1: % Dernierement modifie au 12/09/04 sur holbein
2: % (at)Recherche/Kuramaoto/Draft/YK.tex
3:
4: \documentclass{elsart}
5: % \documentclass[doublespacing]{elsart}
6:
7: % style referencing;
8: \usepackage{natbib}
9:
10: \usepackage[dvips]{graphics}
11: % \usepackage{graphicx}
12: % \usepackage{epsfig}
13:
14:
15: \newcommand{\beq}{\begin{equation}} \newcommand{\eeq}{\end{equation}}
16: \newcommand{\beqa}{\begin{eqnarray}} \newcommand{\eeqa}{\end{eqnarray}}
17: \newcommand{\bde}{\begin{description}} \newcommand{\ede}{\end{description}}
18: \newcommand{\ben}{\begin{enumerate}} \newcommand{\een}{\end{enumerate}}
19: \newcommand{\ind}{\indent } \newcommand{\noic}{\noindent $\circ$}
20: \newcommand{\noi}{\noindent} \newcommand{\noib}{\noindent $\bullet$}
21: \newcommand{\clp}{\clearpage }
22: \newcommand{\hs}{\hspace{1cm} } \newcommand{\vs}{\vspace{3mm} }
23: \newcommand{\la}{\langle} \newcommand{\ra}{\rangle}
24: \newcommand{\AnP}{{A$\{^{\rm T}_{\rm D}\}$P}}
25: \newcommand{\der}[2]{\frac{d {#1}}{d {#2}}} %\der{K}{t}
26: \newcommand{\term}[1]{{\bf \underline{{#1}}}}
27: \newcommand{\gtsim}{\mbox{${}^{>}_\sim$}}
28: \newcommand{\ltsim}{\mbox{${}^{<}_\sim$}}
29: \newcommand{\kT}{\mbox{$k_{\rm B}T$}}
30:
31:
32: %
33: \begin{document}
34: \begin{frontmatter}
35: \title{Autonomous free-energy transducer \\
36: working under thermal fluctuations}
37: \author{Ken Sekimoto}
38: \address{Groupe Physico-Chimie Th\'eorique, ESPCI\\
39: 10 rue Vauquelin, F75231 Paris Cedex 05,
40: France\thanksref{a}}
41: \ead{sekimoto@turner.pct.espci.fr}
42: \thanks[a]{On leave from Universit\'e Louis Pasteur, Strasbourg, France.}
43: \begin{abstract}
44: By a modular combination of mesoscopic detectors and gates,
45: we present a thoughtful pump
46: that transports gas particles against the difference of
47: their density at the expense of the diffusion of another
48: species of gas particles.
49: We also discuss briefly the relevance of the model to the
50: study of structure-function relationship of the
51: biomolecular machines.
52: \end{abstract}
53:
54: \begin{keyword}
55: free energy conversion \sep detector \sep gate \sep
56: thermal fluctuation \sep control \sep stochastic energetics
57: \PACS 05.40.-a % Fluctuation phenomena, random processes, noise, and Brownian
58: % motion
59: \sep 05.70.Ln % Nonequilibrium and irreversible thermodynamics
60: \sep 34.10.+x % General theories and models of atomic and molecular collisions
61: % and interactions (including statistical theories, transition
62: % states, stochastic and trajectory models, etc.)
63: \sep 82.39.-k % Chemical kinetics in biological systems
64: % PACS \PACS code \sep code
65: \end{keyword}
66: \end{frontmatter}
67: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
68:
69:
70:
71:
72: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
73: \section{Introduction}
74: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
75: \label{sec:introduction}
76:
77: Biological molecular machines such as ion pumps, molecular motors (myosin,
78: F$_1$ motor, etc.) or signal transducers (G-proteins) have attracted
79: a lot of interest among physicists as well as biologists
80: \citep{cell}.
81: %
82: The molecular machines works under thermal
83: fluctuations, where the latter
84: serve as an energy source \citep{sekimoto97}
85: for the free energy transduction \citep{eisenberg85}\footnote{
86: Upon a single ATP hydrolysis reaction {\it in vivo},
87: about 70\% of the maximally available work (20\kT with
88: \kT$\simeq$ 4pN$\cdot$nm)
89: is ascribed to the change of the mixing entropy.}
90: as well as for the thermal activation \citep{Wong}.
91: %
92: Moreover, the cycle of free-energy transduction in
93: a single molecular machine requires no external control.
94: %
95: As many of these molecular machines share highly conserved
96: three-dimensional structural modules,
97: it is widely believed that they have been evolved
98: from a common ancestors \citep{ancestor}.
99: %
100: This fact suggests that such molecular machines may be
101: composed of a few modular structures each of which bears
102: some elementary functional role.
103: %
104: With recent rapid development of nano-biology
105: and the structural biology,
106: the biologists have begun to look for such modular
107: structure-function relationships
108: within a single molecular motors
109: \citep{higuchi02,KIF1A04,strain-detector-a}.
110: %
111:
112: From the theoretical physicists side, there is no surprise
113: about the mere fact that a molecular machine
114: without any special forward-backward symmetry
115: can move in one direction under non-equilibrium
116: conditions (e.g. under the ATP hydrolysis),
117: as stated in the Curie's principle \citep{curie}\footnote{
118: {\it ``Lorsque certaines causes produisent certains effets,
119: les \'el\'ements de sym\'etrie
120: des causes doivent se retrouver dans les effets produits.''}}.
121: %
122: With this in mind, the question of our interest is
123: how we can construct,
124: in combining well-defined simple functional modules,
125: an autonomously controlled
126: system that can work under thermal fluctuations.
127: %
128: Since the early works by \citet{feynman66} and \citet{buttiker87},
129: several models of autonomous free energy
130: transducer have been proposed, and their improvements
131: through the incorporation of a ``gate'' have been proposed
132: \citep{sakaguchi,derenyi99} (see also the review by \citet{reimann}).
133: %
134: These models were, however, rather {\it ad hoc} and the
135: modular nature of the construction was not clearly visible.
136: %
137: In the present paper,
138: we would construct a free energy transducer
139: with being more conscious about the modules and
140: taking the object oriented approach \citep{preprint}.
141: %
142: We start by introducing the concept of
143: {``semi-detectors''}, the detectors
144: that can perceive a certain external state, but not all, with
145: arbitrary sureness despite the thermal fluctuation
146: (\S~\ref{sec:detector}).
147: %
148: We then combine these semi-detectors together with
149: the gates, under the designing concept
150: that we call the {``bidirectional control''},
151: to realize an autonomous
152: particle transporter (\S~\ref{sec:transducer}).
153: %
154: Because of the limited space of this special issue,
155: the extensive discussion on the possible
156: relations to the real biological molecular machines
157: will be given elsewhere.
158:
159: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
160: \section{semi-detectors in the fluctuating world}
161: \label{sec:detector}
162: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
163:
164: The detector is the mechanism that correlates the states of the
165: outside to those of the inside of the system in question.
166: %
167: We focus on the case where the detection site admits a single
168: ligand particle coming randomly from the outside.
169: We will limit ourselves to the energetically ``neutral'' detections,
170: involving no change of the total energy.
171: %
172: Let us define the mapping from a
173: state of the outside, $x$, to a subset of the states of the inside,
174: $\Phi(x)$.
175: %
176: In the discrete representation,
177: the states of outside consists of {\sf IN} (i.e. a ligand
178: particle is on the detection site) and {\sf OUT} (otherwise),
179: while those states inside consist of {\sf ON} and {\sf OFF}.
180: The perfect detector would establish the mapping,
181: $\Phi({\sf IN})=\{ {\sf ON}\}$ and $\Phi({\sf OUT})=\{ {\sf OFF}\}$.
182: But these correspondences are too stringent to realize under
183: thermal fluctuations.
184: %
185: We will show below that there can be the physical
186: mechanisms which assure either one of the above correspondences.
187: %
188: We will call the mappings corresponding to such mechanisms
189: the semi-detectors.
190: %
191:
192: %
193: {\it Semi-detector of absence:} This module functions as the mapping
194: $\Phi_{\rm ab}(x)$ that prohibits only the output of
195: {\sf OFF} under the input of {\sf IN}. That is,
196: \beq \label{eq:absence{detector}}
197: \Phi_{\rm ab}(\mbox{{\sf IN}})=\{\mbox{{\sf ON}}\}, \quad
198: \Phi_{\rm ab}(\mbox{{\sf OUT}})=\{\mbox{{\sf ON}},\mbox{{\sf OFF}}\}.
199: \eeq
200: Then the {\sf OFF} surely indicates the
201: absence of the particle ({\sf OUT}).
202: %
203: Such aspect is
204: useful for repressive processes like the suppression of DNA
205: transcription by an repressor \citep{genes8}.
206: %
207: In the continuum representation, one may define $x$ as the position of
208: a ligand particle in a half space $0\!\le \!x \!< \!\infty $, where $x=0$
209: corresponds to the detection site (see Fig.~\ref{38up23.eps} {\it Left}).
210: %
211: The state of the inside, $a$, is assumed to be bounded, say,
212: in the region of $ -1\! \leq\! a\! \leq \! 0$ without loosing generality.
213: %
214: The semi-detector of absence may be realized
215: by the steric repulsion between the ligand
216: particle and a movable object
217: (thick bars in Fig.~\ref{38up23.eps} {\it Left}).
218: %
219: The total energy of the ligand-detector system, $U(x,a)$,
220: is written as $U(x,a)=0$ for $x\! \ge a \!+\! 1$ and $U(x,a)=\infty$ otherwise.
221: %
222: {\sf OFF} and {\sf ON} states correspond, respectively, to
223: $a\! \ge\! -1\!+\!\delta$ and $-1\! \le\! a
224: \! <\! -1\!+\!\delta$ with a small $\delta\! >\! 0$.
225: %
226: %%%%%%%%%%%%%%%%%%%%%%%%%%%
227: \begin{figure}[ht]
228: \hspace*{1cm}
229: \resizebox{12.00cm}{3.00cm}{\includegraphics*{38up23.eps}}
230: \caption{
231: {\it Left:}
232: Schematic representation of the semi-detector of absence in
233: a continuum representation.
234: The movable object (thick bar) and the ligand particle (thick dot)
235: are sterically excluding with each other.
236: $a$ represents the coordinate of the left extremity of the object.
237: $x=0$ is the detection cite.
238: {\it Right:}
239: The shadowed region on the $(x,a)$-plane
240: indicates the accessible phase space, where
241: $U(x,a)$ =0. }
242: \label{38up23.eps}
243: \end{figure}
244: %%%%%%%%%%%%%%%%%%%%%%%%%%%
245:
246: %
247: {\it Semi-detector of presence:} This module functions complementarily to
248: the former semi-detector. The mapping
249: $\Phi_{\rm pr}(x)$ that it defines prohibits only the output of
250: {\sf ON} under the input of {\sf OUT}. That is,
251:
252: \beq \label{eq:presence{detector}}
253: \Phi_{\rm pr}(\mbox{{\sf IN}})=\{\mbox{{\sf ON}},\mbox{{\sf OFF}}\}, \quad
254: \Phi_{\rm pr}(\mbox{{\sf OUT}})=\{\mbox{{\sf OFF}}\}.
255: \eeq
256: Then the {\sf ON} surely indicates the
257: presence of the particle ({\sf IN}).
258: %
259: Such aspect is useful for active processes like the
260: uptake of ATP by a molecular motor.
261: %
262: In the continuum representation,
263: the semi-detector of presence may be realized by the compensation
264: of a strong restoring potential for the movable object,
265: $M\phi(a)$ ($M\!\gg\! \kT$), by the
266: strong attractive interaction energy, $-M\phi(a-x)$,
267: between the ligand particle and
268: the movable object (see Fig.~\ref{38low24.eps} {\it Left}).
269: Here $\phi(z)$ is defined on $-\infty< \! z\!\le\! 0$ so that
270: $\phi(z)=0$ for $z\!\le\! -1$ and that $\phi(z)$ increases
271: monotonically from $\phi(-1)=0$ to $\phi(0)=1$.
272: %
273: %%%%%%%%%%%%%%%%%%%%%%%%%%%
274: \begin{figure}[ht]
275: \hspace*{1cm}
276: \resizebox{12.00cm}{3.00cm}{\includegraphics*{38low24.eps}}
277: \caption{{\it Left:}
278: Schematic representation of the semi-detector of presence.
279: $a$ and $x$ represent, respectively, the position of
280: the movable object (filled square) and that of the ligand particle
281: (thick dot).
282: Unless the particle is within
283: the proximity of the detection cite ($0\le x<\delta$) with
284: small $\delta$, the movable object is constrained at $a\simeq -1$
285: even under thermal agitations.
286: {\it Right:}
287: The accessible phase space
288: is indicated as a shadowed region, where
289: $U(x,a)\sim \kT$.
290: }
291: \label{38low24.eps}
292: \end{figure}
293: %%%%%%%%%%%%%%%%%%%%%%%%%%%
294: Energetically,
295: this movable object can take the value $a \ge -1+\delta$
296: only when the particle is near the binding site,
297: that is $0\le x\le \delta$ with a small $\delta>0$ so that
298: $M \phi(\delta) \ltsim \kT$\footnote{One can verify this by drawing
299: $M\{\phi(a)-\phi(a-x)\}$ {\it vs} $a$ for various values
300: of $x$.}.
301: %
302: Therefore {\sf ON} and {\sf OFF} correspond, respectively, to
303: $a\ge -1+\delta$ and $-1\le a < -1+\delta$.
304: %
305: The compensation mechanism similar to the one
306: discussed here has been discussed and called
307: {\it induced fit}
308: by \citet{koshland73} in the context of the
309: ligand binding by enzymes.
310: %
311:
312:
313: The above two semi-detectors establishes the
314: correlations between the outer world of the detectors
315: and the detectors themselves in the way that the semi-detectors
316: represent certain error-free information despite the thermal
317: fluctuations.
318: %
319: If the state variables of two semi-detectors of presence,
320: say $a_1$ and $a_2$, respectively, are coupled energetically
321: with each other,
322: the resulting system might function to realize the
323: {``cooperative
324: binding''} \citep{monod65,koshland66} or
325: the {``exchange of binding''} \citep{eisenberg85}, depending on
326: the nature of the coupling.
327: %
328:
329:
330: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
331: \section{Free-energy transducer based on the {semi-detector}s of
332: presence and the gates}
333: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
334: \label{sec:transducer}
335:
336: In order to focus on the subject of the autonomous control,
337: we would like to avoide energetic aspects as far as possible.
338: To this end, we will construct a pump of the ideal ``{\sf load}''(L)
339: gas particles from a dilute reservoir, $({\rm L},{\rm l})$,
340: to a dense reservoir, $({\rm L},{\rm h})$,
341: at the expense of the diffusion of the
342: ideal ``{\sf fuel}''(F) particles
343: from a dense reservoir, $({\rm F},{\rm h})$,
344: to a dilute reservoir, $({\rm F},{\rm l})$,
345: see Fig~\ref{16.eps}.
346: %%%%%%%%%%%%%%%%%%%%%%%%%%%
347: \begin{figure}[ht]
348: \hspace*{2.5cm}
349: \resizebox{7.00cm}{4.50cm}{\includegraphics*{16bw.eps}}
350: \caption{Schematic diagram of how the free-energy transducer
351: (the hexagon)
352: pumps up the {\sf load} particles at the expense of
353: the diffusion of the {\sf fuel} particles. See the text.}
354: \label{16.eps}
355: \end{figure}
356: %%%%%%%%%%%%%%%%%%%%%%%%%%%
357: If we denote the chemical potentials of these reservoirs by
358: $\mu_{\rm L,{\rm l}}$, $\mu_{\rm L,h}$, $\mu_{\rm F,h}$ and
359: $\mu_{\rm F,{\rm l}}$, respectively,
360: then the decrease of the total Gibbs' free energy
361: in the course of the transport of
362: $\Delta N_{L,{\rm l}\to{\rm h}}$ of the {\sf load} particles
363: at the expense of $\Delta N_{F,{\rm h}\to{\rm l}}$
364: of the {\sf fuel} particles is
365: $(\mu_{\rm F,h}-\mu_{\rm F,{\rm l}})\Delta N_{F,{\rm h}\to{\rm l}}- $
366: $ (\mu_{\rm L,h}-\mu_{\rm L,{\rm l}}) \Delta N_{L,{\rm l}\to{\rm h}}$.
367: %
368: Below we will present a completely symmetric model
369: in which the roles of the {\sf fuel} and the {\sf load} are
370: totally exchangeable depending on the relative magnitude of
371: $\mu_{\rm F,h}-\mu_{\rm F,{\rm l}}$ and
372: $\mu_{\rm L,h}-\mu_{\rm L,{\rm l}}$\footnote{Under this symmetry,
373: the currents of the active transport of the {\sf load} particles,
374: $J_{\rm L, {\rm l}\to{\rm h}}$, and that of the passive diffusion of
375: the {\sf fuel} particles, $J_{\rm F, {\rm h}\to{\rm l}}$,
376: are given in the linear non-equilibrium thermodynamics as
377: %
378: $J_{\rm L,{\rm l}\to{\rm h}}= $
379: $L_{\rm asy}\Delta \mu_{\rm asy}- L_{\rm sym}\Delta \mu_{\rm sym}$
380: and
381: $J_{\rm F,{\rm h}\to{\rm l}}=$
382: $L_{\rm asy}\Delta \mu_{\rm asy}+ L_{\rm sym}\Delta \mu_{\rm sym}.$
383: Here
384: $L_{\rm asy}\ge 0$ and $L_{\rm sym}\ge 0$
385: are the kinetic coefficients, and $\Delta \mu_{\rm asy}\equiv$
386: $\mu_{\rm F,h}-\mu_{\rm F,{\rm l}}-\mu_{\rm L,h}+\mu_{\rm L,{\rm l}}$
387: and $\Delta \mu_{\rm sym}\equiv$
388: $\mu_{\rm F,h}-\mu_{\rm F,{\rm l}}+\mu_{\rm L,h}-\mu_{\rm L,{\rm l}}$.
389: This pump can work almost reversibly near the stalled condition,
390: $\Delta \mu_{\rm asy}=0$, if $L_{\rm sym}$ is sufficiently small.}.
391:
392: The {\it cyclic} process in the above mentioned pump
393: requires at least two internal degrees of
394: freedom\footnote{This statement does not contradicts
395: with the fact that the existing models
396: of heat engine \citep{feynman66,buttiker87} have assumed
397: only one rotational degree of freedom to lift a load, since such
398: degree of freedom corresponds to two bounded degrees of freedom.}.
399: %
400: To construct such pump with
401: using the semi-detectors of presence as constituent modules,
402: %
403: the most natural way would be to
404: put one semi-detector on the {\sf fuel} side
405: (with the variable $a_{\rm F}$)
406: and the another one on the {\sf load} side
407: (with the variable $a_{\rm L})$,
408: and to have them to control the access of the particles
409: on the opposite sides,
410: via allosteric couplings \citep{monod65}, see Fig.~\ref{27.eps}.
411: %
412: %%%%%%%%%%%%%%%%%%%%%%%%%%%
413: \begin{figure}[ht]
414: \hspace*{2.5cm}
415: \resizebox{8.80cm}{4.00cm}{\includegraphics*{27.eps}}
416: \caption{{\it Left:} Scheme of the bidirectional control.
417: {\it Right:} Typical cycle realized by the
418: two independent internal state variables.
419: {\sf A}-{\sf F} correspond to those in Fig.~\ref{19new.eps}.}
420: \label{27.eps}
421: \end{figure}
422: %%%%%%%%%%%%%%%%%%%%%%%%%%%
423: %
424: We would call this scheme ``bidirectional control''.
425: %
426: We will describe in three steps the details of the model
427: which implements this scheme:\\
428: %
429: \hspace*{5mm} {\it Step-1: Introduction of the ``reaction'' coordinates
430: $\tilde{x}_{\rm L}$ and $\tilde{x}_{\rm F}$.}
431: %
432: Being inspired by Fig.~\ref{38low24.eps},
433: we introduce the convenient coordinate
434: $\tilde{x}$ ($0\le \tilde{x}<\infty$) which describes
435: a ligand particle {\it and} a semi-detector of presence:
436: %
437: $x=\max(\tilde{x}-1,0),\quad$ $ a = -\min(\tilde{x},1)$.
438: %
439: Then we extend this definition of $\tilde{x}$ to the case of
440: two particle reservoirs:
441: %
442: We assume that the high-density [low-density] reservoir
443: occupies the half space $x>0$ [$x<0$], respectively.
444: %
445: We redefine the mapping $\tilde{x}\mapsto (x,a)$
446: so that $\tilde{x}$ can take the values on the entire axis,
447: $-\infty<\tilde{x}<\infty$,
448: with:
449: \beq \label{eq:xtilde}
450: x=({\tilde{x}}/{|\tilde{x}|})\,\max(|\tilde{x}|-1,0),
451: \quad a = -\min(|\tilde{x}|,1).
452: \eeq
453: %
454: A ligand particle is on the detection site if
455: $|\tilde{x}| \le 1$,
456: in the high-density reservoir if $\tilde{x} > 1$, and
457: in the low-density reservoir if $\tilde{x} < -1$.
458: %
459: We interpret $\tilde{x}$ so that the region of
460: $|\tilde{x}|\le 1-\delta$ with small $\delta >0$
461: corresponds to the state {\sf ON} of the semi-detector,
462: while $|\tilde{x}| > 1-\delta$
463: corresponds to the state {\sf OFF}.
464: %
465: Finally we apply this type of mapping for both the
466: {\sf load} side ($\tilde{x}_{\rm L}$) and the {\sf fuel} side
467: ($\tilde{x}_{\rm F}$).
468: %
469: The coordinate plane $(\tilde{x}_{\rm L},\tilde{x}_{\rm F})$
470: can then represent the positioning of one representative
471: {\sf load} particle {\it and} the one {\sf fuel} particle
472: together with the states of the semi-detectors of presence,
473: $a_{\rm L}$ and $a_{\rm F}$\footnote{By the introduction of
474: $\tilde{x}_{\rm L}$ and $\tilde{x}_{\rm F}$, we can
475: avoid to be lost in the four-dimensional space.
476: %
477: The trade-off that we pay is that
478: any value of $a_{\rm F} > -1$ is represented by
479: the two equivalent points,
480: $\tilde{x}_{\rm F}=\pm a_{\rm F}$.
481: The same holds also on the {\sf load} side.}.\\
482: %
483: \hspace*{5mm} {\it Step-2: Definition of the gates' action.}
484: For the {\sf fuel} particles, we construct the gate which allows
485: the access of the particles exclusively from one of their
486: reservoirs at a time.
487: %
488: %%%%%%%%%%%%%%%%%%%%%%%%%%%
489: \begin{figure}[ht]
490: \hspace*{3.5cm}
491: \resizebox{7cm}{4.75cm}{\includegraphics*{36.eps}}
492: \caption{ {\it Top:} Potential profile of the gate that opens
493: to the {\sf fuel} particles in their low-density reservoir.
494: {\it Bottom:} Similar to the above but the accessibility is for
495: the high-density side.
496: The thick horizontal bars indicate the detection site for the
497: {\sf fuel} particles. }
498: \label{36.eps}
499: \end{figure}
500: %%%%%%%%%%%%%%%%%%%%%%%%%%%
501: %
502: In the {\it Top} of Fig.~\ref{36.eps},
503: the potential barrier (the hight $\gg \kT$) is established
504: at $\tilde{x}_{\rm F}=1+\epsilon$ with a small $\epsilon>0$,
505: so that the access of the {\sf fuel} particles
506: from the high-density reservoir is blocked
507: at a distance $\epsilon$ off
508: the detection site ($|\tilde{x}_{\rm F}|\le 1$).
509: %
510: The same architecture is defined for the {\sf load} particles.
511: %
512: Similarly the gate represented in
513: the {\it Bottom} of Fig.~\ref{36.eps}
514: blocks the access of the {\sf fuel} particles from
515: the low-density reservoir at the distance $\epsilon$ off
516: the detection site.\\
517: %
518: \hspace*{5mm} {\it Step-3: Coupling of the semi-detectors to the gates.}
519: We define the bidirectional control
520: by the following symmetric rules:\\
521: %
522: %%%%%%%%%%%%%%%%%%%%%%%%%%%
523: \begin{figure}[ht]
524: \hspace*{2.5cm}
525: \resizebox{9.00cm}{4.00cm}{\includegraphics*{17bw.eps}}
526: \caption{
527: {\it Top:} The control of the {\sf load} particles by the detection
528: of the {\sf fuel} particles.
529: %
530: %
531: {\it Bottom:} The control of the {\sf fuel} particles by the detection
532: of the {\sf load} particles.
533: %
534: The {\sf J}-shaped symbol in, for example,
535: the {\it Top-Left} indicates the exclusive accessibility
536: from the low-density reservoir of the {\sf load} particles,
537: as indicated in the {\it Top} of Fig.~\ref{36.eps}.
538: The other cases would be understood similarly.}
539: \label{17.eps}
540: \end{figure}
541: %%%%%%%%%%%%%%%%%%%%%%%%%%%
542: {[Control by $a_{\rm F}$]} %
543: If a {\sf fuel} particle is detected, i.e.
544: if $|\tilde{x}_{\rm F}| \le 1-\delta$
545: ({\it Top-Left} of Fig.~\ref{17.eps}),
546: only the {\sf load} particles in the low-density reservoir
547: ($\tilde{x}_{\rm L}<-1$) can access to their detection site.
548: %
549: If a {\sf fuel} particle is not detected i.e., if
550: $|\tilde{x}_{\rm F}| > 1-\delta$ ({\it Top-Right}),
551: then only the {\sf load} particles in the high-density reservoir
552: ($\tilde{x}_{\rm L}> -1$) can access to their detection site.\\
553: %
554: {[Control by $a_{\rm L}$]}: %
555: If a {\sf load} particle is detected, i.e. if $|\tilde{x}_{\rm L}|
556: \le 1-\delta$ ({\it Bottom-Left} of Fig.~\ref{17.eps}),
557: only the {\sf fuel} particles in the low-density reservoir
558: ($\tilde{x}_{\rm F}<-1$) can access to their detection site.
559: %
560: If a {\sf load} particle is not detected, i.e. if
561: $|\tilde{x}_{\rm L}|> 1-\delta$ ({\it Bottom-Right}),
562: then only the {\sf fuel} particles in the high-density reservoir
563: ($\tilde{x}_{\rm F}> -1$) can access to their detection site.
564:
565: The consequence of these simple and symmetric combination
566: of the semi-detectors and the gates is
567: immediately seen by the graphical representation
568: in Fig.~\ref{19new.eps}.
569: %
570: %%%%%%%%%%%%%%%%%%%%%%%%%%%
571: \begin{figure}[ht]
572: \hspace*{4.0cm}
573: \resizebox{5.00cm}{5.00cm}{\includegraphics*{19new.eps}}
574: \caption{
575: In the $(\tilde{x}_{\rm L}, \tilde{x}_{\rm F})$-plane,
576: the thick horizontal and vertical bars
577: represent the presence of the potential barriers
578: either for the {\sf fuel} particles (the horizontal bars) or for the
579: {\sf load} particles (the vertical bars).
580: %
581: The shaded regions are where the bidirectional control prohibits
582: to access except for a small ``leak'' due to the finite values of
583: $\epsilon$ and $\delta$. See the text for the details.
584: %
585: The dashed curve shows one representative process (cf.
586: the footnote on the multiplication of the state points).
587: $B$-$E$ on the curve correspond roughly to those in
588: Fig~\ref{27.eps}.
589: }
590: \label{19new.eps}
591: \end{figure}
592: %%%%%%%%%%%%%%%%%%%%%%%%%%%
593: There, the potential barriers of the gates
594: are indicated by the thick horizontal or vertical bars.
595: %
596: For example, the short horizontal bar at $\tilde{x}_{\rm F}=1+\epsilon$
597: in the range of $|\tilde{x}_{\rm L}|<1-\delta$
598: represents the blockade of the access of the {\sf fuel} particles
599: from their high-density reservoir.
600: %
601: We see how the combination of those bars
602: organizes a broad passageway joining the second and the forth
603: quadrants on the $(\tilde{x}_{\rm L},\tilde{x}_{\rm F})$-plane.
604: %
605: This passageway corresponds to the diffusion
606: of a single {\sf fuel} particle from its high-density reservoir to the
607: low-density one {\it accompanying} the active transport of a
608: single {\sf load} particle from its low-density reservoir to
609: the high-density one.
610: %
611: Therefore, if we could neglect the small gaps between the bars
612: due to the finite values of $\epsilon$ and $\delta$, this pump
613: would work tightly with $\Delta N_{F,{\rm h}\to{\rm l}}=$
614: $\Delta N_{L,{\rm l}\to{\rm h}}$, whose mean flow direction
615: is determined by the sign of
616: $(\mu_{\rm F,h}-\mu_{\rm F,{\rm l}})-$
617: $(\mu_{\rm L,h}-\mu_{\rm L,{\rm l}})$.
618: %
619: Actual autonomous pump seems not be able to
620: avoid the ``leak'' due to the above mentioned gaps,
621: but this leak would be of importance only near the stalled state
622: ($\mu_{\rm F,h}-\mu_{\rm F,{\rm l}}
623: =$ $\mu_{\rm L,h}-\mu_{\rm L,{\rm l}}$)
624: as far as $\epsilon$ and $\delta$ are small\footnote{In
625: the context of the the linear non-equilibrium thermodynamics,
626: we expect that $L_{\rm sym}/L_{\rm asy}$ is at most of
627: the order of $\epsilon$ and $\delta$.}.
628: %
629:
630: How would the pump thus constructed
631: looks like for an observer who can survey
632: only the {\sf load} particles?
633: %
634: Let us suppose that $(\mu_{\rm F,h}-\mu_{\rm F,{\rm l}})-$
635: $(\mu_{\rm L,h}-\mu_{\rm L,{\rm l}})$ is positive and enough
636: greater than \kT.
637: %
638: As long as the detection site of the {\sf load} particle is empty,
639: this site is almost always accessible from the low-density reservoir
640: of the {\sf load} particle ({\sf B} along the dashed curve in
641: Fig.~\ref{19new.eps}).
642: %
643: When a {\sf load} particle arrives at the detection site
644: {\it and} is detected ({\sf C,D}), the gate
645: is very likely to reverse its accessibility so that
646: the {\sf load} particle on the detection site can leave
647: now for the high-density reservoir ({\sf E}).
648: %
649: Then as soon as the {\sf load} particle quits the detection
650: site ({\sf F}), the gate comes back to the
651: initial conformation ({\sf B}).
652: %
653: Thus the pump behaves as if it responded by itself to the arrival
654: and the leave of the {\sf load} particle.
655:
656:
657: %\clp
658: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
659: \section{Discussion}
660: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
661: \label{sec:prediction}
662: %
663: We have constructed theoretically an autonomous system
664: that works under thermal fluctuations.
665: %
666: The errors of the detection was avoided
667: by introducing what we call the semi-detectors.
668: %
669: The concept of the semi-detectors on mesoscopic scale
670: might also be of interest in the context of
671: the mesoscopic devices \citep{imry}.
672: %
673: The cyclic process was enabled by what we call
674: the bidirectional control.
675: %
676: The latter idea is applicable also to the macroscopic
677: autonomous processes.
678: %
679: For example, in the operation protocols of vending machines
680: or of pay-phones,
681: the gate on one side (taking up the money) and
682: that on the other side (rendering goods or
683: services in exchange) are controlled by
684: their respective detectors on the opposite sides.
685: %
686:
687: The biological molecular motors are not as simple as
688: we have discussed.
689: %
690: Still, the present analysis might help as a reference frame
691: when we look for the structure-function relationship
692: in those systems.
693: %
694: For example, we might ask if
695: the hydrolysis reaction of ATP ({\sf fuel})
696: in a molecular motor corresponds to changing the accessibility
697: from the reservoir of the ATP to that of
698: the ADP and the inorganic phosphate [Pi].
699: %
700: Also we might wonder if the putative Pi-sensors
701: (``switch loop'' etc., \citet{Vale1996,gilbert}) can be compared
702: to the semi-detector of the ligand in our model.
703: %
704: As for the kinetics of the interaction between
705: the molecular motor and its counterpart filament ({\sf load}),
706: we might ask if the motors have
707: a degree of freedom to detect its own mechanical strain
708: \citep{strain-detector-a,ks-prive},
709: in addition to the degree of freedom to control the
710: binding to the filament \citep{KIF1A04}.
711: %%%
712: %%%%%%%%%%%%%%%%%%%%%%%%%%%
713: \begin{figure}[ht]
714: \hspace*{1.5cm}
715: \resizebox{10.50cm}{5.55cm}{\includegraphics*{29red.eps}}
716: \caption{The pump of the ``wild-type''
717: (left, see Fig.~\protect\ref{19new.eps}) and
718: a ``mutant'' with blocked fuel reactions (right). See the text.}
719: \label{29red.eps}
720: \end{figure}
721: %%%%%%%%%%%%%%%%%%%%%%%%%%%
722: As a qualitative prediction, we might mention
723: the possibility of the ``mutants'' in which
724: the degree of freedom related to the
725: detection of, say, the {\sf load} particles,
726: $a_{\rm L}$, is immobilized, as shown in Fig.~\ref{29red.eps}.
727: %
728: In this case,
729: the rate of passive diffusion of the {\sf load} particles
730: should depend on the
731: density of the (non-consumed) {\sf fuel} particles in
732: their high-density reservoirs.
733: %
734:
735: I would like to thank
736: K. Kitamura, H. Miyashita, E. Muneyuki, H. Noji,
737: I. Ojima, Y. Okada, Y. Oono, and K. Sutoh
738: for stimulating discussions.
739: It is my great pleasure to dedicate this paper to Y. Kuramoto.
740: %
741:
742: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
743:
744:
745: % Parenthetical: \citep{Bai92} produces (Bailyn 1992).
746: % Textual: \citet{Bai95} produces Bailyn et al. (1995).
747: % An affix and part of a reference: \citep[e.g.][Ch. 2]{Bar76}
748: % produces (e.g. Barnes et al. 1976, Ch. 2).
749:
750: \begin{thebibliography}{}
751:
752: %
753: \bibitem[Alberts el al.(1998)]{cell}
754: Alberts, B., Bray, D., Johnson, A., Lewis, J., Raff, M.,
755: Roberts, K. \& Walter, P.
756: {\it Essential Cell Biology: An introduction to the molecular biology
757: of the cell} (1998), \S 12.
758: New York: Garland Pub. Inc.
759:
760: %
761: \bibitem[Brendza et al.(2000)]{gilbert}
762: Brendza, K.M., Sontag, C.A., Saxton, W.M., %\&
763: Gilbert, S.P. (2000)
764: %{A kinesin mutation that uncouples motor domains
765: %and desensitizes the $\gamma$-Phosphate {sensor}}.
766: J. Biol. Chem. {\bf 29}, 22187-22195.
767: %Biophys. J. {\bf 75}, 646-661.
768:
769: %
770: \bibitem[B\"uttiker et al.(1987)]{buttiker87}
771: B\"uttiker, M. (1987)
772: %{Transport as a consequence of state-dependent diffusion}.
773: Zeitschrift der Physik B {\bf 68}, 161-167,
774:
775: %
776: \bibitem[Curie(1894)]{curie}
777: P. Curie (1894)
778: %{\it Sur la sym\'etrie dans les ph\'enom\`enes physiaues,
779: %sym\'etrie d'un champ \'electriaue et d'un champ magn\'etique},
780: J. Phys. (Paris) 3\`eme s\'erie, {\bf 3}, 393-415.
781:
782: %
783: \bibitem[Der\'enyi and Astumian(1999)]{derenyi99}
784: Der\'enyi, I. \& Astumian, R.D. (1999)
785: %{Efficiency of Brownian heat engines}.
786: Phys. Rev. E. {\bf 59}, R6219-22.
787:
788: %
789: \bibitem[Duke(1999)]{duke99}
790: Duke, T.A.J. (1999)
791: %{Molecular model of muscle contraction}.
792: Proc. Natl. Acad. Sci. USA {\bf 96}, 2770-2775.
793:
794: %
795: \bibitem[Eisenberg and Hill(1985)]{eisenberg85}
796: Eisenberg, E. \& Hill, T.L. (1985)
797: %{Muscle Contraction and Free Energy Transduction in Biological Systems}.
798: Science, {\bf 227}, 999-1006.
799:
800: %
801: \bibitem[Feynman et al.(1966)]{feynman66}
802: Feynman, R.P., Leighton, R.B., \& Sands, M. (1966)
803: {\it The Feynman Lectures on Physics}
804: Vol.I, Chap.46. Addison-Wesley, Reading, MA.
805:
806: \bibitem[Higuchi and Endow(2002)]{higuchi02}
807: Higuchi, H. and Endow, S. A. (2002)
808: %{Directionality and processivity of molecular motors}.
809: Current Opinion Cell Biol. {\bf 14}, 50-57.
810:
811: %
812: \bibitem[Imry(1997)]{imry}
813: Imry, Y. (1997)
814: {\it Introduction to mesoscopic physics}.
815: Oxford University Press, UK.
816:
817: %
818: \bibitem[Koshland(1973)]{koshland73}
819: Koshland, D. E. (1973)
820: %{Protein shape and biological control}.
821: Sci. Amer. {\bf 229} 52-64.
822: %Proc. Natl. Acad. Sci. USA {\bf 44}, 98-104.(1958)
823: %Science {\bf 142}, 3599 (1963). induced fit
824: %
825:
826: \bibitem[Koshland et al.(1996)]{koshland66}
827: Koshland, D.E., Nemethy, G., \& Filmer, D. (1966)
828: %{Comparison of experimental binding data and theoretical models
829: %in proteins containing subunits}.
830: Biochem. {\bf 5}, 365-85.
831:
832: \bibitem[Kull et al.(1998)]{ancestor}
833: Kull, F.J., Vale R.D. \& Fletterick, R.J. (1998)
834: %{The case for a common ancestor: kinesin and myosin motor
835: %proteins and G proteins}.
836: {J. Muscle Res. Cell Motil.} {\bf 97}, 877-886.
837:
838: \bibitem[Lewin(2003)]{genes8}
839: Lewin, B. (2003) {\it Genes VIII}. Prentice Hall.
840:
841: \bibitem[Monod et al.(1965)]{monod65}
842: Monod, J, Wyman, J, \& Changeux, J.P. (1965)
843: %{On the nature of allosteric transitions: a plausible model}.
844: J. Mol.Biol. {\bf 12}, 88-118.
845:
846: \bibitem[Nitta et al.(2004)]{KIF1A04}
847: Nitta, R., Kikkawa, M., Okada, Y. \& Hirokawa, N. (2004)
848: %{KIF1A alternately uses two loops to bind microtubules}.
849: Science {\bf 305}, 678-683.
850:
851:
852: \bibitem[Reimann(2002)]{reimann}
853: Reimann, P. (2002)
854: %{\it Brownian motors: noisy transport far from equilibrium},
855: Physics Reports {\bf 361}, 57-265 [condmat/0010237].
856:
857: \bibitem[Sakaguchi(1998)]{sakaguchi}
858: Sakaguchi, H. (1998)
859: %{A Langevin simulation for the Feynman
860: %ratchet model}.
861: J. Phys. Soc. Japan {\bf 67}, 709-712.
862:
863: \bibitem[Sekimoto(1997)]{sekimoto97}
864: Sekimoto, K. (1997)
865: %{Kinetic Characterization of Heat Bath and
866: %the Energetics of Thermal Ratchet Models}.
867: J. Phys. Soc. Japan {\bf 66},1234-1237.
868:
869: \bibitem[Sekimoto(2000)]{ks-prive}
870: Sekimoto, K. (2000) private note to S. Esaki.
871:
872: \bibitem[Sekimoto(2000)]{preprint}
873: Sekimoto, K. (2000)
874: A version of this paper has been presented
875: % as
876: %`{\it Control and Sensing in Thermally Fluctuating Systems}',
877: at the International Conference on
878: Dynamical Aspects of Complex Systems from Cells to Brain
879: at Sendai, Japan.
880:
881: \bibitem[Tsiavaliaris et al.(2004)]{strain-detector-a}
882: Tsiavaliaris, G., Fujita-Becker, S. \& Manstein, D. J. (2004)
883: %{Molecular engineering of a backwards-moving myosin motor}.
884: Nature {\bf 427}, 558-561.
885:
886: \bibitem[Vale(1996)]{Vale1996}
887: Vale, R.D. (1996),
888: %{\it Switches, Latches and Amplifiers: Common Themes of
889: %G Proteins and Molecular Motors}.
890: J. Cell Biol. {\bf 135}, 291-302.
891: %(The ``switch'' in this reference would often correspond to
892: %the ``{detector}'' in the present paper.)
893:
894: \bibitem[Wong et al.(1997)]{Wong}
895: Wong, J. Y., Kuhl, T. L., Israelachvili, J. N., Mullah, N., Zalipsky, S.
896: (1997)
897: %{Direct measurement of a tethered ligand-receptor interaction potential}.
898: Science {\bf 7} 820-822.
899:
900:
901: \end{thebibliography}
902: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
903:
904: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
905:
906: %\tableofcontents
907: \end{document}
908: