Classification: Physical Sciences (Chemistry) Title: Relative hydrophilicities of cis and trans formamides Authors: Yong-Sheng Li1#, Luis Escobar2,3#, Yu-Jie Zhu1#, Yoram Cohen4, Pablo Ballester2,5*, Julius Rebek, Jr.1,6* and Yang Yu1* Author Affiliations: 1 Center for Supramolecular Chemistry & Catalysis and Department of Chemistry, College of Science, Shanghai University, 99 Shang-Da Road, Shanghai 200444, China. 2 Institute of Chemical Research of Catalonia (ICIQ), Av. Països Catalans 16, 43007, Tarragona, Spain. 3 Universitat Rovira i Virgili, Departament de Química Analítica i Química Orgánica, c/Marcel·lí Domingo, 1, 43007, Tarragona, Spain. 4 School of Chemistry, The Sackler Faculty of Exact Sciences, Tel Aviv University, Ramat Aviv, Tel Aviv 69978, Israel. 5 Catalan Institution for Research and Advanced Studies (ICREA), Pg. Lluís Companys 23, 08010, Barcelona, Spain. 6 Skaggs Institute for Chemical Biology and Department of Chemistry, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037, USA. *Corresponding authors: Pablo Ballester: pballester@ICIQ.ES Julius Rebek, Jr.: jrebek@scripps.edu Yang Yu: yangyu2017@shu.edu.cn Keywords: Water-soluble cavitand, Formamide, cis and trans isomers, Hydrophilicity Abstract Secondary formamides are widely encountered in biology and exist as mixtures of both cis and trans isomers. Here we assess hydrophilicity differences between isomeric formamides through direct competition experiments. Formamides bearing long aliphatic chains were sequestered in a water-soluble molecular container having a hydrophobic cavity with an end open to the aqueous medium. NMR spectroscopic experiments reveal a modest preference (<1 kcal/mol) for aqueous 1 solvation of the trans formamide terminals over the cis isomers. With diformamides, the supramolecular approach allows staging of intramolecular competition between short-lived species with subtle differences in hydrophobic properties. Significance statement Many studies have investigated secondary formamides as mixtures of cis and trans isomers. They are widespread in nature, but relatively low energetic barriers to interconversion prevents the isolation and characterization of the pure isomers. Here we determine hydrophilic differences between the isomers by binding in a water-soluble, synthetic molecular container. The container offers a range of environments from hydrophilic to hydrophobic and distinguishes subtle differences in the polarities of formamide isomers. Using NMR methods we find that cis formamides are more hydrophobic than the corresponding trans isomers. The conclusion holds for diformamides, where we staged an intramolecular competition between cis and trans formamides within the container. The advantage of this supramolecular method is that stable and isolable compounds are not required to make the determination. Introduction Decades of structural work with peptides have established that secondary amide bonds generally exist in trans conformations. Experimental1 and computational2 studies on model compounds such as N-methyl acetamide revealed that dipole/dipole and steric effects around the secondary amide bond destabilize the cis isomer (Fig. 1)3. The energetic differences are large enough that the cis isomer is usually present in only ~ 1%. Secondary formamides are different. They are widespread in nature as initiators of protein synthesis in bacteria and mitochondria and as triggers of immune responses in eukaryotic cells4. Typically, secondary formamides are 10-20% cis isomers, and the percentage increases with the bulk of the N-substituent (Fig. 1)5. The trans/cis equilibria can be determined by NMR methods6-8, but the relatively low (15-20 kcal/mol) energetic barriers to interconversion – slow on the NMR chemical shift timescale but fast on the human timescale – prevent the isolation and characterization of the pure isomers. Dipole moments are calculated to be slightly higher for the cis formamides vs. trans (4.2 D vs. 4.0 D)1,2, yet the amount of the cis isomer often increases slightly on transfer from water to organic solvents. Neither experimental9,10 nor theoretical hydration studies offer clear answers2. As a result, the differences in polarity, hydrophilicity and their effects on chemical or biological behavior 11,12 remain elusive. Here we apply synthetic molecular containers (cavitands) to determine their relative hydrophilicities. Results and discussion The cavitand used is 1 (Fig. 2), a water-soluble version of a container introduced by de Mendoza13 and Choi14 for use in organic solvents. The cavitand acts as host with a hydrophobic interior and an open end exposed to water (D2O)15, and guest molecules of suitable size, shape and chemical surface are bound within. The cavity offers a range of chemical environments – from hydrophilic to hydrophobic – as well as magnetic environments: The 8 aromatic panels of the cavitand shield the guest nuclei within from the applied field of the NMR spectrometer. The effects on guest nuclei – calculated by the method of Schleyer16 and mapped experimentally – are summarized in the cartoon of Fig. 2. Nuclei at the hydrophobic bottom experience the largest upfield shifts ( = 4.0 to - 4.3 ppm)17. The magnetic effects gradually diminish as the positions rise to near the top of the rim ( = 0 to -0.5 ppm) and nuclei outside the cavitand, exposed to the aqueous solvent are unaffected. 2 These effects emerge from comparison of the NMR spectra of free and bound N-octyl-formamides (Fig. 3). In the free spectra (Fig. 3A, CDCl3 or D2O solvent) the cis vs trans isomerism affects only signals near the polar end of the molecule, as indicated for the -CO-H and the alpha-CH2NH. Integration gives the relative concentrations of cis vs trans isomers of the free formamides as shown. The signals for CH2 groups near the middle or the methyl end of the free molecule remain unaffected. Brief sonication of N-octyl-formamide (2a) with excess cavitand 1 in D2O produces 1:1 complexes. Two species are present (trans and cis amides) whose spectra is shown in Fig. 3A (see also SI Appendix Fig. 1-1), and the chemical shifts are characteristic of extended conformations of the alkyl chains in the cavitand15. The assignments were obtained by 2D experiments (SI Appendix Fig. 1-2) and place the CH3 (on average) closest to the bottom of the cavitand with  = - 4.2 ppm for the trans isomer of 2a. The other (polar) end of the guest is exposed to D2O; the small shift ( = - 0.7 ppm) of the -CH2-CH2-NH-CHO signal places it near the top of the cavitand with the NH-CHO group outside. The cartoon of Fig. 2b reflects this arrangement. The smaller set of signals for the cis isomer are shifted throughout the spectrum: Nuclei near the amide are shifted upfield indicating that the polar end – on average – is deeper in the hydrophobic cavitand, and nuclei near the methyl group are shifted downfield, indicating that the apolar end – on average – is shallower in the cavitand than the trans formamide. Parallel behavior is seen in the spectra of the longer N-decyl-formamide (2b) (SI Appendix Fig. 2-1 and 22). The trans formamide guest 2a is positioned with the methyl group fixed at the bottom (Fig. 3 and SI Appendix Fig. 11 and SI Appendix Table 7): its signal shows  = - 4.1 ppm which is 98% of the maximum (- 4.2 ppm) seen in this cavitand (Fig. 3C). The signal for the methyl group of the cis amide shows  = - 3.36 ppm or only 80% of the maximum value. Accordingly, this guest spends some 20% of the time in the “upside down” position with the formyl end deep in the cavitand (Fig. 3C). Rapid motion of the guest on the NMR chemical shift timescale between the two arrangements shown is consistent with the signals observed. The motion is facilitated by a coiled shape of the alkyl chain as seen in related cavitands18, but the conformational details of the central atoms are unknown. In any case, the cis formyl group appears less hydrophilic than the trans. We also staged intramolecular competitions for the hydrophobic cavitand with formamides of −diamines as guests. The NMR spectra of diformamides in solution are as expected (SI Appendix). But as guests in 1, the diformamides of (CH2)8 to (CH2)11 (3a-3d) show starkly different signal patterns (Fig. 4A and SI Appendix Fig. 3). The major trans,trans isomers show simplified spectra indicating a time-averaged symmetrical arrangement of guest in the host. The signal for the beta methylene group of the trans,trans amide shows  = - 2.1 ppm which is 50% of the maximum seen in this cavitand (- 4.2 ppm)20 (Fig. 4B and SI Appendix Fig. 4). The minor trans,cis isomers show a superimposed set of signals with the more complex features expected of an unsymmetrical arrangement, in which one end spends more time inside the cavitand than the other. For example, the beta methylene of cis amide end signal shows  = -3.38 ppm or only 80% of the maximum value. The biased arrangement of the unsymmetrical trans,cis guest arises from the differences in relative hydrophilicity and hydrophobicity of the two ends. The detailed assignments of signals were established as follows for the (CH2)9 (1:3b) complex (Fig. 5; for this and other complexes see SI Appendix Figs. 3-10). The spectra were obtained under 3 conditions of excess diformamides in order to identify exchange processes between free and bound guests. Downfield regions of the 2D-EXSY spectrum are shown in Fig. 5 panel a and upfield regions in panel b. The NMR of the trans,cis isomer shows resolved upfield-shifted signals for each methylene group of the alkyl chain. The protons of the terminal formamide groups of the bound trans,cis isomer show chemical exchange cross-peaks with those of the corresponding free diamides (Fig. 5, panel a). The methylene alpha to the cis-amide (H9) appears at -0.23 ppm and shows a selective cross peak to the formyl proton of the cis formamide in the complex at 5.81 ppm (SI Appendix Fig. 6). The cis-formyl proton in the bound cis,trans isomer of (CH2)9 (3b) is significantly upfield-shifted compared to its trans counterpart: it is deeper in the cavitand. Accordingly, the trans formyl prefers more exposure to the aqueous medium and is more hydrophilic than the cis formyl. The guests are not static but move rapidly in the cavitand. The details are unknown but can involve a “yo-yo” like motion21 of folded arrangements in the cavitand22 or the simple tumbling of a compacted conformation. Additional evidence of hydrophilicity differences may be inferred from the integral values of the formyl protons of bound (CH2)9 in the two complexes. The ratio between the trans,trans and trans,cis isomers is 60:40, and indicates a binding preference of the cavitand for the trans,cis isomer. If the two complexes were isoenergetic, the expected ratio should be 82:18 (SI Appendix). The hydrophilicity of stable structures can be directly compared by partitioning between water and a hydrophobic solvent such as octanol. The labile structures here are temporarily partitioned between an aqueous phase and the hydrophobic interior of the cavitand. The competition of the cis,trans isomers by this method is direct since both are present in the same solution and, in the case of the diformamides, in the same molecule. The shape and volume of the cavity can impose selectivity of its own23, but the cis and trans formamides isomers are not expected to differ in size and shape enough to bias the competition. The possibility of hydrogen-bonded, dimeric structures exists in media of low polarity, but the competitive aqueous solvent (outside) and the limited size of the cavitand (inside) prevents self-association of these guests. The present results consistently indicate that the cis formamides are more hydrophobic than their trans isomers. Conclusions The advantage of the present application is that it does not require stable, isolable compounds. Their lifetimes need only to be long enough on the NMR chemical shift timescale – seconds or less – to make the determination. The choices between hydrophilic and hydrophobic offered by the cavitand resemble the workings of the Wilcox torsion balance24 as they can operate on the subkilocalorie scale. Materials and Methods The experimental procedures for the synthesis of cavitand 1, N-octyl-formamide 2a, N-decylformamide 2b, diformamide 3 and NMR spectra of the complexes are available in SI Appendix. Acknowledgments This work was supported by the National Natural Science Foundation of China (Grant No. 21801164), the US National Science Foundation (CHE 1801153) and by Shanghai University 4 (N.13-G210-19-230), Shanghai, China. Dr. Yang Yu thanks the Program for Professor of Special Appointment (Dongfang Scholarship) of the Shanghai Education Committee. Author Contributions Y.S.L and Y.J.Z synthesized the formamides, carried out experiments and analyzed the experimental data. L.E performed 2D-NMR experiment of diformamides, DFT computations and data analysis. Y.S.L, Y.J.Z and L.E contributed equally to this work. Y.C reviewed the manuscript and provided advice. P.B analyzed the 2D-NMR data and revised the manuscript. J.R and Y.Y conceived the project, guided experiments and wrote the manuscript. Conflicts of interest The authors declare no conflict of interest. References 1. A. Radzicka, L. Pedersen, R. Wolfenden, Influences of Solvent Water on Protein Folding: Free Energies of Solvation of Cis and Trans Peptides Are Nearly Identical. Biochemistry 27, 45384541 (1988). 2. W. L. Jorgensen, J. Gao, Cis-Trans Energy Difference for the Peptide Bond in the Gas Phase and in Aqueous Solution. J. Am. Chem. Soc. 110, 4212-4216 (1988). 3. E. V. Anslyn, D. A. Dougherty, Modern Physical Organic Chemistry (University Science Books, 2006), p. 23. 4. E. R. Prossnitz, R. D. Ye, The N-Formyl Peptide Receptor: A Model for the Study of Chemoattractant Receptor Structure and Function. Pharmacol. Ther. 74, 73-102 (1997). 5. L. A. LaPlanche, M. T. Rogers, cis and trans Configurations of the Peptide Bond in NMonosubstituted Amides by Nuclear Magnetic Resonance. J. Am. Chem. Soc. 86, 337-341 (1964). 6. A. G. Martínez, E. T. Vilar, A. G. Fraile, P. Martínez-Ruiz, A Computational and Experimental Study on the Relative Stabilities of Cis and Trans Isomers of N-Alkylamides in Gas Phase and in Solution. J. Phys. Chem. A. 106, 4942-4950 (2002). 7. M. Avalos, R. Babiano, J. L. Barneto, P. Cintas, F. R. Clemente, J. L. Jiménez, J. C. Palacios, Conformation of Secondary Amides. A Predictive Algorithm That Correlates DFT-Calculated Structures and Experimental Proton Chemical Shifts. J. Org. Chem. 68, 1834-1842 (2003). 8. D. M. Pawar, A. A. Khalil, D. R. Hooks, K. Collins, T. Elliott, J. Stafford, L. Smith, E. A. Noe, E and Z Conformations of Esters, Thiol Esters, and Amides. J. Am. Chem. Soc. 120, 21082112 (1998). 9. I. P. Gerothanassis, I. N. Demetropoulos, C. Vakka, Hydration of cis and trans Nmethylformamide as Revealed by the Use of O-17-NMR, Molecular Mechanics, and ab-initio Calculations. Biopolymers 36, 415-428 (1995). 5 10. M. J. Deetz, J. E. Fahey, B. Smith, NMR Studies of Hydrogen Bonding Interactions with Secondary Amides and Urea Groups. J. Phys. Org. Chem. 14, 463-467 (2001). 11. H. Jangra, M. H. Haindl, F. Achrainer, J. Hioe, R. M. Gschwind, H. Zipse, Conformational Preferences in Small Peptide Models: The Relevance of cis/trans-Conformations. Chem. Eur. J. 22,13328-13335 (2016). 12. A. Jabs, M. S. Weiss, R. Hilgenfeld, Non-proline Cis Peptide Bonds in Proteins. J. Mol. Biol. 286, 291-304 (1999). 13. M. H. K. Ebbing, M. J. Villa, J. M. Valpuesta, P. Prados, J. de Mendoza, Resorcinarenes with 2-Benzimidazolone Bridges: Self-Aggregation, Self-Assembled Dimeric Capsules, and Guest Encapsulation. Proc. Nat. Acad. Sci. U.S.A. 99, 4962-4966 (2002). 14. H. J. Choi, Y. S. Park, C. S. Cho, K. Koh, S. H. Kim, K. Paek, Unusually Stable Molecular Capsule Formation of a Tetraphenyleneurea Cavitand. Org. Lett. 6, 443-4433 (2004). 15. K. D. Zhang, D. Ajami, J. V. Gavette, J. Rebek, Alkyl Groups Fold to Fit within a WaterSoluble Cavitand. J. Am. Chem. Soc. 136, 5264-5266 (2014). 16. P. V. R. Schleyer, C. Maerker, A. Dransfeld, H. Jiao, N. J. R. V. E. Hommes, NucleusIndependent Chemical Shifts:  A Simple and Efficient Aromaticity Probe. J. Am. Chem. Soc. 118, 6317-6318 (1996). 17. S. Mosca, Y. Yu, J. V. Gavette, J. Rebek, A Deep Cavitand Templates Lactam Formation in Water. J. Am. Chem. Soc. 137, 14582-4585 (2015). 18. L. Trembleau, J. Rebek, Helical Conformation of Alkanes in Hydrophobic Environments. Science 301, 1219-220 (2003). 19. A typical trans/cis ratio of 4:1 for the formyl group results in a statistical distribution of diformamides: 64% trans,trans; 32% trans,cis and 4% cis,cis. The latter is rarely observed directly due to overlapped signals and low intensities. 20. Y. Yu, J. Rebek, Jr., Reactions of Folded Molecules in Water. Acc. Chem. Res. 51, 3031-3040 (2018). 21. Y. Yu, K. D. Zhang, I. D. Petsalakis, G. Theodorakopoulos, J. Rebek, Asymmetric Binding of Symmetric Guests in a Water-soluble Cavitand. Supramol. Chem. 30, 473-478 (2018). 22. S. Liu, D. H. Russell, N. F. Zinnel, B. C. Gibb, Guest Packing Motifs within a Supramolecular Nanocapsule and a Covalent Analogue. J. Am. Chem. Soc. 135, 4314-4324 (2013). 23. K. Wang, B. C. Gibb, Mapping the Binding Motifs of Deprotonated Monounsaturated Fatty Acids and their Corresponding Esters within Supramolecular Capsules. J. Org. Chem. 82, 4279-4288 (2017). 24. S. Paliwal, S. Geib, S, C. S. Wilcox, Molecular Torsion Balance for Weak Molecular Recognition Forces. Effects of “Tilted-T” Edge-to-Face Aromatic Interactions on 6 Conformational Selection and Solid-State Structure. J. Am. Chem. Soc. 116, 4497-4498 (1994). Fig. 1. Amide geometries and areas of steric clashes that effect cis/trans equilibria. Fig. 2. Structures of cavitands 1 and cartoon for the complex of 1 with 2a: (A) Chemical structure of the cavitand host 1; (B) The cartoon abbreviation for its host/guest complex with N-octyl-formamide (2a). The observed Nucleus Independent Chemical Shifts (NICS)16,17 for typical guest nuclei are given in - ppm. The magnetic effects increase gradually with depth in the cavitand as the chemical environment changes from hydrophilic at the top of the cavity to hydrophobic at the bottom. Fig. 3. Partial 1H NMR spectra (600 MHz, 298K) of free and bound 2a and cartoons of its behavior in 1: (A) The spectra of free N-octyl-formamide (2a) in CDCl3 (top) and D2O (bottom). The cis/trans ratio can be determined from the formyl C-H signals; (B) Upfield region of the spectrum of the formamide’s complex of 2a in 1 (see also Figures S1-1 and 1-2). The low intensity set of signals represents the cis isomer; (C) Cartoon of proposed conformations of cis and trans formamide in the cavitand. Fig. 4. Partial 1H NMR spectra of the complexes of diformamide 3 and 1 and their cartoons in 1: (A) Upfield regions of the 1H NMR spectra (600 MHz, D2O, 298 K) of the diformamide complexes of 1. 1) n = 8 (3a); 2) n = 9 (3b); 3); n = 10 (3c); 4) n = 11 (3d); The larger signal clusters represent the symmetrical trans,trans isomers19 while the smaller set represents the trans,cis isomers. (see also Figure S3); (B) Cartoons of the isomeric complexes and their proposed interconversion. Fig. 5. Complexation of 3b: (Top) The 1H NMR spectrum of a D2O solution of [1] = 1 mM and [3b][(CH2)9] = 2 mM. (Bottom) Selected downfield (panel a) and upfield (panel b) regions of the 2D-EXSY spectrum (mixing time 300 ms) showing the cross-peaks due to chemical exchange between the protons of free and bound (CH2)9 diformamide (see also Figures S5 and S6). Primed numbers are proton signals for the bound trans,cis isomer (black). Doubly primed labels correspond to the trans,trans counterpart (blue). Triply primed labels correspond to the cis,cis diformamide (red).19 7 8