Microfluidic Migration and Wound Healing Assay Based on

Subscriber access provided by UNIVERSITY OF SOUTH CAROLINA LIBRARIES

Article

Microfluidic migration and wound healing assay based on mechanically induced injuries of defined and highly reproducible areas Drago Sticker, Sarah Lechner, Christian Jungreuthmayer, Jürgen Zanghellini, and Peter Ertl Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b03886 • Publication Date (Web): 18 Jan 2017 Downloaded from http://pubs.acs.org on January 25, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Microfluidic migration and wound healing assay based on mechanically induced injuries of defined and highly reproducible areas Drago Sticker‡†, Sarah Lechner‡, Christian Jungreuthmayer§ǁ, Jürgen Zanghellini§¥ and Peter Ertl£*

‡

BioSensor Technologies, AIT Austrian Institute of Technology GmbH, Muthgasse 11,

1190 Vienna, Austria. §

Bioinformatics and High Performance Computing, Austrian Centre of Industrial

Biotechnology, Muthgasse 11, 1190 Vienna, Austria. ǁ

TGM - Technologisches Gewerbemuseum, Wexstraße 19-23, 1200 Vienna, Austria.

¥

Department of Biotechnology, University of Natural Resources and Life Sciences,

Muthgasse 18, 1190 Vienna, Austria. £

Faculty of Technical Chemistry, Vienna University of Technology, Getreidemarkt 9,

1060 Vienna, Austria.

*

Corresponding author:

Email: [email protected] †

Present

address:

Department

of

Pharmacy,

University

of

Copenhagen,

Universitetsparken 2, 2100 Copenhagen, Denmark. Email: [email protected]

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract

All cell migration and wound healing assays are based on the inherent ability of adherent cells to move into adjacent cell-free areas, thus providing information on cell culture viability, cellular mechanisms and multicellular movements. Despite their wide-spread use for toxicological screening, biomedical research and pharmaceutical studies, to date no satisfactory technological solutions are available for the automated, miniaturized and integrated induction of defined wound areas. To bridge this technological gap we have developed a lab-on-a-chip capable of mechanically inducing circular cell-free areas within confluent cell layers. The microdevices were fabricated using off-stoichiometric thiol-ene-epoxy (OSTEMER) polymer resulting in hard-polymer devices that are robust, cost-effective and disposable. We show that the pneumatically-controlled membrane deflection/compression method not only generates highly reproducible (RSD 4%) injuries but also allows for repeated wounding in microfluidic environments. Performance analysis demonstrated that applied surface coating remains intact even after multiple wounding, while cell debris is simultaneously removed using laminar flow conditions. Furthermore, only a few injured cells were found along the edge of the circular cell-free areas, thus allowing reliable and reproducible cell migration of a wide range of surface sensitive anchorage dependent cell types. Practical application is demonstrated by investigating healing progression and endothelial cell migration in the absence and presence of an inflammatory cytokine (TNF-α) and a well-known cell proliferation inhibitor (Mitomycin-C).

2


Page 2 of 29

Page 3 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION Cell migration is the movement of cells in response to biological signals or environmental cues and plays a vital role in a variety of key physiological processes including immune cell recruitment, wound healing, tissue repair and embryonic morphogenesis.1,2 For instance, in the case of infection, immune cells rapidly move from the lymph nodes via the circulatory system towards the infected site.3 During wound healing, however, cells continuously migrate to the injured tissue to repair the damage. In turn, abnormal cell migration resulting in the relocation of cell types to unsuitable tissue sites could lead to serious consequences such as cardiovascular disease, arthritis, tumor formation, cancer metastasis, and, in the case of embryonic cells, fetal malformations.4 It has been recently recognized that a better understanding of the complex processes that govern cell migration is needed to foster the development of novel therapeutic strategies to combat disease.5, 6 Cell migration can be activated by a number of biological, chemical and physical

signals

including

mechano-transduction,

chemical

signaling

and

molecular interactions.7, 8 Due to its complexity, a thorough understanding of cell population movements and collective cell behavior can only be achieved through monitoring under physiologically relevant conditions.9 However, in vivo cell migration studies using state-of-the-art imaging methods is limited due to the ethics associated with animal testing. This means that, to date, the only viable means of studying cell migration involves in vitro migration assays. In fact, in vitro migration assays are extensively used by biologists, pharmacologists, medical researchers and toxicologists for a broad range of screening applications including viability and cytotoxicity studies. Due to the inherent ability of cells to migrate into cell-free areas, the main in vitro methods used to study cell migration are based on either cell exclusion or cell depletion to create cell-free areas within a cell culture layer.8 Cell exclusion uses removable stencils to confine areas of cell growth, is labor intensive, but causes little damage to cells along the edge of the cell-free area, which is important

3



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

because cell injury affects cell signaling, which in turn affects cell migration.9-11 On the other hand, the cell depletion method relies on electrical currents, thermal heating, laser ablation, (bio)chemicals and sharp objects to rupture cell-to-cell junctions and remove cells from a confluent cell culture.12, 13 Also known as the scratch assay, the mechanical damage of cell surface layers is the most commonly used cell depletion approach.5, 14 The scratch assay consists of manually creating a cell-free area within a monolayer by scratching the cell culture surface using either plastic pipette tips or specialized blades. Although the scratch assay is popular due to its simplicity and cost effectiveness, a number of problems are associated with manually inducing injuries. The main problem is lack of reproducible wounding and the resulting irregular cell-free areas containing voids and jagged edges,15, 16 which limit the repeatability of the assay. Another problem is the unintended removal of cell adhesion promoters from the scratched surface, which are essential for cell attachment and spreading.5 Lastly, injury to the cells at the edge of the cellfree area releases intracellular components and signaling molecules known to negatively influence cell migration.10,

11

To address these shortcomings,

microfluidic cell culture systems that also take fluid mechanical shear stress, medium supply and constant waste removal into account,17, 18 have been recently introduced as alternatives for both cell depletion and exclusion approaches. For instance, one microfluidics cell depletion approach creates cell-free areas using enzymatic cell removal strategies that avoid cell injury.21-24 A prominent microfluidic cell exclusion assay, in the other hand, is based on the manual insertion of microstencils to generate a cell-free area,25 while another assay relies on capillary force to create cell-free areas without significant cell damage.26 Despite the achievements of microfluidic migration and wound healing assays thus far, the development of automated, miniaturized and integrated devices capable of reliably creating defined cell-free areas within cell monolayers under reproducible and standardized measurement conditions is still in its infancy. To advance existing microfluidic cell migration assays we have developed two actuator-integrated microdevices; one microdevice induces mechanical wounds based on cell exclusion and the other on cell depletion (see also Fig. 1). Our

4


Page 4 of 29

Page 5 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


devices offer all the advantages but none of the drawbacks of existing wound healing assays in a single device, including: (1)

Creation of reproducible circular cell-free areas;

(2)

Repeatable wounding of the same area without removal of the extra cellular matrix (ECM) coating;

(3)

Reduction of injured cells along the edge of the cell-free area;

(4)

Removal of cell debris during perfusion;

(5)

Integration, miniaturization and automation.

Our cell migration assays employ endothelial cells, which means that proper flow conditions is crucial for wound healing and cell migration, since the proper development of the inner lining of blood vessels (endothelium) after injury requires physiological shear forces.19,20 Wounding will be performed through integrated mechanical stamping, which is the controlled compression of a flexible membrane using pneumatic actuation. Mechanical stamping enables both cell exclusion and cell depletion approaches under robust, reliable and standardized measurement conditions. Pressure profiles and applied wounding areas are experimentally optimized and computationally evaluated using Finite Element Method (FEM) simulations to estimate the critical pressure needed to effectively remove adherent cells. The performance of our microfluidic wound healing assay is compared against the classical scratch assay results based on the following criteria: (a) damages to the cell culture substrate and removal of the applied surface coating, (b) number of injured cells along the wound edge, and (c) assay reproducibility. Practical application of the membrane-integrated microfluidic wound healing assay is demonstrated by investigating the modulating effect of non-cytotoxic compounds, such as cytokines and proliferation inhibitors, on cell migration rates.

EXPERIMENTAL Device designs and microfabrication Two microdevices containing integrated pneumatically-activated actuators were designed for either cell depletion, using a flexible membrane, or cell 5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 29

exclusion, using an integrated microstencil (Fig. 1B). Both devices consisted of three layers: (1) a top pneumatic layer, (2) a middle flexible PDMS membrane, and (3) a microchannel network bottom layer. In the pneumatic layer a 1.5 mm and 2.5 mm dia. circular shaped frame for cell depletion and cell exclusion was placed in the middle of the four 90 µm high, 2.5 mm wide and 7.5 mm long cell culture chambers. Figure 2 provides an overview of the employed fabrication and assembly strategy of the microdevices using dryfilm resist (DFR) TMMF S2045 (Tokyo Ohka Kogyo Co., Ltd) as master molds. Fabrication involved a replica molded dual-cure thermoset method previously described27 followed by the integration of a 250 µm thick PDMS membrane (HT-6240, Silex, UK). The APTES (A3648, Sigma Aldrich) silanized PDMS membrane was carefully applied onto the OSTEMER pneumatic layer and baked over night at 110°C to achieve covalent bonding between thiols and epoxy groups (Fig. 2f). Next, the PDMS membrane with the bonded OSTEMER layer was silanized as shown in Fig. 2a-b. The fluidic layer was casted in the same manner as the pneumatic layer (Fig. 2c-e). Finally, the aligned OSTEMER-PDMSOSTEMER assembly was fixed using an aluminum manifold. Alternating pressure of 25 kPa and -40 kPa during the overnight curing process was applied to prevent the membrane from sticking to the fluidic layer (Fig. 2h). Both the cell depletion and cell exclusion microdevices were fabricated in a similar manner using the same casting and bonding techniques described above. However, while a commercially-available PDMS membrane was used for the cell depletion device, a spin-coated PDMS containing the microstencil was used in the cell exclusion device. Briefly, a 1:10 mixture of PDMS (Sylgard 184) was degassed and spin coated at 350 rpm for 60 sec on a DFR-structured (90 µm high) silicon mold to create a 300 µm thin structured polymer layer after polymerization at 65°C for 3 h. Additionally, glass was used as a substrate for the exclusion device while OSTEMER was used for cell depletion. Cell culture handling and protein surface coating method Lentivirally-transduced

GFP-HUVEC

were

purchased

from

Olaf

pharmaceuticals (Worcester, USA) and cultivated using gelatin coated 25 cm2 cell 6


Page 7 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


culture flasks and maintained in EGM-2 (CC3156, Lonza) supplemented with EGM-2 SingleQuots (CC4176, Lonza). The coating was performed using 1% gelatin solution (9000-70-8, Sigma Aldrich) in DPBS and incubated for 30 min at 37°C. For on chip experiments, the medium was further supplemented with 1% Hepes buffer (J848, AMRESCO). To evaluate surface coating stability, the pretreated (500 mM NaOH for 15 min at RT) and rinsed (PBS) microchambers were loaded with a protein mixture comprising 10 µg/ml fibronectin (F4759, Sigma Aldrich) and 5 µg/µl fibrinogen conjugated with Alexa Fluor 488 (F13191, Thermo Fisher) and incubated for 1 h at 37°C. Live/dead staining of HUVEC was performed by adding 2 µM ethidium homodimer-1 (L3224, Life Technologies) to the cell medium and incubated for 30 min. To investigate the effect of a cytokine and a proliferation inhibitor on healing behavior, cells were further exposed for one hour to a medium supplemented with either 1 ng/ml of TNF-α or 10 µg/ml Mitomycin-C, then flushed with normal culture medium and subsequently the wound was induced. Experimental setup of the mechanically-induced wound healing assay The microfluidic device was placed on a heating plate (37°C) with a temperature controller to ensure temperature stability. Prior to experimentation, the microchip was disinfected using 70% ethanol, rinsed using a 1 M NaOH solution for 15 min and washed with PBS prior to coating with a 1% gelatin solution for 1 h. The pre-prepared cell suspension was gently loaded into microchannels using a plastic syringe (1 ml) under a microscope. Following a cell adhesion period of 1 h in the absence of fluid flow, a constant medium perfusion of 3 µl/min was applied using a syringe pump. As soon as a confluent cell layer (typically 1 to 2 days) was obtained, a circular wound was mechanically induced by pneumatic actuation of the PDMS membrane. Bending was induced by rapidly increasing the pressure load to 150 kPa (differential pressure) and then immediately releasing to 0 kPa using a pressure controller. The medium perfusion was kept on while mechanical damage was being performed.

7



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Fluorescence microscopy Fluorescence images were taken using a Wilovert AFL30 fluorescence microscope (Hund Wetzlar, Germany) equipped with a DS-Qi1Mc digital camera. All conventional fluorescence images were processed using ImageJ (U.S. National Institutes of Health, USA). Simulation of the membrane deflection A set of numerical elastostatic simulations were performed to describe the effect of load pressure on the membrane, substrate and cell culture. The multiphysics simulation toolbox Starccm+ by CD-Adapco was used to create the geometry, generate the mesh, perform the simulations, and visualize the results. First, a geometrical 3D model of the membrane was created (diameter: 1.5 mm, height: 250 µm) along with a mesh of approximately 20,000 polyhedral elements (ESI Fig. S1A). The Starccm+ finite volume solid stress solver was utilized to compute the deformation of the membrane caused by the applied pressure. Because of extensive deformation of the membrane, the mesh morphing feature of Starccm+ was used to update the geometry of the model after each iteration step (see also ESI Fig. S1B). The used simulation model neglected: (1) the lateral movement of the membrane, (2) fluid flows inside the chamber, and (3) additional static pressure distributions (see also ESI for a detailed description). Quantification of cell migration Using Adobe Photoshop v.11, the fluorescence pictures were transformed to grey scale, the contrast increased and the leading edge of the wound selected using the Magic wand tool. This image was then processed using ImageJ by determining the cell-free area with the tool Analyze Particles.

RESULTS AND DISCUSSION Characterization of planar pneumatically-actuated membrane deflection method for mechanical removal of adherent cells

8


Page 8 of 29

Page 9 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The goal of the current work was to develop microdevices capable of confining cell growth or inducing injuries of a defined area within confluent cell layers by mechanically compressing a flexible membrane. The microfluidic and pneumatic layouts shown in Fig. 1 are designed to form a microfluidic cultivation chamber

containing

embedded

circular shaped,

bendable membranes

or

microstencils in the center of the microfluidic top layer. An important component of the PDMS membrane-integrated microdevice is the thermoset plastic OSTEMER, which is stiff, transparent and gas-tight to avoid device deformation during

operation.28

Proof-of-principle

of

device

operation

was

initially

demonstrated using water-soluble dyes, where inflow of the ink is prevented by the presence of the flexible stamp. In turn, negative pressure above -20 kPa applied to the pneumatic lines lifted the stamp resulting in a uniform distribution of the dye. While cell exclusion using pneumatically-removable stamps is possible, reliable and reproducible cell seeding between the four cell chambers was limited by the presence of the stencil inside the microchambers. As a result, only the cell depletion microdevice was further evaluated for assay reproducibility, reliability and robustness in the following sections. The cell depletion microdevice was subsequently evaluated to withstand high internal pressure build up, since cell membrane rupture and cell lysis is estimated at about 200 kPa. Therefore, to determine the robustness of our membrane-integrated microdevice, the bonding strength between PDMS and OSTEMER was investigated using a simple experimental setup shown in Fig. 3A. In an initial comparative analysis, bonding strength of differently treated PDMS membranes to OSTEMER plastic sheets were investigated using a Ø 1 mm drilled hole that was covered with covalently bonded membranes (see Fig. 3A). The applied PDMS surface measures included corona treatment and silanization procedures using thiol- and amio- end groups. In the case of simple corona treatment, the membrane delaminated at applied pressure of less than 10 kPa at 37 °C, thus necessitating additional surface modifications. Additional silanization of PDMS using amio- (APTES) and thiol- (MPTS) end groups significantly increased the delamination pressure to above 200 kPa. Although minor delamination was

9



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

observed at the frame for pressures above 300 kPa, PDMS-OSTEMER bonding withstood pressures up to 400 kPa as shown in Fig. 3A. This means that bonding strength between the APTES modified PDMS membrane and the OSTEMER substrate is high enough to allow robust and repeated operation of the microdevice using actuation pressure up to 200 kPa. To gain a better understanding of membrane deflection behavior the fluidic microchannel was filled with fluorescein and fluorescence intensity was measured across the deflection zone during pneumatic actuation of the membrane.29 Results shown in Fig. 3B (top view) clearly reveal how the membrane deflected towards the bottom of the microfluidic channel, resulting in the displacement of the watersoluble dye fluorescein, thus decreasing the intensity of fluorescence over the deflection zone. Spatial distribution of fluorescence intensity during membrane deflection is shown in Fig. 3C, where membrane movement is visualized using pixel gray values from fluorescence images taken along the fluid flow direction (line A-A’ in Fig. 3B). Fluorescence intensity significantly decreased in the center of the membrane in the presence of increasing load pressure resulting in a circular deflection zone. Using pressure profiles below 30 kPa, the flexible membrane deflected laterally towards the microfluidic layer until the center of the membrane reached the bottom channel. At pressure higher than 40 kPa, the membranes apex deflected horizontally along the microfluidic channel surface. At a load pressure of 150 kPa, the membrane displayed a circular-shaped contact area with a diameter of 0.9 mm. Since load pressure above 150 kPa did not result in significant improvement in surface area, this pressure setting was used for all subsequent experiments. Simulation of membrane deflection behavior and pressure acting on cells It is important to note that in addition to deflecting the flexible PDMS membrane towards the chamber substrate, a high enough force was needed to disrupt the cell membrane. As a consequence, a set of numerical simulations was performed to estimate pressure profiles exerted by the compressed membrane on the cell layer. Initially the membrane’s Young’s modulus was experimentally determined using fluorescent measurements at 30 kPa pressure load to account for 10


Page 10 of 29

Page 11 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


changes of the elastic properties following silanization of the PDMS membrane. The numerically calculated Young’s modulus of the crosslinked membrane was found to be 1357 kPa, which is above the manufacturer’s specification range of 300 to 800 kPa for the untreated membrane material. The estimated decrease in material elasticity was caused by crosslinking, which typically results in an increase of the Young’s modulus and a decrease of the Poisson coefficient. Using the above Young’s modulus, membrane displacement within the microfluidic channel was estimated using computational analysis methods. In-silico results shown in Fig. 4A depicts the membrane’s displacement using an applied loading pressure of 150 kPa. The side view of the membrane bended onto the substrate (gray) shows the solid stress displacement occurring inside the membrane due to the applied pressure from the top. Additionally, Fig. 4B illustrates the pressure acting on the chamber floor for load pressure of 150 kPa, which is the working pressure used to actuate the membrane during cell removal. Computational analysis revealed that the pressure acting on the chamber substrate was higher than the applied input pressure, which is caused by the top-membrane-frame to substrate-contact area ratio of >1. In other words, while the input pressure acts on the entire top surface (Atop=r2*π=1.8 mm2) of the membrane, the pressure on the chamber substrate solely acts on the reduced contact area (Asub

Microfluidic Migration and Wound Healing Assay Based on

Recommend Documents