Structural Properties and Stability of Proteins in Dihydrolevoglucosenone/Water Mixtures

Abstract

Dihydrolevoglucosenone (DHL) shows great promise as an alternative to conventional toxic organic solvents widely used for industrial purposes. In this framework, evaluating the potential of DHL (commercially known as Cyrene) as a solvent for dissolving proteins is of great importance. Here, the effect of DHL/water mixtures on protein stability and solubility has been assessed. Several proteins, namely, hemoglobin, ferritin, ribonuclease, and albumin, were readily dissolved in buffer solutions containing up to 50-60% DHL and were stable at room temperature, as indicated by gel electrophoresis and matrix-assisted laser desorption/ionization analysis. Turbidimetry assays were performed in order to assess the solubility limitations derived from DHL/water mixtures. Finally, protein secondary structures in such mixtures, investigated by attenuated total reflectance Fourier-transform infrared spectroscopy, were found to be comparable to those obtained in phosphate buffer up to 50% DHL/water, with small spectral changes in the case of ribonuclease. DHL/water mixtures may thus represent highly convenient solvents for studies of protein chemistry.

1

Time-course absorbance measurements of protein precipitation in DHL/buffer solutions. Solution mixtures of (a) BSA, (b) HFt, and (c) RNase proteins in phosphate buffer containing DHL concentrations ranging from 40% to 70% were recorded at a fixed wavelength (600 nm) for 1200 s (20 min) and T = 25 °C. Protein concentrations for each measurement were 25 mg/mL (BSA), 2.5 mg/mL (HFt), and 2 mg/mL (RNase).

2

Effect of curcumin, dissolved in pure EtOH (panel a) and in 50% EtOH/DHL (panel b), on the fluorescence emission signals of chromophores in BSA solution (1 mg/mL in pure water). Panel c represents the Stern–Volmer plots for the fluorescence quenching of BSA by curcumin in the two different conditions.

3

ATR-FTIR spectra of DHL/sodium phosphate buffer mixtures in different % of DHL.

4

ATR-FTIR spectra of (a) BSA, (b) HFt, (c) Hb, and (d) RNase in buffer alone (blue line) and 50% DHL/buffer solvent (red line). The green rhomboid symbols correspond to the frequencies 1655 and 1545 cm^–1, assigned to α-helices in the amide I and amide II spectral range, respectively. Spectra are normalized at 1655 cm^–1. The DHL spectral contributions in the 1710–1760 cm^–1 range are due to imperfect solvent subtraction.

5

Second derivatives of ATR-FTIR spectra for (a) BSA, (b) HFt, (c) Hb, and (d) RNase in buffer (blue line) and 50% DHL/buffer (red line) solvent. The green rhomboid symbols correspond to the frequencies 1655 and 1545 cm^–1, assigned to α-helices in the amide I and amide II spectral ranges, respectively.

6

Gaussian fit of IR spectra in the amide I frequency range of proteins in different conditions: (a) BSA in 50% DHL/buffer mixture, (b) BSA in buffer, (c) HFt in 50% DHL/buffer mixture, (d) HFt in buffer, (e) Hb in 50% DHL/buffer mixture, (f) Hb in buffer, (g) RNase in 50% DHL/buffer mixture, and (h) RNase in buffer. Single Gaussian contributions of the amide I signal are in yellow (turn), purple (α-helix), pink (disordered), green and brown (β-sheet). The Gaussian curve in cyan centered at 1580 cm^–1, accounts for the spectral contributions from amide II. The combination of all Gaussian curves (fitting curves represented by red lines) interpolates the experimental data (represented by blue dots).