Engineering prokaryotic channels for control of mammalian tissue excitability

Figures

• 

  Figure 1. Generation of actively conducting human E-Fibs. (a) Voltage-clamp recording of IK1 in E-HDFs. (b) Steady-state IK1–V curves in E-HDF and wt (unexcitable) HDF. (c–e) Recording of INa (c), peak INa–V curve (d) and voltage dependence of INa activation and steady-state inactivation (e) in E-HDFs expressing NavRosD G217A. In b,d,e: n=8–10. Data in a–e recorded at 25 °C. (f) All-or-none AP response in E-HDFs induced by current pulses (Im) of increasing amplitude. (g,h) Stable overexpression of Cx43 in E-HDFs (g) results in formation of functional gap junctions, shown by increased recovery of fluorescence after photobleaching, albeit at a slower rate than that of neonatal rat cardiomyocytes (CM) (h, n=6). (i,j) Representative isochrone maps of AP conduction in electrically stimulated isotropic (i) and anisotropic (j) monolayers of E-HDFs stably co-expressing Kir2.1, NavRosD G217A and Cx43. Pulse signs indicate position of stimulating electrode. Circles denote 504 recording sites. All electrophysiological data obtained at 37 °C, unless otherwise specified. Error bars indicate s.e.m.

• 

  Figure 2. Customization of E-Fib phenotype via mutagenesis. (a) BacNav-based strategy for E-Fib customization. (b) Experimental (n=8) and modelling peak INa–V curves. Cell membrane was held at −80 mV before stepping the voltage from −50 to 60 mV in 10 mV increments. (c) Experimental (n=21) and modelling steady-state IK–V curves. Cell membrane was held at −40 mV before stepping the voltage from −90 to 50 mV in 10 mV increments. (d,e) AP initiation and conduction characteristics of BacNav mutant expressing E-HDFs obtained by computational modeling: upstroke velocity (d) and CV (e). For modelling, parameters derived from voltage-clamp measurements at 25 °C were scaled for the temperature of 37 °C. Highly hyperpolarized inactivation of WT and D60E channels prevented AP initiation. (f,g) Time constants of activation (f) and inactivation (g) of selected NavSheP D60X mutants compared with NavRosD G217A (RosDm), recorded in E-HEK293 cells at 25 °C (n=5). (h–j) When co-expressed with Kir2.1 in E-HEK293 cells, NavSheP mutants give rise to APs (h) with faster upstroke (i; n=10–18) and shorter duration (j; n=10–18) than NavRosD G217A (n=11 for i,j). (k) NavSheP D60X expressions in anisotropic E-HDF…

• 

  Figure 3. Tuning electrical properties of E-Fibs via FACS. (a,b) Steady-state IK1–V curves (a; n=8–12) and peak INa–V curve (b; n=8–12) in three groups (blue, red and green denoting low, medium and high eGFP-expressing cells, respectively), measured at 25 °C. (c) Representative AP traces with inset showing AP upstrokes in the three eGFP-sorted E-HDF groups. (d–f) Resting membrane potential (RMP, d), maximum AP upstroke velocity (e) and APD80 (f) in the three eGFP-sorted E-HDF groups (n=14–18). (g,h) Representative isochrone maps of AP conduction in electrically stimulated anisotropic monolayers of E-HDFs (g) showing that cells with higher eGFP expression had faster CV (h; n=6–10). All E-HDFs were first sorted for high Cx43-mCherry level to ensure strong intercellular coupling. Pulse signs indicate location of stimulating electrode. Circles denote 504 recording sites. (i) CV in electrically stimulated monolayers of three mCherry-sorted E-HDF groups (left, middle and right bars denoting low, medium and high mCherry-expressing cells, respectively; n=5). #P<0.01 in f; *P<0.001 versus low eGFP group in e,h or versus low or medium mCherry group in i. All electrophysiological data obtained at 37…

• 

  Figure 4. Stability and robustness of E-Fib phenotype. (a,b) E-HDFs transduced with NavSheP D60N-P2A-eGFP-Kir2.1 and Cx43-P2A-mCherry lentiviruses retain proliferative potential as shown by Ki-67-positive staining (a) after ∼100-fold expansion (b), 4 weeks post-lentiviral transduction. Growth curve is fit to a power function. (c,d) Conversion of smooth muscle actin (SMA)− E-HDFs (c) to SMA+ myofibroblasts after transforming growth factor (TGF)-β treatment (d). (e–i) Steady-state IK1–V (e) and peak INa–V (f) curves (recorded at 25 °C) as well as resting membrane potential (RMP, g), APD80 (h) and AP upstroke velocity (i) do not differ among control, expanded or TGF-β-treated E-HDFs. (n=5–9). (j–l) RMP (j), maximum AP upstroke velocity (k) and APD80 (l) in different E-Fibs (E-HVF, human ventricular fibroblasts; E-HA, human astrocytes) transduced with the NavSheP D60N-P2A-eGFP-Kir2.1 lentivirus (n=6–8). (m) Steady-state I–V curves in different wt Fibs (n=4–6). ^P<0.05, #P<0.01 versus HVF in l. All electrophysiological data obtained at 37 °C, unless otherwise specified. Error bars indicate s.e.m; statistical significance was determined by one-way analysis of variance, followed by Tukey's…

• 

  Figure 5. Improvement of AP conduction by E-Fibs in a model of cardiac interstitial fibrosis. (a) Schematic depicting use of E-Fibs to improve impaired conduction in an in vitro model of cardiac interstitial fibrosis. (b) Immunostaining corresponding to dashed-line square area in a showing vimentin+ E-HDFs in between sarcomeric α-actinin (SAA)+ neonatal rat cardiomyocytes (CMs). (c–e) Average CV (c), APD80 (d) and impulse wavelength (WL=CVxAPD80, e) values during longitudinal or transverse conduction in CM-only cultures (fibrosis control), CM+wt-HDF co-cultures, CM+NavSheP D60A/Kir2.1-expressing HDF co-cultures, CM+Cx43-expressing HDF co-cultures, CM+E-HDF co-cultures and confluent anisotropic CM cultures (healthy control) (n=5–8). *P<0.001 versus transverse CVs in fibrosis control group. Error bars indicate s.e.m; statistical significance was determined by one-way analysis of variance, followed by Tukey's post-hoc test to calculate P-values.

• 

  Figure 6. Improvement of mammalian AP conduction by BacNav in modeled pathological conditions. (a) Schematic depicting direct expression of BacNav in mammalian excitable tissues containing endogenous Nav1.5, to enhance conduction and prevent conduction block in simulated pathological conditions. (b) Steady-state inactivation (h∞) of NavSheP D60A is more depolarized compared with that of Nav1.5 (n=5). (c,d) Conduction failure due to inactivation of Nav1.5 by membrane depolarization with elevated extracellular K+ (c, n=6) or blocking of intercellular coupling by PA (d, n=12–18) in Ex293 monolayers (black squares) is rescued by co-expression of NavSheP D60A (white circles). (e) Schematic depicting exogenous expression of BacNav in neonatal rat CMs to increase resistance to conduction block in ischaemic conditions. Black dashed circle denotes position of glass coverslip used to induce regional ischaemia in CM monolayer. (f) Progressive conduction slowing until block with time of ischaemia in control (CM) and NavSheP D60A transduced (CM+D60A) monolayers (n=6). (g,h) Representative isochrones maps of a CM monolayer showing conduction slowing after 6 min (g) and complete block after 11 min of…