Figure 2.
Comparison of human, yeast and E. coli Hsp90 proteins. (A) Domain structure of human Hsp90β (hHsp90), S. cerevisiae Hsp82 (yHsp82) and E. coli HtpG (EcHtpG) proteins. NBD, nucleotide binding domain; CL, charged linker; MD, middle domain; DD, dimerization domain. (B) HX kinetics of full-length hHsp90. The mass of hHsp90 and the percentage hydrogen exchange, respectively, is plotted vs. the incubation time in D2O. The data points are averages of up to four independent experiments and the error bars represent the standard error of mean. The solid line is a fit of a tri-exponential equation [Y = Y1·(1-exp(−k1·t)) + Y2·(1−exp(−k2·t)) + Y3·(1−exp(−k3·t))] to the data points. The dashed line represents the theoretical exchange curve including the back-exchange during the desalting process calculated for the completely unfolded Hsp90 protein using the HXPep program (courtesy of Z. Zhang). (C) HX kinetics of full-length hHsp90 (black circles), yHsc82 (green squares), yHsp82 (blue diamonds), and EcHtpG (black triangles; data from (Graf et al., 2009) for comparison). Relative exchange minus the amide hydrogens of the linker [(Ht–Hlinker)/(Htotal–Hlinker)] is plotted vs. the time in D2O. The data points are averages of two to four independent measurements. The error bars represent standard error of mean. The solid lines are the fits of a bi-exponential equation [Y = Y1·(1−exp(−k1·t)) + Y2·(1−exp(−k2·t)) + Y3] to the data. The inset shows the same data with a logarithmic time scale.