Barstar

The dependence of the CD of unfolded barstar on temperature and urea concentration was reported by Nolting et al. (1997). In concentrated urea, the CD at 222 nm shows a linear temperature dependence,... 

Fig. 29. Equilbrium unfolding of C40A/C82A/P27A (pseudo-wild-type) barstar monitored by A R, mean residue circular dichroism. Conditions for near-UV CD were 50 /xM protein in 50 mM Tris-HCl buffer, pH 8, 0.1 M KC1, path length 1 cm. (A) Urea-induced unfolding at 25°C at urea concentrations as indicated. (B) Cold-induced unfolding in...

Flq ure 4.12 Rate constants for the association of bamase and barstar at different ionic strengths. The ionic strength I is varied by changing the concentration of NaCl. 

Active sites of enzymes and binding sites of proteins are a general source of instability because they contain groups that are exposed to solvent in order to bind substrates and ligands and so are not paired with their normal types of partners. Stability-activity trade-off is also seen with residues in the natural polypeptide inhibitor of barnase, barstar, that has evolved to bind as rapidly as possible to barnase,70 and also in the active site of T4 lysozyme.71... 

Consolidation of 2° and 3° more concerted for barstar than barnase... 

The contribution of electrostatic interactions to fast association was analyzed by applying the classical Debye-Hiickel theory of electrostatic interactions between ions to mutants of bamase and barstar whose ionic side chains had been altered by protein engineering (Chapter 14).16 The association fits a two-step model that is probably general (equation 4.84). 

The data on bamase and barstar may be analyzed by simple17 or computer-simulation methods.1819... 

The early stages of folding of barstar have been measured on the microsecond time scale by temperature jumping its cold-denatured state from 2 to 10°C.65,66 There is the fast formation of a folding intermediate (tm 200 fxs) with the peptidy 1-proline 48 bond trans, followed by the formation with ty2 60 ms of a second intermediate that is highly native-like because it binds to and inhibits barnase. The native-like intermediate then undergoes trans cis peptidyl-proline isomerization on the time scale of minutes to give the final native structure (equation 19.2). 

Figure 2 Rearrangement of interfacial side chains in bamase upon binding barstar. Dark (light) color represents the bound (unbound) conformation. Arg-59 is highlighted in ball-and-stick representation, using a lighter color for the free and a darker for the bound conformation. The viewpoint is chosen from bamase across its guanine-binding loop toward barstar.

Figure 7 The 20 highest ranking putative complexes from the best 100 FTDOCK candidates obtained after fine-grained rigid-body docking of the unbound barnase and barstar structures. Barnase is shown in the middle by dashed lines, surrounded by putative clusters of barstar orientations. The dense cluster to the right contains the correct docking orientation (shown in bold).

In our simulations, we performed association studies to elucidate their role by selecting pairs of those hot spot residues that established a barnase-barstar contact in the bound complex. In these simulations, we studied what effect the side-chain conformation had on contact formation for certain hot spot residue pairs. A typical result is shown in Figure 8. Starting from the correct rotameric state for certain hot spot residue pairs with the rest of the proteins in their unbound conformations leads to improved contact formation in the initial docking phase. This effect is not observed when performing equivalent simulations with non-hot-spot interfacial residue pairs. 

Figure 8 The impact of hot-spot residue pair conformations on contact formation during docking of barnase (bn) and barstar (bs). Contact formation (as measured by the percentage of native contacts) during the first 100 ps of torsion dynamics simulations is shown see text for details.

The results of simulating barnase-barstar association after 100 ps of simple multiple copy MD without ensemble enriching are shown in Figure 9. Whereas contact formation in the simple multiple copy MD simulation has leveled off after the initial 100 ps, ensemble enriching with a linear scoring function is able to shift the ensemble population toward a higher number of established native contacts. Compared to the side-chain torsion dynamics... 

Figure 9 Formation of native contacts during a multiple copy MD simulation with and without ensemble enriching after an initial 100 ps multiple copy MD simulation see text for details. Ensembles that each consist of 10 barnase and 10 barstar copies are simulated, giving rise to 100 putative complexes. The thick lines show the mean and standard deviations, along with the best/worst values for the ensemble enriching method. The superposed thin lines show the corresponding data for the plain multicopy MD simulations without ensemble enriching.

---