Guanidinium alters the water landscape

Protein stability studies that rely on the use of cosolutes to effect chemical denaturation typically use either urea or the guanidinium ion (Gdm⁺). Both of these chemicals unfold proteins through a process involving multiple, low-affinity binding events, but guanidinium has the larger effect. While this could be attributed simply to the strength of the interactions, it could also be a result of other chemical properties. In an upcoming paper in the Journal of Physical Chemistry, a team of researchers from the University of Pennsylvania suggest that a reorganization of hydrogen bonds in water may be partially responsible for its greater denaturing power.

The reason water is a liquid rather than a gas is the existence of significant networks of transient hydrogen bonds between water molecules. Typically, water forms these bonds proficiently at a wide range of angles, which means an energetic benefit without a high entropic cost. The range of bond angles in use can be tested by assessing their vibrational frequency using infrared (IR) spectroscopy. The core of this paper comes from an experiment Scott et al. perform in which they measure the dependence of these vibrational frequencies on the concentration of Gdm⁺ and the temperature of the solution.

They find that the presence of Gdm⁺ significantly alters the IR spectrum of water at high temperatures, shifting the overall peak to a lower wavenumber and causing the appearance of a shoulder around 3300 cm^-1, near the main peak of the ice spectrum (Figure 2). The effect of the Gdm⁺ ion appears to be less at lower temperatures; as a result the appearance of the water spectra changes less at high Gdm⁺ concentrations than in the absence of the solute. These spectral characteristics are not observed when another positive ion (potassium) is used instead of Gdm⁺.

The authors interpret these changes in the IR spectrum as an increase in the number of short, linear hydrogen bonds, and back this up with quantum chemical simulations. If this is correct, then it implies that Gdm⁺ rearranges the hydrogen bonding network of water, changing the structure to emphasize strong hydrogen bonds of a particular geometry. Moreover, this change in structure appears to be relatively resistant to changes in temperature. This is significant because the creation of protein tertiary structure is thought to be a mechanism by which systems avoid forming highly-structured networks of water hydrogen bonds. If Gdm⁺ induces significant water structuring, might that entropically favor the unfolding of proteins?

Because urea does not appear to induce this kind of change in water structure, it should be possible to test this experimentally. The effect of Gdm⁺ is clearly most pronounced at higher temperatures. So one could perform a comparison of the temperature dependence of Gdm⁺ and urea denaturation of a protein. Proteins tend to be more stable at lower temperatures, as a result we would expect the denaturant concentration at which half of the protein population is unfolded (D_1/2) to increase as the temperature decreases. If the structuring of water is significant, then the change in D_1/2 should be larger for Gdm⁺ than for urea, because Gdm⁺ is decreasing in effectiveness as the temperature goes down. The direct interaction may also be temperature-dependent, but it may be possible to control for this by measuring the energy released when guanidine interacts with an intrinsically unfolded protein.

Unfolding experiments typically are not interpreted in a way that that depends on the precise mechanism by which Gdm⁺ breaks down protein structure. Nonetheless, the specifics of this process are important if we want to use chemically denatured states as models of in vivo unfolded states. As we gain a better understanding of in vivo water structure, the characteristics of denaturant solvation may be an important consideration in experimental design.

J. Nathan Scott, Nathaniel V. Nucci, Jane M. Vanderkooi (2008). Changes in Water Structure Induced by the Guanidinium Cation and Implications for Protein Denaturation Journal of Physical Chemistry A DOI: 10.1021/jp8058239