Patch-clamp recordings in single-cell expression systems have been traditionally used to study the function of ion channels. However, this experimental setting does not enable assessment of tissue-level function such as action potential (AP) conduction. Here we introduce a biosynthetic system that permits studies of both channel activity in single cells and electrical conduction in multicellular networks. We convert unexcitable somatic cells into an autonomous source of electrically excitable and conducting cells by stably expressing only three membrane channels. The specific roles that these expressed channels have on AP shape and conduction are revealed by different pharmacological and pacing protocols. Furthermore, we demonstrate that biosynthetic excitable cells and tissues can repair large conduction defects within primary 2- and 3-dimensional cardiac cell cultures. This approach enables novel studies of ion channel function in a reproducible tissue-level setting and may stimulate the development of new cell-based therapies for excitable tissue repair.
All cells express ion channels in their membranes, but cells with a significantly polarized membrane that can undergo a transient all-or-none membrane depolarization (action potential, AP) are classified as 'excitable cells'1. The coordinated function of ion channels in excitable cells governs the generation and propagation of APs, which enable fundamental life processes such as the rapid transfer of information in nerves2 and the synchronized pumping of the heart3. For this reason, genetic or acquired alterations in ion channel function or irreversible loss of excitable cells through injury or disease (for example, stroke or heart attack) are often life threatening4,5.
Numerous ion channels (wild type (wt) or mutated) have been studied in single-cell heterologous expression systems to investigate channel structure–function relationships and link specific channel mutations found in patients to associated diseases, such as cardiac arrhythmias or epilepsy6. Typically, the potential implications of these single-cell studies for the observed tissue- or organ-level function are only speculated, often through the use of tissue-specific computational models7,8. Similarly, experimental studies of AP conduction in primary excitable tissues and cell cultures are often limited by low reproducibility, heterogeneous structure and function, a diverse and often unknown complement of endogenous channels, and the non-specific action of applied pharmaceuticals. We therefore set out to develop and validate a simplified, well-defined and reproducible excitable tissue system that would enable direct quantitative studies of the roles that specific ion channels have in AP conduction.
Extensive electrophysiological research over the last century1,9 has revealed that the initiation, shape and transfer of APs in excitable cells are regulated by an extraordinarily diverse set of ion channels, pumps and exchangers. Yet, the classic Hodgkin and Huxley bioelectric model of a giant squid axon10 and other simplified models of biological excitable media11,12 suggest that only a few membrane channels are sufficient to sustain cellular excitability and AP conduction. On the basis of these theoretical concepts, we hypothesized that a small number of targeted genetic manipulations could transform unexcitable somatic cells into an electrically active tissue capable of generating and propagating APs.
In this study, we selected a minimum set of channel genes that, upon stable expression in unexcitable cells, would yield significant hyperpolarization of membrane potential, electrical induction of an all-or-none AP response and robust intercellular electrical coupling to support uniform and fast AP conduction over arbitrarily long distances. We designed experiments to thoroughly characterize the electrophysiological properties of these genetically engineered cells including pharmacological manipulations to establish the roles of each of the expressed channels in membrane excitability and impulse conduction. Furthermore, we explored whether these cells could be used to generate biosynthetic excitable tissues with the ability to restore electrical conduction within large cm-sized gaps in primary excitable cell cultures.
Experimental approach to engineering excitable cells
We tested the hypothesis that human unexcitable somatic cells can be genetically engineered to form an autonomous source of electrically excitable and conducting cells through the stable expression of three genes encoding: the inward-rectifier potassium channel (Kir2.1 or IRK1, gene KCNJ2)13, the pore-forming α-subunit of the fast voltage-gated cardiac sodium channel (Nav1.5 or hH1, gene SCN5A)14 and the connexin-43 gap junction (Cx43, gene GJA1)15. These three channels have critical roles in the generation and propagation of electrical activity in the mammalian heart16,17.
To facilitate the visual identification and monoclonal selection of stable cell lines, we constructed three bicistronic plasmids designed to express each channel (Kir2.1, Nav1.5 or Cx43) with a unique fluorescent reporter18 (mCherry, green fluorescent protein (GFP) or mOrange, respectively) and the puromycin resistance (PacR) gene (Supplementary Fig. S1). The human embryonic kidney 293 (HEK-293) cell line was utilized as a proof-of-concept unexcitable somatic cell source based on its low levels of endogenous membrane currents (for example, outwardly rectifying potassium currents19), uniform shape and growth, and extensive use as a heterologous expression system for studies of ion channel function20. Like most unexcitable cells, wt HEK-293 cells (Fig. 1a) exhibited a relatively depolarized resting membrane potential (RMP) of −24.4±0.8 mV (±s.e.m.; n=10 cells; Fig. 1b). In contrast, the RMP of excitable cells is highly negative due to the action of constitutively open and/or inwardly rectifying potassium channels such as Kir2.1 (ref. 21).
IK1 and INa expression enables AP generation
To induce significant membrane hyperpolarization in the wt HEK-293 cells, we stably transfected them with a plasmid encoding Kir2.1–IRES–mCherry and derived 'Kir2.1 HEK-293' monoclonal lines from cells that displayed bright mCherry fluorescence (Fig. 1c). These stable cell lines exhibited robust barium-sensitive inward-rectifier potassium current, IK1 (Supplementary Fig. S2 and Fig. 2a,b), and hyperpolarized RMP values (Fig. 2c) similar to those of primary excitable cells1. A hyperpolarized RMP is required for cellular excitability, as it allows a sufficient number of voltage-gated sodium channels to open on membrane depolarization and initiate a regenerative AP16,17. In addition, a hyperpolarized RMP, as a natural state of excitable cells, has been shown to distinctly affect the expression level and kinetics of ion channels22,23.
To generate electrically excitable cells, we introduced a voltage-gated sodium channel, Nav1.5, into a monoclonal Kir2.1 HEK-293 cell line by transfection of a plasmid encoding Nav1.5–IRES–GFP (Supplementary Fig. S1b). We then selected cells with strong GFP expression to derive monoclonal 'Kir2.1+Nav1.5 HEK-293' cells (Fig. 1e) that generated large, fast-inactivating sodium current, INa (Fig. 2d-g). This current showed cardiac-specific sensitivity to block by tetrodotoxin (TTX; Fig. 2f) and characteristic voltage-dependent steady-state activation and inactivation properties14 (Fig. 2h). Importantly, much like primary excitable cells, the HEK-293 cells genetically engineered to stably express both IK1 and INa developed membrane excitability, wherein on reaching an excitation threshold by current injection, they reproducibly fired an 'all-or-none' AP (Figs 1f and 2i).
To dissect the contributions of IK1 and INa to an AP time course in these cells, we utilized the AP-clamp technique24 (Fig. 2j). As expected from the IK1 current–voltage relationship (Fig. 2b), outward potassium current activated above the resting potential during the AP upstroke in an attempt to maintain membrane hyperpolarization and also during AP plateau negative to −25 mV to bring the cell back to rest. Inward INa activated only during the AP upstroke, followed by rapid inactivation (Fig. 2j).
Cx43 expression enables rapid AP propagation
We further tested whether the excitable Kir2.1+Nav1.5 HEK-293 cells can support active electrical conduction when cultured in confluent two-dimensional (2D) cell networks. Optical mapping of transmembrane voltage in these cultures, however, revealed either uncoordinated aperiodic electrical activity caused by slowly moving and colliding excitation waves25 or slow AP conduction (5.6±0.2 cm s−1 (±s.e.m.; n=4); Fig. 3a,b). This slow or unorganized AP propagation in the Kir2.1+Nav1.5 HEK-293 cells was supported by weak endogenous cell coupling (Supplementary Fig. S3a,b) that likely originated from expression of connexin-45 (refs 26, 27) rather than connexin-43 gap junctions (Fig. 3c, inset)27,28. Therefore, to enable rapid and uniform AP propagation in the Kir2.1+Nav1.5 HEK-293 cells, we stably transfected the monoclonal line characterized in Figure 2 with a plasmid encoding Cx43–IRES–mOrange (Supplementary Fig. S1c).
We derived 'Kir2.1+Nav1.5+Cx43 HEK-293' monoclonal cell lines with bright mOrange fluorescence (Fig. 1g) and abundant expression of Cx43 gap junctions at cell–cell interfaces (Fig. 3c). One of these excitable and well-coupled Kir2.1+Nav1.5+Cx43 HEK-293 monoclonal lines was named 'Excitable-293' (Ex-293). Functional intercellular coupling in Ex-293 cells was dramatically enhanced compared with endogenous HEK-293 coupling, as demonstrated by fluorescence recovery after photobleaching27,29 (Fig. 3d, Supplementary Fig. S3a,b and Supplementary Table S1) and dual whole-cell patch clamping30 (Supplementary Fig. S3c,d). The average gap junctional conductance in Ex-293 cell pairs (134.1±14.0 nS (±s.e.m.; n=12)) was comparable with that measured in isolated pairs of neonatal rat28 or adult canine31 cardiomyocytes. This Ex-293 monoclonal line, which has retained stable electrophysiological properties (Supplementary Table S2) after multiple freeze–thaw cycles and more than 40 cell passages, was further utilized for the generation and characterization of biosynthetic excitable 2D and three-dimensional (3D) cell networks.
When electrically stimulated, confluent 2D networks (monolayers) of randomly oriented Ex-293 cells supported rapid, spatially uniform AP propagation (Fig. 3e, Supplementary Fig. S4a and Supplementary Movie 1) at excitation rates of up to 26.5 Hz and basic conduction velocities (CVs) of 23.3 cm s−1. The dependence of CV in Ex-293 monolayers on the spatial distribution of gap junctions was explored by utilizing topographical and biochemical cues (microgrooved and micropatterned substrates32) to elongate and align Ex-293 cells within confluent anisotropic networks and thin strands (Supplementary Fig. S4b,c and Supplementary Movie 1). As expected, AP conduction was found to be faster along but slower across the aligned cells, as compared with non-aligned cells (longitudinal, transverse and isotropic velocity in Supplementary Table S2), due to the directional differences in the number of gap junctions (and cell–cell contacts) per unit length3.
Expressed channels govern Ex-293 electrical function
We further characterized how the expressed ion channels contribute to AP conduction in Ex-293 monolayers by using specific channel blockers. At the cellular level (assessed by sharp microelectrode recordings), gradual inhibition of IK1 by increasing doses of barium chloride (BaCl2) significantly increased AP duration (APD80, Fig. 4a, Supplementary Fig. S5a), depolarized RMP (Fig. 4b) and decreased the AP upstroke velocity (Supplementary Fig. S5b), confirming the important role of IK1 in AP repolarization, maintenance of RMP, and through the maintenance of the RMP, regulation of INa availability and AP upstroke. Gradual inhibition of INa by increasing doses of TTX, on the other hand, significantly decreased AP duration (APD80, Supplementary Fig. S5c), upstroke velocity (Fig. 4c and Supplementary Fig. S5d) and amplitude (Fig. 4d) without a significant change in RMP (not shown), thereby confirming the prominent role of INa in AP depolarization, and in conjunction with its opposing action on IK1, regulation of AP duration.
At the tissue level (assessed by macroscopic optical mapping of transmembrane potentials), inhibition of Kir2.1, Nav1.5 or Cx43 channels by increasing doses of BaCl2, TTX or palmitoleic acid (PA), respectively, yielded gradual slowing of propagation (Fig. 4e-g) until conduction failure, thus demonstrating the critical roles of all the three expressed channels in AP conduction. Similarly, reduction of INa by stimulation at progressively increasing rates33 yielded a decrease in both CV and APD80 (Fig. 4h). Furthermore, specific modulations of Nav1.5 channel activity by lidocaine or flecainide22,34 showed distinct effects on CV and APD80 in Ex-293 monolayers during steady 10-Hz stimulation (Supplementary Fig. S6a).
In addition to stimulated AP propagation, confluent Ex-293 monolayers supported self-sustained stationary (anchored to a small acellular obstacle, Fig. 5a–c) or freely drifting (Fig. 5d) spiral waves (Supplementary Movie 2), as is characteristic of various excitable media35 including cardiac tissue36. The average rotational rates of single anchored and freely drifting spirals were 13.8±1.1 Hz (±s.e.m.; n=5) and 18.8±1.6 Hz (±s.e.m.; n=4), respectively. The application of BaCl2, PA or different inhibitors of INa during the rotation of a stationary spiral wave slowed the rotation rate and eventually terminated all electrical activity in Ex-293 monolayers (Fig. 5c and Supplementary Fig. S6b).
Ex-293 cells form 3D biosynthetic excitable tissues
To demonstrate our ability to generate de novo functional excitable tissues, we cultured Ex-293 cells within fibrin-based hydrogel constructs37. Using elastomeric molds and Velcro anchors, we created cm-sized 3D tissue cords (Fig. 6a). The Ex-293 cells in tissue cords compacted the hydrogel, aligned along the direction of passive tension imposed by the two Velcro anchors (Fig. 6b,c), and electrically coupled to form 3D-engineered tissues that supported uniform AP propagation (Fig. 6d and Supplementary Movie 3), with an average CV and APD80 of 18.1±1.7 cm s−1 and 33.5±1.2 ms, respectively (±s.e.m.; n=7 tissue cords).
Ex-293 cells restore active conduction in cardiac cultures
We have previously demonstrated that unexcitable HEK-293 cells genetically engineered to express Cx43 can electrically couple with primary neonatal rat ventricular myocytes (NRVMs)27 and passively connect them (with a long delay) over distances of up to a few hundred microns32. Ex-293 cells, on the other hand, both formed Cx43 gap junctions with NRVMs (Supplementary Fig. S7a) and, through fast active propagation, seamlessly connected remote (cm apart) cardiomyocyte regions in both 2D (Supplementary Fig. S7b,c and Supplementary Movie 4) and 3D (Fig. 6e,f) NRVM networks. Thus, biosynthetic excitable cells and tissues could potentially act as actively conducting 'fillers' to restore fast and uniform AP propagation in diseased, slow-conducting excitable tissues.
This study introduces a methodology for the de novo generation of actively conducting excitable tissues from unexcitable somatic cells by the stable forced expression of only three membrane channels. Specifically, through a stepwise experimental approach (Fig. 1 and Supplementary Fig. S1), we showed that genetic engineering of: significant membrane polarization by introduction of Kir2.1 channels; rapid depolarization capacity by introduction of Nav1.5 channels; and robust intercellular coupling by overexpression of Cx43 gap junctions synergistically enabled electrical excitability and conduction in HEK-293 cells.
Previously, Hsu et al.38 transiently expressed voltage-gated sodium (Nav1.2) and potassium (Shaker H4) channels in CHO cells, but because the average RMP of these cells was −26.8 mV, APs could be induced only under electronically applied current-clamp hyperpolarization. More recently, Cho et al.39 have transiently expressed the Kir2.1 channel and a slow bacterial sodium channel (NaChBac) in HEK-293 cells to yield slow rising and long APs in 5/31 cells, whereas Abraham et al.40 have expressed Cx43 to promote electrical coupling of skeletal myotubes and decrease their arrhythmogenic effects in cocultures with cardiomyocytes. We, on the other hand, have created stable excitable cell lines with reproducible electrical properties including fast rising (dV/dtmax≈150 V s−1) and rapidly propagating (CV≈23 cm s−1) APs. As these cells can be assembled into autonomous, actively conducting tissues, we were able to pharmacologically dissect the specific roles of each expressed channel in the dynamics of AP conduction and demonstrate the ability of these biosynthetic tissues to restore fast conduction within arbitrary size gaps in cardiomyocyte cultures.
Single unexcitable cells have been traditionally used in heterologous expression systems in conjunction with patch-clamp recordings to isolate and study the function of specific ion channels41. The biosynthetic excitable cells described in this study can be additionally utilized as a tissue–scale heterologous expression system to directly investigate not only specific channel activity in single cells but also AP conduction in multicellular networks fabricated from these same cells. For example, in this study, we have isolated and quantified the role of Nav1.5 in AP generation and conduction during control conditions as well as different pharmacological challenges in both single cells and cell networks. In general, biosynthetic excitable tissues as a well-defined, reproducible and versatile platform could complement studies in primary excitable cells by helping elucidate the distinct roles that specific ion channels and gap junctions, their mutations42,43,44,45 auxiliary subunits46, or accessory proteins47,48 have in tissue-level electrophysiology. Furthermore, the potential to efficiently manipulate the gene expression and spatial organization of engineered cells within biosynthetic tissues can allow for well-controlled experiments to validate and advance existing theoretical and computational models of AP conduction as well as facilitate the design of robust and reproducible cell-based biosensors49.
Importantly, we also performed proof-of-concept in vitro studies showing that biosynthetic excitable cells and tissues can establish a fast and seamless electrical connection between distant cardiomyocytes. The implantation of stem cells or somatic cells with genetically engineered electrical properties has been proposed as a potential strategy to improve compromised electrical function in the heart50,51,52,53. For example, unexcitable cells with engineered potassium channels have been applied to locally modify cardiac conduction through coupling with cardiomyocytes54. However, because of their unexcitable nature, these cells would still fail to propagate over distances larger than a few hundred microns32,55. In contrast, the use of autonomously conducting biosynthetic cells with basic electrical properties tailored to mimic those of the host tissue could allow functional repair of arbitrary size conduction defects. For example, slow tortuous conduction at the site of myocardial infarct56 could be enhanced by implanting biosynthetic excitable cells or genetically modifying endogenous fibroblasts to establish direct paths of fast AP propagation. For potential clinical use, however, highly proliferative HEK-293 cells would have to be replaced by another contact-inhibited, terminally differentiated and readily accessible cell source.
Overall, we believe that our approach not only opens the door to novel studies of electrical conduction but also, through further integration of electrophysiology, synthetic biology and tissue engineering, the development of new treatments for excitable tissue disorders.
Cloning of rat connexin-43 gene (GJA1)
Total RNA was extracted and purified from isolated NRVMs (Qiagen, RNeasy Kit). Reverse transcription with oligo(dT) primers was performed to synthesize first-strand complementary DNA (Bioline, cDNA Synthesis Kit). The cDNA was used as the template in a standard PCR reaction. PCR primers were designed based upon the published coding sequence for rat connexin-43 (GenBank no. NM_012567). The amplified PCR product was verified by nucleotide sequencing and subcloned into the pCMV5(CuO)–IRES–mOrange expression plasmid.
Generation of engineered cell lines
The human KCNJ2, SCN5A and rat GJA1 cDNAs were subcloned into a bicistronic expression vector (pCMV5(CuO)–IRES–GFP, Qbiogene). The genes encoding mCherry and mOrange18 were used to replace GFP in the KCNJ2 and GJA1 plasmids, respectively. HEK-293 cells (ATCC, CRL-1573) were cultured, transfected and selected for stable expression using previously described methods27. The Ex-293 line was established by three successive single-plasmid (Kir2.1, Nav1.5 and then Cx43) transfections, with monoclonal isolation and expression analysis before each transfection.
Whole-cell membrane current recordings
HEK-293 cells were gently trypsinized and plated onto clean glass coverslips for recording. Patch electrodes were fabricated with tip resistances of 0.8–1.8 MΩ (for INa) or 1.5–3.5 MΩ (for IK1, AP, current or dual clamp) when filled with internal patch solutions. Whole-cell currents were recorded at room temperature using the Multiclamp 700B amplifier (Axon Instruments), filtered with a 10-kHz Bessel filter, digitized at 40 kHz and stored on a computer. Data was acquired and analyzed using WinWCP software (John Dempster, University of Strathclyde). The internal and external solutions and voltage-clamp protocols for recording of INa and IK1 were similar to other published studies33,57.
Whole-cell membrane voltage and AP voltage-clamp recordings
Cell intracellular potential was recorded at room temperature in current-clamp mode (no holding current) in response to 1-ms depolarizing current steps (from 0.5 nA to 2 nA in 0.1-nA increments) in an attempt to elicit an all-or-none AP response. Extracellular bath solution (Tyrode's) consisted of (mM) 135 NaC1, 5.4 KCl, 1.8 CaCl2, 1.0 MgCl2, 0.33 NaH2PO4, 5 HEPES and 5 glucose (pH 7.4). The membrane current during the course of AP was accessed using the AP-clamp technique24, in which a previously recorded AP waveform (measured in an excitable Ex-293 monolayer at 35 °C with a sharp electrode) was used as the stimulus protocol in whole-cell voltage-clamp mode. Recorded currents in the presence of 1 mM BaCl2 or 5 μM TTX reflected respective contributions of INa and IK1 to the AP waveform.
Dual whole-cell patch-clamp recordings
HEK-293 cell pairs that were formed in sparsely plated 2-day cultures were used for dual whole-cell measurement of gap junctional conductance, as previously described30. Patch electrodes (1.5–3.5 MΩ resistance) were filled with a CsCl-containing solution to block endogenous potassium currents. Each cell in the pair was clamped by a separate patch electrode to a 0-mV holding potential. A 20-mV depolarizing (8-s long) pulse was then applied to one of the two cells. Gap junctional conductance (Gj) was then calculated by dividing the resulting junctional current by the transjunctional voltage step (+20 mV) after accounting for the uncompensated series resistance in each electrode58.
Sharp intracellular recordings
HEK-293 monolayers were placed into a temperature-controlled (35 °C) chamber mounted onto an inverted microscope (Nikon, Eclipse TE2000), perfused with warm Tyrode's solution and stimulated at 1 Hz by a bipolar point electrode connected to a Grass Stimulator (Grass Technologies). Propagated APs were recorded 6 mm away from the stimulus site using a sharp intracellular electrode (50–80 MΩ tip resistance), as previously described27. Data was low-pass filtered through a 4.5-kHz Butterworth filter, sampled at 20 kHz using a Multiclamp 700B amplifier (Axon Instruments) and recorded using WinWCP or Chart software. RMP, AP amplitude, duration at 80% repolarization (APD80) and maximum upstroke velocity (dVm dt−1) were measured using a custom-made MATLAB program59.
Fluorescence recovery after photobleaching
Functional coupling in HEK-293 cells was assessed using fluorescence recovery after photobleaching29, as described previously27. HEK-293 monolayers were loaded with Calcein AM dye (Molecular Probes, 0.5 μM, 20 min at 37 °C) and washed in phosphate-buffered saline. The target cell was photobleached with a 488-nm Argon laser and imaged every 10 s for the following 6 min using an upright confocal microscope (Zeiss, LSM 510). The time course of recovery of calcein fluorescence in the target cell (by inflow from the adjacent cells) was fit by an exponential curve using custom-made MATLAB software. The calculated recovery rate constant29 for each sample was normalized by the number of cells in direct contact with the target cell.
Fabrication of excitable 2D HEK-293 networks
Isotropic excitable HEK-293 (Ex-293) monolayers were made by plating 2.5×105 cells onto fibronectin-coated (15 μg ml−1, Sigma) Aclar (Electron Microscopy Sciences) coverslips (22-mm diameter) and culturing to confluence. Anisotropic Ex-293 monolayers and thin strands were made by culturing cells on fibronectin-coated microgrooved films (Edmund Industrial Optics) or microcontact-printed fibronectin lines32, respectively.
Fabrication of 3D tissue constructs
Fibrin-based hydrogel solution was made using bovine fibrinogen (2 mg ml−1, Sigma), bovine thrombin (0.4 U mg−1 fibrinogen, Sigma) and Matrigel (10% v/v, Becton Dickinson), as previously described37. The 3D tissue cords were fabricated by casting 7×106 cells per ml hydrogel solution inside a 2.5-cm-long silicon half-tube with two Velcro felts (3M Company; St Paul, MN) pinned at the tube ends. The Velcro anchored the hydrogel and, by establishing passive tension during cell-mediated gel compaction, facilitated cell spreading and contact. Aminocaproic acid (1 mg ml−1, Sigma) was added to the culture medium to prevent degradation of fibrin by serum plasmin. To create an NRVM tissue cord with a central Ex-293 bridge, three tips of a multichannel pipette were used to simultaneously inject Ex-293/hydrogel solution in the middle and NRVM/hydrogel solutions at the two ends of the silicon half-tube. With time in culture, NRVMs and Ex-293 cells spread, aligned and coupled while maintaining distinct spatial boundaries, as identified by fluorescence imaging for GFP, mCherry and mOrange.
Optical mapping of impulse propagation
The 2D cell networks and 3D tissue constructs were stained with a voltage-sensitive dye, Di-4 ANEPPS (10 μM, 5–7 min), or a non-ratiometric Ca2+-sensitive dye, X-Rhod-1 (3 μM, 45 min). Transmembrane voltage and intracellular Ca2+ concentration, [Ca2+]i, were optically mapped with 504 optical fibres arranged into a 20-mm diameter hexagonal bundle (Redshirt Imaging), as previously described32,59. Fluorescence signals were converted to voltage by photodiodes and acquired at a 2.4-kHz sampling rate with macroscopic (750 μm) or microscopic (37.5 μm, ×20 magnification) spatial resolution. To prevent motion artefacts, cells were exposed to 10-μM blebbistatin (Sigma). An XYZ-micropositioned platinum point electrode was used to initiate electrical conduction at progressively higher pacing rates (1–30 Hz in 30-s steps)60. Data were processed, displayed and analyzed using custom-made MATLAB software. The longitudinal and transverse CV, velocity anisotropy ratio, APD80, maximum capture rate (where each stimulus was still followed by an AP response) and CV and APD80 rate dependence curves were derived, as previously described59,60.
Studies of spiral activity
Spiral waves were induced by rapid (25–30 Hz) burst pacing from the edge of a cell monolayer and optically mapped60. For anchored (anatomical) spiral wave studies, a 1.6-mm circular obstacle was punched out from the centre of a coverslip before cell seeding. Spiral rotational rate was calculated by averaging activation rate from multiple sites at the coverslip periphery. Phase map and singularity analyses (with 15-ms time delay) were used to track the tip trajectory of drifting spiral waves25,60. During perfusion of channel inhibitors, rotational rate, wave unpinning and changes in spiral wave path length were documented until termination of reentrant activity.
TTX (Sigma), BaCl2 (Alfa Aesar) and PA (Sigma) were added to extracellular bath solution during patch clamping or optical mapping (1 Hz, 1.2× threshold stimulation in monolayers) recordings to selectively inhibit INa, IK1 and gap junctional coupling, respectively. Recordings with lidocaine (50 μM, Sigma) or flecainide (10 μM, Sigma) were performed during 40 pulses at 10-Hz stimulation. All recordings using pharmacological modulation were started 5 min after drug application.
Cells were immunostained, as previously described27. Anti-sarcomeric α-actinin (Sigma, EA-53 mouse monoclonal) and anti-connexin-43 (Zymed, rabbit polyclonal) primary antibodies were applied overnight at 4 °C. Secondary Alexa Fluor 488 (chicken anti-rabbit) and Alexa Fluor 594 (chicken anti-mouse) antibodies were applied for 1 h at room temperature. Nuclei were counterstained with DAPI (Sigma). All fluorescent images were acquired, pseudocoloured and processed using IPLab (BioVision Technologies), as previously described27. Tissue bundles were fixed in 4% PFA overnight, paraffin embedded, cut into 5-μm-thick transverse and longitudinal sections, stained with haematoxylin & eosin (Duke Pathology Department) and imaged using an inverted microscope (Nikon, TE2000) and NIS-Elements software (Nikon).
Data are presented as mean±s.e.m. and were evaluated for statistical significance using linear regression or an analysis of variance (one-way, α-factor of 0.05), followed by Tukey's post hoc test for multiple comparisons. Statistical significance was defined as ^P<0.05, #P<0.01 and *P<0.001.
How to cite this article: Kirkton, R. D. and Bursac, N. Engineering biosynthetic excitable tissues from unexcitable cells for electrophysiological and cell therapy studies. Nat. Commun. 2:300 doi: 10.1038/ncomms1302 (2011).
Hille, B. Ion Channels of Excitable Membranes (Sinauer, 2001).
Catterall, W. A. The molecular basis of neuronal excitability. Science 223, 653–661 (1984).
Kleber, A. G. & Rudy, Y. Basic mechanisms of cardiac impulse propagation and associated arrhythmias. Physiol. Rev. 84, 431–488 (2004).
Kass, R. S. The channelopathies: novel insights into molecular and genetic mechanisms of human disease. J. Clin. Invest. 115, 1986–1989 (2005).
Lloyd-Jones, D. et al. Heart disease and stroke statistics--2010 update: a report from the American Heart Association. Circulation 121, e46–e215 (2010).
Ashcroft, F. M. From molecule to malady. Nature 440, 440–447 (2006).
Tan, H. L. et al. A sodium-channel mutation causes isolated cardiac conduction disease. Nature 409, 1043–1047 (2001).
Barela, A. J. et al. An epilepsy mutation in the sodium channel SCN1A that decreases channel excitability. J. Neurosci. 26, 2714–2723 (2006).
Aidley, D. J. The Physiology of Excitable Cells (Cambridge University Press, 1998).
Hodgkin, A. L. & Huxley, A. F. A quantitative description of membrane current and its application to conduction and excitation in nerve. J. Physiol. 117, 500–544 (1952).
Fitzhugh, R. Impulses and physiological states in theoretical models of nerve membrane. Biophys. J. 1, 445–466 (1961).
Zykov, V. S. & Winfree, A. T. Simulation of Wave Processes in Excitable Media (Manchester University Press, 1987).
Kubo, Y., Baldwin, T. J., Jan, Y. N. & Jan, L. Y. Primary structure and functional expression of a mouse inward rectifier potassium channel. Nature 362, 127–133 (1993).
Gellens, M. E. et al. Primary structure and functional expression of the human cardiac tetrodotoxin-insensitive voltage-dependent sodium channel. Proc. Natl Acad. Sci. USA 89, 554–558 (1992).
Fishman, G. I., Moreno, A. P., Spray, D. C. & Leinwand, L. A. Functional analysis of human cardiac gap junction channel mutants. Proc. Natl Acad. Sci. USA 88, 3525–3529 (1991).
Nerbonne, J. M. & Kass, R. S. Molecular physiology of cardiac repolarization. Physiol. Rev. 85, 1205–1253 (2005).
Grant, A. O. Cardiac ion channels. Circ. Arrhythm. Electrophysiol. 2, 185–194 (2009).
Shaner, N. C. et al. Improved monomeric red, orange and yellow fluorescent proteins derived from Discosoma sp. red fluorescent protein. Nat. Biotechnol. 22, 1567–1572 (2004).
Yu, S. P. & Kerchner, G. A. Endogenous voltage-gated potassium channels in human embryonic kidney (HEK293) cells. J. Neurosci. Res. 52, 612–617 (1998).
Thomas, P. & Smart, T. G. HEK293 cell line: a vehicle for the expression of recombinant proteins. J. Pharmacol. Toxicol. Methods 51, 187–200 (2005).
Dhamoon, A. S. & Jalife, J. The inward rectifier current (IK1) controls cardiac excitability and is involved in arrhythmogenesis. Heart Rhythm 2, 316–324 (2005).
Grant, A. O., Chandra, R., Keller, C., Carboni, M. & Starmer, C. F. Block of wild-type and inactivation-deficient cardiac sodium channels IFM/QQQ stably expressed in mammalian cells. Biophys. J. 79, 3019–3035 (2000).
Kim, T. et al. The biochemical activation of T-type Ca2+ channels in HEK293 cells stably expressing alpha1G and Kir2.1 subunits. Biochem. Biophys. Res. Commun. 324, 401–408 (2004).
Ibarra, J., Morley, G. E. & Delmar, M. Dynamics of the inward rectifier K+ current during the action potential of guinea pig ventricular myocytes. Biophys. J. 60, 1534–1539 (1991).
Gray, R. A., Pertsov, A. M. & Jalife, J. Spatial and temporal organization during cardiac fibrillation. Nature 392, 75–78 (1998).
Butterweck, A., Gergs, U., Elfgang, C., Willecke, K. & Traub, O. Immunochemical characterization of the gap junction protein connexin45 in mouse kidney and transfected human HeLa cells. J. Membr. Biol. 141, 247–256 (1994).
McSpadden, L. C., Kirkton, R. D. & Bursac, N. Electrotonic loading of anisotropic cardiac monolayers by unexcitable cells depends on connexin type and expression level. Am. J. Physiol. Cell Physiol. 297, C339–351 (2009).
Toyofuku, T. et al. Direct association of the gap junction protein connexin-43 with ZO-1 in cardiac myocytes. J. Biol. Chem. 273, 12725–12731 (1998).
Abbaci, M. et al. Gap junctional intercellular communication capacity by gap-FRAP technique: a comparative study. Biotechnol. J. 2, 50–61 (2007).
del Corsso, C. et al. Transfection of mammalian cells with connexins and measurement of voltage sensitivity of their gap junctions. Nat. Protoc. 1, 1799–1809 (2006).
Yao, J. A. et al. Remodeling of gap junctional channel function in epicardial border zone of healing canine infarcts. Circ. Res. 92, 437–443 (2003).
Klinger, R. & Bursac, N. Cardiac cell therapy in vitro: reproducible assays for comparing the efficacy of different donor cells. IEEE Eng. Med. Biol. Mag. 27, 72–80 (2008).
Carboni, M., Zhang, Z. S., Neplioueva, V., Starmer, C. F. & Grant, A. O. Slow sodium channel inactivation and use-dependent block modulated by the same domain IV S6 residue. J. Membr. Biol. 207, 107–117 (2005).
Liu, H., Atkins, J. & Kass, R. S. Common molecular determinants of flecainide and lidocaine block of heart Na+ channels: evidence from experiments with neutral and quaternary flecainide analogues. J. Gen. Physiol. 121, 199–214 (2003).
Winfree, A. T. Spiral waves of chemical activity. Science 175, 634–636 (1972).
Davidenko, J. M., Pertsov, A. V., Salomonsz, R., Baxter, W. & Jalife, J. Stationary and drifting spiral waves of excitation in isolated cardiac muscle. Nature 355, 349–351 (1992).
Bian, W., Liau, B., Badie, N. & Bursac, N. Mesoscopic hydrogel molding to control the 3D geometry of bioartificial muscle tissues. Nat. Protoc. 4, 1522–1534 (2009).
Hsu, H. et al. Slow and incomplete inactivations of voltage-gated channels dominate encoding in synthetic neurons. Biophys. J. 65, 1196–1206 (1993).
Cho, H. C., Kashiwakura, Y., Azene, E. & Marban, E. Conversion of non-excitable cells to self-contained biological pacemakers. Circulation 112, II-307 (2005).
Abraham, M. R. et al. Antiarrhythmic engineering of skeletal myoblasts for cardiac transplantation. Circ. Res. 97, 159–167 (2005).
Rudy, B., Iverson, L. E. & Conn, P. M. Ion Channels (Academic Press, 1992).
Chandra, R., Starmer, C. F. & Grant, A. O. Multiple effects of KPQ deletion mutation on gating of human cardiac Na+ channels expressed in mammalian cells. Am. J. Physiol. 274, H1643–1654 (1998).
Clancy, C. E., Tateyama, M., Liu, H., Wehrens, X. H. & Kass, R. S. Non-equilibrium gating in cardiac Na+ channels: an original mechanism of arrhythmia. Circulation 107, 2233–2237 (2003).
Bendahhou, S. et al. Defective potassium channel Kir2.1 trafficking underlies Andersen-Tawil syndrome. J. Biol. Chem. 278, 51779–51785 (2003).
Gollob, M. H. et al. Somatic mutations in the connexin 40 gene (GJA5) in atrial fibrillation. N. Engl. J. Med. 354, 2677–2688 (2006).
Dhar Malhotra, J. et al. Characterization of sodium channel alpha- and beta-subunits in rat and mouse cardiac myocytes. Circulation 103, 1303–1310 (2001).
Xie, L. H., John, S. A., Ribalet, B. & Weiss, J. N. Phosphatidylinositol-4,5-bisphosphate (PIP2) regulation of strong inward rectifier Kir2.1 channels: multilevel positive cooperativity. J. Physiol. 586, 1833–1848 (2008).
Petitprez, S. et al. SAP97 and dystrophin macromolecular complexes determine two pools of cardiac sodium channels Nav1.5 in cardiomyocytes. Circ. Res. 108, 294–304 (2011).
Rider, T. H. et al. A B cell-based sensor for rapid identification of pathogens. Science 301, 213–215 (2003).
Choi, Y. H. et al. Cardiac conduction through engineered tissue. Am. J. Pathol. 169, 72–85 (2006).
Roell, W. et al. Engraftment of connexin 43-expressing cells prevents post-infarct arrhythmia. Nature 450, 819–824 (2007).
Gepstein, L. Cell and gene therapy strategies for the treatment of postmyocardial infarction ventricular arrhythmias. Ann. N Y Acad. Sci. 1188, 32–38 (2010).
Cho, H. C. & Marban, E. Biological therapies for cardiac arrhythmias: can genes and cells replace drugs and devices? Circ. Res. 106, 674–685 (2010).
Yankelson, L. et al. Cell therapy for modification of the myocardial electrophysiological substrate. Circulation 117, 720–731 (2008).
Miragoli, M., Gaudesius, G. & Rohr, S. Electrotonic modulation of cardiac impulse conduction by myofibroblasts. Circ. Res. 98, 801–810 (2006).
de Bakker, J. M. et al. Slow conduction in the infarcted human heart. 'Zigzag' course of activation. Circulation 88, 915–926 (1993).
de Boer, T. P. et al. Inhibition of cardiomyocyte automaticity by electrotonic application of inward rectifier current from Kir2.1 expressing cells. Med. Biol. Eng. Comput. 44, 537–542 (2006).
Van Rijen, H. V., Wilders, R., Rook, M. B. & Jongsma, H. J. Dual patch clamp. Methods Mol. Biol. 154, 269–292 (2001).
Badie, N. & Bursac, N. Novel micropatterned cardiac cell cultures with realistic ventricular microstructure. Biophys. J. 96, 3873–3885 (2009).
Bursac, N., Aguel, F. & Tung, L. Multiarm spirals in a two-dimensional cardiac substrate. Proc. Natl Acad. Sci. USA 101, 15530–15534 (2004).
We thank N. Badie and J. Scull for assistance with optical mapping setup and software; L. McSpadden for FRAP analysis software; J. Dempster for WinWCP software; S. Hinds for assistance with tissue engineering experiments; Z.S. Zhang for advice with patch-clamp experiments; L. Satterwhite and A. Krol for NRVM isolation; R. Abraham for Kir2.1 cDNA; A.O. Grant and V. Neplioueva for Nav1.5 cDNA; R. Tsien for mCherry and mOrange cDNA; as well as D. Gauthier and L. Satterwhite for critical reading of the manuscript. This work was supported by predoctoral fellowships from the National Science Foundation and the American Heart Association to R.D.K. and research grants to N.B. from the American Heart Association (AHA 0530256N) and National Institutes of Health (HL106203, HL104326, HL083342 and HL095069).
The authors declare no competing financial interests.
Supplementary Figures, Tables and Methods
Supplementary Figures S1-S7, Supplementary Tables S1-S2 and Supplementary Methods (PDF 7131 kb)
Supplementary Movie 1
Electrical propagation in 2-D cultures of monoclonal Ex-293 cells. This movie shows action potential propagation in 1) an isotropic monolayer of Ex-293 cells stimulated at 2 Hz (indicated by white flashes) with a central point electrode, 2) an anisotropic monolayer of Ex-293 cells stimulated at 25 Hz with a central point electrode, and 3) two micropatterned Ex-293 cell strands stimulated at 2 Hz from a common peripheral cellular region outside the field of view. Action potential propagation was optically mapped in monolayers at 1x magnification (20 mm field of view) and in strands at 20x magnification (1 mm field of view). (AVI 3604 kb)
Supplementary Movie 2
Spiral waves in Ex-293 monolayers. This movie shows rotation of 1) a spiral wave anchored to a central acellular obstacle within an isotropic Ex-293 monolayer and 2) freely drifting single and multiple spiral waves within anisotropic and isotropic Ex-293 monolayers, respectively. Rotation of the anchored spiral wave is shown using an isopotential movie (red and blue colours denote peak and rest of the action potential, respectively), while rotations of the freely drifting spiral waves are shown using isophase movies (different colours correspond to different phases of the action potential with red denoting the spiral wave front). (AVI 4416 kb)
Supplementary Movie 3
Action potential propagation in a 3-D Ex-293 tissue cord. This movie shows action potential propagation within a 3-D biosynthetic tissue cord made with Ex-293 cells after two weeks in culture. The tissue cord is stimulated by a point electrode at 2 Hz rate. (AVI 1849 kb)
Supplementary Movie 4
Restoration of conduction between remote cardiomyocyte regions via Ex-293 cells. This movie shows attempts to electrically reconnect two isolated NRVM regions (islands) connected by a 2.5 cm-long 'S' shaped bridge made of either control (wild-type) HEK-293 or excitable Ex-293 cells. Action potential propagation induced by point stimulation (2 Hz) in one NRVM island is blocked at the entrance of the wt HEK-293 bridge. In contrast, a similar bridge made of Ex-293 cells seamlessly conducts action potentials from one NRVM island to the other. Point stimulation (6 Hz) inside the Ex-293 bridge shows propagation outward into the NRVM islands as well. The shorter action potential duration of the Ex-293 cells compared to the NRVMs is evidenced by the shorter wave width in the bridge compared to the two islands. (AVI 5506 kb)
Rights and permissions
About this article
Cite this article
Kirkton, R., Bursac, N. Engineering biosynthetic excitable tissues from unexcitable cells for electrophysiological and cell therapy studies. Nat Commun 2, 300 (2011). https://doi.org/10.1038/ncomms1302
This article is cited by
Engineered bacterial voltage-gated sodium channel platform for cardiac gene therapy
Nature Communications (2022)
Quantification of Contractile Dynamic Complexities Exhibited by Human Stem Cell-Derived Cardiomyocytes Using Nonlinear Dimensional Analysis
Scientific Reports (2019)
Dynamic all-optical drug screening on cardiac voltage-gated ion channels
Scientific Reports (2018)
Generation and customization of biosynthetic excitable tissues for electrophysiological studies and cell-based therapies
Nature Protocols (2018)
Virtual cardiac monolayers for electrical wave propagation
Scientific Reports (2017)
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.