 were found to be more soluble than cisplatin, oxaliplatin and carboplatin in water ( Table 6). The results indicate that the studied complexes should have better bioavailability than clinically used platinum(II) drugs and may be suited for oral administration. The inherent characteristics of platinum(IV) complexes have inspired optimism that an effective orally available prodrug can be designed. Evidence has been demonstrated previously with bis-(acetate)-ammine dichloro-(cyclohexylamine) platinum(IV) (satraplatin), which was the first oral platinum agent that took part in phase III clinical trials but had to be discontinued due to low overall survival benefit [75]. The availability of orally available anticancer prodrugs will undoubtedly improve treatment experience and patient care, as hospitalisation would not be necessary.
In this study, the solubility for 1-6 in d.i.H 2 O at room temperature was determined. All values were expressed in mg/mL and mol/L, as shown in Table 6. Complexes 1-6 were found to be more soluble than cisplatin, oxaliplatin and carboplatin in water ( Table 6). The results indicate that the studied complexes should have better bioavailability than clinically used platinum(II) drugs and may be suited for oral administration.

Lipophilicity Measurements
In conjunction with solubility, lipophilicity also plays a vital role in drug effectiveness, as it directly impacts the diffusion of a drug through cell membranes (or permeability) [77]. Generally, NSAIDs (i.e., NPX and ACE) are highly lipophilic substances because they contain lipophilic groups such as carboxylic acids and aromatic rings [78]. Moreover, the biological activity of NSAIDs is mostly dependent on lipophilicity [79].
Since complexes 1-6 incorporate the NSAIDs, NPX and ACE, it was only appropriate to determine their lipophilicity.
Lipophilicity measurements were undertaken using HPLC as previously described [53,[55][56][57][58][59]. The complexes were eluted at different isocratic ratios. The capacity factor (k) was calculated using the retention times of the complexes. A standard curve was generated to determine the log value of the capacity factor (log k), and this was performed by plotting the log k value at each isocratic ratio versus the percentage of the organic solvent used in solvent ( Figure S61). This produced a linear expression that allowed for the extrapolation of log k w . Log k w is the measure of lipophilicity; therefore, a greater value corresponds to increased lipophilicity.
The rank of complexes by increasing lipophilicity is 1 < 2 < 3 < 4 < 5 < 6 ( Table 7). It is evident that those complexes containing the smaller axial ligand NPX (1-3) are less lipophilic than those derivatives containing the larger axial ligand ACE (4)(5)(6). This trend also corresponds to the reported lipophilicity values of NPX (2.88) and ACE (4.49) [80]. Moreover, the methylation at the heterocyclic systems of the complexes also influenced lipophilicity, since the 5,6-Me 2 phen derivatives (3 and 6) were more lipophilic than the phen derivatives (1 and 4), and this observation is also comparable with the literature data [53,[55][56][57][58][59]. Overall, the axial coordinated NPX and ACE ligands had more influence on the overall lipophilicity of the complexes than the change in methylation in their heterocyclic ligands.

Preliminary Reduction Studies
Platinum(IV) complexes are expected to maintain their octahedral structure in the bloodstream prior to their activation inside cancer cells [81]. The intracellular activation of platinum(IV) complexes is typically initiated by biological reducing agents such as AsA or GSH, which act as catalysts in the reduction of platinum(IV) complexes to their corresponding active platinum(II) species, along with the release of the axial ligands [34,[81][82][83]. Because of this phenomenon, platinum(IV) complexes are considered prodrugs.
The reduction behaviour of 1-6 was monitored by 1 H-NMR and 1D-195 Pt-NMR spectroscopy using previously established methods [53,55]. Prior to the reduction measurements, each metal complex was dissolved with PBS in D 2 O only. Initial 1 H-NMR and 1D-195 Pt-NMR experiments measured at 400 and −2800 ppm (30 min per region) at 37 • C were performed to confirm the purity of the complexes, and most importantly, to show that the PBS does not affect the structural integrity of the complexes and causes reduction.
Based on the initial 1D-195 Pt-NMR spectra acquired ( Figures S62 and S63), the structural integrity of the complexes was retained, and no signs of reduction were observed, thus also confirming that the PBS had no effect on the complexes. Subsequently, AsA was added in each solution containing the metal complex, PBS and D 2 O. Then, sequential 1 H-NMR experiments were undertaken for 1 h, followed by final 1D-195 Pt-NMR experiments measured within the regions of 400 and −2800 ppm (Figures S64-S75). The approximate time points for 1-6 at which an estimated 50 and 100% reduction had occurred, in the presence of AsA, as represented by T 50% and T 100% , are summarised in Table 8. Table 8. A summary of estimated time points in min for 1-6 at which 50 and 100% reductions proceed, expressed as T 50% and T 100% . In summary, the platinum(IV)-NPX derivatives (1-3) reduced gradually in the presence of AsA compared to the platinum(IV)-ACE derivatives (4-6), which reduced rapidly ( Table 8). The differences in the reduction times are more influenced by the axial ligands rather than by the methylation at the heterocyclic ligands of the complexes. The 1 H-NMR reduction spectra obtained for 1 demonstrated notable movement of resonances originating from the protons of the heterocyclic ligand (phen) and the NPX ligand ( Figure 11). The resonances of H2 and H9 protons had shifted upfield and overlapped with the resonances of H4 and H7 protons, as shown in Figure 11. Moreover, the resonances of H3 and H8 protons shifted upfield, while the resonances of H5 and H6 protons shifted downfield ( Figure 11). Finally, small peaks (with slight overlapping) of the uncoordinated NPX ligand were also recorded between 7.2 and 7.7 ppm, as shown in Figure 11. Immediately after the 1 H-NMR experiments, 1D-195 Pt-NMR experiments were measured on the regions 400 and −2800 ppm. After 40 min from the final 1 H-NMR experiment, there was still a platinum peak detected in the 1D-195 Pt-NMR spectrum of 1 at 400 ppm, as shown in Figure 12. A more prominent platinum peak was recorded at −2800 ppm 30 min after the preceding 1D-195 Pt-NMR experiment, which indicates the presence of the corresponding platinum(II) precursor, PHENSS (Figure 12). The results suggest that complex Immediately after the 1 H-NMR experiments, 1D-195 Pt-NMR experiments were measured on the regions 400 and −2800 ppm. After 40 min from the final 1 H-NMR experiment, there was still a platinum peak detected in the 1D-195 Pt-NMR spectrum of 1 at 400 ppm, as shown in Figure 12. A more prominent platinum peak was recorded at −2800 ppm 30 min after the preceding 1D-195 Pt-NMR experiment, which indicates the presence of the corresponding platinum(II) precursor, PHENSS (Figure 12). The results suggest that complex 1 is a potential prodrug of PHENSS because it can successfully reduce back to its platinum(II) precursor. Immediately after the 1 H-NMR experiments, 1D-195 Pt-NMR experiments were measured on the regions 400 and −2800 ppm. After 40 min from the final 1 H-NMR experiment, there was still a platinum peak detected in the 1D-195 Pt-NMR spectrum of 1 at 400 ppm, as shown in Figure 12. A more prominent platinum peak was recorded at −2800 ppm 30 min after the preceding 1D-195 Pt-NMR experiment, which indicates the presence of the corresponding platinum(II) precursor, PHENSS (Figure 12). The results suggest that complex 1 is a potential prodrug of PHENSS because it can successfully reduce back to its platinum(II) precursor. Furthermore, the changes in chemical shifts observed for 1 were also comparable in the 1 H-NMR and 1D-195 Pt-NMR reduction spectra of complexes 2 and 3 ( Figures S66-S69). Furthermore, the changes in chemical shifts observed for 1 were also comparable in the 1 H-NMR and 1D-195 Pt-NMR reduction spectra of complexes 2 and 3 ( Figures S66-S69). The only difference is that 2 and 3 had completely reduced to their corresponding platinum(II) precursors, 5MESS and 56MESS, respectively, at~1.5 h based on the final 1D-195 Pt-NMR experiment that was measured at 400 ppm, due to the absence of platinum(IV) peaks ( Figures S67 and S69). Additionally, the platinum(IV)-ACE derivatives (4-6) reduced rapidly in the presence of AsA, which only occurred within 10 min according to the acquired 1 H-NMR spectra ( Figures S70, S72 and S74).

Platinum(IV) Complexes
During the experiments, it was observed that upon adding and mixing AsA with the sample solutions of these complexes (4, 5 and 6), the solutions turned cloudy within 30 s. For these complexes, T 50% was within 5 min, as shown in Table 8. Complex 6 had completely reduced at 5 min, while for 4 and 5, full reduction was achieved between 10 and 15 min (Table 8). Overall, the results indicate that 1-6 can successfully reduce to their corresponding platinum(II) precursors; thus, it is correct to conclude that they are prodrugs.

Stability Studies
Additional stability studies for 1-6 were undertaken using 10 mM PBS (~7.4 pH) as the aqueous solution, but without the presence of a reducing agent such as AsA. This was only to show whether the complexes would be stable in aqueous solution for long hours. Each complex was dissolved in PBS and incubated at room temperature and at 37 • C for 36 h. After the incubation, all samples were analysed through HPLC. The platinum(IV) derivatives incorporating NPX (1-3) were relatively stable for 36 h in PBS at room temperature and at 37 • C, as shown in Figure S76. Although for 2, some traces of precursor platinum(IV) complexes and the dissociated NPX ligand were evident ( Figure S76). In contrast, the platinum(IV) derivatives incorporating ACE (4-6) were less stable for 36 h in PBS at room temperature and at 37 • C ( Figure S77), compared with 1-3. After 36 h at room temperature, peak traces of platinum(IV) complexes and the platinum(II) precursors, as well as the dissociated ACE ligand, were present, as shown in Figure S77. Treatment of 4-6 at 37 • C for 36 h resulted in a major reduction of the complexes ( Figure S77). From this, it was reasoned that the larger the ligand conjugated to our platinum(IV) scaffold (i.e., ACE), the more susceptible the resulting derivatives (4)(5)(6) are to reduction, as they reduce even without a reducing agent. This parallels the results from the reduction experiments above, where 4-6 were the quickest to reduce in the presence of AsA. Overall, the stability of the complexes was more influenced by the axial ligands, NPX and ACE, more so than the methylation in their heterocyclic ligands.

Growth Inhibition Studies
The reported platinum(IV) complexes (1)(2)(3)(4)(5)(6) and their axial ligands, NPX and ACE, were evaluated for antiproliferative activity in twelve cell lines including HT29 colon, U87 glioblastoma, MCF7 breast, A2780 ovarian, H460 lung, A431 skin, Du145 prostate, BE2C neuroblastoma, SJG2 glioblastoma, MIA pancreas, ADDP ovarian variant, and the non-tumour-derived MCF10A breast line. Since current platinum(II) drugs are generally used to treat genitourinary cancers such as ovarian cancer, A2780 ovarian and the resistant ADDP ovarian cell lines were selected to probe the capacity of 1-6 to reverse and overcome chemoresistance. Compound growth inhibition was assessed using the MTT assay after 72 h of treatment. All determined GI 50 values are summarised in Table 9, including the GI 50 values of the platinum(II) precursors and platinum(IV) scaffolds, as well as cisplatin, oxaliplatin and carboplatin [53,55], to allow for comparison.
The platinum(II) precursors (PHENSS, 5MESS and 56MESS) and platinum(IV) scaffolds (PHENSS(IV)(OH) 2 , 5MESS(IV)(OH) 2 and 56MESS(IV)(OH) 2 ) utilised in this study to create the studied complexes, 1-6, display exceptional potency, substantially greater than cisplatin, oxaliplatin and carboplatin (Table 9). It was our interest to further enhance the biological activity of these complexes by conjugating the NSAIDs, NPX and ACE, as axial ligands to their cores. Notably, GI 50 values ranging between 0.22 and 2570 nM were recorded for 1-6, as summarised in Table 9. This indicates their superiority as anticancer agents, especially when compared with cisplatin, oxaliplatin and carboplatin.
The calculated mean GI 50 values for 1-6 in the entire cell line tested was 32-931 nM, which is significantly lower than the calculated mean GI 50 values for cisplatin, oxaliplatin and carboplatin (1463-32,242 nM) ( Table 9). The platinum(IV) derivatives incorporating the heterocyclic ligands, 5-Mephen (2 and 5) and 5,6-Me 2 phen (3 and 6) were more growth inhibitory than those incorporating the phen ligand (1 and 4), as expected ( Table 9). The trend observed here confirms once again that the methylation of the heterocyclic ligand influences biological activity, and this has been translated in previous studies [16,24,53,[55][56][57][58][59]66]. Furthermore, the platinum(IV) derivatives incorporating the NPX ligand (1-3) were less potent than those platinum(IV) derivatives incorporating the ACE ligand (4)(5)(6). This is likely due to the response of NPX and ACE intracellularly, despite having the same GI 50 values of >50,000 nM (Table 9). Nonetheless, this may also suggest that the ACE ligand has more synergism with platinum than the NPX ligand.
The selectivity cytotoxicity index (SCI) for the metal complexes and compounds (NPX and ACE) was also calculated by dividing the GI 50 values of the complexes or compounds in MCF10A (non-tumour) by their GI 50 in Du145, which is reported in Table 9. The SCI gives a measure of the efficacy of the drug against cancer cells: the greater the SCI the higher the selectivity towards cancer cells [61,62]. The high SCI exhibited by 5 and 6 towards the Du145 cell line of 15.4 and 15.5 are notable (Table 9). In contrast, the platinum(II) precursors (PHENSS, 5MESS and 56MESS), platinum(IV) scaffolds (PHENSS(IV)(OH) 2 , 5MESS(IV)(OH) 2 and 56MESS(IV)(OH) 2 ), complexes 1-4, cisplatin, carboplatin, NPX and ACE showed little selectivity towards Du145, with SCI values ranging between 1.00 and 5.48. From this, it can be inferred that the coordination of the ACE ligand to the platinum(IV) scaffolds 5MESS(IV)(OH) 2 and 56MESS(IV)(OH) 2 had significantly improved the selectivity of the resultant platinum(IV) derivatives 5 and 6 towards prostate cancer cell populations.
Furthermore, platinum(IV) complexes utilising the cores of platinum(II) drugs incorporating NPX as axial ligands have also been investigated for their anticancer properties, most specifically in the MCF7 cell line [37,72]. Tolan et al. synthesised cisplatin-NPX, oxaliplatin-NPX and carboplatin-NPX complexes, which demonstrated GI 50 values of 10,400 ± 790, 9470 ± 750 and 9120 ± 630 nM, respectively, in the MCF7 cell line [72]. These values are still substantially higher than the GI 50 values elicited by our platinum(IV) complexes containing NPX (1)(2)(3) in that cell line (Table 9). For example, 3 (60 ± 0.5 nM) proved to be 173-fold more potent than cisplatin-NPX (10,400 ± 790 nM), 158-fold more potent than oxaliplatin-NPX (9470 ± 750 nM), and 152-fold more potent than carboplatin-NPX (9120 ± 630 nM) in the MCF7 cell line. It is only appropriate to mention that these differences in cytotoxicity are mostly attributed to the biological activity of the platinum(II) cores of these complexes, considering that our platinum(II) scaffolds (PHENSS, 5MESS and 56MESS) are significantly more potent than cisplatin, oxaliplatin and carboplatin.

ROS Potential
ROS is a family of highly reactive chemicals generated in the mitochondria, which are also natural by-products of cellular metabolic processes [84]. The regulation of ROS is critical for normal biological functions; however, if compromised by stress factors (i.e., chemotherapy drugs), it can lead to oxidative stress [85]. Oxidative stress reflects the imbalance between ROS production and the ability of a biological system to detoxify these. Clearly, the abundant accumulation of ROS has deleterious side effects on DNA, proteins, lipids, and other cellular components that are vital for life. While this phenomenon is detrimental to normal healthy cells, this can be applied as a treatment strategy against cancer cells, as conveyed in the literature [86][87][88][89]. Most importantly, there are two roles of ROS activity in cancer. With the right amount of ROS intracellularly, it can promote metastasis and help cancer cells acquire resistance to treatment [90]. On the contrary, if the ROS levels are significantly elevated and prolonged, cytotoxic effects would transpire and therefore induce apoptosis and potentially inhibit the resistance mechanisms of cancer cells to treatment [91].
In this study, the ROS activity of 1-6, together with their axial ligands (NPX and ACE), and cisplatin in the human colon cancer cell line, HT29 at 24, 48 and 72 h are reported (Table 10, Figures 13 and S78). A summary of measured fluorescence in RFU units for the complexes and ligands is summarised in Table 10. Additionally, the ROS potential of the platinum(II) precursors and platinum(IV) scaffolds was determined previously [53,55] and was presented here to allow for comparison (Table 10 and Figure 13). The HT29 colon cell line was selected for testing, as it is one of the cell lines wherein 1-6 demonstrated exceptional in vitro cytotoxicity (Table 9). Treatment with the complexes and ligands was completed at each specific GI 50 concentration.  (Table 10, Figures 13 and S78). A summary of measured fluorescence in RFU units for the complexes and ligands is summarised in Table 10. Additionally, the ROS potential of the platinum(II) precursors and platinum(IV) scaffolds was determined previously [53,55] and was presented here to allow for comparison (Table 10 and Figure 13). The HT29 colon cell line was selected for testing, as it is one of the cell lines wherein 1-6 demonstrated exceptional in vitro cytotoxicity (Table 9). Treatment with the complexes and ligands was completed at each specific GI50 concentration.   A significant increase in ROS production was detected for 1-6 and the axial ligands, NPX and ACE, in HT29 colon cells at 24 and 48 h, relative to the control (Table 10 and Figure 13). This increase also corresponds to the ROS activity exhibited by the platinum(II) precursors and platinum(IV) scaffolds, and cisplatin, as shown in Table 10 and Figure 13. Notably, the RFU values of 1-6 were higher compared with the RFU values of their corresponding platinum(II) precursors and platinum(IV) scaffolds at 24 and 48 h (Table 10 and Figure 13). This confirms that the conjugation of NPX and ACE to our platinum(IV) scaffolds enhanced ROS production. A consistent decline in ROS activity was demonstrated by 1-6 and the NPX and ACE ligands, particularly the drastic decreased ROS activity at 72 h (Table 10 and Figure 13). For the platinum(IV) derivatives incorporating NPX, 1-3, their RFU values at 72 h (1: 63 RFU; 2: 84 RFU; 3: 104 RFU) were lower than the RFU value of the control (131 RFU) (Table 10 and Figure 13). While this trend also paralleled the ROS activity of the platinum(IV) derivatives incorporating ACE, 4-6, their RFU values at 72 h (4: 168 RFU; 5: 192 RFU; 6: 202 RFU) were still higher than the control (131 RFU) (Table 10 and Figure 13). Nonetheless, this consistent decline in ROS production observed for 1-6 and NPX and ACE, was not demonstrated by cisplatin, platinum(II) precursors or platinum(IV) scaffolds, since the ROS activity of these complexes remained significantly high at 24, 48 and 72 h, relative to the control (Table 10).
We conclude that the initial increase in ROS production at 24 h, followed by the decline in ROS activity at 48 and 72 h, exhibited by 1-6, is reflective of the pro-oxidant and antioxidant effects of the coordinated axial ligands (or NSAIDs), NPX and ACE, intracellularly towards ROS. There is sufficient evidence suggesting that NSAIDs can both effectively induce ROS production and scavenge ROS [92][93][94][95]. The results acquired from this study indicate that the studied platinum(IV) complexes, 1-6, can promote cellular damage towards cancer cells by production of ROS. Table 10. Production of ROS upon treatment with 1-6, NPX, ACE, and cisplatin in HT29 colon cells at 24, 48 and 72 h. The data presented for platinum(II) precursors (PHENSS, 5MESS and 56MESS) and platinum(IV) scaffolds (PHENSS(IV)(OH) 2 , 5MESS(IV)(OH) 2 and 56MESS(IV)(OH) 2 ) were obtained from previous studies [53,55].

Mitochondrial Membrane Potential (MtMP)
Mitochondria is the powerhouse of the cell, and most significantly, the main driver in creating energy for metabolic processes to sustain normal cellular functions [96]. Exogenous stressors such as chemotherapy drugs (and other metallodrugs) are known to induce bioenergetic stress in the mitochondria and cause MtMP disruption, which then activates the mitochondrial apoptotic pathway [21,[97][98][99][100][101]. MtMP is a critical predictor of mitochondrial activity that reflects the electrical potential difference between the intracellular and extracellular environment of a cell [102]. Loss in MtMP is indicative of bioenergetic stress that may result in apoptosis.
Here, we report the MtMP changes upon treatment of 1-6, NPX, ACE, the platinum(II) precursors (PHENSS, 5MESS and 56MESS), platinum(IV) scaffolds (PHENSS(IV)(OH) 2 , 5MESS(IV)(OH) 2 and 56MESS(IV)(OH) 2 ), and cisplatin in HT29 colon cells at 24, 48 and 72 h (Table 11 and Figure 14). A summary of measured fluorescence in RFU units for the complexes and ligands is summarised in Table 11. TMRE was used to stain HT29 colon cells, and treatment with the complexes and ligands was completed at each specific GI 50 concentration. The MtMP changes detected are expressed as RFU. Additionally, lower RFU values correspond to a decline in mitochondrial activity.   and progressive cell dysfunction. This observation is also proof that there is a correlation between ROS production and MtMP changes (Figures 13 and 14) [98,103,104]. The findings indicated the ability of 1-6 to cause MtMP changes, by reducing the mitochondrial activity up to 72 h in HT29 colon cells. Although, it is noteworthy that the MtMP changes detected for the complexes were mostly influenced by their corresponding platinum(II) precursors (PHENSS, 5MESS and 56MESS) rather than by the released axial ligands, NPX and ACE. From the results, a decrease in mitochondrial activity was detected upon treatment of cisplatin, the platinum(II) precursors (PHENSS, 5MESS and 56MESS) and platinum(IV) scaffolds (PHENSS(IV)(OH) 2 , 5MESS(IV)(OH) 2 and 56MESS(IV)(OH) 2 ), the studied platinum(IV) complexes, 1-6, and the axial ligands, NPX and ACE, up to 72 h (Table 11 and Figure 14). Particularly for 1-6, the decrease in mitochondrial activity for the duration of the treatment followed the same trend that was observed in the ROS activity detected for these complexes (Table 10, Figures 13 and S78). This may suggest that the decline in mitochondrial activity is due to the accumulation of ROS induced by the complexes, considering that increased ROS production can impair mitochondrial functions and progressive cell dysfunction. This observation is also proof that there is a correlation between ROS production and MtMP changes (Figures 13 and 14) [98,103,104]. The findings indicated the ability of 1-6 to cause MtMP changes, by reducing the mitochondrial activity up to 72 h in HT29 colon cells. Although, it is noteworthy that the MtMP changes detected for the complexes were mostly influenced by their corresponding platinum(II) precursors (PHENSS, 5MESS and 56MESS) rather than by the released axial ligands, NPX and ACE.

Cyclooxygenase-2 (COX) Inhibition
NSAIDs are well recognised as inhibitors of the COX enzymes, COX-1 and COX-2 [37,42,45]. While both COX-1 and COX-2 are responsible in prostaglandin production, they are distinct from one another in terms of their tissue distribution and regulation of gene expression [49,105]. COX-1 is typically expressed in most tissues and is associated with normal physiological functions, particularly by maintaining prostaglandin production at basal levels [49,50]. In contrast, COX-2 is involved in the excessive production of prostaglandin during inflammation (or hypoxia), and extensive studies have shown that COX-2 resultant prostaglandins are overexpressed in tumours [49,50]. Most importantly, the overexpression of COX-2 promotes tumour development and progression [51,106]; therefore, COX-2 is a more valuable target than COX-1. To explore the COX-2 inhibitory properties of the studied complexes (1-6), ELISA was used against the COX-2 enzyme. Here, we report the % of COX-2 inhibitory activity of 1-6, NPX, ACE, the platinum(II) precursors (PHENSS, 5MESS and 56MESS) and cisplatin at 72 h (Table 12, Figures 15 and S79). A summary of the measured COX-2 percentage inhibition for the complexes and ligands is outlined in Table 12.  The greatest COX-2 inhibitory activity was demonstrated by the COX-inhibiting ligands, NPX (83%) and ACE (76%) ( Table 12 and Figure 15). In contrast, the platinum(II) precursors (PHENSS, 5MESS and 56MESS) and cisplatin showed the lowest inhibitory response towards COX-2 (Table 12 and Figure 15), with the COX-2 inhibition ranging from only 5-9%. Upon coordination of NPX and ACE to our platinum(IV) scaffolds, the COX-2 inhibition for 1-6 improved significantly, when compared with the COX-2 inhibition of PHENSS, 5MESS, 56MESS and cisplatin (Table 12 and Figure 15). However, relative to the significant COX-2 inhibition demonstrated by NPX (83%) and ACE (76%), the COX-2 inhibitory activity of 1-6 was comparatively lower (Table 12 and Figure 15). This may be dependent on the reductive state of the complexes 1-6 in vitro and the ability to release COX-inhibiting ligands. Furthermore, the platinum(IV) derivatives of PHENSS and 5MESS (1, 2, 4 and 5) elicited more inhibitory activity towards COX-2 compared to the The greatest COX-2 inhibitory activity was demonstrated by the COX-inhibiting ligands, NPX (83%) and ACE (76%) ( Table 12 and Figure 15). In contrast, the platinum(II) precursors (PHENSS, 5MESS and 56MESS) and cisplatin showed the lowest inhibitory response towards COX-2 (Table 12 and Figure 15), with the COX-2 inhibition ranging from only 5-9%. Upon coordination of NPX and ACE to our platinum(IV) scaffolds, the COX-2 inhibition for 1-6 improved significantly, when compared with the COX-2 inhibition of PHENSS, 5MESS, 56MESS and cisplatin (Table 12 and Figure 15). However, relative to the significant COX-2 inhibition demonstrated by NPX (83%) and ACE (76%), the COX-2 inhibitory activity of 1-6 was comparatively lower (Table 12 and Figure 15). This may be dependent on the reductive state of the complexes 1-6 in vitro and the ability to release COX-inhibiting ligands. Furthermore, the platinum(IV) derivatives of PHENSS and 5MESS (1, 2, 4 and 5) elicited more inhibitory activity towards COX-2 compared to the platinum(IV) derivatives of 56MESS (3 and 6). It is also evident that COX-2 inhibition of the platinum(IV)-NPX derivatives (1-3) was slightly greater than for those derivatives containing ACE (4-6) ( Table 12 and Figure 15). The results suggest that COX-2 inhibition is independent of the cytotoxicity and lipophilicity of the complexes. The most cytotoxic and lipophilic complexes contain the ACE ligand (4-6), which exhibited a reduced response towards inhibiting COX-2, while the least cytotoxic and lipophilic complexes incorporating NPX (1-3) exhibited a marginally improved response towards inhibiting COX-2. These findings confirm that the coordination of the NSAIDs, NPX and ACE, to our platinum(IV) scaffolds generated complexes that are capable of inhibiting COX-2, may potentially be effective in reducing tumour-related inflammation, and may also show selectivity towards tumours that overexpress COX-2 resultant prostaglandins.

Conclusions
Six mono-substituted platinum(IV) complexes (1-6) incorporating the NSAIDs, NPX and ACE, were synthesised, characterised and evaluated for their antitumour activity. A range of spectroscopic and spectrometric techniques confirmed the composition and homogeneity of the complexes. Solubility tests were performed at room temperature in d.i.H 2 O. Complexes 1-6 were found to be more soluble than cisplatin, oxaliplatin and carboplatin. 1 H-NMR and 1D-195 Pt-NMR provided insight on the reduction properties of the complexes in PBS (~7.4 pH) and AsA at 37 • C. Complexes 1-6 were reduced to their corresponding platinum(II) precursors and released the axial ligands, indicating their potential as prodrugs. Those derivatives incorporating the smaller ligand NPX (1-3) reduced slower, while those derivatives incorporating the larger ligand ACE (4-6) reduced rapidly in the presence of AsA. Stability and lipophilicity measurements were also undertaken using HPLC. Complexes 1-3 were stable in PBS at room temperature and at 37 • C for 36 h, unlike 4-6 which exhibited major reduction. According to the log k w values, 4-6 proved to be more lipophilic than 1-3.
The in vitro cytotoxicity of 1-6 in multiple cell lines was assessed via the MTT assay. The most potent platinum(IV) derivatives, 5 and 6, which incorporated ACE, elicited the lowest GI 50 values. Complexes 5 and 6 were found to be the most lipophilic complexes, and their high lipophilicity may have directly contributed to their exceptional potency. All the synthesised platinum(IV) complexes demonstrated exceptional in vitro activity compared with conventional chemotherapeutics, such as cisplatin, oxaliplatin and oxaliplatin. Despite the notable potency exhibited by 1-6, they are not entirely selective towards cancer cells because they were also found to be toxic in the normal breast cancer cell line, MCF10A. Nevertheless, 1-6 may overcome drug resistance mechanisms associated with the cisplatin-resistant ADDP ovarian variant cell line. Complexes 1-6 also demonstrated both antioxidant and pro-oxidant effects, which were predominantly influenced by the intracellular release of the axial ligands, NPX and ACE, in the HT29 colon cells. The impact of the increased accumulation of ROS exhibited by the complexes at 24 h was observed in the mitochondrial activity, with the MtMP changes being significantly reduced. Furthermore, the COX-2 inhibitory activity of 1-6 was also assessed, with the results attesting that these complexes can reduce tumour-related inflammation, despite the lack of correlation between COX-2 inhibition and in vitro cytotoxicity and lipophilicity. Finally, cellular uptake studies for the most potent platinum(IV) complexes, 5 and 6, are scheduled in the near future to further confirm whether enhanced cytotoxicity correlates with increased cellular uptake.