Synthesis and structural characterization, by spectroscopic and computational methods, of two fluorescent 3-hydroxy-4-pyridinone chelators bearing sulphorhodamine B and naphthalene

Abstract

Compounds MRB2 and MRB4 are part of a set of 3-hydroxy-4-pyridinone fluorescent chelators designed to be used as antimicrobial compounds. Their formulae and structures have been characterized by mass spectrometry, elemental analysis, nuclear magnetic resonance and X-ray crystallography. Identification of the most stable conformations of both compounds in an aqueous environment was achieved by molecular dynamics simulations. To the best of our knowledge, this is the first study in which: (a) a crystal structure of a rhodamine labelled 3-hydroxy-4-pyridinone chelator is reported and (b) a prediction of the range of conformations that these chelators may exhibit in solution is performed. The most frequent and energetically favoured geometries adopted by MRB2 and MRB4 have the chelating group facing the benzene and naphthalene groups of each compound, respectively. This molecular arrangement promotes the establishment of π–π interactions between both rings, which stabilize the overall conformations.

---

Introduction

Our group has been working on the design of 3-hydroxy-4-pyridinone (3,4-HPO) ligands to be of use in biomedical and analytical applications.^1–4 Fluorescent chelators have been prepared by conjugating 3-hydroxy-4-pyridinone chelating units with fluorophores derived from different families of molecules such as naphthalene,⁵ naphthalimide,⁶ fluorescein⁷ and rhodamine.^7–10 The different nature and properties of the fluorophores and chelating 3,4-HPO units permits the synthesis of fluorescent chelators that exhibit diverse physicochemical properties, namely the fluorescence emission spectrum, charge at physiological pH and hydrophilic/lipophilic balance.

In view of the importance of iron for the growth of all bacteria, fungi and protozoa,¹¹ consistent work has been performed in order to design 3,4-HPO iron(III) chelators that may find application in the development of new strategies to fight infection based on the concept of iron deprivation.^7,10,12 Numerous examples of the use of iron chelation therapies and the consequently improvement in infection susceptibility of several microorganisms have been reported in the literature.^11,13–21 Mycobacterial infections have been our object of study, in particular those caused by the opportunistic infectious pathogen Mycobacterium avium that affects mainly immune compromised patients.²² Several studies have been reported in the literature considering the use of iron chelators to restrict the iron available for mycobacterial growth,²³ namely for Mycobacterium tuberculosis^24,25 and M. avium.^7,10,12,26 Our latest results have shown that all rhodamine labelled chelators were capable of restricting the intramacrophagic growth of M. avium and their activity is strongly dependent on the presence and type of fluorophore bound to the 3,4-HPO chelating unit. Specifically, the most relevant inhibitory effect was obtained for ligands including in its structure N-ethyl groups on the xanthene ring and a thiourea linkage between the rhodamine and the chelating unit.^10,12 Several studies have been performed to gain insight on structure–activity relationships and we found that a differential interaction of the fluorescent chelators with biological membranes seems to be determinant for their antimycobacterial effect.^10,27,28 Moreover, NMR and MD studies showed that chelator's interaction with bilayers occurs through both the fluorophore and the chelating moieties of the molecular framework.²⁸

To increase our knowledge about the repercussion of the type of fluorophore bound to the 3,4-HPO chelating unit on molecular properties, we designed two new compounds: (i) in MRB2 we used sulphorhodamine B thus maintaining the N-ethyl groups but changing the linkage to the chelating unit and (ii) in MRB4 we used a naphthalene fluorophore bound to the 3,4-HPO chelating unit through a thiourea linkage.

In the present work we report the synthesis and structural characterization of chelators MRB2 and MRB4 by using a combined approach that employs experimental methodologies, such as NMR and X-ray diffraction.

Once the compounds are meant to be used as antibiotics in a cellular environment, structural information in aqueous solution is relevant for the understanding of their activity. Computational methodologies, namely molecular dynamics (MD) simulations were performed to get insight into the most populated conformers of chelators MRB2 and MRB4. MD simulations provide additional insights into the structure of the compounds, through its ability to characterize the conformations that the compounds span in water.

To the best of our knowledge, this is the first study in which: (a) a crystal structure of a rhodamine labelled 3,4-HPO chelator is reported and (b) a prediction of the range of conformations that these molecules may exhibit in solution is performed.

The determination of the most stable conformations may be relevant to predict and understand the tri-dimensional arrangement and get insight about the mode of binding of the chelators with the metal ions in solution, namely iron(III). Therefore, molecular dynamics simulations were also performed in order to better characterize the different conformations and energies of compounds MRB2 and MRB4 expected to occur in an aqueous environment at physiological pH conditions.

Experimental

Materials and methods

General information. Chemicals were obtained from Sigma-Aldrich (grade puriss, p.a.) or Fluka (p.a.) and were used as received unless otherwise specified.

NMR spectra were recorded on a Bruker Avance III 400, operating at 400.15 MHz for ¹H and 100.62 MHz for ¹³C atoms, equipped with pulse gradient units, capable of producing magnetic field pulsed gradients in the z-direction of 50.0 G cm⁻¹ or on a Bruker Avance III two-dimensional ¹H/¹H correlation spectra (COSY), gradient selected ¹H/¹³C heteronuclear single quantum coherence (HSQC) and ¹H/¹³C heteronuclear multiple bond coherence (HMBC) spectra were acquired using the standard Bruker software. NMR analyses were performed at Laboratory of Structural Analysis, Centre of Studies of Materials of University of Porto (CEMUP) (Portugal).

Elemental analysis and Mass spectrometry were performed at the analytical services of University of Santiago (Spain).

Synthesis of 3-hydroxy-4-pyridinone ligands

The fluorescent bidentate ligands were prepared by the coupling of fluorophore (rhodamine or naphthalene derivatives) to a protected 3-hydroxy-4-pyridinone (3,4-HPO) bidentate ligand.¹⁰ The bidentate unit was synthesized in our laboratory following the procedures described in the literature.²⁹ A diagram of the 3,4-HPO bidentate unit and fluorophores used to obtain the final fluorescent ligands synthesized in this work and the synthetic procedures are outlined in Fig. 1.

Fig. 1 Reaction scheme for the synthesis of 3,4-HPO ligands MRB2 and MRB4.

The reactions were performed in the presence of anhydrous N,N-dimethylformamide (DMF) (and triethylamine for ligand 1) to produce 1 and 3, followed by deprotection with BCl₃ to yield the final fluorescent bidentate 3,4-HPO ligands MRB2 and MRB4.

Compound 1. To a solution of sulphorhodamine B acid chloride (0.107 g, 1.86 × 10⁻⁴ mol) in anhydrous DMF (0.7 mL) the 3,4-HPO bidentate unit (0.040 g, 1.54 × 10⁻⁴ mol) was added and the mixture was stirred at room temperature in the dark and under argon atmosphere, for 24 h. The product was purified by gradient flash column chromatography, eluting with chloroform/methanol (9 : 1) and increasing polarity until chloroform/methanol (6 : 4) to afford 1 (0.030 g; 24%) as a purple solid.

MS: calculated for C₄₂H₄₇N₄O₈S₂⁺: MS: 799.28: [M⁺], found: matrix-assisted laser desorption/ionization time of flight MS: 799.27 [M⁺]. ¹H NMR (400.15 MHz, MeOD-d4, ppm): δ 1.31 (t, J 7.2 Hz, 12H, NCH₂CH̲₃); 2.42 (s, 3H, 6′′-CH̲₃); 3.69 (t, J 7.2 Hz, 8H, NCH̲₂CH₃); 3.76 (s, 3H, NCH₃); 4.24 (s, 2H, CH̲₂NH); 5.12 (s, 2H, CH̲₂C₆H₅); 6.44 (s, 1H, H-5′′); 6.94 (d, J 2.4 Hz, 2H, H4 + H5); 7.01–7.04 (dd, J 2.0, 10.4 Hz, 2H, H2 + H7); 7.06 and 7.08 (d, J 10.4 Hz, 2H, H1 + H8); 7.25–7.32 (m, 3H, m-+p- CH₂C₆H̲₅); 7.35–7.38 (m, 2H, o-CH₂C₆H̲₅ + 1H, H6′); 8.02 (dd, J 5.0,1.6 Hz, 1H, H5′); 8.58 (d, J 1.6 Hz, H3′); ¹³C NMR (100.62 MHz, MeOD-d4, ppm): δ 12.8 (NCH₂C̲H₃); 21.0 (6′′-C̲H₃); 37.3 (NC̲H₃); 39.5 (C̲H₂NH); 46.8 (NC̲H₂CH₃); 74.6 (–C̲H₂C₆H₅); 97.0 (C4 + C5); 115.1 (C2 + C7); 119.1 (C5′′); 127.6 (C3′); 129.5 (C5′); 129.6 (m-, p-C̲₆H₅); 130.2(C6′); 133.5(o-C̲₆H₅); 133.8 (C1 + C8); 135.8 (C4′); 138.2 (Cq, –C̲₆H₅); 142.0 (C2′′); 143.0 (C1′); 147.4 (C3′′); 151.5 (C6′′); 157.2 (C3 + C6); 159.4 (C1a + C8a); 174.6 (C4′′).

Compound MRB2. Compound 1 (0.029 g, 3.63 × 10⁻⁵ mol) was dissolved in anhydrous dichloromethane (20 mL), under argon and cooled to 0 °C. BCl₃ (1 mL) was added dropwise and the reaction mixture was kept overnight with stirring at room temperature. Methanol (50 mL) was added and the mixture was stirred for 1 h. The solid product formed was removed by filtration and the solvent was removed under reduced pressure to afford the crude product. Recrystallization of the product from methanol/diethyl ether (2 : 8) afforded the hydrochloride salt MRB2 (0.0244 g, 90%) as a purple solid.

Elemental analysis for (C₃₅H₄₀N₄O₈S₂.1.5C₃H₇NO·3.5H₂O·2CHCl₃·3HCl): calculated C 40.54; H 5.12; N 6.27. Found: C 40.41; H 4.78; N 6.25. MS: calculated for C₃₅H₄₁N₄O₈S₂⁺: MS: 709.23: [M⁺], found: matrix-assisted laser desorption/ionization time of flight MS: 709.13 [M⁺]. ¹H NMR (400.15 MHz, MeOD-d4, ppm): δ 1.30 (m, 12H, NCH₂CH̲₃); 2.67 (s, 1.5H, 6′′-CH̲_3, stereoisomer A); 2.70 (s, 1.5H, 6′′-CH̲_3, stereoisomer B); 3.68 (q, J 10.8, 7.2 Hz, 8H, NCH̲₂CH₃); 4.13 (s, 3H, NCH₃); 4.66 (s, 2H, CH̲₂NH); 6.95 (d, J 2.4 Hz, 2H, H4 + H5); 7.00–7.03 (dd, J 2.4, 11.6 Hz, 2H, H2 + H7); 7.08–7.11 (d, J 11.6 Hz, 2H, H1 + H8); 7.09 (s, 1H, H-5′′); 7.54 (d, J 8.00 Hz, 1H, H6′); 8.02 (dd, J 5.0, 3.0 Hz, 1H, H5′); 8.56 (d, J 1.6 Hz, H3′); ¹³C NMR (100.62 MHz, MeOD-d4, ppm): δ 12.3 (NCH₂C̲H₃); 21.7 (6′′-C̲H_3, stereoisomer A); 34.9 (6′′-C̲H_3, stereoisomer B); 38.2 (C̲H₂NH); 39.3 (NC̲H₃); 46.3 (NC̲H₂CH₃); 96.5 (C4 + C5); 113.5 (C5′′); 114.6 and 114.8 (C2 + C7); 127.2 (C3′); 129.1 (C5′); 132.1 (C6′); 133.1 (C1 + C8); 135.5 (C4′); 139.0 (C2′′); 142.3 (C1′, stereoisomer A); 144.5 (C3′′); 146.8 (C1′, stereoisomer B); 150.5 (C6′′); 156.7 (C3+C6); 157.0 (C9); 158.9(C4a + C5a); 160.0 (C1a + C8a). UV-Vis (λ_max/nm) 566; ε = 3.8 × 10⁴ mol⁻¹ dm³ cm⁻¹; fluorescence (λ_max/nm) 586.

Compound 3. To a solution of 1-naphthyl isothiocyanate (0.034 g, 1.85 × 10⁻⁴ mol) in anhydrous DMF (0.6 mL) and triethylamine (0.01 mL), 3,4-HPO bidentate unit (0.040 g, 1.8 × 10⁻⁴ mol) was added and the mixture was stirred at room temperature in the dark and under argon atmosphere, for 24 h. The product was purified by gradient flash column chromatography, eluting with chloroform/methanol (9 : 1) to afford 3 (0.052 g; 76%) as a white solid.

MS: calculated for C₂₆H₂₆N₃O₂S⁺: MS: 444.17: [M⁺], found: matrix-assisted laser desorption/ionization time of flight MS: 444.19 [M⁺]. ¹H NMR (400.15 MHz, CDCl₃, ppm): δ 2.25 (s, 3H, 6′-CH̲₃); 3.66 (s, 3H, NCH₃); 4.90 (d, J = 5.6 Hz, 2H, CH̲₂NH); 4.99 (s, 2H, CH̲₂C₆H₅); 6.30 (s, 1H, H-5′); 7.16–7.18 (m, 3H, m-+p-CH₂C₆H̲₅); 7.22–7.26 (m, 2H, o-CH₂C₆H̲₅); 7.47 (t, J = 7.8 Hz, 1H, H3); 7.50–7.55 (m, 2H, H6 + H7); 7.59 (d, J = 7.2 Hz, 1H, H2); 7.80 (d, J = 8.4 Hz, 1H, H4); 7.88–7.91 (m, 1H, H5); 8.07 (d, J = 9.2 Hz, 1H, H8); ¹³C NMR (100.62 MHz, CDCl₃, ppm): δ 21.9 (6′-C̲H₃); 38.2 (NC̲H₃); 41.9 (C̲H₂NH); 74.2 (–C̲H₂C₆H₅); 119.8 (C5′); 123.3 (C8); 125.9 (C4); 126.5 (C3); 127.4 (C6); 127.7 (C7); 128.8 (C2); 129.0 (m-+p-C̲₆H₅); 129.3 (o-C̲₆H₅); 129.5 (C5); 130.8 (C8a); 135.4 (C4a); 137.8 (Cq, –C̲₆H₅); 141.7 (C2′); 147.6 (C3′); 148.3 (C6′); 173.9 (C4′); 183.5 (NC̲S).

Compound MRB4. Compound 3 (0.052 g, 1.17 × 10⁻⁵ mol) was dissolved in anhydrous dichloromethane (20 mL) under argon and cooled to 0 °C. BCl₃ (1 mL) was added dropwise and the reaction mixture was kept overnight with stirring at room temperature. Methanol (50 mL) was added and the mixture was stirred for 1 h. The solid product formed was removed by filtration and the solvent was removed under reduced pressure to afford the solid product. Recrystallization of the product from chloroform/hexane (2 : 8) afforded the hydrochloride salt MRB4 (0.044 g, 96%) as a light brown solid.

Elemental analysis for (C₁₉H₂₀N₃O₂S.0.33CH₃OH.1.33HCl): calculated C 56.13; H 5.52; N 10.16. Found: C 56.18; H 5.40; N 9.78. MS: calculated for C₁₉H₂₀N₃O₂S⁺: MS: 354.13: [M⁺], found: matrix-assisted laser desorption/ionization time of flight MS: 354.16 [M⁺]. ¹H NMR (400.15 MHz, MeOD-d4, ppm): δ 2.58 (s, 3H, 6′-CH̲₃); 4.00 (s, 3H, NCH₃); 5.16 (s, 2H, CH̲₂NH); 6.96 (s, 1H, H-5′); 7.49 (dd, J = 1.2, 4.2 Hz, 1H, H2); 7.54 (t, J = 8.4 Hz, 1H, H3); 7.54–7.66 (m, 2H, H6 + H7); 7.91 (d, J = 7.6 Hz, 1H, H4); 7.93–7.96 (m, 2H, H5 + H8); ¹³C NMR (100.62 MHz, MeOD, ppm): δ 20.6 (6′-C̲H₃); 39.6 (NC̲H₃); 41.3 (C̲H₂NH); 113.3 (C5′); 123.1 (C4); 126.1 (C2); 126.4 (C3); 127.1 (C6); 127.4 (C7); 128.9 (C5); 129.0 (C8); 131.1 (C8a); 134.1 (C1); 135.7 (C4a); 140.5 (C2′); 144.4 (C3′); 149.6 (C6′); 160.6 (C4′); 183.4 (NC̲S). UV-Vis (λ_max/nm) 312; ε = 1.5 × 10⁴ mol⁻¹ dm³ cm⁻¹; fluorescence (λ_max/nm) 385.

Electronic spectroscopy: absorption and fluorescence measurements

Electronic absorption measurements were performed in a Perkin Elmer Lambda 25 spectrophotometer equipped with a constant-temperature cell holder, using the conditions T = 25 °C, l = 1 cm cuvettes and wavelength range 225–650 nm.

Fluorescence measurements were performed in a Varian Cary Eclipse spectrofluorimeter, equipped with a constant-temperature cell holder, using the conditions T = 25 °C, l = 1 cm cuvettes. Spectra were recorded at: λ_exc = 568 nm and λ_em from 570 to 700 nm, 600 V, excitation and emission slit widths of 5 nm for ligand MRB2 and λ_exc = 290 nm and λ_em from 300 to 550 nm, 600 V, excitation and emission slit widths of 10 nm for ligand MRB4.

Stock solutions of the different compounds were obtained by preparing a concentrated solution of the compound in dimethylsulfoxide (DMSO). Samples for absorption and fluorescence measurements were prepared by dilution of a known volume of the DMSO stock solution in 3-(N-morpholino)propanesulfonic acid (MOPS) buffer solution (pH 7.4, I = 0.1 M NaCl). The percentage of the DMSO stock solution was always less than 2% in the final volume.

Single-crystal X-ray diffraction

Crystalline material of the compounds 1, MRB2, 3 and MRB4, were manually harvested and an appropriate crystal of each compound mounted on cryoloops using appropriate inert oil.³⁰ Diffraction data were collected at 180.0(2) K on a Bruker X8 Kappa APEX II Charge-Coupled Device (CCD) area-detector diffractometer controlled by the APEX2 software package³¹ (Mo K_α graphite-monochromated radiation, λ = 0.71073 Å), and equipped with an Oxford Cryosystems Series 700 cryostream monitored remotely with the software interface Cryopad.³² Images were processed with the software SAINT+,³³ and absorption effects corrected with the multi-scan method implemented in SADABS.³⁴ The structure were solved by direct or Patterson methods employed in SHELXS-97,^35,36 allowing the immediate location of the Sulphur atoms. Remaining non-H-atoms of the compounds were located from difference Fourier maps calculated by successive full-matrix least-squares refinement cycles on F² using SHELXL-97,^36,37 and have been successfully refined with anisotropic displacement parameters.

Hydrogen atoms attached to carbon atoms were placed at their geometrical positions using appropriate HFIX instructions in SHELXL and incorporated in subsequent refinement cycles in riding-motion approximation with isotropic thermal displacements parameters (U_iso) fixed at 1.2 or 1.5 × U_eq of the parent atom. Furthermore, the hydrogen atoms of the hydroxyl, sulfonamide and water molecules (in compound MRB2) were clearly visible in the difference Fourier maps, and included in subsequent refinement stages with the distances adequately restrained. In the structure of 1, a considerable electron density, mainly due to disordered solvent molecules, was not possible to modulate and refine properly. Searches for the total potential solvent area using the software package PLATON^38,39 revealed the existence of cavities with potential solvent accessible voids. The original data sets were then treated with the SQUEEZE⁴⁰ routine to remove the contribution of these disordered molecules in the solvent-accessible volume, and the calculated solvent-free reflection list was consequently used for the final structure refinement. Crystal and structure refinement data for compounds 1, MRB2, 3 and MRB4 is summarized in Table 1.

Table 1 Crystal and structure refinement data for compounds 1, MRB2, 3 and MRB4

	1	MRB2	3	MRB4
Formula	C42H46N4O8S2	C35H51N4O13S2	C26H25N3O2S	C19H19N3O2S
Mr/g mol^−1	798.95	835.36	443.55	353.43
Crystal morphology	Pink prism	Pink prism	Colourless plate	Colourless needle
Crystal size/mm	0.35 × 0.21 × 0.15	0.16 × 0.11 × 0.04	0.22 × 0.06 × 0.01	0.20 × 0.03 × 0.02
Crystal system	Triclinic	Triclinic	Monoclinic	Triclinic
Space group	P1̄	P1̄	P21/n	P1̄
a/Å	12.027(2)	8.5844(6)	13.8764(5)	7.3122(4)
b/Å	13.158(2)	12.4376(9)	9.0072(4)	10.9139(6)
c/Å	16.691(3)	19.9706(14)	18.6937(8)	11.3222(6)
α/°	111.715(8)	77.945(4)	90	102.644(2)
β/°	95.652(8)	87.615(4)	106.834(2)	95.560(2)
γ/°	94.106(8)	86.940(4)	90	106.190(2)
Volume/Å^3	2425.1(7)	2081.2(3)	2236.4(2)	834.58(8)
Z	2	2	4	2
ρcalculated/g cm^−3	1.094	1.333	1.317	1.406
F(000)	844	884	936	372
μ/mm^−1	0.158	0.257	0.174	0.212
θ range/°	3.65 to 26.32	3.71 to 25.03	3.72 to 26.37	3.74 to 25.03
Index ranges	−15 ≤ h ≤ 14	−9 ≤ h ≤ 10	−17 ≤ h ≤ 17	−8 ≤ h ≤ 8
	−16 ≤ k ≤ 16	−14 ≤ k ≤ 14	−11 ≤ k ≤ 11	−12 ≤ k ≤ 12
	−20 ≤ l ≤ 20	−23 ≤ l ≤ 23	−23 ≤ l ≤ 23	−13 ≤ l ≤ 10
Reflections collected	73 045	32 425	29 679	12 323
Independent reflections	9863 (Rint = 0.0530)	7259 (Rint = 0.0452)	4569 (Rint = 0.0644)	2934 (Rint = 0.0371)
Final R indices [I > 2σ(I)]	R1 = 0.0520; wR2 = 0.1439	R1 = 0.0598; wR2 = 0.1457	R1 = 0.0469; wR2 = 0.0902	R1 = 0.0383; wR2 = 0.0811
Final R indices (all data)	R1 = 0.0638; wR2 = 0.1509	R1 = 0.0869; wR2 = 0.1626	R1 = 0.0959; wR2 = 0.1050	R1 = 0.0588; wR2 = 0.0898
Largest diff. peak and hole/e Å^3	0.386 and −0.271	1.049 and −0.467	0.235 and −0.258	0.199 and −0.258

---

Geometry optimization and molecular dynamics

The initial geometries of both MRB2 and MRB4 molecules were built with the GaussView software.⁴¹ To calculate the optimized geometries and electronic properties, to be employed later in the parameterization of these compounds, we used the Gaussian09 suite of programs.⁴² Restricted Hartree–Fock (RHF) calculations with the 6-31G(d) basis set were performed. The atomic charges were further recalculated using RESP.⁴³ This methodology was chosen to be consistent with that adopted for the parameterization process in the AMBER 12 software.⁴⁴ Molecular dynamics (MD) simulations were performed for each molecule, using the general AMBER force field for small organic molecules (GAFF).⁴⁵ In these simulations, an explicit solvation model with pre-equilibrated TIP3P water molecules was used, filling a truncated rectangular box with a minimum distance of 12 Å between the box edges and any atom of each compound. The MRB2 and MRB4 system sizes were 5804 and 3755 atoms, respectively. Each system was minimized in two stages. In the first stage, the compounds were kept fixed and only the position of the water molecules was optimized. In the second stage, the full system was optimized. Subsequently, an equilibration of 100 ps at constant volume and temperature, and considering periodic boundaries conditions was performed. This was followed by 80 ns simulation (production MD) with the NPT ensemble, in which Langevin dynamics was used (collision frequency of 1.0 ps⁻¹) to control the temperature at 303.15 K.⁴⁶ All simulations were carried out using the AMBER 12.0 simulations package.⁴⁴ Bond lengths involving hydrogen atoms were constrained using the SHAKE algorithm, and the equations of motion were integrated with a 2 fs time step using the Verlet leapfrog algorithm.⁴⁷ The Particle-Mesh Ewald (PME) method⁴⁸ was used to treat long-range interactions, and the non-bonded interactions were truncated with a 10 Å cut-off. The MD trajectories were saved every 2 ps and were analysed with the PTRAJ module of AMBER 12.0.⁴⁴ The clustering analysis of the production MD trajectory was performed by using the average-linkage algorithm.⁴⁹ A total of ten clusters were produced using the pairwise root mean square deviation (RMSd) between frames as a metric comparing the atoms of each chelating compound. The representative structures of each cluster were used for further energy calculations. The latter were determined at the molecular mechanics (MM) level using the AMBER simulation package. The Generalized Born implicit solvent was used to evaluate the effect of the aqueous environment with a dielectric constant of ε = 80. Relevant dihedrals were also analysed being depicted in Fig. 2.

Fig. 2 Representation of molecules MRB2 and MRB4. Depicted are relevant torsions considered in this study (ω1–5) for each compound.

Results and discussion

Synthesis and NMR spectroscopy

The new fluorescent ligands MRB2 and MRB4 were prepared using straightforward synthetic protocols involving the coupling of the commercially available fluorophores, sulphorhodamine B acid chloride or 1-naphthyl isothiocyanate to the protected 3,4-HPO bidentate units. The last step is the reaction for the removal of the protecting benzyl group (deprotection) with BCl₃, as depicted in Fig. 1.

The comparison of the results obtained for the synthesis of 1 and 3 demonstrate that the reaction of sulphorhodamine B acid chloride with the 3,4-HPO ligand allowed obtaining ligand 1 in a 24% yield and in the reaction of 1-naphthyl isothiocyanate with the same bidentate unit we obtained ligand 3 in 76% yield. Although both synthetic procedures are very similar, the yield of the reactions is different and this fact could be related with the functional group of the fluorophore that reacts with the amine group of the 3,4-HPO unit. In the deprotection reactions, no major differences are observed and ligands MRB2 and MRB4 were successfully synthetized and the yield of reactions are respectively 90 and 96%.

The structures of the protected (1 and 3) and de-protected ligands (MRB2 and MRB4) in solution were established by NMR analysis (¹H and ¹³C, 1D (Fig. S1–S8†) and 2D experiments, including COSY, HSQC and HMBC spectra for unequivocal assignment of the most characteristic proton and carbon chemical shifts). The assignment of the resonance signals in ¹³C NMR spectra of the protected and de-protected compounds was achieved by analysis of ¹H/¹³C HSQC and ¹H/¹³C HMBC spectra, which provide one and multiple bond ¹H–¹³C connectivity.

The ¹H and ¹³C NMR spectra of ligand 1 revealed that the resonance signals of the methyl and methylene protons of the rhodamine residue appear at 1.31 and 3.69 ppm and those of H1–H8 protons in the spectra are between 7.06 and 7.08 ppm.

The protons of the methyl group of the pyridinone ring appear at 2.42 ppm and the protons of the methyl linked to the nitrogen of the ring appear at 3.76 ppm. The signal at 4.24 ppm is attributed to the CH̲₂NH protons and show HMBC correlation with a carbon at 142.0 ppm assigned as C2′′and C3′′ that appears at 147.4 ppm. The protons of di-substituted aromatic ring of rhodamine, H3′, H5′ and H6′, appear respectively at 8.58, 8.02 and 7.35–7.38 ppm.

The signals related with the protecting groups are the singlet at 5.12 ppm that corresponds to the protons of the methylene group and the protons of the benzyl ring appear between 7.25–7.38 ppm. The carbon associated to this methylene group is at 74.6 ppm, the quaternary carbon appears at 138.2 ppm and the last 5 carbons of the benzyl ring appear between 129.6 and 133.5 ppm. After the deprotection (ligand MRB2), significant differences in the ¹H and ¹³C spectra are detected as for example the shift of the protons of the methyl group bound to the nitrogen of the pyridinone from 3.76 to 4.13 ppm as also the shift in the protons of the methylene group of the linkage (CH̲₂NH) from 4.24 to 4.66 ppm. Their respective carbons are also dislocated to the low field region, namely NC̲H₃ from 37.3 in the protected ligand and 39.3 in the deprotected form and the shift from 39.5 to 38.2 ppm for the carbon of the linkage (CH̲₂NH).

Considering the ¹H and ¹³C NMR spectra of ligand 3, the characteristic protons of methyl groups of pyridinone ring (6′-CH₃ and NCH₃) appear, respectively, at 2.25 and 3.66 ppm. The doublet at 4.90 is attributed to the protons of the methylene group of the linkage between the 3,4-HPO and the naphthalene.

The signals related with the protecting groups are the singlet at 4.99 ppm that corresponds to the protons of the methylene group and the protons of the benzyl ring appear between 7.16–7.26 ppm. The carbon associated to this methylene group appears at 74.2 ppm, the quaternary carbon at 137.8 ppm and the last 5 carbons of the benzyl ring appear at 129.0 and 129.3 ppm. The carbons of the naphthalene ring appear between 123.3 and 129.5 ppm and the quaternary carbons, C8a and C4a, appear at 130.8 and 135.4 ppm, respectively. The signal at 183.5 ppm is attributed to the carbon of SC̲N group and it shows long range couplings HMBC with carbons C6′ and C5′(141.7 and 147.6 ppm) and with the protons attributed to the methylene group of the linkage (CH̲₂NH) that appear at 4.90 ppm in the ¹H spectrum.

The deprotection reaction was successfully achieved as confirmed by the changes of the chemical shift of characteristics protons and carbons of the ligand MRB4. In the ¹H spectrum of ligand MRB4, of the characteristic protons of methyl groups of pyridinone ring (6′-CH₃ and NCH₃) that appear, respectively, at 2.58 and 4.00 ppm. The resonance signals attributed to the protons of the methylene group of the linkage between the 3,4-HPO and the naphthalene are shifted to the low field region at 5.16 ppm. The carbons of the pyridinone ring that were most directly affected by de deprotection, C3′, C4′ and C5′, are also shifted from 119.8, 173.9 and 147.6 to 113.3, 160.6 and 144.4 ppm, respectively.

The observed differences in the chemical shifts of ¹H and ¹³C nuclei in the spectra of the protected and deprotected ligands are primarily due to the deprotection of the hydroxyl group and the acidic pH of the reaction medium that allow us to obtain the final ligand in enol form as described previously.^5,10

X-ray crystallography

Single-crystals of the rhodamine derivative compounds 1 and MRB2 suitable for X-ray diffraction analyses were obtained by controlled recrystallization from MeOH/CHCl₃ and MeOH/Et₂O solvent mixtures, respectively. The two solid-state crystalline structures were determined in the triclinic space group P1̄, confirming the synthesis of the protected and unprotected forms of the desired ligand (Fig. 3).

Fig. 3 Crystal structures of the rhodamine derivative molecules: protected ligand 1 (a) and ligand MRB2 (b) represented in the ball-and-stick model. Hydrogen atoms were omitted for clarity reasons.

The asymmetric unit (asu) of 1 reveals only a neutral organic molecule (as detailed in the experimental part, some solvent molecules highly disordered were removed from the model), while the asu of MRB2 contains one cationic organic molecule, one chloride anion and five water molecules of crystallization. The structural arrangement of ligand MRB2 shows the xanthene group almost perpendicular with the adjacent phenyl ring and the pyridinone ring [dihedral angle between the average planes defined by these aromatic groups is 84.66(2)° and 75.49(2)°, respectively]. Furthermore, the values of the two most crucial dihedrals (ω1 and ω2 represented in Fig. 2) responsible for the conformational variations during each MD simulation (see next section) are 118.91(1)° and 61.21(1)°, respectively.

The contiguous ligands MRB2 close interact by a strong N–H⋯O hydrogen bond involving the sulfonamide and sulfonate groups of adjacent molecules leading to the formation of one-dimensional supramolecular structure (Fig. 4a; see Table 2 for geometric details about the hydrogen bonding interactions). The extended packing of these supramolecular entities leads to an overall porous structure with one-dimensional tunnels along the [1 0 0] direction of the unit cell (Fig. 4b). The porous are entirely occupied by the crystallization water molecules and the charge-balancing chloride anions, which are also involved in an extensive inter-molecular hydrogen-bonding network, namely O–H⋯O, O–H⋯Cl and N–H⋯O interactions (not represented; Table 2).

Fig. 4 Structural features of the crystal structure of the ligand MRB2: (a) one-dimensional supramolecular structure with the N–H⋯O interactions represented as green dashed lines; (b) crystal packing viewed in the [1 0 0] direction of the unit cell with the Cl⁻ anions and crystallization water molecules occupying the channels.

Table 2 Geometric information (distances in Å and angles in degrees) for the D–H⋯A hydrogen bond interactions of the compound MRB2 crystal structure^a

D–H⋯A	d(H⋯A)	d(D⋯A)	∠(DHA)
a Symmetry transformation used to generate equivalent atoms: (i) x + 1, y, z; (ii) −x + 1, −y + 1, −z + 1; (iii) −x, –y + 2, −z − 1; (iv) x − 1, y, z − 1.
N3–H3⋯O4^i	0.885(19)	1.85(2)	171(4)
O7–H7⋯O3W	0.84(2)	1.94(2)	142.6
O8–H8⋯Cl1^ii	0.84(2)	2.19(3)	176.0
O1W–H1W⋯O5W	0.788(16)	2.22(2)	163(3)
O1W–H2W⋯O2W^iii	0.786(16)	2.23(2)	149(3)
O2W–H3W⋯O1W	1.000(19)	1.96(2)	159(5)
O2W–H4W⋯Cl1^iv	0.97(2)	2.12(2)	175(5)
O4W–H7W⋯O2W	0.96(2)	1.67(3)	164(7)

---

To the best of our knowledge, this is the first report of an X-ray crystal structure of a rhodamine derivative 3-hydroxy-4-pyridinone.

Crystalline materials of the naphthalene derivatived molecules 3 and MRB4 with quality for single-crystal X-ray diffraction analyses were isolated by recrystallization from solution of MeOH/CHCl₃ and CHCl₃/n-hexane, respectively. The crystal structure of compound 3 was determined in monoclinic space group P2₁/n, while the structure of ligand MRB4 was unveiled in the triclinic space group P1̄, confirming unequivocally the preparation of the protected and unprotected forms of the desired ligand (Fig. 5).

Fig. 5 Crystal structures of the naphthalene derivatives ligand: protected 3 (a) and deprotected form, MRB4 (b) represented in the ball-and-stick model.

The asu of the two crystal structures comprise only the respective neutral organic molecule. The structural arrangement of ligand MRB4 reveals the xanthene group practically in the same plane of the pyridinone ring [dihedral angle between the average planes defined by the two aromatic groups is 8.62(1)°]. Additionally, in a similar way to that mentioned previously for ligand MRB2, the values of the two dihedrals are 67.226(6)° for ω1 and 63.499(5)° for ω2 (ω1 and ω2 are defined in Fig. 2).

A pair of O–H⋯O hydrogen bonds involving the hydroxypyridinone groups of two adjacent molecules leads to the formation of dimeric entities with the two molecules positioned in anti-parallel mode (O2–H7⋯O1ⁱ, with the distances H7⋯O1ⁱ of 1.809(1) Å and O2⋯O1ⁱ of 2.633(1) Å, and angle O2–H7–O1ⁱ of 146.6(1)°; Fig. 6a). Furthermore, the organic molecules MRB4 also interact by an extensive network of weak non-covalent interactions (not shown), such as C–H⋯O, C–H⋯S, C–H⋯π and long π⋯π stacking which further strengthen the cohesion of the crystalline packing (Fig. 6b).

Fig. 6 Structural features of the crystalline structure of the ligand MRB4: (a) dimeric supramolecular structure with the O–H⋯O interactions represented as green dashed lines; (b) crystal packing viewed in the [1 0 0] direction of the unit.

Studies in an aqueous environment

The 3,4-HPO chelating units, present in the molecular framework of compounds MRB2 and MRB4, possess two dissociable protons whose characteristic pK_a values are centred at 3.2 and 9.6 as previously described by us.⁵ In Fig. S9† we provide the characteristic dissociation equilibria in which the enol and keto forms are represented.

Ligands MRB2 and MRB4 were isolated in the enol form as hydrochloride salts as a consequence of the synthetic route used. In aqueous solution, and in particular at pH 7.4 characteristic of biological fluids, the ligands are present in the keto form.

According to the field of application of the chelators the photophysical characterization was performed at pH 7.4 and the MD simulations were restrained to compounds in the keto form and in an aqueous environment.

Photophysical properties in solution. The absorption UV-Vis spectra of ligands MRB2 and MRB4 show two sets of bands in different regions of the spectra, which correspond to π → π* transitions of the π systems of 3,4-HPO and the fluorophores, rhodamine and naphthalene, respectively, that constitute the ligand structure. Transitions in the range 281–290 nm are associated with ethylene and benzene bonds of the aromatic ring of the 3,4-HPO. The other set of transitions in the range of 580–650 nm are assigned to the π system of sulphorhodamine B group of ligand MRB2. In the case of ligand MRB4, a shoulder at ca. 312 nm is discernible and is assigned to the π system of 1-naphthyl isothiocyanate. The molar absorptivity values obtained are close to those reported in the literature.^5,7,10

The fluorescence spectra of ligands MRB2 and MRB4 show bands in different regions of the spectra that are characteristic for the fluorophore included in the final structure of each ligand. In the case of ligand MRB2 the band appears in the range of 570–670 nm and for ligand MRB4 the band appears in the range of 335–525 nm. The values obtain are in agreement with those reported in the literature for similar ligands.^5,7,10 The spectral parameters for MRB2 and MRB4 are registered in Table 3.

Table 3 Spectral parameters of the fluorescent chelators MRB2, MRB4 (25 °C, MOPS buffer, I = 0.1 M, pH = 7.4)

Chelator	Absorption		Fluorescence
	λmax/nm	ε/mol^−1 dm^3 cm^−1	λmax/nm
			Excitation	Emission
MRB2	566	3.8 × 10^4	568	586
MRB4	312	1.5 × 10^4	290	385

---

Confirmation of the predominance of the keto form at pH 7.4 is particularly evident for compound MRB4 by observation of the characteristic λ_max of fluorescence emission at 385 nm as previously described by our group for other naphthalene labelled 3,4-HPO ligands.⁵

MD simulations. The MD simulations for each chelating molecule allowed for the sampling of their respective conformational space, in order to identify their most stable conformations in an aqueous environment. Fig. 7 shows the root mean square deviation (RMSd) obtained for compounds MRB2 and MRB4 (as compared to the average structure). Both compounds display stable RMSd values, even though conformational shifts are clearly visible. Both MRB2 and MRB4 oscillate between two different RMSd values (ca. 2.4 and 2.0 Å for MRB2; and ca. 1.0 and 1.7 Å for MRB4).

Fig. 7 RMSd values considering the average structure of the MD simulation of compound MRB2 and MRB4.

The root mean score fluctuation (RMSf) gives a measure of the flexibility of a subset of a system related to its average structure over the whole simulation. These values were also calculated for both chelating molecules (by heavy atom), and are displayed in Fig. 8.

Fig. 8 RMSf normalized results (per heavy atom) obtained for compound MRB2 and MRB4. The RMSf results are depicted as a transparent film along the ball-and-stick representation of the molecules. Green corresponds to less flexible regions, whereas red for more flexible regions.

It was observed that the atoms in the most flexible regions are the methyl groups from the rhodamine unit in compound MRB2 and the hydroxyl and methyl groups that belong to the chelating ring in both compounds.

The values of the most crucial dihedrals (ω1 to ω5 represented in Fig. 2) responsible for the conformational variations during each MD simulation were also analysed and are presented in Fig. 9.

Fig. 9 Distribution of dihedral torsions during the MD simulations of compounds MRB2 and MRB4. The same colouring scheme as in Fig. 2 was adopted for the dihedrals ω1 to ω5.

For MRB2, we can see that ω1 and ω5 dihedrals did not change greatly during the MD simulation, showing a maximum frequency value at around 100° and −100°, respectively. The ω2 and ω3 dihedrals show a wide range frequency distribution (see also ESI, Fig. S10†). The ω4 dihedral has two different frequency maxima, one at ca. 80° and another at 180°. For MRB4, the ω1 dihedral changes from ca. 60° to 140° and these values are maintained throughout the simulation; whilst ω2 and ω3 show frequency maximums at 0° and 180°, respectively. In opposition, the values of ω4 and ω5 dihedrals can adopt a wide range of values (see also ESI, Fig. S10†).

These individual distributions allow to see which are the most stable individual dihedral angles but don't tell us which are the most probable molecular conformations.

These depend on the joint probability for all the dihedrals. To analyse the overall conformational rearrangements of these two compounds, a clustering evaluation of all MD structures was performed.

Fig. 10 shows the energies of the ten representative structures (1 of each cluster) for the MRB2 and MRB4 molecules, respectively.

Fig. 10 Relative internal MM energies versus cluster frequency, considering compound MRB2 and MRB4.

These energies were calculated at the molecular mechanics level in a (continuum) aqueous environment. Table 4 shows the dihedral values obtained for the same optimized structures.

Table 4 Clustering analysis of the MD trajectory of compounds MRB2 and MRB4. The ten clusters' frequency is presented, together with ω1 to ω5 dihedral values of a representative and optimized structure of each cluster. The relative internal energy of these structures is also depicted, relative to the most stable conformation

Cluster (frequency)	ω1/°	ω2/°	ω3/°	ω4/°	ω5/°	ΔE/kcal mol^−1
MRB2
1 (0.31)	103	−51	−86	77	−108	0.0
2 (0.26)	98	62	58	158	−91	0.6
3 (0.16)	95	−122	62	161	−95	3.5
4 (0.11)	106	−88	156	165	−74	0.3
5 (0.09)	106	−88	156	165	−74	2.7
6 (<0.05)	81	103	157	164	−78	6.1
7(<0.05)	91	106	−75	161	−71	5.6
8 (<0.01)	75	76	−86	172	−82	2.9
9 (<0.01)	104	−116	96	83	−111	6.4
10 (<0.01)	81	−80	−104	70	−115	1.7
MRB4
1 (0.35)	99	0	−175	179	−89	0.0
2 (0.32)	101	−5	165	−112	−70	0.3
3 (0.19)	78	1	173	−175	88	0.6
4 (<0.05)	107	3	165	−120	93	2.1
5 (<0.05)	73	1	−172	107	60	1.6
6 (<0.05)	94	3	169	−84	−57	2.6
7(<0.01)	90	−6	180	136	−98	2.2
8 (<0.01)	88	1	178	−76	120	3.5
9 (<0.01)	84	−3	177	80	−121	3.8
10 (<0.01)	116	−10	179	71	53	4.4

---

The energy of the different conformers of chelator MRB2 spans a range of 6.4 kcal mol⁻¹, while ligand MRB4 spans a range of energies of 4.4 kcal mol⁻¹. We can see, that the representative structure of the most populated cluster displays the lowest energy, as expected, for both compounds. The energy difference between the two most populated clusters is also very small: −0.6 kcal mol⁻¹ for compound MRB2 and 0.3 kcal mol⁻¹ for compound MRB4.

Fig. 11 shows the optimized geometries for the most and less favoured structures for the two chelating compounds. The most favoured structure for compound MRB2 reveals that a stereo hindrance approaches the chelating and the benzene rings, allowing a perfectly aligned π–π stacking between these two rings. Considering the 2nd and 3rd most favoured structures, and the less favoured conformation for compound MRB2, any of them displays the stacking observed as in the most favoured conformation; however, the less favoured structure does not present a hydrogen bond between the hydroxyl group and the amide group, which is present in the other two conformations.

Fig. 11 Representation of the optimized geometries for the most and the less favoured structures for compounds MRB2 and MRB4. The energy values (ΔE) are calculated relatively to the most energetically stable structure.

Similarly, the most favoured conformation of compound MRB4 has the chelating ring directed to the naphthalene unit, making a π–π stacking between both ring planes. In opposition, the less favoured conformation assumes an extended geometry, avoiding these important intra-molecular contacts.

Conclusions

Two novel fluorescent 3-hydroxy-4-pyridinone compounds derived from two distinct fluorophores, rhodamine B and naphthalene, MRB2 and MRB4, were synthesized and characterized namely by NMR spectroscopy and single X-ray crystallography. The results confirm the compounds formulae and provide the structure of the molecules, in protected (1 and 3) and deprotected (MRB2 and MRB4) forms respectively.

The analysis of the data obtained from the crystalline structure of the compounds revealed that molecules 1, 3 and MRB4 have neutral charge while compound MRB2 is mono cationic.

In the single crystal structure of ligand MRB2, the xanthene moiety of the rhodamine is almost perpendicular to the adjacent phenyl ring and the pyridinone ring whereas in ligand MRB4 the xanthene group of the naphthalene framework is almost coplanar with the pyridinone ring.

The results also demonstrate that the deprotected ligand MRB2 remains in the enol form as described for other fluorescent pyridinone ligands⁵ while chelator MRB4 crystallizes in the keto form.

Information about the structural features of MRB2 and MRB4 chelators in aqueous solution was obtained from: (a) photophysical characterization of the ligands at pH = 7.4 which revealed the predominance of the keto form of the ligands; (b) MD simulations to determine the most stable conformations of ligands.

The most frequent and energetically favoured geometries adopted by both molecules bear the chelating group facing the benzene and naphthalene groups of ligand MRB2 and MRB4, respectively. This result reinforces the establishment of π–π interactions between those rings. For the naphthalene derived pyridinone MRB4, the most stable conformation predicted by the computational studies is in agreement with the form present in the crystal structure, confirming the π–π stacking between the rings. However, for rhodamine derived ligand MRB2, the relative position of the ring is different in the crystalline structure and in the most stable conformations determined in aqueous solution studies performed by computational methodologies.

These observations are somewhat different from the crystallographic studies, a fact which can be explained taking into account the crystal effects, in which the compounds interact with their replicas.

The overall results allowed getting insight on the most stable conformations expected to be present in an aqueous environment both from spectroscopic and computational studies. Moreover, the present data is useful to improve our knowledge on the coordination modes of these chelators with metal ions in solution.

Acknowledgements

This work received financial support from the European Union (FEDER funds through COMPETE) and National Funds (FCT, Fundação para a Ciência e Tecnologia) through project Pest-C/EQB/LA0006/2013. The Bruker Avance III 400 MHz NMR spectrometer was purchased under the framework of the National Programme for Scientific Re-equipment, contract REDE/1517/RMN/2005, with funds from POCI 2010 (FEDER) and (FCT). It is purchased under the framework of QREN, through project NORTE-07-0162-FEDER-000048 and is part of the National NMR network. To all financing sources the authors are greatly indebted. TM and JTSC also thank FCT their grants (SFRH/BD/79874/2011 and SFRH/BD/87434/2012), respectively; and NFB acknowledges her IF starting grant by FCT (IF/01355/2014).

References

1. T. Moniz, M. J. Amorim, R. Ferreira, A. Nunes, A. Silva, C. Queirós, A. Leite, P. Gameiro, B. Sarmento, F. Remião, Y. Yoshikawa, H. Sakurai and M. Rangel, J. Inorg. Biochem., 2011, 105, 1675–1682.
2. A. Leite, A. M. G. Silva, A. Nunes, M. Andrade, C. Sousa, L. Cunha-Silva, P. Gameiro, B. De Castro and M. Rangel, Tetrahedron, 2011, 67, 4009–4016.
3. M. Rangel, A. Leite, A. M. N. Silva, T. Moniz, A. Nunes, M. J. Amorim, C. Queiros, L. Cunha-Silva, P. Gameiro and J. Burgess, Dalton Trans., 2014, 43, 9722–9731.
4. R. B. R. Mesquita, R. Suárez, V. Cerdà, M. Rangel and A. O. S. S. Rangel, Talanta, 2013, 108, 38–45.
5. A. M. G. Silva, A. Leite, M. Andrade, P. Gameiro, P. Brandão, V. Felix, B. de Castro and M. Rangel, Tetrahedron, 2010, 66, 8544–8550.
6. T. Moniz, C. Queirós, R. Ferreira, A. Leite, P. Gameiro, A. M. G. Silva and M. Rangel, Dyes Pigm., 2013, 98, 201–211.
7. A. Nunes, M. Podinovskaia, A. Leite, P. Gameiro, T. Zhou, Y. Ma, X. Kong, U. E. Schaible, R. C. Hider and M. Rangel, JBIC, J. Biol. Inorg. Chem., 2010, 15, 861–877.
8. C. Queirós, A. Leite, M. G. M. Couto, T. Moniz, L. Cunha-Silva, P. Gameiro, A. M. G. Silva and M. Rangel, Dyes Pigm., 2014, 110, 193–202.
9. A. Leite, A. M. Silva, L. Cunha-Silva, B. de Castro, P. Gameiro and M. Rangel, Dalton Trans., 2013, 42, 6110–6118.
10. T. Moniz, A. Nunes, A. M. G. Silva, C. Queirós, G. Ivanova, M. S. Gomes and M. Rangel, J. Inorg. Biochem., 2013, 121, 156–166.
11. J. J. M. Marx, Best Pract. Res., Clin. Haematol., 2002, 15, 411–426.
12. S. S. Fernandes, A. Nunes, A. R. Gomes, B. de Castro, R. C. Hider, M. Rangel, R. Appelberg and M. S. Gomes, Microbes Infect., 2010, 12, 287–294.
13. C. Ratledge and L. G. Dover, Annu. Rev. Microbiol., 2000, 54, 881–941.
14. R. J. Ward, R. R. Crichton, D. L. Taylor, L. D. Corte, S. K. Srai and D. T. Dexter, J. Neural Transm., 2011, 118, 315–328.
15. G. C. Chan, S. Chan, P. L. Ho and S. Y. Ha, Hemoglobin, 2009, 33, 352–360.
16. M. X. Zhang, C. F. Zhu, Y. J. Zhou, X. Le Kong, R. C. Hider and T. Zhou, Chem. Biol. Drug Des., 2014, 84, 659–668.
17. D.-H. Qiu, Z.-L. Huang, T. Zhou, C. Shen and R. C. Hider, FEMS Microbiol. Lett., 2011, 314, 107–111.
18. B. Xu, X. L. Kong, T. Zhou, D. H. Qiu, Y. L. Chen, M. S. Liu, R. H. Yang and R. C. Hider, Bioorg. Med. Chem. Lett., 2011, 21, 6376–6380.
19. Y.-Y. Xie, M.-S. Liu, P.-P. Hu, X.-L. Kong, D.-H. Qiu, J.-L. Xu, R. Hider and T. Zhou, Med. Chem. Res., 2013, 22, 2351–2359.
20. B. D. Corbin, E. H. Seeley, A. Raab, J. Feldmann, M. R. Miller, V. J. Torres, K. L. Anderson, B. M. Dattilo, P. M. Dunman, R. Gerads, R. M. Caprioli, W. Nacken, W. J. Chazin and E. P. Skaar, Science, 2008, 319, 962–965.
21. T. Zhou, G. Winkelmann, Z.-Y. Dai and R. C. Hider, J. Pharm. Pharmacol., 2011, 63, 893–903.
22. S. Silva-Gomes, S. Vale-Costa, R. Appelberg and M. S. Gomes, Front. Cell. Infect. Microbiol., 2013, 3, 96.
23. C. Ratledge, Tuberculosis, 2004, 84, 110–130.
24. N. Lounis, C. Truffot-Pernot, J. Grosset, V. R. Gordeuk and J. R. Boelaert, J. Clin. Virol., 2001, 20, 123–126.
25. D. G. Russell, Nat. Rev. Mol. Cell Biol., 2001, 2, 569–586.
26. M. S. Gomes, G. Dom, J. Pedrosa, J. R. Boelaert and R. Appelberg, Tuberculosis, 1999, 79, 321–328.
27. T. Moniz, D. Silva, T. Silva, M. S. Gomes and M. Rangel, MedChemComm, 2015, 6, 2194–2203.
28. J. T. S. Coimbra, T. Moniz, N. F. Brás, G. Ivanova, P. A. Fernandes, M. J. Ramos and M. Rangel, J. Phys. Chem. B, 2014, 118, 14590–14601.
29. S. Fakih, M. Podinovskaia, X. Kong, H. L. Collins, U. E. Schaible and R. C. Hider, J. Med. Chem., 2008, 51, 4539–4552.
30. T. Kottke and D. Stalke, J. Appl. Crystallogr., 1993, 26, 615–619.
31. APEX2, Data Collection Software Version 2.1-RC13, Bruker AXS, Delft, The Netherlands, 2006.
32. Cryopad, Remote monitoring and control, Version 1.451, Oxford Cryosystems, Oxford, United Kingdom, 2006.
33. SAINT+, Data Integration Engine v. 7.23a ©, 1997-2005, Bruker AXS, Madison, Wisconsin, USA.
34. G. M. Sheldrick, SADABS v.2.01, Bruker/Siemens Area Detector Absorption Correction Program, 1998, Bruker AXS, Madison, Wisconsin, USA.
35. G. M. Sheldrick, SHELXS-97, Program for Crystal Structure Solution, University of Göttingen, 1997.
36. G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008, 64, 112–122.
37. G. M. Sheldrick, SHELXL-97, Program for Crystal Structure Refinement, University of Göttingen, 1997.
38. A. L. Spek, Acta Crystallogr., Sect. A: Found. Crystallogr., 1990, 46, C34.
39. A. L. Spek, J. Appl. Crystallogr., 2003, 36, 7–13.
40. P. van der Sluis and A. L. Spek, Acta Crystallogr., Sect. A: Found. Crystallogr., 1990, 46, 194–201.
41. C. O. P. Gaussian Inc., Bldg. 6, Pittsburgh, PA 15106, USA.
42. 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, N. J. 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, Gaussian, Inc., Wallingford CT, Revision A.1 edn, 2009.
43. C. I. Bayly, P. Cieplak, W. D. Cornell and P. A. Kollman, J. Phys. Chem., 1993, 97, 10269–10280.
44. T. A. D. D. A. Case, T. E. Cheatham III, C. L. Simmerling, J. Wang, R. E. Duke, R., R. C. W. Luo, W. Zhang, K. M. Merz, B. Roberts, S. Hayik, A. Roitberg, G. Seabra, A. W. G. J. Swails, I. Kolossváry, K. F. Wong, F. Paesani, J. Vanicek, R. M. Wolf, J. Liu, S. R. B. X. Wu, T. Steinbrecher, H. Gohlke, Q. Cai, X. Ye, J. Wang, M.-J. Hsieh, G., D. R. R. Cui, D. H. Mathews, M. G. Seetin, R. Salomon-Ferrer, C. Sagui, V. Babin, T., S. G. Luchko, A. Kovalenko and P. A. Kollman, AMBER 12, University of California, San Francisco, 2012.
45. J. M. Wang, R. M. Wolf, J. W. Caldwell, P. A. Kollman and D. A. Case, J. Comput. Chem., 2004, 25, 1157–1174.
46. J. A. Izaguirre, D. P. Catarello, J. M. Wozniak and R. D. Skeel, J. Chem. Phys., 2001, 114, 2090–2098.
47. J. P. Ryckaert, G. Ciccotti and H. J. C. Berendsen, J. Comput. Phys., 1977, 23, 327–341.
48. U. Essmann, L. Perera, M. L. Berkowitz, T. Darden, H. Lee and L. G. Pedersen, J. Chem. Phys., 1995, 103, 8577–8593.
49. J. Y. Shao, S. W. Tanner, N. Thompson and T. E. Cheatham, J. Chem. Theory Comput., 2007, 3, 2312–2334.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 1434795–1434798. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c5ra23217a

---