Osmolyte-induced perturbations of hydrogen bonding between hydration layer waters: correlation with protein conformational changes

Abstract

Gadolinium vibronic sideband luminescence spectroscopy (GVSBLS) is used to probe osmolyte-induced changes in the hydrogen bond strength between first and second shell waters on the surface of free Gd(3+) and Gd(3+) coordinated to EDTA and to structured calcium binding peptides in solution. In parallel, Raman is used to probe the corresponding impact of the same set of osmolytes on hydrogen bonding among waters in the bulk phase. Increasing concentration of added urea is observed to progressively weaken the hydrogen bonding within the hydration layer but has minimal observed impact on bulk water. In contrast, polyols are observed to enhance hydrogen bonding in both the hydration layer and the bulk with the amplitude being polyol dependent with trehalose being more effective than sucrose, glucose, or glycerol. The observed patterns indicate that the size and properties of the osmolyte as well as the local architecture of the specific surface site of hydration impact preferential exclusion effects and local hydrogen bond strength. Correlation of the vibronic spectra with CD measurements on the peptides as a function of added osmolytes shows an increase in secondary structure with added polyols and that the progressive weakening of the hydrogen bonding upon addition of urea first increases water occupancy within the peptide and only subsequently does the peptide unfold. The results support models in which the initial steps in the unfolding process involve osmolyte-induced enhancement of water occupancy within the interior of the protein.

Figure 1. Water Raman of Gd³⁺ at different coordination as a function of urea

a) Water Raman for aqueous samples of 0.5 M GdCl₃ in the absence of urea ( ) and presence of 2.0 M urea ( ) and 8.0 M MgCl₂ ( ); b) water Raman for aqueous samples of 0.5 M GdCl₃ in the absence of urea ( ), 1 mM apo-mSE3 in 10mM HEPES buffer at pH 7.0 with 1 mM GdCl₃ in the absence of urea ( ) and presence of 8.0 M urea ( ); c) water Raman for aqueous samples of 0.5 M GdCl₃ in the absence of urea ( ), 100 mM EDTA in 10mM HEPES buffer at pH 7.0 with 80 mM GdCl₃ in the absence of urea ( ) and presence of 8.0 M urea ( ).

Figure 2. Water Raman spectra of osmolytes solutions

a) Water Raman spectra for ( ) water and trehalose solution with concentration of ( ) 0.25 M, ( ) 0.5 M and ( ) 1.0 M; b) Difference water Raman spectra for ( ) only water and glycerol solution with concentration (v/v) of ( ) 10%, ( ) 20%, ( ) 30%, ( ) 40%, ( ) 50%, ( ) 60%, ( ) 70%, ( ) 80% and ( ) 90%; c) Difference water Raman spectra for ( ) only water and PEG400 solution with concentration (v/v) of ( ) 10%, ( ) 20%, ( ) 30%, ( ) 40%, ( ) 50%, ( ) 60%, ( ) 70%, ( ) 80%.

Figure 3. Water Raman OH stretching frequencies of PEG400, trehalose and glycerol aqueous solution

a) Bar graph of OH stretch vibration frequency from water Raman of free 0.5 M GdCl₃ aqueous samples as a function of PEG400 concentration (v/v); b) Bar graph of OH stretch vibration frequency from water Raman of 0.5 M GdCl₃ aqueous sample with trehalose; c) Bar graph of OH stretch vibration frequency from water Raman of 0.5 M GdCl₃ aqueous sample with glycerol.

Figure 4. Urea induced VSB shifts of free Gd³⁺ or EDTA-Gd³⁺

a) Normalized OH vibronic side band derived from GVSBLS of 0.5 M free GdCl₃ solution in the absence of urea and presence of 2.0 M urea and 8.0 M urea. b) Normalized OH vibronic side band derived from GVSBLS of 100 mM EDTA with 80 mM Gd³⁺ in 10 mM HEPES buffer at pH 7.0 in the absence and presence of 1 M urea; c) Bar graph of −OH stretching mode vibration frequency shifts induced by additions of varied concentrations of urea derived from: water Raman of free 0.5 M Gd³⁺ solution, GVSBLS of free 0.5 M Gd³⁺ solution or GVSBLS of 100 mM EDTA with 80 mM Gd³⁺ in 10 mM HEPES buffer at pH 7.0

Figure 5. Glycerol induced VSB shifts of Gd³⁺ at different coordination

Bar graph of −OH stretching mode vibration frequency shifts induced by addition of varied concentrations of glycerol derived from: water Raman of 0.5 M GdCl₃ solution, GVSBLS of 0.5 M GdCl₃ solution or GVSBLS of 100 mM EDTA with 80 mM Gd³⁺ in 10 mM HEPES buffer at pH 7.0 and GVSBLS of 1 mM apo-mSE3 in 10mM HEPES buffer at pH 7.0 with 1 mM GdCl₃.

Figure 6. OH VSB shifts of Gd³⁺ at different coordination as a function of tagatose, glucose, sucrose and trehalose

OH stretch mode frequency shifts of 0.5 M free GdCl₃ induced by addition of varied concentrations of tagatose, glucose, sucrose and trehalose. Also depicts the OH stretch mode frequency shifts of 100 mM EDTA with 80 mM Gd³⁺ in 10 mM HEPES buffer at pH 7.0 induced by addition of varied concentrations of tagatose, glucose, sucrose and trehalose. Also compares the OH stretch mode frequency shifts of 1 mM apo-mSE3 in 10mM HEPES buffer at pH 7.0 with 1 mM GdCl₃ induced by addition of varied concentrations of tagatose and trehalose. Inset: enlarged area of the drawn box.

Figure 7. The synergy of urea and trehalose on VSB of free Gd³⁺ and EDTA-Gd³⁺

a) Normalized OH vibronic side band derived from GVSBLS of 0.5 M free GdCl₃ solution in the absence of osmolytes and presence of 2.0 M urea, 1.0 M trehalose and 2.0 M urea with 1.0 M trehalose. b) Normalized OH vibronic side band derived from GVSBLS of 100 mM EDTA with 80 mM Gd³⁺ in 10 mM HEPES buffer at pH 7.0 in the absence of osmolytes and presence of 1.0 M urea, 1.0 M trehalose and 1.0 M urea with 1.0 M trehalose.

Figure 8. Far-UV CD spectra of holo-mSE3 in the condition of urea, glycerol and trehalose

Far-UV CD spectra holo-mSE3 in 10mM HEPES buffer at pH 7.0 in the condition of: alone; 5.0 M urea; 5.5 M urea; 6.0 M urea; 7.5 M urea; 8.0 M urea; 60% (v/v) glycerol and 1.0 M trehalose.

Figure 9. Mean residue weighted molar extinction coefficient at 222 nm of far UV CD of the peptide mSE3 bound with Gd³⁺ as a function of concentration of the peptide

with ● no osmolytes, 50% glycerol, 0.5 M trehalose and 1.0 M tagatose.

Figure 10. The glycerol, trehalose and tagatose induced OH stretching frequency shifts of the first hydration layer waters coordinated to mSE3 bound with Gd³⁺ as a function of percentage of

α-helix contents. The ● glycerol, trehalose and tagatose induced OH stretching frequency shifts of the first hydration layer waters coordinated to mSE3 bound with Gd³⁺ as a function of percentage of α-helix contents calculated as described in method section.

Figure 11. The correlation between Far-UV CD at 220 nm monitored unfolding of mSE3 and SE2 peptides with the vibrational stretching frequencies probed by GVSBLS as a function of urea concentration

Urea denaturation of (○) mSE3 and (□) SE2 peptide. Unfolding was monitored at room temperature using mean residue weighted molar extinction coefficient of far-UV CD at 220 nm. The buffer is 10mM HEPES buffer at pH 7.0. Also depicted in the figure are the vibrational stretching frequencies probed by GVSBLS as a function of urea concentration for ( ) mSE3 and ( ) SE2 peptide, ( ) free GdCl₃ and ( ) EDTA-Gd³⁺ solution. The OH stretching frequency of GdCl3 powder heated for 5 hours is also marked ( ).

Figure 12. OH stretching frequency shifts from VSB of free GdCl₃ solution and EDTA-Gd³⁺ by osmolytes of sucrose, glucose and trehalose on a basis of per molarity as a function of osmolytes limiting diffusion coefficients

OH stretching frequency shifts from VSB of 0.5 M free GdCl₃ solution (filled circles) or 100 mM EDTA with 80 mM Gd³⁺ in 10 mM HEPES buffer at pH 7.0 (unfilled circles) by osmolytes of sucrose, glucose and trehalose on a basis of per molarity as a function of osmolytes limiting diffusion coefficients.