- Email: [email protected]
Molecular dynamics simulation of a cucurbituril based molecular switch triggered by pH changes
Accepted Manuscript Molecular Dynamics Simulation of a Cucurbituril Based Molecular Switch Triggered by pH Changes Lama D. Malhis, Khaled Bodoor, Khaleel I. Assaf, Nada A. Al-Sakhen, Musa I. El-Barghouthi PII: DOI: Reference:
S2210-271X(15)00199-1 http://dx.doi.org/10.1016/j.comptc.2015.05.010 COMPTC 1817
To appear in:
Computational & Theoretical Chemistry
Received Date: Revised Date: Accepted Date:
3 April 2015 5 May 2015 5 May 2015
Please cite this article as: L.D. Malhis, K. Bodoor, K.I. Assaf, N.A. Al-Sakhen, M.I. El-Barghouthi, Molecular Dynamics Simulation of a Cucurbituril Based Molecular Switch Triggered by pH Changes, Computational & Theoretical Chemistry (2015), doi: http://dx.doi.org/10.1016/j.comptc.2015.05.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Molecular Dynamics Simulation of a Cucurbituril Based Molecular Switch Triggered by pH Changes Lama D. Malhis1, Khaled Bodoor2, Khaleel I. Assaf3, Nada A. Al-Sakhen1 and Musa I. ElBarghouthi1,* 1
Department of Chemistry, The Hashemite University, Zarqa 13115, Jordan 2
3
Department of Physics, The University of Jordan, Amman 11942, Jordan
School of Engineering and Science, Jacobs University, Campus Ring 1, D-28759 Bremen, Germany
Abstract Molecular Dynamics (MD) Simulations are performed for a cucurbituril-pseudorotaxane system that functions as a molecular switch in which the movement of the macrocycle (CB) component of the rotaxane between two sites is triggered by an appropriate change in pH. The system consists of a CB bead and a fluorenyltriamine string in aqueous solution. Examining the extracted snapshots reveals that the fluorenyl ring bends toward the CB rim when it locates at the butyl moiety, and it becomes aligned with the rest of the guest when it locates at the hexyl moiety. No movement of the macrocycle between the possible positions for CB6 complexes for the fully protonated or the diprotonated guests was observed during the 20 ns simulation. However, such movement was observed in the case of CB7 and CB8. Umbrella sampling was used to obtain the potential of mean force curves, which demonstrates the existence of an energy barrier between the two positions. The height of the barrier was found to decrease as the size of the CB increases. Molecular Mechanics Poisson-Boltzmann Surface Area results indicate that the host-guest electrostatic interactions have the largest contribution to the stability of the complexes.
Keywords: Cucurbit[n]uril; Simulations; Molecular switches; MM-PBSA. *Corresponding Author: Department of Chemistry, The Hashemite University, P.O. Box 150459, Zarqa 13115, Jordan Tel : +962 (5) 3903333 Fax : +962 (5) 3826613 Email : [email protected]
1
Introduction Cucurbit[n]urils (CB[n]) are a family of macrocyclic molecules that consist of n glycoluril units bridged by 2n methylene groups [1-3]. CBs are available in different sizes (n = 5−8 and 10) and are well known for being good cation receptors owing to their two carbonyl portals, which establish natural docking sites for positively charged residues [1,4,5]. Additionally, CBs are well known to host hydrophobic residues in their cavities. The combination of ion-dipole and hydrophobic interactions enables CBs to bind a wide range of synthetic and naturally occurring molecules with exceptional strength [6-9]. CBs have been used for various applications, including drug delivery, dye tuning, and molecular switching [8,10-31]. Various CB-based rotaxanes that function as molecular switches have recently been synthesized and studied [32-39]. An external stimulus, such as a change in pH or temperature, can be used to control their switching behavior. The ability of CBs to interact preferentially with cationic species compared to neutral species makes amine-containing organic compounds attractive candidates for use as guest molecules in CB-based molecular switches [8, 32, 37, 40]. One can control the pH of the medium and hence the protonation states of the amine groups of the guest molecule, modifying the strength of their interactions with the CB macrocycle. Mock and Pierpont studied the encapsulation of the ligand PhNH(CH 2)6NH(CH2)4NH2 within CB6 [32]. When the aniline nitrogen is protonated, CB6 prefers to bind to the six-carbon (diaminohexane) site, which allows the ammonium groups to reside closer to the center of the carbonyl rims of the CB6 cavity, thereby maximizing the ion-dipole interactions [41]. On the other hand, when the aniline nitrogen is in a deprotonated state, the CB6 moves to the four-carbon (diaminobutane) site in the ligand [32]. Inspired by the above work of Mock and Pierpont, Kim and coworkers designed a fluorescent CB6-based molecular switch, (Scheme 1). The CB6 bead resides at the
2
diaminohexane site at pH 1.0, but shuttles to the diaminobutane site at pH 8.0 as shown by NMR studies [33]. Equilibrium between the two binding sites was observed at intermediate pH values. They found that the shuttling of CB6 between the two sites is slow on the NMR time scale, as separate resonances are observed for each binding mode [33]. Investigation of host-guest complexation processes can be carried out with the help of molecular dynamics (MD) simulations, which have proved to be a powerful tool for identifying and computing the driving forces for complex formation as well as for solving the optimal geometries of the resulting complexes [40, 42-47]. In the present work, we use MD simulations to study the thermodynamics of the (host) CB macrocycle interacting with the (guest) fluorenyltriamine molecule at the two possible sites (Scheme 1). The guest can exist in two protonation states: a diprotonated state with two of the three amine groups protonated, and a fully protonated state in which the aromatic amine is protonated as well (Scheme 1). For each state, we carried out simulations for the complexes formed between the guest at the appropriate site and each of the three CB homologous CB6, CB7 and CB8. The MD trajectories we obtained were analyzed to deduce the forces driving host-guest complexation and the relative stability of each complex at a given site in a given protonation state. Moreover, we used the Molecular Mechanics Poisson-Boltzmann Surface Area (MM-PBSA) method to estimate the free energy of complexation as a sum of contributions from appropriate host-guest interactions. Umbrella sampling was used to compute the height of the energy barrier separating the two presumed binding sites of the macrocycle on the guest.
3
Computational Methods The initial molecular geometries of CB6, CB7, and CB8 were obtained from X-ray diffraction data [2, 3]. The starting geometries of the studied complexes were obtained by manually inserting the host into the desired position in the guest molecule. MD simulations were performed with the sander module of the AMBER 11 program [48] using general force field parameters sets [49, 50]. The atomic charges for the host and guest molecules were obtained using AM1-BCC charges [51]. Each system was solvated in a truncated octahedral periodic box of TIP3P [52] water molecules. Chloride ions were added to maintain the neutrality of each system. The non-bonded cut off was set to 12.0 Å. Each given solvated complex was subjected to energy minimization using the conjugate gradient algorithm prior to starting the MD simulation. The minimized complex was then heated up to 298 K for 60 ps, followed by a 500 ps equilibration step at 298 K and 1 atm. During minimization and MD simulation, the Particle Mesh Ewald method (PME) was employed to treat the long range electrostatic interactions in periodic boundary conditions [53]. All bond lengths involving hydrogen atoms were constrained by means of the SHAKE Algorithm [54]. Production runs were carried out for 20 ns at 298 K and 1.0 atm. A 2-fs time step was used, saving structures every 2 ps, and updating the nonbonded pair list every 25 steps. The MD simulation analysis was performed with PTRAJ module of the AMBER 11 program. The VMD 1.8.6 program was used to visualize the structures [55]. The MM-PBSA and normal mode analysis were performed according to the procedure described elsewhere [43,44]. All snapshots (10,000) were used in the different analyses performed in this study. The reaction coordinate used in umbrella sampling was the distance (r) between the terminal nitrogen (N1) in guests A and B and the center of mass of the oxygen atoms of the far carbonyl
4
portal. The values of r ranged from 5 to 18 Å in 1.0 Å increments. A force constant of 16 kcal/mol.Å2 was applied at each position. The extended form of the guest molecule was considered and hence a restraint on guest geometry was employed. Attempts to conduct umbrella sampling while allowing the guest molecule to move without restraints proved difficult because of the bad overlap between the sampled distance distributions. Each biasing simulation consists of an equilibrium run for 500 ps followed by a 1000 ps production run. The distance data for each simulation were collected in 2.0 fs intervals. The results were post-processed using the Weighted Histogram Analysis Method (WHAM) [56, 57].
Results and Discussion The pseudorotaxane system comprised of a CB6 bead and a fluorenyltriamine string has been studied by Kim and coworkers [33]. The shuttling process in this system was found to be controlled by the pH of the medium. The pKa of the aromatic amine of fluorenyltriamine is 5.7, while for the primary and secondary amines it is greater than 10 [33]. This means that at pH 1 the three amine groups of fluorenyltriamine are protonated, while at pH 8 the aromatic amine is neutral and the alkyl amino groups remain protonated. As NMR studies have shown, CB resides at the diprotonated diaminohexane site at pH 1.0 (position 1, Scheme 1) while at pH 8.0 it resides at the diprotonated diaminobutane site (position 2, Scheme 1). MD simulations were carried out to study the thermodynamics of the complexation process at the two possible sites both for the fluorenyltriamine molecule in the triprotonated, (A), and diprotonated, (B), states (Scheme 2). In addition to CB6, the larger CB homologues, CB7 and CB8, were also investigated. 20 ns MD simulations for the free hosts, guests, and their corresponding molecular complexes were carried out.
5
The guest dynamics inside the host cavity was monitored by the superposition of 25 equally spaced snapshots extracted from the MD trajectory for each complex, superimposed on a representative host structure. The clusters of snapshots for the free guests and their complexes that were obtained are given in the supporting information (Figures S1-S7). Moreover, 5 snapshots corresponding to 4, 8, 12, 16, and 20 ns for CB6 complexes are shown in Figure 1. Inspection of the snapshots shows stable binding of CB at both positions 1 and 2 in each of the CB6/A and CB6/B complexes (Figures 1, S2-S3), with no inter-site movement of CB6 taking place during the 20 ns simulations. This indicates the existence of a high energy barrier to host migration from one site to another in CB6 complexes. This view is supported by 1H NMR study of CB6/A, which shows that the shuttling of CB between the two sites is slow on the NMR time scale [33]. Conducting a 300-ns MD simulation of position-1 CB6/B complex also reveals no migration of CB6 to the presumably more stable position 2.
On the other hand, MD simulations for position-2 CB7/A and CB8/A complexes showed movement of the macrocycle toward position 1. This process was found to occur at the equilibration period of MD simulation for CB8. For CB7, the snapshots sampled at the beginning of the production run revealed that the fluorenyl ring in guest A is bent toward the CB7 rim (Figure 2). Then, 8 ns later, the host starts moving toward N2 while the ring moiety moves away from the CB7 rim. Movement of CB7 continues until it finally reaches position 1 at 11 ns and
6
remains there for the rest of the simulation (Figure 2). Figure 3 shows this movement, where the distance between the center of mass of CB7 and nitrogen atom N3 is plotted against simulation time. Interestingly, no movement in either direction was observed for the CB7/B or CB8/B complexes.
These results point to the presence of an energy barrier for the shuttling process for CB6, and smaller ones for the larger CB7 and CB8. The larger barrier in the case of CB6 can be explained by its small cavity size, which causes the intermediate states for the movement, which involve the presence of the protonated amine (N2) inside the CB6 cavity, to possess high energies. Umbrella sampling was used to examine this point; the results obtained will be discussed later. It should be noted here that all results presented below for CB7/A position-2 complex are just for the first 8 ns of its MD trajectory, during which the snapshots show CB7 to be at position 2. The simulations for CB8/A in positions 1 and 2 after the equilibration period are similar and, therefore, only the results for position 1 are shown. Returning to the snapshots for position 1, the guest molecule assumes an extended form (Figures S2, S4 and S6) compared to the free molecules especially for the hexyl segment (Figure S1). This is probably driven by the enhancement of the size match between the diaminohexane part of the guest and the CB molecule as well as the more favorable positioning of the ammonium groups, resulting in increased ion dipole interaction and hydrogen bonding with the CB portals. A much less constrained form of the guests is observed when CB is at position 2, which appears to allow it to interact with the flourenyl ring of the guest (Figures S3, S5 and S7). Root-mean-
7
square-fluctuation analysis was performed on the 10000 snapshots of guests A and B and their complexes (Figure 4) and demonstrate significant increase in guest flexibility upon complexation at position 2, in agreement with snapshots shown in Figures S3, S5 and S7.
The numbers of water-guest and host-guest hydrogen bonds (HBs) are presented in Tables 1 and 2, respectively. A HB cut distance ≤ 3.0 Å and angle ≥ 120º were used for the analysis. The Free guests form several HBs with the surrounding water molecules (Table 1), with a total of 6.15 and 4.88 for A and B, respectively. A reduction was observed in the number of guest-water HBs upon complexation. This reduction is less pronounced as a result of the decreased steric hindrance to the water molecules in the larger CBs.
S2210-271X(15)00199-1 http://dx.doi.org/10.1016/j.comptc.2015.05.010 COMPTC 1817
To appear in:
Computational & Theoretical Chemistry
Received Date: Revised Date: Accepted Date:
3 April 2015 5 May 2015 5 May 2015
Please cite this article as: L.D. Malhis, K. Bodoor, K.I. Assaf, N.A. Al-Sakhen, M.I. El-Barghouthi, Molecular Dynamics Simulation of a Cucurbituril Based Molecular Switch Triggered by pH Changes, Computational & Theoretical Chemistry (2015), doi: http://dx.doi.org/10.1016/j.comptc.2015.05.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Molecular Dynamics Simulation of a Cucurbituril Based Molecular Switch Triggered by pH Changes Lama D. Malhis1, Khaled Bodoor2, Khaleel I. Assaf3, Nada A. Al-Sakhen1 and Musa I. ElBarghouthi1,* 1
Department of Chemistry, The Hashemite University, Zarqa 13115, Jordan 2
3
Department of Physics, The University of Jordan, Amman 11942, Jordan
School of Engineering and Science, Jacobs University, Campus Ring 1, D-28759 Bremen, Germany
Abstract Molecular Dynamics (MD) Simulations are performed for a cucurbituril-pseudorotaxane system that functions as a molecular switch in which the movement of the macrocycle (CB) component of the rotaxane between two sites is triggered by an appropriate change in pH. The system consists of a CB bead and a fluorenyltriamine string in aqueous solution. Examining the extracted snapshots reveals that the fluorenyl ring bends toward the CB rim when it locates at the butyl moiety, and it becomes aligned with the rest of the guest when it locates at the hexyl moiety. No movement of the macrocycle between the possible positions for CB6 complexes for the fully protonated or the diprotonated guests was observed during the 20 ns simulation. However, such movement was observed in the case of CB7 and CB8. Umbrella sampling was used to obtain the potential of mean force curves, which demonstrates the existence of an energy barrier between the two positions. The height of the barrier was found to decrease as the size of the CB increases. Molecular Mechanics Poisson-Boltzmann Surface Area results indicate that the host-guest electrostatic interactions have the largest contribution to the stability of the complexes.
Keywords: Cucurbit[n]uril; Simulations; Molecular switches; MM-PBSA. *Corresponding Author: Department of Chemistry, The Hashemite University, P.O. Box 150459, Zarqa 13115, Jordan Tel : +962 (5) 3903333 Fax : +962 (5) 3826613 Email : [email protected]
1
Introduction Cucurbit[n]urils (CB[n]) are a family of macrocyclic molecules that consist of n glycoluril units bridged by 2n methylene groups [1-3]. CBs are available in different sizes (n = 5−8 and 10) and are well known for being good cation receptors owing to their two carbonyl portals, which establish natural docking sites for positively charged residues [1,4,5]. Additionally, CBs are well known to host hydrophobic residues in their cavities. The combination of ion-dipole and hydrophobic interactions enables CBs to bind a wide range of synthetic and naturally occurring molecules with exceptional strength [6-9]. CBs have been used for various applications, including drug delivery, dye tuning, and molecular switching [8,10-31]. Various CB-based rotaxanes that function as molecular switches have recently been synthesized and studied [32-39]. An external stimulus, such as a change in pH or temperature, can be used to control their switching behavior. The ability of CBs to interact preferentially with cationic species compared to neutral species makes amine-containing organic compounds attractive candidates for use as guest molecules in CB-based molecular switches [8, 32, 37, 40]. One can control the pH of the medium and hence the protonation states of the amine groups of the guest molecule, modifying the strength of their interactions with the CB macrocycle. Mock and Pierpont studied the encapsulation of the ligand PhNH(CH 2)6NH(CH2)4NH2 within CB6 [32]. When the aniline nitrogen is protonated, CB6 prefers to bind to the six-carbon (diaminohexane) site, which allows the ammonium groups to reside closer to the center of the carbonyl rims of the CB6 cavity, thereby maximizing the ion-dipole interactions [41]. On the other hand, when the aniline nitrogen is in a deprotonated state, the CB6 moves to the four-carbon (diaminobutane) site in the ligand [32]. Inspired by the above work of Mock and Pierpont, Kim and coworkers designed a fluorescent CB6-based molecular switch, (Scheme 1). The CB6 bead resides at the
2
diaminohexane site at pH 1.0, but shuttles to the diaminobutane site at pH 8.0 as shown by NMR studies [33]. Equilibrium between the two binding sites was observed at intermediate pH values. They found that the shuttling of CB6 between the two sites is slow on the NMR time scale, as separate resonances are observed for each binding mode [33]. Investigation of host-guest complexation processes can be carried out with the help of molecular dynamics (MD) simulations, which have proved to be a powerful tool for identifying and computing the driving forces for complex formation as well as for solving the optimal geometries of the resulting complexes [40, 42-47]. In the present work, we use MD simulations to study the thermodynamics of the (host) CB macrocycle interacting with the (guest) fluorenyltriamine molecule at the two possible sites (Scheme 1). The guest can exist in two protonation states: a diprotonated state with two of the three amine groups protonated, and a fully protonated state in which the aromatic amine is protonated as well (Scheme 1). For each state, we carried out simulations for the complexes formed between the guest at the appropriate site and each of the three CB homologous CB6, CB7 and CB8. The MD trajectories we obtained were analyzed to deduce the forces driving host-guest complexation and the relative stability of each complex at a given site in a given protonation state. Moreover, we used the Molecular Mechanics Poisson-Boltzmann Surface Area (MM-PBSA) method to estimate the free energy of complexation as a sum of contributions from appropriate host-guest interactions. Umbrella sampling was used to compute the height of the energy barrier separating the two presumed binding sites of the macrocycle on the guest.
3
Computational Methods The initial molecular geometries of CB6, CB7, and CB8 were obtained from X-ray diffraction data [2, 3]. The starting geometries of the studied complexes were obtained by manually inserting the host into the desired position in the guest molecule. MD simulations were performed with the sander module of the AMBER 11 program [48] using general force field parameters sets [49, 50]. The atomic charges for the host and guest molecules were obtained using AM1-BCC charges [51]. Each system was solvated in a truncated octahedral periodic box of TIP3P [52] water molecules. Chloride ions were added to maintain the neutrality of each system. The non-bonded cut off was set to 12.0 Å. Each given solvated complex was subjected to energy minimization using the conjugate gradient algorithm prior to starting the MD simulation. The minimized complex was then heated up to 298 K for 60 ps, followed by a 500 ps equilibration step at 298 K and 1 atm. During minimization and MD simulation, the Particle Mesh Ewald method (PME) was employed to treat the long range electrostatic interactions in periodic boundary conditions [53]. All bond lengths involving hydrogen atoms were constrained by means of the SHAKE Algorithm [54]. Production runs were carried out for 20 ns at 298 K and 1.0 atm. A 2-fs time step was used, saving structures every 2 ps, and updating the nonbonded pair list every 25 steps. The MD simulation analysis was performed with PTRAJ module of the AMBER 11 program. The VMD 1.8.6 program was used to visualize the structures [55]. The MM-PBSA and normal mode analysis were performed according to the procedure described elsewhere [43,44]. All snapshots (10,000) were used in the different analyses performed in this study. The reaction coordinate used in umbrella sampling was the distance (r) between the terminal nitrogen (N1) in guests A and B and the center of mass of the oxygen atoms of the far carbonyl
4
portal. The values of r ranged from 5 to 18 Å in 1.0 Å increments. A force constant of 16 kcal/mol.Å2 was applied at each position. The extended form of the guest molecule was considered and hence a restraint on guest geometry was employed. Attempts to conduct umbrella sampling while allowing the guest molecule to move without restraints proved difficult because of the bad overlap between the sampled distance distributions. Each biasing simulation consists of an equilibrium run for 500 ps followed by a 1000 ps production run. The distance data for each simulation were collected in 2.0 fs intervals. The results were post-processed using the Weighted Histogram Analysis Method (WHAM) [56, 57].
Results and Discussion The pseudorotaxane system comprised of a CB6 bead and a fluorenyltriamine string has been studied by Kim and coworkers [33]. The shuttling process in this system was found to be controlled by the pH of the medium. The pKa of the aromatic amine of fluorenyltriamine is 5.7, while for the primary and secondary amines it is greater than 10 [33]. This means that at pH 1 the three amine groups of fluorenyltriamine are protonated, while at pH 8 the aromatic amine is neutral and the alkyl amino groups remain protonated. As NMR studies have shown, CB resides at the diprotonated diaminohexane site at pH 1.0 (position 1, Scheme 1) while at pH 8.0 it resides at the diprotonated diaminobutane site (position 2, Scheme 1). MD simulations were carried out to study the thermodynamics of the complexation process at the two possible sites both for the fluorenyltriamine molecule in the triprotonated, (A), and diprotonated, (B), states (Scheme 2). In addition to CB6, the larger CB homologues, CB7 and CB8, were also investigated. 20 ns MD simulations for the free hosts, guests, and their corresponding molecular complexes were carried out.
5
The guest dynamics inside the host cavity was monitored by the superposition of 25 equally spaced snapshots extracted from the MD trajectory for each complex, superimposed on a representative host structure. The clusters of snapshots for the free guests and their complexes that were obtained are given in the supporting information (Figures S1-S7). Moreover, 5 snapshots corresponding to 4, 8, 12, 16, and 20 ns for CB6 complexes are shown in Figure 1. Inspection of the snapshots shows stable binding of CB at both positions 1 and 2 in each of the CB6/A and CB6/B complexes (Figures 1, S2-S3), with no inter-site movement of CB6 taking place during the 20 ns simulations. This indicates the existence of a high energy barrier to host migration from one site to another in CB6 complexes. This view is supported by 1H NMR study of CB6/A, which shows that the shuttling of CB between the two sites is slow on the NMR time scale [33]. Conducting a 300-ns MD simulation of position-1 CB6/B complex also reveals no migration of CB6 to the presumably more stable position 2.
On the other hand, MD simulations for position-2 CB7/A and CB8/A complexes showed movement of the macrocycle toward position 1. This process was found to occur at the equilibration period of MD simulation for CB8. For CB7, the snapshots sampled at the beginning of the production run revealed that the fluorenyl ring in guest A is bent toward the CB7 rim (Figure 2). Then, 8 ns later, the host starts moving toward N2 while the ring moiety moves away from the CB7 rim. Movement of CB7 continues until it finally reaches position 1 at 11 ns and
6
remains there for the rest of the simulation (Figure 2). Figure 3 shows this movement, where the distance between the center of mass of CB7 and nitrogen atom N3 is plotted against simulation time. Interestingly, no movement in either direction was observed for the CB7/B or CB8/B complexes.
These results point to the presence of an energy barrier for the shuttling process for CB6, and smaller ones for the larger CB7 and CB8. The larger barrier in the case of CB6 can be explained by its small cavity size, which causes the intermediate states for the movement, which involve the presence of the protonated amine (N2) inside the CB6 cavity, to possess high energies. Umbrella sampling was used to examine this point; the results obtained will be discussed later. It should be noted here that all results presented below for CB7/A position-2 complex are just for the first 8 ns of its MD trajectory, during which the snapshots show CB7 to be at position 2. The simulations for CB8/A in positions 1 and 2 after the equilibration period are similar and, therefore, only the results for position 1 are shown. Returning to the snapshots for position 1, the guest molecule assumes an extended form (Figures S2, S4 and S6) compared to the free molecules especially for the hexyl segment (Figure S1). This is probably driven by the enhancement of the size match between the diaminohexane part of the guest and the CB molecule as well as the more favorable positioning of the ammonium groups, resulting in increased ion dipole interaction and hydrogen bonding with the CB portals. A much less constrained form of the guests is observed when CB is at position 2, which appears to allow it to interact with the flourenyl ring of the guest (Figures S3, S5 and S7). Root-mean-
7
square-fluctuation analysis was performed on the 10000 snapshots of guests A and B and their complexes (Figure 4) and demonstrate significant increase in guest flexibility upon complexation at position 2, in agreement with snapshots shown in Figures S3, S5 and S7.
The numbers of water-guest and host-guest hydrogen bonds (HBs) are presented in Tables 1 and 2, respectively. A HB cut distance ≤ 3.0 Å and angle ≥ 120º were used for the analysis. The Free guests form several HBs with the surrounding water molecules (Table 1), with a total of 6.15 and 4.88 for A and B, respectively. A reduction was observed in the number of guest-water HBs upon complexation. This reduction is less pronounced as a result of the decreased steric hindrance to the water molecules in the larger CBs.