You are on page 1of 5

Messages diffuse faster than messengers

Bernardo Pando*, Silvina Ponce Dawson‡, Don-On Daniel Mak§, and John E. Pearson¶ʈ
*Department of Physics, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA 02139; ‡Departamento de Fı ´sica, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad Universitaria, Pabello ´ n 1, 1428 Buenos Aires, Argentina; §Department of Physiology, University of Pennsylvania, B400 Richards Buildings, 3700 Hamilton Walk, Philadelphia, PA 19104-6085; and ¶Theoretical Biology and Biophysics, Los Alamos National Laboratory, Los Alamos, NM 87545 Edited by Joseph Schlessinger, Yale University School of Medicine, New Haven, CT, and approved January 24, 2006 (received for review November 3, 2005)

In many cell-signaling pathways, information is transmitted by the diffusion of messenger molecules. Diffusion coefficients characterize the messenger’s spatial range and the characteristic times of signal propagation. Inside cells, particles usually diffuse in the presence of immobile binding sites (or traps). It is well known that binding to traps results in an effective diffusion coefficient that is smaller than the free coefficient in media free of traps. To measure effective diffusion coefficients in cells, ‘‘tagged’’ particles are often used. Radioactive calcium was used in a giant squid axon and in cytosolic extracts of Xenopus laevis oocytes. Fluorescence recovery after photobleaching yields diffusion coefficients from observations of the distribution of fluorescently labeled proteins. In the absence of traps, free diffusion coefficients give both the rate at which single-particle mean square displacements increase and the rate at which information in the form of inhomogeneities in particle concentration spread out with time. We show here that, in the presence of traps, information diffuses faster than single particles. Thus, messages diffuse faster than messengers. Tagged-particle experiments give the single-particle diffusion coefficients and, thus, can underestimate the rate of diffusive signal propagation.
binding ͉ effective diffusion ͉ fluorescence recovery after photobleaching ͉ tagged particles ͉ traps

t is well known that binding to traps results in an effective diffusion coefficient that is smaller than the free coefficient in media free of traps (1). To measure effective diffusion coefficients in cells, ‘‘tagged’’ particles are often used. Radioactive calcium was used in a giant squid axon (2) and in cytosolic extracts of Xenopus laevis oocytes (3). Fluorescence recovery after photobleaching (FRAP) yields diffusion coefficients from observations of the distribution of fluorescently labeled proteins (4–10). In the absence of traps, free diffusion coefficients give both the rate at which single-particle mean square displacements increase and the rate at which information in the form of inhomogeneities in particle concentration spread out with time. We show here that, in the presence of traps, information diffuses faster than single particles and that messages diffuse faster than messengers. Tagged-particle experiments give the single-particle diffusion coefficients and, thus, can underestimate the rate of diffusive signal propagation. To illustrate these effects, we consider the simplest model that incorporates traps and freely diffusing particles. The free particles, P f, bind to an immobile substrate, S , resulting in bound particles, P b, according to the simple reaction scheme L ; P b, Pf ϩ S |
k off kon


Results For specificity we use pseudo-on (kon Seq) and off rates and diffusion coefficient for GFP-tagged glucocorticoid receptors in the nuclei of mouse adenocarcinoma cells (10). The first simulation (see Methods) (Fig. 1A) mimics experiments (2, 3) in which radioactive calcium (45Ca2ϩ) is added to an essentially one-dimensional medium. In the simulation, a bolus of tagged particles is added to the leftmost 1.1 ␮m of the medium in which identical but untagged particles are already in equilibrium with the traps. The equilibrium concentration of free (bound) particles prior to the addition of the bolus is denoted Pbeq. Fig. 1 shows the concentration profiles at t ϭ 0.1 s after u t t evolution from initial concentrations (Pf , Pu b , Pf, Pb) ϭ (Pfeq, u u t t Pbeq, 0.2Pfeq, 0) for x Յ 1.1 ␮m and (Pf , Pb , Pf, Pb) ϭ (Pfeq, Pbeq, 0, 0) for x Ͼ 1.1 ␮m. The deviation of the particle concentrations from Pfeq of total free (thick solid line), total tagged (light solid line), tagged free (dotted line), and untagged free particles (dashed line) are plotted as fractions of Pfeq. The spread of the untagged particles is clearly greater than that of the tagged particles. In Fig. 1B, the deviation from background t u of free particles (Pf ϩ Pf Ϫ Pfeq) is shown as a solid line representing what a calcium-sensitive electrode would record at the MP, a measurement point some distance away from the point at which the bolus is added (IP). In the same graph, the t total tagged particle concentration, Pf ϩ Pt b is also plotted as a dotted line representing the radioactivity observed at the MP. Clearly, the total free-particle concentration is growing much faster there than the tagged particle concentration (the radioactivity). As proven in Methods and illustrated in Fig. 1B, the time it takes for the total free particle concentration at the MP to reach the same threshold value above Pfeq is shorter. The reason is that, as tagged particles diffuse, they compete with the untagged ones for traps. The net effect is that there is a preferential unbinding of untagged particles and binding of tagged ones, because the tagged ones are diffusing into a region with no tagged particles. These additional untagged particles that are released throughout the medium result in the threshold value being reached faster at the MP. The fact that all tagged particles observed at the MP travel the entire distance from IP to MP, in contrast, implies that the effective diffusion coefficient for the population of tagged particles is given by the single-particle effective diffusion coefficient. It follows that the time it takes for inhomogeneities to diffuse throughout the medium is much shorter than the singleparticle diffusion time. The experimental observable is the t ϩ Pt total tagged-particle concentration Pf b, whereas the quantity of interest for cell signaling is the free-particle t u deviation from background: Pf ϩ Pf Ϫ Pfeq. As illustrated in
Conflict of interest statement: No conflicts declared. This paper was submitted directly (Track II) to the PNAS office. Freely available online through the PNAS open access option. Abbreviations: MP, measurement point; IP, point at which the bolus is added; FRAP, fluorescence recovery after photobleaching.


with dissociation constant, K D ϭ k off/ k on. We illustrate the differences between single-particle diffusion and diffusion of inhomogeneities with a sequence of numerical simulations of the appropriate reaction–diffusion equations for all four classes of particles: tagged and untagged bound (P t b and t u Pu b , respectively) and free (P f and P f , respectively).
5338 –5342 ͉ PNAS ͉ April 4, 2006 ͉ vol. 103 ͉ no. 14

whom correspondence should be addressed. E-mail:

© 2006 by The National Academy of Sciences of the USA͞cgi͞doi͞10.1073͞pnas.0509576103

1 ␮m and (Pf. we plot the results of two three-dimensional simulations (see Methods).75 already at t ϭ 1 s.1 s. respectively. left scale). Pb. the measured diffusion coefficients (Dmeas) can be obtained from t1/2 by Dmeas ϭ 2 ͞(9. (B) Time course of particle concentrations at MP. and they agree with Eqs. Seq S eq 1ϩ KD ST [3] PNAS ͉ April 4. First. Pf . 1 A and B. 2006 ͉ vol. Pbeq) for r Ͼ 1. In particular. for values of k off and k on that are 100 times smaller t u than those used in Fig. As shown in Methods.98t1/2). and t ϭ 3 s. left axis). much slower than the spread of inhomogeneities. t t The dotted line denotes the total tagged-particle concentration. there were no tagged particles and the initial conditions were (Pf. Pf u ϩ Pf Ϫ Pfeq.1 ␮m. Fig. and the solid line corresponds to Seq͞ST ϭ 0.1 s. Pbeq) for r Յ 1. which is proven in Methods and illustrated in Fig. in an experiment in which a bolus of tagged particles (such as radioactive calcium) is released at t ϭ 0. the total tagged population spreads at the same rate as the free tagged particles. Pf ϩ Pf Ϫ Pfeq (thick solid). The slopes of these curves are the effective diffusion coefficients. We observe similar results.1 ␮m is the radius of the perturbed Ro region. 1 A and B changing k off and k on but keeping K D and all other parameters constant. and Pf in the FRAP simulation (solid). and t the solid line denotes total free particle concentration above background.1 ␮m. 2.8Pfeq. u t t u u Pb ) ϭ (0.Fig. A ‘‘particle deficit experiment’’ is also shown in which the free particle concentration was reduced to Pf ϭ 0. 2 and 3. (B) Dt͞Du as a function of Seq͞KD. Pf in the particle-deficit t simulation (dotted). Furthermore. P f ϩ Pf Ϫ P feq is 1. 1 A.1 ␮m were untagged at t ϭ 0. 0) for r Ͼ 1. The coefficients thus obtained compare well with the ones deduced in Methods (see Eqs. The ratio of mean square b f displacements is 1. In Fig. Visual comparison of the widths of the concentration peaks and troughs in Fig. (A) Onedimensional simulation of the spread of tagged and untagged particles. 1D shows the integrated recovery curves for the particle-deficit (dashed line) and FRAP-like (solid line) simulations. We repeated the simulations of Fig. Seq 1ϩ KD [2] in all the experiments discussed here. raising the concentration of free particles by 20% in the leftmost 1.097 s for the f luorescence recovery in the FRAP-like simulation. (D) Percent recovery as a function of time for the FRAP-like (solid line) and particle deficit (dashed line) simulation (see Methods). In the t u . MP is the measurement point from which the signals shown in B were obtained. The four lines shown are for Pf ϩ Pf (dashed) and Pf (dashed– u dotted) in the tagged particle bolus simulation. P f that the difference in the rate of spread occurs even before the spreading becomes diffusive (which always occurs in the long time limit in an infinite domain).1 s (solid line. the spread of tagged particles is determined by the same effective diffusion coefficient Dt ϭ Df . The plot shows the free particle recovery at t ϭ 0. 0. 0. Note that the concentration is almost fully recovered in the particle deficit case but not in the FRAP-like case. In the particle-deficit simulation. 14 ͉ 5339 BIOPHYSICS APPLIED MATHEMATICS . whereas the particles in the rest of the 5-␮m t sphere remained tagged. the ratio approaches its asymptotic value: Seq͞ST. our simulations show particle concentration. a FRAP-like simulation is shown in which all particles in a sphere of radius 1. P f . 1 A and B reveals that the effective diffusion coefficient obtained in the FRAP-like simulation is identical to the tagged-particle effective diffusion coefficient obtained before. the initial conditions were (Pf b. Both simulations were performed in a sphere of radius 5 ␮m.4 times t larger than P t ϩ P at MP. 1C).3. the mean square displacement (divided by twice the space dimension) is plotted as a function of t for various quantities in the three simulations discussed. At this time. (A) (͗ x2 ͘ Ϫ ͗ x0 ͘)͞(2d)vst (where d is the space dimension: d ϭ 1 in the tagged-particle bolus simulation discussed in Fig. 2. right axis). Pb) ϭ (Pfeq. 2. Pb ) ϭ (Pfeq. Pf (dotted). For this geometry.1. The dotted line corresponds to Seq͞ST ϭ 0.1 ␮m and (Pf.1 s (dotted line. it spreads at a slower rate than a local inhomogeneity of the unlabeled u . and Pb ϩ Pf (thin solid) at t ϭ 0. denoted by superscripts t and u. The spread of a bolus of untagged particles. The first represents a FRAP-like simulation (solid line. Pbeq) for r Յ 1. Pf in the particle-deficit simulation is spatially uniform to within 3% of its final value. but for later times as k off decreases. The plot shows the recovery of tagged particles. right scale). The t1/2 is 0.022 s for the particle-recovery simulation and 0. Pb ϩ Pf . and the second represents a particle-deficit simulation (dotted line. The slopes of these lines give the effective diffusion coefficients. Therefore. S eq͞K D is the ratio between the mean time a particle stays bound and the mean time between successive trappings (11). The plots show the concentrations as a percentage of the final equilibrium concentrations at t ϭ 0. the dashed line corresponds to Seq͞ST ϭ 0.9. Simulated diffusion of particles in the presence of traps. Pfeq. Pf t ϩ Pb at t ϭ 0. Pbeq. whereas the FRAP-like simulation still shows deviations of 80% from its final equilibrium value. 1C. where Ro ϭ 1. 2 and 3). Fig. 1. The four curves u t u t t t are Pf Ϫ Pfeq (dashed). D t is also the effective diffusion coefficient of a single particle in the medium. (C) Two three-dimensional simulations in a 5-␮m sphere. For large Seq͞KD. is determined by the coefficient Du ϭ Df . Effective diffusion coefficients.1-␮m sphere at t ϭ 0. 2 Fig. Pando et al. In Fig. 1 A. Pt FRAP-like experiment. Pb) ϭ (0. however.8Pfeq in the 1. The black dot corresponds to the parameters used in all simulations in this report. and d ϭ 3 in the FRAP and particle deficit experiments discussed u t t in Fig. 103 ͉ no. Fluorescence in FRAP experiments spreads at the same rate as a bolus of tagged particles that is added to a background of untagged particles and traps in equilibrium.1 ␮m.

tagged particles will compete with untagged ones for the traps. In this case. The difference is largest for large S eq͞K D and small S eq͞S T. The total binding site concentration ST and equilibrium concentration of unoccupied sites Seq were not reported. only a few particles can be tracked at a time. We used the pseudo-on and off rates reported in ref.1. ST. k off. In this case. We used Seq ϭ 1 mM and ST ϭ 10 mM. Analytical Derivation of Effective Diffusion Coefficients.1728 mM. Seq ϭ 1 mM. 2 B) or if they are almost completely occupied (Seq ϭ {Pbeq͞Pfeq} KD Ͻ Ͻ KD as represented by the curves near the left side of Fig. 15). Df. In the past few years. the time evolution of the concentrations is given by Ά i ѨPf i i j i ϭ D fٌ 2P f Ϫ kon P f ST Ϫ Pb ϩ k off P b Ѩt j . In the ‘‘pure-diffusion’’ regime. which describe the behavior of single or tagged molecules. as in the ‘‘full model’’ and ‘‘effective-diffusion’’ regimes defined in ref. the tagged particle measurement gives an accurate assessment of the rate at which inhomogeneities smear out (Dt Ϸ Du Ͻ Ͻ Df in the former case and Dt Ϸ Du Ϸ Df in the latter case). it is not possible to infer D u or the time scale on which perturbations of the native system evolve when S T͞S eq Ͼ Ͼ 1. 1 A) and three space dimensions (FRAP and free-particle deficit simulations shown in Fig. and Pfeq. and dotted lines in Fig. They can also be used more generally to determine the spatial range over which local calcium signals spread (14. there are differences in the dynamics of FRAP-like experiments and the evolution of perturbations of untagged particles in these two cases as well. The net preferential unbinding of untagged particles then results in Dt Ͻ Ͻ Du. In the case of messengers. S T. we expect them to appear as differences between the rates of diffusion of inhomogeneities in statistical quantities. However.. will determine statistical quantities. Methods We use the same letters to represent the species and the concentration of that species (e. Thus. and the time step was 1 ␮s. measurements that resolve trajectories of (many) single molecules can provide complete information and thus yield kon.01 ␮m. at equilibrium.3.5552 mM. most of the particles are bound to traps and for which most of the traps are full. The results were quantitatively robust under space and time mesh refinement. As may be concluded from Eqs.2 ␮m2͞s and the other parameters were Pbeq ϭ 9 mM.9. When there are differences between diffusion coefficients of inhomogeneities and single molecules.1073͞pnas. because of the net preferential unbinding of untagged particles. such as f luorescence loss in photobleaching (FLIP) or f luorescence correlation spectroscopy (FCS). Dt. and D t. which is not a priori unreasonable for nonspecific binding to DNA for example. the time it takes for the signal to arrive at B can be much shorter than the time it takes for the input molecules to traverse the intervening distance. In the so-called ‘‘reaction-dominant’’ regime. Discussion These results have ramifications for the interpretation of data taken by using FRAP (4 –10).pnas. using first order explicit time-stepping. Under the assumption that the total concentration of traps. koff. diffusion coefficient measurements serve to estimate how long it takes for a local increment of the messenger concentration at point A to reach another at point B. 10 for the glucocorticoid receptor: Seqkon ϭ 500 ͞s and koff ϭ 86. and methods that monitor the spread of radioactivity (2. respectively. regardless of the spatial extent of the perturbation. Whether D t and D u coincide or differ for proteins that are studied by using FRAP or related techniques needs to be assessed on a case-bycase basis. The condition S T͞S eq Ͼ Ͼ 1 is encountered in the case of calcium signals. In principle. Pb. In any case. KD ϭ 0. second order centered differencing for the Laplacian and Neumann boundary conditions. we use these same binding rates and diffusion coefficients in the simulations reported in Fig. k on S eq. other related methods. With current technology. Pf ϭ [Pf]). and 0. We chose the ratio Seq͞ST ϭ 0. 10 that correspond to different relations between the reactive and diffusive times. This time scale and 5340 ͉ www. the ratio of bound to free times. 1B) domain. such as mean response times. such complete statistics are currently not available (18). ST ϭ 10 mM. and Pfeq ϭ 1. such as correlations. The space grid spacing was 0. unless k on and S eq can be measured independently. Inferences based on D t rather than D u can lead to erroneous conclusions. the time scales with which small perturbations to equilibrium evolve and the time scales of FRAP experiments are different when ST͞Seq is large. The distinction between collective (Du) and single-particle (Dt) diffusion is especially important in the case for which. and of single particles.g.4 ͞s. dashed. from which D f can be inferred. 3). FRAP experiments provide information on k off. Pbeq. i ѨPb i j i ϭ k on P f ST Ϫ Pb Ϫ k off P b Ѩt j ͩ ͸ ͪ ͩ ͸ ͪ [4] where the superscript i refers to whether the species is tagged (i ϭ t) or untagged (i ϭ u). 10. Dt Ϸ Du Ϸ Df. and the equilibrium concentrations of S. 2 B). The dynamics of inhomogeneities. The illustrative examples in this manuscript were obtained via numerical simulations of Eq. This effect pertains whenever Seq͞KD Ͼ 1 and Seq͞ST Ͻ Ͻ 1 and the spatial spread of the particles can be measured in experiment. Effective diffusion coefficients can be used in theoretical estimates of the speed of intercellular and intracellular calcium waves (13). We denote the total substrate (or trap) concentration by ST. and Pf by Seq. The domain in the three-dimensional simulations was a sphere of radius R ϭ 5 ␮m and in the one-dimensional simulations a line segment of length 5 ␮m. 0. There are two other limiting regimes defined in ref. as opposed to D t. rather than of single molecules. is rescaled by a factor that includes all the particles that can become unbound during the diffusive process. These assumptions are typical of FRAP analyses (10). diffusion coefficients can not be obtained from experimentation. Nevertheless. In these cases. the difference disappears and Dt Ϸ Du. 1 A rather than parameter values that would be appropriate for the binding of calcium. We assume that the traps are immobile and distributed uniformly in space. 17). direct observation of single molecules has begun to provide another window into the cell and single-molecule diffusion coefficients have been constructed from what is tantamount to direct observation of the molecules (16. . For the case in which a small bolus of tagged particles is initially added to a background of Pando et al. The free particle diffusion coefficient used was Df ϭ 9. 12–15.which coincides with the one obtained in the rapid buffering approximation (12). S eq/K D. If there are already trapped messenger molecules between A and B.0509576103 D u are the quantities that characterize the dynamics of signal propagation. because large amounts of calcium ions get bound to buffers inside cells upon their entry in the͞cgi͞doi͞10. is uniform. The solid. 4 in one space dimension (tagged particle bolus simulation shown in Fig. and k on S eq. If the traps are nearly unoccupied (Seq Ϸ ST as represented by the dotted curve in Fig.1 to illustrate the difference between diffusion of tagged and untagged particles. and Du. Seq. For comparison between the various simulations. 2 B show the ratio of D t͞ D u as a function of S eq͞K D for S eq͞S T ϭ 0. Assuming that the spread of tagged and untagged particles proceeds at the same rate can result in potentially incorrect inferences about in vivo biological processes.

0) ϭ Pfeq. t ) Ϫ ( P f ϩ P b) tϭ0) dV ͞ (( P f ϩ P b) tϭϱ Ϫ ( P f ϩ P b) tϭ0). 14 ͉ 5341 BIOPHYSICS ͫͬ ͫ u ˆ u ˆt u ϭ e ˆ u͑q ͒␶ A ˆ͑␶͒ ˆu Z ␾ ˆ t͑q ͒␶ ⅐ A ˆt . 0 ͒ ϭ ␾͑ ␰ ͒ . and R. but also in the time scale on which equilibrium is approached if S T͞S eq is large. Given that the ˆ translates into a similar form for the exponential. and Pf(x. when S T͞S eq Ͼ Ͼ 1. perturbations of the native system evolve differently from FRAP. P b Ϫ P beq)] from equilibrium in the case for which there are no tags. Pu concentration) and noting that. Note that Eqs. where ⌽(x) is a function of compact support to represent the fact that the bolus is added locally in space. To this end. Thus. Leon for useful conversations. 0) ϭ STPfeq͞(Pfeq ϩ KD). 12 and 13 and Eqs. . where q is the conjugate variable to ␰ and we have used hats to denote Fourier transforms. APPLIED MATHEMATICS ͭ Ѩu1 ϭ A ⅐u1 Ѩ␶ u1 ͑ ␰ . A ϵ Au W . the asymptotic state will be approximately equal to the initial equilibrium configuration. block form of A the solution of Eq. respectively. If the added bolus is small or if the spatial domain is large enough. where V is the volume of the perturbed sphere. Goldstein. 14 and 15 can be written in dimensionless form as Ѩu ͞Ѩt ϭ Auu . The recovery curves shown in Fig. We define the vectors u† ϭ (uu. we consider the u initial condition: Pf (x. 8 we find the results presented in Eqs. with Au defined as before. 1D are given by the integral over the perturbed sphere of the difference between the current and initial total particle concentrations divided by the integral of the difference between the final and initial total particle concentrations: 1 ͞ V ͐ (( P f ϩ P b)( r . 7 we deduce that these matrices can be written as ˆi ϭ A ͫ Ϫq 2 Ϫa i ai hi . Thus. Bruno. 6 expanding the solution in powers of ␧. although conceptually quite different. J.untagged particles and traps in equilibrium. The above analysis was for experiments in which a bolus of tagged particles was added as done in refs. Cohen. 0) ϭ ⌽(x). ␾( q ). Our conclusions are indewhere Z ˆ : the matrices Au( q ) and At( q ) pendent of the particular form of Z encode the asymptotic spatiotemporal dynamics of the untagged and tagged concentrations. and S ϭ S T Ϫ P beq ϭ S eq for all time. ˆ (q)␶ ˆ ␶) ϭ e A . For the ␧ order we obtain u0 ϭ 0. So u t t ϩ Pf ϭ P feq. P feq ST ϭ KD S eq which can be written in dimensionless form as Ѩ u ͞ Ѩ t ϭ Atu with At defined as before. P u that from Eq. t t Pb(x. PNAS ͉ April 4. This work was supported by National Institutes of Health Bioengineering Research Partnership Grant R01GM6583001. Thus. 0 ͒ ϭ ␧ ␾͑ ␰ ͒ . 9 can be written as 0 ˆ (␶) is a 2 ϫ 2 matrix (19). and Universidad de Buenos Aires Grant X099. P. For the ␧ order we obtain the system [15] the solution of which can be expressed in Fourier space as: u ˆ 1( q . ␦P b) ϭ ( P f Ϫ P feq. Thus. 2 and 3. i i 0 1 u ϭ ͚iϱ ϭ0 ␧ u . Ϫh i ͬ from which we obtain that their slowest eigenvalues (ruling the asymptotic dynamics) correspond to small q and satisfy ␭ Ϸ Pando et al. B. Pu b (x. [9] Note that the only differences between Eqs. It is then reasonable to describe the dynamics in terms of an expansion around the equilibrium configuration. with 0 At ͫ ͬ Ai ϵ au ϭ at ϵ ͫ ٌ2 Ϫ ai ai hi Ϫh i ͬ Wϵ ͫ 0 0 c Ϫc ͬ [7] ST S eq ϭ P feq ϩ K D K D P feq cϵ KD hu ϵ 1 ϩ h t ϵ 1. [5] b ϵ t t Pb where ␧ is a small quantity that represents the deviation with respect to equilibrium. Not only are the effective diffusion coefficients different in the two cases. at t ϭ 0 P f b ϩ P b and u t S ϭ S T Ϫ P b Ϫ P b are spatially uniform and at equilibrium. can be treated similarly by writing down all five evolution equations (tagged and untagged particles and complexes and substrate u t t ϩ Pf . [11] ␶ ϵ kofft u fu ϵ Pf Ϫ P feq S TP feq u bu ϵ Pb Ϫ P feq ϩ K D ␰ϵ t ft ϵ Pf ͱ k off x Df ␾ ͑ ⅐͒ ϵ ⌽ ͑ ⅐͒ ␧ .E. Eqs. ␾ e ͬͫ ͬ [10] Recovery Curves. we define the variables: Ϫ q 2h i͞ ( a i ϩ h i). thanks W. Thus. Ѩt ͭ [12] Ѩu ϭ A ⅐ u ϩ B͑ u͒ Ѩ␶ u͑ ␰ . ut)† ϭ ( f u b u f t b t) and ␾† ϭ (␾u. Ѩt S eq off b [14] [8] We can solve Eq. J. 2 and 3. 4 we deduce that P f b ϩ P b ϭ P beq. the effective diffusion coefficient obtained in this case coincides with that of the untagged particles in the case of the added bolus. 2006 ͉ vol. 12 and 13 are similar to those that describe the evolution of a small perturbation [( ␦P f. the dimensionless effective diffusion coefficients are Di ϭ 1 1ϩ ai hi . From Eq. A. Agencia Nacional de Promocio ´ n Cientı ´fica y Tecnolo ´ gica (Argentina) Proyecto de Investigacio ´n Cientı ´fica y Tecnolo ´ gica 03-08133. 103 ͉ no. The case of FRAP (10). the solution of the five nonlinear coupled equations can be obtained from t ѨPf t t t ϭ D fٌ 2P f Ϫ konP f S eq ϩ k off P b Ѩt t ѨPb t t ϭ konP f S eq Ϫ k off P b . the effective diffusion coefficient obtained in this case coincides with that of the tagged particles in the case of the added bolus.P. 0) ϭ 0. [6] [13] where B is quadratic in u and A is a linear operator. 14 and 15 are that k off is rescaled by S T͞S eq in the latter. Laboratory-Directed Research and Development Contract X1E8 from Los Alamos National Laboratory. The linearized evolution equations in this case are ST Ѩ␦Pf ϭ Dfٌ2␦Pf Ϫ konS eq␦ P f ϩ k ␦P Ѩt S eq off b ST Ѩ␦Pb ϭ konS eq␦ P f Ϫ k ␦P . ␾t)† ϭ (0 0 ␾ 0) and write the evolution equations as Restoring the units and replacing the corresponding values of ai and hi according to Eq.

Webb.0509576103 Pando et al. (2004) Biophys. (1957) J. 87–91. 11. 321–375.. 4594 – 4598. E. (1976) Proc. & McNally. W. E. D. 331–346. 4. G..1. L. 10. 12. W. S. M. (2003) Neuron 40. & Dawson. (1975) The Mathematics of Diffusion (Clarendon͞Oxford Univ. H. R. E. E Stat. D. Nishimura.. H. J. 86. 253–281. Axelrod.. A. Sci. Physiol. (2002) Phys. (1976) Proc. 5. E... H. Keizer.. (1992) Science 258.. I. D. D. & Pearson. D.. (1976) Biophys. Acad. W. & Havel. Elson. Y. L. 15921–15926. Patterson. M. E. Math. A. W. E. D. Koppel.. P. T. W. Chernomoretz. S. Sprague.. Meyer. & Webb. Vrljic. A. Natl. 7. Stavreva. Schlessinger. Brasselet. Lippincott-Schwartz. E. (1998) Neuron 20. (2002) Biophys. B. W. Koppel. Allbritton.pnas. P... Press. P. D. 544–548. (2002) Science 297. G. 16. R. Wagner. & Lippincott-Schwartz. 9. 17. Natl. Strier. T. J. L. F. 16. 5342 ͉ www. W. 22. & Patterson. 67. Sci.. E. K. D. & Stryer. S. Natl. (1994) Biophys. L. 6. Acad.. 18. J. 19. Schlessinger. USA 73. J. J.1073͞pnas. (1999) Proc. Natl. M.. 1315–1329. & Elson.. Koppel. & Keynes. E. L. Nonlinear Soft Matter Phys. 2409–2413. J. 16. J. 83. Elson. J. 8. Crank. E. 6060–6063. E. Sci. Acad. H. 389–399. Augustine. L. Moerner. Moerner.. W. (2003) Science 300. (1995) Adv. & McConnell. G.. J. J. W. 2681–2692. (2004) Proc. Schlessinger. Acad. N. Appl. USA 96. 1873–1877. Deich. H. E. 2. Axelrod. Judd. S. 3. 65. 046233–046314. Elson. J. Oxford). L.. 447–456. J. & Tanaka. 1055–1069. Sci. &͞cgi͞doi͞10. Najfeld. Axelrod. Neher. E. J.. 13. Koppel. K. Chem. E. (1976) Biophys. McAdams. Santamaria. Pego. Rev. 14. G. Axelrod. J. . Ravdin. Schlessinger. Jacobson. J. (London) 138. 3473–3495. W.. J. Hodgkin.. Dawson. (2003) Trends Anal. F. & Podelski. 16. T. J. D. L.. & Moerner. E. J. 1812–1815. 15. J.. R. D. USA 101. USA 73. W.. Webb. D. & Webb.