Letter Cite This: Org. Lett. 2018, 20, 2420−2423
pubs.acs.org/OrgLett
Diastereoselective Synthesis of a Dyn[3]arene with Distinct Binding Behaviors toward Linear Biogenic Polyamines Marion Donnier-Maréchal,† Jean Septavaux,† Emeric Jeamet,† Alexandre Héloin,† Florent Perret,† Elise Dumont,‡ Jean-Christophe Rossi,§ Fabio Ziarelli,∥ Julien Leclaire,*,† and Laurent Vial*,† †
Institut de Chimie et Biochimie Moléculaires et Supramoléculaires, UMR 5246 CNRS - Université Claude Bernard Lyon1 - CPE Lyon, 43 Boulevard du 11 Novembre 1918, Villeurbanne Cedex 69622, France ‡ Laboratoire de Chimie, UMR 5182 CNRS - Ecole Nationale Supérieure de Lyon, Université Claude Bernard Lyon 1 - CEA, 46 Allée d’Italie, Lyon Cedex 07 69364, France § Institut des Biomolécules Max Mousseron, UMR 5247 CNRS, Université de Montpellier - ENSCM, Place Eugène Bataillon, Montpellier Cedex 5 34296, France ∥ Spectropole d’Aix-Marseille Université - Centrale Marseille - CNRS, Fédération des Sciences Chimiques FR1739, Campus Scientifique de Saint Jérôme, Marseille Cedex 20 13397, France S Supporting Information *
ABSTRACT: The extension of the family of dyn[n]arenes toward a three-membered macrocycle is reported. Through a templated approach, a single diastereoisomer of a dyn[3]arene that bears six carboxyl groups could be isolated by precipitation in 59− 63% yield and excellent purity (≥95%). A combination of experimental and computational experiments in water at physiological pH revealed that the macrocycle could bind parent biogenic polyamines with a unique diversity of surface-binding modes. Whereas no binding event could be accurately measured with 1,3diaminopropane, spermidine formed a classical stoichiometric complex with the dyn[3]arene in the millimolar concentration range. On the other hand, the data obtained for spermine could only be attributed to a more complex binding event with the formation of a 2:1 complex at high [host]/[guest] ratios and redistribution toward a 1:1 complex upon further addition of guest.
A
Scheme 1. Previously Reported Synthesis of Dyn[4]arene 14 (refs 7 and 8) and Proposed Approach To Obtain Larger Dyn[n]arenes 1n
lthough disulfide-linked cyclic oligomers were reported for the first time in the literature more than 40 years ago,1 they only found applications with the emergence of dynamic combinatorial chemistry.2 The introduction of template molecules in dynamic combinatorial libraries (DCLs) of functionalized aromatic bisthiols led, for instance, to the discovery of catalysts,3 post-translational modification binders,4 or self-replicators.5 Among these new macrocycles, dyn[4]arene 14 was previously isolated by preparative HPLC from a DCL made of 2,5dimercaptoterephthalic acid 1 and spermine in water at physiological pH (Scheme 1)6 and was shown to reverse some of the biological effects of the polyamine on DNA.7,8 Recently, we reported that 14 could actually be isolated on a gram scale and without any chromatographic purification by a simple precipitation in an acidic medium/solid−liquid washing procedure,9 thus allowing us to take advantage of its conformational lability for the chirality sensing and discrimination of lysine derivatives by circular dichroism.10 With this purification procedure in hand, we decided to explore DCLs made from 1 and the bulkier polyamine templates 3−7 (Scheme 1) with the aim of obtaining larger dyn[n]arenes for future bioapplications.11 Dynamic combinatorial screening was performed as follows: a solution of building block 1 (4 mM) and a chosen template 2−7 (1 mM) in 200 mM Tris at pH 7.4 was allowed to polymerize through the formation of disulfide bridges and equilibrate for 2 © 2018 American Chemical Society
Received: March 7, 2018 Published: April 6, 2018 2420
DOI: 10.1021/acs.orglett.8b00766 Org. Lett. 2018, 20, 2420−2423
Letter
Organic Letters days in the presence of air. Whereas libraries involving templates 2−6 led to homogeneous mixtures with various ratios of tri-, tetra-, and pentameric cyclic oligomers as determined by HPLC− MS analyses, a white precipitate was observed with 2,4,6-triethyl1,3,5-benzenetrimethanamine 7. The solid, which was isolated by filtration and washed with water, was insoluble in all of the organic or aqueous solvents tested and was therefore analyzed by solidstate 13C and 15N NMR (Figure 1, and Supporting Information
Unfortunately, we were unable to obtain suitable crystals of the macrocycle. Therefore, Molecular Dynamics (MD) simulations, starting from a structure that respects the previous characterizations (i.e., configuration and protonation state), were considered in order to gain insight into the three-dimensional structure of dyn[3]arene 13. MD simulations of 100 ns at 300 K with constant pressure were performed with the Amber 12 software package.13 Ammonium cations NH4+ were added in order to neutralize the system, which was immersed in a truncated octahedral TIP3P water box (see the SI for full computational details). The most representative structure was extracted (Figure 3A,B). The size of the cavity was estimated by subtracting the
Figure 1. Solid-state 13C NMR of the precipitate generated during a library made from 1 and 7 and energy-minimized structure (PM3 level, Spartan 14) of the barrel-like ternary complex.
Figure 3. Most representative 3D structure of dyn[3]arene 13 ((pS)3 configuration): side (A) and top (B) views obtained by MD simulation over 100 ns at 300 K with constant pressure. The displayed structure was present during 36% of the time over the trajectory. Water and ammonium cations have been removed for clarity. (C) Space-filling model and simplified representation of dyn[3]arene 13 used for the estimation of cavity’s size. The radius of the circle r is equal to √3/6 × l.
(SI)). Solid-state 13C polarization inversion (CPPI/MAS) and C−T1 NMR spectra allowed us to assign the signals of nuclei from both the polyamine 7 and building block 1 and to quantify their relative amount ([1]/[7] = 1.5) in the solid. As a unique singlet and a single set of signals were observed on, respectively, the solid-state 15N NMR and 13C NMR spectra, it could be concluded that the assembly obtained was C3-symmetric. These analyses strongly suggested the formation and precipitation of a barrel-type complex between a self-assembled cyclic trimer 13 from three building blocks 1, acting as staves, and two molecules of template 7, acting as heads. As for the synthesis of dyn[4]arene 14,8 we performed an acidic treatment of the solid with aqueous trifluoroacetic acid (pH = 1) in order to protonate the carboxyl functional groups, disassemble the complex, and eventually precipitate dyn[3]arene 13 as a yellow solid (59−63% total yield after optimization of the [1]/[7] ratio at 1.5 in the DCLs). The isolated dyn[3]arene 13 was fully characterized by NMR and mass spectroscopy (see the SI). The 1H NMR spectrum displayed a unique singlet at δ = 8.01 ppm (purity: ≥ 95%), demonstrating the C3-symmetric homochiral (pS) 3/(pR)3 configuration of the macrocycle.12 In addition, pH-metric titrations were performed on the macrocycle by using an automatic titrator (for details, see SI), revealing that 13 could be considered fully deprotonated (i.e., protonation degree Θ < 5%) for pH > 7.4 (Figure 2). 13
radius of a carbon atom (i.e., 1.7 Å) to the radius of the circle conscribed in the equilateral triangle formed by the walls of the three benzene units (i.e., 2.7 Å), giving a maximum size for the cavity of 2.0 Å that is smaller than the van der Waals’ diameter of a single atom of hydrogen (i.e., 2.4 Å, Figure 3C). While dyn[4]arene 14 was able to form pseudorotaxane-type complexes with linear polyamines,8 it was envisioned that dyn[3]arene 13 would not accommodate molecules within its cavity but could bind cationic guests via its anionic crowns.14 Surprisingly, ITC binding experiments with parent biogenic 1,3diaminopropane, spermidine, and spermine in water at pH 7.4 revealed distinct binding behaviors (Figure 4). Whereas no binding event could be accurately measured for 1,3-diaminopropane under these conditions,15 the thermodynamic parameters associated with the formation of a stoichiometric complex with spermidine (Kd = 115 μM, ΔH = −0.81 kcal· mol−1, ΔS = 15.3 cal·mol−1·K−1) were obtained from the 1:1 binding model of the MicroCal ITC Origin software. MD simulations of the complexes using the previous setup (vide supra) revealed thatunlike 1,3-diaminopropane (Figure 5A) spermidine could behave as a multivalent ligand (Figure 5B), resulting in the superior affinity of spermidine for the macrocycle in comparison with the simple 1,3-propanediamine. On the other hand, the data obtained for spermine could only be attributed to a more complex binding event (Figure 4). MD simulations suggested the possible formation of a 2:1 complex at low [spermine]/[13] ratio (Figure 5C) and a redistribution toward a 1:1 complex upon further addition of spermine (Figure 5D). Such a scenario followed the formal eq 1:
Figure 2. Protonation degree Θ versus pH curve for dyn[3]arene 13 in deionized water. The blue region highlights Θ < 5%. 2421
DOI: 10.1021/acs.orglett.8b00766 Org. Lett. 2018, 20, 2420−2423
Letter
Organic Letters
while the law of conservation of mass is displayed in eqs 3 and 4: [spermine]t = ([spermine] + [13 −spermine]) + [13 −spermine−13 ]
(3)
[13 ]t = ([13 ] + [13 −spermine]) + (2 × [13 −spermine−13 ]) (4)
To extract the different thermodynamic parameters associated with these putative binding events, we performed two series of titrations between spermine and dyn[3]arene 13. The limited solubility of 13 in water led us to operate in submillimolar concentrations. Under such conditions and for a limited [dyn[3]arene 13]/[spermine] ratio ( 0.99).
thermodynamic parameters associated with the formation of the [13−spermine] complex (K1 = 74000 M−1, ΔH1 = −2.12 kcal· mol−1, ΔS1 = 15.0 cal·mol−1·K−1). These parameters could then be injected into eqs 2−4 with no approximation in order to fit the experimental data obtained from the reverse titration of a solution of 13 with spermine (Figure 6B). Under these conditions, in which the formation of the ternary complex was comparatively favored, the second set of thermodynamic constants associated with the formation of the complex [13−spermine−13] (K2 = 8100 M−1, ΔH2 = −0.75 kcal·mol−1, ΔS2 = 15.4 cal·mol−1·K−1) could be obtained. Interestingly, both [13−spermine] and [13−spermine− 13] displayed similar entropy of formation, suggesting that the formation of the ternary complex did not result, in this case, in disadvantageous conformational restrictions, in comparison with the stoichiometric one. For both titrations, excellent correlation coefficients R2 (i.e., 0.998 and 0.997 for the fit of the 1:1 and 2:1 model function, respectively) were obtained, strongly supporting our binding scenario between 13 and spermine. Finally, the hypothesis of neglecting the presence of the ternary complex with respect to other species during the titration of spermine with 13 could be validated a posteriori through the speciation generated from the full set of thermodynamic data (Figure 7A) where the ternary complex was consistently dominated by the stoichio-
Figure 5. 3D structures of the various complexes between dyn[3]arene 13 ((pS)3 configuration) and 1,3-diaminopropane (A), spermidine (B), spermine at lower [spermine]/[13] ratio (C), and spermine at higher [spermine]/[13] ratio (D) obtained by MD simulations over 100 ns at 300 K with constant pressure. For each structure, the numer in parentheses represents its percentage (when it is superior to 1%) of presence during the trajectory. Water and ammonium cations have been removed for clarity.
The total dissipated heat Qt within the measurement cell, where V0 was the initial volume of the analyte’s solution, can be expressed as Q t = (V0 × ΔH1 × K1 × [13 ] × [spermine]) + (V0 × ΔH2 × K1 × K 2] × [13 ]2 × [spermine])
(2) 2422
DOI: 10.1021/acs.orglett.8b00766 Org. Lett. 2018, 20, 2420−2423
Letter
Organic Letters
grateful to Prof. Gérald Monard (Université de Lorraine) and the mésocentre EXPLOR for the allocation of GPU resources.
■
(1) Wong, D. T.; Marvel, C. S. J. Polym. Sci., Polym. Chem. Ed. 1976, 14, 1637−1644. (2) (a) Black, S. P.; Sanders, J. K. M.; Stefankiewicz, A. R. Chem. Soc. Rev. 2014, 43, 1861−1872. (b) Corbett, P. T.; Leclaire, J.; Vial, L.; West, K. R.; Wietor, J. L.; Sanders, J. K. M.; Otto, S. Chem. Rev. 2006, 106, 3652− 3711. (3) (a) Vial, L.; Sanders, J. K. M.; Otto, S. New J. Chem. 2005, 29, 1001− 1003. (b) Brisig, B.; Sanders, J. K. M.; Otto, S. Angew. Chem., Int. Ed. 2003, 42, 1270−1273. (4) (a) Pinkin, N. K.; Liu, I.; Abron, J. D.; Waters, M. L. Chem. - Eur. J. 2015, 21, 17981−17986. (b) Pinkin, N. K.; Waters, M. L. Org. Biomol. Chem. 2014, 12, 7059−7067. (c) James, L. I.; Beaver, J. E.; Rice, N. W.; Waters, M. L. J. Am. Chem. Soc. 2013, 135, 6450−6455. (5) Nowak, P.; Colomb-Delsuc, M.; Otto, S.; Li, J. J. Am. Chem. Soc. 2015, 137, 10965−10969. (6) In connection with pillar[n]arenes, the word dyn[n]arene refers to benzene units linked in the para positions by dynamic disulfide bridges. (7) Vial, L.; Ludlow, R. F.; Leclaire, J.; Pérez-Fernández, R.; Otto, S. J. Am. Chem. Soc. 2006, 128, 10253−10257. (8) Other macrocycles were also reported as host molecules for polyamines in water, including cucurbit[n]urils and calix[n]arenes. For two recent examples, see: (a) D'Urso, A.; Brancatelli, G.; Hickey, N.; Farnetti, E.; De Zorzi, R.; Bonaccorso, C.; Purrello, R.; Geremia, S. Supramol. Chem. 2016, 28, 499−505. (b) Parente Carvalho, C.; Norouzy, A.; Ribeiro, V.; Nau, W. M.; Pischel, U. Org. Biomol. Chem. 2015, 13, 2866−2869. (9) Skowron, P. T.; Dumartin, M.; Jeamet, E.; Perret, F.; Gourlaouen, C.; Baudouin, A.; Fenet, B.; Naubron, J. V.; Fotiadu, F.; Vial, L.; Leclaire, J. J. Org. Chem. 2016, 81, 654−661. (10) Vial, L.; Dumartin, M.; Donnier-Maréchal, M.; Perret, F.; Francoia, J.-P.; Leclaire, J. Chem. Commun. 2016, 52, 14219−14221. (11) (a) Perret, F.; Lazar, A. N.; Coleman, A. W. Chem. Commun. 2006, 23, 2425. (b) Perret, F.; Coleman, A. W. Chem. Commun. 2011, 47, 7303−7319. (12) The heterochiral (pS)2(pR)/(pR)2(pS) configuration of 13 would lead to multiple signals in both 1H and 13C NMR spectra, which were not observed. (13) Case, D. A.; Darden, T. A.; Cheatham, T. E., III; Simmerling, C. L.; Wang, J.; Duke, R. E.; Luo, R.; Walker, R. C.; Zhang, W.; Merz, K. M.; Roberts, B.; Hayik, S.; Roitberg, A.; Seabra, G.; Swails, J.; Götz, A. W.; Kolossváry, I.; Wong, K. F.; Paesani, F.; Vanicek, J.; Wolf, R. M.; Liu, J.; Wu, X.; Brozell, S. R.; Steinbrecher, T.; Gohlke, H.; Cai, Q.; Ye, X.; Wang, J.; Hsieh, M.-J.; Cui, G.; Roe, D. R.; Mathews, D. H.; Seetin, M. G.; Salomon-Ferrer, R.; Sagui, C.; Babin, V.; Luchko, T.; Gusarov, S.; Kovalenko, A.; Kollman, P. A. AMBER 12; University of California, San Francisco, 2012. (14) A moderate shielding (i.e., Δδ ≤ 1 ppm) of the signals corresponding to the guest molecules were observed by 1H NMR upon host addition, confirming peripheral complexation events since the formation inclusion complexes with parent dyn[4]arene 14 led to shielding up to 4 ppm (see refs 7 and 9). However, binding events were also accompanied by substantial line shape broadening due to exchange processes that prevented the accurate monitoring of chemical shift variations during titrations. (15) Titration experiments of 0.95 mM 13 with polyamines were carried at 298 K on a Malvern ITC200 microcalorimeter. Under these conditions, the quantification of a binding event is limited to affinities that are superior to 1/[13] ∼ 103 M−1 (Kd > 1 mM). See: https://cmi. hms.harvard.edu/files/cmi/files/microcal-itc200-system-user-manual. pdf. (16) For two recent examples of polyamine sensing using macrocycles, see: (a) Jiang, G.; Zhu, W.; Chen, Q.; Li, X.; Zhang, G.; Li, Y.; Fan, X.; Wang, J. Sens. Actuators, B 2018, 261, 602−607. (b) Reddy Kothur, R.; Anil Patel, B.; Cragg, P. Chem. Commun. 2017, 53, 9078−9080.
Figure 7. Quantitative distribution of the various species in solution during the titration of 0.375 mM of spermine with 13 (A) and the titration of 0.475 mM of 13 with spermine (B).
metric one throughout the titration and the second term of eq 4 represented an average of 10% of the total dissipated heat Qt. In contrast, the appearance of both [13−spermine] and [13− spermine−13] was concomitant during the titration of 13 with spermine (Figure 7B), resulting in the unusual ITC thermogram obtained in Figure 4. In conclusion, we reported the extension of the family of dynarenes with the templated synthesis of a new polyanionic dyn[3]arene 13. The procedure was diastereoselective and did not require any chromatographic purification. Although it was not able to accommodate biogenic polyamines in its very small cavity, and certainly gave us the incentive to further explore the crucial role of the cavity of parent dyn[4]arene 14 in its tremendous affinities previously measured with these guest molecules,8 the dyn[3]arene displayed a unique diversity of surface binding modes. This diversity resulted in a clear selectivity among the guest molecules, and the macrocycle may therefore find future application as biosensor.16 We are currently working in this direction.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.orglett.8b00766. Synthetic, analytical and computational details, NMR and mass spectra (PDF)
■
REFERENCES
AUTHOR INFORMATION
Corresponding Authors
*E-mail:
[email protected]. *E-mail:
[email protected]. ORCID
Jean-Christophe Rossi: 0000-0002-1451-7647 Julien Leclaire: 0000-0001-7984-9055 Laurent Vial: 0000-0002-0357-0573 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by the LABEX iMUST (ANR-10LABX-0064) of Université de Lyon within the program “Investissements d’Avenir” (ANR-11-IDEX-0007) operated by the French National Research Agency (ANR). We are very 2423
DOI: 10.1021/acs.orglett.8b00766 Org. Lett. 2018, 20, 2420−2423
