Theoretical insight into joint photodynamic action of a gold(I) complex and a BODIPY chromophore for singlet oxygen generation

Bruna C. De Simone a, Gloria Mazzone *a, Wichien Sang-aroon b, Tiziana Marino a, Nino Russo a and Emilia Sicilia a
aDipartimento di Chimica e Tecnologie Chimiche, Università della Calabria, Rende, Italy. E-mail: gloria.mazzone@unical.it
bDepartment of Chemistry, Faculty of Engineering, Rajamangala University of Technology Isan, Khon Kaen, Thailand

Received 31st July 2018 , Accepted 29th August 2018

First published on 29th August 2018


Density functional theory is herein employed to provide theoretical insight into the mechanism involved in 1O2 photosensitization by a gold–BODIPY combined complex proposed as a promising photodynamic therapy agent. The protocol is thus used to compute the non-radiative rate constants for the S1 → Tj intersystem crossing transitions. Calculations show that while the incorporation of an iodine atom into the core skeleton of BODIPY enhances the singlet–triplet intersystem crossing (ISC) efficiency due to the occurrence of the singlet–triplet transition between states with different orbital characters (ππ* → πn*), the presence of a gold atom, even if not directly anchored to the chromophore core but through a triplet bond, equally entails an increase of the spin–orbit coupling constant due to the heavy atom effect. In this way, the system is able to generate singlet molecular oxygen, the key cytotoxic agent in PDT. Our results fit well with the experimental singlet oxygen quantum yield and cytotoxicity determination.


Introduction

Alongside the increasingly high use of photodynamic therapy (PDT) in anti-cancer therapy1–4 and other diseases,5,6 the search for new and more efficient photosensitizers has grown significantly in the last few years.7–10 The photophysical mechanism on which PDT is based involves the excitation of a chromophore, a photosensitizer (PS), from the ground S0 to a singlet excited state that can transfer its energy, throughout an intersystem crossing (ISC) process, to a triplet state lying below. The subsequent transfer of energy from the triplet state of the PS to the molecular oxygen present in the irradiated tissues triggers the so-called type II photoreactions that lead to the production of one of the most active cytotoxic agents, 1O2, which immediately begins to destroy the surrounding sick tissues. Drugs for clinical PDT use must fulfill a series of indispensable requirements.9 The most important requirements can be summarized as follows: be non-toxic in the absence of light radiation, generate reasonable amounts of cytotoxic singlet molecular oxygen upon irradiation, be easily removed from the body, possess thermodynamic stability, not be subject to redox reactions, and absorb in the NIR region. Thus, from a photophysical point of view, the factors that are important to take into account in designing a new and effective photosensitizer for PDT include (i) the absorption wavelength, which has to fall within the therapeutic window (500 and 850 nm) so that the radiation can penetrate as deeply as possible into the tissues and, consequently, reduce the number of clinical treatments; (ii) the triplet excited-state energy, which has to be greater than the energy required to convert O2 from a fundamental state (3Σg) to an excited one (1Δg) (0.98 eV);11 and (iii) the amplitude of the spin–orbit relativistic coupling (SOC) between the two states with different spin multiplicities, whose values are mainly responsible for an efficient kinetics for the ISC process, especially when heavy atoms are included in the molecular structures.

A few photosensitizers are currently used in the medical practice of PDT and they are essentially porphyrin derivatives, sometimes complexed with a metal that contributes to enhancing the ISC efficiency.7 In recent years, new photosensitizers with different chemical structures, such as squaraine,12 4,4-difluoro-4-bora-3a,4a-diaza-s-indacene (BODIPY)13 and its derivatives, aza-BODIPY14 and BOIMPY,15 have been proposed and proven to be highly cytotoxic in in vitro tests. Among them, BODIPY derivatives have been extensively studied as alternatives to porphyrin-like PSs due to some specific characteristics, such as large absorption extinction coefficients, sharp emissions, high redox and thermodynamic stability, and relatively simple syntheses, that make them particularly promising for PDT application.13,16,17 Nevertheless, because of their high fluorescence, structural modifications must be made to promote the ISC process. It has been found that the introduction of heavy atoms in appropriate positions on the BODIPY core promotes spin–orbit coupling and then the probability to populate a triplet state,18,19 quenching unwanted radiative processes. The placement of BODIPY as a ligand in transition-metal complexes is another strategy that recently appeared in the literature to improve the ability of a chromophore to undergo ISC processes, due to the well known heavy atom effect.20–22 A recent work20 showed the potential use of a Au(I) complex, [(Ph3P)Au]+, combined with a BODIPY chromophore for singlet oxygen generation. The use of gold in medicine is quite widespread, and its complexes have been used for the treatment of various diseases;23,24 currently, one Au-based compound, named auranofin, is under clinical trial for the treatment of some types of cancer.25–27 Recently, the potential of gold for a joint anticancer/anti-inflammatory action has also been proved.28

Herein, density functional theory (DFT) and its time-dependent formulation (TDDFT) are employed to provide theoretical insight into the mechanism involved in 1O2 photosensitization by a Au–BODIPY combined complex proposed as a promising PDT agent.20 For this purpose, the photophysical properties of intermediates involved in the synthesis of such a promising drug have also been elucidated in order to underline the characteristics that make the gold complex active toward 1O2 photosensitization, compared to a metal-free compound, and compared with iodine-containing BODIPY that was already proved to be an efficient photosensitizer.17 Previous theoretical investigations29–39 have established the reliability of such a tool not only to reproduce the photophysical properties of experimentally proposed photosensitizers for PDT application, but also to predict the properties with reasonable accuracy, helping the understanding of the mechanisms underlying this therapy, especially in metal-containing systems.40–42 This could be of help in selectively orienting the syntheses, the physico chemical characterization and the biological tests of new PDT drugs in a more rapid way and with considerable economic savings. The main aim of our study is to contribute in clarifying the role of gold in promoting the ISC process, correlating the values of the computed spin–orbit coupling with the quantum yields of singlet oxygen production experimentally determined.20 For selected non-radiative processes, the kinetic constants have been provided to better disentangle the most probable pathway for 1O2 photosensitization.

Computational details

Quantum chemical optimizations have been performed at the density functional theory level (DFT) with Gaussian 09 code.43 No symmetry constraints have been imposed during the geometry optimizations performed by using the hybrid B3LYP44,45 exchange–correlation functional in conjunction with the 6-31G(d) basis set for all the atoms, except gold and iodine for which the SDD pseudopotential,46 the most used for Au-containing complexes,47–49 was employed. The choice of the computational protocol was dictated by our previous benchmarks in computing these properties for several compounds.30,34,35 All-real harmonic vibrational frequencies were obtained for the minimum energy optimized structures. The lowest twenty vertical excitation energies were calculated by the time-dependent density functional linear response formulation (TDDFT) on the previously optimized geometries by adding a diffuse function to the basis set. Solvent effects were evaluated by using the non-equilibrium implementation50 of the polarizable continuum model.51 The environment used in the experiments,52 dichloromethane (ε = 8.93), was simulated by means of the polarizable continuum model (PCM).

The rate for the non-radiative transition from the lowest singlet excited state (S1) to each triplet excited state (Tj) lying below, if both electronic states are harmonic, was calculated according to the Fermi golden rule,53 in which the Franck Condon-weighted density of states can be estimated by the Marcus semiclassical approach in the room-temperature regime,54 through the following expression:55,56

 
image file: c8cp04848g-t1.tif(1)
where ωk are the high-frequency (hf) vibrational modes (ħωk > 1000 cm−1); ΔES–T is the energy gap between the S1 and Tj states at their equilibrium geometries, defined as
ΔES−Tj = Eλlf
in which E is the zero-point energy difference and λlf is the reorganization energy contributed by the low-frequency modes (ħωk < 1000 cm−1), which, assuming that all the normal modes are harmonic oscillators, can be written as:
image file: c8cp04848g-t2.tif
where Si, mi and ΔQi are the Huang–Rhys factor, the reduced mass and the equilibrium displacement of the ith low-frequency normal mode ωi. Accordingly, the Huang–Rhys factor Sk and the number of quanta n of the effective high frequency ħωk mode can be defined respectively as:
image file: c8cp04848g-t3.tif

The Huang–Rhys factor Si is obtained by performing FC calculation with Gaussian 09 code. Optimization and frequency calculations of the excited states were performed at the TDDFT level without symmetry constraints. The solvent reorganization energy (λs) is computed by means of the state-specific (SS) approach to estimate which triplet excited state energy difference is calculated with non-equilibrium solvation image file: c8cp04848g-t4.tif and with equilibrium solvation image file: c8cp04848g-t5.tif at the optimized excited state geometry image file: c8cp04848g-t6.tif:

image file: c8cp04848g-t7.tif

Similarly, the intramolecular reorganization energy computed within the SS approach (λSSv) is given by:

image file: c8cp04848g-t8.tif
where image file: c8cp04848g-t9.tif is the energy of the triplet excited state computed with equilibrium solvation at the optimized ground state geometry image file: c8cp04848g-t10.tif. Under the harmonic oscillator approximation, the intramolecular reorganization energy could be estimated as
image file: c8cp04848g-t11.tif
considering all the normal modes ωi.

Spin–orbit matrix elements 〈S1|HSO|Tj〉 were computed with DALTON code57 by using the spin–orbit coupling operators for effective core potentials with an effective nuclear charge58 for systems containing gold and iodine atoms, while the atomic-mean field approximation59 was used in the other cases. For this purpose, the B3LYP functional in conjunction with cc-pVDZ basis set was employed for all the atoms except gold and iodine for which the coupled pseudopotential was considered. All the other terms in kISC were calculated according to vibrationally-resolved electronic spectra implemented in Gaussian 09 package.43

Results and discussion

Selected structural parameters of the optimized geometries at the B3LYP/6-31G(d) level of theory are reported in Fig. 1 while Cartesian coordinates are given in Table S1 of the ESI. As it has already been observed on similar systems at both experimental14 and theoretical34 levels, in all the considered compounds, the BODIPY core appears completely planar with fluorine atoms that come out from the plane and phenyl in the meso position arranged in a pseudo-orthogonal fashion.
image file: c8cp04848g-f1.tif
Fig. 1 Schematic structure of the investigated systems. Main structural parameters are included, distances in Å and angles in degrees.

Looking at the Au-bdp system, a covalent interaction between Au and the alkynyl ligand can be observed; the obtained Au–C distance (2.002 Å) agrees well with the corresponding distance in similar complexes like ClAu–CCH (1.950 Å) and ClAu–CN (1.993 Å) computed at the CCSD(T) level of theory.60 Similarly, the Au–P bond length is close to that previously obtained for analogous gold compounds (2.355 vs. 2.392 Å).61 The phenyl groups of the triphenylphosphine show torsional angles of about 35 degrees resulting from a balance between steric hindrance and electronic delocalization. In all the studied systems, whatever the substituent in the β position of BODIPY, the planar structure of the core and the meso-phenyl ring in perpendicular arrangement (about 89 degrees in all cases, see Fig. 1) remain unaltered.

The computed vertical excitation energies for both the low-lying singlet and triplet states are reported in Table 1. As already found for other BODIPY-based systems29,30,34,35 the values of the excitation singlet energies are overestimated with respect to the corresponding experimental values, by about 0.3 eV.20,62 The origin of this error was well documented37 and applying the suggested correction, based on the use of the coupled cluster method, the agreement with the experimental data improves significantly.

Table 1 Vertical excitation energy, ΔE (eV), λmax (nm), oscillator strength f and main transitions for the studied compounds computed in dichloromethane using B3LYP/6-31+G(d). Experimental value is taken from ref. 20 and 17
Compound State ΔE λ f Transitions Assignment λ exp
bdp S1 2.86 433 0.606 H → L, (98%) ππ* 503
T1 1.54 804 H → L, (100%) ππ*Ph
T2 2.76 450 H−1 → L, (94%) ππ*Ph
Ac-bdp S1 2.72 456 0.591 H → L, (93%) ππ* 518
T1 1.55 799 H → L, (97%) ππ*Ph
T2 2.53 491 H−1 → L, (89%) ππ*Ph
Au-bdp S1 2.45 505 0.550 H → L, (92%) ππ* 550
T1 1.48 838 H → L, (91%) ππ*PPh3
T2 2.20 563 H−1 → L, (85%) ππ*PPh3
I-bdp S1 2.66 466 0.616 H → L, (93%) ππ* 529
T1 1.52 815 H → L, (97%) πn*
T2 2.52 492 H−1 → L, (90%) πn*


The introduction of iodine atoms or a Au moiety into the BODIPY core red-shifts the S1 band. In particular, the inclusion of iodine atoms in β positions of BODIPY increases the wavelength for the maximum absorption by 33 nm with respect to the naked bdp, while the presence of gold causes a more substantial red-shift (72 nm), as experimentally found (26 and 47 nm in the two cases, respectively).

In all cases, the excitation is due to a HOMO → LUMO transition that results in the ππ* nature depicted in Fig. 2 (for Au-bdp and I-bdp) and Fig. S1 (ESI) (for bdp and Ac-bdp). The coordination of a BODIPY molecule to the Au center, through an acetylene linker, contributes to extending the conjugation of the π-system to the triple bond. The computed density difference plots between S1 and S0, which give an indication of the variation of the electronic density upon electronic excitations, in all cases involve the π-system of the BODIPY core (see Fig. S2, ESI) with a small participation of the heavy atoms.


image file: c8cp04848g-f2.tif
Fig. 2 HOMO−1 (H−1), HOMO (H) and LUMO (L) molecular orbital compositions for singlet and triplet excited states of Au-bdp and I-bdp systems.

For all the investigated compounds, below the S1 state, two excited triplet states (T1 and T2) were located. The energy gap between the ground state and first excited triplet state T1, extremely important to evaluate the possibility for molecular oxygen excitation (3Σg1Δg), does not vary significantly, and reaches values close to 1.5 eV, sufficient to promote 1O2 formation. More significant variations were found for the T2 energies for the different compounds, which rank from 2.20 (I-bdp) to 2.76 eV (bdp). The triplet states, T1 and T2, originated from the HOMO → LUMO and HOMO−1 → LUMO transitions, respectively. For bdp and Ac-bdp, both the transitions are ππ* with the π*Ph orbital mainly localized on the meso-phenyl ring (see Fig. S1, ESI). In Au-bdp triplets, the transition is also ππ* in nature but the π* orbital composition involves only the –PPh3 ligand (see Fig. 2). On the contrary, in I-bdp, the nature of the transition, in both T1 and T2 states, is πn* (Fig. 2).

The computed spin orbit coupling constants (〈ΨS1|ĤSO|ΨTj〉) are presented in Table 2, in which the fluorescence lifetime (ΦF) and the singlet oxygen (1O2) quantum yields, experimentally measured,17,20 are also reported. The SOC small values computed for bdp and Ac-bdp suggest a low probability of intersystem crossing between S1 and Tj spin states that justifies the unrevealed 1O2 production by experimental measurement.17 Keeping in mind the El-Sayed63 rules, no significant changes in the composition of the molecular orbitals of the states involved in the transitions were found in these cases. However, in both Au-bdp and I-bdp, the presence of Au or I heavy atoms increases the SOC values by an order of magnitude. Although in the former case, the MO composition remains of the π type, the LUMO orbital of both triplet states is localized on the PPh3 ligand, contrariwise to that of S1, which involves almost exclusively the BODIPY core. This evidence suggests that accordingly the ISC efficiency should increase. SOC data agree with the experimental observations that show the production of singlet oxygen with good yields only if Au-bdp and I-bdp are used as photosensitizers.20 Furthermore, an inspection of the available experimental photophysical data reveals that only the Au- and I-containing BODIPY possess a very low fluorescence lifetime,20 essential to avoid the occurrence of other excited state deactivation pathways.

Table 2 Spin–orbit matrix elements (cm−1) calculated at the B3LYP/cc-pVDZ//M06/6-31G(d) level of theory. Experimental ΦF and ΦΔ values are taken from ref. 17 and 20
bdp Ac-bdp Au-bdp I-bdp
|〈ΨS1|ĤSO|ΨT1〉| 0.02 0.35 2.27 5.30
|〈ΨS1|ĤSO|ΨT2〉| 0.018 0.23 0.49 7.52
Φ F 0.71 0.58 0.05 0.03
Φ Δ n.a n.a. 0.84 0.79


The greatest SOCs found for I-bdp are due to the significant changes in the composition of the molecular orbitals of the states involved in the transitions that, as previously discussed, are πn*. While in Au-bdp, the MO composition, though involving different parts of the molecule (see LUMO composition localized on –PPh3 ligand in Fig. 2), remains of the ππ* type. Indeed, it was already reported that the inclusion of a heavy atom directly on a BODIPY core has a great influence on the SOC values with respect to heavy atoms in peripheral positions.29 In view of the very small difference in the measured experimental values of 1O2 (ΦΔ in Table 2) quantum yield for the two compounds,20 the SOC values appear to be in contrast with what has been observed. However, even if the ISC rate constant considerably depends on the 〈S1|HSO|Tj〉 values, as underlined above, other terms take part in determining the rate of the whole radiationless process. As reported in the computational details section, in the limit of the Franck–Condon approximation in the non-adiabatic regime, the rate of the ISC process leading to the triplet state population, which needs to be populated for triggering 1O2 photosensitization, can be estimated by using the Fermi golden rule of eqn (1). The quantities computed for determining the kISC are presented in Table 3.

Table 3 Effective Huang–Rhys factor (Sk), intramolecular (λv) and solvent (λs) reorganization energies, dipole moment μ of the excited states and ISC rate constants for S1 → Tj, with j = 1, 2, transitions of Au-bdp and I-bdp
S k λ SSv[thin space (1/6-em)]b λ FCv[thin space (1/6-em)]b λ s μ S1[thin space (1/6-em)]c μ Tj[thin space (1/6-em)]c ΔE(S1 → Tj)d k ISC(S1 → Tj)e
a Effective Huang–Rhys factor for the high frequency modes ωj > 1000 cm−1. b cm−1. c Debye. d eV. e s−1.
j = 1
Au-bdp 1.16 1797 1797 5710 17.2433 11.5368 0.97 3.02 × 1011
I-bdp 6.24 8045 8857 4940 7.1184 4.2041 1.14 1.14 × 1012
j = 2
I-bdp 3.41 4609 4738 5180 7.1184 7.6811 0.14 1.86 × 1011


Based on data presented in Table 1, two possible deactivation pathways can be hypothesized:

image file: c8cp04848g-t12.tif

First, the kISC values for both possible non-radiative transitions were calculated only for I-bdp, while for Au-bdp, exclusively the kISC(S1 → T1) value was computed. This is because, in order to properly use the procedure for the vibrationally-resolved electronic spectra used to obtain all the quantities needed to calculate kISC, obviously both states have to be real minima, no imaginary frequencies should be found in any of the involved states. Unfortunately, for Au-bdp, all the attempts to find a real minimum for the T2 state failed. However, looking at kISC(S1 → Tj) computed for I-bdp, it appears clear that despite the substantial larger energy gap between the initial and the final states, S1–T1, with respect to the other, S1–T2, the rate constant is only slightly influenced. Therefore, we are quite sure that pathway (b) can be considered the most probable for both photosensitizers.

Nevertheless, though the kISC(S1 → T2) for Au-bdp is missing, a significant conclusion can be drawn by the comparison between the available data for both compounds. Indeed, looking at the data reported for the S1 → T1 transition, both compounds exhibit an excellent ability to populate the low-lying triplet state. These results agree very nicely with the experimental observation of a singlet oxygen quantum yield close to 0.8 in both cases. Given the small difference between the two systems found and the slight variance in the ΦΔ quantum yields experimentally measured,17,20 it is difficult to establish which of the two systems is more efficient in producing singlet oxygen. In any case, our study clearly indicates that the inclusion of a gold atom into the BODIPY structure, even if not directly on the chromophore core, sensibly enhances the excited singlet oxygen production, contrariwise to what has been previously observed with other heavy atoms.29

Conclusions

In this work, a DFT-based theoretical investigation is devoted to establishing whether the photophysical properties of a BODIPY compound can be affected by the formation of a gold(I) complex in view of its possible use as a photosensitizer in the PDT. Several photophysical properties, i.e. excitation energies, singlet–triplet energy gaps, spin orbit coupling constants and kISC between the low lying excited states, were also determined for an iodine-containing compound already proved to be an efficient PS. The absorption properties were shown to be influenced by the presence of the heavy atoms, iodine or gold, that entail a slight red-shift (a few nanometers) in the maximum absorption wavelengths, with the Q-band falling in the low region of the therapeutic window. For all the investigated compounds, the energy of the low-lying excited triplet state is enough to excite the oxygen molecules from the 3Σg to 1Δg state (0.98 eV). Calculations clearly show that the presence of the heavy atom is essential to enhance the amplitude of the spin–orbit coupling matrix elements and then the ISC rate. In particular, the occurrence of a radiationless singlet–triplet transition between states with different orbital character (ππ* → πn*) has been found for the iodine-incorporated system I-bdp, while the inclusion of a gold atom not directly anchored to the chromophore core skeleton but through a triplet bond equally entails an increase of the spin–orbit coupling constant due to the heavy atom effect. These results agree very nicely with the experimental observation of a singlet oxygen quantum yield close to 0.8 in both cases. Thus, the computed properties, in satisfactory agreement with the available experimental findings, indicate that both Au-bdp and I-bdp can be equally considered as potential photosensitizers in photodynamic therapy.

Conflicts of interest

There are no conflicts to declare.

References

  1. J. M. Dabrowski and L. G. Arnaut, Photochem. Photobiol. Sci., 2015, 14, 1765–1780 RSC.
  2. S. B. Brown, E. A. Brown and I. Walker, Lancet Oncol., 2004, 5, 497–508 CrossRef CAS.
  3. A. P. Castano, P. Mroz and M. R. Hamblin, Nat. Rev. Cancer, 2006, 6, 535–545 CrossRef CAS.
  4. D. E. J. G. J. Dolmans, D. Fukumura and R. K. Jain, Nat. Rev. Cancer, 2003, 3, 380–387 CrossRef CAS.
  5. H. Kim, W. Kim, Y. Mackeyev, G.-S. Lee, H.-J. Kim, T. Tachikawa, S. Hong, S. Lee, J. Kim, L. J. Wilson, T. Majima, P. J. J. Alvarez, W. Choi and J. Lee, Environ. Sci. Technol., 2012, 46, 9606–9613 CrossRef CAS.
  6. O. C. Romero, A. P. Straub, T. Kohn and T. H. Nguyen, Environ. Sci. Technol., 2011, 45, 10385–10393 CrossRef CAS.
  7. J. M. Dabrowski, B. Pucelik, A. Regiel-Futyra, M. Brindell, O. Mazuryk, A. Kyzioł, G. Stochel, W. Macyk and L. G. Arnaut, Coord. Chem. Rev., 2016, 325, 67–101 CrossRef CAS.
  8. C. Mari, V. Pierroz, S. Ferrari and G. Gasser, Chem. Sci., 2015, 6, 2660–2686 RSC.
  9. S. Yanoa, S. Hirohara, M. Obata, Y. Hagiya, S.-i. Ogura, A. Ikeda, H. Kataoka, M. Tanaka and T. Joh, J. Photochem. Photobiol., C, 2011, 12, 46–67 CrossRef.
  10. J. Zhao, W. Wu, J. Sun and S. Guo, Chem. Soc. Rev., 2013, 42, 5323–5351 RSC.
  11. G. Herzberg, Spectra of Diatomic Molecules, Van Nostrand Reinhold, New York, 2nd edn, 1950, pp. 344–346 Search PubMed.
  12. D. Ramaiah, I. Eckert, K. T. Arun, L. Weidenfeller and B. Epe, Photochem. Photobiol., 2002, 76, 672–677 CrossRef CAS.
  13. A. Kamkaew, S. H. Lim, H. B. Lee, L. V. Kiew, L. Y. Chung and K. Burgess, Chem. Soc. Rev., 2013, 42, 77–88 RSC.
  14. A. Loudet and K. Burgess, Chem. Rev., 2007, 107, 4891–4932 CrossRef CAS.
  15. L. J. Patalag, P. G. Jones and D. B. Werz, Angew. Chem., Int. Ed., 2016, 55, 13340–13344 CrossRef CAS.
  16. J. Zhao, K. Xu, W. Yang, Z. Wang and F. Zhong, Chem. Soc. Rev., 2015, 44, 8904–8939 RSC.
  17. W. Wu, H. Guo, W. Wu, S. Ji and J. Zhao, J. Org. Chem., 2011, 76, 7056–7064 CrossRef CAS.
  18. T. Yogo, Y. Urano, Y. Ishitsuka, F. Maniwa and T. Nagano, J. Am. Chem. Soc., 2005, 127, 12162–12163 CrossRef CAS.
  19. S. Ozlem and E. U. Akkaya, J. Am. Chem. Soc., 2009, 131, 48–49 CrossRef CAS.
  20. M. Üçüncü, E. Karakus, E. K. Demirci, M. Sayar, S. Dartar and M. Emrullahoğlu, Org. Lett., 2017, 19, 2522–2525 CrossRef.
  21. F. Nastasi, F. Puntoriero, S. Campagna, S. Diring and R. Ziessel, Phys. Chem. Chem. Phys., 2008, 10, 3982–3986 RSC.
  22. F. Nastasi, F. Puntoriero, S. Serroni, S. Campagna, J.-H. Olivier and R. Ziessel, Dalton Trans., 2014, 43, 17647–17658 RSC.
  23. S. P. Pricker, Gold Bull., 1996, 29, 53–60 CrossRef.
  24. C.-M. Che and R. W.-Y. Sun, Chem. Commun., 2011, 47, 9554–9560 RSC.
  25. C. K. Mirabelli, R. K. Johnson, C. M. Sung, L. Faucette, K. Muirhead and S. T. Crooke, Cancer Res., 1985, 45, 32–39 CAS.
  26. C. Roder and M. J. Thomson, Drugs R&D, 2015, 15, 13–20 CrossRef CAS PubMed.
  27. https://clinicaltrials.gov/ct2/show/NCT01419691?term=auranofin&rank=5 .
  28. A. Trommenschlager, F. Chotard, B. Bertrand, S. Amor, L. Dondaine, M. Picquet, P. Richard, A. Bettaïeb, P. Le Gendre, C. Paul, C. Goze and E. Bodio, Dalton Trans., 2017, 46, 8051–8056 RSC.
  29. B. C. De Simone, G. Mazzone, J. Pirillo, N. Russo and E. Sicilia, Phys. Chem. Chem. Phys., 2017, 19, 2530–2536 RSC.
  30. M. E. Alberto, B. C. De Simone, G. Mazzone, A. D. Quartarolo and N. Russo, J. Chem. Theory Comput., 2014, 10, 4006–4013 CrossRef CAS PubMed.
  31. M. E. Alberto, B. C. De Simone, G. Mazzone, T. Marino and N. Russo, Dyes Pigm., 2015, 120, 335–339 CrossRef CAS.
  32. G. Mazzone, M. E. Alberto, B. C. De Simone, T. Marino and N. Russo, Molecules, 2016, 21, 288–299 CrossRef PubMed.
  33. J. Pirillo, G. Mazzone and N. Russo, Chem. – Eur. J., 2018, 24, 3512–3519 CrossRef CAS PubMed.
  34. G. Mazzone, A. D. Quartarolo and N. Russo, Dyes Pigm., 2016, 130, 9–15 CrossRef CAS.
  35. B. C. De Simone, G. Mazzone, N. Russo, E. Sicilia and M. Toscano, Phys. Chem. Chem. Phys., 2018, 20, 2656–2661 RSC.
  36. G. Mazzone, N. Russo and E. Sicilia, Can. J. Chem., 2013, 91, 902–906 CrossRef CAS.
  37. B. Le Guennic, G. Scalmani, M. J. Frisch, A. D. Laurent and D. Jacquemin, Phys. Chem. Chem. Phys., 2017, 19, 10554–10561 RSC.
  38. C. Adamo and D. Jacquemin, Chem. Soc. Rev., 2013, 42, 845–856 RSC.
  39. E. S. E. Eriksson and L. A. Eriksson, Phys. Chem. Chem. Phys., 2011, 13, 7207–7217 RSC.
  40. M. E. Alberto, N. Russo and C. Adamo, Chem. – Eur. J., 2016, 22, 9162–9168 CrossRef CAS.
  41. M. E. Alberto, J. Pirillo, N. Russo and C. Adamo, Inorg. Chem., 2016, 55, 11185–11192 CrossRef CAS.
  42. M. E. Alberto and C. Adamo, Chem. – Eur. J., 2017, 23, 15124–15132 CrossRef CAS PubMed.
  43. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian 09, Revision D.1, Gaussian, Inc., Wallingford CT, 2009 Search PubMed.
  44. A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652 CrossRef CAS.
  45. C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter Mater. Phys., 1988, 37, 785–789 CrossRef CAS.
  46. D. Andrae, U. Haussermann, M. Dolg, H. Stoll and H. Preuss, Theor. Chim. Acta, 1990, 77, 123–141 CrossRef CAS.
  47. G. Mazzone, N. Russo and E. Sicilia, J. Chem. Theory Comput., 2015, 11, 581–590 CrossRef CAS.
  48. G. Mazzone, N. Russo and E. Sicilia, Organometallics, 2012, 31, 3074–3080 CrossRef CAS.
  49. G. Mazzone, N. Russo and E. Sicilia, J. Chem. Theory Comput., 2010, 6, 2782–2789 CrossRef CAS.
  50. M. Cossi and V. Barone, J. Chem. Phys., 2000, 112, 2427–2435 CrossRef CAS.
  51. J. Tomasi, B. Mennucci and R. Cammi, Chem. Rev., 2005, 105, 2999–3094 CrossRef CAS.
  52. B. S. Brunschwig and N. Sutin, Comments Inorg. Chem., 1987, 6, 209–235 CrossRef CAS.
  53. K. Schmidt, S. Brovelli, V. Coropceanu, D. Beljonne, J. Cornil, C. Bazzini, T. Caronna, R. Tubino, F. Meinardi and Z. Shuai, et al. , J. Phys. Chem. A, 2007, 111, 10490–10499 CrossRef CAS PubMed.
  54. G. Baryshnikov, B. Minaev and H. Ågren, Chem. Rev., 2017, 117, 6500–6537 CrossRef CAS.
  55. G. S. M. Tong, K. T. Chan, X. Changa and C.-M. Che, Chem. Sci., 2015, 6, 3026–3037 RSC.
  56. J. L. Brédas, D. Beljonne, V. Coropceanu and J. Cornil, Chem. Rev., 2004, 104, 4971–5004 CrossRef PubMed.
  57. K. Aidas, C. Angeli, K. L. Bak, V. Bakken, R. Bast, L. Boman, O. Christiansen, R. Cimiraglia, S. Coriani and P. Dahle, et al. , WIREs Comput. Mol. Sci., 2014, 4, 269–284 CrossRef CAS.
  58. S. Koseki, M. W. Schmidt and M. S. Gordon, J. Phys. Chem. A, 1998, 102, 10430–10436 CrossRef CAS.
  59. K. Ruud, B. Schimmelpfennig and H. Ågren, Chem. Phys. Lett., 1999, 310, 215–221 CrossRef CAS.
  60. H.-T. Liu, X.-G. Xiong, P. D. Dau, Y.-L. Wang, D.-L. Huang, J. Li and L.-S. Wang, Nat. Commun., 2013, 4, 2223–2228 CrossRef PubMed.
  61. D. Benitez, N. D. Shapiro, E. Tkatchouk, Y. Wang, W. A. Goddard III and F. D. Toste, Nat. Chem., 2009, 1, 482–486 CrossRef CAS PubMed.
  62. C. M. Marian, Wiley Interdiscip. Rev. Comput. Mol. Sci., 2012, 2, 187–203 CrossRef CAS.
  63. M. A. El-Sayed, Acc. Chem. Res., 1968, 1, 8–16 CrossRef CAS.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c8cp04848g

This journal is © the Owner Societies 2019