Figure 1.
Structure comparison of wt bradavidin, wt streptavidin [PDB: 2BC3] and chicken avidin [PDB: 1AVD].
(A) Cartoon models of tetrameric proteins. Subunits are shown in different colors as follows: I (blue), II (cyan), III (magenta) and IV (yellow). The key biotin-binding pocket occupying residues (wt bradavidin and wt streptavidin) and biotin (chicken avidin) are shown as spheres. The N- and C-termini are indicated by letters. (B) Superimposition of the Cα traces of subunits I of wt bradavidin (blue), wt streptavidin (orange) and chicken avidin (cyan). The Cα trace for C-terminal residues starting from Lys127 of subunit III of wt bradavidin is shown, too. The ligand-binding site occupying residues Glu131-Leu133 of wt bradavidin and Gly151-Pro153 of wt streptavidin, and the biotin ligand of avidin, are shown as sticks. The left and right arrows pinpoint the N- and C-terminal sites, respectively, were major differences are seen between the proteins. Trp5 (left arrow) of wt bradavidin is shown as sticks. The N- and C-termini are indicated and loops are numbered. (C) Monomeric cartoon models. The N-terminus and C-terminus of each protein are indicated in red and blue, respectively, the colouring starting from equal positions in all proteins. The key residues occupying the biotin-binding site are shown as sticks. (D) Loop design. Colouring of subunits are as in (A) and representation of the key residues as in (C). The arrow pinpoints the varying beginnings of the L7,8-loops in the three structures. The L3,4-loop of the wt streptavidin structure is not fully visible (dashed line).
Table 1.
X-ray structure determination statistics for wt bradavidin [PDB: 2Y32].
Figure 2.
Sequence alignment of wt bradavidin, rhodavidin, streptavidin and chicken avidin.
The UniProt [70] accession numbers are shown after the names of the sequences. The Brad-tag sequence is highlighted with orange background, cysteine residues with yellow and the tryptophan residue important for the N-terminus of wt bradavidin with cyan. The secondary structures of wt bradavidin (from the structure reported here) and chicken avidin (from [PDB: 1AVD]) are shown. The truncation site for core-bradavidin is between Leu118 and Leu119 and is shown by a short, vertical dashed line. The conserved residues are indicated by the default colouring scheme of the ESPript program. TT, β-turn; TTT, α-turn; and η1, 3/10-helix. The structural alignment was created in Bodil [66] and the picture using ESPript [68].
Figure 3.
Four tyrosine residues at the intersection of all four subunit interfaces in bradavidin.
A stereo view. The tyrosine residues are shown as stick models with different colouring for the different subunits (subunit I, blue; II, cyan; III, magenta; and IV, yellow). Structural water molecules are shown as red spheres. Electron density map (a weighted 2FO-FC map; sigma level 1) around the water molecules and the side chain oxygen atoms of Tyr90 is shown in blue. Putative hydrogen bonds are indicated with grey dashes.
Figure 4.
Stick model (stereo view) of the ligand-binding site of wt bradavidin.
(A) The carbon atoms of the Brad-tag sequence (subunit I) are shown in blue and the residues around the Brad-tag sequence in magenta (subunit III) and yellow (subunit IV). Structural water molecules near the Brad-tag sequence are shown as red spheres. (B) Superimpositioning of the ligand-binding site of wt bradavidin and chicken avidin [PDB: 1AVD]. Colouring for wt bradavidin as in (A); the carbon atoms of residues of chicken avidin are shown in grey. The residues Asn12, Leu14, Ser16, Tyr33, Trp70, Ser73, Ser75, Thr77, Phe79, Trp97, Leu99 and Asn118 of chicken avidin were superimposed to the equivalent residues Asn9, Tyr11, Ser13, Tyr31, Phe66, Cys69, Ser71, Thr73, Trp75, Trp89, Leu91 and Asp107 of wt bradavidin. For clarity, the Brad-tag sequence is not shown. For chicken avidin, the residue numbers are shown in brackets. BTN, biotin; *, Tyr31 (Tyr33); **, Asn33 (Thr35).
Table 2.
Summary of different avidin-binding peptide tags.
Figure 5.
ITC analysis of ligand binding.
Thermograms of measurements performed at three different temperatures (A) 15°C, (B) 25°C and (C) 40°C are shown. At each temperature, core-bradavidin was first titrated with Brad-tag (1), followed by competitive titration with biotin (2). As a control measurement, core-bradavidin was titrated with biotin only (3). In order to prove that the intrinsic Brad-tag decreases the affinity towards biotin, wt bradavidin was also titrated with biotin (4). In addition, core-bradavidin was titrated with Brad-tag–EGFP at 15 and 25°C (5). (D) Comparison of the binding enthalpies of all measurements at different temperatures. Brad-tag had a clear effect on the binding enthalpy of the competitive titration with biotin at 15°C (endothermic Brad-tag binding) and 40°C (exothermic Brad-tag binding). At 25°C, the enthalpy of competitive titration was equal to that of titration with biotin only (no detectable binding of Brad-tag to core-bradavidin).
Table 3.
Thermodynamic parameters of ligand binding analyzed by ITC.
Figure 6.
Brad-tag–EGFP fusion proteins.
(A) Four different Brad-tag–EGFP fusion protein constructs were used in the current study. Brad-tag was positioned at the N- or C-terminus of the fusion proteins. A His-tag was also included in two of the constructs directly after the sequence of EGFP. (B) Immunoblot analysis using antibody against GFP was used to evaluate the quality and amount of tagged EGFPs. Biotinylated–EGFP was used as a positive control (Vikholm-Lundin et al, unpublished) and core-bradavidin as a negative control. Numbers in brackets indicate different protein productions. Molecular weight markers (M, kDa) are indicated on the left. (C) The fluorescence spectra measured for purified Brad-tag–EGFP–His-tag (12.4 ng/µl) and clarified cellular lysates of other Brad-tag–EGFP constructs.
Figure 7.
Purification of EGFP fusion protein using N-terminal Brad-tag.
(A) Photograph of Brad-tag–EGFP–His-tag bound to core-bradavidin resin under UV-light. First, core-bradavidin was coupled by amine groups to the terminal NHS carboxylates of the linkers (resin–NH–(CH2)5–COONHS). Then, Brad-tag–EGFP–His-tag (prepurified with Ni-NTA column) was incubated with the functionalized resin and the resin was pelleted. In the absence (–) of biotin Brad-tag–EGFP–His-tag concentrates on the resin pellet. The presence (+) of free biotin inhibits the binding. Stoichiometry and the size of compounds in the schematic figure are only speculative. (B) SDS-PAGE analysis of the protein purification experiment for cellular lysates of N-Brad-tag–EGFP-C. Cleared cellular lysate (total) was incubated with core-bradavidin resin. Then the resin was washed with buffer (50 mM Tris-HCl, 100 mM NaCl, pH 7.5) and samples 1 to 5 were eluted. Molecular weight markers (M, kDa) are indicated on the left. (C) The fluorescence spectra measured for cleared cellular lysate (total) and eluted samples 1 to 5 from the protein purification experiment for Brad-tag–EGFP.
Figure 8.
Specificity of core-bradavidin binding to Brad-tag analyzed by biolayer interferometry.
(A) Anti-Penta-HIS biosensors were functionalized with Brad-tag–EGFP–His-tag fusion protein (step 1, arrow in the graph). After a brief wash (10 s) in measurement buffer, biosensors were incubated with a series of different avidin proteins at concentration of 0.06 mg/ml (step 2). Sample without any avidin protein was used as a negative control (buffer). Finally, biosensors were exposed to the measurement buffer leading to dissociation of the bound core-bradavidin proteins (step 3). (B) The measured raw data for buffer is subtracted from the raw data of different proteins.