Ion Specificity at the Polypeptide Backbone: molecular Mechanisms of the Hofmeister Effect

Open Access
Rembert, Kelvin B
Graduate Program:
Doctor of Philosophy
Document Type:
Date of Defense:
October 09, 2014
Committee Members:
  • Paul S Cremer, Dissertation Advisor
  • Scott A Showalter, Committee Chair
  • William George Noid, Committee Member
  • Craig Eugene Cameron, Special Member
  • Hofmeister Series
  • Ions
  • Polymers
  • Water
  • NMR
  • LCST
  • FTIR
The behavior of biomolecules in aqueous solutions is ion specific and follows the Hofmeister series. Discovered in 1888, the Hofmeister series is a recurring trend that originally ranked cations and anions by their ability to precipitate proteins from aqueous salt solutions. Over the course of the 21st century however, the Hofmeister effect was found to extend well beyond protein precipitation studies and occurs in myriad bio-chemical and physico-chemical systems. Today, the field remains an active area of research and has gained the attention of many researchers due to the elusiveness in developing a theoretical framework for predicting ion specific effects in water. Only recently has a unified molecular picture began to emerge. Moreover, utilizing the proper techniques and model systems is key. In this dissertation, we have employed model protein backbones in a set of salt solution studies using NMR, FTIR and thermodynamic phase transition measurements to elucidate both the molecular level and macroscopic details of the Hofmeister effect. First, we investigate the interactions of Hofmeister anions with a thermoresponsive polymer, poly(N,N-diethylacrylamide) (PDEA). This amide based polymer lacks an NH moiety in its chemical structure and serves as a model to directly test if anions can bind to amide moieties in the absence of an NH site. The lower critical solution temperature (LCST) of PDEA was measured as a function of concentration for eleven sodium salts in aqueous solutions and followed a direct Hofmeister series for the ability of anions to precipitate the polymer. More strongly hydrated anions (CO32-, SO42-, S2O32, H2PO4-, F-, Cl-) linearly decreased the LCST of the polymer with increasing salt concentration. Weakly hydrated anions (SCN-, ClO4-, I-, NO3-, Br-) increased the LCST at lower salt concentrations and salted the polymer into solution, but salted the polymer out at higher salt concentrations. Proton NMR was used to probe the mechanism of the salting-in effect and showed apparent binding between weakly hydrated anions (SCN- and I-) and the α-protons of the polymer backbone. Additional experiments performed by FTIR found little change in the amide I band, which is consistent with limited, if any, interactions between the salt ions and the carbonyl moiety of the amide. These results support a molecular mechanism for ion-specific effects on proteins and model amides that does not specifically require an NH group to interact with the anions for salting-in to occur. Next, the binding sites of a 600 residue protein polymer, (VPGVG)120, were elucidated using a combination of NMR and LCST measurements. It was found that the salting-in effect of large soft anions such as SCN− and I− is due to direct interactions at glycine residues within the polypeptide backbone via a hybrid binding site that consists of the amide dipole and the adjacent α-carbon. The hydrocarbon groups at these sites bear a slight positive charge, which enhances anion binding without disrupting specific hydrogen bonds to water molecules. The hydrophobic side chains do not contribute significantly to anion binding or the corresponding salting-in behavior of the biopolymer. Cl− binds far more weakly to the amide dipole/α-carbon binding site, while SO42− is repelled from both the backbone and hydrophobic side chains of the polypeptide. The Na+ counterions are also repelled from the polypeptide. The identification of these molecular-level binding sites provides new insight into the mechanisms of peptide−ion interactions. Moreover, the location of the amide, whether in a protein backbone or on a polymer side chain, has no appreciable influence on the ability of weakly hydrated anions to bind. To further explore the glycine residue binding site interaction with anions, we employed triglycine (GGG) along with two variants of the peptide as model systems. These small peptides allowed for the investigation of the influence of size on ion binding as well as the effect of electrostatic interactions of ions with charged termini. We demonstrated by means of NMR and ATR-FTIR spectroscopy that the Hofmeister series for anions changes from a direct to a reversed series upon uncapping the N-terminus of the peptide. Weakly hydrated anions, such as I- and SCN-, interact with the amide backbone, while strongly hydrated anions like SO42- are repelled from it. In contrast, a reversed order of ion interactions was observed at the positively charged, uncapped N-terminus. This by analogy should also be the case at side chains of positively charged amino acids such as lysine and arginine. Overall, the small GGG peptides displayed over an order of magnitude weaker binding compared to the large ELPs mentioned above. These results demonstrate that the specific chemical and physical properties of peptides and proteins play a fundamental role in ion specific effects and provides a route for exploiting these interactions by tuning protein size and the number of basic amino acid residues on the surface.