Anatomy and Regenerative Biology Faculty Publications Anatomy and Regenerative Biology Integrin Activation by a Cold Atmospheric Plasma Jet Recommended Citation Integrin Activation by a Cold Atmospheric Plasma Jet New Journal of Physics Integrin Activation by a Cold Atmospheric Plasma Jet

Current breakthrough research on cold atmospheric plasma (CAP) demonstrates that CAP has great potential in various areas, including medicine and biology, thus providing a new tool for living tissue treatment. In this paper, we explore potential mechanisms by which CAP alters cell migration and influences cell adhesion. We focus on the study of CAP interaction with fibroblasts and corneal epithelial cells. The data show that fibroblasts and corneal epithelial cells have different thresholds (treatment times) required to achieve maximum inhibition of cell migration. Both cell types reduced their migration rates by ∼30–40% after CAP compared to control cells. Also, the impact of CAP treatment on cell migration and persistence of fibroblasts after integrin activation by MnCl 2 , serum starvation or replating cells onto surfaces coated with integrin ligands is assessed; the results show that activation by MnCl 2 or starvation attenuates cells' responses to plasma. Studies carried out to assess the impact of CAP treatment on the activation state of β1 integrin and focal adhesion size by using immunofluorescence show that fibroblasts have more active β1 integrin on their surface and large focal adhesions after CAP treatment. Based on these data, a thermodynamic model is presented to explain how CAP leads to integrin activation and focal adhesion assembly. 3 Authors to whom any correspondence should be addressed.


Introduction
The migration of cells within tissues plays an important role in many physiological and pathological processes, including embryonic development, wound repair, angiogenesis and metastasis. Various signaling pathways control these processes. Integrins (single-pass type I transmembrane proteins) are a major family of metazoan cell-surface-adhesion receptors playing a key role in the signaling and mechanotransduction mechanisms. Their functions include maintaining cell adhesion, tissue integrity, cell migration and differentiation [1][2][3]. Previous studies revealed that integrins could be affected by the application of cold atmospheric plasma (CAP) [4,5]. Recent studies of CAP jets have shown that they have great potential for use in biomedical applications. Their distinguished physical and chemical properties are defined by the uniqueness of non-thermal non-equilibrium plasmas. Depending on their configuration they can be used for wound healing, sterilization, targeted cell/tissue removal and cancer treatment [6][7][8][9]. It was found that CAP jets can slow down cell migration and change integrin expression on the cell surface, i.e. treatment of cells with CAP jet resulted in a decrease of β1 and αv-integrins ∼24 h after treatment [10]. In addition, the localization effect of cold plasma treatment on cell migration was shown [11]. The plasma effect is determined by the size of the plasma jet and does not alter cell migration outside the treated area. The CAP effect has not only spatial but also temporal dependence since it persists for at least 33 h. Despite the progress made in studying CAP interaction with the living tissue, the question remains: what causes the motility and cell adhesion of cells to change in response to plasma treatment? The answer to this question needs to be obtained to allow application of cold plasma in aiding patients with wound healing [1,12].
The goal of this study is to understand the mechanism by which the CAP jet alters cell migration and influences adhesion. Although the main focus here is on the study of the 3 interaction between the CAP jet and fibroblasts, we also present data comparing fibroblast responses to those of epithelial cells.

Cell culture
Wild-type tertiary passaged mouse fibroblast cells (WTDF 3, the procedures used to obtain cells from mice were authorized and approved by the GWUMC Institutional Animal Care and Use Committee; wt mice (BALB/c) obtained from NCI-Frederick [13]) were cultured in Dulbecco's modified Eagle medium (DMEM; Invitrogen Corp., Carlsbad, CA) enriched with 5% serum, 1% NEAA, 1% l-glutamine, 1% Pen.-Strep. Diluted cells (30% confluence) were plated on multiwell plates or double-well glass slides and treated with plasma on the third day in culture. The human corneal limbal epithelial cell line (HCLE; Dr Ilene Gipson, Harvard Medical School and described in [14]) was grown in HCLE medium (500 ml of Keratinocyte Serum Free Medium with added 1.2 ml of Bovine Pituitary Supplement, 4l µl of Epidermal Growth Factor, 0.17 mM CaCl 2 and 5 ml of Pen.-Strep, Invitrogen Corp., Carlsbad, CA). HCLE cells were used on the second day of culture with a confluence of about 30-40%. During plasma treatment, plates with cells were kept on a heating plate (Boekel Scientific, model 240000, Feasterville, PA) to maintain the temperature of the medium at 37 • C. Serum starvation experiments were performed on WTDF 3 cells and placing the cells in DMEM with 1% serum ∼24 h before the plasma treatment. For studies using MnCl 2 to activate integrins, 0.5 mM MnCl 2 was added to DMEM with 5% serum immediately after cells were subjected to cold plasma treatment.
Some cold plasma treatments were conducted on cells that had been placed in plates that had been pre-coated with extracellular matrix proteins. Coating of tissue culture plates was performed as described in [12]. Fibroblasts were either placed onto plates pre-coated with 10 µg ml −1 fibronectin (FN; BD Bioscience, Pasadena, CA, catalog no. 354008)/1% collagen (CN) type I (Advanced BioMatrix, San Diego, CA, part no. 5005) or with 10 µg ml −1 vitronectin (VN) (BD Bioscience, catalog no. 354238). Plasma treatment was performed 8 h after replating.

Cold atmospheric plasma (CAP) jet
The cold plasma jet (see figure 1) is the dielectric barrier discharge in helium that is described elsewhere [4,15]. The output voltage is about 4.5-5 kV, the frequency is ∼13 kHz, the helium flow rate ∼11-12 liter min −1 and the average jet power ∼4 W. The distance between the jet outlet and the culture plate/slide was always kept at about 20 mm. The depth of the medium was maintained at ∼2 mm. Dermal fibroblasts and epithelial cells were prepared and equivalent numbers were plated onto the culture plastic (for time-lapse studies) or glass slide (for immunostaining) and maintained in DMEM supplemented with 5% serum (fibroblasts) or HCLE medium (epithelial cells). Fresh medium or fresh medium with MnCl 2 was added to cells following plasma treatment.

Time-lapse studies
Time-lapse studies were conducted on an Olympus IX81 research microscope (Olympus America, Center Valley, PA) equipped with a Proscan motorized stage (Prior Scientific Instruments Ltd, Rockland, MA) and placed in a temperature-and CO 2 -controlled chamber (LiveCell Incubation System, Neue Biosciences, Camp Hill, PA). Using relief-contrast optics, images were taken every 10 min for 16 h 40 min (100 images). Images were transferred to a workstation equipped with Metamorph image analysis software (Molecular Devices Corporation, Chicago, IL) where velocities of ten cells were calculated using the track cell module in each tracked location. A more detailed description can be found elsewhere [13]. At least 80 cells were analyzed per conditional change, such as the time point and/or treatment. The tracking experiment was started immediately after treatment.

Immunofluorescent studies
For experiments involving immunofluorescent staining of cells, cells were plated onto two-well glass chamber slides for 24 h prior to CAP treatment. For cold plasma treatment, chamber slides containing cells were placed on the heating plate to maintain the temperature and treated with cold plasma as described above. Fresh medium was added and cells were returned to the 37 • C CO 2 incubator for 3 h. The medium was discarded and cells were fixed in 4% paraformaldehyde (product no. 28908, Thermo Scientific, Rockford, IL) in PBS for 10 min and then rinsed in PBS. Non-specific staining of cells was blocked by incubating the cells for 15 min in blocking buffer (1% bovine serum albumin in 1× PBS). Cells were incubated in an appropriate dilution of each of the primary antibodies below for 1 h, followed by washing (15 min in PBS) and incubation in secondary antibodies. When the primary antibodies used were generated in different species, cells were stained with multiple antibodies simultaneously. For studies of active and total surface integrins, cells were used for immunostaining without permeabilization. However, for studies of vinculin, and the αv receptor within focal adhesions, cells were permeabilized by incubating in 0.1% Triton X-100 for 10 min and then incubated in PBS. Permeabilization is needed because vinculin is an intracellular protein and antibodies to detect it cannot penetrate the cell unless it is permeabilized. Total b1 integrin was detected with the Hamster derived CD 29 monoclonal antibody clone Ha2/5 (BD Bioscience, catalog no. 555004) at a dilution of 1:200. For activated b1 integrin, the rat-derived monoclonal antibody against activated CD 29, clone 9EG7 (BD Bioscience, catalog no. 553715), was used at a dilution of 1:200. αV integrin was visualized using the rat-derived monoclonal antibody against CD51 RMV7 (BD Bioscience, catalog no. 552299) at a dilution of 1:200. Focal adhesions were visualized by staining for the focal adhesion protein vinculin using the mouse-derived monoclonal antibody clone VIIF9 (7F9) at a dilution of 1:200 (catalog no. MAB3574, Chemicon Int., Temecula, CA). Nuclei were visualized using DAPI, dilution 1:2000 (D21490; Invitrogen Corp.). Images were acquired at 20× magnification using a Nikon fluorescent microscope equipped with an RT-Slider SPOT camera (Melville, NY).

Data analysis
Adobe Photoshop 7.0 was used to manage the images. For quantification analysis Image Pro Plus, version 6.2 software (Media Cybernetics Inc.) was used. Data were subjected to statistical analyses using InStat software (Graphpad Software, Inc., San Diego, CA). The differences between groups were determined by two-tailed unpaired t-tests and data were considered significant for p-values < 0.05. When standard deviations between groups were not equal, significance was determined using Welch corrected unpaired t tests and when the distributions of values within a data set were not Gaussian, significance was determined using the Mann-Whitney test. Histograms were generated using mean and standard error of the mean (SEM) values [13].

Dermal fibroblasts and human corneal epithelial cells reduce their migration in response to CAP
While we had reported that dermal fibroblasts migrate more slowly after CAP treatment, we are yet to evaluate a different type of cell, epithelial cells, to determine their response to CAP. Presented in figure 2(A) are relief contrast images of the tracks taken (red lines) by migrating fibroblasts (WTDF 3) and epithelial (HCLE) cells with and without CAP treatment over a 16 h and 40 min time period. The tracks of the cells after plasma treatment (fibroblast cells were treated for ∼60 s and epithelial cells for ∼100 s) are significantly shorter than untreated cells (control). Additional experiments were conducted to assess the impact of increasing plasma treatment times on cell migration rates of fibroblasts and epithelial cells, i.e. to determine the threshold of treatment (threshold here is determined as the treatment dose (duration of treatment) after which the change becomes statistically non-significant (i.e. p > 0.05 between two neighbor points) and no other changes occur). In figure 2(B), cell migration rates are shown as a function of the duration of CAP treatment. Fibroblasts show a maximum drop in the migration rate at about 40 s; increasing treatment times ranging from 40 to 200 s do not alter cell migration rates. The impact of CAP on the migration of epithelial cells is more graded; migration rates of epithelial cells decrease with increasing treatment times from 10 to 100 s; treatment times between 100 and 200 s do not alter epithelial cell migration rates. The decrease in cell migration rates for both cell types is about 30-40%. It was shown earlier that 7 helium treatment alone does not affect the cell migration without additional conditions such as serum starvation or MnCl 2 activation [10,11], so only untreated cells were used as the controls. Figure 2(C) shows the persistence of cell motion as a function of the duration of the plasma treatment. The persistence was measured as the ratio between net and total displacements and thus represents the directionality of cell motion. Neither fibroblasts nor corneal epithelial cells show significant changes in persistence as a function of increasing plasma treatment time.

Activating fibroblast integrins reduces their response to CAP
One of the possible reasons for the decrease in cell migration rate after CAP treatment is the activation of integrins on cell surfaces. Integrins can be activated by adding MnCl 2 to their media [1,16]. This treatment shifts the integrins from a folded state to an unfolded state exposing their ligand-binding sites. Serum starvation of fibroblasts increases their spreading, which induces activation of α5β1 integrin [17]. Figure 3(A) shows the migration rates of fibroblasts: controls (not treated), only helium treated for 100 s and CAP treated (plasma treated for 100 s) followed by incubation in standard medium, medium supplemented with MnCl 2 and low-serum medium.
In the case of no additional treatment, we found a 30-40% drop in the migration rates of the cells as expected. However, under conditions where integrins were activated, no change in migration rate was observed after CAP treatment. The decrease of cell migration rates for all three conditions is 30-40% of the optimal cell migration rate. These data suggest that CAP treatment activates integrins on cell surfaces.
Plating cells onto surfaces coated with integrin ligands activates integrins engaged by those ligands. β1 family integrins bind to FN and CN, whereas αv-family integrins bind to VN. Next we assessed whether plating cells onto FN/CNI-or VN-coated surfaces impacts on their ability to respond to CAP. In figure 3(B), cell migration rates and persistence of migration for fibroblasts plated on un-coated surfaces and surfaces pre-coated with FN/CNI or VN are shown for untreated and for CAP (100 s)-treated cells. Data are shown for only the first 6 h of cell tracking, i.e. right after cells adhered to the surface. There are no significant differences in cell migration rates between cells plated onto uncoated compared with FN/CNI-and VN-coated surfaces for the control cells (p-value > 0.05). In addition, pre-coating surfaces with FN/CNI or VN did not alter their response to plasma treatment and a drop of about 30% in cell migration rate was seen for the three conditions studied. Statistical analyses of these data showed that there greater statistical significance (p-value < 0.001) for fibroblasts plated on the un-coated surface rather than on the surface pre-coated with FN/CNI or VN (p-value < 0.01). No significant changes were found in the persistence of the cells' motion. Note also that helium alone does not effect the cell migration, which was shown earlier [10,11].

β1 integrins are activated by CAP
We have shown so far that integrin activation by MnCl 2 and starvation attenuates the response of cells to CAP but adhering cells to specific integrin ligands had no impact on cells' ability to respond to CAP. To look directly at the impact of CAP on integrin activation, control cells and cells treated with CAP were fixed and used for immunofluorescence studies without permeabilization to assess the ratio of activated to total β1 integrin on their surface. The method requires the use of an antibody that recognizes total β1 integrin regardless of conformation (Ha2/5) and another that recognizes only the active conformation of the β1 integrin (9EG7). Immunofluorescent images are acquired and subjected to quantitative image analysis as described in section 2. Figure 4 shows the results of immunofluorescence studies to assess the activation state of the β1 integrin in the CAP-treated fibroblasts.  The data represent the ratio of 9EG7 and total β1 integrin for the peripheral part of the cells; error bars stand for the standard error of mean, ∼20 cells were analyzed per conditional change. The double asterisk shows the statistical significance (p-value < 0.001) in the increase of the intensity of the 9EG7 over the total β1 in the cells treated with MnCl 2 , plasma or both.

CAP-treated cells have larger focal adhesions
Activated integrins accumulate within focal adhesions, increasing their size and reducing cell migration rates. To determine whether focal adhesions of CAP-treated cells were indeed larger, control cells and cells treated with CAP were fixed, permeabilized and used for immunofluorescence to detect the presence of vinculin as well as αv integrin within focal adhesions; the data are presented in figure 5(A) and quantified in figure 5(B). Vinculin is a focal adhesion protein [1,3] and the amount of protein engaged in the focal adhesion is proportional to the size of the focal adhesion [18]; thus the intensity of the fluorescence antibody will be proportional to the size of the focal adhesion. CAP-treated cells, cells treated with MnCl 2 only or cells treated with both CAP and MnCl 2 show an overall increase in vinculin intensity and thus have larger focal adhesions [18], whereas a statistically significant (p-value< 0.001) increase of ∼20% is seen for vinculin intensity at the periphery of cells treated with MnCl 2 , CAP or both. However, no significant changes were found in the expression of the total αv integrin ∼3 h after treatments.

Discussion
The results shown here demonstrate that CAP activates β1 integrin, decreases cell migration and increases focal adhesion size. The ability of CAP to alter cell migration rates is not specific to fibroblasts but extends to human corneal epithelial cells; both types of cell reduce their cell migration rates by at most 30-40% after CAP treatment. The extent of reduction in cell migration rate after CAP in fibroblasts and epithelial cells was identical to the reduction in cell migration achieved by activating integrins using MnCl 2 or by serum-starving the cells.
In both cell types, CAP decreases cell migration as a function of the length of plasma treatment time up to a threshold above which no further change in cell migration is seen. That threshold is reached at 40 s for fibroblasts and at 100 s for epithelial cells. The difference in threshold between cell types suggests that epithelial cell adhesion is mediated by integrins that are more adhesive and more difficult to activate. Epithelial cells but not fibroblasts express α6β4 integrin, which acts like a cellular glue to mediate the tight adhesions needed between the epidermis and underlying dermis [12]. It is likely that the increased adhesion seen in epithelial cells underlies the higher threshold seen in the length of CAP treatment required to reduce epithelial cell migration by 30-40%. It is important to note that CAP treatment of about 40 s leads to discrimination between fibroblast and epithelial cells, thus allowing differential treatment of various cells present in tissue.
MnCl 2 treatment is known to activate integrins and to reduce cell migration rates; here we show that MnCl 2 alone reduces cell migration rates to the same extent as plasma treatment. The fact that the migration rates of plasma-treated cells and those of plasma-treated cells subsequently treated with 0.5 mM MnCl 2 are similar links plasma treatment to integrin activation. If plasma treatment did not activate integrins, incubating cells with MnCl 2 would further reduce cell migration rates. One caveat to that interpretation of the data is that there may be a lower threshold to cell migration rates below which cell death occurs.
Serum starvation induces cell flattening and activates integrins via mechanotransduction of stress forces to the cells' focal adhesions. The fact that plasma treatment does not alter the migration rates of serum-starved cells can be interpreted in numerous ways. It is possible that the reduction in cell migration after plasma treatment requires nutrients lacking in medium with

12
1% serum; this is unlikely since rapid cell migration demands energy and nutrients. It is also possible that cell migration in 1% medium is reduced to a level that cannot be reduced further without loss of cell viability. Although cell migration rates did not change after CAP treatment of serum-starved cells, it is remarkable that the most observed starved cells did not show an increase in cell death in response to CAP treatment (the analyzing software gives information about the slopes of the velocity change and by the positive slopes we can confirm that ∼90% of the observed cells were alive; data not shown).
Integrins are expressed on cell surfaces as αβ heterodimers. There are two different classes of integrins on fibroblasts: those that adhere primarily to FN and CNs and are primarily β1-containing integrins (α1β1, α2β1, α3β1, α5β1, α7β1, α8β1, α9β1 and α11β1) and those that adhere to VN, which include the αv-containing integrins (αvβ3 and αvβ5) [19]. Epithelial cells are more complex, having an additional integrin heterodimer (α6β4) to mediate attachment to laminin [12,19].
Tissue culture plastic is charged. When cells are placed in dishes with medium containing 5% serum, low concentrations of glycoproteins with charged sugar moieties on their surface that are present in the serum stick to the uncoated tissue culture plastic. FN and VN, two integrin ligands, are among the proteins present in serum that stick to tissue culture plastic. This yields a low concentration of those ligands on uncoated tissue culture plastic and supports cell adhesion and migration by both β1 and αv integrins. When cells adhere to high concentrations of FN-CN type I (FN/CNI), β1 integrins cluster, which induces their activation; adhering cells to VN clusters and induces activation of αv integrins. When cells migrate on FN/CNI-or VN-coated surfaces, they migrate using primarily β1or αv-integrins, respectively.
The purpose of the replating studies was to determine whether cells using primarily β1 integrins to migrate respond to plasma treatment differently than cells using primarily αv integrins or cells using both β1 and αv integrins on uncoated plastic. If plasma treatment acts by increasing β1 integrin activity, then plating cells on FN/CNI would reduce or eliminate the impact of plasma treatment on cell migration rate because β1 integrins would be active prior to plasma treatment; in addition, if plasma treatment acts by increasing β1 integrin activity, plating cells on VN would enhance the plasma effect. Putting cells on VN reduces the cells' surface ligated and activated β1 integrin compared with cells on uncoated plastic. There would be less active β1 integrin present on cells on VN than on cells on uncoated plastic and therefore there would be an increase in the ability of plasma to alter cell migration if plasma operates by increasing β1 integrin activation alone. As shown in figure 3, there is a reduction in the ability of plasma to alter cell migration when cells are plated on FN/CNI compared to uncoated plastic and VN-coating dishes. These results show that plasma treatment operates primarily through β1 integrins but that αv integrins are also involved because plating cells on VN did not increase the ability of plasma treatment to reduce cell migration.
The data obtained for β1 integrin activation using immunofluorescence (see figure 4) confirms that MnCl 2 treatment, as expected, activates β1 integrin but the data were only statistically significant for β1 integrin at the cell periphery where the ratio of active to total integrin increased from 1.1 to 1.4; we did not see an increase in β1 integrin activation at the cell center. The failure to demonstrate β1 integrin activation at the center of cells after MnCl 2 likely reflects the limitations of 9EG7, which binds to its activation-state epitope more readily when the integrin is both ligand bound and active [20]. Because we did not see integrin activation at the cell center of control cells treated with MnCl 2 , we focused our assessment on whether β1 integrin was activated by plasma treatment on changes seen at the cell periphery. The data 13 presented in figure 4 show that without plasma treatment the ratio of active to inactive is 1.1, whereas after plasma treatment it is 1.38, a value that is statistically significant; treating cells with plasma followed by MnCl 2 increases the ratio of active to inactive β1 integrin to 1.5. It is also worth noting that there is a statistically significant difference in the presence of the activated β1 integrin between cells treated with only plasma and only MnCl 2 and between only plasma and MnCl 2 and both. This interesting observation can be the subject of future study, since it is revealing about the different mechanisms involved in integrin activation. The data describing the cells treated by both plasma and MnCl 2 suggest that most likely these two processes run independently of each other and show a cumulative effect.
Integrin activation is one of the conditions required for the formation of focal adhesions [21,22]. The activation of integrins leads to an increase in the size of focal adhesions. The data presented in figures 5(A) and (B) show that plasma-treated cells and cells treated with MnCl 2 both show an increase in the size of the vinculin-positive focal adhesions measured in pixel intensities at the cell periphery of ∼20%. Treating cells with plasma followed by MnCl 2 further increased the size of the focal adhesions, but the increase was small and not mathematically significant. Ligated αv integrin also localizes to focal adhesions. After plasma treatment, focal adhesions increase in size, and activated β1 integrin accumulates within these larger focal adhesions and yet αv integrin does not increase within focal adhesions. If the ratio of αv integrin to vinculin is determined, it is clear that plasma and MnCl 2 treatments both reduce the localization of αv integrin within focal adhesions. Taken together with the cell migration rate data presented in figure 3(B) and the β1 integrin activation data in figure 4, these data show that plasma treatment increases β1 integrin activation, leading to increased accumulation of β1 integrin and reduced accumulation of αv integrin within focal adhesions. These events increase focal adhesion size. The increased size of focal adhesions resists disassembly during cell migration and impedes cell migration rates. Interestingly, the migration rate reduction we see by plasma and MnCl 2 is about ∼30-40%; the low migration rate value seen after these treatments may represent some universal lower threshold for cell migration rate. Besides, the presented data suggest that the decrease in cell migrations induced by plasma should be quite transient (∼2-3 h) compared to the total surface integrin change reported earlier (∼24 h) [10].
Let us discuss a possible physical mechanism leading to integrin activation and thus decrease in cell motility. We propose a mechanism based on the thermodynamic model of mechanotransduction developed by Shemesh et al [23]. This model is based on the thermodynamic argument that the pulling force leads to self-assembly of molecules into an aggregate, increasing the number of focal adhesions in order to decrease the stresses induced on the cell membrane.
This paper considers cold plasma jet interaction with the cell layer. Several mechanisms such as reactive chemical species, charged species or charging can possibly effect cell migration [5][6][7]24]. In the following model, we have demonstrated that cell membrane charge change can affect the formation of the focal adhesions. The change in charge that is caused by the plasma at the cell might lead to conformational changes in the extracellular domain of the integrin ligands [25] and therefore integrin activation. While our model is based on the membrane charge change, there are several effects that CAP can lead to [7]. Consider the direct charge transfer from the plasma to the cell. Recall that while cells are covered by media, plasma jet action leads to the removal of the liquid above the treated cells. However, there are other mechanisms related to intracellular activity that are triggered by the CAP [7]. Charge that is changed at the cell can be calculated based on the floating potential argument [26] Q ∼ 10 4 · e (for Debye length L D ∼ 10 -4 m, cell radius ∼10 -5 m and e = 1.16 × 10 −19 C). The charge change will lead to the creation of repulsive forces and thus the elastic stress will be formed along the cell membrane. According to the thermodynamic argument the aggregation (i.e. focal adhesion centers formation) occurs if the chemical potential in the aggregate is smaller than without focal adhesions. One can estimate the force due to the formed focal adhesions as N A f A , where N A is the number of focal adhesions and f A is the resisting force of the anchor. Thus, the change in the thermodynamic potential can be expressed [23] as where µ(Q, N A ) is the change in the chemical potential and µ 0 is the change in the chemical potential in the absence of induced charge. It can be seen that an increase in the number of focal adhesions leads to a decrease in the chemical potential. Calculations based on this model are presented in figure 6. One can see that according to this model induction of the extra charge at the cell surface will lead to the formation of new focal adhesions (see figure 6). This can explain the decrease of cell motility after plasma treatment that is observed experimentally.

Conclusions
In this paper, we show that CAP treatment decreases fibroblast cell migration by increasing integrin activation. While we see an activation of β1 integrin by CAP in fibroblasts, both fibroblasts and epithelial cells reduce their migration rate by ∼30-40%. However, it takes longer plasma treatment times for epithelial cells to achieve maximum reduction of migration rate; thus differential treatment is possible. No significant changes in the persistence of the cell motion were found. A thermodynamic model is presented that predicts that CAP treatment leads to an increase in focal adhesion size, leading to a decrease in the cell motility in accordance with the experimental results obtained.