Genetically-encoded circuit for remote regulation of cell migration by

Edward S. Rogers Sr. Department of Electrical and Computer Engineering, University of Toronto, 10 King's College Circle, Toronto, Ontario, M5S 3G4, Ca...
2 downloads 4 Views 2MB Size
Download PDF
Research Article Cite This: ACS Synth. Biol. 2018, 7, 718−726

pubs.acs.org/synthbio

Genetically Encoded Circuit for Remote Regulation of Cell Migration by Magnetic Fields Abdullah A. Mosabbir† and Kevin Truong*,†,‡ †

Institute of Biomaterials and Biomedical Engineering, University of Toronto, 164 College Street, Toronto, Ontario M5S 3G9, Canada ‡ Edward S. Rogers Sr. Department of Electrical and Computer Engineering, University of Toronto, 10 King’s College Circle, Toronto, Ontario M5S 3G4, Canada S Supporting Information *

ABSTRACT: Magnetoreception can be generally defined as the ability to transduce the effects of a magnetic field into a cellular response. Magnetic stimulation at the cellular level is particularly attractive due to its ability for deep penetration and minimal invasiveness, allowing remote regulation of engineered biological processes. Previously, a magnetic-responsive genetic circuit was engineered using the transient receptor potential vanilloid 1 (TRPV1) and the iron containing ferritin protein (i.e., the TF circuit). In this study, we combined the TF circuit with a Ca2+ activated RhoA protein (CaRQ) to allow a magnetic field to remotely regulate cell migration. Cells expressing the TF circuit and CaRQ exhibited consistent dynamic protrusions, leading to migration along a porous membrane, directed spreading in response to a magnetic field gradient, as well as wound healing. This work offers a compelling interface for programmable electrical devices to control the migration of living systems for potential applications in cell-based therapy. KEYWORDS: TRPV1, calcium, cell migration, magnet, HEK293 TRPV1 (i.e., αGFP-TRPV1), while chimeric ferritin was the fusion of GFP, ferritin light chain and ferritin heavy chain (i.e., GFP-ferritin). Endogenous iron nanoparticles would form complexes with GFP-ferritin, which would then be anchored to αGFP-TRPV1 by the binding interface of αGFP to GFP. Notably the absence of this binding interface resulted in reduced magnetosensitivity.16 Magnetic stimulation was proposed to mechanically stimulate the assembled complex and cause a Ca2+ influx.15 This proposed mechanism by which Ca2+ influx occurs has been challenged on accounts of physical infeasibility according to theoretical calculations.17 Although the mechanism remains unclear, the phenomenon does appear reproducible as another group created an alternative to the TF circuit through a fusion protein of the TRPV4 channel with ferritin (called Magneto).18 This one-component system appeared to have a somewhat higher level of toxicity. In this paper, we have combined the TF circuit with another chimeric protein, namely a Ca2+ activated RhoA (hereafter CaRQ),19,20 thus allowing a magnetic field to turn on cell migration. TRPV1 has been implicated to produce diverse effects on cell migration, showing a positive,21,22 a negative,23,24 or a facilitative effect on cell migration after it has been initiated by stress, injury, or cell-specific growth factors.25−27 A deeper investigation of the role of TRPV1 shows that the diversity of downstream effects of TRPV1 in cell migration is primarily due

B

ehavioral studies in the past have indicated that organisms such as bacteria, bees, and birds can detect and respond to external magnetic fields.1−3 A recent discovery in C. elegans indicated for the first time the presence of an ion channel responsible for the detection of magnetic fields;4 however, no such magnetoreceptor has been demonstrated in mammals. Magnetic stimulation at the cellular level is particularly attractive due to its ability for deep penetration and minimal invasiveness, allowing for it to act as an ideal stimulator for remote regulation of engineered biological processes. One such process is cell migration, which is often engineered for the purpose of controlled tissue regeneration and directed cell therapy. Current methods of engineering magnetic cell migration in mammalian cells involve loading cells with paramagnetic nanoparticles.5−7 However, nanoparticle injection has limitations such as invasiveness, internalization kinetics, local uptake of cells, and toxicity.8 Thus, we have introduced a genetic based system of magnetosensitive cell migration in mammalian cells using a chimeric transient receptor potential vanilloid 1 (TRPV1) and a chimeric RhoA protein. The TRPV1 receptor is a Ca2+ channel that is best known for its role in nociception and sensory transmission.9,10 A variety of noxious stimuli can activate the receptor such as acid,11 heat,12 and chemicals such as capsaicin.13 Previously a magnetosensitive genetic circuit (hereafter, the TF circuit) was engineered involving a chimeric TRPV1 and a chimeric iron containing ferritin protein.14−16 Specifically, the chimeric TRPV1 was the fusion of a camelid anti-GFP nanobody with © 2018 American Chemical Society

Received: November 19, 2017 Published: January 17, 2018 718

DOI: 10.1021/acssynbio.7b00415 ACS Synth. Biol. 2018, 7, 718−726

Research Article

ACS Synthetic Biology

Figure 1. TF circuit generates a Ca2+ signal in a magnetic field. (A) Schematic diagram of the protein system including αGFP-TRPV1 (in blue), GFP-ferritin (green) and RCaMP (red). Ca2+ influx from the magnetic activation of αGFP-TRPV1 causes an increase in RCaMP fluorescence. (B− E) Fluorescent images taken of TF-RCaMP cells before (B) and after (C) stimulation with a magnetic field, as well as a control case of TF-RCaMP cells before (D) and after (E) images at the same time interval with no magnetic field. (F−I) Fluorescence intensity plots of TF-RCaMP cells in response to ATP (F), magnetic field (G), acid (H), and control media (I). (J) Bar graph quantifying the percent of TF-RCaMP cells that displayed a Ca2+ signal in response to stimuli in (F−I). (K−N) Fluorescence intensity plots of RCaMP cells in response to ATP (K), magnetic field (L), acid (M), and control media (N). (O) Bar graph quantifying the percent of RCaMP cells that displayed a Ca2+ signal in response to stimuli in (K−N). All images are in false color. Fluorescence intensity plot values are in arbitrary units. Asterisks on bar graphs indicate statistically significant values at p < 0.05. “NS” indicates nonsignificant values compared to control. All experiments were repeated n = 6 times. Scale bars represent 25 μm.

(HEK293) cells expressing the TF circuit produce a Ca2+ signal upon exposure to a magnetic field. HEK293 cells were chosen as a model because they have no notable ability for directed migration and thus, any engineered migration is unusual. Cells expressing the TF circuit and CaRQ showed consistent dynamic protrusions, migration along a porous membrane, directed spreading in response to a magnetic

to the specific context in which it activates (i.e., phosphorylation state, lipid environment, interacting proteins and concentration of relevant ligands).28 Therefore, the activation of TRPV1 alone is not sufficient to produce cell migration. To allow robust activation of cell migration via TRPV1, CaRQ was added to rewire the Ca2+ signal generated by TRPV1 to cell migration. In this paper, it was shown that human embryonic kidney 719

DOI: 10.1021/acssynbio.7b00415 ACS Synth. Biol. 2018, 7, 718−726

Research Article

ACS Synthetic Biology

Figure 2. TF circuit with CaRQ exhibits dynamic protrusion in a magnetic field. (A) Schematic diagram of the protein system including αGFPTRPV1 (in blue), GFP-ferritin (green) and CaRQ (cyan). (B−F) YFP images of TF-CaRQ cells exposed to a magnetic field. Each panel represents a time point along a time-lapse experiment where a dynamic protrusion can be seen (indicated by the arrow). (G−K) YFP images of TF-CaRQ cells shown in the absence of a magnetic field. Each panel represents a time point along a time-lapse experiment corresponding to the same time points as the test case. (L−O) Bar graphs quantifying the percentage of cells showing dynamic protrusion in response to ATP and a magnetic field. Data was collected for TF-CaRQ cells (L), TF-RCaMP cells (M), CaRQ cells (N), and Null cells (O). All images are in false color. Asterisks on bars indicate statistically significant values compared to bar values without an asterisk (p < 0.05). “NS” indicates nonsignificant values compared to control. Bars represent the mean. Error bars represent standard deviations above and below the mean. All experiments were repeated n = 6 times. Scale bar represents 10 μm.

influx in response to a magnetic field, a stable HEK293 cell line (hereafter TF-RCaMP cells) expressing αGFP-TRPV1, GFPferritin, and a red Ca2+ indicator RCaMP1.0729 was created by lentiviral infection using the TF-RCaMP transfer vector (Figure S1E). RCaMP was labeled to the plasma membrane (PM) to allow visualization of the boundaries between cells as well as Ca2+ signals near the PM. As the Ca2+ concentration increases, red fluorescence of RCaMP also increases. As control, a stable

gradient, as well as wound healing. We believe that this technology can be a step forward in engineering magnetically responsive cells for potential noninvasive therapeutic applications.

■

RESULTS AND DISCUSSION The TF Circuit Generates a Ca2+ Signal in a Magnetic Field. To test the ability of the TF circuit to produce a Ca2+ 720

DOI: 10.1021/acssynbio.7b00415 ACS Synth. Biol. 2018, 7, 718−726

Research Article

ACS Synthetic Biology

The TF Circuit with CaRQ Exhibits Dynamic Protrusions in a Magnetic Field. To cause dynamic protrusions in response to a magnetic field, a stable HEK293 cell line (hereafter TF-CaRQ cells) expressing αGFP-TRPV1, GFPferritin and CaRQ19 was created by lentiviral infection using the TF-CaRQ transfer vector (Figure S1E). The Ca2+ signal generated by the TF circuit in response to a magnetic field should in turn activate CaRQ-mediated dynamic protrusions (Figure 2A). A time-lapse experiment in the time scale of minutes showed that TF-CaRQ cells exhibited dynamic protrusions in the presence of a magnetic field (Figure 2B− F) (Video S2), whereas a control case (absence of magnetic field) did not show any response (Figure 2G−K). Dynamic protrusions are characterized as the rapid extension and retract of cell protrusions which is more easily observed from a timelapsed recording (Video S2). The electromagnet was centered 10 mm above the sample and when “on” caused a magnetic strength of 10 mT at the position of the sample (Figure S1A). During the experiment, there was no change of temperature at the position of the sample (Figure S1A). In the proposed mechanism for magnetic control, the GFP-ferritin should in theory label the plasma membrane, but instead, our images clearly show that GFP-ferritin is labeled in the cytoplasm (Figure 2B−K). Indeed, even the images of GFP-ferritin by the original authors of the TF circuit show a cytoplasmic distribution.16 To further control for this phenomenon, two additional stable HEK293 cell lines were created by lentiviral infection using their corresponding transfer vectors: CaRQ, and Null cells (Figure S1E). Null cells only expressed Cerulean fluorescent protein labeled to the PM (Figure S1E). As expected, both the magnetic field and ATP cause dynamic protrusions in TF-CaRQ cells (Figure 2L). Interestingly, 82.5 ± 19.25% of TF-CaRQ cells exhibited dynamic protrusions in the presence of a magnetic field, despite only 14.98 ± 4.75% of TFRCaMP cells displaying a visible Ca2+ signal. The previous characterization of CaRQ19 revealed that it induced dynamic protrusions with an IC50 at 27 μM of Ca2+. Since such high levels of Ca2+ are never attained in the cytoplasm, CaRQ was subsequently labeled to the plasma membrane where some CaRQ could be in close enough proximity to Ca2+ channels to attain such high levels of Ca2+ during channel opening. Thus, CaRQ was very sensitive to Ca2+ increases at plasma membrane that may not even be detectable. In our previous papers, lighted-gated channels such as channelrhodopsin-2 and LOVS1K31 generated undetectable Ca2+ signals at the membrane when stimulated with light, but when they are paired with CaRQ, a single pulse of light caused the cells to create membrane protrusions.19,32 Similarly, while we cannot measure global Ca2+ signals in many cells expressing the TF circuit caused by magnetic stimulation, magnetic stimulation does cause sufficient increases in Ca2+ at the membrane to activate CaRQ robustly. When testing CaRQ cells, ATP caused dynamic protrusions, whereas a magnetic field did not (Figure 2N). This is an expected result as ATP induces a Ca2+ signal while the magnetic field, in the absence of the TF circuit does not. When testing TF-RCaMP cells, neither a magnetic field nor ATP caused dynamic protrusions (Figure 2M). This is consistent with the literature, as TRPV1 has been reported to be unable to initiate migration.25−27 Null cells, as expected, shown no signs of dynamic protrusion with all stimuli (Figure 2O). CaRQ-mediated dynamic protrusions can be blocked by adding the Y-27632 chemical which inhibits the Rho/ROCK pathway or by removing the extracellular Ca2+ which prevents

HEK293 cell line expressing only RCaMP was created by lentiviral infection with the RCaMP transfer vector (Figure S1E). Three stimuli and a control were used for this experiment: adenosine triphosphate (ATP), acidic phosphate buffered saline (pH = 6.0), an oscillating magnetic field, and additional media as a negative control. ATP is known to induce Ca2+ signals through IP3-mediated release of Ca2+ from the endoplasmic reticulum (ER) and was used as a positive control.30 As expected, red fluorescence intensity plots showed that stimulation with ATP elicited a Ca2+ signal response in both TF-RCaMP and RCaMP cells (Figure 1F,K). The negative control case did not show any Ca2+ signal in both cases (Figure 1I,N). Acidic phosphate buffered saline showed a Ca2+ signal in TF-RCaMP cells but not in RCaMP cells (Figure 1H,M), which was expected since acid is a stimulant for TRPV1 and not for RCaMP. Similarly, exposure to a magnetic field also elicited a Ca2+ signal from TF-RCaMP cells (Video S1) and not RCaMP cells (Figure 1G,L; also depicted in A−D), which indicated that the Ca2+ signal produced by magnetic field is due to the presence of the TF circuit. The electromagnet was centered 10 mm above the sample and when “on” caused a magnetic strength of 10 mT at the position of the sample (Figure S1A). During the experiment, there was no change of temperature at the position of the sample (Figure S1A). Quantification of the number of cells exhibiting a Ca2+ signal for each stimulus supported this, as RCaMP cells were only activated by ATP whereas the TF-RCaMP cells showed activation from ATP, acid, and a magnetic field (Figure 1J,O). One important observation however was that only 14.98 ± 4.75% of TF-RCaMP cells exhibited Ca2+ signals in a magnetic field, although this was still significantly higher than controls or in RCaMP cells. The ability of the TF circuit to generate globally detectable Ca2+ signals was disappointingly low because it does not occur in most cells and when it does occur, it occurs sporadically within the observation period of 30 min. However, when we stop the magnetic field stimulation for another 30 min, we observe no Ca2+ transients during that time. When we restart the magnetic field again, we observe sporadic Ca2+ transients in the same manner. These sporadic globally detectable Ca2+ signals only happen when there is magnetic field stimulation. While the number of cells that displayed visible Ca2+ signals from a magnetic field stimuli seems low, it does not preclude the possibility that a small undetectable amount of Ca2+ did influx through TRPV1. If true, the TF circuit is still useful in Ca2+ rewiring to cell migration. Our studies with the LOVS1K system31 to generate nearly undetectable local Ca2+ signals at the membrane upon photostimulation showed it was sufficient to robustly activate CaRQ-mediated dynamic protrusions.32 Thus, we can potentially expect a more robust activation of CaRQ-mediated dynamic protrusion than global Ca2+ signal generation by the TF circuit. Lastly, the ability of the TF circuit to generate Ca2+ signals upon magnetic field stimulation was dependent on the presence of αGFP-TRPV1, GFP-Ferritin and holotransferrin media supplementation (Figure S1G). Holotransferrin binds to ubiquitous transferrin receptors to import iron into the cell. All cell lines that included only a partial combination of the three components were unable to generate global Ca2+ signals in response to magnetic field stimulation. The magnetic field stimulation only generates global Ca2+ signals in the case where cells express the full TF circuit in the presence of holotransferrin media supplementation. Holotransferrin was supplemented in all experiments hereafter. 721

DOI: 10.1021/acssynbio.7b00415 ACS Synth. Biol. 2018, 7, 718−726

Research Article

ACS Synthetic Biology

Figure 3. TF-CaRQ cells exhibit wound healing in a magnetic field. (A−F) Bright field images of before and after stimulation overnight. Images include TF-CaRQ cells before (A) and after (B) overnight incubation with no magnetic field (control), before (C) and after (D) overnight incubation with a magnetic field, as well as before (E) and after (F) overnight incubation in ATP. (F−I) Bar graph quantifying the percent wound closure of all wounds in the conditions with or without ATP or a magnetic field. Percent wound closure of TF-CaRQ cells (F), TF-RCaMP cells (G), CaRQ cells (H) or Null cells (I). All images are in false color. Asterisks on bar graph indicate statistically significant values compares to values without an asterisk (p < 0.05). “NS” indicates nonsignificant values compared to control. Bars represent the mean. Error bars represent standard deviations above and below the mean. All experiments were repeated n = 6 times. The scale bar represents 150 μm.

any Ca2+ entering into the cell from the extracellular medium. Both of these conditions inhibit magnetic field-induced dynamic protrusions (Figure S1H). The TF-CaRQ Cells Exhibit Wound Healing in a Magnetic Field. TF-CaRQ cells showed significant wound closure within 24 h in response to a magnetic field and ATP compared to controls (Figure 3A−F). In a wound healing assay, both cell growth and migration contribute to wound closure. Without any stimuli, TF-CaRQ cells closed 18.3 ± 7.6% of the wound distance through growth alone (Figure 3I). With stimulation by a magnet wound closure rises to 49.7 ± 13.4%, which corresponds to 345.8 ± 89 μm traveled across the wound. The electromagnet was centered 3 mm below the sample and when “on” caused a magnetic strength of 27 mT at the position of the sample (Figure S1B). During the experiment, there was no change of temperature at the position of the sample (Figure S1B). The wound closure of TF-CaRQ cells was due to cell migration as the growth rate of TF-CaRQ cells with and without a magnetic field did not differ between the groups (Figure S1F). As further controls, Null cells, CaRQ cells and TF-RCaMP cells were tested using the same wound healing assay. As expected, Null cells did not show significant wound closure greater than controls, while CaRQ cells only showed wound closure in response to ATP (Figure 3H,I). The TF-RCaMP cells did not show significant wound closure in any case as well, which is supportive of the previous finding that the dynamic protrusions was absent in TF-RCaMP cells (Figure 3G). To determine the minimum magnetic field strength required to induce wound closure greater than control (i.e., no magnetic stimulation), the magnetic strength (when “on”) of the wound assay was varied between 0 mT to 27 mT (Figure S1B, I). This was attained by vary the distance to the sample by stacking microscope slides (∼1 mm in height) between the

electromagnet and the sample. At around 9 mT, the wound closure resembled the control without a magnetic stimulation (Figure S1I). Hence, the TF circuit with CaRQ can serve as a first step in promoting tissue repair using a genetically encoded system in order to facilitate tissue repair and regeneration. Being a genetically encoded system offers the benefit of taking advantage of various gene delivery systems to impart this ability into real tissues temporarily (or permanently),33 in order to facilitate faster regeneration via a noninvasive magnetic field. TF-CaRQ Cells Exhibit Transwell Migration and Directed Cell Spreading in a Magnetic Field. The next step was to explore the possibility of TF-CaRQ cells inducing spatial displacement and directionality in migration. Single cells were seeded onto the apical chamber of a Boyden transwell and allowed to migrate across a porous membrane in the presence or absence of a magnetic field. TF-CaRQ cells showed a significant amount of migration across the membrane in the presence of a magnetic field as compared to without one (Figure 4A). The electromagnet was centered 6 mm below the sample and when “on” caused a magnetic strength of 15 mT at the position of the sample (Figure S1C). During the experiment, there was no change of temperature at the position of the sample (Figure S1C). Furthermore, neither the Null cells, CaRQ cells or TF-RCaMP cells showed significant amounts of migration in a magnetic field which is consistent with the previous controls. Next, the ability of TF-CaRQ cells to spread in the directionality of a magnetic field was tested. If the source of the magnetic field is shifted away from the TFCaRQ cells, it should generate a magnetic field gradient that influences the directional spread of the cells. A standard rose diagram was used to represent the experimental findings of directional cell migration. Each cell colony was divided into radial sections to be quantified before and after a 24-h 722

DOI: 10.1021/acssynbio.7b00415 ACS Synth. Biol. 2018, 7, 718−726

Research Article

ACS Synthetic Biology

Figure 4. TF-CaRQ cells exhibit transwell migration and directed cell spreading in a magnetic field, (A) A diagram of the Boyden chamber assay as well as a bar graph quantifying the migration index of cells. (B) Diagram of a circular colony of cells under a magnetic field. TF-CaRQ cells are shown before (C) and after (D) the test case. (E) Rose diagram showing the percentage change in cell number within a given spatial section for the test case. (F) Diagram of a circular colony of cells not under a magnetic field. TF-CaRQ cells are shown before (G) and after (H) the control case. (I) Rose diagram showing the percentage change in cell number within a given spatial section for the control case. The red plus sign indicates the center of the main colony. Clusters of cells that break off are not considered part of the main colony. The p-value is indicated to show a significance in the increase in cell number in one particular direction as opposed to radial growth. Asterisks on bar graphs indicate statistically significant values compares to values without an asterisk (p < 0.05). “NS” indicates nonsignificant values compared to controls. Bars represent the mean. Error bars represent standard deviations above and below the mean. All experiments were repeated n = 6 times.

“on” caused a magnetic strength of 20 mT at the position of the sample (Figure S1D). During the experiment, there was no change of temperature at the position of the sample (Figure S1D). Given that a magnetic field does not induce faster growth (Figure S1F), the results indicate that a magnetic field gradient can be employed to influence the direction of migration. The significance in influencing the direction of migration lies in the potential to control tissue regeneration in a directed manner for more control or in cases where cell homing to a specific site of disease is important, such as in cell therapy. The deeper penetration of a magnetic field facilitates this advantage in live organisms as well.

experimental period. By quantifying the percentage change of cells within each radial section, a measurable quantitation of directionality could be obtained. For cells not under a magnetic field (Figure 4F), it was observed that growth occurred radially, and so there was no significant direction of growth or migration (Figure 4G,H). Quantitation of this phenomenon showed a similar amount of percent increase in cells radially, which was also not significant toward a particular direction (Figure 4I). In a magnetic field gradient however (Figure 4B), cell colonies were oriented toward the direction of the source of the magnetic field in a significant manner (Figure 4C,D). In these colonies, clusters of cells can break off from the main colony in the direction of the magnetic field (Figure 4D). Quantitation of this phenomenon showed a significant amount of increase in cells toward the side nearer to the magnetic field source (Figure 4E; source direction indicated in Figure 4B). The electromagnet was oft-centered 3 mm below the sample and when

■

CONCLUSION Magnetic stimulation is a promising modality to remotely control biological processes due to its ability for deep 723

DOI: 10.1021/acssynbio.7b00415 ACS Synth. Biol. 2018, 7, 718−726

Research Article

ACS Synthetic Biology

electromagnet draws 200 mA when on and undetectable currents when off. The magnetic field strength was measured at particular positions in space using the Lawkiaa WT10A magnetometer. The temperature was measured at particular positions in space using a thermocouple amplifier MAX31855 breakout board with a K-type thermocouple. Magnetic Field Stimulation. The Arduino controlled the electromagnet to be 5 s on and 2 min off. This was repeated for 30 min in the microscope experiments to measure Ca2+ signals and dynamic protrusions. This was repeated overnight in the incubator for the transwell, wound healing and displacement assays. Ca2+ Imaging and Dynamic Protrusions. Cells plated in 10 mm culture dishes (Fisher Scientific, Pittsburgh, PA) and were imaged in Dulbecco’s Modified Eagle’s Medium with 10% supplemented Fetal Bovine Serum (FBS) (Sigma-Aldrich, St. Lois, MO). An electromagnet was placed above the 10 mm dish on top of the lid directly over the inner well. Images were collected once every 2 s. For quantification of Ca2+ images, the number of cells that exhibited an increase in fluorescence was counted. Fluorescence intensity plots were generated using the Live Intensity Plot plugin on the ImageJ open source software. The duration of the signal was measured for 30 min and quantified as ΔF/F0. For quantification of dynamic protrusions, cells that showed membrane protrusions over the observation period were counted. The presence of dynamic protrusions was counted as true if cell extended and retracted greater than 15 times within the observation period of 30 min after the addition of the stimulus. Wound Healing. The basic wound healing protocol was taken from a previously published work.38 A confluent monolayer of cells was grown in a 96-well tissue culture plate (Fisher Scientific, Pittsburgh, PA). A 100 uL pipet tip was used to scratch a wound across the monolayer and imaged immediately afterward and also again 24 h later. Boyden Chamber Assay. A standard 24-well chamber assay was used to assay for cell migration across a porous membrane. Cells were seeded onto the transwells and were imaged 24 h later. For magnet stimulation, an electromagnet was placed underneath the 24-well plate that housed the transwells. For the migration index, cells were counted first and an equal amount of cells were passaged into the trans-well (cell density: 10 000 cells/mL). Fluorescent TF-CaRQ cells or Null controls were counted the following day. The amount of migration equals the percentage of cells that traveled across the membrane. The migration index is a normalization of the data, which takes the amount of migration in all cases and divides that by the amount of migration in the control case (i.e., Null cells without magnet). Thus, the control has a value of 1, and all other values are relative to the control. Rose Diagram. Cell colonies were allowed to grow overnight under a magnetic field. Images were taken before and 24 h after. The rose diagram was quantified by collecting the images of colonies, dividing it into 12 equal portions and counting the number of cells in each section. Each section count was then converted as the percentage of total cells within that section. A subtraction method was used to subtract the percentage of cells from the prestimulation period from the poststimulation period in order to account for growth. The final data was then represented in a radial graph, showing the net displacement of cells within a colony expressed as a percentage of the total number of cells.

penetration and minimal invasiveness. Although the discovery of a natural magnetosensitive receptor in mammals is still elusive, cells can be engineered to be responsive to magnetic fields using synthetic systems like the TF circuit or Magneto. Previously, these systems have been used as a magnetosensitive switches for controlling neuronal activity or turning on gene expression.15,16,18 In this study, by coupling the TF circuit with CaRQ, cell migration was rewired via Ca2+ to be responsive to a magnetic field in wound-healing, transwell and cell spread assays. Furthermore, the TF circuit could be coupled with other natural or synthetic Ca2+ activated proteins to control a variety of cellular processes using a magnetic field. This work could find applications in tissue repair as cells at a wound could be temporarily infected with nonintegrating lentivirus or adenovirus carrying the circuit and subsequently drive wound closure with a magnetic field. Lastly, the spatiotemporal precision of producing magnetic fields from current carrying wires may offer a compelling interface for programmable electrical devices to control the migration of biological systems.

■

EXPERIMENTAL PROCEDURES Plasmid Construction. Using the base structure of our cassette plasmid,34 the transfer vectors for TF-RCaMP, TFCaRQ, Null, RCaMP, and CaRQ cells were synthesized by Genscript (Pescataway, NY) and subcloned into the pUC57Simple vector using EcoRV. Given the construct information provided,14−16 the αGFP-TRPV1 and GFP-ferritin were synthesized. pm-RCaMP or pm-Ceru were synthesized as the tandem fusion of signal sequence of Lyn kinase (1MGCIKSKGKDSA12) with RCaMP1.07 or mCerulean. All genes were highly expressed with the CMVp (cytomegalovirus promoter) in mammalian cells. All plasmid manipulations were performed by Genscript. All plasmids were transformed in E. coli DH5-α and isolated using the Mini-prep kit (Invitrogen). Cell Culture, Transfection and Preparation. Dulbecco’s Modified Eagle’s Medium with 10% supplemented Fetal Bovine Serum (FBS) (Sigma-Aldrich, St. Lois, MO) was used to culture HEK293 cells in T25 flasks (37 °C and 5% CO2). Cells were transiently transfected using Lipofectamine 3000 according to standard protocols (Invitrogen). Using a previously published lentiviral method,35 stable cells were selected using blasticidin (10 μg/mL) or zeocin (200 μg/mL) for 2 weeks. A single cell was selected by limited serial dilution and grown into a population of cells to allow for better control over genetic differences between cells. Cell medium was replaced and holotransferrin (2 mg/mL, Sigma) was added to cells. These cells were then studied 72 h afterward. Illumination and Imaging. An inverted IX81 microscope with Lambda DG4 xenon lamp source and QuantEM 512SC CCD camera with a 10× or 40× objective (Olympus) was used for imaging. Filter excitation (EX) and emission (EM) bandpass specifications were as follows (in nm): CFP (EX: 438/24, EM: 482/32), YFP (EX: 500/24, EM: 542/27), RFP (EX: 580/20, EM: 630/60) (Semrock). Image acquisition and analysis was done with μManager and ImageJ software, respectively.36,37 Magnetic Field and Temperature Measurements. The magnetic field was generated by the Grove electromagnet (Seeedstudio) which requires a power supply of 5 V and 400 mA. To avoid extra components, we powered the electromagnet using the Arduino board, which has a max current of 200 mA. Thus, the magnetic field strength is relatively weak. Using BEST DT-9205 M multimeter, we measured that the 724

DOI: 10.1021/acssynbio.7b00415 ACS Synth. Biol. 2018, 7, 718−726

Research Article

ACS Synthetic Biology

(8) Singh, N., Jenkins, G. J., Asadi, R., and Doak, S. H. (2010) Potential toxicity of superparamagnetic iron oxide nanoparticles (SPION). Nano Rev. 1, 5358. (9) Ho, K. W., Ward, N. J., and Calkins, D. J. (2012) TRPV1: a stress response protein in the central nervous system. Am. J. Neurodegener. Dis. 1, 1−14. (10) Immke, D. C., and Gavva, N. R. (2006) The TRPV1 receptor and nociception. Semin. Cell Dev. Biol. 17, 582−591. (11) Dhaka, A., Uzzell, V., Dubin, A. E., Mathur, J., Petrus, M., Bandell, M., and Patapoutian, A. (2009) TRPV1 is activated by both acidic and basic pH. J. Neurosci. 29, 153−158. (12) Grandl, J., Kim, S. E., Uzzell, V., Bursulaya, B., Petrus, M., Bandell, M., and Patapoutian, A. (2010) Temperature-induced opening of TRPV1 ion channel is stabilized by the pore domain. Nat. Neurosci. 13, 708−714. (13) Yang, F., Xiao, X., Cheng, W., Yang, W., Yu, P., Song, Z., YarovYarovoy, V., and Zheng, J. (2015) Structural mechanism underlying capsaicin binding and activation of the TRPV1 ion channel. Nat. Chem. Biol. 11, 518−524. (14) Stanley, S. A., Gagner, J. E., Damanpour, S., Yoshida, M., Dordick, J. S., and Friedman, J. M. (2012) Radio-wave heating of iron oxide nanoparticles can regulate plasma glucose in mice. Science 336, 604−608. (15) Stanley, S. A., Kelly, L., Latcha, K. N., Schmidt, S. F., Yu, X., Nectow, A. R., Sauer, J., Dyke, J. P., Dordick, J. S., and Friedman, J. M. (2016) Bidirectional electromagnetic control of the hypothalamus regulates feeding and metabolism. Nature 531, 647−650. (16) Stanley, S. A., Sauer, J., Kane, R. S., Dordick, J. S., and Friedman, J. M. (2015) Remote regulation of glucose homeostasis in mice using genetically encoded nanoparticles. Nat. Med. 21, 92−98. (17) Meister, M. (2016) Physical limits to magnetogenetics. eLife, DOI: 10.7554/eLife.17210. (18) Wheeler, M. A., Smith, C. J., Ottolini, M., Barker, B. S., Purohit, A. M., Grippo, R. M., Gaykema, R. P., Spano, A. J., Beenhakker, M. P., Kucenas, S., Patel, M. K., Deppmann, C. D., and Guler, A. D. (2016) Genetically targeted magnetic control of the nervous system. Nat. Neurosci. 19, 756−761. (19) Mills, E., and Truong, K. (2011) Ca2+-mediated synthetic biosystems offer protein design versatility, signal specificity, and pathway rewiring. Chem. Biol. 18, 1611−1619. (20) Mills, E., Pham, E., and Truong, K. (2010) Structure based design of a Ca2+-sensitive RhoA protein that controls cell morphology. Cell Calcium 48, 195−201. (21) Howe, A. K. (2011) Cross-talk between calcium and protein kinase A in the regulation of cell migration. Curr. Opin. Cell Biol. 23, 554−561. (22) Martin, E., Dahan, D., Cardouat, G., Gillibert-Duplantier, J., Marthan, R., Savineau, J. P., and Ducret, T. (2012) Involvement of TRPV1 and TRPV4 channels in migration of rat pulmonary arterial smooth muscle cells. Pfluegers Arch. 464, 261−272. (23) Caprodossi, S., Amantini, C., Nabissi, M., Morelli, M. B., Farfariello, V., Santoni, M., Gismondi, A., and Santoni, G. (2011) Capsaicin promotes a more aggressive gene expression phenotype and invasiveness in null-TRPV1 urothelial cancer cells. Carcinogenesis 32, 686−694. (24) Goswami, C., Schmidt, H., and Hucho, F. (2007) TRPV1 at nerve endings regulates growth cone morphology and movement through cytoskeleton reorganization. FEBS J. 274, 760−772. (25) Ho, K. W., Lambert, W. S., and Calkins, D. J. (2014) Activation of the TRPV1 cation channel contributes to stress-induced astrocyte migration. Glia 62, 1435−1451. (26) Nakanishi, K., Fujimoto, J., Ueki, T., Kishimoto, K., HashimotoTamaoki, T., Furuyama, J., Itoh, T., Sasaki, Y., and Okamoto, E. (1999) Hepatocyte growth factor promotes migration of human hepatocellular carcinoma via phosphatidylinositol 3-kinase. Clin. Exp. Metastasis 17, 507−514. (27) Waning, J., Vriens, J., Owsianik, G., Stuwe, L., Mally, S., Fabian, A., Frippiat, C., Nilius, B., and Schwab, A. (2007) A novel function of

Statistical Analysis. All data with normal distribution and similar variance were analyzed with one-factor ANOVA with Tukey−Kramer posthoc test. For all tests, α was set at 0.01. P values smaller than 0.01 were considered significant and are indicated in text. The data is expressed as mean ± s.d. unless otherwise stated.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acssynbio.7b00415. Figure S1 (PDF) Video S1 (AVI) Video S2 (AVI)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: 416-978-7772. Fax: 416-978-4317. ORCID

Kevin Truong: 0000-0002-9520-2144 Author Contributions

K.T. developed the main concepts behind this paper. K.T. and A.M. planned all experiments. A.M. performed all experiments. A.M. wrote the first draft of the manuscript. K.T. and A.M. edited the manuscript. K.T. supplied all materials needed for experiments through his funding. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was funded by grants from the Canadian Cancer Society Research Institute (#701936) and NSERC (#0532214).

■

REFERENCES

(1) Hsu, C. Y., Ko, F. Y., Li, C. W., Fann, K., and Lue, J. T. (2007) Magnetoreception system in honeybees (Apis mellifera). PLoS One 2, e395. (2) Liedvogel, M., Maeda, K., Henbest, K., Schleicher, E., Simon, T., Timmel, C. R., Hore, P. J., and Mouritsen, H. (2007) Chemical magnetoreception: bird cryptochrome 1a is excited by blue light and forms long-lived radical-pairs. PLoS One 2, e1106. (3) Shaw, J., Boyd, A., House, M., Woodward, R., Mathes, F., Cowin, G., Saunders, M., and Baer, B. (2015) Magnetic particle-mediated magnetoreception. J. R. Soc., Interface 12, 0499. (4) Vidal-Gadea, A., Ward, K., Beron, C., Ghorashian, N., Gokce, S., Russell, J., Truong, N., Parikh, A., Gadea, O., Ben-Yakar, A., and Pierce-Shimomura, J. (2015) Magnetosensitive neurons mediate geomagnetic orientation in Caenorhabditis elegans. eLife, DOI: 10.7554/eLife.07493. (5) Alon, N., Havdala, T., Skaat, H., Baranes, K., Marcus, M., Levy, I., Margel, S., Sharoni, A., and Shefi, O. (2015) Magnetic micro-device for manipulating PC12 cell migration and organization. Lab Chip 15, 2030−2036. (6) Raffa, V., Vittorio, O., Ciofani, G., Pensabene, V., and Cuschieri, A. (2010) Cell creeping and controlled migration by magnetic carbon nanotubes. Nanoscale Res. Lett. 5, 257−262. (7) Xia, B., Huang, L., Zhu, L., Liu, Z., Ma, T., Zhu, S., Huang, J., and Luo, Z. (2016) Manipulation of Schwann cell migration across the astrocyte boundary by polysialyltransferase-loaded superparamagnetic nanoparticles under magnetic field. Int. J. Nanomed. 11, 6727−6741. 725

DOI: 10.1021/acssynbio.7b00415 ACS Synth. Biol. 2018, 7, 718−726

Research Article

ACS Synthetic Biology capsaicin-sensitive TRPV1 channels: involvement in cell migration,. Cell Calcium 42, 17−25. (28) Fiorio Pla, A., and Gkika, D. (2013) Emerging role of TRP channels in cell migration: from tumor vascularization to metastasis. Front. Physiol. 4, 311. (29) Akerboom, J., Carreras Calderon, N., Tian, L., Wabnig, S., Prigge, M., Tolo, J., Gordus, A., Orger, M. B., Severi, K. E., Macklin, J. J., Patel, R., Pulver, S. R., Wardill, T. J., Fischer, E., Schuler, C., Chen, T. W., Sarkisyan, K. S., Marvin, J. S., Bargmann, C. I., Kim, D. S., Kugler, S., Lagnado, L., Hegemann, P., Gottschalk, A., Schreiter, E. R., and Looger, L. L. (2013) Genetically encoded calcium indicators for multi-color neural activity imaging and combination with optogenetics. Front. Mol. Neurosci. 6, 2. (30) Salter, M. W., and Hicks, J. L. (1995) ATP causes release of intracellular Ca2+ via the phospholipase C beta/IP3 pathway in astrocytes from the dorsal spinal cord. J. Neurosci. 15, 2961−2971. (31) Pham, E., Mills, E., and Truong, K. (2011) A synthetic photoactivated protein to generate local or global Ca(2+) signals,. Chem. Biol. 18, 880−890. (32) Mills, E., and Truong, K. (2013) Analysis and regulation of amoeboid-like cell motility using synthetic Ca(2+)-sensitive proteins. Cell Calcium 53, 231−240. (33) Kamimura, K., Suda, T., Zhang, G., and Liu, D. (2011) Advances in Gene Delivery Systems. Pharm. Med. 25, 293−306. (34) Truong, K., Khorchid, A., and Ikura, M. (2003) A fluorescent cassette-based strategy for engineering multiple domain fusion proteins,. BMC Biotechnol. 3, 8. (35) Mosabbir, A. A., and Truong, K. (2016) Genomic integration occurs in the packaging cell via unexported lentiviral precursors. Biotechnol. Lett. 38, 1715−1721. (36) Edelstein, A. D., Tsuchida, M. A., Amodaj, N., Pinkard, H., Vale, R. D., and Stuurman, N. (2014) Advanced methods of microscope control using muManager software. J. Biol. Methods 1, 10. (37) Schneider, L., Cammer, M., Lehman, J., Nielsen, S. K., Guerra, C. F., Veland, I. R., Stock, C., Hoffmann, E. K., Yoder, B. K., Schwab, A., Satir, P., and Christensen, S. T. (2010) Directional cell migration and chemotaxis in wound healing response to PDGF-AA are coordinated by the primary cilium in fibroblasts. Cell. Physiol. Biochem. 25, 279−292. (38) Liang, C. C., Park, A. Y., and Guan, J. L. (2007) In vitro scratch assay: a convenient and inexpensive method for analysis of cell migration in vitro. Nat. Protoc. 2, 329−333.

726

DOI: 10.1021/acssynbio.7b00415 ACS Synth. Biol. 2018, 7, 718−726