Ion-dependent protein-surface interactions from intrinsic solvent response

Abstract

The phyllosilicate mineral muscovite mica is widely used as a surface template for the patterning of macromolecules, yet a molecular understanding of its surface chemistry under varying solution conditions, required to predict and control the self-assembly of adsorbed species, is lacking. We utilize all-atom molecular dynamics simulations in conjunction with an electrostatic analysis based in local molecular field theory that affords a clean separation of long-range and short-range electrostatics. Using water polarization response as a measure of the electric fields that arise from patterned, surface-bound ions that direct the adsorption of charged macromolecules, we apply a Landau theory of forces induced by asymmetrically polarized surfaces to compute protein-surface interactions for two muscovite-binding proteins (DHR10-mica6 and ^C98RhuA). Comparison of the pressure between surface and protein in high-concentration KCl and NaCl aqueous solutions reveals ion-specific differences in far-field protein-surface interactions, neatly capturing the ability of ions to modulate the surface charge of muscovite that in turn selectively attracts one binding face of each protein over all others.

Keywords: Landau theory; electrostatics; soft matter; solution assembly; specific ion effects.

Fig. 1.

(A and B) AFM images of DHR10-mica18 assembly. (A) The 0.1-$\mathrm{\mu }$M DHR10-mica18 in 3 M KCl. (B) The 0.1-$\mathrm{\mu }$M DHR10-mica18 in 3 M NaCl. Insets are fast Fourier transforms. The buffer is 20 mM Tris, pH 7. (C) Schematic representation of our key finding that macroscopic water polarization over muscovite is reversed in the presence of KCl vs. NaCl, which leads to either constructive or destructive interference with the polarization of a protein’s surface, depending on specific ion effects. For DHR10-mica6 (shown) the relative attraction strengths of the protein’s front or back faces to muscovite correlate with reduced orientational specificity observed in AFM (B). (D) Illustration of decomposing the total electric field $\mathbf{E}$ into displacement field $\mathbf{D}$, consisting of background charges and the two surfaces (treated as fixed), and the electric polarization $\mathbf{P}$ due to water orientations, which is our focus here. This relation is written in SI units, where ${\epsilon }_0$ is vacuum permittivity. Simulations of protein and muscovite are conducted separately.

Fig. 2.

For 3 M KCl aqueous conditions, we have (A and B) averaged probability density from atomic coordinates as a function of perpendicular distance from muscovite bedrock (ZDF), normalized to bulk, for (A) muscovite including explicit Al substitutions and (B) muscovite in the mean-field substitution representation. Both are using the INTERFACE force field. Percentages are fraction of cation binding sites occupied, measured by integrating up to the first minimum in the potassium density. (C and D) Average probability density of lateral water oxygen positions a distance of 3.0 $\pm$ 0.3 Åfrom the muscovite surface as measured from the surface-exposed oxygen atoms. Patterning is uneven under explicit Al substitution (C) and organized under the mean-field treatment (D).

Fig. 3.

Lateral probability distribution of water molecules occupying the narrow region of 3.5 $\pm$ 0.1 Å (Top row, 3 M KCl with IFF mean-field muscovite) or 2.1 $\pm$ 0.1 Å (Bottom row, 3 M NaCl with IFF mean-field muscovite). This same probability density is smoothed by a Gaussian of width $\sigma$ (a KDE of probability), in columns Left to Right, from $\sigma$ = 0.0 Å (no smoothing) to $\sigma$ = 4.5 Å in increments of 0.5 Å. Characteristic streaking can be seen from the lower left to upper right corners of each panel, up to a smoothing length of about 2.0 $\sigma$. This illustrates that some distinctive atomistic signatures are inaccessible to far-field smoothing procedures that use Gaussians of width $\sigma =4.5$ Å.

Fig. 4.

Demonstration of converting ${\rho }_b^{\sigma }\left(z\right)$ obtained from LMF-processed MD trajectories into polarization-induced pressure between muscovite and the front/back faces of DHR10-mica6. (A) A concatenated bound charge density profile is made by adjoining partial ${\rho }_b^{\sigma }\left(z\right)$ profiles from trajectories of protein or muscovite in isolation at the bulk, where solution conditions are identical. (B) We compute $P_{\widehat{z}}^{\sigma }\left(z\right)$ from ${\rho }_b^{\sigma }\left(z\right)$ via Eqs. 3 and 4 as in the main text and then fit the functional form of order parameter $\eta \left(z\right)$ (Eq. 6) to polarization on the region [−30 Å, +30 Å] that is solvent accessible (fits are dashed lines). (C) Fitting $\eta \left(z\right)$ to $P_{\widehat{z}}^{\sigma }\left(z\right)$ determines the phenomenological parameters needed to evaluate pressure $\mathrm{\Pi }$ (Eq. 9) as a function of separation $D$, where a positive $\mathrm{\Pi }$ indicates repulsion. In A–C, red coloration indicates 3 M KCl solution, while blue indicates 3 M NaCl.

Fig. 5.

Selected far-field polarization-induced pressure curves demonstrating impacts of ${\mathrm{K}}^{+}$/Na${}^{+}$ exchange. The eight traces presented are to be thought of as four pairs, each representing a given muscovite–macromolecule pair under both 3 M KCl and 3 M NaCl conditions. These macromolecule pairings are muscovite–muscovite (as reference, in black), muscovite–${\mathrm{D}\mathrm{H}\mathrm{R}}_{\text{binding}}$ (red), muscovite–${\mathrm{D}\mathrm{H}\mathrm{R}}_{\text{back}}$ (blue), and muscovite–${}^{\mathrm{C}98}$Rhu${\mathrm{A}}_{\text{binding}}$ (green). Pairs of traces are indicated by a shared color (dark vs. light shade for KCl vs. NaCl, respectively) and corresponding symbols, (, ), (, ), (, ), (, ), listed in KCl/NaCl order. For example, the red pair of traces (, ) corresponds to those in Fig. 4C. With $\mathrm{\Pi }>0$ indicating repulsion we see muscovite–muscovite is repulsive under both KCl and NaCl, muscovite–${}^{\mathrm{C}98}$Rhu${\mathrm{A}}_{\text{binding}}$ is attractive in both cases, and the muscovite–DHR interaction can be modulated to attractive or repulsive depending on which protein face is presented and the salt treatment. See SI Appendix, Fig. S2 for renderings of the two proteins to illustrate the notional front and back faces.