Single Partical Tracking Review

P1: NGM/rpk P2: sny/rpk QC: rpk/BS T1: rpk
March 31, 1997 19:49 Annual Reviews AR031-14 AR031-14
Annu. Rev. Biophys. Biomol. Struct. 1997. 26:373–99

Copyright c 1997 by Annual Reviews Inc. All rights reserved
SINGLE-PARTICLE TRACKING:
Applications to Membrane Dynamics
Annu. Rev. Biophys. Biomol. Struct. 1997.26:373-399. Downloaded from arjournals.annualreviews.org
Michael J. Saxton
Institute of Theoretical Dynamics, University of California, Davis, California 95616;
email: mjsaxton@ucdavis.edu
Ken Jacobson
by National Taiwan University on 03/13/07. For personal use only.
Department of Cell Biology and Anatomy, University of North Carolina at Chapel

Hill, Chapel Hill, North Carolina 27599; email: frap@med.unc.edu
KEY WORDS: single-particle tracking, fluorescence recovery after photobleaching, lateral

diffusion, membrane dynamics, cell membrane
ABSTRACT
Measurements of trajectories of individual proteins or lipids in the plasma mem-
brane of cells show a variety of types of motion. Brownian motion is ob-
served, but many of the particles undergo non-Brownian motion, including di-
rected motion, confined motion, and anomalous diffusion. The variety of motion
leads to significant effects on the kinetics of reactions among membrane-bound
species and requires a revision of existing views of membrane structure and
dynamics.
CONTENTS
PERSPECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
Capabilities of SPT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
Modes of Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376
EXPERIMENTAL TECHNIQUES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376
DATA ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
APPLICATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
Classification of Modes of Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
Anomalous and Normal Diffusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381
Confined Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
Directed Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389
SPT and FRAP: Effects of the Label . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
WHAT DIFFUSION TELLS US ABOUT MEMBRANE STRUCTURE . . . . . . . . . . . . . . . . 391
TECHNICAL PRIORITIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393
373
1056-8700/97/0610-0373$08.00
374 SAXTON & JACOBSON
PERSPECTIVES
In single-particle tracking (SPT), computer-enhanced video microscopy is used
to track the motion of proteins or lipids on the cell surface. Individual molecules
or small clusters are observed, with a typical spatial resolution of tens of
nanometers and a typical time resolution of tens of milliseconds. Some general
questions addressed by the technique are as follows:
(a) How do particles move on the cell surface? To what extent does the mo-
tion of various particles deviate from pure diffusion? How is that motion
controlled, and what is its function?
(b) How is the cell surface organized? To what extent do membranes deviate
from the fluid mosaic model? Is a fractal time model a useful description of
the cell surface (36, 63)? How are structures on the cell surface assembled?
Does compartmentation prevent crosstalk of receptors (30)? What regional
or global control over cell membrane dynamics exists (85)?
(c) What are the effects of heterogeneous motion in a heterogeneous environ-

ment on kinetics and equilibrium (3, 20, 44, 91, 99)?
More specifically, SPT may help to answer questions about particle motion
raised by fluorescence recovery after photobleaching (FRAP) measurements.
First, FRAP experiments show that diffusion coefficients for proteins in a cell
membrane are 5–100 times lower than the values for proteins in an artificial
bilayer (28, 103). Many mechanisms may be involved: obstruction by mobile
or immobile proteins, transient binding to immobile or mobile species, confine-
ment by membrane skeletal corrals, binding or obstruction by the extracellular
matrix, and hydrodynamic interactions. These mechanisms have been difficult
to sort out, in large part because some or all of them may occur simultaneously,
and their relative importance may depend on the protein and the cell type (30).
Second, a significant fraction of protein and lipid is immobile on the time scale
of a FRAP experiment. For artificial bilayers and rhodopsin in the rod outer
segment, recovery is close to 100%, but in the plasma membrane, recovery is
typically 25% to 80% (30). The increased resolution of SPT ought to make it
possible to understand the FRAP immobile fraction. Third, in FRAP exper-
iments, the distribution of observed diffusion coefficients D is much broader
than expected from experimental error (28, 51, 98). Values of D vary around
twofold among different points on a single cell, and tenfold among cells (51).
This suggests significant heterogeneity in the membrane, a view supported by
other evidence (7, 28, 29).
SPT IN MEMBRANES 375
Capabilities of SPT
SPT has several advantages over FRAP measurements. The spatial resolu-
tion is approximately two orders of magnitude higher than FRAP, so that with
sufficient time resolution (65) motion in small domains can be characterized.
Typically the time resolution is similar to FRAP, so the minimum detectable
diffusion coefficient is lowered by approximately two orders of magnitude. Fur-
thermore, FRAP averages over hundreds or thousands of diffusing molecules,
but SPT measures individual trajectories. Thus, different subpopulations indis-

tinguishable by FRAP can be resolved. SPT provides the ultimate specificity
in measurement of motion of membrane components, particularly if the in-
dividual particle tracked could be characterized in terms of, for example, its
phosphorylation state.
Modes of Motion
A major advantage of SPT is the ability to resolve modes of motion of in-
dividual molecules, and a major result of the technique is that motion in the
membrane is not limited to pure diffusion. Several modes of motion have
been observed: immobile, directed, confined, tethered, normal diffusion, and
anomalous diffusion. In an ensemble average, the time dependence of the
mean-square displacement (MSD) for pure modes of motion is much different
(Figure 1) so the motion can be classified readily.
Figure 1 The mean-square displacement hr 2 i as a function of time t for simultaneous diffusion

and flow, pure diffusion, diffusion in the presence of obstacles, and confined motion.
Two important phenomena have been observed that are related to the classi-
fication of modes of motion. First, correlated motion can provide convincing
evidence that apparently non-Brownian motion is in fact non-Brownian mo-
tion and not merely a fluctuation in Brownian motion. Second, practically all
experimental results show apparent transitions among modes of motion. If a
transition is real, it could result from partition of the mobile species into dif-
ferent microdomains or from an active control mechanism such as transient
binding to a cytoskeletal motor (76, 90).
History
In the first SPT experiment on cell membranes, Barak & Webb (5) tracked
a fluorescent-labeled low density lipoprotein receptor (see also 42, 45). De
Brabander et al developed the technique of nanovid microscopy, in which
a highly scattering colloidal gold label is used with bright-field microscopy

(24). They applied the technique to endocytosis and protein motion on the cell
surface (25). Sheetz and collaborators developed techniques using differen-
tial interference contrast microscopy to determine particle coordinates with
nanometer resolution, and applied this to the motion of motor molecules and
membrane proteins (41, 81, 89). This combination of techniques led to current
SPT work on gold-labeled membranes.
EXPERIMENTAL TECHNIQUES
Video microscopy is reviewed in (48, 49, 92), and SPT techniques, resolution,
and error analysis are discussed in several reviews (6, 41–43, 65, 78, 81, 88,
89).
Nanometer-scale SPT is possible because the center of a small particle can
be located with a precision well below the wavelength of light, even though
two particles at that separation cannot be resolved (43, 81, 88). The particle
is much smaller than the wavelength of light, so its image is an Airy disk, and
two nearby particles give partially overlapping Airy disks. According to the
Rayleigh criterion (49), if the particles are too close, the pair cannot be resolved.
But this unresolved spot is more intense than the spot for a single particle, so
the number of particles can be determined, at least well enough to distinguish
multiple particles from a single particle. For a wavelength of 546 nm and a
numerical aperture of 1.4, the radius of the Airy disk is 238 nm.
The limiting spatial accuracy in an SPT measurement is set by the mechanical
stability of the apparatus and is obtained from trajectories of stationary parti-
cles. The scatter in position is 1–30 nm, yielding a minimum observable D of
5 × 10−14 to 5 × 10−13 cm2 s−1 . For mobile particles, the spatial accuracy is
decreased by the motion of the particle during the acquisition time of the image,
and it is therefore a function of D (81). The acquisition time depends on the
label. For gold labels, images are usually obtained at the standard video rate,
so the image is integrated over 1/30 or 1/25 s. For fluorescent labels, typical
acquisition times are 1–10 s, although the fastest reported so far is 5 ms (78).
Camera lag and interlacing must also be considered because they degrade the
time resolution (21a, 48, 88, 100).
Colloidal gold, latex beads, and fluorescent particles have been used as labels.
Colloidal gold is a strong light scatterer that acts as a light sink rather than a light
source. Light is scattered out of the objective, so, after background subtraction
and contrast enhancement, the label appears darker than the surrounding image.
The diameter d of the gold particle is much less than the wavelength of light, so
the particle is a Rayleigh scatterer, for which the scattering ∝ d 6 . The minimum
detectable diameter is ∼15 nm and the typical diameter used is 30–40 nm. Gold
particles are much stronger scatterers than organelles are, so the organelles are
almost invisible in bright-field microscopy (93). The use of gold labels is

reviewed in References 21 and 23.
Fluorescent labels used include fluorescent microspheres, typically of diam-
eter 30–100 nm (37, 46); phycobiliproteins (104); virus labeled with fluorescent
lipid analogs (2); low-density lipoprotein labeled with the carbocyanine lipid
analog diI (diI-LDL) (4, 42, 43); diI-LDL conjugated to immunoglobulin E
(IgE) (95); and tetramethylrhodamine conjugated to individual lipid molecules
(78, 79). Advantages and disadvantages of the labels are discussed in the
references.
There are several potential difficulties associated with different labels. First,
most labels are large, so that drag from the interaction of the label with the
extracellular matrix may be significant (59, 60, 95). Second, labels are often
multivalent and can crosslink binding sites. Crosslinking lowers D through
hydrodynamic effects (1) and may trigger biological responses such as trans-
membrane signaling and interactions with the cytoskeleton. Furthermore, if
diffusion is restricted by corrals, crosslinking yields aggregates less likely to
cross corral walls (34, 68). Third, perturbations caused by antibody binding can
affect interactions of the labeled protein with other proteins (19, 52). Finally,
during a measurement, a particle may disappear as a result of moving out of
the focal plane, endocytosis, detachment from the membrane, or photobleach-
ing (2). (For a detailed quantitative discussion of photobleaching of single
fluorophores, see 78.)
DATA ANALYSIS
The goal of SPT data analysis is to sort trajectories into various modes of
motion and to find the distribution of quantities characterizing the motion,
such as the diffusion coefficient, velocity, anomalous diffusion exponent, cor-
ral size, and escape probability. The difficulty is that in a pure random walk
the randomness yields trajectories that suggest other modes of motion. This
problem is made worse by the experimental limits on the duration of trajectories
measured (65, 72).
It is instructive to calibrate (or uncalibrate) one’s intuition by writing a simple
random walk program and looking at a few dozen pure random walks. One
will see apparent diffusion, directed motion, trapping, and transitions (66, 70)
because our nervous systems are wired to see patterns. But observation of mul-
tiple trajectories with the same apparently nonrandom behavior provides strong
evidence that the nonrandom behavior is real. The most striking experimental
examples are of directed motion (43) and motion among corrals (67).
In addition to the usual cellular, biochemical, and instrumental controls, it
is necessary to do controls for data analysis using a pure random walk as a
reference. The minimum test for a classification algorithm is to try it on pure
random walks of the appropriate number of time steps. A more rigorous test
requires both experiment and simulation. For example, consider the case of
corralled motion. First, the experimental corralled trajectories are identified by
some criterion. Then the criterion is applied to pure random walks to see how
many pure random walks are falsely classified as corralled, and to corralled
random walks to see how many corralled random walks are falsely identified as
free. Some corralled trajectories are necessarily rejected because their residence
times are by chance very low, so the average escape time is biased toward
higher escape times. To be able to do such tests, it is necessary to use some
algorithm to find quantities such as the initial slope, rather than finding them
by eye.
When reporting classifications of trajectories based on some parameter, in-
clusion of a histogram of the parameter for the experimental data and pure
random walks (57, 67) is useful to show whether the classification is based on a
somewhat-arbitrary dividing line in a unimodal distribution or a minimum in a
multimodal distribution. Similarly, if multiple parameters are used, it is useful
to show them as a scatter plot (68).
To reduce the noise in an experimental trajectory, the data points within a
single trajectory are averaged, yielding the mean-square displacement (MSD)
for that trajectory (65). The MSD for a given time lag can be defined as the
average over all independent pairs of points with that time lag (42), or all pairs
of points with that time lag. These averages are discussed in detail elsewhere
(MJ Saxton, manuscript submitted). Briefly, for time lags less than ∼ 14 of
the total number of points in the trajectory, the two averages agree, but the aver-
age over all points is less noisy. When the time lag is a substantial fraction of the
length of the trajectory, neither average is useful because there are simply not
enough data points, as shown by the formulas for the standard deviations (see
65). The short-range MSD is accurately determined, but the long-range MSD
is noisy, yielding good short-range diffusion coefficients but highly scattered
long-range ones. Averaging should not be done automatically because it may

obscure transitions between diffusive and nondiffusive segments of a trajectory.
The analytical forms of the curves of MSD versus time for the different
modes of motion (Figure 1) form the basis of various classification methods.
hr 2 i = 4Dt normal diffusion (1)

hr 2 i = 4Dt α
anomalous diffusion (2)

hr i = 4Dt + (V t)
2 2
directed motion with diffusion (3)
®£ ¡ ± ®¢¤
hr 2 i ' rC2 1 − A1 exp −4A2 Dt rC2 corralled motion (4)
In Equation 2, α < 1, so strictly speaking this is anomalous subdiffusion (11,

36). In Equations 3 and 4, V is velocity, hrC2 i is the corral size, and A1 and A2
are constants determined by the corral geometry. Equation 4 is based on the
first two terms of the exact series solutions for square corrals (57) and circular
corrals (70).
The probability density p(r, t)dr is the probability that a particle at the origin
at time zero is at position r at time t. For pure diffusion in two dimensions (65),
1
p(r, t)dr = exp(−r 2 /4Dt)2πr dr, (5)
4π Dt
and for diffusion with simultaneous flow along the x-axis with velocity V ,
1
p(x,y, t,V )d x d y = exp([−(x − V t)2 + y 2 ]/4Dt)d x d y. (6)
4π Dt
For corralled motion, the probability density depends on the initial position in
the corral, and is complicated (57, 70).
Webb and collaborators (36, 95) assume that the probability density is the
standard two-dimensional form of Equation 5 but with a time-dependent diffu-
sion coefficient:
D = (1/4)0t α−1 , (7)
or, equivalently,
hr 2 i = 4Dt = 0t α , (8)
with α < 1. Diffusion is free at short times but slowed at longer times as the
effect of barriers becomes dominant. The physical basis for Equation 7 is the
idea of the membrane as a random array of continuously changing traps with
a distribution of energies so broad that there is no average residence time (36).
The continuous-time random walk (CTRW) model (63) gives the same form
for hr 2 i at long times.
Next, we summarize methods used to classify trajectories. Whatever the

method, the longer the run, the more reliable the classification unless the particle
changes its mode of motion.
Cherry and colleagues (2, 104) use the shape of the hr 2 (t)i curve to classify
trajectories. They calculate the experimental MSD and determine which analyt-
ical expression yields the best fit. These workers also construct an experimental
probability density and fit sums of standard forms of p(r ) to it.
Kusumi and colleagues (57) characterize the shape of the hr 2 (t)i curve in
terms of the relative deviation (RD). In effect, one draws a straight line through
the origin with the observed initial slope, which is known precisely and is not
affected significantly by nondiffusive motion. Then one extrapolates the MSD
to a prescribed time and takes RD to be the ratio of the observed MSD to the
extrapolated MSD. If RD > 1, then the motion is directed; if RD < 1, the
motion is confined. This approach reduces the shapes of the different curves of
Figure 1 to a single parameter. The distribution of RD can then be calculated for
a pure random walk and non-Brownian motion. The overlap of the distributions
is a measure of how well non-Brownian motion can be distinguished from pure
Brownian motion when using this parameter (57, 72).
Webb and collaborators (36, 95) use the anomalous diffusion exponent from
Equation 8. For each trajectory, log hr 2 i is plotted versus log t, α is found from
the initial slope, and the trajectory is classified according to α.
The radius of gyration tensor is a well-known tool to characterize random
walks (66), which yields the asymmetry parameter a2 and the radius of gyration
2 2
Rgyr , a measure of the extent of the random walk. The joint distribution of Rgyr
and a2 may be used to classify trajectories (70). A related approach (93)
combines the observed D, the shape of the hr 2 (t)i curve, and the values of
2
Rgyr and a2 to sort trajectories into mobile, slowly diffusing, corralled, and
immobile.
APPLICATIONS
Results of SPT experiments, and the corresponding FRAP experiments when
available, are summarized in the tables. Table 1 includes artificial bilayers;
Table 2, lipids and GPI-linked (glycosylphosphatidylinositol) proteins in cells;
and Table 3, selected transmembrane proteins in cells. We believe the tables
include all the results to date for which both SPT and FRAP data are available.
Classification of Modes of Motion

FRAP measurements are generally interpreted as showing only mobile and
immobile fractions. SPT makes a much more detailed classification possible.
Practically all experimentalists report different modes of motion and transitions
Table 1 Artificial bilayersa
Membrane Bilayer
component Label composition D(SPT)b D(FRAP) Ref.
Lipid analogs
TMR-POPE None POPC 140 ± 23 77 ± 13 78
Fi-PE 30 nm Au 87% egg PC 30 (MV) 133 ± 33 58
Ab-Fl 13% Chol 70 (PV)
Biotin-PE 30 nm FM 80% egg PC 30 — 37

St 20% Chol
GPI-linked proteins
DAF (CD55) 30 nm FM 80% egg PC 25 — 37
St-biotin-Ab 20% Chol
Fcγ RIIIB (CD16) same same 56 — 37
a
Ab, antibody; Chol, cholesterol; Fl, fluorescein; FM, fluorescent microsphere; GPI, glycosylphos-
phatidylinositol; MV, multivalent; PC, phosphatidylcholine; PE, phosphatidylethanolamine; POPC,
palmitoyloleoyl PC; POPE, palmitoyloleoyl PE; PV, paucivalent; St, streptavidin; TMR, tetramethyl-
rhodamine.
b
All diffusion coefficients D in units 10−10 cm2 s−1 .
among the modes. Often, simple diffusion is observed in only a minority of

trajectories. Some data on modes of motion are given in the tables, but methods
of classification differ enough among laboratories that a table of modes of
motion is not useful.
Anomalous and Normal Diffusion
One of the most important results of SPT to date is the observation and mea-
surement of anomalous diffusion in cell membranes. Anomalous diffusion can
be used as a probe of membrane organization. Furthermore, anomalous diffu-
sion implies slow diffusional mixing and therefore affects reaction rates in the
membrane (3).
What is the cause of this nonclassical behavior? In the most general terms,
anomalous diffusion results from a deviation from the central limit theorem,
resulting from pathologically broad distributions of jump times or jump lengths,
or strong correlations in diffusive motion (11). In cell membranes, anomalous
diffusion is most likely the result of both obstacles to diffusion and traps with
a distribution of binding energies or escape times.
For diffusion in the presence of random point obstacles (71), diffusion is
anomalous at short times and normal at long times: hr 2 i ∼ t α for t tC R
and hr 2 i ∼ t for t tC R , where tC R is the crossover time and α < 1. As the
obstacle concentration approaches the percolation threshold, α decreases and
tC R increases, that is, diffusion becomes more anomalous for a longer time.
At the percolation threshold there is no crossover, and diffusion is anomalous
at all times because the percolation cluster is self-similar. For diffusion in the
P1: NGM/rpk
March 31, 1997
Table 2 Lipids and GPI-linked proteins in cellsa

382
Membrane Apparent D(SPT)

P2: sny/rpk
component Cell Label for mobile fraction D(FRAP) Comments Ref.

19:49
Lipid analogs
Fi-PE Fibroblasts 30 nm Au 12 ± 7 54 ± 27 23% with D < 4 (lamella) 60
Ab lamella (69% mobile) 71% with D < 4 (nucleus)
QC: rpk/BS
biotin-PE Fibroblasts 100 nm FM 8 — Unspecified fraction stationary 46

avidin and not extractable by detergent
Fi-PE Neurons 500 nm latex 24 ± 6.8 — Analyzed as 1d diffusion 22
SAXTON & JACOBSON
Ab-Fl along axon length

T1: rpk
Annual Reviews
Glycolipid Fibroblasts 40 nm Au 7.1 ± 1.3 — 18% slow diffusion 84

GM-1 cholera with D ∼ 1
toxin B 38% corralled diffusion
GPI linked proteins
Thy-1 Fibroblasts 100 nm FM 6.1 ∼30–40 39% slow diffusion 46
Ab (corralled) (∼50% mobile) 50
AR031-14
D(slow) = 0.057
Thy-1 Fibroblasts 40 nm Au 7.2 ± 1.0 — 31% slow diffusion 84
Ab D(slow) ∼ 0.2
38% corralled
NCAM 125 Myoblasts 30 nm Au 0.89 ± 0.18 3.7 ± 0.4 33% slow diffusion b
Ab ∼70% mobile D(slow) ∼ 0.04

AR031-14
28% hybrid
MHC I (Qa2) HEPA-OVA 40 nm Au 2.1 ± 0.3 2–4 33
cells Ab
a
See footnotes a and b of Table 1. NCAM, neural cell adhesion molecule; MHC, major histocompatibility complex.
b
R Simson, SE Moore, P Doherty, FS Walsh & KA Jacobson, submitted.
P1: NGM/rpk
March 31, 1997
Table 3 Transmembrane proteins in cellsa

P2: sny/rpk
19:49
Membrane Apparent D(SPT)

component Cell Label for mobile fraction D(FRAP) Comments Ref.
Cell adhesion molecules

QC: rpk/BS
E-cadherin Cultured 40 nm Au 0.16 0.34 High Ca+2 medium 57

epidermal Ab 75% mobile 28% random diffusion
cells 64% corralled
ensemble average D(SPT)
T1: rpk
Annual Reviews
NCAM180 Fibroblasts 30 nm Au 3.7 ± 0.18 18.3 ± 3.1 14% slow diffusion 93

Ab 60% mobile D(slow) = 0.5
21% corralled
NCAM140 Muscle 30 nm Au 1.1 ± 0.1 4.7 ± 0.5 22% slow diffusion 93
Ab 70% mobile D(slow) = 0.06
14% hybrid
AR031-14
Receptors
Transferrin Fibroblasts 40 nm Au 10 (short-range) — Most receptors confined 68
receptor transferrin 0.24 (long-range) to ≈500 nm for
≈30 s durations
FcRI Rat basophilic diI-LDL ≈1 5.4 Time-dependent D 36
leukemia cells (see comment) 80% mobile evaluated at 1 s
AR031-14
Major histocompatibility antigens

H-2Da Class I HEPA-OVA 40 nm Au 1.3 ± 0.2 2–4 33
SPT IN MEMBRANES
cells Ab
HLA-DR Class II Fibroblasts R-phycoerythrin 0.001–0.05 — Also corralled diffusion 104
Ab and directed motion
383
a
See footnotes a and b of Table 1 and footnote a of Table 2. LDL, low-density lipoprotein.
presence of traps, or obstacles that bind the mobile species, the behavior is
complex (11, 74). Other anomalous diffusion models, such as those discussed
in the next paragraph, have no crossover, so determining whether a crossover
time exists restricts possible models of anomalous diffusion.
The effect of anomalous diffusion on FRAP experiments has been exam-
ined. Nagle (63), using a one-dimensional CTRW model with no crossover,
found that if a FRAP recovery curve is generated from the CTRW model but
fit by the standard FRAP recovery equation, the diffusion coefficient and the
mobile fraction change dramatically depending on the time scale of the FRAP
experiment. Feder et al (36) used a time-dependent D (Equation 7) in the usual
two-dimensional probability density function (Equation 5) to obtain a FRAP
recovery equation of standard form but with t replaced by t α . For the exper-
imentally accessible time range, FRAP data could be fit equally well by the
anomalous diffusion equation (parameters 0 and α) and the standard equation

(parameters D and mobile fraction), but no statistically significant improve-
ment in the fit was obtained with the anomalous diffusion equation (36). SPT is
more sensitive to anomalous diffusion than FRAP is because in SPT every tracer
is tested for anomalous diffusion individually, but FRAP averages over many
tracers, some of which may be diffusing normally, and others anomalously.
To analyze SPT trajectories, log hr 2 i or log hr 2 i/t is plotted against log t
(42, 71, 93). The initial slope yields the exponent α, but simulations suggest
that it is difficult to see the crossover in an individual trajectory (MJ Saxton, in
preparation). One is more likely to see a crossover from anomalous diffusion to
noise than a crossover from anomalous diffusion to normal. Various factors can
cause crossovers, which suggests caution in interpreting observed crossovers.
Experimental results suggest that anomalous and normal diffusion may occur
in distinct domains, producing an apparent crossover when the tracer leaves
a domain. An apparent crossover might also occur as a result of biological
modulation, such as a change in phosphorylation state.
SPT data for various proteins and cells show a significant amount of anoma-
lous diffusion. Ghosh (42, 43) was the first to observe anomalous diffusion
in cell membranes, in measurements on the LDL receptor labeled with diI-
LDL. Trajectories were classified according to α, and ∼ 50% of the trajectories
showed anomalous diffusion, with values of α between 0.2 and 0.9. Transitions
were observed from anomalous to normal to anomalous diffusion in the trajec-
tory of an individual particle. If anomalous diffusion in this system were caused
by random point obstacles, diffusion would remain normal after the crossover.
The observed behavior suggests domain structure in the membrane.
Slattery (95) measured diffusion of the high-affinity IgE receptor FcRI
in rat basophilic leukemia cells, using an antibody-conjugated diI-LDL label
elegantly shown not to crosslink receptors under experimental conditions. The
average α was 0.22–0.48, and for 48–60% of the trajectories, α was between
0.1 and 0.9. The FRAP results of Feder et al (36) were obtained in the same
system.
Interestingly, Ghosh (42) found that if the cells were treated with azide and
2-deoxyglucose, the fraction of particles showing normal diffusion increased
from 26% to 43%, which suggests that the constraints to diffusion require
metabolic energy. But Slattery (95) found no effect of metabolic inhibitors in
similar experiments on his system. It will be important to determine whether
the structures that produce anomalous diffusion generally require metabolic

energy.
Simson (93) examined the mobility of neural cell adhesion molecules
(NCAM) labeled with antibody adsorbed to 40-nm gold particles. Short-term
(6.6 s total) SPT measurements were made on fibroblasts, which have no en-
dogenous NCAM. For both GPI-linked and transmembrane isoforms, 50% of

the particles were classified as mobile and showed normal diffusion, but the
slow (15%) and corralled (20%) fractions showed anomalous diffusion with
α = 0.29–0.50. GPI-anchored and transmembrane isoforms of human NCAM
were expressed in cultured mouse muscle cells, along with endogenous NCAM,
predominantly a transmembrane isoform. Long-term trajectories (90 s) showed
that no particles were actually stationary and that corralled particles escaped
and diffused normally before being corralled again, yielding hybrid trajec-
tories. For both transfected and endogenous isoforms, the mobile fraction
(40%–60%) diffused normally, but diffusion was anomalous in the slow frac-
tion and the confined portions of the hybrid trajectories. Diffusion was also
found to be anomalous for the GPI-linked protein Thy-1, the lipid analog Fl-PE
(fluorescein-conjugated phosphatidylethanolamine), and the glycolipid GM1
in those trajectories that exhibited transient confinement or slow diffusion (84).
It will be important to see whether anomalous diffusion occurs in fixed regions
of the cell surface, and if so, the lifetimes of those regions.
Confined Motion
Confined motion may result from corrals formed by cytoskeletal proteins near
the membrane (56, 85a), from tethering to immobile species, or from restrictions
to motion imposed by lipid domains. Evidence for confined motion comes from
the interpretation of FRAP (29, 31), SPT, and laser trap experiments. In SPT
measurements, several workers have reported a significant fraction of confined
motion, with corral sizes in the range of 250–1500 nm, and average residence
times in the range of 3–35 s (85). The range of sizes reflects instrumental factors
and should not be taken as real lower or upper limits. To see small corrals, high
resolution in space and time is needed; to see large corrals, trajectories must be
measured for long times before confinement becomes apparent (65, 67).
Some of the strongest experimental evidence for confined motion comes from
the work of Kusumi and collaborators (56). A remarkable figure of Sako &
Kusumi (67) shows long-term trajectories (1000 images at 3 images/s) of the
transferrin receptor in fibroblasts. Each trajectory was divided into segments by
eye, each segment corresponding to a compartment. The particle appeared to
diffuse within a compartment, escape to an adjacent compartment and diffuse
there, and so on. Quantitative analysis showed that the average compartment
radius was 280 nm, the average escape time was 29 s, and the short-range
diffusion coefficient was 0.1 µm2 s−1 (67). The relation hr 2 i = 4Dt gives a
root-mean-square displacement for a freely diffusing particle of 3400 nm in 29 s,
which suggests confinement. A test for confined motion yields a probability
∼10−20 that a freely diffusing particle with the observed D would remain in a
region of that size for that time (73).
Kusumi and collaborators (57, 67) used their RD parameter to classify tra-
jectories, reporting histograms of RD for their experimental values and Monte
Carlo values for a pure random walk. The distributions are clearly differ-
ent. Furthermore, the criterion for confined motion was chosen so that only
2.5% of pure random walks are falsely classified as confined, and the ob-
served fraction of confined motion was much greater: 30–65% for E-cadherin
in keratinocytes and 80–90% for the transferrin and α2 -macroglobulin recep-
tors in fibroblasts. These results clearly established that for a significant
fraction of trajectories, motion was non-Brownian. Analysis of the hr 2 (t)i
curves showed distinct plateaus for 80–90% of the transferrin receptors (A
Kusumi, private communication), which implies confined motion, not anoma-
lous diffusion.
Simson et al (94) analyzed subtrajectories of experimental trajectories using
a test (70) to see whether a diffusing particle stays in one region longer than
a freely diffusing particle is expected to do. Subtrajectories were classified as
confined if they were not likely to have resulted from a pure random walk and the
residence time was above a prescribed minimum (94). In measurements of two
GPI-linked proteins, Thy-1 in fibroblasts (84) and NCAM-125 in cultured mus-
cle cells (94), 28% of the experimental trajectories showed periods of transient
confinement, but only 1.5% of pure random walks were scored as showing tran-
sient confinement. Similar results were obtained for transmembrane isoforms
of NCAM in muscle cells. Schmidt et al (80) developed a related subtrajec-
tory test and showed confinement of lipids in an artificial bilayer, which they
attributed to defects in the supported planar bilayer.
Others have argued for the existence of corrals using various tests: by fitting
the hr 2 (t)i curve to various functional forms (2), by probability density tests
(25, 104), and by the RD parameter (46). Some results are summarized in the
Tables above. A key SPT experiment would be to test for permanent corrals
by trying to observe two particles in the same corral at different times and to
observe two distinctly labeled particles in the same corral simultaneously.
Further experimental evidence on corrals is obtained from laser trap experi-
ments that measure the barrier-free path (BFP), the distance a labeled particle
can be moved by the trap before it encounters a barrier strong enough to force
it to escape the trap (33). The laser trap (laser tweezers) provides a controlled,
movable potential well with nanometer dimensions and piconewton forces (8,
96, 97). The forces used in the BFP experiments are in the range 0.05–1
pN, which are sufficient to stretch a polymer, less than the force exerted by a
motor molecule, and insufficient to break even a single hydrogen bond (35).
Boal (9, 10) has used Monte Carlo calculations to evaluate the BFP for a model
membrane skeleton consisting of a network of flexible, nonpenetrating polymer
chains attached to the membrane.
Edidin et al (33) measured the BFP for the class I major histocompatibility
complex (MHC), comparing homologous GPI-linked and transmembrane iso-
forms. At 34◦ C, the transmembrane isoform had a BFP of 3.5 µm, and the
GPI-linked isoform, 8.5 µm, which is much larger than the typical corral size
measured by SPT. A temperature decrease from 34◦ C to 23◦ C led to a large
change in BFP but a small change in D, which suggests that the measured bar-
riers do not control diffusion. Experiments on truncated forms of class I MHC
showed a doubling in the BFP for a sufficiently short cytoplasmic domain (34).
To characterize the barriers, Sako & Kusumi (68) measured the BFP as a
function of trap strength for the transferrin receptor in fibroblasts. Corralling
and tethering can be distinguished by laser trap measurements but not nec-
essarily by SPT (68). Approximately 10% of the particles were classified as
tethered, with a BFP < 300 nm and a low diffusion coefficient. The remainder
were classified as confined and had a larger diffusion coefficient. At the highest
trap force, these particles had a BFP ≥ 2 µm, the maximum displacement at-
tempted, but at lower trap forces they often escaped at distances corresponding
to the corral size measured by SPT for free diffusion. Particles were also ob-
served to escape the trap and rebound, implying that the boundaries are elastic.
The force required to move a particle across the boundary was 0.1 pN for a
40-nm gold label and 0.6 pN for a 210-nm latex label. The increase in force
was attributed to increased crosslinking of receptors by the larger label. If the
width of the corral boundary is 10 nm and the force is 0.5 pN, then the energy
required for a particle to cross the boundary (68) is 5 × 10−21 J ' 2.5( 12 kT ).
The BFP measurements are highly interesting, but there are some complica-
tions in their interpretation. First, the trap may exert forces perpendicular to the
membrane, in addition to the in-plane forces (WW Webb, private communica-
tion), though the BFP is still sensitive to truncation of the cytoplasmic domain of
the protein (34; M Edidin, private communication). Second, moving a labeled
protein may stretch the barrier (68; MJ Saxton, in preparation). Third, the mea-
surements involve energies of order kT, and the number of particles measured
is limited, so the results are inherently noisy.
The hindrances constraining lateral diffusion are thought to fluctuate in space
and time (27, 36, 56), but there is disagreement about the length scale of the bar-
riers. Kusumi and collaborators (56) consider a network of membrane skeleton
barriers to be a basic feature of plasma membranes. Long-range diffusion oc-
curs by crossing corral walls as a result of fluctuations in the separation between
the membrane and the membrane skeleton, and as a result of the association-
dissociation equilibrium of membrane skeleton proteins. The corral walls im-
pose a preferred length scale of 300–600 nm, and the fractal time model (63)
applies only to diffusion over shorter lengths. In contrast, the time-dependent
diffusion coefficient of Webb and collaborators (36) requires barriers on all
length scales and no preferred length. Hindrances fluctuate in space and time
over the full range of experimental values, 0.3 to ∼600 s and 30 nm to ∼5 µm,
as determined by SPT (42, 95) and FRAP (36).
It may be possible to reconcile these views. The distribution of escape times
from corrals is wide (around 1.5 orders of magnitude), as shown by Monte
Carlo calculations for uniform circular corrals with a fixed escape probability
for each collision with the wall (73). Presumably the distribution would be
even broader for a more realistic corral model, such as a membrane skeleton of
polymers with conformational fluctuations (9, 10), especially when association-
dissociation equilibria are included. Fluctuations in association are likely to
occur on longer time scales than fluctuations in conformation would because
the energies involved are greater. The distribution of escape times would be
further broadened by the distribution in corral sizes. The corrals would thus
contribute to a time-dependent diffusion coefficient over times greater than the
corral escape time.
Establishing a connection between a domain as defined by SPT trajectories
and a biochemically defined domain is of great interest. One possible example
involves local lipid compositional differences in the membrane. Many GPI-
linked proteins preferentially co-isolate at 4◦ C with a detergent-inextractable
fraction enriched in glycosphingolipids and cholesterol (14, 64). The more
ordered lipids of this fraction may optimally accommodate the highly saturated
acyl chains of the GPI anchor (83). Schnitzer et al (82) have isolated from
lung tissue detergent-resistant vesicles with diameters that are consistent with
the size of the SPT confinement regions observed for several GPI-anchored
proteins (93) (ED Sheets, R Simson, GM Lee & K Jacobson, unpublished
data). The gel-like microphases existing within these putative lipid domains
may be the obstacles producing anomalous diffusion within such regions.
Directed Motion
Directed motion has been observed frequently in SPT experiments. In the uni-
form flow model (72, 90), the entire membrane moves with a constant velocity,
and the particle diffuses with respect to the membrane. In the conveyor belt
model (39, 62, 72, 76, 90), the mobile particle is reversibly attached to a cy-
toskeletal motor. When the particle is bound, it moves with a constant velocity
along the cytoskeletal element; when it is unbound, it diffuses freely. In the
similar “conveyor rope” model, when a particle is bound to the conveyor rope,
it moves with constant velocity along the rope. But it also diffuses over a short
distance in the perpendicular direction due to the flexibility of the cytoskele-
tal element (42, 57), or it diffuses isotropically over short distances from its
attachment point due to the flexibility of the attachment (90).
The uniform flow model describes directed bulk flows within the cell frame of
reference, but no such flow could be measured by SPT with gold-labeled lipids
in keratocytes (60) or by FRAP in leukocytes (61). These studies suggested that
the bilayer moves forward passively with the cell, most likely using membrane
accumulated as the cell retracts its trailing edge. In these experiments, most of
the gold-labeled lectin receptors in the plasma membrane also moved forward
with the keratocyte, showing no net directed movement in the cell frame of
reference (55). These studies seem to rule out a retrograde lipid flow, at least in
non-neuronal cells, although the original proponent remains unconvinced (13).
In contrast, retrograde lipid flow does occur in some axons as membrane
is added at the growth cone and retrieved at the neuronal cell body (12, 40).
Under certain conditions, vesicular transport along the axon and insertion at
the growing tip appears to be the most efficient way to get material to the
site of growth. Recent SPT experiments support this view. In dorsal root
ganglia, membrane lipid flows rearward along the axon as determined by SPT
of Fl-PE labeled with latex beads (22). When a laser trap was used to pull
membrane tethers from bead-labeled axons, the beads moved a significantly
greater distance on the growth cone side of the tether than on the cell body side.
The results of this elegant experiment suggest that the source of the membrane
was at the base of the growth cone.
Although it is well known that various membrane components that are suf-
ficiently crosslinked will move rearward in motile fibroblasts, lymphocytes,
macrophages and neuronal growth cones, SPT studies have shown unexpected
forms of directed transport, which may be attributed to a conveyor belt mecha-
nism. Bead-labeled neuronal antigens were transported forward to the leading
edges of growth cones (87). Similarly, concanavalin A–coated beads were
moved to the leading edge of locomoting fish keratocytes (54), as were inte-
grins in motile fibroblasts (77). In cells undergoing cytokinesis, fluorescent
beads connected to the actin cytoskeleton are drawn to the cleavage furrow
(102).
The conveyor belt model was used by Schmidt et al (76, 77) to analyze
the motion of integrin in moving fibroblasts. Observed trajectories fitted to
Equation 3 yielded the instantaneous velocity and D. The rate constants for
binding and unbinding were obtained from computer simulations in which the
calculated average velocity and the fraction of directed transport were required
to match the observed values.
SPT and FRAP: Effects of the Label

No general, quantitative correspondence has yet been found between SPT
modes of motion and the FRAP immobile fraction, although efforts have been
made to reconcile the data from measurements on the same system (57, 93).
An alternative approach uses a time-dependent D with no immobile fraction to

fit both experiments (36).
Some striking differences between FRAP and SPT measurements suggest
that our understanding of the effects of the label is incomplete. For lipid
components in supported planar bilayers, D(FRAP) values are a factor of two
to four times larger than D(SPT). A notable exception is the elegant work of
Schmidt et al (78–80), tracking individual rhodamine-labeled lipid analogs in
planar supported bilayers. Here no gold or fluorescent bead is used, and the
agreement between FRAP and SPT is much closer, with the SPT values higher.
This result is attributed to lateral heterogeneity in the bilayer. SPT detects free,
rapid diffusion within small regions, but FRAP measurements sample both free
diffusion and diffusion obstructed by defects in the bilayer.
The general diminution of D values in SPT relative to FRAP is attributable to
the valency and size of the label. The effect of these two factors has not yet been
resolved. However, Lee et al (58) measured the diffusion of Fl-PE labeled with
30-nm gold microspheres in planar supported bilayers, using both a multivalent
label, coated completely with relevant (anti-Fl) antibodies, and a paucivalent la-
bel, coated with a 20:1 mixture of irrelevant and relevant antibodies. Compared
with D(FRAP) for Fl-PE, D(SPT) was reduced by a factor of ∼4 for the multiva-
lent label but only by a factor of 2 for the paucivalent label. Moreover, increasing
the viscosity of the aqueous phase by a factor of 2 did not change D, which sug-
gests that it was the higher valency that reduced D. Valency affects D because
it affects both the cluster size and the distance between the lipids anchoring
the label. Depending on the degree of penetration by the lipid solvent into the
gold-tagged lipid cluster, hydrodynamic effects lower D by a factor of 1.5–2.0.
Measurements on gold-labeled avidin-biotin-phosphatidylethanolamine (PE)
by scanning concentration correlation spectroscopy (53) gave D values be-
tween those found by Lee et al (58) for the two antibody labels. Fein et al (37)
found similar D values for PE derivatives and GPI-linked proteins in planar
supported bilayers, but in this work only 10–50% of the particles were mobile.
Whether this is due to the label or to structural irregularities in the bilayer is not
known. When SPT and FRAP values for the highly mobile fraction of various
lipids and GPI-anchored proteins in plasma membranes are compared, again
the FRAP values are 2- to 7-fold greater than the SPT values (Table 2).
Table 3 lists transmembrane proteins measured by SPT and FRAP. The label
can have a large effect on mobility measurements. FRAP measurements on
FcRI labeled with fluorescent antibody gave D values of 5 × 10−10 cm2 s−1 ,
with an 80% mobile fraction. Measurements using the antibody conjugated to

diI-LDL gave similar D values, but a mobile fraction of <20%, perhaps due to
interaction of the label with the extracellular matrix (36).
Mobility measurements on the LDL receptor itself also reveal a pronounced
effect of the label (95). LDL bound to its receptor in the plasma membrane has a
diffusion coefficient two orders of magnitude lower than that of the unliganded
receptor in the plasma membrane or the liganded receptor in a bleb. Presumably,
the bleb membrane is not tethered to the cytoskeleton and has little associated
extracellular matrix material. In the plasma membrane, the label could cause
drag by interacting with the extracellular matrix, or its binding could cause the
receptor to engage the membrane-associated cytoskeleton, markedly reducing
diffusion.
For the class II MHC antigen, D(SPT) (104) is two orders of magnitude
less than D(FRAP) (101). The cause could be biological: Different MHCs
are expressed in different cells and may have different interactions with the
cytoskeleton. But particles that move as slowly as those observed in the SPT
experiments would clearly be part of the immobile fraction in standard FRAP
experiments. This work is the first step in examining the dynamics of the FRAP
immobile fraction and shows that even in the immobile fraction, different modes
of motion are observed.
WHAT DIFFUSION TELLS US ABOUT

MEMBRANE STRUCTURE
One can take two approaches to modeling diffusion in the membrane. In the
mechanistic approach, one starts with some biological structure and tries to
model its effects on diffusion. It is ultimately necessary to identify and model
all the structures and processes, and to find their combined effect. To test the
model it is necessary to unravel these processes experimentally. Alternatively,
in the phenomenological approach of Feder et al (36), photobleaching and SPT
experiments are analyzed in terms of the parameters of the time-dependent
diffusion coefficient (Equation 7).
To see how the lateral diffusion of membrane proteins is controlled (86), one
must consider three regions of the protein: cytoplasmic, transmembrane, and
extracellular. The influence of these regions may vary among proteins and cell
types.
In some cases, a major factor affecting diffusion is the cytoplasmic domain
of the membrane protein and the membrane skeleton with which the domain
interacts. Truncation of the cytoplasmic domain of the class II MHC receptor
or erythrocyte band 3 increases the diffusion coefficient (27, 105). Kusumi and
collaborators (56) argue that the membrane skeleton is a universal feature of
cells, which defines short-range diffusion and controls long-range diffusion.
Webb and collaborators (103) have shown for a variety of proteins that lateral
diffusion in blebs is much faster than in intact membranes, suggesting direct or
indirect control of diffusion by the cytoskeleton. For a number of other proteins,
however, the membrane skeleton has little effect on diffusion. Severe truncation
of the cytoplasmic domain, or replacement of the transmembrane segment by
a GPI linkage, has little effect on diffusion coefficients (27, 105). In such
cases, either the membrane skeleton has little effect or the effect is indirect,
involving other membrane proteins that do interact with the membrane skeleton
directly (85).
In the transmembrane region, the area fraction of protein is likely to be
well below the percolation threshold, because many membrane proteins have
large cytoplasmic and extracellular domains, but only a single transmembrane
helix. Hammer, Koch, and collaborators (15–18, 26) evaluated the combined
effect of obstruction (68a) and hydrodynamic interactions with the obstacles.
They showed that low concentrations of immobile protein reduce the diffusion
coefficient of mobile proteins significantly. Their model assumes a macroscopic
object moving in a fluid continuum. The model is therefore restricted to proteins
with multiple membrane-spanning regions to ensure that the protein solute is
larger than the lipid solvent that forms the continuum. It will be important
to see whether the hydrodynamic model can be extrapolated to proteins with
a single transmembrane domain. In an unobstructed system, the Saffman-
Delbrück model predicts little difference in D for large and small proteins (1),
so it will be interesting to see whether or not there is a significant difference in
D as a result of hydrodynamic interactions with immobile obstacles. If these
interactions are significantly different, the diffusion of multichain proteins may
be controlled by hydrodynamic effects in the bilayer, but the diffusion of single-
chain and GPI-linked proteins in the same membrane may be controlled by the
extracellular domains. The effect of immobile domains of gel-phase lipid would
be similar to that of immobile proteins.
In some cases, there is compelling evidence of the importance of extracellular
domains in diffusion (31, 86, 105). Given the large size of the extracellular
domains, obstruction and percolation effects are likely to be dominant in this
region and are determined by the area fraction of extracellular domains and
extracellular matrix. More precisely, the effect of obstructions is determined

by the excluded area fraction, which depends on the radius of the particular
mobile protein (69). If the protein-protein interaction is simple electrostatic
repulsion, the effective excluded area fraction increases as either temperature
or ionic strength decreases. However, there may be specific complications.
Removal of charge can lower D, presumably by allowing association (32).
In addition to steric and hydrodynamic effects, protein-protein interactions
(62a) in any of the regions are likely to have a major effect on diffusion. Tran-
sient association is favored in membranes (32, 105), and diffusion is hindered

by the association of mobile species with either mobile or immobile species.
Anomalous diffusion is emerging as a key property in membrane dynamics,
both because it is a probe of membrane microstructure and because it has a major
influence on reaction kinetics within the membrane. Anomalous diffusion has
been observed by SPT, at least in local domains (42, 93, 95). It has not yet been
established experimentally whether anomalous diffusion in cell membranes
shows a crossover to normal diffusion at large times, or is anomalous at all
times. If diffusion is anomalous at all times, it could be the result of a single
mechanism, such as a singular distribution of trap depths as found in amorphous
semiconductors (75). Alternatively, it could originate from various biological
structures and processes with different length and time scales. For example,
over short distances, anomalous diffusion could result from obstruction and
binding. Over intermediate distances, it could result from corrals with a broad
distribution of escape times. Over long distances and times, it could result from
biological modulation, such as turnover of membrane components by cycles of
endocytosis and exocytosis (86), or changes in phosphorylation state.
In summary, the main factors controlling diffusion in the plasma membrane
are likely to be fluctuations in the membrane skeleton, obstruction and hydro-
dynamic interaction in the transmembrane region, and obstruction and perco-
lation in the extracellular region. These interactions depend on the structure of
the protein, including its size, charge, mode of membrane anchorage, degree of
oligomerization, and degree of glycosylation. All of these factors may combine
to produce the non-Brownian behavior that characterizes much of the lateral
motion in the membrane.
TECHNICAL PRIORITIES
If SPT is to reach its full potential, several technical issues need to be addressed:
(a) Calibration. Further work on the relation of FRAP and SPT is important.
A systematic study of D as a function of label size and valency in well-
defined synthetic bilayers and in cells would provide a firmer basis for the
interpretation of experimental results in cells.
(b) Cross-calibration. One of the problems in the field is that each laboratory
uses different cell types, membrane proteins, labels, and methods of data
analysis. It is therefore difficult to compare results among laboratories. Ex-
periments in all laboratories should include measurements of one standard
protein with one standard label in one standard cell line. In addition, where
possible, the motion of the standard protein and label should be measured
in the cell lines used in each laboratory, to provide a basis for comparison
among cell lines of the mechanisms that constrain diffusion. Similar mea-
surements with a standard lipid label should also be made. Laboratories

should exchange data analysis programs and use one anothers’ methods in
addition to their own. It would also be useful for the laboratories to agree
on standard methods for classifying modes of motion and report the results
of those methods in addition to their own classification.
(c) Hydrodynamic interactions. Hammer, Koch, and colleagues (15–18, 26)

have shown theoretically that hydrodynamic interactions have a major effect
on lateral diffusion coefficients. A small area fraction of immobile protein
reduces the diffusion coefficient significantly. The hydrodynamic effect
of immobile protein is much greater than that of mobile protein, just as
for obstruction (68a). Both effects are therefore much greater in plasma
membranes than in organelle or artificial membranes. An experimental
test of these results in a model system is crucial to our understanding of
membrane dynamics and would involve both FRAP and SPT.
(d) Improved data analysis and modeling. One area of particular interest is
lateral diffusion of proteins obstructed by the membrane skeleton, modeled
as a network of flexible polymers (9, 10). Another is Brownian dynamics
calculations of the hydrodynamic interactions just mentioned, to see their
effect on SPT trajectories and anomalous diffusion. Development of new
methods of data analysis and systematic comparison of existing methods
are also essential.
(e) Multiple-labeling experiments. The use of distinct fluorophores may make

it possible to track two reactants simultaneously, or a diffusing species
and an obstacle or receptor. This approach could also be used to exam-
ine whether corrals or regions of anomalous diffusion are semipermanent
features of the cell surface that different proteins can enter.
ACKNOWLEDGMENTS
We thank our colleagues for helpful discussions of various aspects of SPT
and membranes, and for supplying preprints, answering questions about their
results, and criticizing parts of this review: BG Barisas, K Berland, DH Boal,

RJ Cherry, M Edidin, H Geerts, DA Hammer, A Kusumi, DA Pink, H Qian,
Th Schmidt, ED Sheets, MP Sheetz, R Simson, and WW Webb. This work
was supported by National Institutes of Health grants GM38133 (MJS) and
GM41402 (KJ).
Visit the Annual Reviews home page at

http://www.annurev.org.
Literature Cited
1. Almeida PFF, Vaz WLC. 1995. Lat- 11. Bouchaud J-P, Georges A. 1990. Anoma-
eral diffusion in membranes. In Struc- lous diffusion in disordered media: sta-
ture and Dynamics of Membranes, ed. R tistical mechanisms, models and phys-
Lipowsky, E Sackmann, 1A:305–57. Am- ical applications. Phys. Rep. 195:127–

sterdam: Elsevier Sci. 293
2. Anderson CM, Georgiou GN, Morrison 12. Bray D. 1996. Membrane biophysics—
IEG, Stevenson GVW, Cherry RJ. 1992. the dynamics of growing axons. Curr.
Tracking of cell surface receptors by Biol. 6:241–43
fluorescence digital imaging microscopy 13. Bretscher MS. 1996. Moving membrane
using a charge-coupled device camera. up to the front of migrating cells. Cell
Low-density lipoprotein and influenza 85:465–67
virus receptor mobility at 4◦ C. J. Cell Sci. 14. Brown DA, Rose JK. 1992. Sorting
101:415–25 of GPI-anchored proteins to glycolipid-
3. Argyrakis P, Kopelman R. 1986. Fractal enriched membrane subdomains during
aspects of heterogeneous chemical reac- transport to the apical cell surface. Cell
tions. In Advances in Chemical Reaction 68:533–44
Dynamics, ed. PM Rentzepis, C Capellos, 15. Bussell SJ, Hammer DA, Koch DL. 1994.
pp. 339–44. Dordrecht: D Reidel The effect of hydrodynamic interactions
4. Barak LS, Webb WW. 1981. Fluorescent on the tracer and gradient diffusion of in-
low density lipoprotein for observation tegral membrane proteins in lipid bilay-
of dynamics of individual receptor com- ers. J. Fluid Mech. 258:167–90
plexes on cultured human fibroblasts. J. 16. Bussell SJ, Koch DL, Hammer DA. 1992.
Cell Biol. 90:595–604 The resistivity and mobility functions for
5. Barak LS, Webb WW. 1982. Diffusion a model system of two equal-sized pro-
of low density lipoprotein-receptor com- teins in a lipid bilayer. J. Fluid Mech.
plex on human fibroblasts. J. Cell Biol. 243:679–97
95:846–52 17. Bussell SJ, Koch DL, Hammer DA.
6. Beaurepaire E, Webb WW. 1995. Single 1995. Effect of hydrodynamic interac-
particle tracking: optimizing the local- tions on the diffusion of integral mem-
ization of fluorescence labeled molecules. brane proteins: tracer diffusion in or-
Biophys. J. 68:A288 (Abstr.) ganelle and reconstituted membranes.
7. Bergelson LD, Gawrisch K, Ferretti JA, Biophys. J. 68:1828–35
Blumenthal R, eds. 1995. Domain orga- 18. Bussell SJ, Koch DL, Hammer DA. 1995.
nization in biological membranes. Mol. Effect of hydrodynamic interactions on
Membr. Biol. 12(1):1–162 the diffusion of integral membrane pro-
8. Block SM. 1990. Optical tweezers: a new teins: diffusion in plasma membranes.
tool for biophysics. See Ref. 38, pp. 375– Biophys. J. 68:1836–49
402 19. Che A, Cherry RJ. 1995. Loss of rota-
9. Boal DH. 1994. Computer simulation of tional mobility of band 3 proteins in hu-
a model network for the erythrocyte cy- man erythrocyte membranes induced by
toskeleton. Biophys. J. 67:521–29 antibodies to glycophorin A. Biophys. J.
10. Boal DH, Boey SK. 1995. Barrier-free 68:1881–87
paths of directed protein motion in the 20. Cherry RJ, Georgiou GN, Morrison
erythrocyte plasma membrane. Biophys. IEG. 1994. New insights into the struc-
J. 69:372–79 ture of cell membranes from single
particle tracking experiments. Biochem. 33. Edidin M, Kuo SC, Sheetz MP. 1991. Lat-
Soc. Trans. 22:781–84 eral movements of membrane glycopro-
21. Cogswell CJ. 1995. Imaging immuno- teins restricted by dynamic cytoplasmic
gold labels with confocal microscopy. In barriers. Science 254:1379–82
Handbook of Biological Confocal Mi- 34. Edidin M, Zúñiga MC, Sheetz MP. 1994.
croscopy, ed. JB Pawley, pp. 507–13. New Truncation mutants define and locate
York: Plenum. 2nd ed. cytoplasmic barriers to lateral mobility
21a. Crocker JC, Grier DG. 1996. Methods of membrane glycoproteins. Proc. Natl.
of digital video microscopy for colloidal Acad. Sci. USA 91:3378–82
studies. J. Colloid Interface Sci. 179:298– 35. Evans E, Merkel R, Ritchie K, Tha S,
310 Zilker A. 1994. Picoforce method to probe
22. Dai J, Sheetz MP. 1995. Axon membrane submicroscopic actions in biomembrane
flows from the growth cone to the cell adhesion. In Studying Cell Adhesion, ed.
body. Cell 83:693–701 P Bongrand, PM Claesson, ASG Curtis,
23. De Brabander M, Geerts H, Nuydens R, pp. 125–39. Berlin: Springer-Verlag
Nuydens R, Cornelissen F. 1993. Nanovid 36. Feder TJ, Brust-Mascher I, Slattery JP,
microscopy: imaging and quantification Baird B, Webb WW. 1996. Constrained
of colloidal gold labels in living cells. See diffusion or immobile fraction on cell sur-
Ref. 92, pp. 141–55
faces: a new interpretation. Biophys. J.

24. De Brabander M, Geuens G, Nuydens 70:2767–73
R, Moeremans M, De Mey J. 1985. 37. Fein M, Unkeless J, Chuang FYS, Sas-
Probing microtubule-dependent intracel- saroli M, da Costa R, et al. 1993. Lat-
lular motility with nanometre particle eral mobility of lipid analogues and GPI-
video ultramicroscopy (nanovid ultrami- anchored proteins in supported bilayers
croscopy). Cytobios 43:273–83 determined by fluorescent bead tracking.
25. De Brabander M, Nuydens R, Ishihara A, J. Membr. Biol. 135:83–92
Holifield B, Jacobson K, Geerts H. 1991. 38. Foskett JK, Grinstein S, eds. 1990. Non-
Lateral diffusion and retrograde move- invasive Techniques in Cell Biology. New
ments of individual cell surface com- York: Wiley-Liss. 423 pp.
ponents on single motile cells observed 39. Fredrickson GH. 1995. Analytical solu-
with Nanovid microscopy. J. Cell Biol. tion of a model of integrin-cytoskeletal
112:111–24 interactions in migrating fibroblasts. J.
26. Dodd TL, Hammer DA, Sangani AS, Phys. II (Paris) 5:369–76
Koch DL. 1995. Numerical simulations
of the effect of hydrodynamic interactions 40. Futerman AH, Banker GA. 1996. The
on diffusivities of integral membrane pro- economics of neurite outgrowth—the ad-
teins. J. Fluid Mech. 293:147–80 dition of new membrane to growing ax-
27. Edidin M. 1990. Molecular associa- ons. Trends Neurosci. 19:144–49
tions and membrane domains. Curr. Top. 41. Gelles J, Schnapp BJ, Sheetz MP.
Membr. Transp. 36:81–96 1988. Tracking kinesin-driven move-
28. Edidin M. 1992. Translational diffusion ments with nanometre-scale precision.
of membrane proteins. In The Structure Nature 331:450–53
of Biological Membranes, ed. P Yeagle, 42. Ghosh RN. 1991. Mobility and cluster-
pp. 539–72. Boca Raton: CRC ing of individual low-density lipoprotein
29. Edidin M. 1993. Patches and fences: receptor molecules on the surface of hu-
probing for plasma membrane domains. man skin fibroblasts. PhD thesis. Cornell
J. Cell Sci. Suppl. 17:165–69 Univ., Ithaca. 260 pp.
30. Edidin M. 1994. Lateral mobility of 43. Ghosh RN, Webb WW. 1994. Auto-
membrane proteins—a journey from het- mated detection and tracking of individ-
erokaryons to laser tweezers. In The ual and clustered cell surface low density
Legacy of Cell Fusion, ed. S Gordon, pp. lipoprotein receptor molecules. Biophys.
101–14. Oxford: Oxford Univ. Press J. 66:1301–18
31. Edidin M. 1994. Fluorescence photo- 44. Grasberger B, Minton AP, DeLisi C,
bleaching and recovery, FPR, in the anal- Metzger H. 1986. Interaction between
ysis of membrane structure and dynam- proteins localized in membranes. Proc.
ics. In Mobility and Proximity in Biolog- Natl. Acad. Sci. USA 83:6258–62
ical Membranes, ed. S Damjanovich, J 45. Gross D, Webb WW. 1986. Molecular
Szöllősi, L Trón, M Edidin, pp. 109–35. counting of low-density lipoprotein par-
Boca Raton: CRC ticles as individuals and small clusters
32. Edidin M. 1996. Getting there is only half on cell surfaces. Biophys. J. 49:901–
the fun. Curr. Top. Membr. 43:1–13 11
46. Hicks BW, Angelides KJ. 1995. Tracking 60. Lee GM, Zhang F, Ishihara A, McNeil CL,
movements of lipids and Thy1 molecules Jacobson KA. 1993. Unconfined lateral
in the plasmalemma of living fibroblasts diffusion and an estimate of pericellular
by fluorescence video microscopy with matrix viscosity revealed by measuring
nanometer scale precision. J. Membr. the mobility of gold-tagged lipids. J. Cell
Biol. 144:231–44 Biol. 120:25–35
47. Deleted in proof 61. Lee J, Gustafsson M, Magnusson K-
48. Inoué S. 1986. Video Microscopy. New E, Jacobson K. 1990. The direction
York: Plenum. 584 pp. of membrane lipid flow in locomoting
49. Inoué S. 1989. Imaging of unresolved polymorphonuclear leukocytes. Science
objects, superresolution, and precision 247:1229–33
of distance measurement with video mi- 62. ÃL uczka J, Niemiec M, Hänggi P. 1995.
croscopy. Methods Cell Biol. 30:85–112 First-passage time for randomly flashing
50. Ishihara A, Hou Y, Jacobson K. 1987. The diffusion. Phys. Rev. E 52:5810–16
Thy-1 antigen exhibits rapid lateral dif- 62a. Mouritsen OG, Bloom M. 1993. Mod-
fusion in the plasma membrane of rodent els of lipid-protein interactions in mem-
lymphoid cells and fibroblasts. Proc. Natl. branes. Annu. Rev. Biophys. Biomol.
Acad. Sci. USA 84:1290–93 Struct. 22:145–71
51. Jacobson K, O’Dell D, August JT. 1984. 63. Nagle JF. 1992. Long tail kinetics in bio-
Lateral diffusion of an 80,000-dalton physics? Biophys. J. 63:366–70

glycoprotein in the plasma membrane 64. Parton RG, Simons K. 1995. Digging into
of murine fibroblasts: relationships to caveolae. Science 269:1398–99
cell structure and function. J. Cell Biol. 65. Qian H, Sheetz MP, Elson EL. 1991. Sin-
99:1624–33 gle particle tracking. Analysis of diffu-
52. Knowles DW, Chasis JA, Evans EA, sion and flow in two-dimensional sys-
Mohandas N. 1994. Cooperative ac- tems. Biophys. J. 60:910–21
tion between band 3 and glycophorin 66. Rudnick J, Gaspari G. 1987. The shapes
A in human erythrocytes: immobiliza- of random walks. Science 237:384–
tion of band 3 induced by antibodies 89
to glycophorin A. Biophys. J. 66:1726– 67. Sako Y, Kusumi A. 1994. Compartmen-
32 talized structure of the plasma membrane
53. Koppel DE, Morgan F, Cowan AE, Car- for receptor movements as revealed by a
son JH. 1994. Scanning concentration nanometer-level motion analysis. J. Cell
correlation spectroscopy using the confo- Biol. 125:1251–64
cal laser microscope. Biophys. J. 66:502– 68. Sako Y, Kusumi A. 1995. Barriers for lat-
7 eral diffusion of transferrin receptor in the
54. Kucik DF, Elson EL, Sheetz MP. 1989. plasma membrane as characterized by re-
Forward transport of glycoproteins on ceptor dragging by laser tweezers: fence
leading lamellipodia in locomoting cells. versus tether. J. Cell Biol. 129:1559–74
Nature 340:315–17 68a. Saxton MJ. 1990. Lateral diffusion in a
55. Kucik DF, Elson EL, Sheetz MP. 1990. mixture of mobile and immobile parti-
Cell migration does not produce mem- cles: a Monte Carlo study. Biophys. J. 58:
brane flow. J. Cell Biol. 111:1617–22 1303–6
56. Kusumi A, Sako Y. 1996. Cell surface 69. Saxton MJ. 1993. Lateral diffusion in an
organization by the membrane skeleton. archipelago: dependence on tracer size.
Curr. Opin. Cell Biol. 8:566–74 Biophys. J. 64:1053–62
57. Kusumi A, Sako Y, Yamamoto M. 1993. 70. Saxton MJ. 1993. Lateral diffusion in
Confined lateral diffusion of membrane an archipelago: single-particle diffusion.
receptors as studied by single parti- Biophys. J. 64:1766–80
cle tracking (nanovid microscopy). Ef- 71. Saxton MJ. 1994. Anomalous diffusion
fects of calcium-induced differentiation due to obstacles: a Monte Carlo study.
in cultured epithelial cells. Biophys. J. Biophys. J. 66:394–401
65:2021–40 72. Saxton MJ. 1994. Single-particle track-
58. Lee GM, Ishihara A, Jacobson KA. 1991. ing: models of directed transport. Bio-
Direct observation of Brownian motion of phys. J. 67:2110–19
lipids in a membrane. Proc. Natl. Acad. 73. Saxton MJ. 1995. Single-particle track-
Sci. USA 88:6274–78 ing: effects of corrals. Biophys. J.
59. Lee GM, Johnstone B, Jacobson K, Cater- 69:389–98
son B. 1993. The dynamic structure of the 74. Saxton MJ. 1996. Anomalous diffusion
pericellular matrix on living cells. J. Cell due to binding: a Monte Carlo study. Bio-
Biol. 123:1899–907 phys. J. 70:1250–62
75. Scher H, Shlesinger MF, Bendler JT. and dynamic domains in fluid plasma
1991. Time-scale invariance in transport membranes. Annu. Rev. Biophys. Biomol.
and relaxation. Phys. Today 44(1):26–34 Struct. 22:417–31
76. Schmidt CE, Chen T, Lauffenburger DA. 87. Sheetz MP, Baumrind NL, Wayne DB,
1994. Simulation of integrin-cytoskeletal Pearlman AL. 1990. Concentration of
interactions in migrating fibroblasts. Bio- membrane antigens by forward transport
phys. J. 67:461–74 and trapping in neuronal growth cones.
77. Schmidt CE, Horwitz AF, Lauffen- Cell 61:231–41
burger DA, Sheetz MP. 1993. Integrin- 88. Sheetz MP, Elson EL. 1993. Measurement
cytoskeletal interactions in migrating of membrane glycoprotein movement by
fibroblasts are dynamic, asymmetric, single-particle tracking. In Optical Mi-
and regulated. J. Cell Biol. 123:977– croscopy: Emerging Methods and Appli-
91 cations, ed. B Herman, JJ Lemasters, pp.
78. Schmidt Th, Schütz GJ, Baumgartner W, 285–94. San Diego: Academic
Gruber HJ, Schindler H. 1995. Char- 89. Sheetz MP, Kuo SC. 1993. Tracking
acterization of photophysics and mobil- nanometer movements of single motor
ity of single molecules in a fluid lipid molecules. Methods Cell Biol. 39:129–36
membrane. J. Phys. Chem. 99:17662– 90. Sheetz MP, Turney S, Qian H, Elson EL.
68 1989. Nanometre-level analysis demon-
79. Schmidt Th, Schütz GJ, Baumgartner W, strates that lipid flow does not drive mem-
Gruber HJ, Schindler H. 1996. Imaging brane glycoprotein movements. Nature
of single molecule diffusion. Proc. Natl. 340:284–88
Acad. Sci. USA 93:2926–29 91. Shiozawa JA, Brandts JF, Jacobson BS.
80. Schmidt Th, Schütz GJ, Gruber HJ, 1989. Binding of plasma membrane gly-
Schindler H. 1996. Non-random motion coproteins to the cytoskeleton during
in a lipid membrane observed on a sin- patching and capping is consistent with
gle molecule level. Biophys. J. 70:A426 an entropy-enhancement model. Biochim.
(Abstr.) Biophys. Acta 980:361–66
81. Schnapp BJ, Gelles J, Sheetz MP. 92. Shotton D. 1993. Electronic Light Mi-
1988. Nanometer-scale measurements us- croscopy. New York: Wiley-Liss. 355 pp.
ing video light microscopy. Cell Motil. 93. Simson R. 1994. Different modes of lat-
Cytoskel. 10:47–53 eral mobility for neural cell adhesion
82. Schnitzer JE, McIntosh DP, Dvorak AM, molecules as observed by nanovid mi-
Liu J, Oh P. 1995. Separation of caveolae croscopy and single particle tracking.
from associated microdomains of GPI- Diplomarbeit [thesis]. Technische Univ.
anchored proteins. Science 269:1435– München. 78 pp.
39 94. Simson R, Sheets ED, Jacobson K. 1995.
83. Schroeder R, London E, Brown D. Detection of temporary lateral confine-
1994. Interactions between saturated acyl ment of membrane proteins using single-
chains confer detergent resistance on particle tracking analysis. Biophys. J.
lipids and glycosylphosphatidylinositol 69:989–93
(GPI)-anchored proteins: GPI-anchored 95. Slattery JP. 1995. Lateral mobility of
proteins in liposomes and cells show sim- FcRI on rat basophilic leukemia cells as
ilar behavior. Proc. Natl. Acad. Sci. USA measured by single particle tracking us-
91:12130–34 ing a novel bright fluorescent probe. PhD
84. Sheets ED, Jacobson K. 1996. GPI- thesis. Cornell Univ., Ithaca. 153 pp.
anchored proteins and glycosphingolipids 96. Svoboda K, Block SM. 1994. Biological
exhibit transient lateral confinement to applications of optical forces. Annu. Rev.
small domains: a single particle tracking Biophys. Biomol. Struct. 23:247–85
study. Biophys. J. 70:A335 (Abstr.) 97. Svoboda K, Block SM. 1994. Optical
85. Sheets ED, Simson R, Jacobson K. 1995. trapping of metallic Rayleigh particles.
New insights into membrane dynamics Opt. Lett. 19:930–32
from the analysis of cell surface inter- 98. Thomas J, Webb WW. 1990. Fluores-
actions by physical methods. Curr. Opin. cence photobleaching recovery: a probe
Cell Biol. 7:707–14 of membrane dynamics. See Ref. 38, pp.
85a. Sheetz MP. 1983. Membrane skeletal dy- 129–52
namics: role in modulation of red cell de- 99. Thompson TE, Sankaram MB, Biltonen
formability, mobility of transmembrane RL, Marsh D, Vaz WLC. 1995. Effects
proteins, and shape. Semin. Hematol. of domain structure on in-plane reac-
20:175–88 tions and interactions. Mol. Membr. Biol.
86. Sheetz MP. 1993. Glycoprotein motility 12:157–62
100. Tsay TT, Inman R, Wray B, Herman B, 103. Webb WW, Barak LS, Tank DW, Wu
Jacobson K. 1990. Characterization of ES. 1981. Molecular mobility on the cell
low-light-level cameras for digitized surface. Biochem. Soc. Symp. 46:191–
video microscopy. J. Microsc. (Oxford) 205
160:141–59 104. Wilson KM, Morrison IEG, Smith PR,
101. Wade WF, Freed JH, Edidin M. 1989. Fernandez N, Cherry RJ. 1996. Single
Translational diffusion of class II ma- particle tracking of cell-surface HLA-DR
jor histocompatibility complex molecules molecules using R-phycoerythrin labeled
is constrained by their cytoplasmic do- monoclonal antibodies and fluorescence
mains. J. Cell Biol. 109:3325–31 digital imaging. J. Cell Sci. 109:2101–9
102. Wang YL, Silverman JD, Cao LG. 1994. 105. Zhang F, Lee GM, Jacobson K. 1993.
Single particle tracking of surface recep- Protein lateral mobility as a reflection

tor movement during cell division. J. Cell of membrane microstructure. BioEssays
Biol. 127:963–71 15:579–88
Annual Review of Biophysics and Biomolecular Structure
Volume 26, 1997
CONTENTS
WHATEVER HAPPENED TO THE FUN? An Autobiographical
Investigation, Frederic M. Richards 1
STRUCTURAL AND MECHANISTIC DETERMINANTS OF
AFFINITY AND SPECIFICITY OF LIGANDS DISCOVERED OR
ENGINEERED BY PHAGE DISPLAY, Bradley A. Katz 27
CALCIUM IN CLOSE QUARTERS: Microdomain Feedback in
Excitation-Contraction Coupling and Other Cell Biological Phenomena,
Eduardo Ríos, Michael D. Stern 47
HISTONE STRUCTURE AND THE ORGANIZATION OF THE

NUCLEOSOME, V. Ramakrishnan 83
HIERARCHY AND DYNAMICS OF RNA FOLDING, Philippe
Brion, Eric Westhof 113

FLEXIBILITY OF RNA, Paul J. Hagerman 139
STRUCTURAL PERSPECTIVES OF PHOSPHOLAMBAN, A

HELICAL TRANSMEMBRANE PENTAMER, Isaiah T. Arkin, Paul
D. Adams, Axel T. Brünger, Steven O. Smith, Donald M. Engelman 157
BIOMOLECULAR DYNAMICS AT LONG TIMESTEPS: Bridging

the Timescale Gap Between Simulation and Experimentation, Tamar
Schlick, Eric Barth, Margaret Mandziuk 181
MOLECULAR MECHANISM OF PHOTOSIGNALING BY
ARCHAEAL SENSORY RHODOPSINS, Wouter D. Hoff, Kwang-
Hwan Jung, John L. Spudich 223
MODULAR PEPTIDE RECOGNITION DOMAINS IN
EUKARYOTIC SIGNALING, John Kuriyan, David Cowburn 259
EUKARYOTIC TRANSCRIPTION FACTOR-DNA COMPLEXES,
G. Patikoglou, S. K. Burley 289
NANOSECOND TIME-RESOLVED SPECTROSCOPY OF
BIOMOLECULAR PROCESSES, Eefei Chen, Robert A. Goldbeck,
David S. Kliger 327
LESSONS FROM ZINC-BINDING PEPTIDES, Jeremy M. Berg,
Hilary Arnold Godwin 357
SINGLE-PARTICLE TRACKING: Applications to Membrane
Dynamics, Michael J. Saxton, Ken Jacobson 373
BACTERIOPHAGE DISPLAY AND DISCOVERY OF PEPTIDE
LEADS FOR DRUG DEVELOPMENT, H. B. Lowman 401
SOLVATION: HOW TO OBTAIN MICROSCOPIC ENERGIES
FROM PARTITIONING AND SOLVATION EXPERIMENTS, Hue
Sun Chan, Ken A. Dill 425
THE STRUCTURAL AND FUNCTIONAL BASIS OF ANTIBODY
CATALYSIS, Herschel Wade, Thomas S. Scanlan 461
ADVANCED EPR SPECTROSCOPY ON ELECTRON TRANSFER
PROCESSES IN PHOTOSYNTHESIS AND BIOMIMETIC MODEL
SYSTEMS, H. Levanon, K. Möbius 495
USE OF SURFACE PLASMON RESONANCE TO PROBE THE

EQUILIBRIUM AND DYNAMIC ASPECTS OF INTERACTIONS
BETWEEN BIOLOGICAL MACROMOLECULES, Peter Schuck 541
OPTICAL DETECTION OF SINGLE MOLECULES, Shuming Nie,
Richard N. Zare 567
PROTEIN FOLDS IN THE ALL-ß AND ALL alpha- Classes, Cyrus

Chothia, Tim Hubbard, Steven Brenner, Hugh Barns, Alexey Murzin 597
SITE-SPECIFIC DYNAMICS IN DNA: Experiments, Bruce H.
Robinson, Colin Mailer, Gary Drobny 629


Single Partical Tracking Review

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Single Partical Tracking Review

Uploaded by

Copyright:

Available Formats

P1: NGM/rpk P2: sny/rpk QC: rpk/BS T1: rpk

March 31, 1997 19:49 Annual Reviews AR031-14 AR031-14

Annu. Rev. Biophys. Biomol. Struct. 1997. 26:373–99

Department of Cell Biology and Anatomy, University of North Carolina at Chapel

KEY WORDS: single-particle tracking, fluorescence recovery after photobleaching, lateral

374 SAXTON & JACOBSON

(c) What are the effects of heterogeneous motion in a heterogeneous environ-

SPT IN MEMBRANES 375

but SPT measures individual trajectories. Thus, different subpopulations indis-

Figure 1 The mean-square displacement hr 2 i as a function of time t for simultaneous diffusion

376 SAXTON & JACOBSON

a highly scattering colloidal gold label is used with bright-field microscopy

SPT IN MEMBRANES 377

almost invisible in bright-field microscopy (93). The use of gold labels is

378 SAXTON & JACOBSON

SPT IN MEMBRANES 379

long-range ones. Averaging should not be done automatically because it may

hr 2 i = 4Dt normal diffusion (1)

anomalous diffusion (2)

In Equation 2, α < 1, so strictly speaking this is anomalous subdiffusion (11,

380 SAXTON & JACOBSON

Next, we summarize methods used to classify trajectories. Whatever the

Classification of Modes of Motion

SPT IN MEMBRANES 381

Table 1 Artificial bilayersa

Biotin-PE 30 nm FM 80% egg PC 30 — 37

among the modes. Often, simple diffusion is observed in only a minority of

Table 2 Lipids and GPI-linked proteins in cellsa

Membrane Apparent D(SPT)

component Cell Label for mobile fraction D(FRAP) Comments Ref.

biotin-PE Fibroblasts 100 nm FM 8 — Unspecified fraction stationary 46

Ab-Fl along axon length

Glycolipid Fibroblasts 40 nm Au 7.1 ± 1.3 — 18% slow diffusion 84

Ab ∼70% mobile D(slow) ∼ 0.04

Table 3 Transmembrane proteins in cellsa

Membrane Apparent D(SPT)

Cell adhesion molecules

E-cadherin Cultured 40 nm Au 0.16 0.34 High Ca+2 medium 57

NCAM180 Fibroblasts 30 nm Au 3.7 ± 0.18 18.3 ± 3.1 14% slow diffusion 93

Major histocompatibility antigens

384 SAXTON & JACOBSON

anomalous diffusion equation (parameters 0 and α) and the standard equation

SPT IN MEMBRANES 385

the structures that produce anomalous diffusion generally require metabolic

dogenous NCAM. For both GPI-linked and transmembrane isoforms, 50% of

386 SAXTON & JACOBSON

SPT IN MEMBRANES 387

388 SAXTON & JACOBSON

SPT IN MEMBRANES 389

390 SAXTON & JACOBSON

SPT and FRAP: Effects of the Label

An alternative approach uses a time-dependent D with no immobile fraction to

SPT IN MEMBRANES 391

with an 80% mobile fraction. Measurements using the antibody conjugated to

WHAT DIFFUSION TELLS US ABOUT

392 SAXTON & JACOBSON

SPT IN MEMBRANES 393

extracellular matrix. More precisely, the effect of obstructions is determined

sient association is favored in membranes (32, 105), and diffusion is hindered

394 SAXTON & JACOBSON

surements with a standard lipid label should also be made. Laboratories

(c) Hydrodynamic interactions. Hammer, Koch, and colleagues (15–18, 26)

(e) Multiple-labeling experiments. The use of distinct fluorophores may make

SPT IN MEMBRANES 395