An integrated characterization of contractile, electrophysiological, and structural cardiotoxicity of Sophora tonkinensis Gapnep. in human pluripotent stem cell-derived cardiomyocytes
Cardiotoxicity remains an important concern in drug discovery and clinical medication. Meanwhile, Sophora tonkinensis Gapnep. (S. tonkinensis) held great value in the clinical application of traditional Chinese medicine, but cardiotoxic effects were reported, with matrine, oxymatrine, cytisine, and sophocarpine being the primary toxic components.
In this study, impedance and extracellular field potential (EFP) of human-induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) were recorded using the cardio non-labeled cell function analysis and culture system (Cardio-NLCS). The effects of matrine, oxymatrine, cytisine, and sophocarpine (2, 10, 50 μM) on cell viability; level of lactate dehydrogenase (LDH), creatine kinase MB isoenzyme (CK-MB), and cardiac troponin I (CTn-I); antioxidant activities; production of reactive oxygen species (ROS) and malondialdehyde (MDA); and disruption of intracellular calcium homeostasis were also added into the integrated assessment.
The results showed that matrine and sophocarpine dose-dependently affected both impedance and EFP, while oxymatrine and cytisine altered impedance significantly. Our study also indicated that cardiotoxicity of matrine, oxymatrine, cytisine, and sophocarpine was related to the disruption of calcium homeostasis and oxidative stress. Four alkaloids of S. tonkinensis showed significant cardiotoxicity with dose dependence and structural cardiotoxicity synchronized with functional changes of cardiomyocytes.
This finding may provide guidance for clinical meditation management. Furthermore, this study introduced an efficient and reliable approach, which offers alternative options for evaluating the cardiotoxicity of the listed drugs and novel drug candidates.
KeywordsCardiotoxicity Sophora tonkinensis hiPSC-CMs Cardio-NLCS
Cardio non-labeled cell function analysis and culture system
Cell Counting Kit 8
Comprehensive in vitro Proarrhythmia Assay
Creatine kinase MB isoenzyme
Cardiac troponin I
Extracellular field potential
Field potential duration
Green fluorescence protein
Human embryonic stem cell-derived cardiomyocytes
Human-induced pluripotent stem cell-derived cardiomyocytes
Long QT syndrome
Nanion CardioExcyte 96 Sensor Plate
Reactive oxygen species
- S. tonkinensis
Sophora tonkinensis Gapnep.
Torsade de pointes
Cardiotoxicity is still one of the primary factors restricting the development of drugs. Drug development is a time-consuming and expensive process, and past statistics show that up to 90% of compounds that passed pre-clinical screening were terminated in a clinical trial, with approximately 45% of those failures being due to cardiotoxicity . In addition, many drugs that were originally on the market have been withdrawn due to clinical cases of cardiotoxicity. The HERG channel in Chinese hamster ovary or human embryonic kidney cells has been found to be closely related to QT prolongation and torsade de pointes (TdP) and has become an important component of early drug screening and safety pharmacology experiments [2, 3]. However, the assessment of cardiotoxicity cannot be completely dependent on these simple model cell lines, as they do not reproduce the integrated cardiomyocyte system and had species differences . For example, there were cases in which compounds that interacted with hERG may not have caused TdP, e.g., verapamil . Besides, cardiac safety was evaluated by some complementary assays such as multi-electrode arrays (MEA) and impedance, which the current needs were still not meet . Therefore, there is an urgent need for a method to comprehensively evaluate cardiotoxicity in both pre-drug development stage and post-evaluation stage of listed drugs.
Human-induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs)/human embryonic stem cell-derived cardiomyocytes (hESC-CMs) expressing cardiac-specific factors and structural proteins provide an alternative model for drug screening in vitro . These cells are derived from human cells, have similar mechanical and electrical activities as adult cardiomyocytes, and are more reliable in evaluating cardiotoxicity than other cell lines . Cardio non-labeled cell function analysis and culture system (Cardio-NLCS) combined hiPSC-CMs is a hybrid device supporting highly resolved impedance-based contractility measurements and electric field potential recordings that will become a more powerful and reliable means of detecting cardiotoxicity .
This article will conduct more in-depth research in the cardiotoxicity of matrine, oxymatrine, cytisine, and sophoridine. We detected impedance and potentially related indicators of hiPSC-CMs in real time, which could provide a basis for the study of the relationship among quantity-time-toxicity of S. tonkinensis, which could be greatly beneficial for promoting rational drug use in clinical practice. Drug-induced structural cardiotoxicity manifests morphological damage or loss of subcellular components of the heart leading to loss of contractility . Therefore, measurement of structural cardiotoxicity including cell viability, oxidative stress, lactate dehydrogenase (LDH) leakage, and intracellular calcium is equally important for the comprehensive evaluation for cardiotoxicity of S. tonkinensis . In conclusion, the contractile, electrophysiological, and structural cardiotoxicity of matrine, oxymatrine, cytisine, and sophocarpine was evaluated through human cardiomyocytes in order to provide guidelines for clinical medication management of S. tonkinensis.
Chemicals and reagents
Matrine, oxymatrine, cytisine, sophocarpine, and aconitine were purchased from Shanghai Winherb Medical Science Co., Ltd. (Shanghai, China) with purities higher than 98%. Cardiomyocyte recovery and maintaining medium were purchased from CELLAPYBIO (Beijing, China).
hiPSC-CMs and GCAMP hESC-CMs were obtained from CELLAPYBIO (Cat# CA2201106, Cat# CA2206106, Beijing, China) and are a well-validated cell line. Briefly, each well of Nanion CardioExcyte 96 Sensor Plate (NSP-96) and corning-96 was pre-coated with 100 ml of a 1:100 diluted Matrigel solution (BD) and maintained at 37 °C overnight. The cells were thawed rapidly from liquid nitrogen into cardiomyocytes plating medium following recommended procedures and were incubated in maintaining medium at 37 °C, 5% CO2.
After plating for about 3 days, CMs matured and began to beat. At this time, different compounds were added, and experiments were performed 24 h after treatment. The experiments were grouped as follows: control group (normal medium); matrine, oxymatrine, sophocarpine, and cytisine treatment (2, 10, 50 μM) groups; and aconitine treatment (1 μM) group. The doses of toxicity studies above were determined by the effective therapeutic doses reported and included or exceeded the therapeutic dose [9, 12, 13, 14].
Cardio non-labeled cell function analysis and culture system operation
Cardio-NLCS (Optoprobe Science LTD, Canada) is a turn-key system for efficient impedance and extracellular field potential (EFP) measurements of cell monolayers or tissue preparations. NSP-96 dedicated for Cardio-NLCS should be coated with BD and the wells should not run dry before adding the cell suspension. After seeding, the medium was not exchanged for 12 h. Subsequently, the medium exchange was completed once a day for 2–3 days and every 2 days thereafter. Adding compounds required a complete solution exchange to allow for exact dosing. After the cells were allowed to re-equilibrate for 2–3 h, the 2× concentrated compounds were added by exchanging half of the medium per well. In this experiment, several indexes (base impedance, amplitude, beat rate, and FPD) were recorded at the start of the experiment (control), then every 5 min for 4 h after compound application and every 30 min the same as cell growth period detection, lasting a minimum of 24 h. In this way, both time and concentration-dependent effects could be monitored. Impedance and EFP data analyses were taken every 4 h after compound administrations.
Cell Counting Kit 8 assay
CMs were incubated in 10% CCK-8 (Dojindo, Kumamoto, Japan) that was diluted in normal culture medium at 37 °C until a visual color conversion occurred. The absorbance of each well was measured using a microplate reader (Spectrafluor, TECAN, Sunrise, Austria) at 450 nm . Three individual experiments were performed. All samples were operated in duplicate.
Typan blue exclusion assay
A 96-well plate was used to seed hiPSC-CMs at a concentration of 5 × 104 cells per well. The cultured cells were treated with different concentrations of compounds and were incubated for another 48 h. CMs were digested with trypsin, and a small amount of cell suspension was added into the same amount of trypan blue solution. Countstar Automated Cell Counter IC1000 (Elite Life Sciences, Shanghai, China) was used to count the number of total and living cells. Data were presented as the mean ± SEM, n ≥ 3.
LDH leakage assay
Cell treatment method was the same as above. After reaching the predetermined time, 120 μl of the supernatant of each well was taken for measuring LDH leakage using a commercially available kit (Beyotime Biotechnology, Shanghai, China) following the manufacturer’s instructions . Data were presented as the mean ± SEM, n ≥ 3.
Creatine kinase MB isoenzyme and cardiac troponin I ELISA assay
Cell treatment method was the same as above. After the administration, the cells were digested and then destroyed by repeated freeze-thaw and sonication to release the components in the cells. Centrifuged at 2500 rpm for 20 min, the supernatant was collected for the detection of creatine kinase MB isoenzyme (CK-MB) and cardiac troponin I (CTn-I) ELISA kit (Expandbiotech, Beijing, China) following the manufacturer’s instructions. Data were presented as the mean ± SEM, n ≥ 3.
Superoxide dismutase, GSH, and malondialdehyde assay
Cell treatment was the same as discussed in the “Compound treatment” section. After reaching the predetermined time, the cells were washed with 4 °C physiological saline and later were added into the lysate and appropriately pipetted. After centrifuging at 1.2 × 104 rpm for 5 min, the supernatant was obtained to measure the levels of SOD, GSH and MDA using a commercially available kit (Beyotime Biotechnology, Shanghai, China) following the manufacturer's instructions . Data were presented as the mean ± SEM, n ≥ 3.
Reactive oxygen species detection on hiPSC-CMs
A 96-well plate was used to seed hiPSC-CMs at a concentration of 5 × 104 cells per well. The cultured cells were treated with different concentrations of compounds and were incubated for another 48 h. Cells were incubated in a maintaining medium containing 10 μM DCFH-DA (Beyotime Biotechnology, Shanghai, China) for 30 min at 37 °C in the dark . The measurement of ROS was performed with IncuCyte™ S3 ZOOM cell imaging system (Essen BioScience, Ann Arbor, MI). TissueQuest 6.0 was used for the quantitative analysis of ROS formation. Data were presented as the mean ± SEM, n ≥ 3.
Intracellular calcium on GCaMP hESC-CMs
The hESC-CMs in this experiment can express the GCaMP reporter gene by gene editing, and green fluorescence marker can be observed by a fluorescence microscope. The fluorescence intensity is proportional to Ca2+ concentration. The CMs were seeded into 96-well plates at 5000/well and were administered the same as above . The fluorescence intensities of hESC-CMs were monitored using an IncuCyte™ S3 ZOOM cell imaging system (Essen BioScience, Ann Arbor, MI). Data were presented as the mean ± SEM, n ≥ 3.
Data analysis and statistics
All experiments were repeated at least three times. Data are expressed as the mean ± SD and analyzed by one-way ANOVA followed by Newman-Keuls multiple comparison test as appropriate (GraphPad Prism version 5 software). P < 0.05 was considered statistically significant.
Optimization of hiPSC-CMs plating density on Cardio-NLCS
Effect of matrine on the impedance and EFP signal in hiPSC-CMs
Effect of oxymatrine on the impedance and EFP signal in hiPSC-CMs
Effect of cytisine on the impedance and EFP signal in hiPSC-CMs
Effect of sophocarpine on the EFP signal in hiPSC-CMs
Effect of matrine, oxymatrine, cytisine, and sophocarpine on cell viability and cell cytotoxicity in hiPSC-CMs
Furthermore, membrane damage that leads to the increase of LDH, CK-MB, and CTnI is generally considered irreversible, so the three indicators reflected cell cytotoxicity to some extent . As shown in Fig. 7d–f, aconitine induced a large amount of LDH, CK-MB, and CTnI in the supernatant, and the addition of four compounds also led to higher LDH, CK-MB, and CTnI levels in a dose-dependent fashion. The results were consistent with CCK-8 and trypan blue exclusion assay, which indicated medium and high doses of four compounds elicited cardiotoxicity.
Effect of matrine, oxymatrine, cytisine, and sophocarpine on oxidative stress in hiPSC-CMs
Effect of matrine, oxymatrine, cytisine, and sophocarpine on intracellular calcium in hESC-CMs with GFP
Early and effective cardiac safety pharmacology evaluation is important to guarantee measures for advancing the candidate drug . The Comprehensive in vitro Proarrhythmia Assay (CiPA) was established to develop a new paradigm for assessing proarrhythmic risk after the two guidelines ICH S7B and ICH E14, early single cardiac safety assessment methods [3, 4, 28]. This new strategy represented a shift in the pharmacology evaluation model of cardiac safety, which replaced the previous method of assessing cardiac safety using single parameters (hERG) and indirect indicators (QT prolongation) [5, 29], prone to false positives or false negatives. For example, phenobarbital led to QT prolongation without causing a fatal arrhythmia. Historically, effects on cardiac contractility were evaluated in animal studies to determine the suitability of a candidate drug to continue on a development path . Recently, the use of hiPSC-CMs based on CiPA could avoid species differences and evaluate the effects of drugs on cardiac function and structure, thereby allowing a physiologically accurate dissection of the adverse action on drugs [31, 32]. We detected cardiac contractility and some features about EFP by Cardio-NLCS, such as field potential duration (FPD)—which is analogous to the QT interval in vivo—and torsadogenic risk based on observations of early afterdepolarization and triggered activity, providing a more comprehensive assessment of cardiac function and toxicity. The combination of Cardio-NLCS and hiPSC-CMs better reflects vivo cardiotoxicity testing by offering relevant data for in vitro recordings from intact, synchronously beating networks of cardiomyocytes.
In this study, we evaluated the potential use of hiPSC-CMs for S. tonkinensis-mediated cardiotoxicity in humans by investigating some functional indicators . The clinical application of S. tonkinensis is extensive, including chronic hepatitis B, arrhythmia, and gastric ulcer, but there are few studies on its cardiotoxicity and its mechanism. To ensure the safety of medication of S. tonkinensis, we selected its main toxic component, matrine, oxymatrine, cytisine, and sophocarpine, and studied their cardiac safety to provide some guidance for clinical rational drug use. In Cardio-NLCS, the EFP reflects the population action potential, hence the crossing of ions over the cell membrane, while the impedance reflects the contraction, or the intracellular calcium flux determining the duration and strength of contraction. The four components did not change much in the base impedance relative to the control group, except for the high-dose group, the resistance decreased significantly. Our data showed that matrine affected the contraction and EFP of CMs and caused a concentration-dependent decrease in amplitude both impedance and EFP. Previous research demonstrated that the therapeutic dose of matrine and oxymatrine were able to act as hERG channel activators to suppress long QT syndrome (LQTS) . The effect on FPD, positive correlation with QT interval, was greater within 16 h after matrine administration, indicating that excessive dose of matrine also led to QT prolongation and the effect of matrine on potential was rapidly produced after administration. Several studies have shown that matrine had a positive inotropic effect and a negative frequency effect on the heart [34, 35]. For beat rate, at low, therapeutically relevant concentrations (2 μM) of matrine, a decrease in the beat rate can be detected, while at higher concentrations (10, 50 μM), an increase in beat rate can be detected in comparison with the low-dose group, indicating negative frequency effect of matrine and increased cardiotoxicity with dose increasing. Oxymatrine was reported to have a cardioprotective effect against doxorubicin-induced cardiotoxicity by inhibiting apoptosis and oxidative stress . However, in our study, oxymatrine had a large fluctuation in the amplitude (EFP) and beat rate dose-independently with a similar trend to matrine, indicating its cardiotoxicity mainly expressed by potential and pulsation. Cytisine had a greater impact on contraction and amplitude (EFP); however, it had no significant effect on FPD similar to oxymatrine, implying that both alkaloids have little effect on QT interval. Sophocarpine was another alkaloid component that had a significant impact on both impedance and EFP. A high dose (50 μM) of sophocarpine showed significant differences in each indicator compared other groups, especially FPD. FPD immediately increased in a jump after high-dose (50 μM) administration sophocarpine, and the gap gradually narrowed as the administration time increased, and there was no significant difference from the other groups after 24 h of administration, which indicated sophocarpine-induced cardiotoxicity on EFP produced quickly fast but lasted for a short time.
Several structural and mechanism-related indicators also collectively indicated the degree of cardiotoxicity. Cell viability using CCK-8 and trypan blue exclusion assay reduced with increasing concentrations of four alkaloids, and the high concentration of them and aconitine, the positive drug showed a great significance. The concentration of LDH and CK-MB in cellular supernatant reflected the degree of cell membrane and integrity and also served as the diagnostic markers of myocardial damage . In our experiment, four compounds increased the level of LDH and CK-MB dose-dependently and aconitine increased immensely the level of LDH and CK-MB. An additional cellular assay that could be useful to further understand the cellular cytotoxicity is the level of ROS, lipid accumulation, and antioxidant enzymes. SOD is a metalloprotein and accomplishes its antioxidant functions by enzymatically detoxifying the peroxides and superoxide anion, and GSH is an ubiquitous sulfhydryl-containing molecule in cells that is responsible for maintaining cellular oxidation-reduction homeostasis . Changes in GSH homeostasis can be monitored as an indication of cell damage . In the study, the activities of two antioxidant enzymes both decreased drastically with dose dependence. However, we found that the level of MDA, which is one of the several low-molecular-weight end products formed via the decomposition of certain primary and secondary lipid peroxidation products, dose-dependently increased after administration. Several physiological functions, including cell survival, growth, differentiation, and metabolism, are mediated by tightly regulated low levels of ROS. However, the presence of ROS is a double-edged sword, and excessive production of ROS can damage macromolecules, including DNA, proteins, and lipids . Thus, aberrant ROS generation constitutes a major mechanism of pathological cell death. Conforming to the predicted results, the fluorescence intensity of ROS in the high-dose groups elevated compared with the control group, which is consistent with the above results, indicating that the CMs were toxic after high-dose administration. It was reported that matrine-type alkaloids had protective effects by inhibiting excessive ROS production in vivo ; however, our results showed matrine and others induced ROS production in excess when the dose was sufficiently high and aconitine also induced excessive ROS production. The above results indicated that S. tonkinensis-induced cardiomyocyte toxicity was associated with oxidative stress.
Drug-induced long QT is most commonly induced by affecting the rapid potassium current Ikr by binding to the hERG ion channel. In addition to potassium channels, the activity of myocardial cells is largely dependent on sodium and calcium channels, whereas activation of sodium channels is required for the generation of an action potential, inward Ca2+ currents through (L-type) calcium channels counterbalance outward potassium currents in the repolarization phase . Therefore, intracellular calcium plays a key role in maintaining cardiac excitation-contraction coupling. The intracellular Ca2+ signal was determined by calcium fluorescence intensity. It was previously reported that matrine significantly attenuated the inhibition of [Ca2+]i by homocysteine in the isolated heart  and upregulated ICa-L density and α1C/Cav1.2 expression in mice . In our study, the images in Fig. 10 showed the intensity of calcium increased significantly after high-dose (50 μM) treatment, which showed the production of calcium overload and proved matrine and other three alkaloids could increase [Ca2+]i on hiPSC-CMs.
Conclusive analysis of matrine, oxymatrine, cytisine, and sophocarpine on contractile, electrophysiological, and structural cardiotoxicity
Cell viability (μM)
ROS level (μM)
LDH leakage (μM)
10 μM (16 h)
10 μM (4 h)
50 μM (20 h)
50 μM (8 h)
10 μM (4 h)
50 μM (4 h)
10 μM (12 h)
50 μM (12 h)
10 μM (4 h)
50 μM (16 h)
50 μM (8 h)
50 μM (4 h )
50 μM (16 h)
50 μM (0–12 h)
A limitation is that sample size in our study is a little low. But we used at least three samples for statistical analysis and found some interesting results. Therefore, the next step is to verify the results that have been completed by expanding the sample size, and further study will be carried out to gain a deeper understanding of the mechanism of cardiotoxicity.
In summary, our study indicated that matrine, oxymatrine, cytisine, and sophocarpine could induce cardiotoxicity through changing the cardiac impulse function or/and extracellular field potential of cardiomyocytes related to oxidative stress and disruption of calcium homeostasis, but the influence of each component on the contractile function and field potential was inconsistent. Mechanism and specific targets will be investigated in our further research, aiming at providing a basis for the clinical dose of S. tonkinensis and avoiding unnecessary toxic side effects effectively. In addition, this study used an efficient and effective approach to study cardiotoxicity, which is also helpful to evaluate the cardiotoxicity of listed drugs and novel drug candidates.
This study was supported by the Special Project for the Special Research Project for TCM (Grant No.201507004) and CAMS Innovation Fund for Medical Sciences (CIFMS, Grant No. 2016-I2M-1-012).
Availability of data and materials
Please contact author for data requests.
GS, XS, and MW designed the study. RW, MW, and SW performed the experiments and edited the manuscript. KY, PZ, QC, and XX analyzed the data. JY, GS, and XS supervised the study. All authors read and approved the final manuscript.
Ethics approval and consent to participate
hiPSC-CMs and hESC-CMs were obtained from Beijing Cellapy Biotechnology Co., Ltd.
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.