1: %\documentclass{kluwer} % Specifies the document style.
2: %\documentclass[a4paper,twoside,11pt]{article}
3: \documentclass[pre,aps,twocolumn]{revtex4}
4: \usepackage{graphicx}
5: %\newdisplay{guess}{Conjecture}
6: \newcommand{\be}{\begin{equation}}
7: \newcommand{\ee}{\end{equation}}
8: \newcommand{\ba}{\begin{eqnarray}}
9: \newcommand{\ea}{\end{eqnarray}}
10: \begin{document}
11:
12:
13: \title{Kinesin as an electrostatic machine}
14: \author{A. Ciudad\dag, J.M. Sancho\dag and G.P. Tsironis\dag\ddag}
15: %\runningauthor{A.Ciudad}
16: \title{Kinesin as an electrostatic machine}
17: \address{
18: \dag\ Departament d'Estructura i Constituents de la Mat\`eria,
19: Facultat de F\'{\i}sica,
20: Universitat de Barcelona, Diagonal 647, E-08028 Barcelona, Spain\\
21: \ddag\ Department of Physics, University of Crete
22: and Institute of Electronic Structure and Laser, FORTH,
23: P.O. Box 2208, Heraklion 71003, Crete, Greece.
24: }
25: \date{\today}
26: \begin{abstract}
27: Kinesin and related motor
28: proteins utilize ATP fuel to propel themselves along
29: the external surface of microtubules in a processive and directional fashion. We show that
30: the observed step-like motion is possible
31: through time varying charge distributions
32: furnished by the ATP hydrolysis circle while the static
33: charge configuration on the microtuble provides the guide for motion.
34: Thus, while the chemical hydrolysis energy induces appropriate local conformational
35: changes, the motor translational energy is fundamentally electrostatic.
36: Numerical simulations of the mechanical equations of motion
37: show that processivity and directionality are direct consequences of the ATP-dependent electrostatic interaction between the different charge distributions of kinesin and microtubule.
38: \end{abstract}
39:
40: \keywords{kinesin electrostatics, tubulin dipole moment, neck domain, ATP hydrolysis, microtubule directionality, kinesin processivity, kinesin substep}
41: \maketitle
42:
43:
44:
45:
46: \section{Introduction}
47: %Paragraph 1:
48: %Conformational changes.
49:
50: The kinesin family is a set of motor proteins that move on the
51: surface of microtubules and shuttle various
52: cargo molecules to different parts of the cell\cite{Tuz}-\cite{parallel}.
53: The energy for kinesin
54: transport motion is provided by hydrolysis of ATP molecules that
55: are in its vicinity and attach to a specific site of the protein.
56: Despite much progress in the area, the precise, microscopic fashion of ATP
57: action is not known. Kinesin motion is step-like with one step for each ATP
58: hydrolysis while processivity and directionality depend on specific neck properties of each particular type of the family. These features of motor protein motion led several years ago in the introduction of mechanical ratchet models that could provide some
59: insight on the phenomenology of individual \cite{Reimann,Strat,Fisher,bier,sancho} and collective \cite{julicher,nishinari} motion.
60: Although many issues and especially those
61: related to the random aspects of the walk where addressed successfully by these
62: models, fundamental questions on the energetics and nature of forces that enable
63: the walk have not been discussed. In the present work we focus on the latter;
64: specifically we use electrostatic information of the microtubule and kinesin and
65: evaluate the full electric force that binds the latter on the former. We then show that
66: the actual kinesin walk is caused by charge distribution changes on kinesin
67: enabled by the action of ATP. The walk that is a combination of
68: conformational changes accompanied by kinesin charge reshuffling is found to be
69: fundamentally governed by electrostatic forces.
70: \section{Kinesin electrostatic model}
71: Kinesins walk on microtubules; the latter are made by
72: alpha and beta of tubulin proteins that
73: are highly charged. Molecular modeling shows that, in the presence of
74: water, tubulins have additionally a permanent dipole moment that points
75: towards the cylindrical symmetry axis of the microtubule with
76: a non-zero component parallel to the protofilaments towards their minus
77: end\cite{Tuz}. Electrostatically thus a microtubule has negative surface
78: charge accompanied by a positive charge distribution in its immediate interior
79: with a slanted polarization vector that is larger in the beta than the
80: alpha subunit (Figure \ref{1}). This symmetry breaking induced by the tubulin dipole moment direction
81: plays a key role in the specificity of the attachment of a kinesin head onto a
82: beta tubulin subunit\cite{decoration}.\\
83:
84: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
85: \begin{figure}[ht]
86: \begin{center}
87: \includegraphics{b1.eps}
88: \end{center}
89: \caption{ a) Microtubule electrostatic model with arrows indicating local dipole moments.
90: The $\alpha$ tubulin subunits (dark) have smaller dipole moment that
91: the $\beta$ units (lighter color).
92: b) Electrostatic configuration of kinesin and tubulin made protofilament prior to ATP
93: hydrolysis. Angle $\theta$ is polar while $\phi$ is azimuthal.
94: The central-neck charge sign depends on the type of molecular protein while the head
95: charges depend on the ATP hydrolysis circle. Dipolar lenghts $d_{\alpha}$, $d_{\beta}$ and dipolar angles $\omega_{\alpha}$, $\omega_{\beta}$ are different
96: in $\alpha$ and $\beta$ subunits respectively. }
97: \label{1}
98: \end{figure}
99:
100: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
101:
102: Structural analysis of kinesin and related proteins suggests
103: that there are three charge domains,
104: one in each head region and a third
105: one in the central neck linker domain. The head charge distribution depends strongly
106: on the ATP or ADP presence while the neck charge is specific to the protein.
107: For modeling purposes it is sufficient to consider kinesin and similar motor proteins
108: as a body containing three charges located on the apexes of a triangle with angle
109: $\gamma$ subtending from the neck charge to the two head charges (Figure \ref{1}b)
110: These charge distributions interact with tubulin multipole fields and determine
111: at each instant of time the dynamic state of the molecule.
112:
113: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
114:
115: In our analysis of kinesin walk to be detailed below, we place
116: motor proteins and microtubules in an overdamped medium with thermal energy
117: $k_BT=4.1pNnm$ and motor drag force $10^{-6}pNs/rad$ per rotational degree of freedom. For the sake of simplicity, we assume a relative electric permitivity equal to 80 everywhere and a Debye length $l_D$ above $3nm$. Although Debye-H\"uckel theory predicts $l_D\sim1$, the interaction range may be enhanced by the channeling of the electric field along the interior of the proteins, which is not accesible by diffusing ions or water molecules. Then it is reasonable to take $l_D$ of the order of the size of the kinesin.
118: We take consecutive tubulin unit distance to be $8nm$, $4nm$ per $\alpha$ or $\beta$ subunit. The lateral space between two protofilaments is $5nm$ while the pitch of $1nm$
119: between consecutive protofilamens is also included. For a processive plus-ended kinesin
120: and its non-processive-minus-ended chimera ncd we use for the head-to-head distance the value of $6 nm$ while we consider the heads to be charged with $+2 e$ in the absence of nucleotides, $-2e$ with ATP attachment and $-1e$ with ADP.
121: These head charge values are consistent with the charges involved in hydrolysis reaction
122: \be
123: ATP^{4-} + OH^{1-}\rightarrow ADP^{3-}+H_2PO_3^{2-}.
124: \ee
125:
126: The value of the central motor charge changes for different proteins; as a result
127: we consider different motors with charges in the range $[-2e:+2e]$.
128:
129: For the electrostatic distribution of microtubules we assign negative surface
130: charge $q= -27 e$ per tubulin subunit while include a positive charge distribution
131: in the interior leading to dipole moment magnitude of
132: $ p = 5000D \simeq 100 Cnm$\cite{Tuz},
133: or $d \simeq 4nm$ ($p=qd$). Finally, we use a
134: dipolar tilt with length and angle values equal to
135: $d_{\alpha}\simeq 2nm$, $\omega_{\alpha}\simeq 0.07 rad$
136: and $d_{\beta}\simeq 4nm$,
137: $\omega_{\beta}\simeq 0.14 rad$, for the $\alpha$ and $\beta$
138: subunits respectively (Figure \ref{1}b).
139:
140: The microtubule-induced kinesin interaction potential is
141: given by
142: \be
143: V(\vec{r}_i)=\frac{-1}{4\pi\epsilon_0\epsilon_r(1+ka)}\sum_{j=1}^{N}
144: \frac{q_iq_j}{|\vec{r}_i -\vec{r}_j|}e^{k(a-|\vec{r}_i -\vec{r}_j|)},
145: \label{eq2}
146: \ee
147: where $\vec{r}_i$ is the position vector labeling the charges on kinesin
148: ($i=1,2,3$), while $\vec{r}_j$ is the location of the N microtubule charges $q_j$
149: on the $\alpha$ and $\beta$ subunits. $k$ is the inverse of the Debye length, which we take around $\sim3.5 nm$, and $a\sim1 nm$ is the excluding volume radius as described in Ref. \cite{Levin}. A regime with a greater $a$ and lower Debye length $l_D$ is also operative.
150: We considered a flat microtubule with five protofilaments; due to the rapid decay of the force out from protein volumes, we include in total the $N=10$ closest tubulin charges to kinesin. For the simplest case when neck and head charges are aligned we have
151: $\gamma=\pi$ and the protein reduces to a triply charged rigid rod.
152:
153: We have
154: tested the model for several values of $\gamma$ smaller than $\pi$; we
155: have found that the model is not fully opperational for
156: $\gamma \lesssim 175^0$. We have also tested the model
157: considering that the angle $\gamma$ has some elasticity as well.
158: We found that in case the angle is very soft,
159: the raising head rapidly loses its power to pull the motor forward.
160: For the purposes of the model, the motor core and the neck need to be stiff enough
161: to maintain the values of $\gamma$ in the range discussed above even
162: at maximum load conditions.
163: In this simpler rod-like configuration the polar angle $\theta$ and the azimuthal angle $\phi$ are
164: sufficient for describing the motor rotation. We can describe the motion using
165: the following overdamped equations of
166: motion for kinesin:
167:
168: \be
169: \lambda\dot{\theta}=-\frac{1}{L}\frac{dV(\vec{r})}{d\theta}+\xi_{\theta}(t)
170: \label{eq3}
171: \ee
172: and
173: \be
174: \lambda\dot{\phi}=-\frac{1}{L}\frac{dV(\vec{r})}{d\phi}+\xi_{\phi}(t)
175: \label{eq4}
176: \ee
177: where $\lambda$ is the drag coefficient, $L$ the head-to-head distance
178: and
179: $V(\vec{r})$ the total microtubule electrostatic potential of Eq. (\ref{eq2}) at the
180: Cartesian location $\vec{r}$.
181: The environment is simulated through the thermal forces
182: $\xi_{\theta}(t)$ and $\xi_{\phi}$; for each we have
183: $\langle\xi(t)\rangle=0$ and $\langle\xi(t)\xi(t')\rangle=2\lambda
184: k_BT\delta(t-t')$. In order to integrate Eqns. (\ref{eq3}, \ref{eq4})
185: we need to perform at each instant of time the Cartesian-to-polar
186: transformation
187: $\theta=\theta(\vec{r},L)$. For the rotation we consider the
188: attached head to coincide with the origin of the coordinate system;
189: the latter is shifted by $8nm$ each time a step is completed. The simple
190: Larmor-like rotation of the protein for $\gamma = \pi$ becomes a
191: more complex rigid body rotation for $\gamma<\pi$.
192: Extensive numerical simulations with the model described previously for
193: the kinesin-microtubule complex led to the following quantitative picture for the motion.
194: In equilibrium (parked) state the nucleotide-free head is positively
195: charged and thus attached to the beta subunit of the negative microtubule surface
196: while the other head containing ADP is negatively charged and tethered through the neck.
197: Due to the slanted microtubule dipole moment, the kinesin axis is tilted but in a direction
198: determined by the central charge sign; for wild type kinesin
199: points to the fast-growing-end while for ncd points to the minus end ( Figure \ref{2}).
200: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
201: \begin{figure}[ht]
202: \begin{center}
203: \includegraphics[width=8 cm]{b2.eps}
204: \end{center}
205: \caption{ Electrostatics driven motor walk:
206: Wild-type kinesin (+ states) and ncd (- states) stepping process (left column)
207: and numerically determined
208: binding protein-microtubule electrostatic potential as a function of local polar angle
209: $\theta$ (right column).
210: (0) Before ATP hydrolysis, both kinesin (positive neck) and ncd motors (negative neck) are in parked configuration pointing in opposite directions due to the difference in the central charge. The corresponding
211: equilibrium angles are determined for the minimum potentials.
212: (+1) ATP entry in the kinesin (attached) head pocket, with an accompanying charge
213: change while the ADP at the other (tethered) head becomes unstable.
214: (+2) The reversal and shift of the
215: interaction potential of the previous state leads to
216: falling of the tethered head deterministically towards the plus end. Since
217: the length of the motor is not sufficient for reaching the next tubulin subunit we
218: have (+3) a detachment and rising of the trailing head in such a way that allows the other head to slide to the next binding site.
219: The ncd motor protein cycle proceeds similarly (- states) but the parked state is tilted towards the minus-end. Moreover, the negative-central charge induces a potential shift which is opposite to the plus-ended case. The falling of ncd motors is slower than positive-charged-neck motors, so the probability that the attached heads begins the rising before the tethered heads completes the falling is greater, leading to non-processivity.}
220: \label{2}
221: \end{figure}
222: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
223:
224:
225: All experimental data agree that
226: after nucleotide entry in the head pocket, the binding of the attached head to the microtubule
227: and the ADP binding to the tethered head become unstable, signaling the onset of a cascade of events leading to the walk.
228: In our model this critical juncture occurs because in the absence of nucleotide the positively charged head becomes negative
229: acquiring $-4e$ charges after ATP binding. Although the entry of a negatively
230: charged molecule in the head pocket is necessary for the commencement of the walk
231: process, the release of the attached head is not immediate, since a head-tubulin chemical bond needs to be broken. ATP hydrolysis energy thus is used for
232: a local conformational change that captures the negative charge after
233: dephosphorylation. Although charged ADP nucleotides in the medium compete with ATP entering also in the pocket, they do not hydrolyze and, as a result, cannot induce the local conformation that
234: will trap them. In our electrostatic model thus hydrolysis energy induces
235: primarily a specific local conformational change, being the power stroke electrostatic in nature.
236: \section{Kinesin dynamics}
237:
238: Based on this picture, the protein parked configuration
239: is a result of the microtubule-empty head attraction, microtubule-tethered
240: head repulsion while the kinesin angle wrt microtubule
241: is due to the tubulin dipolar tilts. This stable configuration is
242: maintained during a random and [ATP]-load-dependent dwell time
243: until a new ATP binds into the attached head.
244: When this occurs, the local charge changes
245: and the ADP trapped in the tethered head experiences an additional repulsion,
246: becoming unstable and eventually opening the pocket and exiting.
247: After ADP expulsion the tethered head becomes positive, the attached head charge is negative
248: and the protein becomes electrostatically unstable. As a result the tethered head
249: collapses onto the microtubule ("falling processes"), while the attached head
250: is forced to detach from the surface ("rising process") and a new, shifted, park state is
251: reached.
252:
253: Every kinesin step thus includes
254: falling of the leading head onto the microtubule followed by
255: rising of the trailing head ( Figure \ref{2}).
256: We performed extensive 3D numerical simulations and found that the
257: electrostatic force field generated by the microtubule is able to
258: drive these two sequential processes while keeping the motor faithful to
259: a given protofilament.
260: We note that, energetically, the parked state represents a potential energy minimum for the
261: interaction between free head charge and microtubule.
262: When ADP is released and the electrostatic charge distribution of the
263: motor changes this minimum becomes a potential
264: maximum, i.e. the repulsion of the tethered head turns into attraction.
265:
266: For a null-charged-neck case, even though the motor is tilted in the parked state, there is no motion directionality since the falling of the tethered head does not have
267: a preferred collapsing side. The directionality features of the motor enter through the
268: charge distributions of the neck region, which is responsible of the potential shift between stable and unstable equilibrium points. Specifically, when the central charge
269: is positive, not only the protein is more processive due to the attraction
270: with the negatively charged tubulin surfaces, but it also walks towards
271: the plus-end of the microtubule; this is precisely the case with wild-type kinesin.
272: In ncd, on the other hand, the central charge is negative, the protein is
273: non-processive and walks in the opposite direction towards the minus end.
274: Additionally, in proteins with no charge in the neck region,
275: the walk is known not to be deterministic but random\cite{Mutante}.
276: We note further that our simulations
277: are compatible with the experiments in reference \cite{parallel}
278: whereby the
279: falling motor direction is parallel to the protofilament axis, with
280: some variations and latteral collapses leading to protofilament changes
281: due to thermal fluctuations.
282: The lateral periodicity of the microtubule lattice is $5nm$ while the axial
283: one only $4nm$; these values determine crucially
284: the motion after the destabilization of the parked state since the electric force is stronger in the axial direction.
285: We find that this electrostatic model captures fully the directionality and
286: processivity features of the known motors.
287:
288:
289: Two different time-scales characterize the actual kinesin motion.
290: While dwell times between
291: two steps, modulated by ATP concentration and the external load
292: are of the order of milliseconds\cite{Visscher},
293: the action of the step itself lasts only some
294: microseconds \cite{substepsno}; this is the reason for the stepping
295: appearance of the trajectories.
296: In the relatively short period of the step time, several
297: processes must occur. First, falling of the leading head
298: after ADP release, a process activated by ATP binding on the attached head.
299: Subsequently the two heads remain attached onto the microtubule for a
300: time not longer than 30 microseconds \cite{substepsno}.
301: Recent measurements were able to track bead-movements at the
302: microsecond scale \cite{substepsno}. The recording of a single step, in our interpretation, reveals two different regimes that may be separated by a quick stage at which both kinesin heads are attached to the microtubule.
303: Kinesin crytallization data
304: show that the separation
305: between heads is smaller than 8 nm\cite{cristalina}, i.e.
306: slightly smaller than
307: the protofilament spatial period and, as a result,
308: after falling the leading head
309: cannot reach the beta subunit location before the trailing head detaches.
310: Thus, sliding may take place while the trailing head rises to reach
311: the new parked state. This aspect is portrayed in
312: Figure \ref{3} in a sequential form although in practice both
313: sliding and rising could occur simultaneously.
314:
315: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
316: \begin{figure}[ht]
317: \begin{center}
318: \includegraphics{b3.eps}
319: \end{center}
320: \caption{Numerical results for a kinesin step at (a) $ms$ time-scale and (b) $\mu s$
321: time-scale, for $\gamma=\pi$; $x$-axis denotes time while $y$-axis kinesin
322: position..
323: Occurrence of non-processivity and substeps are in competition since for the former
324: detachment of the trailing head occurs before than the falling of tethered head, while for substeps to occur the falling process must be done before detachment, leading to a processive motion.
325: The millisecond time-scale numerical results shown in (a) are detailed
326: at a microsecond time-scale in (b) where the
327: aforementioned competition is seen.
328: }
329: \label{3}
330: \end{figure}
331: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
332:
333: In order to address motor protein processivity
334: we consider the two sets of chemical reactions that are activated by ATP binding at the trailing head. The first group is related to a cascade of rapid reactions involving
335: ATP hydrolysis, dephosphorylation and subsequent head detachment
336: with an estimated time 500
337: $\mu s$ for each\cite{chemistry}.
338: Secondly, ATP activates ADP release in the other head, although
339: the typical time of this process is not known. There is a
340: competition between these two reactions; if hydrolysis and
341: detachment occur faster, the whole motor detaches from the microtubule
342: and processivity ceases. If, on the other hand, ADP release and falling is the fastest
343: process, then a small substep occurs while both heads are attached.
344: Processivity thus is related to the competition of these two processes.
345: In the case of non-processive ncd, for instance, the existence of a
346: negative central charge has the effect of slowing
347: down the falling process while
348: speeding up the rising one resulting, thus, in a non-processive motor. This effect has been tested experimentally in Ref. \cite{Vale}.\\
349:
350: We point out that when ATP enters into the attached head,
351: the system becomes electrostatically unstable.
352: However, detachment of the head occurs some time later triggered by
353: the phosphate release. This fact implies that the chemical bonding
354: that binds the motor to the microtubule, while
355: stable in absence of ATP, becomes unstable when the
356: nucleotide arrives. The time delay between ATP entry and bond dissociation
357: appears to be crucial for motor processivity.
358: In the Figure \ref{2}, we have assumed that both possesive and
359: non-possesive motors may perform the full detachment-attachment cycle,
360: although, in the case of ncd, the chemical bond may be actually broken
361: before the falling head reaches the next binding site.
362: The time duration of a single kinesin step
363: is very small compared to the parked dwell times, and, as a result,
364: global observables such as the mean velocity or randomness
365: can be predicted using only chemical kinetics.
366: This has been shown in reference \cite{Ciudad} where a
367: kinetic approximation was shown to be sufficient
368: in fitting the measured data, even in the presence of an external load.
369: However, our present approach shows, that thermal fluctuations alone
370: may not produce directly the $16 nm$ long displacements in a $\mu s$ time-scale
371: necessary for the walk.
372:
373:
374: \section{Conclusions}
375:
376: In this work we attempted to analyze the complex motion of molecular motor
377: proteins from a mesoscopic point of view placing emphasis on the fundamental
378: interactions that enable the motion. We found that the motion is primarily driven
379: by the electrostatic interaction between the charged microtubule surface and
380: the fluctuating motor head charges. The nonequilibrium aspect of the walk is
381: provided by the ATP hydrolysis cycle that furnishes appropriate
382: charge motor distributions that make the walk possible. We made several
383: assumptions in this work, most of which have been motivated by the
384: current status of knowledge in the area. We considered the motor as a relatively
385: rigid one; this assumption is presently well founded,
386: however it may be lifted through an
387: improved motor model that includes additionally protein eleasticity.
388: Furtheremore, our model does not address at all the structual changes
389: in the head pockets effected by ATP hydrolysis, the mode of the
390: local energy relase as
391: well as the mechanism for the ADP detachment from the thethered head.
392: These issues that may
393: involve complex conformational changes as well as possibly charge or energy
394: transfer processes must be addressed through a more foundamental,
395: microscopic model. Finally,
396: it is crucial to evaluate the range of Debye screening in the vicinity of
397: a highly charged surface such as the microtubule. While our motor model
398: remains fully operational to a Debye length range of approximately $3 nm$, one
399: certainly needs to address the complex nature of the electrostatic shielding in the
400: vicinity of the microtubule and assess the true range of the electrostatic
401: forces involved in the walk.
402: We note that our model provides a consistent dynamical
403: picture for the kinesin walk that is
404: based on two premises, viz.
405: (a) that the ATP hydrolysis energy is used for a head
406: local conformational change that captures locally charges and (b) the motor
407: motion is driven by electrostatic forces. The emerging qualitative and
408: quantitative picture for the walk is fully compatible with all known
409: experimental data, while, furthermore, it is testable experimentally.
410: If electrostatics, charge transfer as well as capture
411: indeed link the chemistry of ATP and the mechanics of
412: kinesin, this may hold true for other ATP-dependent processes as well.
413:
414:
415:
416:
417:
418:
419:
420:
421:
422:
423:
424:
425:
426: \acknowledgements
427: We thank Marta Iba\~nes for discussions.
428: This work was supported by the Ministerio de Educaci\'on y Ciencia
429: (Spain) under Project No. $FIS2006-11452-C03-01$, Grant No. $BES-2004-3208$,
430: by grant $2006PIV10007$ of the Generalitat de Catalunya
431: and Grant "Pythagoras II" $KA-2102-TDY-25$ of the Ministry of
432: Education of Greece
433: and the European Union.
434:
435:
436:
437: \begin{thebibliography}{}
438:
439: \bibitem[1]{Tuz} J.~A. Tuszynski, T. Luchko, E.~J. Carpenter and E. Crawford.
440: \newblock{Electrostatic Properties of Tubulin and Their Consequences for Microtubules}.
441: \newblock{\em Journal of Computational and Theoretical Nanoscience}, 1:1--6, 2005.
442:
443: \bibitem[2]{decoration} M. Thorm\"ahlen, A. Marx, S.~A. M\"uller, Y.~H. Song, E.~M. Mandelkow, U. Aebi and E. Mandelkow.
444: \newblock{Interaction of Monomeric and Dimeric Kinesin with Microtubules}.
445: \newblock{\em Journal of Molecular Biology}, 275:795--809, 1998.
446:
447: \bibitem[3]{Mutante} S.~A. Endow and H. Higuchi.
448: \newblock{A mutant of the motor protein kinesin that moves in both directions on microtubules}.
449: \newblock{\em Nature}, 406:913--916, 2000.
450:
451: \bibitem[4]{Visscher} K. Visscher, M.~J. Schnitzer and S.~M. Block.
452: \newblock{Single kinesin molecules studied with a molecular force clamp}.
453: \newblock{\em Nature}, 400:184--189, 1999.
454:
455: \bibitem[5]{review} A.~A. Kasprzak and A. Hajdo.
456: \newblock{Directionality of kinesin motors}.
457: \newblock{\em Acta Biochimica Polonica}, 49(4):813--821, 2002.
458:
459: \bibitem[6]{substepsno} N.~J. Carter and R.~A. Cross.
460: \newblock{Mechanics of the kinesin step}.
461: \newblock{\em Nature}, 435(19):308--312, 2005.
462:
463: \bibitem[7]{cristalina} F. Kozielski, S. Sack, A. Marx, M. Thorm\"ahlen, E. Sch\"onbrunn, V. Biou, A. Thompson, E.~M. Mandelkow and E. Mandelkow.
464: \newblock{The Crystal Structure of Dimeric Kinesin and Implications for Microtubule-Dependent Motility}.
465: \newblock{\em Cell}, 91:985--994, 1997.
466:
467: \bibitem[8]{chemistry} R.A. Cross.
468: \newblock{The kinetic mechanism of kinesin}.
469: \newblock{\em TRENDS in Biochemical Sciences}, 29(6):301-309, 2004.
470:
471: \bibitem[9]{parallel} S. Ray, E. Meyhfer, R.~A. Milligan and J. Howard.
472: \newblock{Kinesin Follows the Microtubule's Protofilament Axis}.
473: \newblock{\em The Journal of Cell Biology}, 121(5):1083--1093, 1993.
474:
475: \bibitem[10]{Reimann} P. Reimann.
476: \newblock{Brownian motors: noisy transport far from equilibrium}.
477: \newblock{\em Physics Reports}, 361:57--265 , 2002 and references therein.
478:
479: \bibitem[11]{TL} G.~P. Tsironis and K. Lindenberg.
480: \newblock{}.
481: \newblock{\em Adv. in Structural Bio.}, 5(eds):271--281 , 1998.
482:
483: \bibitem[12]{Strat} G. Stratopoulos, T. Dialynas and G.~P. Tsironis.
484: \newblock{Directional Newtonian motion and reversals of molecular motors}.
485: \newblock{\em Physics Letters A}, 252:151--156, 1999.
486:
487: \bibitem[13]{Fisher} M.~E. Fisher and A.~B. Kolomeisky.
488: \newblock{Simple mechanochemistry describes the dynamics of kinesin molecules}.
489: \newblock{\em PNAS}, 98:7748--7753, 2001.
490:
491: \bibitem[14]{bier} M. Bier.
492: \newblock{Processive Motor Protein as an Overdamped Brownian Stepper}.
493: \newblock{\em Physical Review Letters}, 91(14):148104, 2003.
494:
495: \bibitem[15]{sancho} A. Ciudad, A.~M. Lacasta and J.~M. Sancho.
496: \newblock{Physical analysis of a processive molecular motor: The conventional kinesin}.
497: \newblock{\em Physical Review E}, 72:031918, 2005.
498:
499: \bibitem[16]{julicher} F. J\"ulicher, A. Ajdari and J. Prost.
500: \newblock{Modeling molecular motors}.
501: \newblock{\em Review of Modern Physics}, 69:1269--1282, 1997.
502:
503: \bibitem[17]{nishinari} K. Nishinari, Y. Okada, A. Schadschneider and D. Chowdhury.
504: \newblock{Intracellular Transport of Single-Headed Molecular Motors KIF1A}.
505: \newblock{\em Physical Review Letters}, 95:118101, 2005.
506:
507: \bibitem[18]{Levin} Yan Levin.
508: \newblock{Electrostatic correlations: from plasma to biology}.
509: \newblock{\em Reports on progress in physics}, 65:1577--1632, 2002.
510:
511: \bibitem[19]{Vale} K.~S. Thorn, J.~A. Ubersax and R.~D. Vale.
512: \newblock{Engineering the Processive Run Length of the Kinesin Motor}.
513: \newblock{\em The Journal of Cell Biology}, 151(5):1093--1100, 2000.
514:
515: \bibitem[20]{Ciudad} A. Ciudad and J.~M. Sancho.
516: \newblock{External mechanical force as an inhibition process in kinesin's motion}.
517: \newblock{\em Biochemical Journal}, 390:345--349, 2005.
518:
519:
520:
521: \end{thebibliography}
522:
523: \end{document}
524: