This work presents an animal model of endothelial-to-mesenchymal transition-induced fibrosis, as seen in congenital cardiac defects such as critical aortic stenosis or hypoplastic left heart syndrome, which allows for detailed histological tissue evaluation, the identification of regulatory signaling pathways, and the testing of treatment options.
Endocardial fibroelastosis (EFE), defined by subendocardial tissue accumulation, has major impacts on the development of the left ventricle (LV) and precludes patients with congenital critical aortic stenosis and hypoplastic left heart syndrome (HLHS) from curative anatomical biventricular surgical repair. Surgical resection is currently the only available therapeutic option, but EFE often recurs, sometimes with an even more infiltrative growth pattern into the adjacent myocardium.
To better understand the underlying mechanisms of EFE and to explore therapeutic strategies, an animal model suitable for preclinical testing was developed. The animal model takes into consideration that EFE is a disease of the immature heart and is associated with flow disturbances, as supported by clinical observations. Thus, the heterotopic heart transplantation of neonatal rat donor hearts is the basis for this model.
A neonatal rat heart is transplanted into an adolescent rat's abdomen and connected to the recipient's infrarenal aorta and inferior vena cava. While perfusion of the coronary arteries preserves the viability of the donor heart, flow stagnation within the LV induces EFE growth in the very immature heart. The underlying mechanism of EFE formation is the transition of endocardial endothelial cells to mesenchymal cells (EndMT), which is a well-described mechanism of early embryonic development of the valves and septa but also the leading cause of fibrosis in heart failure. EFE formation can be macroscopically observed within days after transplantation. Transabdominal echocardiography is used to monitor the graft viability, contractility, and the patency of the anastomoses. Following euthanasia, the EFE tissue is harvested, and it shows the same histopathological characteristics as human EFE tissue from HLHS patients.
This in vivo model allows for studying the mechanisms of EFE development in the heart and testing treatment options to prevent this pathological tissue formation and provides the opportunity for a more generalized examination of EndMT-induced fibrosis.
Endocardial fibroelastosis (EFE), defined by the accumulation of collagen and elastic fibers in the subendocardial tissue, presents as a pearly or opaque thickened endocardium; EFE undergoes most active growth during the fetal period and early infancy1. In an autopsy study, 70% of cases with hypoplastic left heart syndrome (HLHS) were associated with the presence of EFE2.
Cells expressing markers for fibroblasts are the main cell population in EFE, but these cells also concomitantly express endocardial endothelial markers, which is an indication of the origin of these EFE cells. Our group previously established that the underlying mechanism of EFE formation involves a phenotypical change of endocardial endothelial cells to fibroblasts through endothelial-to-mesenchymal transition (EndMT)3. EndMT can be detected using immunohistochemical double-staining for endothelial markers such as cluster of differentiation (CD) 31 or vascular endothelial (VE)-cadherin (CD144) and fibroblast markers (e.g., alpha-smooth muscle actin, α-SMA). Furthermore, we also previously established the regulatory role of the TGF-ß pathway in this process with activation of the transcription factors SLUG, SNAIL, and TWIST3.
EndMT is a physiological process that occurs during embryonic cardiac development and leads to the formation of the septa and valves from endocardial cushions4, but it also causes organ fibrosis in heart failure, kidney fibrosis, or cancer and plays a key role in vascular atherosclerosis5,6,7,8. EndMT in cardiac fibrosis is mainly regulated through the TGF-β pathway, as we and others have reported3,9. Various stimuli have been described to induce EndMT: inflammation10, hypoxia11, mechanical alterations12, and flow disturbances, including alterations of the intracavitary blood flow13, and EndMT may also be a consequence of a genetic disease14.
This animal model was developed using the key components of cardiac EFE development, which are immaturity and alterations of the intracavitary blood flow, specifically flow stagnation. Immaturity was fulfilled by using neonatal rat hearts as donors, since neonatal rats are known to be developmentally immature immediately after birth. Heterotopic heart transplantation offered the provision of intracavitary flow restriction15.
From a clinical point of view, this animal model allows for better investigating the impact of EndMT on the growing left ventricle (LV). The growth restriction imposed on the fetal and neonatal heart through EndMT-induced EFE formation16 precludes patients with left ventricular outflow tract obstructions (LVOTO) such as congenital critical aortic stenosis and hypoplastic left heart syndrome (HLHS) from curative anatomical biventricular surgical repair17. This animal model facilitates the study of the cellular mechanisms and regulation of tissue formation through EndMT and allows for the testing of pharmacological treatment options3,18.
Transabdominal echocardiography is used to monitor the graft viability, contractility, and the patency of the anastomoses. Following euthanasia, EFE formation can be macroscopically observed within 3 days after transplantation. EFE tissue shows the same histopathological characteristics as human EFE tissue from patients with LVOTO.
Hence, this animal model, though developed for pediatric use in the spectrum of HLHS, can be applied when studying various diseases based on the molecular mechanism of EndMT.
All the animal procedures were conducted in accordance with the National Research Council. 2011. Guide for the Care and Use of Laboratory Animals: Eighth Edition. The animal protocols were reviewed and approved by the Institutional Animal Care and Use Committee at Boston Children's Hospital.
Prior to surgery, all the surgical instruments are steam-autoclaved, and modified Krebs-Henseleit buffer, with a final concentration of 22 mmol/L KCl, is prepared as a cardioplegic solution (Table 1). The solution is filter-sterilized and stored at 4 °C overnight. A surgical microscope (12.5x) is required for the heterotopic neonatal rat heart transplantation procedure.
1. Preparation and anesthesia
2. Surgical preparation and heterotopic transplantation of the neonatal donor heart in the recipient rat
3. Harvesting of the neonatal donor heart
4. Recovery of the recipient and graft monitoring
Graft viability and beating
In this work, the graft viability was visually assessed after all the clamps had been removed, and an approximate reperfusion time of 10-15 min was allowed with an open abdomen for observation of the graft. The same scoring system to objectively verify graft viability was used for visual assessment at the end of surgery and for the echocardiography on POD 1, POD 7, and POD 14.
0 = no organ function; 1 = (rest) organ function, only minimal contraction; 2 = weak or partial organ function; 3 = contractile rate or intensity reduced, but homogenous organ function; 4 = optimal atrium and ventricle contraction (120-160 beats/min). A score of 3 or 4 was rated a success. Palpatory evaluation of the abdominal donor graft was used to monitor the graft viability between the time points of echocardiographic assessment.
Mortality and graft viability success rate
The procedure was introduced to a new surgical team at the study center between October 2022 and December 2022, and 19 neonatal heterotopic rat heart transplantations were performed at the study center during this period. The immediate operative survival rate was 79%, and the graft viability success rate (displaying a viable, beating donor heart) was 84%. The procedure characteristics are presented in Table 3.
Among the 12 surviving animals, 2 required euthanasia prior to the 2 week study endpoint, 1 due to an ileus (n = 1), and the other due to pain unrelieved with pain medication (n = 1), and 2 were euthanized by design 1 week after surgery.
In three rats, the applied modified Stanford score increased from 3 to 4 between immediate postoperative visual grading and echocardiographic evaluation on POD 1. Among the eight surviving rats at the 14 days endpoint, the modified Stanford scores at echocardiography were four for seven animals and three for one animal. The most common cause of death in this series was hemodynamic failure due to excessive blood loss as a consequence of the very immature heart and, thus, fragile donor vessels for anastomosis or long anesthesia times.
Histological assessment of EFE tissue
Following CO2 euthanasia of the recipient rat, a re-laparotomy was performed under sterile preparation. The donor graft was excised and immediately placed in a physiologic saline solution on ice for further processing. A horizontal slice was resected at the mid-ventricular level of the right and left ventricle, placed in optimal cutting temperature (OCT) embedding medium, and frozen in liquid nitrogen (Figure 4A). All the other tissue was snap-frozen with liquid nitrogen and stored in a −80°C freezer for further analysis. Images were acquired using an inverted microscope (Figure 4B–D).
Immunohistochemical staining as the gold standard to identify EndMT was performed using 4',6-diamidino-2-phenylindole (DAPI) (blue), VE-Cadherin as an endothelial marker (red), and α-SMA as a fibroblast marker (green). Phosphorylated SMAD proteins and the transcription factor SLUG/SNAIL were also stained in the EFE tissue (Figure 5A–E)3,20.
Figure 1: Oblique shelf for intubation. The rat is placed on its back, with the front teeth secured with a string and the head facing toward the surgeon. Please click here to view a larger version of this figure.
Figure 2: Echocardiographic long-axis view of the LV. (A) Native rat heart indicating normal filling during diastole. (B) Donor graft with flow stagnation within the LV. Diminished volume loading during diastole. Abbreviations: LV = left ventricle; MV = mitral valve; LVOT = left ventricular outflow tract. Please click here to view a larger version of this figure.
Figure 3: Anastomoses and EFE evaluation. (A) Echocardiographic color Doppler study indicating patent arterial (red arrow) and venous (blue arrow) anastomoses. (B,C): Echo-bright endocardial surface within the LV cavity indicative of EFE (white arrows). Abbreviations: LV = left ventricle; EFE = endocardial fibroelastosis. Please click here to view a larger version of this figure.
Figure 4: Macroscopic and microscopic tissue evaluation. (A) Mid-ventricular cross-section through the LV and RV. The white arrows point toward the EFE tissue. (B) Hematoxylin-eosin, (C) Masson's trichrome (MTS), and (D) Elastin van Gieson (EVG) staining. The large magnification indicates that the EFE tissue (black arrows) contains high amounts of organized collagen (blue in MTS) and elastin fibers (black in EVG). Abbreviations: LV = left ventricle; RV = right ventricle; EFE = endocardial fibroelastosis. Please click here to view a larger version of this figure.
Figure 5: Comparison of histological and immunohistological images. (A) Hematoxylin-eosin staining. (B–E) Immunohistochemical staining; EFE tissue double-stained for (B,C) VE-Cadherin and α-SMA, (D) CD31 and phospho-SMAD2/SMAD3 (colocalized with the nuclei stained with DAPI in blue), and (E) CD31 and SLUG/SNAIL (colocalized with the nuclei stained with DAPI in blue), indicative of EndMT, as shown by the white arrows. Abbreviations: LV = left ventricle; EFE = endocardial fibroelastosis; EndMT = endothelial-to-mesenchymal transition. Please click here to view a larger version of this figure.
1 liter of sterilized, distilled water | |
NaCl | 118 mmol/L |
KCl | 22 mmol/L |
KH2PO4 | 1.2 mmol/l |
MgSO4 | 1.2 mmol/L |
NaHCO3 | 25 mmol/L |
Glucose | 11 mmol/L |
CaCl2 | 2.5 mmol/L |
Table 1: Composition of the Modified Krebs-Henseleit buffer. High-potassium (22 mmol/L KCl) cardioplegic solution is prepared, filter-sterilized, and stored at 4 °C overnight.
Common failures and troubleshooting | ||
Graft does not start to beat/coronary arteries do not fill after releasing the clamps | Check for thrombus formation at arterial anastomosis | |
Check for ischemia time (=total arrest time) (should not exceed 100 minutes) | ||
Long wake-up time or rat does not wake up after surgery | Monitor pulse strength and frequency during surgery and reduce isoflurane inhalation, if hemodynamics are weak | |
Immediate postoperative livid or necrotic intestines are suspicious for reduced intraoperative hemodynamics often due to long anesthesia time | ||
Weak hemodynamics right after laparotomy | Adjust isoflurane flow for anesthesia | |
Evaluate intubation and proper chest motion: unilateral intubation, pneumothorax, obstructed endotracheal lumen are common failures in the beginning. | ||
Rat wakes up but dies in first 24 hours | Extensive blood loss during surgery | |
If increased amount of blood is found at autopsy in abdomen, it is most likely due to failure of the anastomosis |
Table 2: Common failures and troubleshooting. Thorough monitoring and re-evaluation of unsuccessful procedures are crucial to achieving a high survival rate in this model.
Recipient rat weight in grams, median [IQR] | 150 [50] |
Donor age in days, median [IQR] | 3 [1] |
Donor weight in grams, median [IQR] | 9 [2] |
Graft ischemia time in minutes, median [IQR] | 100 [25] |
Postoperative success rate, n | 16/19 (=84%) |
Table 3: Procedure characteristics. Recipient and donor selection, graft ischemia time, and survival rate. Abbreviation: IQR = interquartile range.
This animal model of heterotopic transplantation of a neonatal donor rat heart into the recipient's abdomen creates the possibility to study EndMT-derived fibrosis through detailed histological tissue evaluation, identify regulatory signaling pathways, and test treatment options. Since EndMT is the underlying mechanism for fibrotic diseases of the heart, this model has great value in the field of pediatric cardiac surgery and beyond. In this model, many factors can negatively influence the outcome of the procedure. Thus, proper handling of the very fragile tissue due to the immaturity of the donor heart, proper animal handling during anesthesia, and high-level microsurgical skills are basic requirements for the success of this model. An optimal technical setup, including a surgical microscope, small animal ventilator, and microsurgical instruments, should be used when performing these experiments. Although not essential, basic monitoring of the heart rate or body temperature could be beneficial, especially for inexperienced surgeons, in order to monitor the hemodynamics and depth of anesthesia.
Important surgical aspects to bear in mind include the immaturity of the neonatal donor hearts, which makes the tissue very fragile and leaves the ascending aorta and pulmonary trunk vulnerable to tears. Thus, any handling should be undertaken with great care. Due to the small vessels used for anastomosis, it is recommended to perform arterial anastomosis with interrupted stitches and intermittent flushing of the anastomosis site with heparinized saline, which helps to avoid thrombus formation. Selection of appropriately aged neonatal rats is required to overcome the issue of using hearts that are too immature and, therefore, highly susceptible to anastomosis rupture. On the other hand, after a certain age of about 7 days, EndMT can no longer be shown reproducibly in this animal model15.
EndMT has been identified as the central mechanism for various kinds of cardiac fibrosis and atherosclerosis, but research has been hampered due to a lack of in vivo models8. The main developments in the field of EndMT research are restricted to cell culture models, which have inherent limitations3,8,9. Furthermore, studies on endocardial endothelial cells are even more restricted. As an alternative, coronary artery endothelial cells are often used as a substitute, as they have been reported to originate in part from endocardial cells21. Hence, this animal model can be used not only for cardiac fibrosis but to study important pathomechanisms of flow-induced EndMT in atherosclerosis. For congenital heart disease, we have shown the ability to reproduce the transition from healthy endocardium to EFE tissue through EndMT in our rat model, with EFE that structurally resembles human EFE tissue. There is some controversy regarding the cellular origin of mesenchymal cells within the EFE tissue. Clark et al.22 reported that epicardial cells contribute to EFE, but our data indicated that the majority of EFE tissue is derived through endocardial endothelial cells undergoing EndMT3. Experiments on a single-cell level are currently underway to ascertain the cellular origin of EFE tissue.
Through this in vivo model, the regulatory pathways of EndMT can be studied. An imbalance, specifically an increase in the TGF-ß pathway and impaired bone morphogenetic protein (BMP) signaling, has been shown to play a major role in endocardial cells expressing transcription factors regulating EndMT. Alternatively, Jagged/NOTCH signaling and Wnt/ß-Catenin have also been reported to induce EndMT3,23. The TGF-ß pathway induces the activation of transcription factors like SLUG, SNAIL, and TWIST via SMAD proteins, thereby regulating EndMT20,24. In this animal model, we have been able to recapitulate these mechanisms, which have been confirmed by immunohistochemical staining.
The stimulating factors for EndMT-induced fibrosis in this animal model are immaturity and flow stagnation, whereas other models are designed to induce EndMT through genetic modifications, hypertension, or dietary restrictions9,25. Compared to other species, neonatal rats are very immature at birth, and therefore, they are particularly susceptible to undergoing EndMT.
We and others have used mice to better study the origins of EFE via transgenic lineage tracing, but several limitations need to be discussed3,22. First, due to the model's complexity, mortality rates are higher in mice compared to rats, and the presentation of EFE is more heterogenous; therefore, the rat model is more reliable and reproducible. Echocardiographic measures are crucial to assess the graft function throughout the study period, and we have shown that with these measures, as well as assessing the pulsatility and patency of the anastomoses, graft function, and contractility can also be studied. With more experience, even more advanced analyses of the transplanted heart, such as strain analysis of the LV, could be performed in rat models. It is currently unclear whether the same pathophysiological condition can be induced in larger animals other than rodents, and this requires further investigation.
In conclusion, this pediatric animal model mimics the human disease of EndMT and can be useful to determine the regulation of EndMT and to study pharmacological interventions to inhibit this pathological process.
The authors have nothing to disclose.
This research was funded by Additional Ventures – Single Ventricle Research Fund (SVRF) and Single Ventricle Expansion Fund (to I.F.) and a Marietta Blau scholarship of the OeAD-GmbH from funds provided by the Austrian Federal Ministry of Education, Science and Research BMBWFC (to G.G.).
Advanced Ventilator System For Rodents, SAR-1000 | CWE, Inc. | 12-03100 | small animal ventilator |
aSMA | Sigma | A2547 | Antibody for Immunohistochemistry |
Axio observer Z1 | Carl Zeiss | inverted microscope | |
Betadine Solution | Avrio Health L.P. | 367618150092 | |
CD31 | Invitrogen | MA1-80069 | Antibody for Immunohistochemistry |
DAPI | Invitrogen | D1306 | Antibody for Immunohistochemistry |
DemeLON Nylon black 10-0 | DemeTECH | NL76100065F0P | 10-0 Nylon suture |
ETFE IV Catheter, 18G x 2 | TERUMO SURFLO | SR-OX1851CA | intubation cannula |
Micro Clip 8mm | Roboz Surgical Instrument Co. | RS-6471 | microvascular clamps |
Nylon black monofilament 11-0 | SURGICAL SPECIALTIES CORP | AA0130 | 11-0 Nylon |
O.C.T. Compound | Tissue-Tek | 4583 | Embedding medium for frozen tissue specimen |
p-SMAD2/3 | Invitrogen | PA5-110155 | Antibody for Immunohistochemistry |
Rodent, Tilting WorkStand | Hallowell EMC. | 000A3467 | oblique shelf for intubation |
Silk Sutures, Non-absorbable, 7-0 | Braintree Scientific | NC9201231 | Silk suture |
Slug/Snail | Abcam | ab180714 | Antibody for Immunohistochemistry |
Undyed Coated Vicryl 5-0 P-3 18" | Ethicon | J493G | 5-0 Vicryl |
Undyed Coated Vicryl 6-0 P-3 18" | Ethicon | J492G | 6-0 Vicryl |
VE-Cadherin | Abcam | ab231227 | Antibody for Immunohistochemistry |
Zeiss OPMI 6-SFR | Zeiss | Surgical microscope | |
Zen, Blue Edition, 3.6 | Zen | inverted microscope software |