Table 1.
Data collection and processing.
Table 2.
Structure refinement.
Figure 1.
Alignment of amino acid sequences of Banana thaumatin-like protein (1Z3Q from Musa acuminata) and of the three Vitis vinifera thaumatin-like proteins F2/4JRU, I/4L5H, and H2/4MBT.
The multiple alignments were performed with the alignment function of the STRAP software [55]. The overall secondary structure for the fold predicted by STRAP software is shown.
Figure 2.
Three-dimensional structure of protein F2/4JRU.
A) Ribbon diagram of the overall three-dimensional structure of protein F2/4JRU. The protein consists of three structural domains: a central core domain I built from a β-sandwich of two sheets of six (front) and five (back) β-strands, flanked by two shorted domains II (where the 5 α-helixes and β-strand 11 and 12 are located) and III (where two β-strands - 5 and 6 - and a turn form two looping regions). The eight disulphide bridges are shown in yellow. B) Superposition of the backbone representations of the grape TLP F2/4JRU (in orange) and banana TLP 1Z3Q (in blue). TM-align data showed that the two proteins have RMSD of 0.78 Å, and a TM-score of 0.98 [41]. Significant differences in loops' structures are indicated by arrows, and numbers denote residues in F2/4JRU. C) and D) Mapping of the surface hydrophobicity of the grape TLP F2/4JRU, front view (same as Figure 2A) and back view (rotated approximately 180°) respectively. Hydrophobicity continuum from orange to blue representing hydrophobic to hydrophilic. E) Topology of protein F2/4JRU showing the residues included in each β-strand or α-helix.
Table 3.
Summary of the physical properties of F2/4JRU, I/4L5H and H2/4MBT.
Figure 3.
Comparison of the three-dimensional structures of grape TLP.
A) Mapping of the electrostatic potentials on the molecular surface of the grape TLP F2/4JRU. The negative potentials and positive potentials are colored red and blue, respectively. Neutral surfaces are white. Arrow indicates the acidic cleft region located in between domains I and II. B) Superposition of the backbone representations of chain A (in red) and B (blue) of I/4L5H. TM-align data showed that the two chains of I/4L5H have RMSD of 0.14 Å, and a TM-score of 0.99925. C) Superposition of the backbone representations of chain A (in red) and B (blue) of H2/4MBT. TM-align data showed that the two chains of H2/4MBT have RMSD of 0.13 Å, and a TM-score of 0.99939. D) Superposition of the secondary structure of I/4L5H chain A (in purple) and F2/4JRU (in blue). Arrows indicate the two loop regions showing the largest differences between the two structures. RMSD between 193 atom pairs is 0.505 Å (calculated with UCSF Chimera MatchMaker function). E) Detail of differences in loop regions in Domain I. F) Detail of differences in loop region in Domain II. Cysteine are shown in yellow, and disulfide bonds are formed between residues 140–213 and 183–178 in F2/4JRU, and residues 136–210 and 180–173 in I/4L5H, as indicated by the asterisks.
Figure 4.
Haze developed upon heat test of proteins F2/4JRU, I/4L5H and H2/4MBT.
Haze was measured after treatment at 80°C for 2 h, followed by 4°C for 2 h. Assays were carried out at 50 mg/L in model wine.
Figure 5.
SDS-PAGE of purified proteins under reducing and non-reducing conditions.
Proteins (∼2 µg per lane) were reduced with 5% BME or loaded on a 12% gel in non-reducing conditions and subjected to SDS-PAGE.
Figure 6.
Comparison of the hydrophobicity of proteins F2/4JRU and I/4L5H.
Hydrophobicity plot based on Kyte & Doolittle's scale for the alignment of proteins F2/4JRU (blue) and I/4L5H (red) obtained with AlignMe software http://www.bioinfo.mpg.de/AlignMe/AlignMe.html [56]. Window size used for analysis: 9 residues.