4004 Biophysical Journal Volume 86 June 2004 4004–4014

Colloid Surface Chemistry Critically Affects Multiple Particle Tracking

Measurements of Biomaterials

M. T. Valentine,* Z. E. Perlman,yz M. L. Gardel,* J. H. Shin,§ P. Matsudaira,{

T. J. Mitchison,y and D. A. Weitz*
*Department of Physics and Division of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts 02138;
y
Department of Cell Biology and Institute of Chemistry and Cell Biology, Harvard Medical School, Boston, Massachusetts 02115;
z
Program in Biophysics, Harvard University, Cambridge, Massachusetts 02138; §Department of Mechanical Engineering,
Massachusetts Institute of Technology, Cambridge, Massachusetts 02142; and {Department of Biology, Massachusetts Institute
of Technology and the Whitehead Institute for Biomedical Research, Cambridge, Massachusetts 02142

ABSTRACT Characterization of the properties of complex biomaterials using microrheological techniques has the promise of
providing fundamental insights into their biomechanical functions; however, precise interpretations of such measurements are
hindered by inadequate characterization of the interactions between tracers and the networks they probe. We here show that
colloid surface chemistry can profoundly affect multiple particle tracking measurements of networks of fibrin, entangled F-actin
solutions, and networks of cross-linked F-actin. We present a simple protocol to render the surface of colloidal probe particles
protein-resistant by grafting short amine-terminated methoxy-poly(ethylene glycol) to the surface of carboxylated microspheres.
We demonstrate that these poly(ethylene glycol)-coated tracers adsorb significantly less protein than particles coated with
bovine serum albumin or unmodified probe particles. We establish that varying particle surface chemistry selectively tunes the
sensitivity of the particles to different physical properties of their microenvironments. Specifically, particles that are weakly
bound to a heterogeneous network are sensitive to changes in network stiffness, whereas protein-resistant tracers measure
changes in the viscosity of the fluid and in the network microstructure. We demonstrate experimentally that two-particle
microrheology analysis significantly reduces differences arising from tracer surface chemistry, indicating that modifications of
network properties near the particle do not introduce large-scale heterogeneities. Our results establish that controlling colloid-
protein interactions is crucial to the successful application of multiple particle tracking techniques to reconstituted protein
networks, cytoplasm, and cells.

INTRODUCTION
Understanding the unique microscopic mechanical proper- network. This provides the opportunity to probe many
ties of biopolymer networks is essential to understanding aspects of complex biomaterials; however, because the de-
their biological function. For example, fibrin networks tails of network architecture and molecular composition are
assemble at wounds and provide the critical scaffolding often not known a priori, these multiple sensitivities can
upon which healing and tissue repair occur, and actin also lead to ambiguities in the interpretation of these data.
networks in the cytoskeleton guide cell shape, motility, and In one simplifying limit, when the chemical interactions
division (Clark, 2001; Howard, 2001; Boal, 2002). The me- between the tracers and polymers are negligible, it is possible
chanical properties of networks such as these are determined to characterize tracer motions based only on the size of the
by structures and dynamics on the scale of microns, and so probe relative to the native structural sizes of the network.
precise tools to probe network response on these length Varying probe diameter with respect to these native lengths
scales are required. As a result, microrheology techniques gives qualitatively different information about network
have been developed to measure the microscopic viscoelastic mechanics. When thermally driven embedded probe par-
and mechanical properties of soft materials by analyzing the ticles are large compared to all structural length scales, as
motion of embedded tracers driven by thermal (Mason and illustrated in the sketch in Fig. 1 A, the ensemble-averaged
Weitz, 1995; Gittes et al., 1997; Mason et al., 1997a,b), mag- mean-squared displacement (MSD) is directly related to the
netic (Amblard et al., 1996b; Schmidt et al., 1996; Ziemann linear frequency-dependent viscous and elastic moduli using
et al., 1994), or optical forces (Valentine et al., 1996; Hough a generalized Stokes-Einstein relation (Mason and Weitz,
and Ou-Yang, 1999). In these measurements, tracer displace- 1995; Gittes et al., 1997; Mason et al., 1997a,b; Dasgupta
ments are sensitive to local viscous and elastic forces, as well et al., 2002). By contrast, when the embedded particles are
as chemical and steric interactions between the particle and approximately equal to or smaller than the structural length
scales of the material, as sketched in Fig. 1 B, particles move
within small, mechanically distinct microenvironments and
Submitted November 29, 2003, and accepted for publication February 23, their dynamics are no longer directly related to the bulk
2004. viscoelastic response (Valentine et al., 2001; Wong et al.,
Address reprint requests to Prof. David A Weitz, Physics Department, 2004). Rather, their Brownian movements are sensitive to
Harvard University, 29 Oxford St., Cambridge, MA 02138. Tel.: the viscosity of the solvent, the effects of macromolecular
617-496-2842; E-mail: [email protected].

2004 by the Biophysical Society
0006-3495/04/06/4004/11 $2.00 doi: 10.1529/biophysj.103.037812
Colloid-Protein Interactions Affect MPT 4005

crowding, and steric and hydrodynamic interactions with the ities. Alternatively, for measurements of the microenviron-
network (Jones and Luby-Phelps, 1996; Luby-Phelps, 2000; ments of a heterogeneous material with tracers that are
Chen et al., 2003). Since many biopolymer networks are smaller than the structural length scales, even a small amount
structured on micron length scales, measurements can be of protein adsorption can cause particles to adhere to cavity
extremely sensitive to even small changes in the tracer dia- walls, preventing them from fully exploring small pores and
meter or the mesh size of the network. Careful studies of possibly even inducing local changes in network structure,
tracer displacements as a function of probe size are thus often leading to uncertainties in the interpretation of particle dy-
required. namics, as sketched in Fig. 1, C and D.
In general, chemical interactions between tracers and One approach that has been developed to address some of
protein networks cannot be ignored and introduce large these limitations is two-particle microrheology, in which the
ambiguities in the interpretation of network responses. Thus, correlated motion of pairs of particles is analyzed to measure
the ability to characterize and modify surface interactions, as the long wavelength deformation of a network (Crocker et al.,
well as tracer size, is crucial to proper interpretation of 2000; Levine and Lubensky, 2000). Correlated motions arise
measurements of complex biomaterials. For microrheology only from fluctuations on length scales greater than the
measurements of network viscoelasticity for example, the interparticle separation distance; thus two-particle micro-
particles must not only be large in comparison to all rheology is insensitive to local heterogeneities and promises
structural length scales, but must also be sufficiently resistant to be independent of the coupling of the particle to the
to protein adsorption to prevent the local modification of medium. The ability to isolate the long wavelength network
network architecture and introduction of small heterogene- response is a significant advance. However to fully character-
ize the microscopic properties of complex materials, the
ability to probe network mechanics at the length scale of the
particle is desirable. Therefore, a full study of the effect of
particle surface chemistry on tracer movements in bio-
polymer gels is required.
Ideally, it would be possible to tune tracer surface chem-
istry to prevent undesirable protein adsorption yet allow
specific and controllable binding; however, such surfaces are
not commercially available, and studies of the effect of
protein binding on particle mobility are limited. Designing
tracers with tunable adsorption properties requires knowl-
edge of the extremely complex and poorly understood
microscopic surface interactions between colloids and
proteins (Haynes and Norde, 1994). Since the molecular
level details are unknown, one common approach to reduce
the nonspecific adsorption of a particular protein is to simply
preincubate the colloids with another protein solution that is
known to have a high surface affinity, such as bovine serum
albumin (BSA), to physically block any available protein
binding sites. Protein coatings such as these have been
shown to reduce the adsorption of some proteins to colloidal
surfaces, and in many cases BSA-coated particles provide an
FIGURE 1 Sketch illustrating several physical scenarios for the way
excellent alternative to bare tracers. For example, previous
colloidal particles can be embedded in a biopolymer network. (A) When
chemically inert particles of radius a
network mesh size j are used, the measurements have demonstrated that BSA-coated particles
mean-squared displacement is directly related to the linear viscoelastic are more mobile than bare carboxylate-modified latex
moduli. (B) When particles are resistant to protein adsorption and a # j, (CML) spheres when used in micromechanical measure-
tracers move within small microenvironments and their movements are ments of cross-linked networks of actin; however, the
sensitive to the viscosity of the solvent and hydrodynamic interactions with
adsorbed BSA monolayer is patchy and does not render the
the network, but do not reflect the bulk viscoelasticity. (C) For a # j,
‘‘sticky’’ tracers adsorb protein and recruit polymer strands to their surface, particles completely resistant to additional protein adsorption
possibly modifying the local polymer concentration near the sphere. In this (McGrath et al., 2000). Moreover, because the protein layer
case, particle movements do reflect network fluctuations; however, the is simply adsorbed, and not covalently bound, some de-
tracers may sample unusually and artifactually stiff regions of the network, sorption of the BSA from the surface is possible, com-
leading to uncertainty in the interpretation of relationship of the particle
plicating interpretations of particle dynamics (McGrath et al.,
dynamics to the network dynamics. (D) For a # j, even a small amount of
protein adsorption can cause particles to adhere to cavity walls, leading to 2000).
unusual hydrodynamic interactions with the adsorbed network, and As an alternative to BSA coating, a dense monolayer of
uncertainty in the interpretation of these data. poly(ethylene glycol) (PEG) has been shown empirically to

Biophysical Journal 86(6) 4004–4014

4006 Valentine et al.

reduce the adsorption of a wide variety of proteins when approximate physiologically relevant actin gels, which are
grafted onto flat surfaces (Prime and Whitesides, 1991, typically cross-linked or bundled by actin-binding proteins
1993; Ostuni et al., 2001; Harder et al., 1998). The origin in vivo.
of protein resistance in PEG-coated surfaces is generally With these measurements, we demonstrate that, by
attributed to steric repulsion effects arising from both the varying particle surface modifications, we can selectively
loss of conformational entropy of the polymer as the brush tune the sensitivity of the particles to the different physical
is compressed and the disfavorable desolvation of the properties of the biomaterials they probe. In particular, we
chains as water molecules are expelled from the polymer find that particles that are weakly bound to a heterogeneous
layer (de Gennes, 1987; Jeon et al., 1991; Grunze et al., network are sensitive to changes in local stiffness, whereas
1998). Despite the success of PEG-coated flat surfaces, the the protein resistant PEG-coated tracers are sensitive to
adsorption properties of alternative geometries have not changes in viscosity and microstructure. Additionally, we
been fully explored, as PEG-coated colloidal particles are provide an experimental demonstration that local variations
not readily available commercially and are not widely used in network mechanics introduced by protein adsorption are
(Weisbecker et al., 1996; Liu et al., 1997; Shay et al., uncorrelated over large distances, and thus two-particle
2000, 2001; De Sousa Delgado et al., 2001; McGrath et al., microrheology essentially eliminates the differences due to
2003). surface chemistry effects.
In this article, we develop a simple and robust protocol to
graft short amine-terminated methoxy-poly(ethylene glycol) MATERIALS AND METHODS
to the surface of carboxylate-modified colloids using com-
Preparing PEG-coated particles
mercially available reagents. We demonstrate that PEG-
coated colloids adsorb significantly less protein than bare To create the PEG-coated particles, we attach amine-terminated methoxy-
CML spheres or probes coated with physisorbed BSA, and poly(ethylene glycol), (mPEG-NH2), NH2-(CH2-CH2-O)n-OCH3, where, on
average, n ¼ 16, resulting in an average molecular mass of 750 Da (Rapp
use multiple particle tracking techniques to measure the Polymere, Tübingen, Germany), to CML particles (Molecular Probes,
Brownian movements of these tracers embedded in bio- Eugene, OR, or Interfacial Dynamics, Portland, OR) using standard
polymer networks. We choose three protein systems that carbodiimide coupling chemistry. In this reaction, the carboxylic acids are
display different adsorption characteristics and whose activated by the formation at low pH of a reactive N-hydroxysuccinimide
mechanical properties are critical to their biological function. (NHS) ester, proceeding through the formation of a less stable ester with
1-[3-(dimethylamino)propyl]-3-ethylcarbodiimide (EDC) (Sigma-Aldrich,
We first choose fibrinogen, a globular protein that has St. Louis, MO). The activated esters are then mixed with mPEG-NH2 at
a very strong surface affinity and is thus commonly used higher pH to accelerate deprotonation of the amine, which reacts with the
to investigate protein-surface interactions (Prime and NHS-ester to yield a stable amide bond.
Whitesides, 1993; Harder et al., 1998; Lahiri et al., 1999; To minimize the probability of colloidal aggregation, we perform all
Alcantar et al., 2000; Chapman et al., 2000; Ostuni et al., buffer exchanges slowly using dialysis tubing, rather than the more
traditional centrifugation or filtration techniques. Also, we foreshorten the
2001; Valiokas et al., 2001; Chirakul et al., 2002). Because timing of the low-pH ester formation step, since we observe significant
fibrinogen is found in blood plasma and is involved in the aggregation when this reaction is allowed to proceed for .30 min, which we
formation of clots, characterizing and controlling its in- attribute to the loss of electrostatic repulsion. All reactions and washes are
teractions with foreign surfaces in vivo is critical to the suc- performed under constant slow stirring, and with the buffer volume
cessful biomedical implantation of therapeutic devices and exceeding the particle solution volume at least 100-fold.
Spheres are loaded into dialysis tubing (SpectraPor, 10 kD cutoff; Spec-
prostheses. In vitro, fibrinogen self-assembles into a cross- trum, Rancho Dominguez, CA) at number densities of ;1011–1013 particles/
linked and branched semiflexible fibrin network upon mL; we observe that higher number densities result in aggregation and poor
activation by thrombin (Doolittle, 1981). The second sam- coupling efficiency. The bags are submerged in MES buffer (100 mM
ple is filamentous actin (F-actin), a well-studied and im- 2-(N-morpholino)ethanesulfonic acid) at pH 6.0 for 2 h. The bags are
portant semiflexible biopolymer that is found in cellular then rinsed with deionized (DI) water and submerged in a solution
containing 15 mM EDC, 5 mM NHS, and a 10-fold excess of mPEG-NH2 in
cytoplasm and plays an essential role in determining the MES buffer for 30 min. Both the EDC and NHS are in vast excess, so the
strength and shape of living cells (Maggs, 1997a,b; Morse, concentrations may be varied slightly without loss of coupling efficiency;
1998; Gittes and Mackintosh, 1998; Howard, 2001; Alberts however, using significantly higher concentrations of either reagent may
et al., 2002; Boal, 2002). Reconstituted F-actin networks accelerate the loss of charge stability and promote aggregation. The mPEG-
have been studied extensively with microrheology tech- NH2 is added at this step primarily to ensure its immediate presence in the
dialysis bag after the pH is raised; a 100-fold increase in concentration has
niques, and understanding and controlling the surface inter- no effect on the reaction efficiency or protein resistivity of the particles.
actions of F-actin with colloids is imperative for properly The bags are then submerged into borate buffer (50 mM boric acid,
interpreting those data (Ziemann et al., 1994, Amblard et al., 36 mM sodium tetraborate) at pH 8.5, with NHS and mPEG-NH2 at the same
1996a, Schmidt et al., 1996, McGrath et al., 2000; Keller concentrations as above. The reaction is allowed to proceed for at least 8 h
et al. 2001, Gardel et al., 2003). The final sample, a under constant gentle stirring, and is repeated twice with fresh buffer and
reagents. After the third reaction, the particles are washed in pure borate
composite protein network composed of F-actin and the buffer for at least 2 h to remove any unreacted reagents and polymer. The
actin-bundling protein scruin, is chosen to investigate the particles are then recovered and stored at 4C; they remain stable against
effect of surface interactions in networks that more closely aggregation and protein-resistant for at least several months.

Biophysical Journal 86(6) 4004–4014

Colloid-Protein Interactions Affect MPT 4007

Preparation of BSA-coated particles San Diego, CA) onto S-VHS videotape or are directly digitized in real-time
using custom-written image analysis software (Keller et al., 2001). Video
Bovine serum albumin (Sigma-Aldrich) is slowly dissolved in phosphate frames are acquired to obtain tracer positions with 30 Hz temporal
buffered saline (PBS) (Gibco, Invitrogen Life Technologies, Carlsbad, CA) resolution. In each frame, the positions of the particles are identified by
at room temperature at concentrations of 5 mg/mL. CML particles, with finding the brightness-averaged centroid position with a subpixel accuracy
diameters of either 1.0 mm or 0.84 mm, are incubated with the BSA solutions of ;10–20 nm (Crocker and Grier, 1996). Positions are then linked in time
for 1 h at room temperature, or overnight at 4C. The suspensions are to create two-dimensional particle trajectories.
centrifuged at low speed for 30 min, the supernatant is discarded, and the
particles are resuspended in fresh PBS. This washing is repeated three or
more times to remove any unbound BSA. Particles are stored at 4C and One-particle microrheology
used within 48 h of preparation, to reduce the likelihood of the BSA
desorbing from the particle surface. To measure network viscoelasticity, we embed small colloidal particles into
a complex fluid, record their thermally activated Brownian motions, and
calculate the ensemble averaged MSD, ÆDx2 ðtÞæ ¼ Æjxðt 1 tÞ  xðtÞj2 æ; as
Characterization of protein adsorption using a function of lag time, t, where the angled brackets indicate an average over
fluorescent BSA many starting times t and the ensemble of particles in the field of view. For
spherical tracers that are embedded in a homogeneous and incompressible
Undyed particles of each surface chemistry are incubated with bovine- medium, the MSD is directly related to the viscoelastic response of the
derived albumin, labeled with tetramethylrhodamine isothiocyanate surrounding material (Mason and Weitz, 1995; Gittes et al., 1997; Mason
(R-BSA) (Sigma-Aldrich), then observed with fluorescence microscopy to et al., 1997a,b; Levine and Lubensky, 2000). Physically, this can be
determine the amount of protein adsorption. The R-BSA is dissolved in PBS understood by considering two limiting cases: a purely viscous fluid and
at a concentration of 1 mg/mL and stored at 4C for up to 48 h. For each of a completely elastic solid. For a purely viscous fluid, the d-dimensional
the differently treated particles, we add 50 mL of 0.84 mm particles at MSD will increase linearly with lag time ÆDx2 ðtÞæ ¼ 2dDt: The viscosity
a volume fraction of ;0.3% to 500 mL of the R-BSA solution and incubate h ¼ kB T=6pDa can be calculated from the diffusion coefficient D, where
the solution overnight at 4C under constant slow rotation. Volume fractions a is the particle radius. For a purely elastic material, the MSD will reach an
of the BSA- and PEG-coated particles are determined by visually comparing average plateau value ÆDxp2 æ that is independent of lag time and is determined
the diluted particle suspensions with optical microscopy to solutions of CML by the elastic modulus of the material. Equating the thermal energy to the
particles at known volume fractions. The suspensions are then centrifuged at elastic deformation energy, we estimate the elastic plateau modulus
low speed for 30 min, the supernatant is discarded, and the particles are Gp ; kB T=ÆDxp2 æa: In general, the full frequency dependence of the
resuspended in fresh PBS. This washing is repeated three or more times to viscoelastic moduli can be obtained from the MSD using a generalized
remove any unbound R-BSA. Particles are observed with a Leica DM-IRB Stokes-Einstein equation: x̃2 ðsÞ ¼ kB T=psaG̃ðsÞ; where x̃2 ðsÞ is the Laplace
inverted microscope with a 1003 magnification, oil-immersion objective, transform of ÆDx2 ðtÞæ; and G̃ðsÞ is the viscoelastic response as a function of
with numerical aperture 1.4. Using Metamorph acquisition software (Uni- the Laplace frequency s (Mason and Weitz, 1995; Mason et al., 1997a,b).
versal Imaging, Downingtown, PA), bright-field images are obtained with However, many biological materials have heterogeneities on micron length
a Hamamatsu C2400 charge-coupled device (CCD) camera, and fluores- scales, due to native microstructure or irregularities introduced by the
cence images are obtained using a Hamamatsu EB-CCD intensified camera inclusion of small colloids, which preclude the application of one-particle
(Hamamatsu, Bridgewater, NJ). Images are analyzed using Adobe Photo- microrheology.
shop.

Two-particle microrheology
Multiple particle tracking
Two-particle microrheology uses the correlated movements of pairs of
For visualization of particle dynamics, samples are loaded into microscope
separated particles to measure the long-wavelength macroscopic response of
observation chambers that consist of three 1-mm spacers that are positioned
materials that are inhomogeneous on the length scale of a single tracer but
in a ‘‘U’’ shape and sandwiched between a glass slide and a No. 1.5 glass
homogeneous on the length scale of several particles (Crocker et al., 2000;
cover slip. The slide and cover slip are rinsed with water and methanol, and
Levine and Lubensky, 2000). We calculate the ensemble-averaged tensor
air dried before use. All pieces are held in place with ultraviolet-cured optical
product of the tracer displacements: Dab ðr; tÞ ¼ ÆDrai ðt; tÞDrbj ðt; tÞ
glue (No. 81 or No. 61, Norland, Cranbury, NJ). To reduce the amount of
d½r  Rij ðtÞæi6¼j;t ; where i and j label different particles, a and b label
nonspecific protein adsorption onto the glass surfaces, a 5 mg/mL BSA
different coordinates, and Rij is the distance between particles i and j. In the
solution is loaded into the chamber and incubated at room temperature for at
limit, r
a; for an incompressible medium, Drr ðr; sÞ ¼ kB T=2prsG̃ðsÞ;
least 1 h before use. After incubation, the chambers are rinsed exhaustively
with no dependence on the particle size, shape, or boundary conditions.
with DI water to remove any remaining unbound BSA and the chambers are
Using only the Rij where Drr ;1/r, to ensure that the medium can be treated
air-dried before use. Sample volumes of 30–50 mL are loaded into the
as a coarse-grained homogeneous continuum, we extrapolate the correlated
observation chambers, and the open side is sealed with high vacuum grease
motion to the length scale a and define the two-particle mean-squared
to prevent evaporation. Care is taken to prevent any air bubbles from
displacement ÆDr2 ðtÞæD ¼ 2r=aDrr ðr; tÞ:
contacting the sample as such bubbles can lead to slow leakage and
contribute to macroscopic drift of the embedded particles. Once loaded, the
chamber is left undisturbed for at least 30 min at room temperature to allow Preparation of biopolymer gels
the protein networks to form. We observe no differences in particle
dynamics for waiting times of 60 min or more, suggesting that the To form the fibrin network, fibrinogen monomers, at a concentration of
micromechanical properties of the sample have reached a steady state during 0.44 mg/mL, are rapidly thawed from 70C, placed at room temperature,
this time. The slides are then gently transferred to an inverted research and used within 4 h. Concentrated thrombin is rapidly thawed from 70C,
microscope (Leica DM-IRB) for observation. For two-particle micro- stored on ice for up to 2 h, and diluted to 0.6 mM immediately before use.
rheology measurements, we image at least 100 microns into the sample to The network is formed by mixing 50 mL 0.44 mg/mL fibrinogen in Tris-
minimize hydrodynamic interactions with the walls. buffered saline (145 mM NaCl, 20 mM Trizma base, 0.3 mM sodium azide,
Particles are imaged with bright-field or epifluorescence microscopy, and 0.01% Tween 20; pH ¼ 7.4), 0.5 mL 1M CaCl2, 0.5 mL of the particle
the particle movements are recorded with a CCD camera (Cohu Electronics, solution, and 0.5 mL of 0.6 mM thrombin. Once the thrombin is added, the

Biophysical Journal 86(6) 4004–4014

4008 Valentine et al.

fibrinogen solution quickly polymerizes to form the fibrin network, and must
be transferred immediately into the measurement chamber to prevent
mechanical disruption of the network due to pipetting.
To form the entangled F-actin networks, lyophilized globular actin
(G-actin) is thawed from 20C, dissolved in DI water, and dialyzed against
fresh G-buffer (2 mM Tris HCl, 0.2 mM ATP, 0.2 mM CaCl2, 0.2 mM
dithiothreitol, 0.005% sodium azide) at pH 8.0 and 4C for 24 h, during
which time G-buffer is replaced every 8 h. Solutions of G-actin are kept at
4C and used within 7 days of preparation. G-actin is mixed with the
colloidal particle suspensions and fresh G-buffer to adjust final concentra-
tion. Polymerization is initiated by addition of 1/10 of the final sample
volume of 103 F-buffer (20 mM Tris HCl, 20 mM MgCl2, 1 M KCl, 2 mM
dithiothreitol, 2 mM CaCl2, 5 mM ATP, pH 7.5). Samples are mixed gently
for 10 s, then loaded into observation chambers and allowed to equilibrate
for 60 min at room temperature before particle movements are recorded.
To form composite F-actin gels, we use scruin, an actin-bundling protein
found uniquely in the sperm of horseshoe crabs, where it bundles F-actin in
the acrosome (Schmid et al., 1995; Tilney et al., 1996). Scruin purification is
performed as described by Sun et al. (1997) with minor modifications
(J. Shin, M. Gardel, L. Mahadevan, P. Matsudaira, and D. Weitz,
unpublished). The integrity of the protein is checked by SDS-PAGE gel
electrophoresis before each experiment to ensure low level of proteolysis or FIGURE 2 Bright-field (A) and fluorescence (B) images of 0.84-mm CML
degradation. The concentrations are determined either by the Bradford and BSA- and PEG-coated particles that have been incubated with R-BSA.
assay, using BSA as a standard, or by absorbance at 280 nm (Bradford, The fluorescence intensity indicates the binding capacity of each particle.
1976). Cross-linked and bundled actin networks are prepared by adding The CML and BSA-coated particles adsorb a significant amount of protein,
G-actin to the mixture of scruin, particles, and 10 3 F-buffer. Samples are whereas the adsorption on the PEG-coated particles is very small. There is
allowed to equilibrate for 1 h before observation. a shift in the fields of view between the bright-field and fluorescence images
due to different optics along the two paths; in some cases particles have
diffused slightly between image acquisitions. (C) Normalized fluorescence
intensity of the CML and BSA- and PEG-coated particles incubated with
R-BSA. The intensity of the BSA-coated particles is 60% of that of the CML
RESULTS AND DISCUSSION
particles, indicating that the BSA coating does prevent some protein
Determination of particle protein-binding capacity adsorption, but does not render the particles completely inert. The intensity
of the PEG-coated particles is only 2% of that of the CML particles,
To characterize the amount of protein adsorbed onto particles indicating a significant improvement in protein resistivity.
with different surface modifications, we incubate 0.84-mm
undyed CML and BSA- and PEG-coated particles with
bovine-derived albumin, labeled with rhodamine and ob- mg/mL BSA, the surface coverage of physisorbed BSA
serve the particles with bright-field and fluorescence never exceeds 50% of the total particle surface area, and thus
microscopy. Since the particles are not inherently fluores- cannot completely prevent protein adsorption (McGrath
cent, only those that have adsorbed a considerable amount of et al., 2000). By contrast, the PEG-coated particles are
fluorescently labeled protein will be visible with fluores- significantly more protein resistant than either the CML or
cence imaging. Bright-field images of particles are used to BSA-coated spheres; the intensity of the PEG-coated
identify the positions of the tracers, as shown in Fig. 2 A. particles is only 2% of that of the untreated CML probes.
Both CML and BSA-coated particles fluoresce, indicating
that these surface chemistries allow significant protein
adsorption, as shown in Fig. 2 B. By contrast, the PEG-
Mechanical properties of fibrin
coated particles show negligible fluorescence, indicating that
very little protein has adsorbed. To quantify the amount of To explore the effect of surface chemistry on particle
adsorption, we calculate the relative fluorescence intensities mobility in a biopolymer network, we form a cross-linked
of the particles by measuring the total intensity of the image, network of fibrin, and record the movements of 1-mm CML
subtracting the average background intensity, which is and BSA- and PEG-coated particles embedded in the gel.
measured in the absence of the particles, and normalizing The mesh size of the 0.43 mg/mL fibrin network is ;5–10 mm,
by the number of particles in the field of view. The intensity significantly larger than the diameter of the tracers, as
of the BSA-coated particles is only 60% of the intensity of observed with fluorescence microscopy of networks formed
the CML particles, as shown in Fig. 2 C, indicating that the with rhodamine-dyed fibrinogen (data not shown). Thus the
BSA coating does prevent some amount of additional protein network is heterogeneous on the scale of individual particles,
adsorption, but does not render the particles completely inert. and we observe significant variations in particle dynamics
This partial reduction in adsorption is consistent with due to differences in tracer surface chemistry. The CML and
previous measurements of BSA coating on CML spheres, BSA-coated particles are constrained and barely move,
which indicate that even with coating solutions of up to 400 whereas the PEG-coated particles are much more mobile and

Biophysical Journal 86(6) 4004–4014

Colloid-Protein Interactions Affect MPT 4009

the shape of their trajectories suggests a random walk, as To improve our statistical accuracy, we calculate the
shown in Fig. 3. ensemble-averaged MSD of these 38 mobile PEG-coated
To better quantify the particle motions, we calculate the particles, and use this to determine the local viscosity h
one-dimensional MSD. In no cases do the CML and BSA- within the pores. We measure h ¼ 1.7 mPa-s, consistent with
coated tracers move detectably, and the value of the MSD is the viscosity of the buffer, which we independently measure
a reflection only of our error in identifying particle centers. to be ;1 mPa-s; this small increase may be due to enhanced
Because the mesh size is large compared to the tracer dia- hydrodynamic drag as the particles move through the
meter, it is unlikely that the particles are merely constrained polymer network (Valentine et al., 2001; Jones and Luby-
by the elastic mesh; rather, these data suggest that these Phelps, 1996). The remaining eight particles move subdif-
protein-adsorbing probes are adhering to a stiff elastic ma- fusively with some reaching a saturated plateau value at long
terial. Although we are unable to accurately determine the lag times. For these particles, we are unable to distinguish
MSD of either the CML or BSA-coated particles, we are able experimentally between particles that are constrained due to
to place an upper bound on the MSD; from the squared a small amount of adsorption of fibrin, and particles that are
measurement error of the particle positions, we estimate not bound, but merely reside in an unusually dense region of
ÆDxp2 æ # 4 3 104 mm2 : For particles moving in a heteroge- the gel where the local mesh size is approximately equal to or
neous network, characterized by an average mesh size j, we smaller than the tracer. For all measurements, the network is
estimate the elastic plateau modulus to be Gp ; kB T=ÆDxp2 æj heterogeneous on the length scale of the particles, and in no
(Krall and Weitz, 1998). Using our upper bound for the MSD case do we measure the bulk rheological response.
and j ; 5 mm, we estimate Gp $ 5 Pa. Because the fibrin
gels form very rapidly upon addition of thrombin and are
Surface chemistry effects on the microrheology
extremely fragile, we are unable to reproducibly load the gels
of entangled F-actin networks
into a traditional mechanical rheometer for comparison.
By contrast, the MSD of each PEG-coated particle is well To further explore the effect of surface chemistry on the
above our resolution limit, and thus accurately reflects the protein adsorption and mobility of colloids in biomaterials,
local microenvironment surrounding the tracer. By examin- we measure the movements of 1.0-mm CML and BSA- and
ing the one-dimensional MSDs of 46 individual PEG-coated PEG-coated particles in an 11.9 mM entangled F-actin
particles, we observe a variety of different behaviors, as network. F-actin networks have been studied extensively
shown in Fig. 4. To collect sufficient statistics on individual with microrheology techniques (Ziemann et al., 1994;
particles, we analyze the trajectories only of particles that Amblard et al., 1996a; Schmidt et al., 1996; Keller et al.,
remain in focus for at least 10 s. Of the 46 particles, 38 move 2001; Gardel et al., 2003), and the surface chemistry effects
diffusively for lag times up to 1 s, with MSDs increasing of CML and BSA-coated particles in actin networks have
linearly in t, suggesting that because the PEG-coated been previously reported in detail (McGrath et al., 2000).
particles resist protein adsorption, most are nearly un- Here, we calculate the ensemble-averaged MSD for the
perturbed by the surrounding polymer strands. Moreover, we PEG-coated particles, and compare to those of the CML and
measure no difference in the one-dimensional MSD obtained BSA-coated particles. There is a measurable reduction in the
using displacement data collected along two orthogonal plateau value of the MSD, ÆDxp2 æ; of the untreated CML
directions, indicating that particles are not sensitive to any particles, by roughly a factor of 2, in comparison to that of
mechanical anisotropy that may be present in the network. At the BSA- and PEG-coated particles, as shown in Fig. 5; this
longer lag times, the particle motion may be influenced by is in agreement with previous work (McGrath et al., 2000).
the network; however, because the PEG-coated particles are This reduction may indicate that the CML particles are more
able to diffuse through the network and out of our focal tightly bound to the polymer network, or that by binding to
volume, we are unable to measure the MSD of the tracers at actin filaments, the CML particles introduce local cross-
these timescales. linking that slightly increases the local elastic modulus.

FIGURE 3 Trajectories of 1-mm CML

and BSA- and PEG-coated particles in
a 0.43 mg/mL fibrin network with a/j ;
0.1–0.2. The BSA-coated and CML
particles adhere to the stiff fibrin network
and are completely immobile within the
resolution of the measurement; circles
indicate their static positions. By con-
trast, the PEG-coated particles are re-
sistant to the nonspecific protein
adsorption and remain mobile with
trajectories that resemble random walks.

Biophysical Journal 86(6) 4004–4014

4010 Valentine et al.

actin gels and cytoplasm (McGrath et al., 2000). To further

investigate these effects, we compare the movements of
1.0-mm CML, and BSA- and PEG-coated particles in
F-actin networks that are cross-linked and bundled by the
actin-bundling protein scruin. Scruin is found in the acrosomal
bundle of horseshoe crab sperm, where it decorates individ-
ual actin filaments at a 1:1 stoichiometric ratio (Schmid
et al., 1995). In vivo, scruin-scruin interactions align F-actin
in a tight parallel array of neighboring filaments. In vitro,
FIGURE 4 A sampling of the mean-squared displacements of individual actin-scruin composite networks are used as model systems
PEG-coated particles moving in a fibrin network. Most particles diffuse, but to investigate the mechanics and microstructure of cross-
some are locally constrained, leading to a plateau in their MSDs at long lag linked F-actin gels; in this case, the ratio of scruin to G-actin,
times. R, is varied from 1:30 to 1 (J. Shin, M. Gardel, L. Mahadevan,
P. Matsudaira, and D. Weitz, unpublished). Over these ratios,
Unlike the measurements of fibrin, we observe no measur- scruin serves to both cross-link and bundle actin filaments into
able difference between the value of ÆDxp2 æ for the BSA- and a three-dimensional network. For small R, infrequent scruin-
PEG-coated particles, suggesting that the binding affinity of scruin interactions cross-link neighboring actin filaments that
actin to the particles is weak in comparison to fibrinogen. remain randomly oriented without promoting bundle forma-
Overall, the effect of varying surface chemistry for entangled tion. The network structure is only slightly perturbed from
actin solutions is small. that of a purely entangled actin network, and is characterized
pffiffiffiffiffi
by a mesh size in microns j ¼ 0:3= cA ; where cA is the
concentration of actin in mg/mL (Schmidt et al., 1989). At
One- and two-particle microrheology larger R, bundles form and as R is increased for a constant
measurements of F-actin/scruin networks actin concentration, the bundle thickness increases, leading
to a concomitant increase in the mesh size. Independent
We measure only weak effects of particle surface modifica-
measurements have shown that for R ranging from 1:30 to 1,
tions in pure actin solutions; however, physiological actin
j varies from ;0.5 mm to 8 mm (J. Shin et al., unpublished).
networks are stiffened by actin-binding proteins that cross-
To investigate the role of scruin in the surface interactions
link, bundle, and nucleate filaments; moreover, actin
between the tracer colloids and the gel, and to study the
networks in cells form in the presence of numerous globular
transition from a weakly cross-linked to a highly bundled
proteins in the cytosol, as well as other cytoskeletal filaments
network, we embed CML and BSA- and PEG-coated
(Alberts et al., 2002). Actin-binding proteins may display
particles into the composite actin-scruin networks and ob-
a stronger binding affinity for surfaces, leading to more
serve the resultant particle displacements. In contrast to the
striking differences in the mobility of particles with different
solutions of purely entangled F-actin, we see dramatic dif-
surface modifications, and complicating the interpretation of
ferences in both the magnitude and time dependence of the
particle tracking measurements in physiologically relevant
MSDs of particles with different surface modifications. At
the lowest concentration of cross-linkers, R ¼ 1:30, the
CML particles aggregate and no single tracers are observed,
precluding measurements of particle dynamics. At higher
cross-linker densities, this aggregation worsens and the CML
particles form large clumps that sediment out of solution. By
contrast, the BSA- and PEG-coated particles remain dis-
persed, consistent with their ability to better resist protein
adsorption compared to unmodified spheres.
To further investigate the effect of surface chemistry on
particle dynamics, we examine the MSDs of the BSA- and
PEG-coated particles at different scruin/actin ratios. The
concentration of actin is fixed at 11.9 mM (0.5 mg/mL), and
FIGURE 5 Ensemble averaged mean-squared displacements of 1.0-mm the amount of scruin is varied to achieve the desired
CML (h), BSA-coated (s), and PEG-coated (4) particles in an entangled stoichiometric ratio. As shown in Fig. 6, for each R, the MSD
F-actin solution. There is measurable decrease in the plateau value of the of the BSA-coated particles (solid symbols) increases at short
MSD of the CML particles as compared to those of the BSA- or PEG-coated lag times, and saturates at a plateau value ÆDxp2 æ at ; t ¼
particles; however, the overall effect is small, suggesting that the binding
affinity of actin to bare CML particle surfaces is weak. We observe no 0.3 s; the value of ÆDxp2 æ decreases as the cross-linker density
difference between the plateau value of the MSDs for the BSA- and PEG- is increased from R ¼ 1:30 (solid squares), to R ¼ 1:15 (solid
coated tracers. circles), to R ¼ 1:5 (solid triangles). By contrast, the

Biophysical Journal 86(6) 4004–4014

Colloid-Protein Interactions Affect MPT 4011

diameter d increases, causing an increase in the bending

modulus B of the bundled actin strands; for thin isotropic
rods, B ; d4. Because the BSA-coated particles adhere to
these bundles, they are sensitive to changes in local stiffness
and move less with increasing amounts of scruin. By
contrast, the protein-resistant PEG-coated particles are less
likely to adhere to the network, and are instead merely
constrained by the elastic cage formed by the surrounding
polymer strands. At low and intermediate scruin concen-
trations, we observe no change in the plateau value of the
MSD measured by the PEG-coated particles with increasing
amounts of scruin, indicating little change in the local micro-
environment. At larger concentrations of scruin, j increases
FIGURE 6 The ensemble averaged MSDs for BSA-coated particles (solid beyond the fixed particle radius, allowing the PEG-coated
symbols) and PEG-coated particles (open symbols) moving in actin networks particles to move more.
cross-linked and bundled with the actin-binding protein scruin, at various
ratios of scruin/actin: R ¼ 1:30 (h), 1:15 (s), 1:10 (,), 1:5 (n), 1:2.5 ()),
Because we measure similar dynamics for the BSA- and
and 1 (9). Representative error bars are shown for each surface coating. The PEG-coated particles in pure entangled actin networks, we
BSA-coated particles are constrained for each R, reaching a plateau ÆDxp2 æ surmise that the differences we observe in the actin-scruin
that decreases with increasing amounts of scruin. The ensemble averaged networks arise because the BSA-coated particles bind to
MSDs for PEG-coated particles show a completely different trend for R scruin. This suggests that the BSA-coated particles are
ranging from 1:30 to 1. At the lower concentrations of cross-linkers, the
particles are constrained; however, the particle mobility increases with
preferentially distributed to the scruin-rich regions of the gel,
increasing amounts of scruin, until at R ¼ 1, the particles are nearly diffusive or alternatively that the BSA-coated particles recruit scruin
at short lag times. to their surfaces, thereby altering the distribution of cross-
linkers in the sample. By increasing the effective concentra-
tion of cross-linkers near the tracer surface, the BSA-coated
PEG-coated particles (open symbols) show a completely beads create locally stiffer, more cross-linked regions of the
different trend. At the lowest cross-linker density, R ¼ 1:30 gel near the particles, causing a reduction in the amplitude of
(open squares), the MSD saturates to a plateau value by a lag their thermal fluctuations. By contrast, the PEG-coated
time of ;0.5 s; however, this value is more than five times particles resist the adsorption of scruin and are more
larger than the plateau value observed with the BSA-coated randomly distributed among the native mechanical micro-
particles at the same R. When the amount of scruin is environments of the gel. The possibility that there is a
increased slightly to R ¼ 1:15 (open circles) and R ¼ 1:10 differential distribution of partially sticky and inert tracers in
(downward open triangles), there is no change in the MSDs different mechanical microenvironments of the network is
of the PEG-coated particles despite the threefold change in supported by the differences in ÆDxp2 æ for the BSA- and PEG-
cross-linker concentration. At higher cross-linker densities of coated particles at low and intermediate scruin concen-
R ¼ 1:5 (upward open triangles), or greater, the ensemble- trations. We find that ÆDxp2 æ for the PEG-coated particles is
averaged MSD no longer reaches a true plateau, and the a factor of 5–10 greater than that of the BSA-coated tracers in
value of the MSD at long lag times is larger than that of this regime where particles with both surface modifications
particles moving in less cross-linked samples or the purely are constrained by the network. Alternatively, it has been
entangled F-actin solutions. At the highest cross-linker suggested that during the initial stages of gelation, before
density of R ¼ 1 (leftward open triangles), the MSD is nearly a percolated network has formed, nonbinding particles will
linear at short lag times. From this short time data, we migrate to the weaker regions of heterogeneous gels as this
calculate h  3 mPa-s, similar to the viscosity of the separation maximizes particle motions and thus is entropi-
background aqueous fluid, which we independently measure cally favored (McGrath et al., 2000). We are unable to
to be ;1 mPa-s. experimentally distinguish among these possibilities, or pre-
The differences in the movements of the BSA-coated and cisely interpret ÆDxp2 æ for the constrained particles in these
PEG-coated particles reflect the differences in their protein- composite networks.
binding capacities, and demonstrate our ability to differen- To further explore these differences and reveal the
tially probe a heterogeneous network using particles with underlying polymer dynamics, we use two-particle micro-
carefully chosen surface modifications. The BSA-coated rheology to measure the long wavelength fluctuations of the
particles, although better able to resist protein adsorption actin-scruin networks and determine the bulk viscoelastic
than the bare CML spheres, are still somewhat ‘‘sticky’’ and response. For a network with R ¼ 1:30, we observe a slight
adhere to the network; as a result, their movements reflect the difference between the two-particle MSDs measured with
thermal fluctuations of the polymer strands of the network. the BSA-coated particles (solid symbols) and PEG-coated
As we increase the amount of scruin, the average bundle tracers (open symbols), as shown in Fig. 7 A; however, this

Biophysical Journal 86(6) 4004–4014

4012 Valentine et al.

difference is significantly smaller than that measured with coated particles are less likely to bind to the network, and
the one-particle MSDs, and is within the sample/sample vari- a smaller proportion of their Brownian motion reflects the
ation. For a more strongly cross-linked network with R ¼ 1:15, correlated motions. This leads to a much poorer signal/noise
we measure a small decrease in the two-particle MSDs as ratio for the PEG-coated particles than for the BSA-coated
compared to the data for R ¼ 1:30, and good agreement particles, leading to noisier data in the two-particle MSD.
between the data for the BSA- and PEG-coated particles as
shown in Fig. 7 B. In all cases, the two-particle MSDs plateau
SUMMARY
at long lag times, and the plateau values for ÆDr 2 ðtÞæD are
lower than those measured by the one-particle techniques The examples of the fibrin, F-actin, and F-actin-scruin
using either the BSA- or PEG-coated particles. Furthermore, composite networks illustrate the rich information about
we find that the two-particle analysis eliminates virtually all local viscosity, elasticity, and microstructure available at
the differences between the BSA- and PEG-coated tracers. a range of length scales with multiple particle tracking
This suggests that although there are local differences in the measurements, and demonstrate the challenges of interpret-
coupling of the BSA- and PEG-coated particles to the network ing particle movements in complex heterogeneous biomate-
that give rise to differences in the one-particle MSDs, the rials. To achieve the full potential of these methods, the
particles do not induce large length scale inhomogeneities; delicate interactions between the embedded probe colloids
moreover, the macroscopic stress relaxation measured by the and the biopolymer networks must be understood and
two particle types is similar. To our knowledge, this is the first controlled. In particular, to understand the interplay between
demonstration that two-particle microrheology methods mechanics and structure in complex biomaterials, and to
successfully eliminate the local variations in mechanical address important biological questions in vivo, it is critical to
response caused by differences in tracer surface chemistry. identify which aspect of the multi-component system is
Interestingly, for both samples, the two-particle MSD being probed.
calculated from the data for the PEG-coated particles is In this article, we employ two types of particles with
noisier than that calculated using the data from the BSA- commonly used surface modifications, CML particles, and
coated particles. In both cases, the particle dynamics are CML particles that have been coated with physisorbed BSA,
a superposition of local fluctuations, which are uncorrelated and compare their adsorption properties and mobility to
and unrelated to the long wavelength polymer fluctuations, those of novel PEG-coated tracers. The binding capacity of
and long wavelength modes due to the thermal fluctuations the particles varies depending on the biopolymer we probe;
of the network itself, which lead to the correlated motions however, in all cases, we find the CML particles adsorb the
measured by the two-particle technique. Because the BSA- most protein, whereas the PEG-coated particles adsorb the
coated particles are physically coupled to the polymer least. Moreover, we find a correlation between increased
filaments, a greater proportion of their Brownian dynamics protein binding capacity and decreased particle mobility.
reflects the fluctuations of the network. By contrast, the PEG- Thus, differences in particle dynamics measured with
multiple particle tracking techniques may reflect differences
in the particles’ adsorption characteristics, as well as
differences the local viscoelastic response or changes in
local microstructure. When probing a new material, a careful
study of the effect of both particle size and surface chemistry
is absolutely required for the proper interpretation of
multiple particle tracking experiments.
We have further demonstrated that binding and non-
binding particles are sensitive to different physical properties
of heterogeneous networks. Our measurements of particle
displacements in fibrin networks and cross-linked and
bundled actin gels suggest that particles that are weakly
bound to the network are sensitive to changes in local
stiffness but not microstructure, whereas the protein resistant
FIGURE 7 Two-particle mean-squared displacements for composite PEG beads are sensitive to changes in viscosity and mesh
actin-scruin networks with (A) R ¼ 1:30 and (B) R ¼ 1:15; ;70 particles size and less sensitive to changes in elastic modulus. Thus,
are included in this calculation. The long wavelength elastic modes are judicious choice of colloid surface chemistry should allow us
unaffected by local coupling of the particles to the network, and differences to selectively probe these different material properties of
between the BSA-coated particles (solid symbols) and PEG-coated particles
complex biomaterials. For two-particle microrheology
(open symbols) measured with one-particle techniques are eliminated. In all
cases, the MSDs show a plateau at long lag times, and the plateau values of measurements, differences arising from surface chemistry
ÆDxp2 æ are significantly smaller than those measured by one-particle are eliminated, but the signal/noise ratio of the measurement
techniques. increases when tracers adhere to the network. In cases such

Biophysical Journal 86(6) 4004–4014

Colloid-Protein Interactions Affect MPT 4013

as this, where a small amount of adsorption is desired, the Chirakul, P., V. H. Pérez-Luna, H. Owen, G. P. López, and P. D. Hampton.
2002. Synthesis ad characterization of amine-terminated self-assembled
uncontrolled synthesis, ill-defined characterization, and short monolayers containing diethylene glycol linkages. Langmuir. 18:4324–
shelf-life of the BSA-coated particles may be impractical. 4330.
With slight modifications to the protocol described here, it is Clark, R. A. F. 2001. Fibrin and wound healing. Ann. N. Y. Acad. Sci.
possible to create designer particles with dense methoxy- 936:355–367.
terminated PEG brushes that prevent nonspecific adsorption, Crocker, J. C., and D. G. Grier. 1996. Methods of digital video microscopy.
but bind to a particular target protein by the addition of J. Colloid Interface Sci. 179:298–310.
a small number of reactive PEG chains with terminal groups Crocker, J. C., M. T. Valentine, E. R. Weeks, T. Gisler, P. D. Kaplan, A. G.
Yodh, and D. A. Weitz. 2000. Two-point microrheology of inhomoge-
that present an antibody or other specific linkage (Lahiri neous soft materials. Phys. Rev. Lett. 85:888–891.
et al., 1999). In addition to providing surfaces with tunable Dasgupta, B. R., S.-Y. Tee, J. C. Crocker, B. J. Frisken, and D. A. Weitz.
adsorption properties, such designer particles will allow 2001. Microrheology of polyethylene oxide using diffusing wave
mechanical measurements of a single component of a com- spectroscopy and single scattering. Phys. Rev. E. 65:051505.
plex composite structure, such as in the cytoplasm of living de Gennes, P. G. 1987. Polymers at an interface: a simplified view. Adv.
Colloid Interface Sci. 27:189–209.
cells.
De Sousa Delgado, A., M. Léonard, and E. Dellacherie. 2001. Surface
properties of polystyrene nanoparticles coated with dextrans and dextran-
This work was supported in part by grants from the National Science PEO copolymers. Effect of polymer architecture on protein adsorption.
Foundation (DMR-0243715), the Materials Research Science and Engi- Langmuir. 17:4386–4391.
neering Center through the auspices of the National Science Foundation Doolittle, R. F. 1981. Fibrinogen and fibrin. Sci. Am. 245:126–135.
(DMR-0213805), and the National Aeronautics and Space Administration
Gardel, M. L., M. T. Valentine, J. C. Crocker, A. R. Bausch, and D. A.
(NAG3-2284).
Weitz. 2003. Microrheology of entangled F-actin solutions. Phys. Rev.
The authors also gratefully acknowledge support from the National Science Lett. 91:158302.
Foundation Graduate Research Fellowship (J.H.S.), Lucent Graduate Gittes, F., and F. C. MacKintosh. 1998. Dynamic shear modulus of a semi-
Research Program for Women Fellowship (M.L.G.), and Howard Hughes flexible polymer network. Phys. Rev. E. 58:R1241–R1244.
Medical Institute Predoctoral Fellowship (Z.E.P.). We thank J. Weitz, J. Gittes, F., B. Schnurr, P. D. Olmsted, F. C. MacKintosh, and C. F. Schmidt.
Rischke, and A. Lazier (McMaster University, Hamilton, Canada) for 1997. Microscopic viscoelasticity: shear moduli of soft materials
supplying the fibrin and thrombin and for technical support in handling determined from thermal fluctuations. Phys. Rev. Lett. 79:3286–3289.
these gels, E. Sackmann and A. R. Bausch (Technical University, Munich, Harder, P., M. Grunze, R. Dahint, G. M. Whitesides, and P. E. Laibinis.
Germany) for supplying the actin samples, G. Waller (Whitehead Institute 1998. Molecular conformation in oligo(ethylene glycol)-terminated self-
for Biomedical Research, Cambridge, MA) for providing technical assembled monolayers on gold and silver surfaces determines their
assistance in scruin purification, and J. Jiang and X. Qian (Dept. of ability to resist protein adsorption. J. Phys. Chem. B. 102:426–436.
Chemistry and Chemical Biology, Harvard University, Cambridge, MA) for Haynes, C. A., and W. Norde. 1994. Globular proteins at solid/liquid
useful discussions, insights, and technical advice. interfaces. Colloid. Surface. B. 2:517–566.
Hough, L. A., and H. D. Ou-Yang. 1999. A new probe for mechanical
testing of nanostructures in soft materials. J. Nanoparticle Res. 1:495–
REFERENCES 499.
Howard, J. 2001. Mechanics of Motor Proteins and the Cytoskeleton.
Alberts, B., A. Johnson, J. Lewis, M. Raff, K. Roberts, and P. Walter. 2002. Sinauer, Sunderland, MA.
Molecular Biology of the Cell. Garland, New York. Jeon, S. I., J. H. Lee, J. D. Andrade, and P. G. de Gennes. 1991. Protein
Alcantar, N. A., E. S. Aydil, and J. N. Israelachvili. 2000. Polyethylene surface interactions in the presence of polyethylene oxide. 1. Simplified
glycol-coated biocompatible surfaces. J. Biomed. Mater. Res. 51:343– theory. J. Colloid. Interface Sci. 142:149–166.
351. Jones, J. D., and K. Luby-Phelps. 1996. Tracer diffusion through F-actin:
effect of filament length and cross-linking. Biophys. J. 71:2742–2750.
Amblard, F., A. C. Maggs, B. Yurke, A. N. Pargellis, and S. Leibler. 1996a.
Subdiffusion and anomalous local viscoelasticity in actin networks. Phys. Keller, M., J. Schilling, and E. Sackmann. 2001. Oscillatory magnetic bead
Rev. Lett. 77:4470–4473. rheometer for complex fluid microrheometry. Rev. Sci. Instrum.
72:3626–3634.
Amblard, F., B. Yurke, A. Pargellis, and S. Leibler. 1996b. A magnetic
manipulator for studying local rheology and micromechanical properties Krall, A. H., and D. A. Weitz. 1998. Internal dynamics and elasticity of
of biological systems. Rev. Sci. Instrum. 67:818–827. fractal colloidal gels. Phys. Rev. Lett. 80:778–781.
Lahiri, J., L. Issacs, J. Tien, and G. M. Whitesides. 1999. A strategy for the
Boal, D. 2002. Mechanics of the Cell. Cambridge University Press, generation of surfaces presenting ligands for studies of binding based on
Cambridge. an active ester as a common reactive intermediate: a surface plasmon
Bradford, M. M. 1976. Rapid and sensitive method for quantitation of resonance study. Anal. Chem. 71:777–790.
microgram quantities of protein utilizing the principle of protein-dye Levine, A. J., and T. C. Lubensky. 2000. One- and two-particle
binding. Anal. Chem. 72:248–254. microrheology. Phys. Rev. Lett. 85:1774–1777.
Chapman, R. G., E. Ostuni, L. Yan, and G. M. Whitesides. 2000. Liu, J., L. M. Gan, C. H. Chew, C. H. Quek, H. Gong, and L. H. Gan. 1997.
Preparation of mixed self-assembled monolayers (SAMs) that resist The particle size of latexes from dispersion polymerization of styrene
adsorption of proteins using the reaction of amines with a SAM that using poly(ethylene oxide) macromonomer as a polymerizable stabilizer.
presents interchain carboxylic anhydride groups. Langmuir. 16:6927– J. Appl. Polym. Sci. 35:3575–3583.
6936. Luby-Phelps, K. 2000. Cytoarchitecture and physical properties of the
Chen, D. T., E. R. Weeks, J. C. Crocker, M. F. Islam, R. Verma, J. Gruber, cytoplasm: volume, viscosity, diffusion, intracellular surface area. Int.
A. J. Levine, T. C. Lubensky, and A. G. Yodh. 2003. Rheological Rev. Cytol. 192:189–221.
microscopy: local mechanical properties from microrheology. Phys. Rev. Maggs, A. C. 1997a. Micro-bead mechanics with actin filaments. Phys.
Lett. 90:108301. Rev. E. 57:2091–2094.

Biophysical Journal 86(6) 4004–4014

4014 Valentine et al.

Maggs, A. C. 1997b. Two-plateau moduli for actin gels. Phys. Rev. E. Schmidt, F. G., F. Ziemann, and E. Sackmann. 1996. Shear field mapping
55:7396–7400. in actin networks by using magnetic tweezers. Eur. Biophys. J. 24:348–
353.
Mason, T. G., K. Ganesan, J. H. Van Zanten, D. Wirtz, and S. C. Kuo.
1997a. Particle tracking microrheology of complex fluids. Phys. Rev. Shay, J. S., R. J. English, R. J. Spontak, C. M. Balik, and S. A. Khan. 2000.
Lett. 79:3282–3285. Dispersion polymerization of polystyrene latex with novel grafted
poly(ethylene glycol) macromers in 1-propanol/water. Macromolecules.
Mason, T. G., H. Gang, and D. A. Weitz. 1997b. Diffusing-wave
33:6664–6671.
spectroscopy measurements of viscoelasticity of complex fluids. J. Opt.
Soc. Am. 14:139–149. Shay, J. S., S. R. Raghavan, and S. A. Khan. 2001. Thermoreversible
gelation in aqueous dispersions of colloidal particles bearing grafted
Mason, T. G., and D. A. Weitz. 1995. Optical measurements of the poly(ethylene oxide) chains. J. Rheol. 45:913–927.
frequency-dependent linear viscoelastic moduli of complex fluids. Phys.
Rev. Lett. 74:1250–1253. Sun, S., M. Footer, and P. Matsudaira. 1997. Modification of Cys-837
identifies an actin-binding site in the beta propeller protein scruin. Mol.
McGrath, J. L., N. J. Eungdamrong, C. I. Fisher, F. Peng, L. Mahadevan, Biol. Cell. 8:421–430.
T. J. Mitchison, and S. C. Kuo. 2003. The force-velocity relationship for the
actin-based motility of Listeria monocytogenes. Curr. Biol. 13:329–332. Tilney, L. G., M. S. Tilney, and G. M. Guild. 1996. Formation of actin
filament bundles in the ring canals of developing Drosophila follicles.
McGrath, J. L., J. H. Hartwig, and S. C. Kuo. 2000. The mechanics of J. Cell Biol. 133:61–74.
F-actin microenvironments depend on the chemistry of probing surfaces.
Biophys. J. 79:3258–3266. Valentine, M. T., L. E. Dewalt, and H. D. Ou-Yang. 1996. Forces on
a colloidal particle in a polymer solution: a study using optical tweezers.
Morse, D. C. 1998. Viscoelasticity of concentrated isotropic solutions of J. Phys. Condens. Matter. 8:9477–9482.
semiflexible polymers. Macromolecules. 31:7030–7067.
Valentine, M. T., P. D. Kaplan, D. Thota, J. C. Crocker, T. Gisler, R. K.
Ostuni, E., R. G. Chapman, R. E. Holmlin, S. Takayama, and G. M. Prud’homme, M. Beck, and D. A. Weitz. 2001. Investigating the
Whitesides. 2001. A survey of structure-property relationships of microenvironments of inhomogeneous soft materials with multiple
surfaces that resist the adsorption of protein. Langmuir. 17:5605–5620. particle tracking. Phys. Rev. E. 64:061506.
Prime, K. L., and G. M. Whitesides. 1993. Adsorption of proteins onto Valiokas, R., S. Svedham, M. Östblom, S. C. T. Svensson, and B. Liedberg.
surfaces containing end-attached oligo(ethylene oxide): a model systems 2001. Influence of specific intermolecular interactions on the self-
using self-assembled monolayers. J. Am. Chem. Soc. 115:10714–10721. assembly and the phase behavior of oligo(ethylene glycol)-terminated
Prime, K. L., and G. M. Whitesides. 1991. Self-assembled organic alkanethiolates on gold. J. Phys. Chem. B. 105:5459–5469.
monolayers: model systems for studying adsorption of proteins at Weisbecker, C. S., M. V. Merritt, and G. M. Whitesides. 1996. Molecular
surfaces. Science. 252:1164–1167. self-assembly of aliphatic thiols on gold colloids. Langmuir. 12:3763–
Schmid, M. F., J. M. Agris, J. Jakana, P. Matsudaira, and J. W. Chiu. 1995. 3772.
Three-dimensional structure of a single filament in the Limulus Wong, I. Y., M. L. Gardel, D. R. Reichman, E. R. Weeks, M. T. Valentine,
acrosomal bundle: scruin binds to homologous helix-loop-beta motifs A. R. Bausch, and D. A. Weitz. 2004. Anomalous diffusion probes
in actin. J. Cell Biol. 124:341–350. microstructure dynamics of entangled F-actin networks. Phys. Rev. Lett.
Schmidt, C. F., M. Baermann, G. Isenberg, and E. Sackmann. 1989. Chain 92:178101.
dynamics, mesh size, and diffusive transport in networks of polymerized Ziemann, F., J. Rädler, and E. Sackmann. 1994. Local measurements of
actin: a quasielastic light scattering and microfluorescence study. viscoelastic moduli of entangled actin networks using an oscillating
Macromolecules. 22:3638–3649. magnetic bead micro-rheometer. Biophys. J. 66:2210–2216.

Biophysical Journal 86(6) 4004–4014