CTCM maintains heart tissue physiological stretch for 12 days in culture
Hemodynamic mechanical cues play a critical role in preserving the functionality of cardiomyocytes (CM) in vitro 26–28. In the current manuscript, we developed a CTCM (Fig. 1a) that can emulate the adult cardiac milieu by inducing simultaneous electrical and mechanical stimulation at the physiological frequency (1.2Hz, 72 beats per minute). To avoid over stretching the tissue during the diastolic phase the tissues were oversized by 25% using a 3D-printed apparatus (Fig. 1b). To fully reproduce the cardiac cycle, electrical stimulation induced by a C-PACE system was synchronized to initiate at 100ms prior to the systolic phase using a data acquisition system (DAQ). The tissue culture system uses a programmable pneumatic driver (LB engineering, Germany) to cyclically distend a flexible silicone membrane inducing the stretch of the heart slices in the chambers above. The system is connected to an external air line with a pressure probe allowing fine pressure tuning (± 1 mmHg) and timing (± 1 ms) (Fig. 1c). Using one pneumatic driver we can operate 4 CTCM devices, each accommodating 6 tissue slices (Fig. 1d). Within the CTCM, the air pressures in the air chambers are translated into synchronized pressures in the fluid chamber and produce physiological stretch of the heart slices (Fig. 2a, supplementary movie 1). Assessment of the tissue stretch at 80 mmHg resulted in 25% stretch of the tissue slice (Fig. 2b). This percent stretch has been shown to correspond with a physiological sarcomere length, 2.2-2.3 um, for normal contractility of heart slices 20,22,29. Tissue movements are assessed using a custom camera setup (Supplementary Fig. 1). The tissue movement amplitude and speed rate (Fig. 2c-d) are consistent with the stretching during the cardiac cycle and the timing during systole and diastole (Fig. 2b). The heart tissue stretch and speed during contraction and relaxation remained consistent over the 12 days in culture (Fig. 2e).
Heart slices cultured in CTCM maintain full viability but not structural integrity over 12 days.
In our previous static biomimetic heart slice culture system 18,19, we were able to maintain the viability, functionality, and structural integrity of heart slices for 6 days with the application of electrical stimulation and optimized composition of the culture medium. However, after day 10 and 12, there was a sharp decline of these parameters. We will refer to slices cultured in our previous static biomimetic culture system 18,19 as control condition (Ctrl) and we will refer to the slices cultured under synchronized mechanical and electrical stimulation (CTCM) using our previously optimized culture media as MC condition. Interestingly, by introducing physiological mechanical stimulation using the CTCM, the viability of 12-day heart slices was maintained similar to fresh heart slices in MC condition but not in the Ctrl condition as shown by MTT assay (Fig. 3a). This indicates that the mechanical stimulation and simulation of the cardiac cycle can maintain the viability of tissue slices for twice the time reported in our previous static culture system. However, assessment of the structural integrity of the tissue slices as assessed by immunolabeling for cardiac Troponin-T and Connexin 43 demonstrated that connexin 43 expression of day 12 MC tissue was significantly higher than the same day control, but the uniform expression of connexin43 and z-disc formation were not fully maintained. Using an artificial intelligence (AI) based framework to quantify tissue structural integrity 30, MC tissue showed an improved structural similarity to day 0 compared to the static control slices. Furthermore, Masson’s trichrome stain showed a significantly lower percent area of fibrosis with the MC condition compared to the control condition at day 12 of culture. While the CTCM did improve the viability of day 12 heart tissue slices to levels similar to fresh heart tissue, it did not significantly improve the structural integrity of the heart slices.
Small molecule screening to improve the heart tissue viability and structural integrity in CTCM
We hypothesized that by incorporating small molecules into the culture media cardiomyocyte integrity could be improved and the development of fibrosis during the CTCM culture could be reduced. Therefore, we performed a small molecule screening using our static control culture 18,19 because of the lower number of confounding factors associated with it. Dexamethasone (Dex), tri-iodothyronine (T3), and SB431542 (SB) were selected for this screening. These small molecules have been previously used in hiPSC-CM cultures to induce cardiomyocyte maturation by improving sarcomere length, t-tubules, and conduction velocity 31,32. Additionally, both Dex, a glucocorticoid, and SB are known to suppress inflammation 33,34. Therefore, we tested whether incorporating one or a combination of these small molecules would improve the structural integrity of the heart slices. For the initial screening the dosage of each compound was selected based on the concentration typically used in cell culture models (1 µM Dex 31, 100nM T3 31, and 2.5 µM SB 35). Following 12 days in culture, the combination of T3 and Dex resulted in the best cardiomyocyte structural integrity and the lowest fibrotic remodeling (Supplementary Fig. 2-3). In addition, using either half or double of these concentrations of T3 and Dex had detrimental effects compared to the normal concentrations (Supplementary Fig. 4).
The combination of T3 and Dex with the CTCM fully maintained pig heart slices similar to fresh heart slices for 12 days.
Following the initial screening, we performed head-to-head comparisons of 4 culture conditions; Ctrl: heart slices cultured in our previously described static culture with our optimized culture media 18,19; TD: Ctrl with T3 and Dex added to the culture media; MC: heart slices cultured in CTCM using our previously optimized culture media; and MT: CTCM with T3 and Dex added to the culture media. After 12 days in culture, the viability of MC and MT tissues was maintained similar to fresh tissue as assessed by the MTT assay. Interestingly, the addition of T3 and Dex to the transwell culture (TD) did not significantly improve viability compared to Ctrl condition, implying the vital role of mechanical stimulation in maintaining heart slice viability.
Metabolic shift from fatty acid oxidation to glycolysis is a hallmark for cardiomyocyte dedifferentiation. Immature cardiomyocytes primarily utilize glucose for ATP production and have underdeveloped mitochondria with few cristae 6 36. Glucose utilization assay demonstrated that under MC and MT conditions, glucose utilization was similar to day 0 tissue. However, Ctrl samples showed a significant increase in glucose utilization compared to fresh tissue. This indicates that the combination of CTCM and T3+Dex improved the viability and metabolic activities in heart slices cultured for 12 days.
To analyze the overall impact of the CTCM and T3+Dex on the global transcriptional landscape of the heart slice tissue, we performed RNAseq on heart slices from all 4 different culture conditions (Supplementary source data). Interestingly, MT slices showed high transcriptional similarity to the fresh heart tissue with only 16 out of 13,642 genes differentially expressed. However, as we demonstrated before 19, the Ctrl slices showed 1,229 differentially expressed gene after 10-12 days in culture (Fig. 5D). Interestingly, the Ctrl slices showed downregulation of the cardiac genes and cell cycle genes and an upregulation of inflammatory gene programs. These data indicate that the dedifferentiation that normally occurs following long-term culture, was completely attenuated under the MT condition (Supplementary Fig. 5). A closer examination of the sarcomeric genes revealed that only the MT conditions that preserved the sarcomeric genes from downregulation seen in Ctrl, TD, and MC conditions (Fig. 5E). These data indicate that with the combination of mechanical and humoral stimulation (T3+Dex), the transcriptome of heart slices could be maintained similar to fresh heart slices for 12 days in culture.
These transcriptional findings were confirmed by the fact that the structural integrity of the cardiomyocytes in heart slices was best preserved for 12 days in the MT condition as shown by the intact and localized gap junction protein, connexin 43 (Fig. 5A). Furthermore, fibrosis in heart slices in MT condition was significantly reduced compared to Ctrl and was similar to the fresh heart slices (Fig. 5B). These data indicate that the combination of mechanical stimulation and T3+Dex treatment was able to effectively preserve heart slice cardiac structure for extended time in culture.
Inducing cardiac hypertrophy through overstretching the tissue in the CTCM
Lastly, the ability of the CTCM to model overload cardiac hypertrophy was assessed by increasing the cardiac tissue stretch amplitude. In the CTCM, the peak pressure in the air chamber was increased from 80 mmHg (normal stretch) to 140 mmHg (Fig. 6a). This corresponds to an increase in the preload stretch of 32% (Fig. 6b), which has been previously shown to be the appropriate percent stretch necessary for a heart slice to achieve a sarcomere length similar to that seen in hypertrophy 20,22,29. The heart tissue stretch and speed during contraction and relaxation remained consistent over the 6 days of culture (Fig. 6c). Heart slices tissue from the MT conditions were either subjected to normal preload stretch (MT (Norm)) or overload conditions (MT (OL)) for 6 days. As early as 4 days in culture, the hypertrophic biomarker, NT-proBNP, was significantly increased in the culture media in MT (OL) conditions compared to the MT (Norm) conditions (Fig. 7a). Furthermore, following 6 days in culture, the cell size in MT (OL) (Fig. 7b) was significantly increased compared to MT (Norm) heart slices. These results show the progressive development of pathological remodeling following overload and provide a proof of concept that the CTCM device can be used as a platform to induce overload cardiac hypertrophy.