Glycine insertion modulates the fluorescence properties of Aequorea victoria green fluorescent protein and its variants in their ambient environment

Abstract

The green fluorescent protein (GFP) derived from Pacific Ocean jellyfish is an essential tool in biology. GFP-solvent interactions can modulate the fluorescent property of GFP. We previously reported that glycine insertion is an effective mutation in the yellow variant of GFP, yellow fluorescent protein (YFP). Glycine insertion into one of the β-strands comprising the barrel structure distorts its structure, allowing water molecules to invade near the chromophore, enhancing hydrostatic pressure or solution hydrophobicity sensitivity. However, the underlying mechanism of how glycine insertion imparts environmental sensitivity to YFP has not been elucidated yet. To unveil the relationship between fluorescence and β-strand distortion, we investigated the effects of glycine insertion on the dependence of the optical properties of GFP variants named enhanced-GFP (eGFP) and its yellow (eYFP) and cyan (eCFP) variants with respect to pH, temperature, pressure, and hydrophobicity. Our results showed that the quantum yield decreased depending on the number of inserted glycines in all variants, and the dependence on pH, temperature, pressure, and hydrophobicity was altered, indicating the invasion of water molecules into the β-barrel. Peak shifts in the emission spectrum were observed in glycine-inserted eGFP, suggesting a change of the electric state in the excited chromophore. A comparative investigation of the spectral shift among variants under different conditions demonstrated that glycine insertion rearranged the hydrogen bond network between His148 and the chromophore. The present results provide important insights for further understanding the fluorescence mechanism in GFPs and suggest that glycine insertion could be a potent approach for investigating the relationship between water molecules and the intra-protein chromophore.

Keywords: Fluorescence spectroscopy; ethanol concentration-dependence; pH-dependence; pressure-dependence; temperature-dependence.

Figure 1

Optical properties of no-, one-, or three-glycine-inserted eGFP variants. (A) Schematic drawing of the glycine insertion method. The star indicates the glycine insertion site between Asn144 and Tyr145. (B–D) Absorbance (left) and emission (right) spectra of no-, one-, or three-glycine-inserted eYFP (B), eGFP (C), and eCFP (D). Insets in D display a section of the spectra on an expanded scale. Solid lines are spectra that are normalized by the maximum value of each spectrum. Broken lines are normalized by the maximum value of no-glycine variants (gray-curves). Traces represent the average of four trials. (E) Summary of spectral peak shifts of glycine-inserted variants compared to each no-glycine-inserted variant. (F) The ratio of the sum emission intensity divided by the absorbance intensity at the excitation wavelength versus peak shift in the emission spectra.

Figure 2

pH dependence of no-, one-, or three-glycine-inserted eGFP variants. (A, B, C) Absorbance (broken lines) and emission (solid lines) spectra of no- (left), one- (middle), or three (right)-glycine-inserted eYFP (A), eGFP (B), or eCFP (C) at pH 6 (gray), 7 (orange, light green, or cyan), and 8 (red, green, or blue). Insets in C show enlarged graphs. The intensities were normalized by that observed at pH 8. Traces represent the average of four trials. (D) pH dependence of the ratio of sum emission intensity and absorbance intensity at the excitation wavelength of the no- (0G, gray), one- (1G, orange, light green, or cyan) or three-glycine (3G, red, green, or blue) inserted variants. Values were normalized by the value of the no-glycine variant observed at pH 8. (E, F) The spectral peak of absorbance (E) and emission (F) of the no- (0G, gray), one- (1G, orange, light green, or cyan) or three-glycine (3G, red, green, or blue) inserted variants.

Figure 3

Temperature dependence of no-, one-, or three-glycine inserted eGFP variants. (A) Emission spectra of no- (left), one- (middle), or three (right)-glycine inserted eYFPs measured at temperatures ranging from 5°C (magenta) to 45°C (red). The fluorescence intensities were normalized with respect to the peak values observed at 25°C. Traces represent the average of four trials. (B) Temperature dependence of the sum emission intensity at 500–650 nm for eGFPs and eYFPs and 460–600 nm for eCFPs. The values represent ratios between the sum emission intensities at each temperature and at 25°C. Error bars represent standard deviation. (C) Summary of the effect of temperature on the sum emission intensity. The slope value was obtained by fitting the data in B to a linear equation by least squares. (D) Temperature dependence of the peak shift. The values represent the distance between spectral peaks at each temperature and at 25°C. Error bars represent standard deviation. (E) Summary of the effect of temperature on the spectral peak shift. The slope value was obtained by fitting the data in D to a linear equation by least squares.

Figure 4

Pressure dependence of no-, one-, or three-glycine inserted eGFP variants. (A) Emission spectra of eYFP (left), eGFP (middle), or eCFP (right) measured at pressures ranging from 0.1 MPa (magenta) to 50 MPa (red). The fluorescence intensities were normalized with respect to peak values at 0.1 MPa. Traces represent the average of four trials. (B) Pressure dependence of the sum emission intensity at 500–650 nm for eGFPs and eYFPs and 460–600 nm for eCFPs. The values represent ratios between the sum emission intensities at each pressure and at 0.1 MPa. Error bars represent standard deviation. (C) Summary of the effect of pressure on the sum emission intensity. The slope value was obtained by fitting the data in B to a linear equation by least squares. (D) Pressure dependence of the peak shift. The values represent the distance between spectral peaks at each pressure and at 0.1 MPa. Error bars represent standard deviation. (E) Summary of the effect of pressure on the spectral peak shift. The slope value was obtained by fitting the data in D to a linear equation by least squares.

Figure 5

Ethanol concentration dependence of no-, one-, or three-glycines inserted eGFP variants. (A) Emission spectra of no- (left), one- (middle), or three (right)-glycine inserted eYFPs measured at ethanol concentrations ranging from 0% (magenta) to 30% (red). The fluorescence intensities were normalized with respect to the peak values at 0%. Traces represent the average of four trials. (B) Ethanol concentration dependence of the sum emission intensity at 500–650 nm for eGFPs and eYFPs and 460–600 nm for eCFPs. The values represent ratios between the sum emission intensities at each ethanol concentration and at 0%. Error bars represent standard deviation. (C) Summary of the effect of dehydration induced by ethanol on the sum emission intensity. The slope value was obtained by fitting the data in B to a linear equation by least squares. (D) Ethanol concentration dependence of the peak shift. The values represent the distance between spectral peaks at each ethanol concentration and at 0%. Error bars represent standard deviation. (E) Summary of the effect of dehydration induced by ethanol on the spectral peak shift. The slope value was obtained by fitting the data in D to a linear equation by least squares.