Subscriber access provided by MCGILL UNIV
Analytical Chemistry is published by the American Chemical Society. 1155 SixteenthStreet N.W., Washington, DC 20036Published by American Chemical Society. Copyright © American Chemical Society.However, no copyright claim is made to original U.S. Government works, or worksproduced by employees of any Commonwealth realm Crown government in the courseof their duties.
Technical Note
A microfluidic chamber to study the dynamics ofmuscle contraction specific molecular interactions
Horia Nicolae Roman, David Juncker, and Anne-Marie LauzonAnal. Chem., Just Accepted Manuscript • DOI: 10.1021/ac503963r • Publication Date (Web): 28 Jan 2015
Downloaded from http://pubs.acs.org on February 4, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are postedonline prior to technical editing, formatting for publication and author proofing. The American ChemicalSociety provides “Just Accepted” as a free service to the research community to expedite thedissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscriptsappear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have beenfully peer reviewed, but should not be considered the official version of record. They are accessible to allreaders and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offeredto authors. Therefore, the “Just Accepted” Web site may not include all articles that will be publishedin the journal. After a manuscript is technically edited and formatted, it will be removed from the “JustAccepted” Web site and published as an ASAP article. Note that technical editing may introduce minorchanges to the manuscript text and/or graphics which could affect content, and all legal disclaimersand ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errorsor consequences arising from the use of information contained in these “Just Accepted” manuscripts.
1
A microfluidic chamber to study the dynamics of muscle contraction
specific molecular interactions
Horia Nicolae Romana,c
, David Juncker c, Anne-Marie Lauzon*
a,b,c,d.
a: Meakins-Christie Laboratories, McGill University, 3626 St-Urbain, Montréal, H2X 2P2, Québec,
Canada.
b: Department of Medicine, McGill University, Royal Victoria Hospital, 687 Pine avenue, Montréal,
H3A 1A1, Québec, Canada.
c: Department of Biomedical Engineering, McGill University, Duff Medical Building, 3775 University,
Montréal, H3A 2B4 Québec, Canada.
d: Department of Physiology, McGill University, McIntyre Medical Building, 3655 Sir William Osler,
Montréal, H3G 1Y6, Québec, Canada.
Corresponding Author
* Anne-Marie Lauzon, Ph.D., Meakins-Christie Laboratories, McGill University, 3626 St-Urbain street,
Montreal, H2X 2P2, Quebec, Canada, Telephone: (514) 398-3864, Fax: (514) 398-7483, Email: anne-
Page 1 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
2
ABSTRACT
In vitro motility and laser trap assays are commonly used for molecular mechanics measurements.
However, chemicals cannot be added during measurements because they create flows that alter the mo-
lecular mechanics. Thus, we designed a microfluidic device that allows the addition of chemicals with-
out creating bulk flows. Biocompatibility of the components of this device was tested. A micro-channel
chamber was created by photolithography with the patterns transferred to polydimethylosiloxane
(PDMS). The PDMS chamber was bound to a polycarbonate membrane which itself was bound to a
molecular mechanics chamber. The micro-channels assured rapid distribution of the chemicals over the
membrane whereas the membrane assured efficient delivery to the mechanics chamber while preventing
bulk flow. The biocompatibility of the materials was tested by comparing the velocity (νmax) of propul-
sion by myosin of fluorescently labeled actin filaments to that of the conventional assay; no difference
in νmax was observed. To estimate total chemical delivery time, labeled bovine serum albumin was in-
jected in the channel chamber and TIRF was used to determine the time to reach the assay surface
(2.7±0.1 s). Furthermore, the standard distance of a trapped microsphere calculated during buffer diffu-
sion using the microfluidic device (14.9±3.2 nm) was not different from that using the conventional as-
say (15.6±5.3 nm, p=0.922). Finally, νmax obtained by injecting adenosine triphosphate (ATP) in the mi-
cro-channel chamber (2.37±0.48 µm/s) was not different from that obtained when ATP was delivered
directly to the mechanics chamber (2.52±0.42 µm/s, p=0.822). This microfluidic prototype validates the
design for molecular mechanics measurements.
Page 2 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
3
INTRODUCTION
Our understanding of molecular motors has greatly advanced since the development of techniques such
as the in vitro motility1 and the laser trap assays
2. The general setup for these assays uses a flow-through
chamber3,4
made up of a microscope slide and a coverslip separated by shims of a given thickness
(Fig.1A). Buffers and proteins are sequentially perfused through those chambers to establish the protein
interactions3,5,6
. However, once the measurements have been initiated, there is no possibility of altering
the buffer composition. Indeed, these molecular mechanics measurements are exquisitely sensitive to
the addition of reagents because single molecule interactions would be disturbed by bulk flow. However,
several protocols could take advantage of chambers that would allow injections of chemicals at specific
points in time during the measurements. For example, the latch-state is a condition of force maintenance
observed in tonic smooth muscle7. It has been postulated several years ago that this force maintenance
occurs when myosin gets deactivated (dephosphorylated) while attached to actin7. That theory could
never be verified at the molecular level because it is not possible to add enzymes to dephosphorylate
during force measurements with the laser trap. Other examples of protocols that would be enhanced by
changes in buffers during molecular mechanical measurements are the studies of the behavior of groups
of motors5,8-10
.
Thus, in the current study, we developed a microfluidic device (Fig. 1B and Fig.2) to allow the introduc-
tion of chemicals in a molecular mechanics flow-through chamber, without creating any bulk flow that
would otherwise disturb the molecular mechanics measurements. This microfluidic device complements
the traditional laser trap assay (Fig. 1A) by adding (Fig.2): 1) a chamber made up of horizontal micro-
channels built in polydimethylsiloxane (PDMS) to distribute the chemicals rapidly and uniformly over a
membrane; 2) a membrane that allows the chemicals to access the contractile proteins in the molecular
mechanics chamber without creating any bulk flow. The proof of principle of this microfluidic device is
presented. The final design was tested by injecting ATP in the micro-channel chamber to activate the
motility in the molecular mechanics flow-through chamber. Uniform actin propulsion by myosin was
indeed observed, validating the applicability of this microfluidic device to molecular mechanics assays.
Page 3 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
4
METHODS
Microfabrication of the micro-channel chamber: The micro-channel chamber was constructed by stand-
ard photolithography11
as follows. A resin (SU-8 2050, Microchem, Newton, MA) was spin-coated on a
6 inch silicon wafer at 1,600 rpm for 30s and then pre-baked for 5 min at 65°C and for 15 min at 95°C.
Patterns were then created in the resin using a mask (chrome deposition on glass, Front Range Photo-
mask, resolution 10 µm/feature) and exposed to ultraviolet (UV) light (35 mW/cm2) for 34 s. The pat-
terns were fixed to the substrate by baking at 65°C for 5 min and at 95°C for 10 min followed by an
immersion in SU-8 developer for 10 min. To eliminate tensions in the structure, a final hard bake at
150°C was performed for 5 min. The patterns obtained were used as a mold to carve channels in PDMS
(Sylgard 184, Dow Corning). The PDMS was mixed at a 10:1 ratio with the curing agent (Sylgard 184
Silicone Elastomer Curing Agent, Dow Corning) and treated overnight at 60°C.
Membrane: To allow diffusion of chemicals from the micro-channel chamber to the flow-through mo-
lecular mechanics chamber, a polycarbonate membrane (0.1 µm pore diameter, 4.18% porosity, Mili-
pore, Billerica, MA) was bound between the two chambers, using a previously described procedure12
.
Briefly, the PDMS micro-channel chamber and the polycarbonate membrane were cleaned using deion-
ized water and dried under nitrogen jet and were then plasma treated (process that consists in bombard-
ing the surface of interest with ions to increase its adhesion properties) in an oxygen plasma chamber
(PVA TePLA AG, Germany) at 60 W for 1 min. The membrane was then immersed in 1% v/v aqueous
APTES (Sigma Aldrich) solution at room temperature for 20 min, washed with deionized water, dried
under nitrogen jet and brought in conformal contact with the freshly plasma treated micro-channel de-
vice so that they bounded together.
Molecular mechanics chamber: The molecular mechanics chamber was created by binding a nitrocellu-
lose coated glass coverslip (50 x 22 x 1,5 mm, Premium Glass, Fisher Scientific) below the membrane
using 40 x 5 mm double sided carbon coated conductive tape (Tedd Pella, Redding, CA). As a reference,
the conventional molecular mechanics flow-through chamber3 was built as follows: a space between the
microscope coverslips was created with plastic shims (40mm x 3mm, 0.125 mm thick) spaced 5mm
apart, glued using liquid adhesive (NOA, Norland Products, Cranbury, NJ) cured using ultra-violet
light5. Note that this type of gluing was not possible for binding the micro-mechanics chamber to the
membrane of the microfluidic device because the liquid adhesive clogged the upper chamber channels
and the membrane pores; thus double coated carbon conductive tape was used.
Protein preparation: Actin was purified from chicken pectoralis as described by Pardee and Spudich13
and fluorescently labeled by overnight incubation with tetramethylrhodamine isothiocyanate (TRITC)-
phalloidin (P1951, Sigma-Aldrich Canada). Myosin was purified from chicken pectoralis following a
previously published protocol14
but using a high ionic strength (0.6 M KCl) acto-myosin extraction
buffer (Dr. A. Sobieszek, personal communication).
Buffers: Actin buffer (25 mM KCl, 25 mM imidazole, 1 mM EGTA, 4 mM MgCl2, and 30 mM DTT,
with an oxygen scavenger system consisting of 0.25 mg/ml glucose oxidase, 0.045 mg/ml catalase, and
5.75 mg/ml glucose; pH adjusted to 7.4); Myosin buffer (300 mM KCl, 25 mM imidazole, 1 mM
EGTA, 4 mM MgCl2, and 30 mM DTT; pH adjusted to 7.4); Motility assay buffer: The in vitro motility
assay buffer consisted of actin buffer to which were added MgATP (2 mM) and methylcellulose (0.5%,
viscous solution to favor the interactions between myosin and actin). Laser trap assay buffer: Same as
motility buffer except that no MgATP or methylcellulose were added.
Page 4 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
5
In vitro motility assay: Our in vitro motility assay has been described5. Briefly, degraded myosin was
removed by ultracentrifugation (Optima ultracentrifuge L-90K and 42.2 Ti rotor, Beckman Coulter,
Fullerton, CA) at 500 µg/ml with equimolar filamentous actin and 1 mM MgATP, in myosin buffer. My-
osin was then perfused in the flow-through chamber at a concentration of 200 µg/ml and allowed to ran-
domly attach to the nitrocellulose for 2 min. The following solutions were then perfused successively in
the flow- through chamber: BSA (0.5 mg/ml in actin buffer), unlabeled G-actin (1.33 µM in actin buff-
er) to bind to any remaining non-functional myosin, MgATP (1 mM in actin buffer) to release the unla-
belled actin from the functional myosin, actin buffer (two washes), TRITC labeled actin (0.03 µM in
actin buffer), incubated for 1 min, and assay buffer. Motility was assessed using an inverted microscope
(IX70, Olympus, Melville, NY) equipped with a high numerical aperture objective (X100 magnification
Ach 1.25 numerical aperture, Olympus, Melville, NY) and rhodamine epifluorescence. A charge cou-
pled device (CCD) camera (KP-E50, Hitachi Kokusai Electric, Woodbury, NY, 720 x 480 resolution,
68.6 µm x 45.7 µm real frame size, 29.94 frame/s, 8 bit grayscale) was used to visualize and record the
actin filament movement on computer (Custom Built by Norbec Communication, Montreal, QC) using a
data acquisition card (Pinnacle Studio AV/DV V.9 PCI Card, Corel Corp., Ottawa, ON) and image cap-
turing software (AMCap software V9.20) at 29.94 Hz. νmax was determined from the total path de-
scribed by the filaments divided by the elapsed time using our automated version15
of the National Insti-
tutes of Health tracking software (NIH macro in Scion Image 4.02, Scion) coded in Matlab (R2011a).
Only filaments present for at least 20% of the recorded video time (15 s) and describing a path of at
least 3µm were analyzed.
Laser trap assay: Our laser trap has been described before5. Briefly, it consists of a Laser Tweezers
Workstation (Cell Robotics, Albuquerque, NM) and the motility assay described above. The trap was
created by a diode pumped Nd:YAG solid-state laser (TEM00, 1.5 W, 1064 nm). The microspheres (3
µm polystyrene, Polybead, Polysciences, Warrington, PA) were visualized in bright field by a second
CCD camera (XC-75, Sony Corporation of America, New York, NY) and the images were recorded on
computer with a frame grabber (Aexeon Quattro, dPict Imaging, Indianapolis, IN).
The trapped microsphere displacement was tracked using our Matlab customized software for optimal
fitting of a reference image5. Briefly, a segment of the first frame of the video (720x480 resolution,
68.6x40.3 µm real frame size, 29.94 frame/s, 8 bit grayscale) that contained the image of the micro-
sphere was matched to a segment of analogous size in the subsequent frames. This reference image of
the microsphere was centered on the center of mass of the largest area of connected pixels above a
threshold gray value. The position of the microsphere in the frame analyzed was the position at which
the sum of the absolute differences in pixel grey values between the current frame and the reference
image was minimized. Interpolation was used to obtain sub-pixel resolution5. The microsphere position
distribution was evaluated by computing the standard distance16
, a spatial statistic tool that allows the
assessment of the data distribution around a center and is calculated as the square root of the sum of the
variance of the microsphere’s coordinates.
TIRF: To estimate the time required for chemicals to travel from the inlet of the micro channel chamber
to the molecular mechanics bottom surface, fluorescent bovine serum albumin (FITc labeled BSA #BS1-
FC-1, NANOCS, New York, NY) was injected and its presence at the motility surface was measured by
total internal reflection fluorescence (TIRF) microscopy (Diskovery TIRF, Spectral Applied Research,
Richmond Hill, ON). A region of approximately 100 nm from the coverslip surface was excited by a
488 nm diode pumped solid state (DPSS) laser through a X63 magnification, 1.47 numerical aperture,
oil immersed objective (Leica HCX PL APO, correction collar objective) of an inverted microscope
(Leica DMI 600B, Leica Microsystems, Wetzlar, Germany). To adapt the microfluidic chamber to the
TIRF microscope, the coverslip was replaced by a cell culture dish (Fluorodish FD 35-100, World Preci-
Page 5 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
6
sion Instruments, Saratosa, FL). Before the image acquisition, the molecular mechanics chamber was
loaded with PBS buffer (1.37 M NaCl, 26 mM KCl, 280mM Na2PO4, 15 mM KH2PO4, pH 7.2) and
the whole microfluidic device was placed on the microscope stage. Then, the inlet of the microfluidic
device was connected to a syringe pump (KDS 210, kdScientific, Holliston, MA) by small diameter
tubing (PE-10, Warner Instruments, Hamden, CT) already filled with FITc labeled BSA (500 µM) in
PBS buffer with 2mM sodium azide. Images were recorded at 10 Hz with an electron multiplying CCD
camera (ImagEM, Hamamatsu Corp., Bridgewater, NJ) using data acquisition software (MetaMorph,
Molecular Devices, Sunnyvale, CA). The acquired videos were comprised of a 60 s period before the
syringe pump triggering, which allowed the assessment of baseline fluorescence of PBS, followed by a
120 s recording period of the fluorescence level after the pump started delivering the FITc BSA at a con-
stant flow rate of 0.7 ml/min. Images were analyzed using NIH image processing and analysis software
(ImageJ).
Page 6 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
7
RESULTS
The performance of the microfluidic device was tested at several intermediate steps of its construction.
First, several channel designs (Fig. 3A) were created and compared by injecting a dye in the chamber
inlet (syringe pump at a constant flow rate of 0.7 ml/min). Because all patterns tested showed very simi-
lar filling time, the choice of the design was based on the highest channel surface area available, thus
promoting diffusion, without compromising the surface area of the walls of the channel, thus promoting
better binding to the membrane.
The binding of the membrane was tested manually by attempting to detach it from the micro-channel
PDMS chamber. The maximal channel width was also limited by the collapsing of the membrane into
the molecular mechanics chamber. Thus the pattern circled in figure 3 was chosen (channel dimensions:
5.4 mm x 1mm x 0.1mm (length x width x height); wall width: 0.15 mm; overall channel chamber di-
mensions: 18mm x 18mm; and the thickness of the PDMS varied between 5-10 mm).
Next, the effect of the membrane on the filling time of the channels of the micro-channel chamber was
assessed. No statistically significant difference in the filling time was observed if the bottom of the mi-
cro-channel chamber was made of plasma treated membranes (1.02 ± 0.15 s, mean ± SE) or plasma
treated glass coverslips (1.12 ± 0.18 s, p = 0.765; Fig. 3B).
The bio-compatibility of the materials used to build the microfluidic device was assessed using the in
vitro motility assay. Firstly, the biocompatibility of the carbon coated double sided tape used to bind the
molecular mechanics chamber to the membrane was assessed. νmax obtained from the conventional flow-
through chambers built using the double coated carbon tape (6.24 ± 0.36 µm/s, mean±SE) was not sig-
nificantly different from that obtained with the plastic shims conventional chambers (5.97 ± 0.31 µm/s,
p = 0.610, Fig. 4A). Next, the biocompatibility of the membrane treated with the adhesion promoter
(APTES) was assessed. There was no significant difference in νmax between the chambers that used the
APTES treated membranes (3.41 ± 0.32 µm/s) and the untreated membranes (4.14±0.51; p = 0.287, Fig.
4B). However, the membrane per se decreased νmax (compare Fig.4A and 4B) and led to a stop and go
movement of the filaments (data not shown).
To verify whether diffusion through the membrane induced any mechanical perturbations in the me-
chanics chamber, the microsphere position in the laser trap was recorded while buffers were injected in
the micro-channel chamber inlet and compared to that obtained with the conventional assay. The micro-
sphere standard distance calculated during buffer diffusion using the mirofluidic device (14.9 ± 3.2 nm,
mean ± SE, N = 5) was not statistically different from that using the conventional assay (15.6 ± 5.3 nm,
N = 3, p=0.922, Fig.5).
To determine the delivery time of chemicals to the molecular mechanics surface, FITc labeled BSA was
infused in the inlet of the microfluidic device and the time required to change the fluorescence level
within 100 nm of the mechanics chamber surface was assessed by TIRF microscopy. Figure 6 shows
that 2.7 ± 0.1 s (mean ± SE) were required to detect changes in the level of fluorescence, at the molecu-
lar mechanics surface.
To verify that this microfluidic chamber was adequate to alter the assay conditions during a motility or
laser trap experiment, a final performance test was done which consisted in infusing MgATP in the inlet
of the micro-channel chamber and recording νmax in the molecular mechanics chamber. This νmax was
compared to that obtained when MgATP was injected from the side of the mechanics chamber, as for the
Page 7 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
8
standard assay. νmax obtained with ATP injected from the inlet of the micro-channel chamber (2.37 ±
0.48 µm/s, N=3) was not statistically different from that obtained with ATP injected from the side of the
flow-through chamber (2.52 ± 0.42 µm/s, N=3; p = 0.822). Note that the assays performed here were
done under laser trapping conditions which are not the same as the motility conditions (i.e. no unlabeled
G-actin and MgATP steps and no methylcellulose). Therefore, these values of νmax cannot be compared
to the results reported figure 4.
Page 8 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
9
DISCUSSION
In this study, we provided a proof of concept prototype of a microfluidic device to efficiently introduce
chemicals in a molecular mechanics flow-through chamber while preventing bulk flow that would oth-
erwise disturb the single molecule assays. The study also addressed biocompatibility of the components.
The final channel design was obtained by trial and error as several solutions were possible. The filling
time of the channels (Fig.3) was not affected significantly by the channel architecture whereas the bind-
ing of the membrane was somewhat more sensitive to the surface area. The biocompatibility of the ma-
terials used to construct the device was similar to that of the conventional chambers (Fig.4). However, a
decrease in νmax was observed when the glass coverslip was replaced by the membrane (compare figs.4A
and 4B). This is most likely due to the binding of myosin to the membrane, thus decreasing the myosin
density on the motility surface. Such a decrease in myosin density would also explain the stop and go
behavior observed for the actin filaments in the presence of the membrane. It is also possible that some
back diffusion through the membrane occurred, thereby altering the conditions of the motility assay.
Thus, further optimization may still be required for the laser trap assay conditions (e.g. working at high-
er myosin concentrations, etc.) when using the microfluidic device.
The diffusion of chemicals through the membrane does not introduce mechanical perturbations in mo-
lecular mechanics assays (Fig.5). The standard distance of a trapped microsphere during diffusion was
not significantly different from that recorded in a conventional assay, without any diffusion. Moreover,
the standard distance obtained (~15 nm) is below the spatial resolution of our camera (1 pixel ~ 100 nm)
and so was obtained by interpolation. The standard distance is a measure of the noise in our assay and is
mostly thermal and optical. If this microfluidic device is to be used for measurements in which actin
filaments are to be in contact with myosin on a pedestal, then the system stiffness will increase, leading
to an even smaller influence of external factors such as diffusion through the membrane.
The time required for a chemical or protein to reach the molecular mechanics surface was less than 3
seconds (Fig.6). Thus this chamber would certainly be useful to study such phenomena as the latch-state
where the events occur over a period of minutes. The behavior of several molecular motors is also usu-
ally studied over several seconds so once again this microfluidic chamber would be an excellent tool to
study their kinetics.
The final test performed was aimed at verifying if a chemical could be injected in the micro-channel
chamber to alter the conditions in the molecular mechanics chamber. Thus, we chose to initiate the mo-
tility in the molecular mechanics chamber by injecting MgATP in the micro-channel chamber. However,
the conditions that were used were those typical for laser trap assays because this is the ultimate use of
this chamber. That is, no methylcellulose was added because it would have slowed down the diffusion
of the MgATP. Because methylcellulose favors the binding of myosin to actin, it improves the motility
assay. Also, the unlabeled G- actin infusion followed by the MgATP wash were also skipped because of
the risk of contaminating the molecular mechanics chamber with MgATP. These steps are meant to re-
move the effect of degraded myosin. Thus, these less than optimal motility assay conditions led to a
further decrease in νmax. Nonetheless, the design of the chamber was successfully demonstrated because
νmax obtained when the MgATP was injected in the micro-channel chamber was not significantly differ-
ent from that obtained when the MgATP was injected directly in the flow-through chamber, under the
same experimental conditions.
Page 9 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
10
CONCLUSION
In conclusion, these results provide a proof of concept for a microfluidic device aimed at injecting
chemicals in a laser trap assay without creating bulk flows that would otherwise affect the molecular
mechanics measurements.
Page 10 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
11
FIGURE CAPTIONS
Figure 1 (A) Conventional molecular mechanics assay chamber. Molecular mechanics flow-through
chamber made up of 2 coverslips (one coated with nitrocellulose to bind the myosin molecules) separat-
ed by plastic shims bounded by a curable glue. The detailed view of the laser trap assay shows polysty-
rene microspheres captured in the laser traps. The surface of these microspheres is biochemically altered
so that a fluorescently labeled actin filament can be attached between them and be brought in the prox-
imity of a glass pedestal coated with nitrocellulose and myosin molecules. (Dimensions are not to
scale). (B) Micro-fluidic device. Prototype micro-fluidic device to replace the conventional flow-
through chamber in order to introduce chemicals without creating any bulk flow that would disturb the
mechanics measurements. (See figure 2 for the details on microfluidic device).
Figure 2: Microfluidic device: The design of the microfluidic device includes 1) a micro-channel
chamber (horizontal channels) built in polydimethylosiloxane (PDMS) to distribute the chemicals rapid-
ly and uniformly over a membrane; 2) a polycarbonate membrane to allow the chemicals to reach the
molecular mechanics chamber without creating any bulk flow; 3) a molecular mechanics chamber.
(Dimensions are not to scale).
Figure 3: A. Channel design: Several patterns of channels were tested to ensure uniform and efficient
chemical distribution. The choice of the design was based on the best compromise between channel vol-
ume for bulk flow and binding surface for the membrane. The pattern circled was chosen (see text for
dimensions). B. Channel filling time: The time needed for uniform filling of the channels was meas-
ured using a dye for the micro-channel chamber sealed by plasma treated glass coverslips (N = 4) or by
membranes (N = 4).
Figure 4: Biocompatibility test: Velocity (νmax) of fluorescently labelled actin filaments when pro-
pelled by myosin randomly adhered to the coverslip. A) The measurements were performed using the
conventional assay in the presence of plastic shims (N = 3) or double sided carbon conductive tape
strips (N = 3). B) The measurements were performed with untreated membranes (N = 3) or APTES
treated membranes (N = 3), bound to the micro-channel chambers.
Figure 5: Microsphere position standard distance: The trapped microsphere standard distance when
using the conventional assay (N=3) and the microfluidic device (N=5), during diffusion.
Figure 6: Time needed for proteins to reach the assay: The time needed for FITc labelled BSA to
reach the assay surface was measured as the time elapsed between the triggering of the syringe pump
and the moment when the fluorescence level changed at the assay surface. The syringe pump was trig-
gered at time = 0 s. The fluorescence level for all three experiments was normalized between their min-
imal and maximal level recorded during the entire length of the video recording (180 s); N=3.
Page 11 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
12
ACKNOWLEDGMENT
This study was supported by the Natural Sciences and Engineering Council of Canada (NSERC). We
thank the personnel of McGill Microfab facility for their constant support in constructing the microflu-
idic device. We also thank Drs. Claire Brown, Erika Tse-Luen Wee, Aurelie Cleret-Buhot for their help
with the TIRF microscopy experiments performed at the McGill University Life Sciences Complex Ad-
vanced BioImaging Facility (ABIF), supported by the Canadian Foundation for Innovaiton (CFI) and
the Ministère du Développement économique, innovation et exportation Québec (MDEIE).
Page 12 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
13
REFERENCES
(1) Kron, S. J.; Spudich, J. A. Proc. Natl. Acad. Sci. U. S. A. 1986, 83, 6272-6276.
(2) Finer, J. T.; Simmons, R. M.; Spudich, J. A. Nature 1994, 368, 113-119.
(3) Warshaw, D. M.; Desrosiers, J. M.; Work, S. S.; Trybus, K. M. J. Cell Biol. 1990, 111, 453-463.
(4) Work, S. S.; Warshaw, D. M. Anal. Biochem. 1992, 202, 275-285.
(5) Roman, H. N.; Zitouni, N. B.; Kachmar, L.; Ijpma, G.; Hilbert, L.; Matusovskiy, O.; Benedetti, A.;
Sobieszek, A.; Lauzon, A. M. Biochim. Biophys. Acta 2013, 1830, 4634-4641.
(6) Shirinsky, V. P.; Biryukov, K. G.; Hettasch, J. M.; Sellers, J. R. J. Biol. Chem. 1992, 267, 15886-
15892.
(7) Dillon, P. F.; Aksoy, M. O.; Driska, S. P.; Murphy, R. A. Science 1981, 211, 495-497.
(8) Debold, E. P.; Walcott, S.; Woodward, M.; Turner, M. A. Biophys. J. 2013, 105, 2374-2384.
(9) Hilbert, L.; Cumarasamy, S.; Zitouni, N. B.; Mackey, M. C.; Lauzon, A. M. Biophys. J. 2013, 105,
1466-1474.
(10) Placais, P. Y.; Balland, M.; Guerin, T.; Joanny, J. F.; Martin, P. Physical review letters 2009, 103,
158102.
(11) Campo, A.; Greiner, C. J. Micromech. Microeng 2007, 17, R81-95.
(12) Sunkara, V.; Park, D. K.; Hwang, H.; Chantiwas, R.; Soper, S. A.; Cho, Y. K. Lab on a chip 2011,
11, 962-965.
(13) Pardee, J. D.; Spudich, J. A. Methods Enzymol. 1982, 85, 164-181.
(14) Sobieszek, A. In Airways smooth muscle: Biochemical Control of Contraction and Relaxation,
Giembycz, D. R. a. M. A., Ed.; Birkhauser: Bassel, 1994, pp 1-29.
(15) Hilbert, L.; Bates, G.; Roman, H. N.; Blumenthal, J. L.; Zitouni, N. B.; Sobieszek, A.; Mackey, M.
C.; Lauzon, A. M. PLoS computational biology 2013, 9, e1003273.
(16) Levine, N. J. Am. Plann. Assoc. 1996, 62, 381-391.
Page 13 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Figure 1
254x190mm (300 x 300 DPI)
Page 14 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Figure 2
254x190mm (300 x 300 DPI)
Page 15 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Figure 3
200x99mm (300 x 300 DPI)
Page 16 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Figure 4
218x104mm (300 x 300 DPI)
Page 17 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Figure 5
144x109mm (299 x 299 DPI)
Page 18 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Figure 6
398x273mm (72 x 72 DPI)
Page 19 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
For TOC Only
Page 20 of 20
ACS Paragon Plus Environment
Analytical Chemistry
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960