The epicardium plays a crucial role in the development and repair of the heart by providing cells and growth factors to the myocardial wall. Here, we describe a method to culture human primary epicardial cells that enables the study and comparison of their developmental and adult characteristics.
The epicardium, an epithelial cell layer covering the myocardium, has an essential role during cardiac development, as well as in the repair response of the heart after ischemic injury. When activated, epicardial cells undergo a process known as epithelial to mesenchymal transition (EMT) to provide cells to the regenerating myocardium. Furthermore, the epicardium contributes via secretion of essential paracrine factors. To fully appreciate the regenerative potential of the epicardium, a human cell model is required. Here we outline a novel cell culture model to derive primary epicardial derived cells (EPDCs) from human adult and fetal cardiac tissue. To isolate EPDCs, the epicardium is dissected from the outside of the heart specimen and processed into a single cell suspension. Next, EPDCs are plated and cultured in EPDC medium containing the ALK 5-kinase inhibitor SB431542 to maintain their epithelial phenotype. EMT is induced by stimulation with TGFβ. This method enables, for the first time, the study of the process of human epicardial EMT in a controlled setting, and facilitates gaining more insight in the secretome of EPDCs that may aid heart regeneration. Furthermore, this uniform approach allows for direct comparison of human adult and fetal epicardial behavior.
The epicardium, a single-cell epithelial layer that envelopes the heart, is of vital importance for cardiac development and repair (reviewed in Smits et al.1). Developmentally, the epicardium arises from the proepicardial organ, a small structure located at the base of the developing heart. Around developmental day E9.5 in mouse, and 4 weeks post-conception in human, cells start to migrate from this cauliflower structure and cover the developing myocardium2. Once a single epithelial cell layer is formed, a portion of the epicardial cells undergoes epithelial to mesenchymal transition (EMT). During EMT, cells lose their epithelial characteristics, such as cell-cell adhesions, and obtain a mesenchymal phenotype which gives them the capacity to migrate into the developing myocardium. The formed epicardial derived cells (EPDCs) can differentiate into several cardiac cell types including fibroblasts, smooth muscle cells, and potentially cardiomyocytes and endothelial cells3, although differentiation of the latter two cell populations remains subject to debate (reviewed in Smits et al.4). Furthermore, the epicardium provides instructive paracrine signals to the myocardium to regulate its growth and vascularization5,6,7,8. Multiple studies have demonstrated that impaired epicardial formation leads to developmental defects in cardiac muscle9,10, vasculature11, and conduction system12, emphasizing the essential contribution of the epicardium to the formation of the heart.
Although in the adult heart the epicardium is present as a dormant layer, it becomes reactivated upon ischemia13. Epicardial reactivation post-injury recapitulates several of the processes described for cardiac development, including proliferation and EMT14, albeit less efficiently. Interestingly, although the exact mechanism is not fully understood, the epicardial contribution to repair can be improved by treatment with, e.g., Thymosin β415 or modified VEGF-A mRNA16, resulting in ameliorated cardiac function after myocardial infarction. The epicardium is therefore considered an interesting cell source to enhance endogenous repair of the injured heart.
Mechanisms of cardiac development are often recapitulated during injury, although in a less efficient manner. In search of epicardial activators, it is paramount that we can determine and compare the full capacity of the fetal and adult epicardium. Moreover, from a therapeutic point of view, it is important that, in addition to animal experiments, we extend knowledge regarding the response of the human epicardium. Here, we describe a method to isolate and culture human adult and fetal epicardial derived cells (EPDCs) in an epithelial-cell-like morphology and to induce EMT. With this model, we aim to explore and compare adult and fetal epicardial cell behavior.
The main advantage of this protocol is the use of human epicardial material, which has not been thoroughly studied. Importantly, the described isolation and cell culture protocol provides a single uniform method to derive both fetal and adult cobble EPDCs, enabling a direct comparison between these two cell sources. Additionally, since the epicardium is isolated based on its location, it is ensured that the cells are actually epicardially derived17.
While human EPDC isolation methods have been established previously, these mostly rely on outgrowth protocols where pieces of cardiac or epicardial tissue are plated onto a cell culture dish18,19. This approach thereby selects specifically for cells that partially lose their epithelial phenotype in order to migrate, and that are more prone to undergo spontaneous EMT. In the current protocol, the epicardium is first processed into a single cell solution which allows the isolated EPDCs to maintain their epithelial state. This method therefore provides a solid in vitro model to study epicardial EMT.
All experiments with human tissue specimens were approved by the ethics committee of the Leiden University Medical Center and conforms to the Declaration of Helsinki. All steps are performed with sterile equipment in a cell culture flow cabinet.
1. Preparations
2. Retrieval and Storage of Adult and Fetal Heart Specimens
3. Isolation of the Epicardial Layer
4. Culture of EPDCs
5. Induction of EMT in EPDCs
Here, we outline a straightforward protocol to isolate EPDCs from human adult and fetal cardiac tissue (Figure 1). This protocol takes advantage of the easily accessible location of the epicardium on the outside of the heart (Figure 1A). Staining of the heart auricle after dissection demonstrates that the WT1+ epicardium is removed while the underlying subepicardial extracellular matrix and myocardial tissue remain intact (Figure 1J). Extensive characterization has been performed before, demonstrating that EPDCs express epicardial markers, e.g., ALDH1A2, TBX18, KRT8, KRT19 and lack expression of other heart-resident cell types, e.g., PECAM1, ISL1, CD34, and TNNT2.17
Both adult (Figure 1K) and fetal (Figure 1L) EPDCs cultured in the presence of ALK5 kinase inhibitor SB431542 show a cobblestone morphology. However, depending on the donor and culture conditions, fetal EPDCs can undergo EMT despite the presence of SB. This can result in spindle-shaped cells within the population of cobblestone cells (see arrows in Figure 1L). Please be aware that the isolation of cobblestone-shaped cells from fetal tissue is not always feasible, and may result in the derivation of mostly spindle-shaped cells. Since these cells grow faster, they will rapidly overgrow other cell types.
EMT can be induced by incubating with TGFβ320 for 5 days in both adult and fetal cells, as demonstrated by a clear morphological transition to spindle-shaped cells (Figure 2A). As demonstrated with the untreated control ("Empty"), fetal EPDCs will undergo spontaneous EMT upon removal of SB, while adult EPDCs will only undergo EMT upon stimulation with TGFβ3. To validate EMT, EPDCs were immunostained with alpha-smooth muscle actin (αSMA), a mesenchymal marker (Figure 2B). Furthermore, EMT was confirmed in adult EPDCs using qRT-PCR showing downregulation of the epicardial marker WT1 and upregulation of the mesenchymal markers POSTN, αSMA, Collagen 1A1, MMP3, and N-Cadherin (Figure 2C). Comprehensive experiments regarding adult and fetal epicardial EMT are published before17.
Tissue | Well type | Cell Growth Area (cm2) | Volume of medium (mL) | Addition of SB |
Fetal | 24 | 1.9 | 0.5 | Directly |
Adult small (<4 cm) | 12 | 3.8 | 1 | After 1st passage |
Adult large | 6 | 9.6 | 2 | After 1st passage |
(>4 cm) |
Table 1: Guideline for selecting the appropriate cell culture plate. The required well size depends on the amount of epicardial cells isolated. These guidelines can be used as a rule of thumb to select the appropriate cell culture plate.
Gene | Sequence |
WT1 Forward | CAG CTT GAA TGC ATG ACC TG |
WT1 Reverse | TAT TCT GTA TTG GGC TCC GC |
N-Cadherin Forward | CAG ACC GAC CCA AAC AGC AAC |
N-Cadherin Reverse | GCA GCA ACA GTA AGG ACA AAC ATC |
POSTN Forward | GGA GGC AAA CAG CTC AGA GT |
POSTN Reverse | GGC TGA GGA AGG TGC TAA AG |
SMA Forward | CCG GGA GAA AAT GAC TCA AA |
SMA Reverse | GAA GGA ATA GCC ACG CTC AG |
MMP3 Forward | TGG ATG CCG CAT ATG AAG |
MMP3 Reverse | CAG AAA TGG CTG CAT CGA |
COL1A1 Forward | CCA GAA GAA CTG GTA CAT CAG CA |
COL1A1 Reverse | CGC CAT ACT CGA AAT GGG AAT |
GAPDH Forward | AGC CAC ATC GCT CAG ACA C |
GAPDH Reverse | GCC CAA TAC GAC CAA ATC C |
B2M Forward | ACA CTG AAT TCA CCC CCA CT |
B2M Reverse | GCT TAC ATG TCT CGA TCC CAC T |
Table 2: Primer sequences used for validation of EMT in EPDCs. Sequences of forward and reverse primers used for qRT-PCR to determine the expression of several EMT related genes in adult EPDCs.
Figure 1: Isolation of Epicardial Derived Cells (EPDCs). (A) Depicted is an adult auricle removed from the human heart with a thin outer membranous layer, the epicardium, which is removed. The black arrow points to the epicardium. (B-I) Visual representation of the isolation method of EPDCs. The black arrow points to the cell pellet. (J) Immunofluorescent staining of a heart auricle with and without dissection of the epicardium. Scale bar: 50 µm. (K) Representative pictures of two different adult EPDC isolations cultured with SB. (L) Representative pictures of two different fetal EPDC isolations cultured with SB. Black arrows indicate mesenchymal-like cells in fetal EPDC culture. Scale bar: 200 µm. Please click here to view a larger version of this figure.
Figure 2: Validation of EMT in human adult and fetal Epicardial Derived Cells (EPDCs). Adult and fetal EPDCs were cultured with SB, not treated (Empty) or stimulated with TGFβ3 for 5 days. (A) Representative bright field pictures. Scale bar: 200 µm. (B) Immunostaining for DAPI and αSMA. Scale bar: 100 µm. (C) mRNA levels of EMT-related genes, determined in adult EPDCs using qRT-PCR. Measured values were normalized to GAPDH and B2M expression. Values are depicted as mean + SD 2^-ΔΔct (n = 2). Abbreviations: WT1:Wilms’ tumor 1, MMP3: Matrix Metalloproteinase 3, POSTN:Periostin, αSMA: alpha Smooth Muscle Actin, Col1A1:Collagen 1A1. Please click here to view a larger version of this figure.
Here we describe a detailed protocol to isolate and culture primary epicardial cells derived from human adult and fetal hearts. Extensive characterization of these cells has been previously published17. We have shown that both cell types can be maintained as epithelial cobblestone-like cells when cultured with the ALK5 kinase inhibitor SB431542. EMT is an integral part of epicardial activation in vivo during both development and the post-injury response. EMT can be studied using this method by addition of TGFβ. Importantly, we previously observed that fetal EPDCs rapidly undergo spontaneous EMT when SB is removed, while adult EPDCs only undergo EMT upon stimulation17. Studying these processes in fetal EPDCs may aid in understanding how to optimally activate the adult epicardium after damage.
The presented method relies on patient material, which is obtained during surgery. Therefore, one can expect several variations in the state of the material isolated, which are either due to patient variability or the speed at which the material is collected in the operating theater. This variation may explain differences in the procedure: 1) how easily the epicardium can be dissected from the myocardium, 2) adherence of isolated cells to the plate, 3) the ability to prevent or undergo EMT, and 4) proliferation speed. In general, a quick isolation is preferable for cell survival. The main critical point is peeling the epicardium from the myocardium. If the epicardium is strongly adhered to the underlying tissue, a 15-min pre-treatment of the cardiac tissue with trypsin can help to remove the epicardium more easily. Furthermore, the efficacy of the trypsin treatment depends on several factors, including the trypsin activity and the composition of the tissue. Therefore, if a low yield is observed, the incubation time with trypsin can be adjusted. Additionally, if no cell pellet is visible after spinning down, the solution might not have been completely dissociated before running through the filter. Mixing the solution before passing the solution through the syringe or using an extra, smaller syringe could be useful to dissociate the cells. The well size for seeding EPDCs should be considered carefully to enable cells to maintain their epithelial state. Table 1 gives an indication, but one can deviate from this guideline when, for example, only a small part of fetal heart can be used and plating in a smaller well is more convenient.
Since fetal cells will undergo EMT immediately without SB17, fetal EPDCs should always be plated with SB. However, adult EPDCs tend to maintain their epithelial morphology and therefore can be plated without SB during the first 48 h of the first passage, to promote cell adherence. After the first passage, both adult and fetal EPDCs are continuously cultured with SB to prevent spontaneous EMT. In addition, low cell density is an important trigger for EMT. EPDCs should therefore never be cultured below 50% confluency. On the other hand, if EMT is desired, make sure that EPDCs are not seeded too densely, and stay below 70% confluency. Though this protocol utilizes TGFβ3, stimulation with TGFβ1 and -2 can induce EMT in EPDCs as well (observations unpublished).
In this protocol, cells are split directly without spinning down, since we observe a lower cell survival when using the centrifuge. Therefore, it is vital to use low volumes of trypsin to ensure its deactivation when serum containing medium is added.
After ~6 – 8 passages, EPDCs with an epithelial morphology will either stop growing or will undergo EMT spontaneously. Therefore, for experiments, we use cobble EPDCs between passage 3 and passage 6. In contrast, EPDCs with a mesenchymal morphology are less vulnerable and can be cultured up to passage 20. In experiments, however, we have never used spindle EPDCs after the 10th passage.
Since the epicardium is located at the outside of the heart and is easy to separate from underlying tissue, the authenticity of the cells is evident, and the chance of significant contamination is low. Although adipose tissue or blood vessels sometimes stick to the isolated epicardial layer, the majority of those cells will not pass the strainer during the isolation, and otherwise cannot survive in EPDC cell culture. Furthermore, as mentioned before, using human material provides a unique model to investigate human epicardial cell behavior. It is known that, for instance, epicardial adipose tissue is different between species, emphasizing the necessity of human epicardial cell models21.
It should be noted that the heart auricles were obtained during surgery on diseased hearts, and therefore the differences in disease, used medication, age, and gender of the donor may influence the reproducibility of the experiments. Experiments should therefore be performed on several isolations to obtain valid results and allow solid conclusions to be drawn. Furthermore, depending on the research question, one could choose specifically to only use epicardium from patients with ischemic disease or patients with non-ischemic valvular disease, which are expected to behave differently.
It has been suggested that atrial epicardium may have distinct characteristics from ventricle epicardium2, thereby questioning if epicardium derived from the atrial heart auricle provides a valid model for epicardial behavior. In this context, it is important to point out that we could not find differences between fetal atrial and ventricular derived EPDCs (unpublished data). However, using epicardium from the human adult ventricle would be the ultimate cell source to verify this. Yet, collecting large specimens of human adult ventricles for EPDC isolation is highly invasive for the patient, and is therefore currently not feasible in this hospital.
We observed that SB is not always sufficient to prevent EMT, mainly in fetal EPDCs. When fetal EPDCs undergo EMT in the presence of SB, we exclude them from experiments. As a consequence, cells used for experiments are selected for their ability to maintain an epithelial phenotype in response to SB. We hypothesize that fetal cells can already be beyond a certain threshold in the process of EMT, and that inhibition with SB is not able to stop this.
This epicardial cell model has several applications, since both the developing and the adult epicardium can be investigated. In our lab, we focus on improvement of the epicardial regenerative response after cardiac damage. Adult EPDCs can be used to test compounds which induce EMT, aiming to find a potential therapeutic drug for an ameliorated regenerative response of the epicardium. In addition, it is possible to measure factors secreted by EPDCs to comprehend the paracrine signaling to the (re)generating myocardium. Furthermore, since we observed that fetal EPDCs are more prone to undergo EMT spontaneously compared to adult EPDCs, we investigate the differences between the fetal and adult EPDCs. Determining the underlying mechanism of increased fetal activation could provide a cue to improve epicardial activation in the adult heart.
The authors have nothing to disclose.
This research is supported by The Netherlands Organization for Scientific Research (NWO) (VENI 016.146.079) and a LUMC Research fellowship both to AMS, and LUMC Bontius Stichting (MJG).
Dulbecco’s modified Eagle’s medium + GlutaMAX | Gibco | 21885-025 | |
Medium 199 | Gibco | 31150-022 | |
Fetal Bovine Serum | Gibco | 10270-106 | |
Trypsin 0.25% | Invitrogen | 25200-056 | |
Penicillin G sodium salt | Roth | HP48 | |
Streptomycin sulphate | Roth | HP66 | |
Trypsin 1:250 from bovine pancreas | Serva | 37289 | |
EDTA | Sigma | E4884 | |
Gelatin from porcine skin | Sigma-Aldrich | G1890 | |
Culture plates 6 well | Greiner bio-one | 657160 | |
Culture plates 12 well | Corning | 3512 | |
Culture plates 24 well | Greiner bio-one | 662160 | |
SB 431542 | Tocris | 1614 | |
Dimethyl Sulfoxide (DMSO) | Merck | 102931 | |
100-1000µL Filtered Pipet Tips | Corning | 4809 | |
10-ml pipet | Greiner bio-one | 607180 | |
5-ml pipet | Greiner bio-one | 606180 | |
Cell culture dish 100/20 mm | Greiner bio-one | 664160 | |
PBS | Gibco | 10010056 | Or home-made and sterilized |
Eppendorf tubes 1.5 mL | Eppendorf | 0030120086 | |
15-ml centrifuge tubes | Greiner bio-one | 188271 | |
50-ml centrifuge tubes | Greiner bio-one | 227261 | |
10 mL Syringe | Becton Dickinson | 305959 | |
Needles 19 Gauge | Becton Dickinson | 301700 | |
Needles 21 Gauge | Becton Dickinson | 304432 | |
EASYstrainer Cell Sieves, 100 µm | Greiner bio-one | 542000 | |
TGFβ3 | R&D systems | 243-B3 | |
Monoclonal Anti-Actin, α-Smooth Muscle | Sigma | A2547 | |
Anti-Mouse Alexa Fluor 555 | Invitrogen | A31570 | |
Alexa Fluor 488 Phalloidin | Invitrogen | A12379 | |
Equipment | |||
Name | Company | Catalog Number | Comments |
Pipet P1,000 | Gilson | F123602 | |
Pipet controller | Integra | 155 015 | |
Stereomicroscope | Leica | M80 | |
Inverted Light Microscope | Olympus | CK2 | |
Centrifuge | Eppendorf | 5702 | |
Waterbath | GFL | 1083 |