Thank you for visiting Nature.com. The browser version you are using has limited CSS support. For the best experience, we recommend that you use an updated browser (or disable Compatibility Mode in Internet Explorer). In the meantime, to ensure continued support, we will render the site without styles and JavaScript.
There is a need for a reliable in vitro system that can accurately reproduce the physiological environment of the heart for drug testing. The limited availability of human heart tissue culture systems has led to inaccurate interpretations of cardiac drug effects. Here, we have developed a cardiac tissue culture model (CTCM) that electromechanically stimulates heart slices and undergoes physiological stretch during the systolic and diastolic phases of the cardiac cycle. After 12 days of culture, this approach partially improved the viability of heart sections, but did not fully preserve their structural integrity. Therefore, after small molecule screening, we found that the addition of 100 nM triiodothyronine (T3) and 1 μM dexamethasone (Dex) to our medium maintained the microstructure of the sections for 12 days. In combination with T3/Dex treatment, the CTCM system maintained transcriptional profiles, viability, metabolic activity, and structural integrity at the same level as fresh heart tissue for 12 days. In addition, excessive stretching of cardiac tissue in culture induces hypertrophic cardiac signaling, providing evidence for the ability of CTCM to mimic hypertrophic conditions induced by cardiac stretch. In conclusion, CTCM can model the physiology and pathophysiology of the heart in culture over long periods of time, enabling reliable drug screening.
Prior to clinical research, reliable in vitro systems are needed that can accurately reproduce the physiological environment of the human heart. Such systems should mimic altered mechanical stretch, heart rate, and electrophysiological properties. Animal models are commonly used as a screening platform for cardiac physiology with limited reliability in reflecting the effects of drugs in the human heart1,2. Ultimately, the Ideal Cardiac Tissue Culture Experimental Model (CTCM) is a model that is highly sensitive and specific for various therapeutic and pharmacological interventions, accurately reproducing the physiology and pathophysiology of the human heart3. The absence of such a system limits the discovery of new treatments for heart failure4,5 and has led to drug cardiotoxicity as a major reason for exiting the market6.
Over the past decade, eight non-cardiovascular drugs have been withdrawn from clinical use because they cause QT interval prolongation leading to ventricular arrhythmias and sudden death7. Thus, there is an increasing need for reliable preclinical screening strategies to assess cardiovascular efficacy and toxicity. The recent use of human-induced pluripotent stem cell-derived cardiomyocytes (hiPS-CM) in drug screening and toxicity testing provides a partial solution to this problem. However, the immature nature of hiPS-CMs and the lack of multicellular complexity of cardiac tissue are major limitations of this method. Recent studies have shown that this limitation can be partly overcome by using early hiPS-CM to form cardiac tissue hydrogels shortly after the onset of spontaneous contractions and gradually increasing electrical stimulation over time. However, these hiPS-CM microtissues lack the mature electrophysiological and contractile properties of the adult myocardium. In addition, human cardiac tissue has a more complex structure, consisting of a heterogeneous mixture of different cell types, including endothelial cells, neurons, and stromal fibroblasts, interconnected by specific sets of extracellular matrix proteins. This heterogeneity of non-cardiomyocyte populations11,12,13 in the adult mammalian heart is a major barrier to modeling cardiac tissue using individual cell types. These major limitations emphasize the importance of developing methods for culturing intact myocardial tissue under physiological and pathological conditions.
Cultured thin (300 µm) sections of the human heart have proven to be a promising model of intact human myocardium. This method provides access to a complete 3D multicellular system similar to human heart tissue. However, until 2019, the use of cultured heart sections was limited by the short (24 h) culture survival. This is due to a number of factors including the lack of physical-mechanical stretch, the air-liquid interface, and the use of simple media that do not support the needs of cardiac tissue. In 2019, several research groups demonstrated that incorporating mechanical factors into cardiac tissue culture systems can extend culture life, improve cardiac expression, and mimic cardiac pathology. Two elegant studies 17 and 18 show that uniaxial mechanical loading has a positive effect on cardiac phenotype during culture. However, these studies did not use the dynamic three-dimensional physico-mechanical loading of the cardiac cycle, since heart sections were loaded with either isometric tensile forces 17 or linear auxotonic loading 18 . These methods of tissue stretching resulted in the suppression of many cardiac genes or the overexpression of genes associated with abnormal stretch responses. Notably, Pitoulis et al. 19 developed a dynamic heart slice culture bath for cardiac cycle reconstruction using force transducer feedback and tension drives. Although this system allows for more accurate in vitro cardiac cycle modeling, the complexity and low throughput of the method limits the application of this system. Our laboratory has recently developed a simplified culture system using electrical stimulation and an optimized medium to maintain the viability of sections of porcine and human heart tissue for up to 6 days20,21.
In the current manuscript, we describe a cardiac tissue culture model (CTCM) using sections of the porcine heart that incorporates humoral cues to recapitulate three-dimensional cardiac physiology and pathophysiological distension during the cardiac cycle. This CTCM can increase the accuracy of pre-clinical drug prediction to a level never before achieved by providing a cost-effective, mid-throughput cardiac system that mimics the physiology/pathophysiology of the mammalian heart for pre-clinical drug testing.
Hemodynamic mechanical signals play a critical role in maintaining cardiomyocyte function in vitro 22,23,24. In the current manuscript, we have developed a CTCM (Figure 1a) that can mimic the adult cardiac environment by inducing both electrical and mechanical stimulation at physiological frequencies (1.2 Hz, 72 beats per minute). To avoid excessive tissue stretch during diastole, a 3D printing device was used to increase tissue size by 25% (Fig. 1b). Electrical pacing induced by the C-PACE system was timed to start 100 ms before systole using a data acquisition system to fully reproduce the cardiac cycle. The tissue culture system uses a programmable pneumatic actuator (LB Engineering, Germany) to cyclically expand a flexible silicone membrane to cause expansion of the heart slices in the upper chamber. The system was connected to an external air line through a pressure transducer, which made it possible to accurately adjust the pressure (± 1 mmHg) and time (± 1 ms) (Fig. 1c).
a Attach the tissue section to the 7 mm support ring, shown in blue, inside the culture chamber of the device. The culture chamber is separated from the air chamber by a thin flexible silicone membrane. Place a gasket between each chamber to prevent leaks. The lid of the device contains graphite electrodes that provide electrical stimulation. b Schematic representation of the large tissue device, guide ring and support ring. The tissue sections (brown) are placed on the oversized device with the guide ring placed in the groove on the outer edge of the device. Using the guide, carefully place the support ring coated with tissue acrylic adhesive over the section of cardiac tissue. c Graph showing the time of electrical stimulation as a function of air chamber pressure controlled by a programmable pneumatic actuator (PPD). A data acquisition device was used to synchronize electrical stimulation using pressure sensors. When the pressure in the culture chamber reaches the set threshold, a pulse signal is sent to the C-PACE-EM to trigger electrical stimulation. d Image of four CTCMs placed on an incubator shelf. Four devices are connected to one PPD via a pneumatic circuit, and pressure sensors are inserted into the hemostatic valve to monitor the pressure in the pneumatic circuit. Each device contains six tissue sections.
Using a single pneumatic actuator, we were able to control 4 CTCM devices, each of which could hold 6 tissue sections (Fig. 1d). In CTCM, the air pressure in the air chamber is converted to synchronous pressure in the fluid chamber and induces physiological expansion of the heart slice (Figure 2a and Supplementary Movie 1). Evaluation of tissue stretch at 80 mm Hg. Art. showed stretching of tissue sections by 25% (Fig. 2b). This percentage stretch has been shown to correspond to a physiological sarcomere length of 2.2–2.3 µm for normal cardiac section contractility17,19,25. Tissue movement was assessed using custom camera settings (Supplementary Figure 1). The amplitude and velocity of tissue movement (Fig. 2c, d) corresponded to stretch during the cardiac cycle and time during systole and diastole (Fig. 2b). Stretch and velocity of cardiac tissue during contraction and relaxation remained constant for 12 days in culture (Fig. 2f). To evaluate the effect of electrical stimulation on contractility during culture, we developed a method for determining active deformity using a shading algorithm (Supplementary Fig. 2a,b) and were able to distinguish between deformities with and without electrical stimulation. The same section of the heart (Fig. 2f). In the movable region of the cut (R6-9), the voltage during electrical stimulation was 20% higher than in the absence of electrical stimulation, which indicates the contribution of electrical stimulation to contractile function.
Representative traces of air chamber pressure, fluid chamber pressure, and tissue movement measurements confirm that chamber pressure changes fluid chamber pressure, causing a corresponding movement of the tissue slice. b Representative traces of percent stretch (blue) of tissue sections corresponding to percent stretch (orange). c The measured motion of the cardiac slice is consistent with the measured speed of motion. (d) Representative trajectories of cyclic motion (blue line) and velocity (orange dotted line) in a slice of the heart. e Quantification of cycle time (n = 19 slices per group, from different pigs), contraction time (n = 19 slices per group), relaxation time (n = 19 slices per group, from different pigs), tissue movement (n = 25 ). slices)/group from different pigs), peak systolic velocity (n = 24(D0), 25(D12) slices/group from different pigs) and peak relaxation rate (n=24(D0), 25(D12) slices/group from different pigs). Two-tailed Student’s t-test showed no significant difference in any parameter. f Representative strain analysis traces of tissue sections with (red) and without (blue) electrical stimulation, ten regional areas of tissue sections from the same section. The bottom panels show the quantification of the percentage difference in strain in tissue sections with and without electrical stimulation in ten areas from different sections. (n = 8 slices/group from different pigs, Two-tailed Student t-test is performed; ****p < 0.0001, **p < 0.01, *p < 0.05). (n = 8 slices/group from different pigs, Two-tailed Student t-test is performed; ****p < 0.0001, **p < 0.01, *p < 0.05). (n = 8 срезов/группу от разных свиней, проводится двусторонний t-критерий Стьюдента; ****p<0,0001, **p<0,01, *p<0,05). (n = 8 sections/group from different pigs, two-tailed Student’s t-test; ****p<0.0001, **p<0.01, *p<0.05). （n = 8 片/组，来自不同的猪，进行双尾学生t 检验；****p < 0.0001，**p < 0.01，*p < 0.05）。 （n = 8 片/组，来自不同的猪，进行双尾学生t 检验；****p < 0.0001，**p < 0.01，*p < 0.05）。 (n = 8 срезов/группу, от разных свиней, двусторонний критерий Стьюдента; ****p <0,0001, **p <0,01, *p <0,05). (n = 8 sections/group, from different pigs, two-tailed Student’s t-test; ****p<0.0001, **p<0.01, *p<0.05). Error bars represent the mean ± standard deviation.
In our previous static biomimetic heart slice culture system [20, 21], we maintained the viability, function, and structural integrity of heart slices for 6 days by applying electrical stimulation and optimizing the medium composition. However, after 10 days, these figures dropped sharply. We will refer to sections cultured in our previous static biomimetic culture system 20, 21 control conditions (Ctrl) and we will use our previously optimized medium as MC conditions and culture under simultaneous mechanical and electrical stimulation (CTCM). called . First, we determined that mechanical stimulation without electrical stimulation was insufficient to maintain tissue viability for 6 days (Supplementary Fig. 3a,b). Interestingly, with the introduction of physio-mechanical and electrical stimulation using STCM, the viability of 12-day heart sections remained the same as in fresh heart sections under MS conditions, but not under Ctrl conditions, as shown by MTT analysis (Fig. 1). 3a). This suggests that mechanical stimulation and simulation of the cardiac cycle can keep tissue sections viable for twice as long as reported in our previous static culture system. However, assessment of the structural integrity of tissue sections by immunolabeling of cardiac troponin T and connexin 43 showed that connexin 43 expression was significantly higher in MC tissues on day 12 than in controls on the same day. However, uniform connexin 43 expression and Z-disc formation were not fully maintained (Fig. 3b). We use an artificial intelligence (AI) framework to quantify tissue structural integrity26, an image-based deep learning pipeline based on troponin-T and connexin staining43 to automatically quantify the structural integrity and fluorescence of heart slices in terms of strength of localization. This method uses a Convolutional Neural Network (CNN) and a deep learning framework to reliably quantify the structural integrity of cardiac tissue in an automated and unbiased manner, as described in the reference. 26. MC tissue showed improved structural similarity to day 0 compared to static control sections. In addition, Masson’s trichrome staining revealed a significantly lower percentage of fibrosis under MS conditions compared to control conditions on day 12 of culture (Fig. 3c). While CTCM increased the viability of heart tissue sections at day 12 to a level similar to that of fresh heart tissue, it did not significantly improve the structural integrity of the heart sections.
a Bar graph shows quantification of the MTT viability of fresh heart slices (D0) or heart slices culture for 12 days either in static culture (D12 Ctrl) or in CTCM (D12 MC) (n = 18 (D0), 15 (D12 Ctrl), 12 (D12 MC) slices/group from different pigs, one way ANOVA test is performed; ####p < 0.0001 compared to D0 and **p < 0.01 compared to D12 Ctrl). a Bar graph shows quantification of the MTT viability of fresh heart slices (D0) or heart slices culture for 12 days either in static culture (D12 Ctrl) or in CTCM (D12 MC) (n = 18 (D0), 15 (D12 Ctrl ), 12 (D12 MC) slices/group from different pigs, one way ANOVA test is performed; ####p < 0.0001 compared to D0 and **p < 0.01 compared to D12 Ctrl). the histogram shows the quantification of the viability of MTT fresh heart sections (D0) or culture of heart sections for 12 days in either static culture (D12 control) or CTCM (D12 MC) (n = 18 (D0), 15 (D12 control). ) ), 12 (D12 MC) sections/group from different pigs, a one-way ANOVA test is performed; ####p < 0,0001 по сравнению с D0 и **p < 0,01 по сравнению с D12 Ctrl). ####p < 0.0001 compared to D0 and **p < 0.01 compared to D12 Ctrl). a 条形图显示在静态培养(D12 Ctrl) 或CTCM (D12 MC) (n = 18 (D0), 15 (D12 Ctrl) 中新鲜心脏切片(D0) 或心脏切片培养12 天的MTT 活力的量化)，来自不同猪的12 (D12 MC) 切片/组，进行单向ANOVA 测试；与D0 相比，####p < 0.0001，与D12 Ctrl 相比，**p < 0.01)。 a 条形图显示在静态培养(D12 Ctrl) 或CTCM (D12 MC) (n = 18 (D0), 15 (D12 Ctrl) 中新鲜心脏切片(D0) ，来自不同猪的12 (D12 MC) 切片/组，进行单向ANOVA 测试；与D0 相比，####p < 0.0001，与D12 Ctrl 相比，**p。) histogram showing the quantification of MTT viability in fresh heart sections (D0) or heart sections cultured for 12 days in static culture (D12 control) or CTCM (D12 MC) (n = 18 (D0), 15 (D12 control)) , 12 (D12 MC) sections/group from different pigs, one-way ANOVA test; ####p < 0,0001 по сравнению с D0, **p < 0,01 по сравнению с D12 Ctrl). ####p < 0.0001 compared to D0, **p < 0.01 compared to D12 Ctrl). b Troponin-T (green), connexin 43 (red) and DAPI (blue) in freshly isolated heart sections (D0) or heart sections cultured under static conditions (Ctrl) or CTCM conditions (MC) for 12 days) of representative immunofluorescence images (blank scale = 100 µm). Artificial intelligence quantification of the heart tissue structural integrity (n = 7 (D0), 7 (D12 Ctrl), 5 (D12 MC) slices/group each from different pig, one-way ANOVA test is performed; ####p < 0.0001 compared to D0 and ****p < 0.0001 compared to D12 Ctrl). Artificial intelligence quantification of the heart tissue structural integrity (n = 7 (D0), 7 (D12 Ctrl), 5 (D12 MC) slices/group each from different pigs, one-way ANOVA test is performed; ####p < 0.0001 compared to D0 and ****p < 0.0001 compared to D12 Ctrl). Количественная оценка структурной целостности сердечной ткани искусственным интеллектом (n = 7 (D0), 7 (D12 Ctrl), 5 (D12 MC) срезов/групп от разных свиней, проводится однофакторный тест ANOVA; ####p < 0,0001 по сравнению с D0 и ****p < 0,0001 по сравнению с D12 Ctrl). Quantification of the structural integrity of cardiac tissue by artificial intelligence (n = 7 (D0), 7 (D12 Ctrl), 5 (D12 MC) sections/groups from different pigs, one-way ANOVA test performed; ####p < 0.0001 vs. with D0 and ****p < 0.0001 compared to D12 Ctrl).人工智能量化心脏组织结构完整性（n = 7 (D0), 7 (D12 Ctrl), 5 (D12 MC) slices/group each of different pig, one-way ANOVA test;####p < 0.0001 与D0 相比，****p < 0.0001 与D12 Ctrl 相比）。人工智能量化心脏组织结构完整性（n = 7 (D0), 7 (D12 Ctrl), 5 (D12 MC) slices/group each of different pigs, one-way ANOVA test;####p < 0.0001 与D0相比，****p < 0.0001 与D12 Ctrl 相比）。 Искусственный интеллект для количественной оценки структурной целостности сердечной ткани (n = 7 (D0), 7 (D12 Ctrl), 5 (D12 MC) срезов/группу каждой из разных свиней, односторонний тест ANOVA; ####p <0,0001 vs. D0 Для сравнения ****p < 0,0001 по сравнению с D12 Ctrl). Artificial intelligence to quantify the structural integrity of cardiac tissue (n = 7 (D0), 7 (D12 Ctrl), 5 (D12 MC) sections/group each of different pigs, one-way ANOVA test; ####p<0.0001 vs .D0 For comparison ****p < 0.0001 compared to D12 Ctrl). c Representative images (left) and quantification (right) for heart slices stained with Masson’s trichrome stain (Scale bare = 500 µm) (n = 10 slices/group each from different pig, one-way ANOVA test is performed; ####p < 0.0001 compared to D0 and ***p < 0.001 compared to D12 Ctrl). c Representative images (left) and quantification (right) for heart slices stained with Masson’s trichrome stain (Scale bare = 500 µm) (n = 10 slices/group each from different pigs, one-way ANOVA test is performed; #### p < 0.0001 compared to D0 and ***p < 0.001 compared to D12 Ctrl). c Репрезентативные изображения (слева) и количественная оценка (справа) срезов сердца, окрашенных трихромным красителем Массона (масштаб без покрытия = 500 мкм) (n = 10 срезов/группу от разных свиней, выполняется односторонний тест ANOVA; #### p < 0,0001 по сравнению с D0 и ***p < 0,001 по сравнению с D12 Ctrl). c Representative images (left) and quantification (right) of heart sections stained with Masson’s trichrome stain (uncoated scale = 500 µm) (n = 10 sections/group from different pigs, one-way ANOVA test performed; #### p < 0 .0001 compared to D0 and ***p < 0.001 compared to D12 Ctrl). c 用Masson 三色染料染色的心脏切片的代表性图像（左）和量化（右）（裸尺度= 500 µm）（n = 10 个切片/组，每组来自不同的猪，进行单向ANOVA 测试；#### p < 0.0001 与D0 相比，***p < 0.001 与D12 Ctrl 相比）。 C 用 masson 三 色 染料 的 心脏 切片 的 代表性 （左 左） 量化 （右） 裸尺度 裸尺度 裸尺度 裸尺度 = 500 µm） （n = 10 个 切片 组 每 组 来自 不同 猪 ， 进行 单向 单向 Anova 测试；#### p < 0.0001 与D0 相比，***p < 0.001 与D12 Ctrl 相比）。 c Репрезентативные изображения (слева) и количественный анализ (справа) срезов сердца, окрашенных трихромным красителем Массона (чистая шкала = 500 мкм) (n = 10 срезов/группа, каждый от другой свиньи, протестировано с помощью однофакторного дисперсионного анализа ;### #p < 0,0001 по сравнению с D0, ***p < 0,001 по сравнению с D12 Ctrl). c Representative images (left) and quantitation (right) of heart sections stained with Masson’s trichrome stain (blank = 500 µm) (n = 10 sections/group, each from a different pig, tested by one-way analysis of variance ;### # p < 0.0001 compared to D0, ***p < 0.001 compared to D12 Ctrl). Error bars represent the mean ± standard deviation.
We hypothesized that by adding small molecules to the culture medium, cardiomyocyte integrity could be improved and fibrosis development reduced during CTCM culture. We therefore screened for small molecules using our static control cultures20,21 due to the small number of confounding factors. Dexamethasone (Dex), triiodothyronine (T3), and SB431542 (SB) were chosen for this screen. These small molecules have previously been used in hiPSC-CM cultures to induce maturation of cardiomyocytes by increasing sarcomere length, T-tubules, and conduction velocity. In addition, both Dex (a glucocorticoid) and SB are known to suppress inflammation29,30. Therefore, we tested whether the inclusion of one or a combination of these small molecules would improve the structural integrity of heart sections. For initial screening, the dose of each compound was selected based on the concentrations commonly used in cell culture models (1 μM Dex27, 100 nM T327, and 2.5 μM SB31). After 12 days of culture, the combination of T3 and Dex resulted in optimal cardiomyocyte structural integrity and minimal fibrous remodeling (Supplementary Figures 4 and 5). In addition, use of double or double these concentrations of T3 and Dex produced deleterious effects compared to normal concentrations (Supplementary Fig. 6a,b).
After initial screening, we performed a head-to-head comparison of 4 culture conditions (Figure 4a): Ctrl: heart sections cultured in our previously described static culture using our optimized medium; 20.21 TD: T3 and Ctrl s Added Dex on Wednesday; MC: heart sections cultured in CTCM using our previously optimized medium; and MT: CTCM with T3 and Dex added to the medium. After 12 days of cultivation, the viability of MS and MT tissues remained the same as in fresh tissues assessed by MTT assay (Fig. 4b). Interestingly, the addition of T3 and Dex to transwell cultures (TD) did not result in a significant improvement in viability compared to Ctrl conditions, indicating an important role of mechanical stimulation in maintaining the viability of heart sections.
a Experimental design diagram depicting the four culture conditions used to evaluate the effects of mechanical stimulation and T3/Dex supplementation on medium for 12 days. b Bar graph shows quantification of viability 12 days post culture in all 4 culture conditions (Ctrl, TD, MC, and MT) compared to fresh heart slices (D0) (n = 18 (D0), 15 (D12 Ctrl, D12 TD and D12 MT), 12 (D12 MC) slices/group from different pigs, one-way ANOVA test is performed; ####p < 0.0001, ###p < 0.001 compared to D0 and **p < 0.01 compared to D12 Ctrl). b Bar graph shows quantification of viability 12 days post culture in all 4 culture conditions (Ctrl, TD, MC, and MT) compared to fresh heart slices (D0) (n = 18 (D0), 15 (D12 Ctrl, D12 TD and D12 MT), 12 (D12 MC) slices/group from different pigs, one-way ANOVA test is performed; ####p < 0.0001, ###p < 0.001 compared to D0 and **p < 0.01 compared to D12 ctrl). b Гистограмма показывает количественную оценку жизнеспособности через 12 дней после культивирования во всех 4 условиях культивирования (контроль, TD, MC и MT) по сравнению со свежими срезами сердца (D0) (n = 18 (D0), 15 (D12 Ctrl, D12 TD и D12 MT), 12 (D12 MC) срезов/группу от разных свиней, проводится односторонний тест ANOVA; ####p < 0,0001, ###p < 0,001 по сравнению с D0 и **p < 0,01 по сравнению с D12 Ctrl). b The bar graph shows the quantification of viability at 12 days post culture in all 4 culture conditions (control, TD, MC, and MT) compared to fresh heart sections (D0) (n = 18 (D0), 15 (D12 Ctrl, D12 TD, and D12 MT), 12 (D12 MC) sections/group from different pigs, one-way ANOVA test; ####p < 0.0001, ###p < 0.001 vs. D0 and **p < 0.01 by compared to D12 Ctrl). b 条形图显示所有4 种培养条件（Ctrl、TD、MC 和MT）与新鲜心脏切片(D0) (n = 18 (D0)、15 (D12 Ctrl、D12 TD 和D12 MT)，来自不同猪的12 (D12 MC) 切片/组，进行单向ANOVA 测试；####p < 0.0001，###p < 0.001 与D0 相比，**p < 0.01 与D12控制）。 b 4 12 (D12 MC) b Гистограмма, показывающая все 4 условия культивирования (контроль, TD, MC и MT) по сравнению со свежими срезами сердца (D0) (n = 18 (D0), 15 (D12 Ctrl, D12 TD и D12 MT), от разных свиней 12 (D12 MC) срезы/группа, односторонний тест ANOVA; ####p <0,0001, ###p <0,001 по сравнению с D0, **p <0,01 по сравнению с контролем D12). b Histogram showing all 4 culture conditions (control, TD, MC and MT) compared to fresh heart sections (D0) (n = 18 (D0), 15 (D12 Ctrl, D12 TD and D12 MT), from different pigs 12 (D12 MC) sections/group, one-way ANOVA test; ####p<0.0001, ###p<0.001 vs. D0, **p<0.01 vs. control D12). c Bar graph shows the quantification of glucose flux 12 days post culture in all 4 culture conditions (Ctrl, TD, MC, and MT) compared to fresh heart slices (D0) (n = 6 slices/group from different pigs, one-way ANOVA test is performed; ###p < 0.001, compared to D0 and ***p < 0.001 compared to D12 Ctrl). c Bar graph shows the quantification of glucose flux 12 days post culture in all 4 culture conditions (Ctrl, TD, MC, and MT) compared to fresh heart slices (D0) (n = 6 slices/group from different pigs, one-way ANOVA test is performed; ###p < 0.001, compared to D0 and ***p < 0.001 compared to D12 Ctrl). c Гистограмма показывает количественную оценку потока глюкозы через 12 дней после культивирования во всех 4 условиях культивирования (контроль, TD, MC и MT) по сравнению со свежими срезами сердца (D0) (n = 6 срезов/группу от разных свиней, односторонний Выполняется тест ANOVA; ###p < 0,001 по сравнению с D0 и ***p < 0,001 по сравнению с D12 Ctrl). c Histogram shows quantification of glucose flux 12 days post-culture under all 4 culture conditions (control, TD, MC and MT) compared to fresh heart sections (D0) (n = 6 sections/group from different pigs, one-way ANOVA test performed ; ###p < 0.001 compared to D0 and ***p < 0.001 compared to D12 Ctrl). c 条形图显示所有4 种培养条件（Ctrl、TD、MC 和MT）与新鲜心脏切片(D0) 相比，培养后12 天的葡萄糖通量定量（n = 6 片/组，来自不同猪，单向执行ANOVA 测试；###p < 0.001，与D0 相比，***p < 0.001 与D12 Ctrl 相比）。 C 条形图 显示 所有 4 种 条件 （（ctrl 、 td 、 mc 和 mt） 新鲜 心脏 切片 切片 切片 相比 ， 培养 后 后 12 天 的 通量 定量 （n = 6 片/组 ， 来自 猪 ， ， ， ， ， ， ， ， 猪 猪单向执行ANOVA 测试；###p < 0.001，与D0 相比，***p < 0.001 与D12 Ctrl 相比）。 c Гистограмма, показывающая количественную оценку потока глюкозы через 12 дней после культивирования для всех 4 условий культивирования (контроль, TD, MC и MT) по сравнению со свежими срезами сердца (D0) (n = 6 срезов/группа, от разных свиней, односторонний Были проведены тесты ANOVA; ###p < 0,001 по сравнению с D0, ***p < 0,001 по сравнению с D12 (контроль). c Histogram showing quantification of glucose flux at 12 days post-culture for all 4 culture conditions (control, TD, MC, and MT) compared to fresh heart sections (D0) (n = 6 sections/group, from different pigs, unilateral Were ANOVA tests were performed, ###p < 0.001 compared to D0, ***p < 0.001 compared to D12 (control). d Strain analysis plots of fresh (blue), day 12 MC (green), and day 12 MT (red) tissues at ten regional tissue section points (n = 4 slices/group, one-way ANOVA test; there was no significant difference between groups). e Volcano plot showing differentially expressed genes in fresh heart sections (D0) compared to heart sections cultured under static conditions (Ctrl) or under MT conditions (MT) for 10-12 days. f Heatmap of sarcomere genes for heart sections cultured under each of the culture conditions. Error bars represent the mean ± standard deviation.
Metabolic dependence on the switch from fatty acid oxidation to glycolysis is a hallmark of cardiomyocyte dedifferentiation. Immature cardiomyocytes primarily use glucose for ATP production and have hypoplastic mitochondria with few cristae5,32. Glucose utilization analyzes showed that under MC and MT conditions, glucose utilization was similar to that in day 0 tissues (Figure 4c). However, Ctrl samples showed a significant increase in glucose utilization compared to fresh tissue. This indicates that the combination of CTCM and T3/Dex enhances tissue viability and preserves the metabolic phenotype of 12-day cultured heart sections. In addition, strain analysis showed that strain levels remained the same as in fresh heart tissue for 12 days under MT and MS conditions (Fig. 4d).
To analyze the overall impact of CTCM and T3/Dex on the global transcriptional landscape of cardiac slice tissue, we performed RNAseq on cardiac slices from all four different culture conditions (Supplementary Data 1). Interestingly, MT sections showed high transcriptional similarity to fresh heart tissue, with only 16 differentially expressed out of 13,642 genes. However, as we showed earlier, Ctrl slices displayed 1229 differentially expressed genes after 10–12 days in culture (Fig. 4e). These data were confirmed by qRT-PCR of heart and fibroblast genes (Supplementary Fig. 7a-c). Interestingly, the Ctrl sections showed downregulation of cardiac and cell cycle genes and activation of inflammatory gene programs. These data suggest that dedifferentiation, which normally occurs after long-term culturing, is completely attenuated under MT conditions (Supplementary Fig. 8a,b). Careful study of sarcomere genes showed that only under MT conditions are the genes encoding the sarcomere (Fig. 4f) and ion channel (Supplementary Fig. 9) preserved, protecting them from suppression under Ctrl, TD, and MC conditions. These data demonstrate that with a combination of mechanical and humoral stimulation (T3/Dex), the heart slice transcriptome can remain similar to fresh heart slices after 12 days in culture.
These transcriptional findings are supported by the fact that the structural integrity of cardiomyocytes in heart sections is best preserved under MT conditions for 12 days, as shown by intact and localized connexin 43 (Fig. 5a). In addition, fibrosis in heart sections under MT conditions was significantly reduced compared to Ctrl and similar to fresh heart sections (Fig. 5b). These data demonstrate that the combination of mechanical stimulation and T3/Dex treatment effectively preserves cardiac structure in heart sections in culture.
a Representative immunofluorescence images of troponin-T (green), connexin 43 (red), and DAPI (blue) in freshly isolated heart sections (D0) or cultured for 12 days in all four heart section culture conditions (scale bar = 100 µm). ). Artificial intelligence quantification of the heart tissue structural integrity (n = 7 (D0 and D12 Ctrl), 5 (D12 TD, D12 MC and D12 MT) slices/group from different pigs, one-way ANOVA test is performed; ####p < 0.0001 compared to D0 and *p < 0.05, or ****p < 0.0001 compared to D12 Ctrl). Artificial intelligence quantification of the heart tissue structural integrity (n = 7 (D0 and D12 Ctrl), 5 (D12 TD, D12 MC and D12 MT) slices/group from different pigs, one-way ANOVA test is performed; #### p < 0.0001 compared to D0 and *p < 0.05, or ****p < 0.0001 compared to D12 Ctrl). Количественная оценка структурной целостности ткани сердца с помощью искусственного интеллекта (n = 7 (D0 и D12 Ctrl), 5 (D12 TD, D12 MC и D12 MT) срезов/группу от разных свиней, проведен однофакторный тест ANOVA; #### p < 0,0001 по сравнению с D0 и *p < 0,05 или ****p < 0,0001 по сравнению с D12 Ctrl). Quantification of the structural integrity of heart tissue using artificial intelligence (n = 7 (D0 and D12 Ctrl), 5 (D12 TD, D12 MC and D12 MT) sections/group from different pigs, one-way ANOVA test performed; #### p < 0.0001 compared to D0 and *p < 0.05 or ****p < 0.0001 compared to D12 Ctrl).对不同猪的心脏组织结构完整性（n = 7（D0 和D12 Ctrl）、5（D12 TD、D12 MC 和D12 MT）切片/组）进行人工智能量化，进行单向ANOVA 测试；#### p < 0.0001 与D0 和*p < 0.05 相比，或****p < 0.0001 与D12 Ctrl 相比）。对 不同 猪 的 心脏 结构 完整性 （n = 7 （d0 和 d12 ctrl） （5 （d12 td 、 d12 mc 和 d12 mt） 组） 人工 智能量 化 进行 单向 单向 单向 测试 ； ########## p < 0.0001 与D0 和*p < 0.05 相比，或****p < 0.0001 与D12 Ctrl 相比）。 Quantification of the structural integrity of cardiac tissue using artificial intelligence in different pigs (n = 7 (D0 and D12 Ctrl), 5 (D12 TD, D12 MC and D12 MT) sections/group) with a one-way ANOVA test; #### p < 0,0001 по сравнению с D0 и *p < 0,05 или ****p < 0,0001 по сравнению с D12 Ctrl). #### p < 0.0001 compared to D0 and *p < 0.05 or ****p < 0.0001 compared to D12 Ctrl). b Representative images and quantification for heart slices stained with Masson’s trichrome stain (Scale bar = 500 µm) (n = 10 (D0, D12 Ctrl, D12 TD, and D12 MC), 9 (D12 MT) slices/group from different pigs, one-way ANOVA test is performed; ####p < 0.0001 compared to D0 and ***p < 0.001, or ****p < 0.0001 compared to D12 Ctrl). b Representative images and quantification for heart slices stained with Masson’s trichrome stain (Scale bar = 500 µm) (n = 10 (D0, D12 Ctrl, D12 TD, and D12 MC), 9 (D12 MT) slices/group from different pigs, one-way ANOVA test is performed; ####p < 0.0001 compared to D0 and ***p < 0.001, or ****p < 0.0001 compared to D12 Ctrl). b Репрезентативные изображения и количественная оценка срезов сердца, окрашенных трихромным красителем Массона (масштабная линейка = 500 мкм) (n = 10 (D0, D12 Ctrl, D12 TD и D12 MC), 9 (D12 MT) срезов/группу от разных свиней, выполняется односторонний тест ANOVA; ####p < 0,0001 по сравнению с D0 и ***p < 0,001 или ****p < 0,0001 по сравнению с D12 Ctrl). b Representative images and quantification of heart sections stained with Masson’s trichrome stain (scale bar = 500 µm) (n = 10 (D0, D12 Ctrl, D12 TD and D12 MC), 9 (D12 MT) sections/group from different pigs, performed one-way ANOVA; ####p < 0.0001 vs. D0 and ***p < 0.001 or ****p < 0.0001 vs. D12 Ctrl). b 用Masson 三色染料染色的心脏切片的代表性图像和量化（比例尺= 500 µm）（n = 10（D0、D12 Ctrl、D12 TD 和D12 MC），来自不同猪的9 个（D12 MT）切片/组，进行单因素方差分析；####p < 0.0001 与D0 相比，***p < 0.001，或****p < 0.0001 与D12 Ctrl 相比）。 b 用 masson 三 色 染料 的 心脏 切片 的 代表性 和 量化 （比例 尺 尺 尺 = 500 µm） （n = 10 （d0 、 d12 ctrl 、 d12 td 和 d12 mc） 来自 不同 的 9 个 d12 mt 切片 切片 切片 切片 切片 切片 切片 切片 切片 切片 切片 切片 切片 切片 切片 切片 切片 切片/组，进行单因素方差分析；####p < 0.0001 与D0 相比，***p < 0.001，或****p < 0.0001 与D12 Ctrl 相比）。 b Репрезентативные изображения и количественная оценка срезов сердца, окрашенных трихромом Массона (масштабная линейка = 500 мкм) (n = 10 (D0, D12 Ctrl, D12 TD и D12 MC), 9 (D12 MT) срезов от разных свиней / группы, один- способ ANOVA; ####p < 0,0001 по сравнению с D0, ***p < 0,001 или ****p < 0,0001 по сравнению с D12 Ctrl). b Representative images and quantification of heart sections stained with Masson’s trichrome (scale bar = 500 µm) (n = 10 (D0, D12 Ctrl, D12 TD and D12 MC), 9 (D12 MT) sections from different pigs/group, one ANOVA method; ####p < 0.0001 compared to D0, ***p < 0.001 or ****p < 0.0001 compared to D12 Ctrl). Error bars represent the mean ± standard deviation.
Finally, the ability of CTCM to mimic cardiac hypertrophy was assessed by increasing cardiac tissue stretch. In CTCM, peak air chamber pressure increased from 80 mmHg to 80 mmHg. Art. (normal stretch) up to 140 mmHg Art. (Fig. 6a). This corresponds to a 32% increase in stretch (Fig. 6b), which was previously shown as the corresponding percentage stretch required for heart sections to achieve a sarcomere length similar to that seen in hypertrophy. Stretch and velocity of cardiac tissue during contraction and relaxation remained constant during six days of culture (Fig. 6c). Cardiac tissue from MT conditions was subjected to normal stretch (MT (Normal)) or overstretch conditions (MT (OS)) for six days. Already after four days in culture, the hypertrophic biomarker NT-ProBNP was significantly elevated in the medium under MT (OS) conditions compared to MT (normal) conditions (Fig. 7a). In addition, after six days of culturing, the cell size in MT (OS) (Fig. 7b) significantly increased compared to sections of MT heart (normal). In addition, NFATC4 nuclear translocation was significantly increased in overstretched tissues (Fig. 7c). These results show the progressive development of pathological remodeling after hyperdistension and support the concept that the CTCM device can be used as a platform to study stretch-induced cardiac hypertrophy signaling.
Representative traces of air chamber pressure, fluid chamber pressure, and tissue movement measurements confirm that chamber pressure changes fluid chamber pressure, causing a corresponding movement of the tissue slice. b Representative stretch percentage and stretch rate curves for normally stretched (orange) and overstretched (blue) tissue sections. c Bar graph showing cycle time (n = 19 slices per group, from different pigs), contraction time (n = 18-19 slices per group, from different pigs), relaxation time (n = 19 slices per group, from different pigs) ), amplitude of tissue movement (n = 14 slices/group, from different pigs), peak systolic velocity (n = 14 slices/group, from different pigs) and peak relaxation rate (n = 14 (D0), 15 (D6) ) sections/groups) from different pigs), two-tailed Student’s t-test showed no significant difference in any parameter, indicating that these parameters remained constant during 6 days of culture with overvoltage. Error bars represent the mean ± standard deviation.
a Bar graph quantification of NT-ProBNP concentration in culture media from heart slices cultured under MT normal stretch (Norm) or overstretching (OS) conditions (n = 4 (D2 MTNorm), 3 (D2 MTOS, D4 MTNorm, and D4 MTOS) slices/group from different pigs, Two-way ANOVA is performed; **p < 0.01 compared to normal stretch). a Bar graph quantification of NT-ProBNP concentration in culture media from heart slices cultured under MT normal stretch (Norm) or overstretching (OS) conditions (n ​​= 4 (D2 MTNorm), 3 (D2 MTOS, D4 MTNorm, and D4 MTOS) slices/group from different pigs, Two-way ANOVA is performed; **p < 0.01 compared to normal stretch). Quantitative histogram of NT-ProBNP concentration in culture medium from heart slices cultured under conditions of normal MT stretch (norm) or overstretch (OS) (n = 4 (D2 MTNorm), 3 (D2 MTOS, D4 MTNorm, and D4).MTOS) slices /group from different pigs, two-factor analysis of variance is performed; **p < 0,01 по сравнению с нормальным растяжением). **p < 0.01 compared to normal stretch). a 在MT 正常拉伸(Norm) 或过度拉伸(OS) 条件下培养的心脏切片培养基中NT-ProBNP 浓度的条形图量化（n = 4 (D2 MTNorm)、3（D2 MTOS、D4 MTNorm 和D4 MTOS）来自不同猪的切片/组，进行双向方差分析；**与正常拉伸相比，p < 0.01）。 a Quantification of NT-ProBNP concentration in heart slices cultured under MT normal stretch (Norm) or overstretch (OS) conditions (n ​​= 4 (D2 MTNorm), 3 (D2 MTOS, D4 MTNorm和D4 MTOS) from different 猪的切片/组，可以双向方方发发动; **compared with normal stretching, p < 0.01). histogram Quantification of NT-ProBNP concentrations in heart slices cultured under conditions of normal MT stretch (norm) or overstretch (OS) (n = 4 (D2 MTNorm), 3 (D2 MTOS, D4 MTNorm) and D4 MTOS) slices/group from different pigs, two-way analysis of variance; **p < 0,01 по сравнению с нормальным растяжением). **p < 0.01 compared to normal stretch). b Representative images for heart slices stained with troponin-T and WGA (left) and cell size quantification (right) (n = 330 (D6 MTOS), 369 (D6 MTNorm) cells/group from 10 different slices from different pigs, Two-tailed Student t-test is performed; ****p < 0.0001 compared to normal stretch). b Representative images for heart slices stained with troponin-T and WGA (left) and cell size quantification (right) (n = 330 (D6 MTOS), 369 (D6 MTNorm) cells/group from 10 different slices from different pigs, Two- tailed Student t-test is performed; ****p < 0.0001 compared to normal stretch). b Репрезентативные изображения срезов сердца, окрашенных тропонином-Т и АЗП (слева) и количественного определения размера клеток (справа) (n = 330 (D6 MTOS), 369 (D6 MTNorm) клеток/группу из 10 разных срезов от разных свиней, два- проводится хвостовой t-критерий Стьюдента; ****p < 0,0001 по сравнению с нормальным растяжением). b Representative images of heart sections stained with troponin-T and AZP (left) and cell size quantification (right) (n = 330 (D6 MTOS), 369 (D6 MTNorm) cells/group from 10 different sections from different pigs, two- tailed Student’s t-test was performed; ****p < 0.0001 compared to normal strain). b 用肌钙蛋白-T 和WGA（左）和细胞大小量化（右）染色的心脏切片的代表性图像（n = 330（D6 MTOS），来自不同猪的10 个不同切片的369（D6 MTNorm）细胞/组，两进行有尾学生t 检验；与正常拉伸相比，****p < 0.0001）。 b Representative images of heart slices stained with calcarein-T and WGA (left) and cell size (right) (n = 330 (D6 MTOS), 369 from 10 different slices (D6 MTNorm)) Cells/组，两方法有尾学生t test；compared with normal stretching，****p < 0.0001). b Репрезентативные изображения срезов сердца, окрашенных тропонином-Т и АЗП (слева) и количественная оценка размера клеток (справа) (n = 330 (D6 MTOS), 369 (D6 MTNorm) из 10 различных срезов от разных свиней Клетки/группа, двусторонние критерий Стьюдента; ****p < 0,0001 по сравнению с нормальным растяжением). b Representative images of heart sections stained with troponin-T and AZP (left) and quantification of cell size (right) (n = 330 (D6 MTOS), 369 (D6 MTNorm) from 10 different sections from different pigs) Cells/group, two-tailed criterion Student’s t; ****p < 0.0001 compared to normal strain). c Representative images for day 0 and day 6 MTOS heart slices immunolabeled for troponin-T and NFATC4 and quantification of the translocation of NFATC4 to the nuclei of CMs (n = 4 (D0), 3 (D6 MTOS) slices/group from different pigs, Two-tailed Student t-test is performed; *p < 0.05). c Representative images for day 0 and day 6 MTOS heart slices immunolabeled for troponin-T and NFATC4 and quantification of the translocation of NFATC4 to the nuclei of CMs (n = 4 (D0), 3 (D6 MTOS) slices/group from different pigs , Two-tailed Student t-test is performed; *p < 0.05). c Репрезентативные изображения для срезов сердца 0 и 6 дней MTOS, иммуномеченых для тропонина-Т и NFATC4, и количественная оценка транслокации NFATC4 в ядра кавернозных клеток (n = 4 (D0), 3 (D6 MTOS) срезов/группу от разных свиней , выполняется двусторонний t-критерий Стьюдента; *p < 0,05). c Representative images for heart sections at 0 and 6 days MTOS, immunolabeled for troponin-T and NFATC4, and quantification of NFATC4 translocation in the nucleus of cavernous cells (n = 4 (D0), 3 (D6 MTOS) slices/group from different pigs) performed two-tailed Student’s t-test; *p < 0.05). c 用于肌钙蛋白-T 和NFATC4 免疫标记的第0 天和第6 天MTOS 心脏切片的代表性图像，以及来自不同猪的NFATC4 易位至CM 细胞核的量化（n = 4 (D0)、3 (D6 MTOS) 切片/组, 进行双尾学生t 检验；*p < 0.05)。 c Representative images of calcanin-T and NFATC4 immunolabeling 第0天和第6天MTOS heart slices, and NFATC4 from different NFATC4 易位至CM cell nucleus的quantity化 (n = 4 (D0), 3 (D6 MTOS) 切礼/组, 时间双尾学生et 电影；*p < 0.05). c Репрезентативные изображения срезов сердца MTOS на 0 и 6 день для иммуномаркировки тропонином-Т и NFATC4 и количественная оценка транслокации NFATC4 в ядра CM от разных свиней (n = 4 (D0), 3 (D6 MTOS) срез/группа, два- хвостатый t-критерий Стьюдента; *p < 0,05). c Representative images of MTOS heart slices at day 0 and 6 for troponin-T and NFATC4 immunolabeling and quantification of NFATC4 translocation in the nucleus of CM from different pigs (n = 4 (D0), 3 (D6 MTOS) slices/group, two-tailed t -criterion Student’s; *p < 0.05). Error bars represent mean ± standard deviation.
Translational cardiovascular research requires cellular models that accurately reproduce the cardiac environment. In this study, a CTCM device was developed and characterized that can stimulate ultrathin sections of the heart. The CTCM system includes physiologically synchronized electromechanical stimulation and T3 and Dex fluid enrichment. When porcine heart sections were exposed to these factors, their viability, structural integrity, metabolic activity, and transcriptional expression remained the same as in fresh heart tissue after 12 days of culture. In addition, excessive stretching of cardiac tissue can cause hypertrophy of the heart caused by hyperextension. Overall, these results support the critical role of physiological culture conditions in maintaining a normal cardiac phenotype and provide a platform for drug screening.
Many factors contribute to creating an optimal environment for the functioning and survival of cardiomyocytes. The most obvious of these factors are related to (1) intercellular interactions, (2) electromechanical stimulation, (3) humoral factors, and (4) metabolic substrates. Physiological cell-to-cell interactions require complex three-dimensional networks of multiple cell types supported by an extracellular matrix. Such complex cellular interactions are difficult to reconstruct in vitro by co-culture of individual cell types, but can be easily achieved using the organotypic nature of heart sections.
Mechanical stretch and electrical stimulation of cardiomyocytes are critical for maintaining cardiac phenotype33,34,35. While mechanical stimulation has been widely used for hiPSC-CM conditioning and maturation, several elegant studies have recently attempted mechanical stimulation of heart slices in culture using uniaxial loading. These studies show that 2D uniaxial mechanical loading has a positive effect on the phenotype of the heart during culture. In these studies, sections of the heart were either loaded with isometric tensile forces17, linear auxotonic loading18, or the cardiac cycle was recreated using force transducer feedback and tension drives. However, these methods use uniaxial tissue stretch without environmental optimization, resulting in the suppression of many cardiac genes or overexpression of genes associated with abnormal stretch responses. The CTCM described here provides a 3D electromechanical stimulus that mimics the natural cardiac cycle in terms of cycle time and physiological stretch (25% stretch, 40% systole, 60% diastole, and 72 beats per minute). Although this three-dimensional mechanical stimulation alone is not sufficient to maintain tissue integrity, a combination of humoral and mechanical stimulation using T3/Dex is required to adequately maintain tissue viability, function, and integrity.
Humoral factors play an important role in modulating the adult heart phenotype. This was highlighted in HiPS-CM studies in which T3 and Dex were added to culture media to accelerate cell maturation. T3 may influence the transport of amino acids, sugars and calcium across cell membranes36. In addition, T3 promotes MHC-α expression and MHC-β downregulation, promoting the formation of fast twitch myofibrils in mature cardiomyocytes compared to slow twitch myofibrils in fetal CM. T3 deficiency in hypothyroid patients results in the loss of myofibrillar bands and a reduced rate of tone development37. Dex acts on glucocorticoid receptors and has been shown to increase myocardial contractility in isolated perfused hearts; 38 this improvement is thought to be related to the effect on calcium deposit-driven entry (SOCE) 39,40. In addition, Dex binds to its receptors, causing a broad intracellular response that suppresses immune function and inflammation30.
Our results indicate that physical mechanical stimulation (MS) improved overall culture performance compared to Ctrl, but failed to maintain viability, structural integrity, and cardiac expression over 12 days in culture. Compared to Ctrl, addition of T3 and Dex to CTCM (MT) cultures improved viability and maintained similar transcription profiles, structural integrity, and metabolic activity with fresh heart tissue for 12 days. In addition, by controlling the degree of tissue stretch, a hyperextension-induced cardiac hypertrophy model was created using STCM, illustrating the versatility of the STCM system. It should be noted that although cardiac remodeling and fibrosis usually involve intact organs whose circulating cells can provide the appropriate cytokines as well as phagocytosis and other remodeling factors, sections of the heart can still mimic the fibrotic process in response to stress and trauma. into myofibroblasts. This has been previously evaluated in this cardiac slice model. It should be noted that CTCM parameters can be modulated by changing pressure/electrical amplitude and frequency to simulate many conditions such as tachycardia, bradycardia, and mechanical circulatory support (mechanical unloaded heart). This makes the system a medium throughput for drug testing. The ability of CTCM to model overexertion-induced cardiac hypertrophy paves the way for testing this system for personalized therapy. In conclusion, the present study demonstrates that mechanical stretch and humoral stimulation are critical for maintaining the culture of cardiac tissue sections.
Although the data presented here suggest that CTCM is a very promising platform for modeling intact myocardium, this culture method has some limitations. The main limitation of CTCM culture is that it imposes continuous dynamic mechanical stresses on the slices, which precludes the ability to actively monitor cardiac slice contractions during each cycle. In addition, due to the small size of cardiac sections (7 mm), the ability to evaluate systolic function outside of culture systems using traditional force sensors is limited. In the current manuscript, we partially overcome this limitation by evaluating optical voltage as an indicator of contractile function. However, this limitation will require further work and may be addressed in the future by introducing methods for optical monitoring of the function of heart slices in culture, such as optical mapping using calcium and voltage-sensitive dyes. Another limitation of CTCM is that the working model does not manipulate physiological stress (preload and afterload). In CTCM, pressure was induced in opposite directions to reproduce 25% physiological stretch in diastole (full stretch) and systole (length of contraction during electrical stimulation) in very large tissues. This limitation should be removed in future CTCM designs by adequate pressure on the cardiac tissue from both sides and by applying the exact pressure-volume relationships that occur in the chambers of the heart.
The overstretch-induced remodeling reported in this manuscript is limited to mimicking hypertrophic hyperstretch signals. Thus, this model can help in the study of stretch-induced hypertrophic signaling without the need for humoral or neural factors (which do not exist in this system). Further studies are needed to increase the multiplicity of CTCM, for example, co-culturing with immune cells, circulating plasma humoral factors, and innervation when co-culturing with neuronal cells will improve the possibilities of disease modeling with CTCM.
Thirteen pigs were used in this study. All animal procedures were performed in accordance with institutional guidelines and were approved by the University of Louisville Institutional Animal Care and Use Committee. The aortic arch was clamped and the heart was perfused with 1 L of sterile cardioplegia (110 mM NaCl, 1.2 mM CaCl2, 16 mM KCl, 16 mM MgCl2, 10 mM NaHCO3, 5 U/mL heparin, pH up to 7.4); the hearts were preserved in ice-cold cardioplegic solution until transported to the lab on ice which is usually <10 min. the hearts were preserved in ice-cold cardioplegic solution until transported to the lab on ice which is usually <10 min. сердца хранили в ледяном кардиоплегическом растворе до транспортировки в лабораторию на льду, что обычно занимает <10 мин. hearts were stored in ice-cold cardioplegic solution until transport to the laboratory on ice, which usually takes <10 min.将心脏保存在冰冷的心脏停搏液中，直到冰上运送到实验室，通常<10分钟。将心脏保存在冰冷的心脏停搏液中，直到冰上运送到实验室，通常<10分钟。 Держите сердца в ледяной кардиоплегии до транспортировки в лабораторию на льду, обычно <10 мин. Keep hearts on ice cardioplegia until transport to the laboratory on ice, usually <10 min.
The CTCM device was developed in SolidWorks computer-aided design (CAD) software. The culture chambers, dividers and air chambers are made of CNC clear acrylic plastic. The 7mm diameter back-up ring is made of high density polyethylene (HDPE) in the center and has an o-ring groove to accommodate the silicone o-ring used to seal the media underneath. A thin silica membrane separates the culture chamber from the separation plate. The silicone membrane is laser cut from 0.02″ thick silicone sheet and has a hardness of 35A. The bottom and top silicone gaskets are laser cut from 1/16″ thick silicone sheet and have a hardness of 50A. 316L stainless steel screws and wing nuts are used for fastening the block and creating an airtight seal.
A dedicated printed circuit board (PCB) is designed to be integrated with the C-PACE-EM system. The swiss machine connector sockets on the PCB are connected to graphite electrodes by silver-plated copper wires and bronze 0-60 screws screwed into the electrodes. The printed circuit board is placed in the cover of the 3D printer.
The CTCM device is controlled by a programmable pneumatic actuator (PPD) that creates a controlled circulatory pressure similar to a cardiac cycle. As the pressure inside the air chamber increases, the flexible silicone membrane expands upward, forcing the medium under the tissue site. The area of ​​tissue will then be stretched by this expulsion of fluid, mimicking the physiological expansion of the heart during diastole. At the peak of relaxation, electrical stimulation was applied through graphite electrodes, which reduced the pressure in the air chamber and caused contraction of tissue sections. Inside the pipe is a hemostatic valve with a pressure sensor to detect the pressure in the air system. The pressure sensed by the pressure sensor is applied to a data collector connected to the laptop. This allows continuous monitoring of the pressure inside the gas chamber. When the maximum chamber pressure was reached (standard 80 mmHg, 140 mmHg OS), the data acquisition device was ordered to send a signal to the C-PACE-EM system to generate a biphasic voltage signal for 2 ms, set to 4 V.
Heart sections were obtained and culture conditions in 6 wells were performed as follows: Transfer the harvested hearts from the transfer vessel to a tray containing cold (4° C.) cardioplegia. The left ventricle was isolated with a sterile blade and cut into pieces of 1-2 cm3. These tissue blocks were attached to tissue supports with tissue adhesive and placed in a vibrating microtome tissue bath containing Tyrode’s solution and continuously oxygenated (3 g/L 2,3-butanedione monooxime (BDM), 140 mM NaCl (8.18 g) . ), 6 mM KCl (0.447 g), 10 mM D-glucose (1.86 g), 10 mM HEPES (2.38 g), 1 mM MgCl2 (1 ml 1 M solution), 1.8 mM CaCl2 ( 1.8 ml 1 M solution), up to 1 L ddH2O). The vibrating microtome was set to cut 300 µm thick slices at a frequency of 80 Hz, a horizontal vibration amplitude of 2 mm, and an advance rate of 0.03 mm/s. The tissue bath was surrounded by ice to keep the solution cool and the temperature was maintained at 4°C. Transfer tissue sections from the microtome bath to an incubation bath containing continuously oxygenated Tyrode solution on ice until enough sections are obtained for one culture plate. For transwell cultures, tissue sections were attached to sterile 6 mm wide polyurethane supports and placed in 6 ml of optimized medium (199 medium, 1x ITS supplement, 10% FBS, 5 ng/ml VEGF, 10 ng/ml FGF-alkaline and 2X antibiotic-antifungal ). Electrical stimulation (10 V, frequency 1.2 Hz) was applied to the tissue sections through the C-Pace. For TD conditions, fresh T3 and Dex were added at 100 nM and 1 μM at each medium change. The medium is saturated with oxygen before replacement 3 times a day. Tissue sections were cultured in an incubator at 37°C and 5% CO2.
For CTCM cultures, tissue sections were placed on a custom-made 3D printer in a Petri dish containing modified Tyrode’s solution. The device is designed to increase the size of the slice of the heart by 25% of the area of ​​the support ring. This is done so that the sections of the heart do not stretch after being transferred from Tyrode’s solution to the medium and during diastole. Using histoacrylic glue, sections 300 µm thick were fixed on a support ring 7 mm in diameter. After attaching the tissue sections to the support ring, cut off the excess tissue sections and place the attached tissue sections back into the bath of Tyrode solution on ice (4°C) until enough sections have been prepared for one device. The total processing time for all devices should not exceed 2 hours. After 6 tissue sections were attached to their support rings, the CTCM device was assembled. The CTCM culture chamber is pre-filled with 21 ml pre-oxygenated medium. Transfer the tissue sections to the culture chamber and carefully remove any air bubbles with a pipette. The tissue section is then guided into the hole and gently pressed into place. Finally, place the electrode cap on the device and transfer the device to the incubator. Then connect the CTCM to the air tube and the C-PACE-EM system. The pneumatic actuator opens and the air valve opens the CTCM. The C-PACE-EM system was configured to deliver 4 V at 1.2 Hz during biphasic pacing for 2 ms. The medium was changed twice a day and the electrodes were changed once a day to avoid accumulation of graphite on the electrodes. If necessary, tissue sections can be removed from their culture wells to expel any air bubbles that may have fallen under them. For MT treatment conditions, T3/Dex was added fresh with each medium change with 100 nM T3 and 1 μM Dex. The CTCM devices were cultured in an incubator at 37°C and 5% CO2.
To obtain stretched trajectories of heart slices, a special camera system was developed. A SLR camera (Canon Rebel T7i, Canon, Tokyo, Japan) was used with a Navitar Zoom 7000 18-108mm macro lens (Navitar, San Francisco, CA). Visualization was performed at room temperature after replacing the medium with fresh medium. The camera is positioned at a 51° angle and video is recorded at 30 frames per second. First, open source software (MUSCLEMOTION43) was used with Image-J to quantify the motion of heart slices. The mask was created using MATLAB (MathWorks, Natick, MA, USA) to define regions of interest for beating heart slices to avoid noise. Manually segmented masks are applied to all images in a frame sequence and then passed to the MUSCLEMOTION plug-in. Muscle Motion uses the average intensity of the pixels in each frame to quantify its movement relative to the reference frame. Data were recorded, filtered and used to quantify cycle time and assess tissue stretch during the cardiac cycle. The recorded video was post-processed using a first-order zero-phase digital filter. To quantify tissue stretch (peak-to-peak), peak-to-peak analysis was performed to distinguish between peaks and troughs in the recorded signal. In addition, detrending is performed using a 6th order polynomial to eliminate signal drift. Program code was developed in MATLAB to determine global tissue motion, cycle time, relaxation time, and contraction time (Supplementary Program Code 44).
For strain analysis, using the same videos created for mechanical stretch assessment, we first traced two images representing motion peaks (the highest (upper) and lowest (lower) points of motion) according to the MUSCLEMOTION software. We then segmented the tissue regions and applied a form of shading algorithm to the segmented tissue (Supplementary Fig. 2a). The segmented tissue was then divided into ten subsurfaces, and the stress on each surface was calculated using the following equation: Strain = (Sup-Sdown)/Sdown, where Sup and Sdown are the distances of the shape from the top and bottom shadows of the fabric, respectively (Supplementary Fig. .2b).
Heart sections were fixed in 4% paraformaldehyde for 48 hours. Fixed tissues were dehydrated in 10% and 20% sucrose for 1 h, then in 30% sucrose overnight. The sections were then embedded in optimum cutting temperature compound (OCT compound) and gradually frozen in an isopentane/dry ice bath. Store OCT embedding blocks at -80 °C until separation. Slides were prepared as sections with a thickness of 8 μm.
To remove OCT from heart sections, heat the slides on a heating block at 95 °C for 5 min. Add 1 ml PBS to each slide and incubate for 30 minutes at room temperature, then permeate the sections by setting 0.1% Triton-X in PBS for 15 minutes at room temperature. To prevent non-specific antibodies from binding to the sample, add 1 ml of 3% BSA solution to slides and incubate for 1 hour at room temperature. The BSA was then removed and the slides were washed with PBS. Mark each sample with a pencil. Primary antibodies (diluted 1:200 in 1% BSA) (connexin 43 (Abcam; #AB11370), NFATC4 (Abcam; #AB99431) and troponin-T (Thermo Scientific; #MA5-12960) were added over 90 minutes, then secondary antibodies (diluted 1:200 in 1% BSA) against mouse Alexa Fluor 488 (Thermo Scientific; #A16079), against rabbit Alexa Fluor 594 (Thermo Scientific; #T6391) for an additional 90 minutes Washed 3 times with PBS To distinguish target staining from background, we used only the secondary antibody as a control. Finally, DAPI nuclear stain was added and the slides were placed in a vectashield (Vector Laboratories) and sealed with nail polish. -x magnification) and Keyence microscope with 40x magnification.
WGA-Alexa Fluor 555 (Thermo Scientific; #W32464) at 5 μg/ml in PBS was used for WGA staining and applied to fixed sections for 30 minutes at room temperature. The slides were then washed with PBS and Sudan black was added to each slide and incubated for 30 minutes. The slides were then washed with PBS and vectashield embedding medium was added. Slides were visualized on a Keyence microscope at 40x magnification.
OCT was removed from the samples as described above. After removing the OCT, immerse the slides in Bouin’s solution overnight. The slides were then rinsed with distilled water for 1 hour and then placed in a Bibrich aloe acid fuchsin solution for 10 minutes. Then the slides were washed with distilled water and placed in a solution of 5% phosphomolybdenum/5% phosphotungstic acid for 10 minutes. Without rinsing, transfer the slides directly into the aniline blue solution for 15 minutes. Then the slides were washed with distilled water and placed in a 1% acetic acid solution for 2 minutes. Slides were dried in 200 N ethanol and transferred to xylene. Stained slides were visualized using a Keyence microscope with a 10x objective. Fibrosis area percentage was quantified using Keyence Analyzer software.
CyQUANT™ MTT Cell Viability Assay (Invitrogen, Carlsbad, CA), catalog number V13154, according to the manufacturer’s protocol with some modifications. In particular, a surgical punch with a diameter of 6 mm was used to ensure uniform tissue size during MTT analysis. Tissues were individually plated into the wells of a 12-well plate containing MTT substrate according to the manufacturer’s protocol. The sections are incubated at 37° C. for 3 hours and the living tissue metabolizes the MTT substrate to form a purple formazan compound. Replace the MTT solution with 1 ml DMSO and incubate at 37 °C for 15 minutes to extract purple formazan from heart sections. Samples were diluted 1:10 in DMSO in 96-well clear bottom plates and purple color intensity measured at 570 nm using a Cytation plate reader (BioTek). Readings were normalized to the weight of each slice of the heart.
Heart slice media was replaced with media containing 1 μCi/ml [5-3H]-glucose (Moravek Biochemicals, Brea, CA, USA) for glucose utilization assay as previously described. After 4 hours of incubation, add 100 µl of medium to an open microcentrifuge tube containing 100 µl of 0.2 N HCl. Then the tube was placed in a scintillation tube containing 500 μl of dH2O to evaporate [3H]2O for 72 hours at 37°C. Then remove the microcentrifuge tube from the scintillation tube and add 10 ml of scintillation fluid. Scintillation counts were performed using a Tri-Carb 2900TR liquid scintillation analyzer (Packard Bioscience Company, Meriden, CT, USA). Glucose utilization was then calculated taking into account [5-3H]-glucose specific activity, incomplete equilibrium and background, dilution of [5-3H]-to unlabeled glucose, and scintillation counter efficiency. The data are normalized to the mass of sections of the heart.
After tissue homogenization in Trizol, RNA was isolated from heart sections using the Qiagen miRNeasy Micro Kit #210874 according to the manufacturer’s protocol. RNAsec library preparation, sequencing and data analysis were performed as follows:
1 μg of RNA per sample was used as starting material for the preparation of the RNA library. Sequencing libraries were generated using the NEBNext UltraTM RNA Library Preparation Kit for Illumina (NEB, USA) following the manufacturer’s recommendations, and index codes were added to the attribute sequences for each sample. Briefly, mRNA was purified from total RNA using magnetic beads attached with poly-T oligonucleotides. Fragmentation is carried out using divalent cations at high temperature in NEBNext First Strand Synthesis Reaction Buffer (5X). First strand cDNA was synthesized using random hexamer primers and M-MuLV reverse transcriptase (RNase H-). The second strand cDNA is then synthesized using DNA polymerase I and RNase H. The remaining overhangs are converted to blunt ends by exonuclease/polymerase activity. After adenylation of the 3′ end of the DNA fragment, a NEBNext Adapter with a hairpin loop structure is attached to it to prepare it for hybridization. For selection of cDNA fragments of preferred length 150-200 bp. library fragments were purified using the AMPure XP system (Beckman Coulter, Beverly, USA). Then, 3 μl USER Enzyme (NEB, USA) with size-selected cDNA ligated with an adapter was used for 15 minutes at 37°C and then for 5 minutes at 95°C before PCR. PCR was then performed using Phusion High-Fidelity DNA polymerase, universal PCR primers, and Index (X) primers. Finally, PCR products were purified (AMPure XP system) and library quality assessed on an Agilent Bioanalyzer 2100 system. The cDNA library was then sequenced using a Novaseq sequencer. Raw image files from Illumina were converted to raw reads using CASAVA Base Calling. Raw data is stored in FASTQ(fq) format files that contain read sequences and corresponding base qualities. Select HISAT2 to match filtered sequencing reads to the Sscrofa11.1 reference genome. In general, HISAT2 supports genomes of any size, including genomes larger than 4 billion bases, and default values ​​are set for most parameters. Splicing reads from RNA Seq data can be efficiently aligned using HISAT2, the fastest system currently available, with the same or better accuracy than any other method.
The abundance of transcripts directly reflects the level of gene expression. Gene expression levels are assessed by the abundance of transcripts (sequencing count) associated with the genome or exons. The number of reads is proportional to gene expression levels, gene length, and sequencing depth. FPKM (fragments per thousand base pairs of transcript sequenced per million base pairs) were calculated and P-values ​​of differential expression were determined using the DESeq2 package. We then calculated the false discovery rate (FDR) for each P value using the Benjamini-Hochberg method9 based on the built-in R-function “p.adjust”.
RNA isolated from heart sections was converted to cDNA at a concentration of 200 ng/μl using the SuperScript IV Vilo Master mix from Thermo (Thermo, cat. no. 11756050). Quantitative RT-PCR was performed using an Applied Biosystems Endura Plate Microamp 384-well transparent reaction plate (Thermo, cat. no. 4483319) and microamp optical adhesive (Thermo, cat. no. 4311971). The reaction mixture consisted of 5 µl Taqman Fast Advanced Master mix (Thermo, cat # 4444557), 0.5 µl Taqman Primer and 3.5 µl H2O mixed per well. Standard qPCR cycles were run and CT values ​​were measured using an Applied Biosystems Quantstudio 5 real-time PCR instrument (384-well module; product # A28135). Taqman primers were purchased from Thermo (GAPDH (Ss03375629_u1), PARP12 (Ss06908795_m1), PKDCC (Ss06903874_m1), CYGB (Ss06900188_m1), RGL1 (Ss06868890_m1), ACTN1 (Ss01009508_mH), GATA4 (Ss03383805_u1), GJA1 (Ss03374839_u1), COL1A2 (Ss03375009_u1 ), COL3A1 (Ss04323794_m1), ACTA2 (Ss04245588_m1) The CT values ​​of all samples were normalized to the housekeeping gene GAPDH.
Media release of NT-ProBNP was assessed using the NT-ProBNP kit (pig) (Cat. No. MBS2086979, MyBioSource) according to the manufacturer’s protocol. Briefly, 250 µl of each sample and standard was added in duplicate to each well. Immediately after adding the sample, add 50 µl of Assay Reagent A to each well. Gently shake the plate and seal with sealant. Then the tablets were incubated at 37°C for 1 hour. Then aspirate the solution and wash the wells 4 times with 350 µl of 1X wash solution, incubating the wash solution for 1-2 minutes each time. Then add 100 µl of Assay Reagent B per well and seal with plate sealant. The tablet was gently shaken and incubated at 37°C for 30 minutes. Aspirate the solution and wash the wells 5 times with 350 µl of 1X wash solution. Add 90 µl of substrate solution to each well and seal the plate. Incubate the plate at 37°C for 10-20 minutes. Add 50 µl Stop Solution to each well. The plate was immediately measured using a Cytation (BioTek) plate reader set at 450 nm.
Power analyses were performed to choose the group sizes which will provide >80% power to detect a 10% absolute change in the parameter with a 5% Type I error rate. Power analyzes were performed to choose the group sizes which will provide >80% power to detect a 10% absolute change in the parameter with a 5% Type I error rate. Анализ мощности был выполнен для выбора размеров групп, которые обеспечат >80% мощности для обнаружения 10% абсолютного изменения параметра с 5% частотой ошибок типа I. Power analysis was performed to select group sizes that would provide >80% power to detect 10% absolute parameter change with a 5% Type I error rate.进行功效分析以选择将提供> 80％功效以检测参数中10％绝对变化和5％I型错误率的组大小。进行功效分析以选择将提供> 80％功效以检测参数中10％绝对变化和5％I型错误率的组大小。 Был проведен анализ мощности для выбора размера группы, который обеспечил бы > 80% мощности для обнаружения 10% абсолютного изменения параметров и 5% частоты ошибок типа I. A power analysis was performed to select a group size that would provide >80% power to detect 10% absolute parameter change and 5% type I error rate. Tissue sections were randomly selected before the experiment. All analyzes were condition blind and samples were decoded only after all data had been analyzed. GraphPad Prism software (San Diego, CA) was used to perform all statistical analyses. For all statistics, p-values were considered significant at values <0.05. For all statistics, p-values ​​were considered significant at values ​​<0.05. Для всей статистики p-значения считались значимыми при значениях <0,05. For all statistics, p-values ​​were considered significant at values ​​<0.05.对于所有统计数据，p 值在值<0.05 时被认为是显着的。对于所有统计数据，p 值在值<0.05 时被认为是显着的。 Для всей статистики p-значения считались значимыми при значениях <0,05. For all statistics, p-values ​​were considered significant at values ​​<0.05. Two-tailed Student’s t-test was performed on the data with only 2 comparisons. One-way or two-way ANOVA was used to determine significance between multiple groups. When performing post hoc tests, Tukey’s correction was applied to account for multiple comparisons. RNAsec data has special statistical considerations when calculating FDR and p.adjust as described in the Methods section.
For more information on study design, see the Nature Research Report abstract linked to this article.