E S1). This motor appears capable of binding microtubules and taking several steps in one directionProtein KCBP (884?253) 6-His-KCBP (884?225) KCBP (884?244) KCBP (820?225) KIC (1?35), Egta or Ca2+ KIC (29?35), Egta or Ca2+Predicted MW 42 kDa 42 kDa 41 kDa 46 kDa 15 kDa 12 kDaMeasured MW 7262 kDa 3662 kDa 41.460.5 kDa 7261 kDa 3461 kDa 2461 kDadoi:10.1371/journal.pone.0066669.tDimerization of KCBP at C-TerminusFigure 5. Differential interference contrast microscopy of microtubules in the presence of different constructs of KCBP. Amicrotubules alone; B- microtubules plus KCBP (884?253); C- microtubules plus KCBP (884?225); D- microtubules plus Ca2+-KIC; E- microtubules plus KCBP (884?253) and Ca2+-KIC; F- microtubules plus KCBP (884?225) and Ca2+-KIC. doi:10.1371/journal.pone.0066669.gbefore reversing and taking several steps in the opposite direction. All microtubules in these samples showed this robust back-andforth behavior, suggesting that this effect is not due to insufficient blocking of the glass surface. Further analysis did not demonstrate a net plus or minus end bias to these movements. Thus, the neck mimic present in KCBP (876?261) does not act as a determinant of the directionality, unlike the true necks in the kinesins [21], and the C-terminal dimers lack directed motility. The N-terminal dimer of KCBP without the regulatory helix, KCBP (820?225), moved toward the minus end of the microtubules (Movie S2). Walking occurred in bursts between pauses resulting in average velocity of 361.5 mm/min. The large standard error in this measured velocity accurately reflects the 23148522 stop-and-go kind of motility we observed. Given the differences in conditions and constructs for these assays, the velocity of 361.5 mm/min is reasonably close to the velocity of 860.6 mm/ min measured previously [9] for N-terminally GST tagged KCBP (820?261) containing both the dimerization 223488-57-1 site domain at the Nterminus and the intact regulatory domain. Expressed and purifiedKCBP (820?261) was unstable and tended to aggregate disallowing assay of the purified protein. Thus, only the N-terminal dimerization domain is required for motility of 1407003 KCBP. Deletion of the regulatory domain Fexinidazole site reduced, but did not abolish KCBP motility and did not affect the directionality of movement.DiscussionIn this study, we have discovered a previously unsuspected selfassociation of KCBP into a dimer via its C-terminal regulatory domain. We have found that the negative coil of KCBP, a part of its regulatory domain, stabilizes KCBP dimers, which are formed by hydrophobic interactions between the residues of the calmodulin binding helices, another part of its regulatory domain. In our experiments, dimerization of KCBP via regulatory domain was observed both in crystals and in solution. In crystals, dimerization was observed for both Arabidopsis KCBP presented here and potato KCBP presented in our earlier works [12,18] and re-analyzedDimerization of KCBP at C-TerminusFigure 6. Structural similarity between KCBP and kinesin-1. A- Structure of Arabidopsis KCBP dimer. Color coding: grey ?motor core, red ?regulatory domain, blue ?ADP. B- Structure of conventional human kinesin-1 (PDB ID 3KIN). Color coding: dark grey ?motor core, green ?neck, blue ?ADP. C- Structural alignment of KCBP and kinesin-1 monomers. D ?Structural alignment of KCBP and kinesin-1 dimers. doi:10.1371/journal.pone.0066669.ghere. In solution, Arabidopsis KCBP dimers were recognized using both size exclusion chromatography and ana.E S1). This motor appears capable of binding microtubules and taking several steps in one directionProtein KCBP (884?253) 6-His-KCBP (884?225) KCBP (884?244) KCBP (820?225) KIC (1?35), Egta or Ca2+ KIC (29?35), Egta or Ca2+Predicted MW 42 kDa 42 kDa 41 kDa 46 kDa 15 kDa 12 kDaMeasured MW 7262 kDa 3662 kDa 41.460.5 kDa 7261 kDa 3461 kDa 2461 kDadoi:10.1371/journal.pone.0066669.tDimerization of KCBP at C-TerminusFigure 5. Differential interference contrast microscopy of microtubules in the presence of different constructs of KCBP. Amicrotubules alone; B- microtubules plus KCBP (884?253); C- microtubules plus KCBP (884?225); D- microtubules plus Ca2+-KIC; E- microtubules plus KCBP (884?253) and Ca2+-KIC; F- microtubules plus KCBP (884?225) and Ca2+-KIC. doi:10.1371/journal.pone.0066669.gbefore reversing and taking several steps in the opposite direction. All microtubules in these samples showed this robust back-andforth behavior, suggesting that this effect is not due to insufficient blocking of the glass surface. Further analysis did not demonstrate a net plus or minus end bias to these movements. Thus, the neck mimic present in KCBP (876?261) does not act as a determinant of the directionality, unlike the true necks in the kinesins [21], and the C-terminal dimers lack directed motility. The N-terminal dimer of KCBP without the regulatory helix, KCBP (820?225), moved toward the minus end of the microtubules (Movie S2). Walking occurred in bursts between pauses resulting in average velocity of 361.5 mm/min. The large standard error in this measured velocity accurately reflects the 23148522 stop-and-go kind of motility we observed. Given the differences in conditions and constructs for these assays, the velocity of 361.5 mm/min is reasonably close to the velocity of 860.6 mm/ min measured previously [9] for N-terminally GST tagged KCBP (820?261) containing both the dimerization domain at the Nterminus and the intact regulatory domain. Expressed and purifiedKCBP (820?261) was unstable and tended to aggregate disallowing assay of the purified protein. Thus, only the N-terminal dimerization domain is required for motility of 1407003 KCBP. Deletion of the regulatory domain reduced, but did not abolish KCBP motility and did not affect the directionality of movement.DiscussionIn this study, we have discovered a previously unsuspected selfassociation of KCBP into a dimer via its C-terminal regulatory domain. We have found that the negative coil of KCBP, a part of its regulatory domain, stabilizes KCBP dimers, which are formed by hydrophobic interactions between the residues of the calmodulin binding helices, another part of its regulatory domain. In our experiments, dimerization of KCBP via regulatory domain was observed both in crystals and in solution. In crystals, dimerization was observed for both Arabidopsis KCBP presented here and potato KCBP presented in our earlier works [12,18] and re-analyzedDimerization of KCBP at C-TerminusFigure 6. Structural similarity between KCBP and kinesin-1. A- Structure of Arabidopsis KCBP dimer. Color coding: grey ?motor core, red ?regulatory domain, blue ?ADP. B- Structure of conventional human kinesin-1 (PDB ID 3KIN). Color coding: dark grey ?motor core, green ?neck, blue ?ADP. C- Structural alignment of KCBP and kinesin-1 monomers. D ?Structural alignment of KCBP and kinesin-1 dimers. doi:10.1371/journal.pone.0066669.ghere. In solution, Arabidopsis KCBP dimers were recognized using both size exclusion chromatography and ana.