Near-Infrared Optical Imaging of Necrotic Cells by ... - ACS Publications

Communication pubs.acs.org/JACS

Near-Infrared Optical Imaging of Necrotic Cells by Photostable Lanthanide-Based Metallacrowns Ivana Martinić,† Svetlana V. Eliseeva,*,† Tu N. Nguyen,‡ Vincent L. Pecoraro,*,‡ and Stéphane Petoud*,§,† †

Centre de Biophysique Moléculaire, CNRS UPR 4301, 45071 Orléans Cedex 2, France Department of Chemistry, Willard H. Dow Laboratories, University of Michigan, Ann Arbor, Michigan 48109, United States

‡

S Supporting Information *

of partial or complete removal of the autofluorescence signal, which increases the signal-to-background ratio and the corresponding detection sensitivity.10−13 Therefore, the development of novel NIR imaging agents with improved photophysical properties that can address specific biological problems is in high demand. Lanthanide (Ln3+)-based probes are garnering high attention owing to their attractive optical properties for the detection in biological conditions: (i) strong resistance to photobleaching, (ii) sharp emission bands covering a broad range of wavelengths from the visible to the NIR, the positions of which are not affected by the local microenvironment (temperature, pH and hydrophobic/hydrophilic character of biomolecules), and (iii) large difference between excitation and emission wavelengths.14−20 Despite several examples of in vitro imaging with NIR-emitting Ln3+-based complexes21,22 and nanoparticles,23,24 to date, none of them have been used for the imaging of necrotic cells. Metallacrowns (MCs) are the most highly luminescent molecular complexes for Ln-based NIR emission. These molecules are formed by repeating [Metal-N-O] subunits in a manner analogous to classical crown ethers.25,26 In 2011, a sandwich structure was reported that assembled 16 picolinehydroxamate ligands (picHA 2− ) (Figure 1), 16 Zn2+ as constituents of the MC ring metals, and a single Ln3+ as a central ion.27 It was demonstrated that highly efficient sensitization of NIR emission and protection of the Ln3+ from sources of nonradiative deactivations were achieved.

ABSTRACT: Sensitive detection of cell necrosis is crucial for the determination of cell viability. Because of its high resolution at the cellular level and sensitivity, optical imaging is highly attractive for identifying cell necrosis. However, challenges associated with this technique remain present such as the rapid photobleaching of several types of organic fluorophores and/or the interference generated by biological autofluorescence. Herein, we synthesized novel biologically compatible Zn2+/Ln3+ metallacrowns (MCs) that possess attractive near-infrared (NIR) emission and are highly photostable. In addition, these MCs have the ability to label differentially necrotic HeLa cells from living cells. This work is also the first demonstration of (i) the use of the NIR emission arising from a single lanthanide(III) cation for optical biological imaging of cells under single photon excitation, (ii) the first example of a lanthanide(III)-based NIR-emitting probe that can be targeted to a specific type of cell.

T

o monitor cell viability as a feedback to various environmental factors or drug treatments,1 it is essential to develop simple to operate, rapid, highly sensitive and efficient methods for the detection of unprogrammed cell death, necrosis.2 In the last few decades, the use of optical fluorescence imaging has significantly expanded due to the unique advantage of the sensitive detection of small quantities of fluorescent probes with high resolution at the cellular level (nM scale).3 However, the number of fluorescent probes used for the detection of cell necrosis is limited. Existing methods for the detection of cell necrosis by optical imaging rely on visible-emitting, nonpermeable fluorescent probes that gain entry into necrotic cells through the disrupted plasma membranes.4 One of the most widely used probes suitable for the determination of cell viability through selective labeling of necrotic cells is propidium iodide (PI).5 The main drawback of this organic fluorophore and of other organic molecules is their rapid photobleaching that causes difficulties in setting up real-time, repeated and quantitative experiments.6 In addition, most of these probes emit in the visible domain, thus limiting their range of applications due to the detrimental contribution of unwanted biological background fluorescence (autofluorescence) to the collected signal preventing accurate quantification and unambiguous interpretation of results. The detection in the NIR region7−9 provides the unique advantage © 2017 American Chemical Society

Figure 1. Crystal structure representation of Ln3+[12-MCZn(II),pyzHA4]2[24-MCZn(II),pyzHA-8] and structural formulas of hydroxamic acids.27,28 Received: February 14, 2017 Published: June 14, 2017 8388

DOI: 10.1021/jacs.7b01587 J. Am. Chem. Soc. 2017, 139, 8388−8391

Communication

Journal of the American Chemical Society These MCs possess a high energy excitation wavelength (31 250 cm−1), which is detrimental to the biological systems and strongly limits their potential use. More recently, we have reported a second generation of Zn2+/Ln3+ MCs based on quinaldichydroxamic acid (H2quinHA) (Figure 1) that possess excitation wavelengths shifted toward lower energy and outstanding photophysical properties in the solid state and deuterated methanol.28 However, for imaging applications under physiological conditions, we needed more water-soluble and hydrophilic derivatives (Figure S9). To address this requirement, we have developed a new chromophoric group based on pyrazinehydroxamic acid (H2pyzHA) (Figure 1) that leads to the formation of Zn2+/Ln3+ MCs possessing a significantly higher water solubility while retaining the outstanding photophysical properties of the previous generations of NIR-emitting MCs. Taking advantage of Ln3+[Zn(II)MCpyzHA] MCs (Ln3+ = Yb, Nd), our first goal was to establish a first proof-of-principle that a metallamacrocycle can operate as a NIR imaging agent for biological applications. It is worth noting that these new biologically compatible MCs are significantly smaller in size and contain only one emissive Ln3+ in contrast to the previously reported poly-Ln3+ NIR probes suitable for optical microscopy under single photon excitation, MOFs or polyamidoamine dendrimers. In addition, the ability of these MCs to accumulate preferentially in necrotic HeLa cells, while not penetrating into living ones, can be exploited for the specific labeling of necrotic cells for NIR microscopy experiments. The Zn 2+ /Ln 3+ MCs, Ln 3+ [12-MC Zn(II),pyzHA -4] 2 [24MCZn(II),pyzHA-8] (hereafter Ln3+[Zn(II)MCpyzHA], Ln3+ = Yb, Nd or Y), were synthesized according to a modified procedure.28 According to the X-ray analysis, they are isostructural to MCs formed with picHA2− and quinHA2− (Figure S5, Table S1, Supporting Information).27,28 The presence of the 16 pyzHA2− ligands significantly enhances the water solubility of the resulting Ln3+[Zn(II)MCpyzHA] compared to those formed with the picHA2− and quinHA2− ligands. A solubility limit of ∼5 mM at room temperature for both Yb3+ and Nd3+ analogues was obtained. Moreover, Nd3+[Zn(II)MCpyzHA] and Yb3+[Zn(II)MCpyzHA] are stable in water as indicated by ESI-MS studies performed immediately after the synthesis and 1 month later, which do not reveal the presence of other species (Figures S1, S2, Supporting Information). Similarly, the 1 H NMR spectra of the diamagnetic model Y3+[Zn(II)MCpyzHA] in D2O solution further substantiated the solution stability of this family of MCs (Figure S4, Supporting Information). The absorption spectra of the Yb3+[Zn(II)MCpyzHA] and Nd3+[Zn(II)MCpyzHA] MCs in aqueous solutions (Figure 2a) exhibit nearly superimposable absorption bands in the range of 200−460 nm with the low-energy maxima located at 360 nm (ε ∼ 4.8 × 104 M−1 cm−1). This band is not observable in the absorption spectrum of the free H2pyzHA ligand and can, therefore, be assigned to an inter- or intraligand charge transfer state (CT) that can only be created upon formation of the MC scaffold. The high value of the molar absorption coefficient can be explained by the π*←π transitions associated with the presence of the 16 aromatic pyzHA2−. Upon excitation at 370 nm (Figure 2b), Yb3+[Zn(II)MCpyzHA] and Nd3+[Zn(II)MCpyzHA] in solid state and aqueous solutions exhibit the characteristic sharp emission bands in the NIR, arising from the f−f transitions, and centered at 980 nm (2F5/2→2F7/2) for Yb3+

Figure 2. (a) Absorption spectra of the H2pyzHA ligand (multiplied by a factor 16 to match the total number of ligands present in each complex) and of Ln3+[Zn(II)MCpyzHA] in water (150 μM, room temperature). (b) Normalized excitation (dashed traces, λem(Nd3+) = 1070 nm, λem(Yb3+) = 980 nm) and emission (solid traces, λex = 370 nm) spectra of Ln3+[Zn(II)MCpyzHA] (solid or 200 μM solution, room temperature).

or 900, 1070 and 1360 nm (4F3/2→4IJ (J = 9/2, 11/2, 13/2)) for Nd3+, respectively (Figure 2b). The excitation spectra of the MCs collected upon monitoring the emission of Yb3+ at 980 nm or Nd3+ at 1070 nm are comparable and are dominated by ligand-centered broad bands in the range of 250−480 nm for aqueous solutions and 250− 550 nm for solid state samples (Figure 2b). This observation shows that the MC scaffold is able to sensitize efficiently both NIR-emitting Ln3+ through antenna effect. As the Yb3+ does not possess any electronic levels in the UV and visible ranges, the only possibility to sensitize its excited states with such wavelengths is to take advantage of the chromophoric MC scaffold, i.e., through the antenna effect. The fact that the excitation spectra of both Yb3+ and Nd3+ MCs are similar indicates that the energy is following the same path in both complexes and that the antenna effect originates from the same electronic structure localized on the chromophores. In addition to broad bands, the excitation spectrum of the Nd3+ MC recorded in the solid state reveals the presence of several sharp features (up to 850 nm) that can be assigned to f−f transitions, indicating the possibility of direct excitation of this cation. Quantitative photophysical data (luminescence lifetimes (τ) and Ln3+-centered quantum yield values (Q)) are summarized in Table 1. Experimental luminescence decays of Ln3+[Zn(II)MCpyzHA] (Ln3+ = Yb, Nd) in the solid state or in aqueous solutions are best fitted with a monoexponential function, thus reflecting the presence of a unique Ln3+-containing emissive species. The comparative analysis of the τ values obtained in H2O and D2O solutions allows one to estimate the number of solvent molecules directly coordinated to Ln3+ using phenomenological equations.29,30 Values of q = 0 (Yb3+) and q = 0.1 (Nd3+) were obtained, indicating that no water molecule is directly bound to the Ln3+ in these MC structures and that the protection is not dependent on the sizes of the Ln3+ as the ionic radius of Nd3+ is 12.6% larger than for Yb3+ (1.109 vs 0.985 Å for a coordination number of 8 according to Shannon).31 8389

DOI: 10.1021/jacs.7b01587 J. Am. Chem. Soc. 2017, 139, 8388−8391

Communication

Journal of the American Chemical Society

Table 1. Photophysical Properties of Ln3+[Zn(II)MCpyzHA] (Ln3+ = Yb, Nd) in the Solid State and in Aqueous Solutions (200 μM)a τ (μs)b

a

Q (%)c

Metallacrown

Solid

H2O

D2 O

Solid

H2O

D2 O

Yb3+[Zn(II)MCpyzHA] Nd3+[Zn(II)MCpyzHA]

45.6(3) 1.71(1)

5.57(1) 0.214(4)

81.3(1) 1.29(1)

0.659(4) 0.444(9)

1.12(7)·10−2 7.7(1)·10−3

0.257(3) 6.17(9)·10−2

At room temperature; 2σ values are given between parentheses. Experimental errors: τ, ±2%; Q, ±10%. bλex = 355 nm. cλex = 370 nm.

Nevertheless, the Q values are significantly lower in H2O compared to the values recorded for the samples in D2O or in the solid state, reflecting a high probability of nonradiative deactivations through vibrations from molecules located in a second-sphere of coordination of NIR-emitting Ln3+. Despite this fact, τ and Q values obtained for Ln3+[Zn(II)MCpyzHA] (Ln3+ = Yb, Nd) are comparable with the highest ones observed to date for Ln3+ complexes formed with ligands containing C−H bonds.32,33 Yb3+[Zn(II)MCpyzHA] was chosen as a candidate for the initial biological studies because it has a superior quantum yield value in comparison to Nd3+[Zn(II)MCpyzHA]. The cell viability was evaluated by the alamarBlue assay.34 A value superior to 90% of viability of HeLa cells was observed upon incubation concentrations up to 45 μM Yb3+[Zn(II)MCpyzHA] for 24 and 48 h (Figure S6, Supporting Information). Therefore, epifluorescence microscopy experiments could be performed on samples treated with a 45 μM solution of MC due to the high cell viability at this concentration. To test the ability of Yb3+[Zn(II)MCpyzHA] to accumulate preferentially in necrotic cells, necrosis was induced by the incubation of HeLa cancer cells in glucose depleted media for 24 h.35 The resulting cell culture containing both necrotic and living HeLa cells was incubated with a 45 μM solution of MCs in Opti-MEM cell culture medium with a 2% serum supplement. Epifluorescence microscopy images collected upon excitation using a 447 nm band-pass 60 nm filter upon monitoring the Yb3+ emission demonstrated that the Yb3+[Zn(II)MCpyzHA] is a NIR-emitting probe that does not enter living cells (nonpermeable) while being able to label both the cytoplasm and the nucleus of necrotic cells (Figure 3). Such properties of MCs were also confirmed by comparative experiments carried out with the widely used commercially available probe, propidium iodide (PI). In addition to the labeling of necrotic cells, PI is often used for the identification of late-stage apoptotic cells since the integrity of their membranes is also disrupted. 36 To demonstrate the ability of MC to accumulate preferentially in late-stage apoptotic cells, in addition to the necrotic ones, we performed epifluorescence microscopy experiments in which the apoptosis was induced in the HeLa cells upon incubation with a 20 μM etoposide solution for 48 h.37 The simultaneous incubation of apoptotic cells with PI or MC and commercial Annexin V conjugate (e.g., Annexin V-FITC probe) allows one to differentiate between early and late phases of apoptosis.36 Therefore, we observed that cells labeled with PI or MC and Annexin V-FITC are in a late-stage of apoptosis whereas the ones labeled only with the Annexin V-FITC could be assigned to an early apoptotic stage (Figure S8, Supporting Information). The resistance to photobleaching is of major importance for quantitative biological experiments, in particular when they require longer periods of time or repetitions, and to avoid any risk of false negative results. The primary photobleaching

Figure 3. Images obtained from epifluorescence microscopy experiments performed on HeLa cells (top) incubated with a 45 μM solution of Yb3+[Zn(II)MCpyzHA] for 15 min followed by a 5 min incubation with a 3 μM solution of PI and (bottom) control cells incubated in glucose-containing media in absence of Yb3+[Zn(II)MCpyzHA]. White arrows identify the necrotic cells in the culture. (a) Brightfield. (b) NIR signal arising from Yb3+[Zn(II)MCpyzHA] (λex, 447 nm band-pass 60 nm; λem, long pass 805 nm; exposure time, 10 s). (c) Visible signal arising from PI (λex, 550 nm band-pass 25 nm; λem, 605 nm band-pass 70 nm; exposure time, 100 ms). (d) Merged between b and c images. (e) Merged between a, b and c images. 40× magnification objective.

experiments were performed for Ln3+[Zn(II)MCpyzHA] (Ln3+ = Yb, Nd) in water solutions (Figure S7, Supporting Information). In addition, the photostability of Yb3+[Zn(II)MCpyzHA] in HeLa cells was confirmed with epifluorescence microscopy experiments (Figure 4). The intensity of the NIR signal arising from Yb3+ remained unchanged upon continuous illumination for 8 min with light selected by a 447 nm bandpass 60 nm filter, whereas the signal from PI significantly decreased upon similar continuous exposure to the light selected with a 550 nm band-pass 25 nm filter.

Figure 4. Images obtained from photobleaching epifluorescence microscopy experiments performed on necrotic HeLa cells incubated with a 3 μM solution of PI for 5 min or a 45 μM solution of Yb3+[Zn(II)MCpyzHA] for 15 min after continuous excitation with light selected by a 550 nm band-pass 25 nm filter or a 447 nm band-pass 60 nm filter, respectively, for (a) 10 s, (b) 50 s, (c) 100 s, (d) 200 s, (e) 500 s. (Top) PI, visible emission (λex, 550 nm band-pass 25 nm; λem, 605 nm band-pass 70 nm; exposure time, 100 ms). (Bottom) Yb3+[Zn(II)MCpyzHA], NIR emission arising from Yb3+ (λex, 447 nm band-pass 60 nm; λem, long pass 805 nm; exposure time, 10 s). 63× magnification objective. 8390

DOI: 10.1021/jacs.7b01587 J. Am. Chem. Soc. 2017, 139, 8388−8391

Communication

Journal of the American Chemical Society We have shown in this work a first proof-of-principle of the application of Ln3+-based NIR emission arising from Zn2+/Ln3+ MCs for optical biological imaging. We have previously established NIR imaging of living cells under single photon excitation by using polymetallic Ln3+-based macromolecules or nanomaterials containing a large number and high density of Ln3+ and sensitizers per unit volume.23,24 The MCs created in this project contain only one emissive lanthanide(III) ion but possess significantly larger quantum yield values than previously reported Ln3+-based NIR emitting systems used for optical imaging. In addition to outstanding photophysical properties and in comparison to previous generations of Zn2+/Ln3+ MCs,27,28 this family of MCs is (i) noncytotoxic at the concentrations required for sensitive optical detection, allowing their use during experiments requiring longer incubation times (hours, days), (ii) sufficiently soluble and robust in water and cell culture media being both chemically stable and photostable under biological conditions. Another element of novelty is the ability of MCs to accumulate preferentially in necrotic cells. Because of a combination of significantly higher photostability, efficient emission in the NIR, and large differences between excitation and emission maxima, these MCs have great potential to be an alternative to the widely used visible fluorescent marker of cell necrosis, propidium iodide. These compounds and achievements are promising for the development of future probes for the specific NIR imaging of necrosis ex vivo (in tissues) or in vivo (in small animals) where the biological autofluorescence is currently a major limitation.

■

acknowledges support from the Institut National de la Santé et de la Recherche Médicale (INSERM).

■

(1) Liu, W.; Sun, P.; Yang, L.; Wang, J.; Li, L.; Wang, J. Microfluid. Nanofluid. 2010, 9, 717. (2) Majno, G.; Joris, I. Am. J. Pathol. 1995, 146, 3. (3) Ntziachristos, V. Nat. Methods 2010, 7, 603. (4) Vanden Berghe, T.; Grootjans, S.; Goossens, V.; Dondelinger, Y.; Krysko, D. V.; Takahashi, N.; Vandenabeele, P. Methods 2013, 61, 117. (5) Sawai, H.; Domae, N. Biochem. Biophys. Res. Commun. 2011, 411, 569. (6) Resch-Genger, U.; Grabolle, M.; Cavaliere-Jaricot, S.; Nitschke, R.; Nann, T. Nat. Methods 2008, 5, 763. (7) Luo, S.; Zhang, E.; Su, Y.; Cheng, T.; Shi, C. Biomaterials 2011, 32, 7127. (8) Guo, Z.; Park, S.; Yoon, J.; Shin, I. Chem. Soc. Rev. 2014, 43, 16. (9) Pansare, V.; Hejazi, S.; Faenza, W.; Prud’homme, R. K. Chem. Mater. 2012, 24, 812. (10) Weissleder, R.; Ntziachristos, V. Nat. Med. 2003, 9, 123. (11) Frangioni, J. V. Curr. Opin. Chem. Biol. 2003, 7, 626. (12) Weissleder, R. Nat. Biotechnol. 2001, 19, 316. (13) Martinić, I.; Eliseeva, S. V.; Petoud, S. J. Lumin. 2017, 189, 19. (14) Dong, H.; Du, S. R.; Zheng, X. Y.; Lyu, G. M.; Sun, L. D.; Li, L. D.; Zhang, P. Z.; Zhang, C.; Yan, C. H. Chem. Rev. 2015, 115, 10725. (15) Hemmer, E.; Venkatachalam, N.; Hyodo, H.; Hattori, A.; Ebina, Y.; Kishimoto, H.; Soga, K. Nanoscale 2013, 5, 11339. (16) Sy, M.; Nonat, A.; Hildebrandt, N.; Charbonniere, L. J. Chem. Commun. 2016, 52, 5080. (17) Prodi, L.; Rampazzo, E.; Rastrelli, F.; Speghini, A.; Zaccheroni, N. Chem. Soc. Rev. 2015, 44, 4922. (18) Bünzli, J.-C. G. J. Lumin. 2016, 170, 866. (19) Amoroso, A. J.; Pope, S. J. Chem. Soc. Rev. 2015, 44, 4723. (20) Eliseeva, S. V.; Bünzli, J.-C. G. Chem. Soc. Rev. 2010, 39, 189. (21) Bui, A. T.; Grichine, A.; Brasselet, S.; Duperray, A.; Andraud, C.; Maury, O. Chem. - Eur. J. 2015, 21, 17757. (22) D’Aléo, A.; Bourdolle, A.; Brustlein, S.; Fauquier, T.; Grichine, A.; Duperray, A.; Baldeck, P. L.; Andraud, C.; Brasselet, S.; Maury, O. Angew. Chem., Int. Ed. 2012, 51, 6622. (23) Foucault-Collet, A.; Gogick, K. A.; White, K. A.; Villette, S.; Pallier, A.; Collet, G.; Kieda, C.; Li, T.; Geib, S. J.; Rosi, N. L.; Petoud, S. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 17199. (24) Foucault-Collet, A.; Shade, C. M.; Nazarenko, I.; Petoud, S.; Eliseeva, S. V. Angew. Chem., Int. Ed. 2014, 53, 2927. (25) Lah, M. S.; Pecoraro, V. L. J. Am. Chem. Soc. 1989, 111, 7258. (26) Mezei, G.; Zaleski, C. M.; Pecoraro, V. L. Chem. Rev. 2007, 107, 4933. (27) Jankolovits, J.; Andolina, C. M.; Kampf, J. W.; Raymond, K. N.; Pecoraro, V. L. Angew. Chem., Int. Ed. 2011, 50, 9660. (28) Trivedi, E. R.; Eliseeva, S. V.; Jankolovits, J.; Olmstead, M. M.; Petoud, S.; Pecoraro, V. L. J. Am. Chem. Soc. 2014, 136, 1526. (29) Beeby, A.; Faulkner, S.; Parker, D.; Williams, J. A. G. J. Chem. Soc. Perkin Trans. 2 2001, 1268. (30) Faulkner, S.; Beeby, A.; Carrié, M.-C.; Dadabhoy, A.; Kenwright, A. M.; Sammes, P. G. Inorg. Chem. Commun. 2001, 4, 187. (31) Shannon, R. D. Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr. 1976, 32, 751. (32) Bünzli, J.-C. G.; Eliseeva, S. V. In Comprehensive Inorganic Chemistry II; Yam, V. W.-W., Ed.; Elsevier B.V.: Amsterdam, 2013; Vol. 8, p 339. (33) Bünzli, J.-C. G. Coord. Chem. Rev. 2015, 293−294, 19. (34) O’Brien, J.; Wilson, I.; Orton, T.; Pognan, F. Eur. J. Biochem. 2000, 267, 5421. (35) El Mjiyad, N.; Caro-Maldonado, A.; Ramirez-Peinado, S.; Munoz-Pinedo, C. Oncogene 2011, 30, 253. (36) Rieger, A. M.; Nelson, K. L.; Konowalchuk, J. D.; Barreda, D. R. J. Visualized Exp. 2011, 50, e2597. (37) Jiang, H.; Luo, S.; Li, H. J. Biol. Chem. 2005, 280, 20651.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b01587. Experimental details (PDF) Video recorded during photobleaching (AVI) Video recorded during photobleaching (AVI)

■

REFERENCES

AUTHOR INFORMATION

Corresponding Authors

*[email protected] *[email protected] *[email protected] ORCID

Vincent L. Pecoraro: 0000-0002-1540-5735 Stéphane Petoud: 0000-0003-4659-4505 Present Address §

Department of Inorganic, Analytical and Applied Chemistry, University of Geneva, CH-1211 Geneva 4, Switzerland Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research has received funding from the European Community’s Marie Curie Seventh Framework Programme (FP7/2007-2013: no. 316906 (ITN Luminet) and no. 611488 (IRSES Metallacrowns)) and in part by La Ligue Contre le Cancer, La Region Centre, Réseau Canaux Ioniques du Cancéropôle du Grand Ouest, Agence Nationale de la Recherche (Lumiphage ANR-13-BSV5-009 and Lumzif ANR12-BS07-0012) and the U.S. National Science Foundation (CH-1361779). I.M. thanks University of Orléans and S.P. 8391

DOI: 10.1021/jacs.7b01587 J. Am. Chem. Soc. 2017, 139, 8388−8391