Actin-binding domain of Rng2 sparsely bound on F-actin strongly inhibits actin movement on myosin II.

We report a case in which sub-stoichiometric binding of an actin-binding protein has profound structural and functional consequences, providing an insight into the fundamental properties of actin regulation. Rng2 is an IQGAP contained in contractile rings in the fission yeast Schizosaccharomyces pombe Here, we used high-speed atomic force microscopy and electron microscopy and found that sub-stoichiometric binding of the calponin-homology actin-binding domain of Rng2 (Rng2CHD) induces global structural changes in skeletal muscle actin filaments, including shortening of the filament helical pitch. Sub-stoichiometric binding of Rng2CHD also reduced the affinity between actin filaments and muscle myosin II carrying ADP and strongly inhibited the motility of actin filaments on myosin II in vitro. On skeletal muscle myosin II-coated surfaces, Rng2CHD stopped the actin movements at a binding ratio of 11%. Rng2CHD also inhibited actin movements on myosin II of the amoeba Dictyostelium, but in this case, by detaching actin filaments from myosin II-coated surfaces. Thus, sparsely bound Rng2CHD induces apparently cooperative structural changes in actin filaments and inhibits force generation by actomyosin II.

Structure and Regulation of the Movement of Human Myosin VIIA.

Human myosin VIIA (HM7A) is responsible for human Usher syndrome type 1B, which causes hearing and visual loss in humans. Here we studied the regulation of HM7A. The actin-activated ATPase activity of full-length HM7A (HM7AFull) was lower than that of tail-truncated HM7A (HM7AΔTail). Deletion of the C-terminal 40 amino acids and mutation of the basic residues in this region (R2176A or K2179A) abolished the inhibition. Electron microscopy revealed that HM7AFull is a monomer in which the tail domain bends back toward the head-neck domain to form a compact structure. This compact structure is extended at high ionic strength or in the presence of Ca(2+). Although myosin VIIA has five isoleucine-glutamine (IQ) motifs, the neck length seems to be shorter than the expected length of five bound calmodulins. Supporting this observation, the IQ domain bound only three calmodulins in Ca(2+), and the first IQ motif failed to bind calmodulin in EGTA. These results suggest that the unique IQ domain of HM7A is important for the tail-neck interaction and, therefore, regulation. Cellular studies revealed that dimer formation of HM7A is critical for its translocation to filopodial tips and that the tail domain (HM7ATail) markedly reduced the filopodial tip localization of the HM7AΔTail dimer, suggesting that the tail-inhibition mechanism is operating in vivo. The translocation of the HM7AFull dimer was significantly less than that of the HM7AΔTail dimer, and R2176A/R2179A mutation rescued the filopodial tip translocation. These results suggest that HM7A can transport its cargo molecules, such as USH1 proteins, upon release of the tail-dependent inhibition.

Nucleotide-dependent displacement and dynamics of the α-1 helix in kinesin revealed by site-directed spin labeling EPR.

In kinesin X-ray crystal structures, the N-terminal region of the α-1 helix is adjacent to the adenine ring of the bound nucleotide, while the C-terminal region of the helix is near the neck-linker (NL). Here, we monitor the displacement of the α-1 helix within a kinesin monomer bound to microtubules (MTs) in the presence or absence of nucleotides using site-directed spin labeling EPR. Kinesin was doubly spin-labeled at the α-1 and α-2 helices, and the resulting EPR spectrum showed dipolar broadening. The inter-helix distance distribution showed that 20% of the spins have a peak characteristic of 1.4-1.7 nm separation, which is similar to what is predicted from the X-ray crystal structure, albeit 80% were beyond the sensitivity limit (>2.5 nm) of the method. Upon MT binding, the fraction of kinesin exhibiting an inter-helix distance of 1.4-1.7 nm in the presence of AMPPNP (a non-hydrolysable ATP analog) and ADP was 20% and 25%, respectively. In the absence of nucleotide, this fraction increased to 40-50%. These nucleotide-induced changes in the fraction of kinesin undergoing displacement of the α-1 helix were found to be related to the fraction in which the NL undocked from the motor core. It is therefore suggested that a shift in the α-1 helix conformational equilibrium occurs upon nucleotide binding and release, and this shift controls NL docking onto the motor core.

Biochemical characterization of the novel rice kinesin K23 and its kinetic study using fluorescence resonance energy transfer between an intrinsic tryptophan residue and a fluorescent ATP analogue.

We previously demonstrated that the rice kinesin K16, which belongs to the kinesin-7 subfamily, has unique enzymatic properties and atomic structure within key functional regions. In this study, we focused on a novel rice plant kinesin, K23, which also belongs to the kinesin-7 subfamily. The biochemical characterization of the K23 motor domain (K23MD) was studied and compared with the rice kinesin K16 and other related kinesins. K23 exhibits ∼45-fold (1.3 Pi mol(-1) site mol(-1) s(-1)) lower microtubule-dependent ATPase activity than conventional kinesins, whereas its affinity for microtubules is comparable with conventional kinesins. MgADP-free K23 is unstable compared with the unusually stable MgADP-free K16MD. The enzymatic properties of K23MD are somewhat different from those of K16. We used a fluorescent ATP analogue 2'(3')-O-(N'-methylanthraniloyl)-ATP (mant-ATP) for the kinetic characterization of K23. The fluorescence of mant-ATP was not significantly altered during its hydrolysis by K23. However, significant fluorescence resonance energy transfer (FRET) between mant-ATP and W21 in the motor domain was observed. The kinetic study using FRET revealed that K23 has unique kinetic characteristics when compared with other kinesins.

Photocontrol of calmodulin interaction with target peptides using azobenzene derivative.

Calmodulin (CaM), a physiologically important Ca(2+)-binding protein, participates in numerous cellular regulatory processes. It is dumbbell shaped and contains two globular domains connected by a short alpha-helix. Each of the globular domains has two Ca(2+)-binding sites, the EF hands. CaM undergoes a conformational change upon binding to Ca(2+), which enables it to bind to specific proteins for specific responses. Here, we successfully photocontrolled CaM binding to its target peptide using the photochromic compound N-(4-phenylazophenyl) maleimide (PAM), which reversibly undergoes cis-trans isomerization upon ultraviolet (UV) and visible (VIS) light irradiation. In order to specifically incorporate PAM, CaM mutants having reactive cysteine residues in the functional region were prepared; PAM was stoichiometrically incorporated into the cysteine residues in these mutants. Further, we prepared the target peptide, M13, fused with yellow fluorescent protein (YFP) to monitor the CaM-M13 peptide interaction. The binding of the PAM-CaM mutants, N60C, D64C and M124C, to M13-YFP was reversibly photocontrolled upon UV-VIS light irradiation at appropriate Ca(2+) concentrations.

Conformational dynamics of loops L11 and L12 of kinesin as revealed by spin-labeling EPR.

The EPR spectra of the spin labels attached to loops L11 and L12 of kinesin were resolved into slow (rotational correlation time, tau=10-45 ns) and fast (tau=2 ns) components. The fraction of the slow component increased considerably when kinesin was complexed with a microtubule (MT). On MT binding and in the presence of nucleotides ADP and AMPPNP, the spin labels on L11, particularly at A252C and L249C, significantly decreased the fraction of the slow component. Moreover, dipolar EPR detected a wide distribution in distance range, 1-2 nm between the two spin labels attached to T242C/A252C or A247C/A252C; this distribution was slightly narrower in the presence of MTs than in their absence. These results suggested that the L11 residues undergo conformational transition on the binding of nucleotides and MT, while these residues remained to fluctuate over a nanometer range.

Photocontrol of kinesin ATPase activity using an azobenzene derivative.

Azobenzene is a photochromic molecule that undergoes rapid and reversible isomerization between the cis- and trans-forms in response to ultraviolet (UV) and visible (VIS) light irradiation, respectively. Here, we introduced the sulfhydryl-reactive azobenzene derivative 4-phenylazophenyl maleimide (PAM) into the functional region of kinesin to reversibly regulate the ATPase activity of kinesin by photoirradiation. We prepared five kinesin motor domain mutants, A247C, L249C, A252C, G272C and S275C, which contained a single reactive cysteine residue in loops L11 and L12. These loops are considered to be key regions for the functioning of kinesin as a motor protein. PAM was stoichiometrically incorporated into the cysteine residues in the loops of the mutants. The PAM-modified S275C mutant exhibited reversible alterations in ATPase activity accompanied by cis-trans isomerization upon UV and VIS light irradiation. The ATPase activity exhibited by the cis-isomer of the PAM bound to the mutant was two times higher than that of the trans-isomer. Further, the PAM-modified L249C mutant exhibited reversible alterations in ATPase activity on UV-VIS light irradiation but exhibited the opposite effect on UV and VIS light irradiation. Using a photochromic azobenzene derivative, we have demonstrated that the ATPase activity of the motor protein kinesin is photoregulated.