Nonadditive Interactions Mediated by Water at Chemically

Subscriber access provided by READING UNIV

Article

Non-Additive Interactions Mediated by Water at Chemically Heterogeneous Surfaces: Non-ionic Polar Groups and Hydrophobic Interactions Chenxuan Wang, Chi-Kuen Derek Ma, Hongseung Yeon, Xiaoguang Wang, Samuel H. Gellman, and Nicholas L. Abbott J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b08367 • Publication Date (Web): 11 Nov 2017 Downloaded from http://pubs.acs.org on November 11, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of the American Chemical Society is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Non-Additive Interactions Mediated by Water at Chemically Heterogeneous Surfaces: Non-ionic Polar Groups and Hydrophobic Interactions Chenxuan Wang†,‡, Chi-Kuen Derek Ma†, Hongseung Yeon†, Xiaoguang Wang†, Samuel H. Gellman‡∗, and Nicholas L. Abbott†∗ †Department of Chemical and Biological Engineering, University of Wisconsin-Madison, 1415 Engineering Drive, Madison, Wisconsin 53706, USA, ‡Department of Chemistry, University of Wisconsin-Madison, 1101 University Avenue, Madison, Wisconsin 53706, USA ABSTRACT: We explore how two non-ionic polar groups (primary amine and primary amide) influence hydrophobic interactions of neighboring non-polar domains. We designed stable β-peptide sequences that generated globally amphiphilic (GA) helices, each with a non-polar domain formed by six cyclohexyl side chains arranged along one side of the 14-helix. The other side of the helix was dominated by three polar side chains, from β3-homolysine (K) and/or β3homoglutamine (Q) residues. Variations in this polar side chain array included exclusively β3-hLys (GA-KKK) and β3hLys/β3-hGln mixtures (e.g., GA-QKK and GA-QQK). Chemical force measurements in aqueous solution vs. methanol allowed quantification of the hydrophobic interactions of the β-peptide with the non-polar tip of an atomic force microscope (AFM). At pH 10.5, where the K side chain is deprotonated, we measured hydrophobic adhesive interactions mediated by GA-KKK to be 0.61 ± 0.04 nN, by GA-QKK to be 0.54 ± 0.01 nN, by GA-QQK to be 0 ± 0.01 nN. This finding suggests that replacing an amine group (K side chain) with a primary amide group (Q side chain) weakens the hydrophobic interaction generated by the six cyclohexyl side chains. AFM studies with solid-supported mixed monolayers containing an alkyl component (60%) and a component bearing either a terminal amide or amine group (40%) revealed analogous trends. These observations from two distinct experiment systems indicate that proximal non-ionic polar groups have pronounced effects on hydrophobic interactions generated by a neighboring non-polar domain, and that the magnitude of the effect depends strongly on polar group identity.

INTRODUCTION Hydrophobic interactions are water-mediated attractions between non-polar molecules or surfaces.1-4 These interactions provide the driving force for many self-assembly and molecular recognition processes, such as receptor-ligand binding, protein folding and formation of lipidic assemblies (e.g., vesicles).5,6 Most biological and technological systems, however, do not present non-polar domains/surfaces in isolation; such domains are typically flanked by a diverse range of uncharged polar and/or ionic groups.7 Although simulations have been used to predict the ways in which chemical heterogeneity can impact hydrophobic interactions, experimental validation of these predictions is lacking.7-12 Furthermore, the potential functions used in these simulations often omit key features of real molecules (e.g., polarizability of atoms and dissociation of water). Recently, to provide insight into the influence of nanoscale chemical heterogeneity on hydrophobic interactions, we reported experiments performed using two independent systems, sequence-specific oligopeptides that adopt rigid and predictable conformations and mixed self-assembled monolayers formed on

the surfaces of gold films.13,14 We used these two systems to explore how immobilized charged groups (ammonium or guanidinium) influence hydrophobic interactions.13

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 11

array of six cyclohexyl side chains, introduced as (1S,2S)-2aminocyclohexane carboxylic acid (ACHC) residues, which constituted a non-polar domain. In addition to providing a non-polar domain, the ACHC residues strongly preorganize the β-peptide backbone for 14-helical folding.17-19 The opposing face of the helix displayed three cationic side chains, from either β3-homolysine (β3-hLys or K) or β3-homoarginine (β3-hArg or R). The sequences and the predicted globally amphiphilic conformations of these β-peptides, GA-KKK and GA-RRR, are shown in Fig. 1a and 1b.13 We covalently immobilized each β-peptide onto a surface and quantified the influence of the cationic groups on hydrophobic adhesion between the ACHC-rich face of single oligopeptides and the tip of an atomic force microscope (AFM) that was made non-polar by a coating with gold and adsorbing a monolayer of dodecanethiol (Fig. 1c).13,14,20 When using GA-KKK, we measured protonation of the side chains of β3-hLys to increase the strength of the hydrophobic adhesion between the AFM tip and the ACHC-rich domain. In contrast, studies with GA-RRR indicated that the guanidinium-containing β3hArg side chains eliminated measurable hydrophobic interactions. This observation led us to the important hypothesis that charged groups immobilized within ~1 nm of a non-polar domain can modulate the strength of the hydrophobic interaction mediated by the domain, and that the structure of the cationic group (ammonium vs. guanidinium) was crucial, rather than simply net charge. The same conclusions were reached when we measured adhesion between a non-polar AFM tip and a mixed monolayer presenting alkyl and either ammonium (Am) or guanidinium (Gdm) groups.13

Figure 1. Influence of charged and polar side chains on hydrophobic interactions of globally amphiphilic (GA) βpeptides. (a, b) The chemical structures (a) and helical representations (b) of globally amphiphilic (GA) β-peptides, GAKKK, GA-RRR, GA-QKK, and GA-QQK at pH 10.5. The side 3 3 chain of β -homolysine is coded as red, β -homoarginine is 3 coded as green, and β -homoglutamine is coded as blue. (c) Schematic illustration of an alkyl-terminated AFM tip interacting with an immobilized GA-KKK peptide. The colors of GA-KKK side chains are coded to match (a) and (b). (d) and (e) Schematic illustrations of two possible mechanisms by which charged (d) or polar groups (e) can influence the organization of water molecules near the non-polar face of the globally amphiphilic (GA) β-peptide. The red circles represent cross-sections of the helical GA β-peptides. The solid red disk with a white plus sign in (d) represents the side 3 chains of β -homolysine residues. The blue disc with a white 3 “P” sign in (e) represents the side chains of β homoglutamine residues. Light blue spheres represent the interfacial water molecules.

The first experimental system was based on β-amino acid oligomers ("β-peptides") that fold into globally amphiphilic (GA) helices.13 These helices contain 14-atom hydrogen-bonded rings and have approximately three residues per turn.15,16 One face of the helix presented an

Two possible mechanisms by which proximal charged groups might influence hydrophobic interactions are sketched in Fig. 1d. The first mechanism is a “direct” charge-dipole interaction that occurs between the charged groups on one face of the helix and water molecules adjacent to the non-polar domains (shown as “Direct mechanism” in Fig. 1d). A possible second mechanism involves an “indirect” interaction. Specifically, perturbations to the structure of water near charged groups on one face of the helix propagate, via water-water interactions (e.g., hydrogen bonding), from the site of the ions to the water adjacent to the non-polar domain (shown as “Indirect mechanism” in Fig. 1d). These perturbations may include local fluctuations in the density of water, which simulations have suggested to play a role in hydrophobic interactions.8 Given that ammonium and guanidinium ions have divergent effects on hydrophobic interactions mediated by non-polar domains yet possess the same net charge at pH 7, the “direct” mechanism involving a simple charge-dipole interaction appears unlikely to account for our past experimental observations. The difference detected between ammonium and guanidinium groups, instead, suggests a role for the “indirect” mechanism (propagation of perturbations to solvent structure).12 If such an indirect mechanism is dominant, however, one would predict that immobilized polar but non-ionic groups would also modulate hydrophobic interactions in a


Page 3 of 11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


group-specific manner (Fig. 1e). In this paper, we report experiments that explore that prediction. Our experimental approach for exploring the effects of non-ionic polar side chains leveraged our prior experiments and understanding of the effects of β3-hLys residues on hydrophobic interactions. Specifically, we explored the influence of replacing β3-hLys residues in GAKKK with β3-homoglutamine (β3-hGln or Q) residues on hydrophobic interactions of globally amphiphilic βpeptides by performing AFM measurements with GAQKK and GA-QQK (Fig. 1a and 1b; GA-QQQ is not soluble and therefore could not be evaluated). The side chain of Q presents a primary amide group, which is polar but uncharged under all conditions explored in our study (pH between 7 and 10.5). At high pH, where the amino group of K is also not charged, our experimental design permits direct comparison of the effects of two non-ionic polar groups, primary amine versus primary amide, on hydrophobic interactions.13 To provide additional comparative insight, we prepared mixed monolayers presenting alkyl and either amine- or amide-terminated components, and explored the relative effects of these two non-ionic polar groups on hydrophobic interactions involving an AFM tip.

but not 60 vol% MeOH in aqueous TEA) is identified as the hydrophobic interaction mediated by the β-peptide.13 We measured the mean pull-off forces arising from hydrophobic interactions between the non-polar ACHC-rich domains and the alkyl-terminated AFM tip to be 0.61 ± 0.04 nN at pH 10.5, and to increase to 1.07 ± 0.01 nN at pH 7 (Fig. 2n). These results confirm that protonation of the β3-hLys side chains increases the hydrophobic adhesion mediated by the non-polar ACHC groups on the opposite face of the helix.

RESULTS AND DISCUSSION Force Measurements with GA β-Peptides. Prior to performing experiments with β-peptides containing β3-hGln, we confirmed our prior results obtained with GA-KKK because this peptide serves as an important reference against which we compare results obtained with GA-QKK and GA-QQK. The spring constants of AFM tips after gold deposition and chemical functionalization were calibrated to be 0.031 ± 0.006 N/m (nominally 0.01 N/m) and 0.088 ± 0.014 N/m (nominally 0.03 N/m) using the thermal tuning method on a Nanoscope V Multimode AFM.20 Following the protocol described in detail previously, we measured the pull-off force between single surface-immobilized GAKKK molecules and a non-polar AFM tip in either aqueous triethylammonium (TEA) buffer or aqueous TEA to which 60 vol% MeOH was added (Fig. 2a-d).13 Our past work established that use of 60 vol% MeOH eliminates the majority of hydrophobic interactions but does not measurably change screened Coulomb (electrical double layer) interactions in these systems.13 Accordingly, pull-off forces measured in the 60 vol% MeOH/40 vol% 10 mM TEA buffer are identified as being dominated by van der Waals and electrical double layer interactions. Consistent with our past measurements, we measured the mean pulloff forces in 60 vol% MeOH to decrease from 0.52 ± 0.02 nN to 0.35 ± 0.03 nN as the pH increased from 7 to 10.5 (Fig. 2m). This change in force is dominated by electrical double layer interactions between the side chain ammonium ions of GA-KKK and the AFM tip.13 To identify the hydrophobic interactions, we fitted the pull-off forces measured in aqueous 10 mM TEA buffer to two Gaussian distributions, one of which was the same as that measured in 60 vol% MeOH (Fig 2a-d). The second distribution of forces (i.e., those forces present in aqueous TEA

Figure 2. Pull-off forces measured using immobilized βpeptides. (a)-(l) Histograms of adhesive forces measured using immobilized GA-KKK (a-d), GA-QKK (e-h), and GA-QQK (i-l) in either aqueous 10 mM TEA buffer (red) or 60 vol% MeOH (blue). Arrows indicate the frequencies of non-adhesive contacts. (m, n) The dependence on pH of pull-off forces measured in 60 vol% MeOH (m) or hydrophobic interactions measured in aqueous 10 mM TEA (n) using GA-KKK (blue), GA-QKK (red), and GA-QQK (green). Force histograms were obtained using an alkylterminated AFM tip and 1,805 – 3,606 pull-off force curves from 2-4 independent samples (Table S1). Data show mean ± s.e.m. Lines are drawn to guide the eye.

Next, we measured the adhesive forces generated by GAQKK in aqueous TEA and then in 60 vol% MeOH, and identified the hydrophobic contribution by using the methodology described for GA-KKK (Fig. 2 e-h). For GAQKK, the mean pull-off force in 60 vol% MeOH was 0.30 ± 0.02 nN at pH 10.5, and increased to 0.37 ± 0.01 nN at



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

pH 7 (Fig. 2m) whereas the hydrophobic interaction was 0.54 ± 0.01 nN at pH 10.5, and increased to 0.73 ± 0.03 nN at pH 7 (Fig. 2n). Here we make two preliminary observations based on a comparison of results obtained with GAKKK and GA-QKK. First, at pH 7, we observe the replacement of K by Q in the peptide to weaken the adhesion measured in 60 vol% MeOH, consistent with the presence of interactions encoded by K that are chargerelated. For example, at pH 7, the force in 60 vol% MeOH generated by GA-KKK is 0.52 ± 0.02 nN, whereas for GAQKK the force decreases to 0.37 ± 0.01 nN. Second, and most interestingly, although the number of non-polar ACHC groups is identical in GA-QKK and GA-KKK, as compared to GA-KKK at the same pH, we measure a weaker hydrophobic interaction using GA-QKK (e.g., at pH 7, the hydrophobic interaction mediated by GA-KKK is 1.07 ± 0.01 nN whereas for GA-QKK this interaction is 0.73 ± 0.03 nN). We return to these points in the section below addressing the influence of Q on hydrophobic interactions. When an additional β3-hLys was substituted by β3-hGln, to generate GA-QQK, we measured the adhesion forces mediated by a single GA-QQK molecule to be the same in 60 vol% MeOH and aqueous TEA (Fig. 2i-l). This result indicates that GA-QQK does not generate a measurable hydrophobic interaction in our measurements (for pH 7 to 10.5, as shown in Fig. 2n). It would have been informative to perform measurements with GA-QQQ, but such studies were not feasible because of the insolubility of this β-peptide in solvents typically used for purification.

Page 4 of 11

Figure 3. Influence of polar side-chain patterning on pull-off forces measured using globally amphiphilic β-peptides. (a, b) Helical representations the globally amphiphilic β-peptides, GAQKK, GA-KQK, GA-KKQ (a), GA-QQK, GA-QKQ, and GAKQQ (b). The ACHC side chains are coded as black, K is coded as red, and Q is coded as blue. (c) and (e) The dependence on pH of pull-off forces measured in 60 vol% MeOH (c) or hydrophobic interactions measured in aqueous 10mM TEA (e) using GA-KKQ (blue), GA-KQK (red), and GA-QKK (green). (d) and (f) The dependence on pH of pull-off forces measured in 60 vol% MeOH (d) or hydrophobic interactions measured in aqueous 10mM TEA (f) using GA-KQQ (blue), GA-QQK (red), and GA-QKQ (green). The histograms are comprised of at least 1000 pull-off curves from 3 or 4 independent samples (Table S1). Data show mean ± s.e.m. Lines are drawn to guide the eye.

The effect of sequence on hydrophobic interactions. We performed two additional sets of measurements to provide further insight into the observations with GAKKK, GA-QKK, and GA-QQK. First, to determine whether any of the adhesion measurements involving GA-QKK or GA-QQK were dependent on the specific patterning of the K and Q side chains, we synthesized two sequence isomers of each β-peptide, GA-KKQ and GA-KQK (for GA-QKK), and GA-QKQ and GA-KQQ (for QQK), and measured adhesion in aqueous TEA before and after addition of 60 vol% MeOH (as shown in Fig. S1 and S2). Inspection of Fig. 3c and 3e (GA-QKK, GA-KQK, GA-KKQ) and Fig 3d and 3f (GA-QQK, GA-QKQ, GA-KQQ) reveals that, at the same pH values, the order of presentation of K and Q on the polar face of the peptide does not significantly impact the strength of the hydrophobic adhesion or the adhesion measured in 60 vol% MeOH. Second, we prepared a set of fully folded but non-globally amphiphilic β-peptides, iso-GA-KKK, iso-GA-QKK, and isoGA-QQK (Fig. 4a and 4b), which are isomers, respectively, of GA-KKK, GA-QKK and GA-QQK.21 Previously, we reported that the iso-GA-KKK does not mediate measurable hydrophobic interactions with the AFM tip (as confirmed in measurements at pH 7 described in Fig. 4c), because this β-peptide does not present a well-defined non-polar domain.14 To determine whether the absence of a well-defined non-polar domain on β-peptides containing Q also eliminates measurable hydrophobic interactions, we measured the adhesive interactions of iso-GAQKK and iso-GA-QQK at pH 7 (Fig. 4d and 4e). These results confirmed the absence of hydrophobic interactions. Importantly, these comparisons of GA vs. iso-GA sequence isomer pairs allow us to attribute the hydrophobic interactions measured with the GA peptides containing Q to the presence of a well-defined non-polar surface formed by ACHC side chains that are grouped together on one side of the 14-helix. Fig. 4 f-i shows measurements of adhesive interactions of iso-GA-KKK, iso-GAQKK, and iso-GA-QQK at pH 10.5 (Fig. 4 f-i), conditions under which both side-chains are expected to be deprotonated.. Overall, the measurements reported indicate that the introduction of Q into the GA peptides leads to changes in hydrophobic interactions of the non-polar ACHC do-


Page 5 of 11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


mains of the GA peptides with the non-polar AFM tip. We find also that Q changes the interactions mediated by GA peptides in 60 vol% MeOH (i.e., electrical double layer and van der Waals interactions). Below, we discuss these measurements and provide insight into the origins of several observations through additional adhesion measurements involving mixed monolayers of alkanethiols formed on gold films.

and charged K side chains (details provided in SI, Section 1.1).13 Here we focus our discussion on the force measurements in Fig. 2 that were performed at pH 10.5 because the side chain of K is not expected to be substantially protonated at this pH (SI section 2.8). Additional discussion of the results shown in Fig. 2 at pH values less than 10.5 is presented in the SI. At pH 10.5, our current results indicate that the replacement of K by Q leads to changes in interactions in 60 vol% MeOH that cannot be directly related to the charge status of the peptides. Specifically, we measure the adhesive forces in 60 vol% MeOH to decrease from 0.35 ± 0.03 nN (GA-KKK), to 0.30 ± 0.02 nN (GA-QKK), and 0.23 ± 0.01 nN (GA-QQK) (Fig. 2m, Table 1). These differences suggest that the primary amine group of K and the primary amide group of Q encode van der Waals interactions that differ in strength. We will return to this topic in the context of additional measurements of adhesion performed with monolayers presenting amine and amide groups.

Figure 4. Influence of homoglutamine side chain on hydrophobic interactions of globally amphiphilic (GA) and non-globally amphiphilic (iso-GA) β-peptides. (a, b) The chemical structures (a) and helical representations (b) of non-globally amphiphilic βpeptides, iso-GA-KKK, iso-GA-QKK, and iso-GA-QQK. (c)-(h) Histograms of pull-off forces measured using immobilized isoGA-KKK (c, f), iso-GA-QKK (d, g), and iso-GA-QQK (e, h) in either 10 mM TEA buffer (red) or 60 vol% MeOH (blue). Arrows indicate the frequencies of non-adhesive contacts. (i) The magnitudes of mean pull-off forces measured in 60 vol% MeOH using iso-GA β-peptides at pH 10.5. The histograms are comprised of 2,563 – 2,852 pull-off force curves from 3 independent samples (Table S1). Data show mean ± s.e.m.

Influence of Q on Interactions Measured in 60 vol% MeOH. In our previous measurements with GA-KKK, at pH 7, we detected charge-related interactions caused by an electrical double layer at the surface of the AFM tip

Figure 5. Histograms of pull-off forces measured in either pure methanol or 60 vol% MeOH using immobilized β-peptides. Monolayer surface formed from HS-C11-EG4-OH (a), GA-KKK (b), GA-QKK (c), GA-QQK (d), and iso-GA-QQK (e) in 60 vol% MeOH at pH 10.5 (blue) and in MeOH (grey). Arrows indicate the frequencies of non-adhesive contacts. (f) Comparison of pulloff forces measured in 60 vol% MeOH at pH 10.5 (blue) and in MeOH (grey). The statistical information is reported in Table S1. Data show mean ± s.e.m.

In the experiments, we used 60 vol % MeOH because some of the experimental conditions (e.g., pH 7 with K) introduced charge interactions and we established previously that TEA and TEA containing 60 vol% MeOH medi-



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ate the same charge interactions.13 However, at pH 10.5, the side chains of K residues are largely or entirely in the uncharged amino state, and thus adhesion between the AFM tip and β-peptide in 60 vol% MeOH is interpreted as arising from van der Waals interactions (no or negligible contribution from electrical double layer interactions). Past studies have demonstrated that van der Waals interactions mediated by MeOH and water are similar,13,22 and thus to test our interpretation of pull-off forces as arising from van der Waals interactions in 60% MeOH, we measured pull-off forces with GA-KKK, GA-QKK, GA-QQK, and iso-GA-QQK in pure methanol (Fig. 5) and compared the results to the previous measurements in 60% MeOH. In control experiments performed with pure MeOH, we measured the absence of pull-off forces when peptides were not immobilized on the surfaces (Fig 5a). When GAKKK was immobilized, the mean pull-off force in pure MeOH was measured to be 0.36 ± 0.04 nN, a value that is indistinguishable from that measured in 60 vol% MeOH (0.35 ± 0.03 nN; Fig. 5b and 5f). Similarly, we measured in pure MeOH the forces generated by GA-QKK (0.32 ± 0.02 nN), GA-QQK (0.22 ± 0.01 nN), and iso-GA-QQK (0.02 ± 0.01 nN) to be the same as those measured in 60 vol% MeOH at pH 10.5 by GA-QKK (0.30 ± 0.02 nN), GA-QQK (0.23 ± 0.01 nN), iso-GA-QQK (0.02 ± 0.01 nN), as shown in Fig. 5 c-f. Overall, our comparison of the pull-off forces measured in pure MeOH and 60 vol% MeOH at pH 10.5 indicates that both solvents mediate similar interactions, consistent with our interpretation of the pull-off forces being dominated by van der Waals forces (i.e, neither hydrophobic nor charge interactions). This conclusion extends to non-globally amphiphilic peptides (iso-GAQQK), which would be expected if the pull-off forces are dominated by van der Waals interactions. We compared measurements using iso-GA-KKK, iso-GAQKK, and iso-GA-QQK (Fig. 4a and 4b). At pH 10.5, we measured the pull-off forces in 60 vol% MeOH to decrease from 0.31 ± 0.01 nN (iso-GA-KKK) to 0.10 ± 0.01 nN (iso-GA-QKK) and then 0.02 ± 0.01 nN (iso-GA-QQK) (Fig. 4i, Table 1). This result confirms that replacement of K by Q leads to a substantial decrease in the adhesive forces measured in 60 vol% MeOH, as concluded from the GA isomers. We return below to discuss the interactions encoded by K versus Q in 60 vol% MeOH in the context of monolayers presenting amine and amide groups. We note, however, that the difference between iso-GA-KKK and iso-GA-QKK (0.21 ± 0.01 nN) is substantially greater than the difference between GA-KKK and GA-QKK (0.05 ± 0.03 nN). This result emphasizes that the spatial presentation of the ACHC, Q and K side chains influences solvation with 60 vol% MeOH. Finally, we comment on the forces mediated by GA isomers with distinct patterning of the K and Q side chains. In 60 vol% MeOH at pH 10.5, we observe that changes in the order of presentation of the K and Q along the polar faces of the peptides have little measurable effect on the interactions between the AFM tip and the β-peptide in 60 vol% MeOH (Fig. 3a and 3b). This observation indicates

Page 6 of 11

that interactions in 60 vol% MeOH mediated by K and Q side chains are not strongly dependent on patterning along the polar face of the β-peptide. Thus, while the change from GA-QKK to iso-GA-QKK, which alters the arrangement of the polar side chains around the periphery of the helix, has a pronounced effect on forces measured in 60 vol% MeOH (compare Fig. 2m to Fig. 4g at pH 10.5, Table 1), changes in the positions of the Q and K residues within a 'stripe' along one side of the helix (GAQKK vs. GA-KQK vs. GA-KKQ) have little effect on forces measured between the AFM tip and β-peptide under the same solvent conditions (Fig. 3c; pH 10.5). A key point emerging from these considerations is that replacement of K by Q influences the strength of van der Waals interactions between the peptide and AFM tip in 60 vol% MeOH in a manner that depends strongly on whether the peptide is globally amphiphilic (GA series) or not (iso-GA series). This point is relevant to the following section. Influence of Q on Hydrophobic Interactions. At pH 10.5, the hydrophobic interactions mediated by GA-KKK were measured to be 0.61 ± 0.04 nN, by GA-QKK to be 0.54 ± 0.01 nN, and by GA-QQK to be 0 ± 0.01 nN (as shown in Fig. 2n, Table 1). These observations reveal that the effects of proximal amine and amide groups on hydrophobic interactions are distinct, with the amide groups causing a substantial weakening of the hydrophobic interaction relative to the amine groups. Table 1. Pull-off forces arising from the interactions of β-peptide molecules with a non-polar AFM tip at pH 10.5. The second column shows the contribution of hydrophobic interactions to the pull-off forces.

We propose that the differential impact of primary amine vs. primary amide on hydrophobic interactions of neighboring non-polar domains reflects changes in the structure of water in the vicinity of the peptide, which in turn is influenced by hydrogen bonding and van der Waals interactions between the water and polar groups on the peptide side chains. Table 2 compares pertinent properties of primary amine and primary amide groups. The permanent dipole of ethylamine (1.22 D in the gas phase) is smaller than that of acetamide (3.76 D in the gas phase), and the dipole of butylamine (1.00 D in the gas phase) is smaller than that of propanamide (3.85 D in the gas phase).23,24 The geometries of the two groups are distinct because the amine nitrogen is sp3-hybridized while the amide nitrogen atom is sp2-hybridized. Primary amine


Page 7 of 11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


and primary amide groups have different abilities to form hydrogen bonds with the surrounding water molecules. Specifically, the experimental value of the hydration free energy of acetamide (-40.8 kJ/mol) is more favorable than the hydration free energy of ethylamine (-18.8 kJ/mol), and the hydration free energy of propanamide (-39.4 kJ/mol) is more favorable than that of butylamine (-17.8 kJ/mol).25,26 This trend in hydration free energies likely arises because a primary amide forms more hydrogen bonds with water molecules than does a primary amine. In particular, the primary amide N-H units are better hydrogen bond donors than primary amine N-H units;27 experimental values of the free energy of hydrogen bond donation to water are not available, but simulations suggest -28.4 kJ/mol for HCOHN-H…OH2 and -6.3 kJ/mol for H2N-H…OH2.27,28 The relative ability of primary amide oxygen and primary amine nitrogen to serve as hydrogen bond acceptors does not appear to have been clearly elucidated. Overall, we conclude that a primary amide group should form more favorable hydrogen bonds with nearby water molecules than does a primary amine group; therefore, the primary amide group should be more effective at decreasing the interfacial energy and perturbing the structure of water adjacent to a proximal nonpolar domain. Additional results reported below in the context of mixed monolayers presenting amide and amine groups support this conclusion. Table 2. Key differences in the physical properties and hydrogen bonding properties of primary amine and primary amide groups

GA-KKK, at pH 10.5, can engage in hydrophobic interactions. To explore further the influence of primary amide groups on hydrophobic interactions, we prepared mixed monolayers presenting both amide and alkyl groups (40% amide-decanethiol/60% decanethiol) (Fig. 6e). For comparison, we also prepared pure alkyl-terminated surfaces (100% decanethiol), pure primary amide-terminated surfaces (100% amide-decanethiol), pure primary amineterminated surfaces (100% amine-undecanethiol), and mixed amine/alkyl terminated surfaces (40% amineundecanethiol/60% decanethiol), as shown in Fig. 6b and Fig. 6e. These surfaces were prepared by incubating goldcoated silicon wafers in 1 mM ethanolic solutions of the appropriate thiols. We confirmed the compositions of the mixed monolayers using X-ray photoelectron spectroscopy (XPS) by comparing the peak area of N1s and S2p (Fig. S3 and Fig. S4). Our previous study demonstrates that the statistical adsorption process leading to the formation of a mixed monolayer containing an average of 40% polar group results in the presence of nanoscopic domains of non-polar species that can mediate hydrophobic interactions.13 Accordingly, mixed monolayers and designed GApeptides are two independent systems that can be used to investigate the influence of polar groups on hydrophobic interactions. The hydrophobic interactions generated by mixed monolayer surfaces were identified by comparing the pull-off forces in either 10 mM TEA aqueous solution or pure MeOH (Fig. 6c). At pH 10.5, the pull-off force in TEA buffer on alkyl-terminated monolayer surfaces is 24.7 ± 0.6 nN, and decreases to 6.0 ± 0.6 nN in MeOH (Fig. 6c). We interpret the residual adhesion force measured in MeOH to be due to the van der Waals interactions between the two alkyl-terminated surfaces and between the alkyl-terminated surface and solvent (Fig 6a).13 In general, the magnitude of a pull-off force is influenced by the work of adhesion between the two surfaces that are separated during pull-off.29 The Johnson-Kendall-Roberts (JKR) theory relates the adhesive force (Fad) to the effective radius of an AFM tip (R) and the work of adhesion (Wad) as: ଷ

Fad= πRWad ଶ

(1)

As shown in Fig. 6a, Wad is related to the excess free energy densities (γ) of the interfaces of the system before and after separation as:30 Wad = γsample-solvent + γtip-solvent - γtip-sample Mixed Monolayers Presenting Amide and Amine Groups. A key hypothesis that emerges from the experiments described above using β-peptides is that the replacement of a β3-hLys residue by a β3-hGln residue adjacent to a non-polar domain results in a decrease in the strength of the hydrophobic interaction mediated by the non-polar domain. Thus, for GA-QQK, the presence of the two β3-hGln residues eliminates measurable hydrophobic interactions. In contrast, the non-polar domain of

(2)

where γsample-solvent is the interfacial energy of the sample surface in contact with solvent, γtip-solvent is the interfacial energy of the AFM tip in contact with solvent, γtip-sample is the interfacial energy of the contact area formed between the AFM tip and sample interface. Importantly, Equations (1) and (2) emphasize that the pull-off force is influenced not only by direct interactions between the tip and sample, but also by interactions between the solvent and both



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the sample and tip surfaces. This understanding provides a framework for the discussion of pull-off forces. We measured the adhesion force between the alkylterminated AFM tip and an amide-terminated monolayer surface using aqueous TEA (Fig. 6c). At pH 10.5, adhesion measured in TEA buffer (0.2 ± 0.1 nN) was small and indistinguishable from that measured in MeOH (0.1 ± 0.1 nN), leading to the conclusion that the pure amideterminated surface does not generate hydrophobic interactions (Fig. 6d). In contrast to the alkyl groups, amide groups can participate in dipole-dipole interactions and hydrogen bonds with both methanol and water, thus generating low interfacial energies between amideterminated surfaces and these two solvents [γ100 amide-TEA in TEA vs. γ100 amide-MeOH in MeOH; details provided in SI (Sections 1.3 and 1.4)]. It is these low interfacial energies that lead to the small pull-off forces measured with the amideterminated surfaces (0.1 ± 0.1 nN) relative to alkylterminated surfaces (6.0 ± 0.6 nN) in MeOH.

Figure 6. Effects of amide versus amine groups on adhesive forces measured using mixed monolayer surfaces. (a) Schematic illustration of changes in interfacial energy associated with a processes that separates a non-polar AFM tip from a sample surface in a solvent. (b) Schematic illustration of monolayer surfaces; from left to right, alkyl-terminated monolayer, amide-terminated monolayer, and amine-terminated monolayer. (c) Pull-off forces measured using alkyl-terminated monolayer surfaces, amideterminated monolayer surfaces or amine-terminated monolayer surfaces, each interacting with a non-polar AFM tip. Red bars indicate adhesion measured in 10 mM TEA buffer at pH 10.5, and blue bars indicate adhesive forces measured in MeOH. (d) Contribution of hydrophobic interactions to pull-off forces measured using alkyl-terminated monolayer surfaces, amide-terminated monolayers, or amine-terminated monolayer surfaces in aqueous solution. (e) Schematic illustration of mixed amide/alkyl mono-

Page 8 of 11

layer surfaces (top) and mixed amine/alkyl monolayer surfaces (bottom). (f) Pull-off forces measured using 40% amideterminated mixed monolayers or 40% amine-terminated mixed monolayers, both interacting with a non-polar AFM tip in 10 mM TEA buffer at pH 10.5 (red) or in MeOH (blue). (g) Contribution of hydrophobic interaction to mean pull-off forces measured using 40% amide-terminated mixed monolayer or 40% amineterminated mixed monolayer surface in aqueous 10 mM TEA solution. Histograms of the distribution of pull-off forces are shown in Fig. S5. Statistical information is reported in Table S1. Data show mean ± s.e.m.

We conducted parallel studies with amine-terminated surfaces to enable comparisons with the observations based on amide-terminated surfaces (Fig. 6c and 6d). Neither type of surface generates hydrophobic interactions with the alkyl-terminated AFM tip at pH 10.5, where the amine groups are expected to be largely deprotonated.13 At this pH, the adhesion generated by amide-terminated surfaces (0.2 ± 0.1 nN) was measured to be smaller than for amine-terminated surface (1.3 ± 0.3 nN) (Fig. 5c).13 The difference between the pull-off forces generated by the amide- and amine-terminated surfaces can be understood by using equation (2) to describe the work of adhesion between the AFM tip and either surface in aqueous solution at pH 10.5 as: Wad tip-TEA-100 amide = γ100 amide-TEA + γtip-TEA - γtip-100 amide

(3)

Wad tip-TEA-100 amine = γ100 amine-TEA + γtip-TEA - γtip-100 amine

(4)

As discussed in the context of Table 2, both van der Waals interactions (e.g., resulting from the interactions of fluctuating dipoles) and hydrogen bonding with aqueous TEA are more pronounced for amide groups than for amine groups. Correspondingly, at pH 10.5, interfacial energies are expected to rank as γ100 amide-TEA < γ100 amine-TEA and γtip-100 amide < γtip-100 amine. The experimental observation that Wad tipTEA-100 amine > Wad tip-TEA-100 amide is consistent with equation (3) and (4) only if γ100 amide-TEA < γ100 amine-TEA but not if γtip-100 amide < γtip-100 amine. This result suggests that the interactions of the solvent with the pure amide/amine-terminated surfaces at pH 10.5 dominate the experimental trends in adhesion energies. These results are also consistent with our observations with GA peptides that the substitutions of K by Q weaken the adhesive forces in 60 vol% MeOH at pH 10.5. The connection between adhesion and interactions between polar groups and water suggests that amide groups are promising for design of interfaces that display low adhesion in water. Our findings are consistent also with a previous report that the substitution of alkylterminated groups at a self-assembled monolayer surface with amide-terminated groups appears to render the surface less adhesive to proteins in aqueous buffer.31 Measurements involving mixed monolayers containing alkyl groups and either primary amide or primary amine groups revealed that amide and amine groups within the mixed monolayers have distinct effects on hydrophobic interactions. Specifically, adhesion measurements on 40% amide-terminated surfaces in aqueous TEA solution and


Page 9 of 11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


in MeOH led us to the conclusion that the hydrophobic interactions generated by mixed monolayer containing 40% amide groups were 3.1 ± 0.4 nN (Fig. 6f and 6g). The adhesion force measured in MeOH, 0.8 ± 0.4 nN (Fig. 6f), arises from van der Waals interactions between the mixed monolayer and non-polar tip, interactions between the mixed monolayer and solvent, and interactions between non-polar tip and solvent. To compare the influence of amine versus amide on interactions mediated by the mixed monolayers, we show in Fig. 6f and 6g, results of adhesion force measurements performed using mixed monolayers of alkyl groups and either amide or amine groups at pH 10.5, aqueous TEA or MeOH. At pH 10.5, the adhesion force measured with the mixed monolayers containing amine groups is 9.0 ± 0.4 nN in TEA solution, and decreases to 1.4 ± 0.1 nN in MeOH. This trend indicates that amine groups within the mixed monolayer do not change hydrophobic interactions in the same manner as do amide groups. Specifically, at pH 10.5, the hydrophobic interaction mediated by the mixed monolayer containing 40% amine is 7.6 ± 0.4 nN, which is 2.5 times as much as the strength of the hydrophobic interaction generated by 40% amide mixed monolayers, 3.1 ± 0.4 nN. This key result reinforces our conclusion that proximal amide and amine groups have divergent influences on hydrophobic interactions of neighboring non-polar surfaces. The adhesive force in MeOH generated by the mixed monolayer containing 40% amide groups (0.84 ± 0.42 nN) is larger than for the pure amide-terminated surface (0.12 ± 0.05 nN). This observation is consistent with contact angle measurements on pure and mixed amideterminated monolayer surfaces (details provided in SI, Sections 1.5 and 1.6). For the pure amide-terminated monolayer surface, the effects of dipole-dipole interactions and hydrogen bonds formed between amide groups and MeOH (γ100 amide-MeOH) dominate the contribution of the energy of the interface between the amide-terminated monolayer and non-polar AFM tip (γtip-100 amide) to the work of adhesion. For the mixed monolayer, the loss of 60% amide groups on the surface substantially weakens the solvent interactions with the mixed monolayer surface, leading to an increase in the energy of the interfaces formed to the solvent and thus to relatively high adhesive forces in MeOH. We also performed measurements using mixed monolayer surfaces presenting 60% amide or amine groups at pH 10.5 (details provided in SI, Section 1.8). The different hydrophobic interactions measured using 60% amide vs. 60% amine-terminated surfaces confirm the divergent roles of amide and amine groups on hydrophobic interactions at non-polar surfaces. CONCLUSIONS Measurement of the adhesive interactions of a non-polar AFM tip with either β-peptides or mixed monolayers leads to the conclusion that primary amide groups immobilized near nonpolar domains decrease the strength of hydrophobic interactions involving those non-polar groups relative to primary amine groups immobilized near the non-polar domain. This result is most evident in our comparison of hydrophobic inter-

actions of GA-KKK and GA-QQK with the AFM tip at pH 10.5. Whereas we measured a hydrophobic interaction with GA-KKK, no measurable hydrophobic interaction was detected with GA-QQK. A comparable conclusion may be drawn from our measurements with mixed monolayers presenting alkyl and either amide or amine groups at pH 10.5. These results, when correlated with the hydrogen bonding characteristics and relative hydration free energies of these functional groups, generate the hypothesis that the divergent effects of amide and amine groups on hydrophobic interactions arise from the relative propensities of these two groups to perturb the structure of water in contact with adjacent non-polar domains. Table 3. Key differences in the hydrogen bonding properties of ammonium and guanidinium groups and the influence of these groups on hydrophobic interactions mediated by proximal non-polar domains.

Our observations that primary amide and primary amine groups differ greatly in their influence on hydrophobic interactions involving neighboring non-polar groups provide a fresh perspective for understanding the watermediated interactions that control associations with chemically heterogeneous surfaces and self-assembly in chemical and biological systems. More broadly, our results highlight the non-additive nature of the interactions mediated by water at chemically heterogeneous surfaces. Specifically, the interactions mediated by the non-polar ACHC domains of the GA β-peptides in our studies depend the local structure of water, which in turn is influenced by interactions of water molecules with adjacent polar functional groups (i.e., sharing of solvation shells).12 Our findings complement our earlier finding that the structure of a cationic group influences the hydrophobic interactions of neighboring non-polar domains. We observed immobilized guanidinium ions to have a substantially greater effect than ammonium ions on hydrophobic interactions mediated by proximal, non-polar domains, a result that correlates closely with the relative magnitudes of the hydration free energies of the two cations (Table 3).13,32 Our results suggest a correlation between hydration free energy and hydrophobic interaction for comparisons between two non-ionic groups (amine versus amide) or comparisons between two ionic groups (ammonium versus guanidinium). The correlation does not exist, however, for comparisons between a non-ionic group (amine) and an ionic group (ammonium). The latter may reflect contributions to the hydration free energy of an ion in water that are not related to local water structure. Over-



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

all, these fundamental studies draw attention to the complexity of the non-covalent networks that govern intermolecular associations, and the inadequacy of reducing these associations to sets of pairwise interactions. The context dependence of hydrophobicity points to the failure of context-independent measures (e.g., hydropathy scales) to quantify hydrophobicity patterns on a protein surface. Additional studies are urgently needed to elucidate fully these effects, and thus place our understanding of water-mediated interactions beyond assumptions of simple additivity.33-35

ASSOCIATED CONTENT Supporting Information. The supporting information is available free of charge via the Internet at http://pubs.acs.org. Experimental details including materials and methods; measurements of the influence of electrical double layers on pull-off forces; measurements of the influence of pH on hydrophobic interactions; additional discussion of adhesion and interfacial energies.

Page 10 of 11

13. Ma, C. D.; Wang, C.; Acevedo-Vélez, C.; Gellman, S. H.; Abbott, N. L. Nature 2015, 517, 347-350. 14. Acevedo-Vélez, C.; Andre, G.; Dufrêne, Y. F.; Gellman, S. H.; Abbott, N. L. J. Am. Chem. Soc. 2011, 133, 3981-3988. 15. Appella, D. H.; Christianson, L. A.; Karle, I. L.; Powell, D. R.; Gellman, S. H. J. Am. Chem. Soc. 1996, 118, 13071-13072. 16. Cheng, R. P.; Gellman, S. H.; DeGrado W. F. Chem. Rev. 2001, 101, 3219-3232. 17. Appella, D. H.; Barchi, J. J.; Durell, S. R.; Gellman, S. H. J. Am. Chem. Soc. 1999, 121, 2309-2310. 18. Raguse, T. L.; Lai, J. R.; Gellman, S. H. J. Am. Chem. Soc. 2003, 125, 5592-5593. 19. Lee, M.; Raguse, T. L.; Schinnerl, M.; Pomerantz, W. C.; Wang, X.; Wipf, P.; Gellman, S. H. Org. Lett. 2007, 9, 1801-1804. 20. Ma, C. D.; Acevedo-Vélez, C.; Wang, C.; Gellman, S. H.; Abbott, N. L. Langmuir 2016, 32, 2985–2995. 21. Pomerantz, W. C.; Grygiel, T. L. R.; Lai, J. R.; Gellman, S. H. Org. Lett. 2008, 10, 1799–1802. 22. Yeon, H.; Wang, C.; Van Lehn, R. C.; Abbott, N. L. Langmuir 2017, 33, 4628–4637. 23. Nelson, R. D., Jr.; Lide, D. R., Jr.; Maryott, A. A. Selected Values of Electric Dipole Moments for Molecules in the Gas Phase. In National Standard Reference Data Series; NSRDS-NBS 10; National Bureau of Standards: Gaithersburg, MD, 1967.

AUTHOR INFORMATION Corresponding Author * [email protected] * [email protected]

24. Lide, D. R. CRC Handbook of Chemistry and Physics, 85th Edition; CRC Press: Boca Raton, FL, 2004-2005.

Notes The authors declare no competing financial interest.

25. Cabani, S.; Gianni, P.; Mollica, V.; Lepori, L. J. Solution Chem. 1981, 10, 563–595.

ACKNOWLEDGMENT

26. No, K. T.; Kim, S. G.; Cho, K. H.; Scheraga, H. A. Biophys. Chem. 1999, 78, 127-145.

This research was primarily supported by the ARO (144481500-4-AAB1263; W911NF-14-1-0140 and W911NF-11-1-0251). Partial support by the Wisconsin Materials Research Science and Engineering Center is also acknowledged (NSF Grant DMR-1121288).

28. Johansson, A.; Kollman, P.; Rothenberg, S.; McKelvey, J. J. Am. Chem. Soc. 1974, 96, 3794–3800.

REFERENCES

29. Johnson, K. L.; Kendall, K.; Roberts, A. D. Proc. R. Soc. Lond. A 1971, 324, 301–313.

1. Chandler, D. Nature 2005, 437, 640–647. 2. Meyer, E. E.; Rosenberg, K. J.; Israelachvili, J. Proc. Natl. Acad. Sci. USA 2006, 103, 15739–15746. 3. Ducker, W. A.; Mastropietro, D. Curr. Opin. Colloid Interface Sci. 2016, 22, 51-58. 4. Garde, S. Nature 2015, 517, 277–279. 5. Whitesides, G. M.; Grzybowski, B. Science 2002, 295, 2418– 2421. 6. Dyson, H. J.; Wright, P. E.; Scheraga, H. A. Proc. Natl. Acad. Sci. USA 2006, 103, 13057–13061. 7. Giovambattista; N., Lopez, C. F.; Rossky, P. J.; Debenedetti, P. G. Proc. Natl. Acad. Sci. USA 2008, 105, 2274–2279. 8. Acharya, H.; Vembanur, S.; Jamadagni, S. N.; Garde, S. Faraday Discuss. 2010, 146, 353–365.

27. Scheiner, S. Hydrogen Bonding: A Theoretical Perspective 1st edn; Oxford University Press: New York, 1997.

30. Hiemenz, P. C.; Rajagopalan, R. Principles of Colloid and Surface Chemistry, 3rd edn, Revised and Expanded; Marcel Dekker, Inc: New York, 1997. 31. Kane, R. S.; Deschatelets, P.; Whitesides, G. M. Langmuir 2003, 19, 2388-2391. 32. Marcus, Y. Ions in Solution and their Solvation 1st edn; Wiley: Hoboken, NJ; 2015. 33. Batista, C. A. S.; Larson, R. G.; Kotov, N. A. Science, 2015, 350, 1242477. 34. Porel, M.; Thornlow, D. N.; Artim; C. M.; Alabi, C. A. ACS Chem. Biol., 2017, 12, 715–723. 35. Kwona, O.; Yooa, T. H.; Othona, C. M.; Van Deventera, J. A.; Tirrell, D. A.; Zewail, A. H. Proc. Natl. Acad. Sci. USA 2010, 107, 17101–17106.

9. Huang, K.; Gast, S.; Ma, C. D.; Abbott, N. L.; Szlufarska, I. J. Phys. Chem. B 2015, 119, 13152–13159. 10. Patel, A. J.; Varilly, P.; Jamadagni, S. N.; Hagan, M. F.; Chandler, D.; Garde S. J. Phys. Chem. B 2012, 116, 2498–2503. 11. Giovambattista, N.; Debenedetti, P. G.; Rossky, P. J. J. Phys. Chem. C 2007, 111, 1323–1332. 12. Li, L.; Fennell, C. J.; Dill, K. A. J. Phys. Chem. B 2014, 118, 6431– 6437.


Page 11 of 11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment

11

Nonadditive Interactions Mediated by Water at Chemically

Recommend Documents