Long-term functional and structural preservation of precision-cut human myocardium under continuous electromechanical stimulation in vitro

Abstract

In vitro models incorporating the complexity and function of adult human tissues are highly desired for translational research. Whilst vital slices of human myocardium approach these demands, their rapid degeneration in tissue culture precludes long-term experimentation. Here, we report preservation of structure and performance of human myocardium under conditions of physiological preload, compliance, and continuous excitation. In biomimetic culture, tissue slices prepared from explanted failing human hearts attain a stable state of contractility that can be monitored for up to 4 months or 2000000 beats in vitro. Cultured myocardium undergoes particular alterations in biomechanics, structure, and mRNA expression. The suitability of the model for drug safety evaluation is exemplified by repeated assessment of refractory period that permits sensitive analysis of repolarization impairment induced by the multimodal hERG-inhibitor pentamidine. Biomimetic tissue culture will provide new opportunities to study drug targets, gene functions, and cellular plasticity in adult human myocardium.

Fig. 1

Biomimetic culture system. a Schematics of the culture chamber depict attachment of tissue slice via plastic triangles (1) to spring wire with magnetic tip (2) and to linear drive (3). Changes in magnetic field are detected by an integrated sensor (4). Field stimulation is provided by graphite electrodes (5) connected to a constant current pulse generator (6). b-c Implementations based on standard dish and machined POM. d Detailed view of tissue slice fixation and stimulation electrodes. e Schematic of controller electronics. Signals from the magnetic field sensors are processed by a microcontroller that also provides bipolar stimulation pulses via a multiplexed current source. f Control unit and eight biomimetic culture chambers (BMCCs) are combined on a circuit board that can be operated in a standard CO₂ incubator. g Recording software. Data are transmitted via USB to an external PC for recording and scheduled execution of stimulation protocols. Black bars represent 10 mm, the green bar 100 mm distances

Fig. 2

Force development of cardiac slices before and during biomimetic culture. a Maximum twitch force and diastolic contracture of cardiac slices prepared after various durations of hypothermic tissue storage. Measurements were taken under isometric conditions in an organ bath. Data of immediately processed tissue ((1), n = 8) have been assessed in hypertrophic myocardium in a previous study. Tonic contracture increased in slices prepared after prolonged storage (40-55 h, n = 12) in comparison to standard transport (18–32 h, n = 15, *P<0.05, t test). Tissue did not tolerate 72 h of cooling (n = 2). Data are displayed as mean ± SEM. b High resolution recording of twitch force in BMCC. c Continuous contractility recording over 4 months. Periodic breakdowns of contraction force corresponded to medium exchange intervals (36–48 h). Positive spikes of contractility were produced by stimulation protocols employed for the assessment of refractory periods. d Time course of twitch amplitude and preload during the initial 5 weeks of cultivation (mean ± SEM of 13 samples taken from nine specimen). e Long-term effect on contractility of β₁-receptor stimulation initiated on the 2nd day of culture (denopamine, 1 µM, representative example of five independent experiments). f Contractility during omission of electrical stimulation or medium agitation for 30 h and 15 min, respectively (representative examples). Source data of 2a and 2d are provided as Source Data file

Fig. 3

Biomechanics of myocardial tissue before and after long-term biomimetic cultivation. a Twitch force at 1 mN preload, as determined under isometric conditions in an organ bath in freshly prepared (n = 10) and cultured tissues (n = 6) before and after application of isoprenaline (1 µM) (*P<0.05 control vs. isoprenaline). b–c Response of twitch force (b) and diastolic force (c) to tissue stretch (^#P<0.05 fresh vs. cultured, n = 10 for fresh tissues, n = 6 for cultured tissues). d Response of twitch force to variation of beating rate (n = 9 for fresh tissues, n = 6 for cultured tissues, *P<0.05 fresh vs. cultured, bpm = beats per minute). Data are presented as mean ± SEM. Effects were evaluated by two-way ANOVA. Source data are provided as Source Data file

Fig. 4

Myocardial tissue structure after long-term cultivation. Fresh samples of failing myocardium (1st row) were compared with tissues cultured for five or more weeks (2nd row). Staining of structural cardiomyocyte proteins (α-actinin, connexin-43) and connective tissue markers (vimentin, α-smooth muscle actin (SMA)) are indicated in column headings. Accumulation of mesenchymal cells expressing either SMA or vimentin at the cut surfaces of cultured tissues and distinct localization of N-cadherin at intercalated discs are demonstrated in the 3rd row. DNA is depicted in blue. White and black bars represent 100 µm, grey bars 10 µm distances. Pictures exemplify three independent stainings

Fig. 5

Assessment of refractory period and effects of hERG-channel blockade. a Stimulation protocol that applied paired stimulation pulses with successively decreasing time-lag. Loss of the subsequent contraction indicated equality with the refractory period (1). Phenomena of post-rest potentiation (2) and attenuation of premature beats (3) were likewise observed. b Dose-dependent increase of refractory period by the hERG-channel blocker dofetilide (n = 5, mean ± SEM). c Delayed increase of refractory period by the multimodal channel blocker pentamidine (*P<0.05 vs. baseline, two-way ANOVA, n = 5, mean ± SEM). d–g Intracellular recordings of action potentials in tissue slices after 3 weeks biomimetic cultivation either under standard conditions (d), or after a 1-week exposure to pentamidine (e–g) (representative observations). f, g Slices chronically pretreated with pentamidine were exposed acutely to combinations of the hERG blocker dofetilide and the K_ATP-channel activator cromakalim. P_m = membrane potential, RMP = resting membrane potential, APD₉₀ = action potential duration at 90% repolarization. Source data of panels 5b, c are provided as Source Data file