Photoactivatable platinum anticancer complex can generate tryptophan radicals† †Electronic supplementary information (ESI) available: Experimental procedures and additional results from photoactivation studies. See DOI: 10.1039/c8cc06496b

l-Tryptophan (Trp), melatonin (MLT) and the Trp-peptide pentagastrin quenched the formation of azidyl radicals generated on irradiation of the anticancer complex trans,trans,trans-[Pt(pyridine)2(N3)2(OH)2] with visible light, giving rise to C3-centred indole radicals which were characterized for Trp and MLT using an EPR spin-trap.


Supplementary Tables
Sample tubes were then positioned in the EPR cavity so that the sample solution filled the entire length of the cavity. Sample preparation was done under dim controlled lightning conditions and transfer to the EPR spectrometer was in the dark to prevent the photoactivation of complex 1 prior to the beginning of the experiment.

Quantification of spin adducts
The quantification of the spin adducts was performed by using a calibration curve obtained from standard solutions of 4-hydroxyl-2,2,6,6-tetramethyl-piperidine-1-oxyl (TEMPOL), whose concentrations were checked by optical absorption as previously reported. 2 The EPR spectrum of each solution of TEMPOL was acquired. The spectra were baseline corrected and simulated with EasySpin, 3 with double integration performed on the simulated spectra.

EPR spectroscopy
All EPR spectra were recorded on an X-band Bruker EMX CW EPR spectrometer at ambient temperature (ca. 295 K) using a TM110 cavity (ER 4103TM). A 2,2-diphenyl-1-picrylhydrazyl (DPPH) standard was used for calibration of the g-factor. Sweep time was approximately 13 s per scan and modulation depth was set to 0.1 mT.

Irradiation
The LED was inserted at the end of a plastic tube which was clamped to a support. The TM110 cavity is equipped with a grid on the front to allow optical access to the sample (ca. 80% transmission). The tube was therefore placed in contact with the grid of the EPR cavity in order to convey all the light into it (Fig. S 1). In this work, the position of the LED was maintained throughout all the irradiation experiments. The LED was connected to a current generator, which was switched on at the beginning of the irradiation. Either a 465 nm blue light (LED465E, Thorlabs, FWHM 25 nm) or a 525 nm green light (LED528E, Thorlabs, FWHM 40 nm) LED were used for irradiating the samples. Under the operating conditions used the radiation power was measured with a power meter and was found to be 7.1 mW cm −2 for the 465 nm diode and 5.4 mW cm −2 for the 525 nm diode.

EPR simulations
EPR spectral simulations were performed in Matlab using the EasySpin package. 3 The garlic routine (appropriate for the fast-motional regime) was used for all the experiments. Spectral parameters were determined by using EasySpin's esfit routine with the Nelder-Mead simplex algorithm.
Simulations of the MNP-tryptophan (MNP-Trp) and the MNP-melatonin (MNP-MLT) spin adducts were performed by including only the hyperfine coupling arising from the nitroxidic nitrogen, which was considered to be fully isotropic. Simulation of the MNP-α-hydroxy-ethyl adduct was performed including the hyperfine couplings of both the nitroxidic nitrogen and the α-proton. Simulation of the DMPO-N3 nitrone spin adduct was performed considering couplings to the nitroxidic nitrogen, the β-proton of the spin trapping agent and the αnitrogen of the trapped azidyl radical.
EPR parameters of the MNP di-adduct di-tert-butyl nitroxide (DTBN) were obtained by fitting a spectrum acquired from a solution of MNP which had been illuminated overnight with the 465 nm LED, in order to promote the formation of DTBN. Hyperfine couplings arising from both the nitroxidic nitrogen and statistical abundance of nearest neighbour 13 C nuclei were included and considered to be fully isotropic.
An isotropic g-tensor was used for all the simulations and dynamic effects were neglected.

Supplementary Tables
Table S1 EPR hyperfine couplings (mT) and g-values for the trapped L-Trp radical (MNP-Trp) and the MNP di-adduct di-tert-butyl nitroxide (DTBN), comparing the parameters determined in this study with those previously published.