Exploiting azide-alkyne click chemistry in the synthesis, tracking and targeting of platinum anticancer complexes.

Click chemistry is fundamentally important to medicinal chemistry and chemical biology. It represents a powerful and versatile tool, which can be exploited to develop novel Pt-based anticancer drugs and to better understand the biological effects of Pt-based anticancer drugs at a cellular level. Innovative azide–alkyne cycloaddition–based approaches are being used to functionalise Pt-based complexes with biomolecules to enhance tumour targeting. Valuable information in relation to the mechanisms of action and resistance of Pt-based drugs is also being revealed through click-based detection, isolation and tracking of Pt drug surrogates in biological and cellular environments. Although less well-explored, inorganic Pt-click reactions enable synthesis of novel (potentially multimetallic) Pt complexes and provide plausible routes to introduce functional groups and monitoring Pt-azido drug localisation.


Introduction
Of the cancer patients who are treated with chemotherapy, around 50% receive a Pt(II)-based medicine such as cisplatin, carboplatin or oxaliplatin (Figure 1a) [1]. The primary mechanism of action of cisplatin and carboplatin results from their ability to cross-link nuclear DNA; the Pt-DNA adducts interrupt transcription, generate DNA damage responses and ultimately induce apoptosis. Pt(II) anticancer drugs also react with a range of other nucleophiles, including RNA, mitochondrial DNA and proteins [2,3]. Oxaliplatin is used clinically to treat stage III colorectal cancer and exhibits a fundamentally different cytotoxic profile to cisplatin and carboplatin. Although DNA platination does occur, other effects including induction of immunogenic cell death [1,4] and ribosome biogenesis stress are thought to dominate the anticancer mechanism of action of oxaliplatin [5].
The clinical effectiveness of Pt anticancer agents is hampered by toxic side effects and both intrinsic and acquired resistance [6]. Therefore, there has been a continued drive to develop novel classes of more effective and better-tolerated Pt(II) and Pt(IV) drug candidates, as well as to better understand the precise cellular and immunological effects of Pt complexes [7]. The development of innovative techniques to synthesise, label and track Pt(II)-and Pt(IV)-based complexes is anticipated to greatly aid this enterprise. Click chemistry is widely used throughout synthetic chemistry and biology, showing tremendous versatility, whilst being atom efficient and in some cases, bio-orthogonal.
The Cu(I)-catalysed [3þ2] azideealkyne cycloaddition (CuAAC), Figure 1b, is synonymous with click chemistry [8,9]. Reaction of an azide with a terminal alkyne generates the corresponding 1,4-disubstituted 1,2,3triazole with excellent selectivity and in high yield. CuAAC has been successfully and routinely used in the syntheses of 1,2,3, triazoles,[10*] many of which have interesting biomedical applications [11e13]. Triazoles are attractive pharmacophores which can potentially intercalate, participate in hydrogen bonding and can act in some respects as a substitute for amides [14]. CuAAC has also been used to develop triazole-based ligands [10*,15], and for chemical conjugations including for labelling in biological systems, though reactions in living systems are limited by Cu(I)-associated toxicity [16].
The widespread popularity of click chemistry is certain. This article will outline, Figure 1c, how click chemistry, specifically azideealkyne cycloaddition, is an important chemical tool for: the functionalisation of Pt-based drug candidates the development of trackable Pt drugs the development of Pt triazole drug candidates.

Functionalisation
Click chemistry represents an excellent conjugation strategy for the functionalisation of Pt complexes with, for example, targeting moieties/delivery systems, fluorescent reporters and secondary chemotherapeutics, Figure 2a. A variety of Pt(II) and Pt(IV) click templates have been developed to date, which possess alkyne or azide ligand-based click handles, Figure 2b.

Targeting agents
Selective coupling of chemotherapeutics to human serum albumin (HSA), the most abundant blood serum protein, has proved to be a successful strategy for enhanced tumour targeting because HSA can act as a long-circulating delivery vehicle which accumulates passively in the tumour tissue. Yao Figure 2c. The Pt NIR-AZA conjugate was nontoxic and dispersed relatively uniformly throughout the cytoplasm. Inevitably, as for any derivative the properties associated with the preclicked Ptefluorophore conjugate were strongly influenced by the physical and chemical properties of the organic fluorophore that impact uptake and guide localisation of the complex. It is however still valuable to be able to compare and contrast the proprties of the trackable derivatives with the original drugs in terms of global properties (uptake, efflux etc.), and DNA adduct formation and metabolic processing. There is also the potential for the trackable derivative to be developed as a novel drug molecule in its own right, if the biological properties are sufficiently promising.

Trackable Pt-based drugs
Click chemistry has a very important role to play in the isolation of Pt-bound biomolecules and detection of Ptbased drugs in a biological setting. In particular, DeRose and Bierbach have pioneered robust and efficient CuAAC-and SPAAC-based methods (i) for posttreatment analysis of Pt biomolecular interactions, [25,26,29,30**] and (ii) to track Pt-based drugs in cells [21,22]. Numerous clickable fluorescent dyes featuring a variety of click partners including azide, alkyne, dibenzocyclooctyne and bicyclononyne are commercially available. A thorough understanding of molecular properties, the uptake, cellular localisation preferences and efflux of the selected complementary fluorophore click derivative are prerequisites for any study.

Post-treatment fluorescent analysis
CuAAC and SPAAC clickederivatised Pt(II) complexes have been successfully used for the convenient detection of Pt(II)eDNA, Pt(II)eRNA and Pt(II)eprotein   Figure 3a.
Significantly recently azidoplatin, Figure 2b (5), was used in a novel chemical proteomic method to label and isolate platinated proteins in S. cerevisiae.[30**] One hundred and fifty two Pt(II)-bound proteins were identified including multiple proteins implicated in the ER stress response,[30**] highlighting the importance of postlabelling/chemical proteomic strategies in identifying biomolecules implicated in the mechanisms of on-or off-target activity of Pt(II)-based drugs.
CuAAC and SPAAC have been validated as important tools for post-treatment labelling of Pt(II) biomolecular adducts. Future work is likely to focus on labelling Pt adducts extracted from cancer cells, particularly involving high-throughput studies to isolate and identify important Pt(II) protein adducts and related celledeath pathways.

Track and analyse
In cellulo fluorescent detection of a Pt drug, using confocal microscopy, facilitates tracking of the drug, providing vital information concerning uptake, cellular transport, subcellular/organellular localisation and efflux. Fluorescent detection can be undertaken through in cellulo postbinding bio-orthogonal ligation strategies, Figure 3c.
Use of Pt-based drug derivatives which possess relatively innocent bio-orthogonal click handles tethered to the stable amine carrier ligands enables retention of the essential chemical and biological properties of the parent drug.
The use of click chemistry (CuAAC and SPAAC) has been shown to be a powerful method for mapping the subcellular localisation of post labelled Pt drug surrogates in fixed cancer cells [32,33], including the cell cycle specific localisation [22]. Nonetheless, future developments in the tracking of Pt-based drugs are likely to focus on real-time tracking of Pt click templates in live human cancer cells using Cu-free click reactions that initiate a turn-on fluorescence. Challenges involve the solubility of SPAAC-based fluorophores and the innocence and stability of click handles.

Pt-azido-based azide-alkyne cycloadditions
The term 'iClick' describes cycloaddition reactions between metal azide and metal acetylene groups [34] and has recently been broadened to include the cycloaddition of organic substrates with either a metal azide or metal acetylide [35]. Pt(II) azides undergo cycloadditions with a range of functional groups (e.g. acetylides, isocyanides, isonitriles, nitriles, carbon disulphides and isothiocyanates) [36]. Here, we focus on cycloadditions between azides and acetylene compounds where at least one group is directly Pt bound, reflecting on the potential biological applications of this reactivity.
These reactions enable synthesis of novel complexes and the assembly of multimetallic architectures and polymers. They also provide potential routes for late-stage introduction of sensitive functional groups to Pt complexes and the opportunity for monitoring uptake and subcellular distribution of Pt-azido drugs in cellulo.
Successful Pt-azidoealkyne cycloadditions with organic alkynes typically involve internal alkynes (RChCR) because terminal alkynes (HChCR) can undergo azideeacetylene ligand exchange at the metal [37]. For Pt-azidoealkyne cycloadditions to occur, activation of the alkyne with a catalyst or with electron-withdrawing groups or strain promotion (SPAAC) is necessary.

Cu-and Au-mediated azide-acetylene cycloaddition reactions
Typical advantages of CuAAC reactions are the rapid rate of reaction, high yields and the regiospecificity of the resulting product (Figure 4a i,ii). Examples include an in-chain polymerisation of an N 3 -organic spacer-Pt-CChR type monomer, giving a novel class of metallopolytriazolates (Figure 4a iii) [38]. Au(I) is believed to play a similar role to Cu(I) in traditional CuAAC [38]; if a Aueacetylide is used, cycloaddition with a Pt-azide can proceed in good yield in the absence of a Cu catalyst (Figure 4a iv). For example, reaction of cis-[Pt II (PPh 3 ) 2 (N 3 ) 2 ] with Au I PPh 3 (ChCeC 6 H 4 NO 2 ) affords the corresponding Pt(II)/Au(I) heterotrimetallic complex 9, with two Pt(II) triazole ligands each coordinating a Au(I) phosphine in 92% isolated yield (Figure 4a v) [34]; the PEt 3 derivative 10 which bears some similarity to auranofin d was also synthesised (obtained in 60% yield) [39].

Electronic-and strain-promoted azide-acetylene click reactions
An advantage of electronic-and strain-promoted methods is biocompatibility; disadvantages include extended reaction times, a restriction on the acetylenes which can be used and if the alkyne is asymmetrical d formation of a mixture of regioisomers.

Pt(II) complexes
Although a considerable number of Pt(II) azido cycloadditions have been reported [36], biological investigations of the resultant Pt-triazoles are scarce. A recent example is complex 11 (Figure 4b), which demonstrated an IC 50 value of 2.4 AE 0.1 mM as compared with 0.9 AE 0.1 mM for cisplatin (GaMG cell line, 72 h contact) [35].

Pt(IV) complexes
Judicious ligand choice enables Pt(IV) azides to be nontoxic in the dark; irradiation with visible light induces reduction to Pt(II) species and release of N 3 and OH radicals and so on causing potent cytotoxicity in cancer cells. This concept forms the basis for their ongoing development as photochemotherapeutics.[40*] The first reported cycloadditions of a Pt(IV) azido complex explored the reactivity of complex 14, Figure 4b, with a range of electron-deficient or strain-promoted acetylenes. Pt(IV) azides are less reactive to azideeacetylene cycloaddition with electron-deficient alkynes than their Pt(II) counterparts, attributed to the higher oxidation state of Pt(IV) which reduces the electron density in the azide ligand. Both Pt(IV) triazolato monoazido and Pt(IV) bis triazolato complexes were synthesised through reaction with electron-deficient (1,4-diphenyl-2-butyne-1,4-dione, dimethyl acetylenedicarboxylate) and strained acetylenes (bicyclo[6.1.0]non-4-yn-9ylmethyloxycarbonyl]-1,8-diamino-3,6-dioxaoctane and dibenzocyclooctyne-amine (16-mono and 16-bis) [41].
For Pt(IV) triazoles synthesised in this way, there is the potential for additional reactivity with the triazole ligand through the axial (OH) ligand. For complex 15 a reversible interaction with the CO group of the triazole was observed in MeCN (complex 15a/15b) [42]. More recently, studies of the di-Pt(IV) triazole complex 17-[N1,N3] (Figure 4b) demonstrated that irradiation with visible light (l irr 452 nm) in the presence of 5 0 -GMP results in the formation of new Pt(IV) and Pt(II) species as well as radical species (N 3 , OH ), in H 2 O and cellfree lysate, confirming that monoazido Pt(IV) complexes retain their potential for photocytotoxicity, and click chemistry therefore opens a wide range of latestage derivation possibilities [43**].
Post-click reactivity of metal-triazole complexes Whilst Pt(II) triazoles undergo solvent-dependent N1e N2 isomerisation [35], this has not been observed for Pt(IV)eN1-coordinated triazoles, presumably because of increased kinetic inertness. Different synthetic routes can therefore be envisaged (Figure 4b).

Conclusions
Azideealkyne organic click reactions have been used to prederivatise the ligands of Pt complexes, to include targeting agents and fluorophores. A series of Pt complexes containing ligand-based azide or alkyne click handles have also been validated as important templates for functionalisation with complementary partners. Click chemistryeenabled post-treatment labelling of Pt-bound biomolecules (mtDNA, rRNA, tRNA and proteins) and in cellulo tracking of Pt drug surrogates has helped highlight the importance of such approaches in establishing non-nuclear and cytoplasmic effects of Pt-based drugs. The speed of innovations in bio-orthogonal chemistry is anticipated to improve sensitivity and facilitate the real-time tracking of Pt drug surrogates in cellulo. The biological application of cycloadditions involving azides or acetylenes coordinated directly to Pt is in its infancy and ripe for development.

Conflict of interest statement
Nothing declared.