Figure 4 - full size

Figure 4. Figure 4 Ribbon diagram of the tubulin dimer showing -tubulin with bound GTP (top), and -tubulin containing GDP and taxotere (bottom). Labels for strands (in the -subunit) and helices (in the -subunit) are included. The arrow indicates the direction of the protofilament and microtubule axis. a, Stereo front view from the putative outside of the microtubule; b, back view from the putative inside of the microtubule; c, side view. Figures produced with AVS (Advanced Visual; ribbon module from M. Carson and A. Shah). The in-out orientation was determined by reference to reconstructions of intact microtubules9. Such reconstructions show prominent longitudinal ridge on the outside, which in our model would be formed by H11, H12 and the loop between H10 and B9, and shallow inside grooves giving the protofilament a bumpy appearance, corresponding in our model to H1, B3 and the long loops in the N-terminal domain. This represents the most likely arrangement of the dimer, because it buries the nucleotide that is at the non-exchangeable site in (see text). For the nucleotide in to be exchangeable at the plus end of a microtubule, the bottom of the figure would correspond to the plus end. We previously presumed the opposite orientation, based on a comparison of the zinc sheets in negatively stained, stain-glucose, and tannin-glucose embedding, with projection maps of open microtubules of known polarity in negative stain9. Some ambiguity in that determination may be introduced by uncertainty about the exact rotational alignment of the protofilament in the sheets with respect to those in open microtubules and by stain artefacts. The polarity with the plus end down would be consistent with experiments that located the -subunit at the plus end of the microtubule^29 and the -subunit at the minus end^30. Circles in b indicate the positions of Cys 241 and Cys 356, separated by about 8 Å.