MASS PRODUCTION OF HUMAN PLURIPOTENT STEM CELL DERIVED CARDIAC STROMAL CELL

20230279357 · 2023-09-07

Assignee

Inventors

Cpc classification

International classification

Abstract

The application describes a method for producing a population of cardiac stromal cells from pluripotent stem cells. Specifically, the method relates to (i) inducing epithelial-mesenchymal transition of pluripotent stem cell derived epicardial cells and (ii) amplifying the number of cardiac stromal cells in serum-free conditions. These cardiac stromal cells can be mass produced according to the described method and said cells maintain the expression of CD90, CD73 and CD44 in at least 80% of the cardiac stromal cells. Furthermore, the application relates to a population of cardiac stromal cells, which are pluripotent stem cells derived and wherein at least 80% of the cardiac stromal cells express CD90, CD73 and CD44. Said cardiac stromal form the basis for several in vitro and in vivo applications such as the production of engineered organ tissue and the support of, for example, heart repair. Also, a serum- free culture medium for the amplification of cardiac stromal cells is provided herein.

Claims

1. A method for producing a population of cardiac stromal cells from pluripotent stem cells, the method comprising the steps of: i. Inducing epithelial-mesenchymal transition of epicardial cells obtained by differentiation of pluripotent stem cells, wherein the epicardial cells express Wilms tumor antigen (WT-1), wherein by inducing epithelial-mesenchymal transition the epicardial cells are differentiated into cardiac stromal cells, wherein inducing epithelial-mesenchymal transition comprises a1) culturing said epicardial cells under suitable conditions in the presence of a first extracellular matrix protein in a serum-free basal medium; followed by a2) culturing the cells of step (i) a1) under suitable conditions in the presence of a second extracellular matrix protein in a serum-free basal medium; wherein at least about 80 % of the cells of the obtained population of cardiac stromal cells express CD90, CD73, and CD44; and ii. Amplifying the number of said cardiac stromal cells by culturing said population of cardiac stromal cells of step (i) in the presence of at least one third extracellular matrix protein in a serum-free basal medium, wherein at least 80 % of the cells of the cardiac stromal cell population maintain the expression of CD90, CD73, and CD44.

2. The method of claim 1, wherein step (ii) stably amplifies the population of cardiac stromal cells as determined by the maintained expression of at least 80% of CD90, CD73, and CD44, preferably at least 81%, more preferably at least 82%, even more preferably at least 83 %, even more preferably at least 84%, even more preferably at least 85%, even more preferably at least 86%, even more preferably at least 87%, even more preferably at least 88%, even more preferably at least 89%, and most preferably at least 90%, as determined by flow cytometry.

3. The method of claim 1, wherein the method comprises the steps of: i. Inducing epithelial-mesenchymal transition of epicardial cells obtained by differentiation of pluripotent stem cells, wherein the epicardial cells express Wilms tumor antigen (WT-1), wherein by inducing epithelial-mesenchymal transition the epicardial cells are differentiated into cardiac stromal cells, wherein inducing epithelial-mesenchymal transition comprises a1) culturing said epicardial cells under suitable conditions in the presence of an first extracellular matrix protein in a serum-free basal medium comprising effective amounts of (a) FGF2, (b) vascular endothelial growth factor (VEGF), (c) glutamine and (d) a GSK-3 inhibitor, wherein said amounts result in the expression of CD90 in at least 50% of the cells obtained by step (i) a1), the expression of CD73 in at most 50% of the cells obtained by step (i) a1), and the expression of CD44 in at most 30% of the cells obtained by step (i) a1); followed by a2) culturing the cells of step (i) a1) under suitable conditions in the presence of a second extracellular matrix protein in a serum-free basal medium comprising effective amounts of (a) FGF2, (b) VEGF, and (c) glutamine; wherein said amounts result in the expression of CD90, CD73, and CD44 in at least 80 % of the obtained population cardiac stromal cells; and ii. Amplifying the number of said cardiac stromal cells by culturing said population of cardiac stromal cells of step (i) in the presence of at least one third extracellular matrix protein in a serum-free basal medium comprising effective amounts of (a) FGF2, (b) VEGF, and (c) glutamine, wherein said amounts result in the maintained expression of CD90, CD73, and CD44 in at least 80 % of said cardiac stromal cell population obtained by step (ii).

4. The method of claim 1, wherein the serum-free basal medium in step (i) a1) comprises a final concentration of 10-200 ng/ml FGF2, preferably 15-100 ng/ml, more preferably 20-80 ng/ml, even more preferably 30-70 ng/ml, most preferably 40-60 ng/ml, and most preferably about 50 ng/ml; wherein serum-free basal medium in step (i) a1) comprises a final concentration of 5-100 ng/ml VEGF, preferably 7-50 ng/ml, more preferably 10-40 ng/ml, even more preferably 15-35 ng/ml, most preferably 20-30 ng/ml, and most preferably about 25 ng/ml VEGF; wherein serum-free basal medium in step (i) a1) comprises a final concentration of 0.2-20 mM glutamine, preferably, 0.5-10 mM glutamine, more preferably 0.75-5 mM glutamine, more preferably 1-3 mM glutamine, more preferably 1.5-2.5 mM glutamine, even more preferably about 2 mM glutamine, and most preferably wherein the glutamine is present as a L-alanine-L-glutamine dipeptide; and/or wherein the basal medium in step (i) a1) comprises a final concentration of 0.1-10 .Math.M CHIR99021, preferably 0.2-9 .Math.M, more preferably 0.3-8 .Math.M, even more preferably 0.4-7 .Math.M, still more preferably 0.5-6 .Math.M, more preferably 0.6-5 .Math.M, more preferably 0.7-4 .Math.M, more preferably 0.8-3 .Math.M, most preferably 0.9-2 .Math.M, and even most preferably about 1 .Math.M CHIR99021.

5. The method of claim 1, wherein the serum-free basal medium in step (i) a2) comprises a final concentration of 10-200 ng/ml FGF2, preferably 15-100 ng/ml, more preferably 20-80 ng/ml, even more preferably 30-70 ng/ml, most preferably 40-60 ng/ml, and most preferably about 50 ng/ml; wherein the serum-free basal medium in step (i) a2) comprises a final concentration of 5-100 ng/ml VEGF, preferably 7-50 ng/ml, more preferably 10-40 ng/ml, even more preferably 15-35 ng/ml, most preferably 20-30 ng/ml, and most preferably about 25 ng/ml VEGF; and/or wherein the serum-free basal medium in step (i) a2) comprises a final concentration of 0.2-20 mM glutamine, preferably, 0.5-10 mM glutamine, more preferably 0.75-5 mM glutamine, more preferably 1-3 mM glutamine, more preferably 1.5-2.5 mM glutamine, even more preferably about 2 mM glutamine, and most preferably wherein the glutamine is present as a L-alanine-L-glutamine dipeptide.

6. The method of claim 1, wherein the serum-free basal medium in step (ii) comprises a final concentration of 10-200 ng/ml FGF2, preferably 15-100 ng/ml, more preferably 20-80 ng/ml, even more preferably 30-70 ng/ml, most preferably 40-60 ng/ml, and most preferably about 50 ng/ml; wherein the serum-free basal medium in step (ii) comprises a final concentration of 5-100 ng/ml VEGF, preferably 7-50 ng/ml, more preferably 10-40 ng/ml, even more preferably 15-35 ng/ml, most preferably 20-30 ng/ml, and most preferably about 25 ng/ml VEGF; and/or wherein the serum-free basal medium in step (ii) comprises a final concentration of 0.2-20 mM glutamine, preferably, 0.5-10 mM glutamine, more preferably 0.75-5 mM glutamine, more preferably 1-3 mM glutamine, more preferably 1.5-2.5 mM glutamine, even more preferably about 2 mM glutamine, and most preferably wherein the glutamine is present as a L-alanine-L-glutamine dipeptide.

7. The method of claim 1, wherein the first extracellular matrix protein comprises laminin, vitronectin, collagen, in particular gelatine, fibronectin, elastin, preferably wherein the first extracellular matrix protein comprises laminin, more preferably wherein the extracellular matrix protein is laminin, most preferably wherein laminin is laminin-521; wherein the second extracellular matrix protein is different from the first extracellular matrix protein and comprises vitronectin, laminin, collagen, in particular gelatine, fibronectin, and elastin, preferably wherein the second extracellular matrix protein is vitronectin; and/or wherein the at least one third extracellular matrix protein is selected from vitronectin, laminin, collagen, in particular gelatine, fibronectin, and elastin, preferably wherein the extracellular matrix protein is vitronectin.

8. The method of claim 1, wherein the epicardial cells express Wilms tumor antigen 1 (WT-1) as determined by fluorescent microscopy; wherein the epicardial cells express Wilms tumor antigen 1 (WT-1) RNA at least 2-fold more than TBP (TATA-binding protein), preferably 3-fold, more preferably 4-fold, even more preferably at least 5-fold, most preferably at least 6-fold; and at most 15-fold, as determined by qRT-PCR; and/or wherein the epicardial cells are obtained as described in Schlick (2018), Doctoral Thesis, November 2018, University of Göttingen, Witty et al. Nat Biotechnology 32, 1026-1035 (2014), or any other suitable method for providing epicardial cells.

9. The method of claim 1, wherein the method comprises the steps of: i*. Inducing mesodermal differentiation by culturing said pluripotent stem cells under suitable conditions on laminin coated substrate in a serum-free basal medium comprising effective amounts of (a) bone morphogenetic protein 4 (BMP4), (b) Activin A (ActA), (c) a GSK-3 inhibitor, (d) basic fibroblast growth factor (FGF2), (e) glutamine, and (f) a serum-free supplement comprising albumin, transferrin, ethanol amine, selenium or a bioavailable salt thereof, L-carnitine, fatty acid supplement, and triodo-L-thyronine (T3), wherein said amounts result in the expression of CD90 in at least 90% of cells obtained by step (i*), the expression of CD73 in at most 10% of cells obtained by step (i*), and the expression of CD44 in at most 10% of the cells obtained by step (i*), as determined by flow cytometry; i**. Inducing epicardial differentiation by culturing the cells of step (i*) under suitable conditions on laminin coated substrates in a serum-free basal medium comprising effective amounts of (a) BMP4, (b) retinoic acid (RA), (c) a GSK-3 inhibitor, (d) insulin, (e) glutamine and (f) the serum-free supplement as in (i); whereby the obtained cells express Wilms tumor antigen 1 (WT-1), as determined by fluorescent microscopy i. Inducing epithelial-mesenchymal transition by a1) culturing the cells of step (i**) under suitable conditions in the presence of an first extracellular matrix protein in a serum-free basal medium comprising effective amounts of (a) FGF2, (b) vascular endothelial growth factor (VEGF), (c) glutamine and (d) a GSK-3 inhibitor, wherein said amounts result in the expression of CD90 in at least 50% of the cells obtained by step (i) a1), the expression of CD73 in at most 50% of the cells obtained by step (i) a1), and the expression of CD44 in at most 30% of the cells obtained by step (i) a1); followed by a2) culturing the cells of step (i) a1) under suitable conditions in the presence of a second extracellular matrix protein in a serum-free basal medium comprising effective amounts of (a) FGF2, (b) VEGF, and (c) glutamine; wherein said amounts result in the expression of CD90, CD73, and CD44 in at least 80 % of the obtained population of cardiac stromal cells; and ii. Amplifying the number of said cardiac stromal cells by culturing said population of cardiac stromal cells of step (i) in the presence of at least one third extracellular matrix protein in a serum-free basal medium comprising effective amounts of (a) FGF2, (b) VEGF, and (c) glutamine, wherein said amounts result in the maintained expression of CD90, CD73, and CD44 in at least 80 % of said cardiac stromal cell population.

10. An isolated population of cardiac stromal cells, wherein the cardiac stromal cells have been obtained by differentiation of pluripotent stem cells and wherein at least about 80 % of the cells of the population of cardiac stromal cells express CD90, CD73, and CD44.

11. An engineered organ tissue comprising a population of cardiac stromal cells as defined in claim 10.

12. Use of the population of cardiac stromal cells (cStC) obtained by the method according to claim 1, in an in vitro model for drug screening, preferably in vitro model for drug efficacy screening or in an in vitro model for drug toxicity screening.

13. Use of the population of cardiac stromal cells (cStC) obtained by the method according to claim 1, in an in vitro production of an engineered organ tissue, preferably of a human engineered organ tissue, more preferably wherein the engineered human organ tissue is engineered human myocardium or engineered human connective tissue.

14. The population of cardiac stromal cells (cStC) obtained by the method according to claim 1, for use in organ repair, preferably heart repair or soft tissue repair.

15. A serum-free cell culture medium suitable for amplification of cardiac stromal cells comprising (a) a serum-free basal medium, (b) 10-200 ng/ml FGF2, (c) 5-100 ng/ml VEGF, (d) 0.2-20 mM glutamine, and (e) an eukaryotic cell culture medium supplement comprising 6.6-165 .Math.g/ml ascorbic acid, 2-50 .Math.g/ml insulin, 1.1-27.5 .Math.g/ml transferrin, 1660-41500 .Math.g/ml albumin, and 11-145 nM selenium.

16. (canceled)

17. Use of the population of cardiac stromal cells (cStC) according to claim 10 in an in vitro production of an engineered organ tissue, preferably of a human engineered organ tissue, more preferably wherein the engineered human organ tissue is engineered human myocardium or engineered human connective tissue.

18. The population of cardiac stromal cells (cStC) according to claim 10 for use in organ repair, preferably heart repair or soft tissue repair.

19. Use of the population of cardiac stromal cells (cStC) according to- claim 10, in an in vitro model for drug screening, preferably in vitro model for drug efficacy screening or in an in vitro model for drug toxicity screening.

20. Use of the engineered organ tissue as defined in claim 11 in an in vitro model for drug screening, preferably in vitro model for drug efficacy screening or in an in vitro model for drug toxicity screening.

Description

DESCRIPTION OF THE FIGURES

[0398] FIG. 1: (A) Scalable stromal cell production with stable phenotype. Overview and chronological sequence of the differentiation/amplification method starting with epicardial cells and resulting in cardiac stromal cells. In step (i) a1), the epithelial-mesenchymal transition (EMT) was induced by culturing the epicardial cells in the presence of a first extracellular matrix protein (for example laminin) in a serum-free basal medium (for example KO DMEM) comprising (a) FGF, (b) VEGF, (c) Glutamin (Gln) and a GSK-3 inhibitor (for example 1 .Math.M CHIR). This step took between 2-8 days, preferably 4 days. After completion of said step at least 50% of the cells expressed CD90, at most 50% of the cells expressed CD73 and at most 30% of the cells expressed CD44. Said step was followed by step (i) a2), wherein the cells were passaged and further cultivated in the presence of a second extracellular matrix protein (such as vitronectin) in a basal medium (such as KO DMEM) comprising (a) FGF, (b) VEGF, (c) Glutamin (Gln). Said step took between 10-20 days, preferably 13 days, and the cells were usually passaged once after the completion of 5-10 days. After completion of step (i) a2), the cells were differentiated into cardiac stromal cells so that at least 80% of the cells expressed CD90, CD73 and CD44. Step (ii) amplified the cardiac stromal cells under serum-free conditions. Specifically, the number of cardiac stromal cells was amplified by culturing the cells in the presence of at least a third extracellular matrix protein (such as vitronectin) in a serum-free basal medium (KO DMEM) comprising (a) FGF, (b) VEGF, (c) Glutamin (Gln). Cardiac stromal cells can be amplified in said medium repeatedly and the cardiac stromal cells show a maintained expression of CD90, CD73 and CD44 in at least 80% of the cardiac stromal cell. (B) Transillumination of human undifferentiated induced pluripotent stem cells on day -14 before the differentiation of human pluripotent stem cells into epicardial cells; day 0 of the method (stage of epicardial cells); and day 17 of the method (stage of cardiac stromal cells); scale 100 .Math.m (day -14) and 20 .Math.m (day 0 and day 17), respectively. (C) Immunofluorescence images of the differentiation and amplification protocol on day 0 (before step (i) a1) of the method). The left panel showed the expression of F-Actin, the middle panel showed the expression of Wilms tumor antigen 1 (WT1), and in the right panel the nuclei were stained.

[0399] FIG. 2: Scaling of the differentiation protocol and stromal cell amplification. (A) Exemplary procedure (refer to FIG. 8 for additional details) of iPSC differentiation to cardiac stromal cells (obtained at passage 4 by completion of protocol steps (i*,i**, and i) and the amplification of cardiac stromal cells from passage 4 to passage 9 (i.e., 5 passages) following method step (ii). For example, the method start was with the seeding of ~1×10E6 iPSC per one T75 culture flask (growth surface area: 75 cm.sup.2) 96 h before the start of mesoderm induction (step (i*) for 6 days). After passaging of obtained cells onto six T225 culture flasks (growth surface area: 225 cm.sup.2 per T225 flask) on culture day -7, formation of (pro)epicardium-like cells was induced for 7 days until culture day 0 (step (i**)), followed by induction of EMT (step (i)). Step (i) comprised of two sub-steps: step (i) a1) until culture day 4 followed by step (i) a2) with a 1:2 passage of cells obtained from step (i)a1) onto twelve T225 culture flasks on culture day 4 and subsequent 1: 1 passage onto another twelve T225 culture flasks on culture day 11. Step (i) a2) was completed on culture day 17-21 with a 1:2 passage of the obtained cardiac stromal cells onto twenty-four T225 culture flasks. The following amplification step (ii) comprised a passaging every seven days (weekly), with an average population doubling every 2.1 days, i.e., a 3.3-fold amplification per passage. Average output for cardiac stromal cells was 1×10E5 cells per cm.sup.2 after passage 4; reseeding was performed with 3×10E4 cells per cm.sup.2. Note that for practical reasons, i.e., for example limited plate handling capacity and a defined cardiac stromal cell quantity required, fewer than the maximal plate numbers are typically used in a manual cell amplification process. Surplus cardiac stromal cells can be cryo-preserved using standard protocols at the end of each passage without compromising the cardiac stromal cell phenotype. Depending on the cryo-preservation protocol used, there may be differences in cardiac stromal cell retrieval. A retrieval is typically 60-80% of the cryo-preserved cardiac stromal cell quantity. (B) Data on population doubling level obtained during stroma cell amplification (step (ii)) under serum-free conditions. At an average population doubling level between passages of 3.3, according to an average doubling time of 2.1±0.4 days and considering an effective transfer of ~90% of the cells from passage to passage a 3-fold increase of cardiac stromal cell yield per passage can be anticipated and is considered in the calculation of cardiac stromal cell amplification growth area for 5 passages in panel (A); the effective cell doubling over 5 passages was 15-times. 15 passages with a constant cardiac stromal cell phenotype (such as demonstrated in FIG. 3D) would e.g. allow a 50-fold increase of cardiac stromal cell number.

[0400] FIG. 3: Monitoring the phenotype during differentiation and amplification of cardiac stromal cells and precursors thereof. (A) Characterization of cardiac stromal cells by flow cytometry. Expression of markers (Collagen 1, Vimentin, CD90, CD73, and CD44 ) after 17 days in culture according to steps (i) a1 and (i) a2) (and optionally steps (i)* and (i**) as set out in FIG. 8) of the disclosed method. 95% of the cells were CD90 positive; 99% of the cells were CD44 positive; 99% of the cells were CD73 positive; 97% of the cells were collagen-1 positive; 98% of the cells were vimentin positive. The negative control showed only 3% background signal. (B) Characterization of cardiac stromal cells by immunofluorescence. Expression of Vimentin, anti-human fibroblast antibody (clone TE-7, Millipore), and collagen-1 (Abcam) in the top panel. The corresponding nuclei labelling is presented in the bottom panel. Bars: 50 .Math.m. (C) Expression of surface markers CD90, CD73, and CD44 as well as the intermediate filament VIM (vimentin) at indicated steps of the differentiation protocol. After completion of step (i*) (culture day -7; please refer to FIG. 8) more than 90% of the cells expressed CD90, less than 10% of the cells expressed CD73, less than 10% expressed CD44 and more than 40% expressed vimentin. After completion of sub-step (i a1 - culture day 4) more than 50% of the cells expressed CD90; less than 50% of the cells expressed CD73; less than 30% of the cells expressed CD44, and more than 80% of the cells expressed vimentin. On culture day 17 (during step (i) a2)) the markers CD90, CD73, and vimentin were expressed in more than 90% of the cells, the marker CD44 was expressed in more than 80% of the cells. After completion of step (ia2 - culture day 17) the markers CD90, CD73, CD44, and vimentin were expressed in more than 90% of the cells. (D) Expression of surface markers CD90, CD73, and CD44 were stable in >90% of the cells during step (ii), i.e., amplification. Data was obtained at passage 9, i.e., 5 passages under the amplification step (ii), demonstrating the stability of the cardiac stromal cell phenotype after extensive passaging manifested in maintained expression (for comparison, data shown at passage 4 in FIGS. 3A and 3C; Data for FIG. 3A was obtained from independent experiments to the data obtained in FIGS. 3C and 3D.

[0401] FIG. 4: Expression profiles of different cell populations. (A) Gene expression assessed by RNA sequencing with transcript abundance displayed as Reads Per Kilobase per Million mapped reads (RPKM) of CD140a, vimentin (VIM) and collagen-1 (COL1A1) in undifferentiated iPSC, iPSC derived cardiomyocytes (CM), iPSC-derived stromal cells, and heart derived primary cardiac fibroblasts (as control). Average RPKM values are indicated in the panels. N=2-5 per group. (B) Gene expression assessed by RNA sequencing with transcript abundance displayed as Reads Per Kilobase per Million mapped reads (RPKM) of CD90, CD73 and CD44 in undifferentiated iPSC, iPSC derived cardiomyocytes (CM), iPSC-derived stromal cells (StC), and heart derived primary cardiac fibroblasts (as control). Average RPKM values are indicated in the panels. N=2-5 per group.

[0402] FIG. 5: Characterization of cardiac stromal cells by transcriptome analysis. (A) The transcriptomes of cardiac stromal cells (cStC, n=2), primary cardiac fibroblasts (n=3), neonatal skin fibroblasts (n=2), and gingival fibroblasts (n=3) were compared. (A) Principal component analysis indicated that 46% of total variance could be attributed to principal component 1. (B) Detailed analysis of the top 100 variance-causing genes of PC1 as well as characteristic genes from literature indicated that the expression patterns of stromal cells were more comparable to primary cardiac fibroblasts than skin and gingiva fibroblasts. This analysis shows that the expression pattern of primary cardiac fibroblasts and cardiac stromal cells according to the differentiation/amplification method were remarkably similar. (C) Upper panel: Induction of pro-collagen protein and smooth muscle actin and CTGF by TGFβ1 and/or Angiotensin II. Lower panel: Western Blot analysis showing that profibrotic stimuli induce myofibroblast markers (smooth muscle acting and CTGF) in cardiac stromal cells. Expression of myofibroblast proteins in cardiac stromal cells after incubation with Angiotensin II (Ang II) and/or TGFβ1 at the indicated concentrations for 24 hrs. Stable expression of periostin (POSTN) demonstrates a cardiac fibroblast specific phenotype. Tubulin was used as loading control. Furthermore, cardiac stroma cells under non-stimulated conditions serve as controls (see first and last lane of the Western Blot).

[0403] FIG. 6: Use of cardiac stromal cells in engineered heart myocardium and engineered connective tissue. (A) Cardiac stromal cells support generation of functional engineered human myocardium (EHM). Generated (iPSC derived) cardiac stromal cells (iPSC-cStC) and primary cardiac fibroblasts (cardiac fibs; acquired from Lonza) were compared as supplements to EHM preparation according to Tiburcy et al. 2017, i.e., EHM comprised of 70% cardiomyocytes and 30% cardiac fibroblasts or cardiac stromal cells (n=4/group). Force of contraction (FOC) of EHM was assessed under isometric conditions, at extracellular calcium concentration of 2 mmol/L, and under electrical field stimulation at 1.5 Hz using standard protocols (Tiburcy et al. 2017). Mean ± Standard Error of the mean is indicated in the columns. (B) Stiffness of engineered connective tissue (ECT) made from primary cardiac fibroblasts or cardiac stromal cells. ECT were treated with 5 ng/ml TGFb1 for 5 days. n=3 per group, *p>0.05 by 1-way ANOVA and Tukey’s post-hoc test. Stiffness was measured by destructive tensile stress test using an RSA-G2 rheometer (TA Instruments) and calculated from the slope of the stress-strain curve as Young’s modulus (E; Santos et al. 2019). Mean ± Standard Error of the mean is indicated in the columns.

[0404] FIG. 7: Amplification step (ii) requires an extracellular matrix protein. The cells were cultivated for passage 5 and passage 6 (during the amplification step (ii)) on either vitronectin, laminin (LN-521) gelatin, fibronectin, Matrigel or on uncoated plates. Growth was detected for all coatings, but no growth was detected on uncoated plates. The comparison shows that coating with an extracellular matrix protein is essential for cardiac stromal cell amplification. The coating should preferably be by making use of a defined substance, such as but not limited to vitronectin, Laminin-521, gelatin, fibronectin; but also undefined coatings such as Matrigel may be applied. The images were obtained by bright field microscopy. Bars: 50 .Math.m

[0405] FIG. 8: Scalable stromal cell production with stable phenotype. Overview and chronological sequence of an example for the derivation of cardiac stromal cells from a human pluripotent stem cell (hPSC) source. For inducing mesodermal differentiation, step (i*), the cells were cultured on laminin (LN-521) coated plates in RPMI medium, supplemented with B27 without insulin, ascorbic acid and pyruvate. The medium further comprised 10 ng/ml BMP4, 3 ng/ml ActA, 1 .Math.M CHIR and 5 ng/ml FGF2. This step (i*) took 6 days followed by passaging to the next step (i**). For inducing the epicardial differentiation, step (i**), the cells were cultured on laminin (LN-521) coated plates in RPMI medium, supplemented with B27 with insulin, ascorbic acid, and pyruvate. The medium further comprised 50 ng/ml BMP4, 1 .Math.M CHIR and 4 .Math.M retinoic acid. This step (i**) takes 7 days (days 6-13), without additional passaging into the next step (i a1). For inducing epithelial-mesenchymal transition (EMT) (step i a1), the culture medium was switched to KO DMEM medium comprising FGF2, VEGF, Glutamine and Eukaryotic Cell Culture Medium (ECCM) supplement comprising Ascorbic Acid, Insulin, Transferrin, Albumin, and Selen. An especially preferred ECCM comprises the ingredients as listed in Table 2. This step took in total 17 days, but may have been extended to up to 21 days (days 0-17 or up to 0-21 of the method) and could be subdivided into two sub-steps, taking 4 days (days 0-4 of the method; step i a1) and 13-17 days (days 4-17 or up to 4-21; step i a2), respectively: During the first step (i a1), the medium is further comprising 1 .Math.M CHIR. After 4 days (completion of step (i) a1), cells were passaged to continue the now passage 2 culture on vitronectin (VTN)-coated plates without further 1 .Math.M CHIR supplementation to complete the EMT process (step (i) a2). The cells were passaged again onto vitronectin (VTN)-coated plates (passage 3) on day 24 of the method. For amplifying the cell number (amplification step ii), the cells were cultured on vitronectin (VTN)-coated plates in KO DMEM medium comprising ECCM, FGF2, VEGF, and glutamine (Gln); from passage 4 onwards.

[0406] FIG. 9: TGFbeta inhibition stabilizes cardiac stromal cell growth. TGFbeta inhibition prevents conversion of proliferating mammalian fibroblasts to mammalian myofibroblasts with reduced proliferation rate. Cardiac stromal cells from Rhesus macaque were investigated, because of their particular propensity for myofibroblast conversion and to exemplify the effect of TGFbeta inhibition on mammalian iPSC-derived cardiac stromal cells. Passage 4 stromal cell morphology 1 day and 6 days after passage in the (A) absence or (B) presence of TGFbeta inhibitor SB431542 (10 .Math.M). Bars: 20 .Math.m. (C) Non-human primate cardiac stromal cells were immunostained with anti-alpha smooth muscle actin antibody and analyzed by flow cytometry. Histograms of alpha smooth muscle actin signal in cells cultured in medium comprising 10 .Math.M TGFbeta inhibitor SB431542 (+SB431542) and cells cultured without TGFbeta inhibitor SB431542 (-SB431542) in comparison to cells with isotype IgG2a control. The table summarizes the mean fluorescence intensity (MFI) of alpha smooth muscle actin per cell. (D) Cumulative population doubling level (PDL) obtained during amplification (steps (i) a2) and (ii)) of rhesus macaque stromal cells in presence or absence of TGFbeta inhibitor SB431542 (10 .Math.M).

EXAMPLES

[0407] The following examples are intended to illustrate the invention further, but are not limited to it. The examples describe technical features, and the invention also relates to combinations of the technical features presented in this section.

Example 1: Generation of Cardiac Stromal Cells in High Quantity, High Quality and Under Serum-Free Conditions

[0408] It has been demonstrated that cardiac stromal cells are essential for the function of engineered heart tissue/engineered myocardium. Protocols to successfully obtain engineered heart tissue are known in the art from e.g. Zimmermann et al. (2006) and Tiburcy et al. (2017). However, the generation of engineered heart tissues for clinical use is dependent on the availability of cardiac stromal cells and myocytes, which are mixed during the generation engineered heart muscle. The bottle neck for these protocols is often the provision of a large quantity of cardiac stromal cells. The inventors optimised the cardiac differentiation/amplification method in order to be able to provide not only a large quantity of cardiac stromal cells but also a high quality of cardiac stromal cells. For clinical use, reproducibility of the method is essential. Reproducibility is best achieved if the method is serum-free so that all media components are known. Furthermore, serum-free conditions allow clinical use, as contaminations from undefined sources, such as serum or matrigel, can be excluded. Such a method was sought for a long time. Especially the amplification of cardiac stromal cells posed a particular difficulty to scientists in the field, as it has been unclear how serum supplementation and matrigel coating could be replaced with defined component. Thus, the inventors developed the first serum-free differentiation and amplification method, wherein cardiac stromal cells can be generated from human pluripotent stem cells in high quantity, quality and under serum-free conditions. In this procedure, a specific temporal sequence of active substances (small molecules as well as inhibitors and stimulators) was used, which induce the differentiation of epicardial cells (optionally starting from human pluripotent stem cells) into cardiac stromal cells and allow mass amplification of these cardiac stromal cells in order to obtain large quantities.

[0409] The serum-free differentiation/amplification method is depicted in FIGS. 1A and 8. A detailed step-by-step method is provided in the following. Details on stock preparations, medium preparations, storage conditions, and suppliers are listed in the Materials section at the end of Example 1 below. It is also generally noted for all steps that the skilled person is able to determine the amount of medium required for a given size of tissue culture flask. When culturing cells, the cells were always covered with the given medium.

[0410] Epicardial cells can be obtained as also disclosed herein or by any other method known in the art. Exemplary protocols in the art to obtain epicardial cells can be found for example in Witty et al. (2014) or as described in Schlick (2018) Doctoral Thesis, November 2018, University of Göttingen. For further assistance, a detailed protocol in order to obtain epicardial cells is also described herein.

[0411] When starting from epicardial cells (day 0) the epithelial-mesenchymal transition is induced comprising two sub-steps: (i) a1) and (i) a2). In step (i) a1), the medium was StC-EM+C medium and the cells were on cultivated in laminin coated plates. The medium was replaced with fresh StC-EM+C on day 1 and the cells were incubated with said medium until day 4 and thereby the first sub-step (i) a1) was completed. The second sub-step started on day 4 with passaging the cells onto vitronectin coated plates. Specifically, accutase was warmed to room temperature and versene was warmed to at 37° C. The cells were washed once with PBS (6 ml for a T75 flask), and then incubated with accutase (6 ml for a T75 flask) for 10-20 minutes. Then, versene (6 ml for a T75 flask) was added to the accutase and incubated for 5 min. StC-EM medium (12 ml for a T75 flask) was added to the pool cells and the cells were centrifuged at 300xg for 5 min at room temperature. The cell number was determined and the cells were diluted with StC-EM medium in order to obtain a density of 3x10E6 cells in 15 ml so that the cells can be plated. Media concentrations are described in detail in the materials section below. On days 7 and 9, the medium was replaced by fresh StC-EM medium. Due to cell division, the cells were passaged again on day 11 on vitronectin plates, as just described. On days 14 and 16 the medium was again replaced by fresh StC-EM medium. On day 17, step (i) a2) of the protocol was completed and cardiac stromal cells were generated. These cardiac stromal cells can then be harvested and used for tissue engineering such an engineered human myocardium (EHM) and engineered connective tissue (ECT) or can be frozen. However, in order to obtain a large quantity of cardiac stromal cells, the cells can be amplified further.

[0412] For amplifying the cell number of cardiac stromal cells, step (ii), the cells were passaged using accutase and versene as described for step (i) a2) of the protocol on day 17 and cultured in StC-EM medium. The medium was exchanged every second day and further passaging was performed every week during step (ii). At least six passages for cardiac stromal cell amplification can be performed but step (ii) can be principally performed indefinitely. Cardiac stromal cells can also be harvested and used for tissue engineering after every passage and were characterized by the maintained expression of CD90, CD44 and CD73 in at least 90% of the cells.

[0413] In order to obtain epicardial cells from pluripotent stem cells, the following protocol can be used: The human pluripotent stem cells were seeded on day -17 on Laminin (LN-521) coated plates and cultivated in the presence of iPS Brew supplemented 10 .Math.M Rock Inhibitor. The cells were plated in order to obtain a confluent layer of cells on day -13. The optimal cell count for seeding must be determined individually for each cell line. The medium was exchanged on day -14 with iPS-Brew medium without Rock inhibitor.

[0414] For inducing mesodermal differentiation of the pluripotent stem cells (step (i*)), the cells were washed once with RPMI medium and then further cultivated in StC-IM medium (as described in detail in the materials section below). On days -12, -11, and -10, the medium was replaced by fresh StC-IM medium.

[0415] For inducing epicardial differentiation (step (i**)), the cells were passaged on day -7 onto Laminin LMN-521 coated plates. In order to do so, TrypLE was warmed to 37° C., the medium was removed, cells were washed with TrypLE (4 ml for a T75 flask), and then incubated in fresh TrypLE (4 ml for a T75 flask) for 3-5 minutes. Then, qStC-SM medium was added (e.g. 9 ml) and the cells were centrifuged at 300×g for 5 min at room temperature. The cell number was determined and the cells were diluted with StC-SM medium in order to obtain a density of 2×10E6 cells in 15 ml so that the cells can be plated. Media concentrations are described in detail in the materials section below. On days -5 and -3, the medium was replaced by fresh StC-SM medium. The completion of step (i**) (epicardial differentiation) has been experimentally assessed by the expression of Wilms Tumor antigen 1 (WT1) using immunofluorescence (FIG. 1C). A clear expression of WT-1 was detected in the nuclei of the cells.

[0416] One example providing exact media volumes for a small scale and large scale generation of cardiac stromal cells is provided in the following table:

TABLE-US-00001 Day Medium Small scale Large scale -17 Split cells on Laminin-coated plate in iPS Brew+10 .Math.M Rock Inhibitor; 1×T75 with LMN coating in 24 ml 1×T75 -14 Medium change iPS-Brew 1×T75 with 12 ml 1×T75 with 12 ml -13 Wash once with 15 ml RPMI, then add 21 ml StC-IM 1×T75 with 21 ml 1×T75 with 21 ml -12 StC-IM 1×T75 with 21 ml 1×T75 with 21 ml -11 StC-IM 1×T75 with 21 ml 1×T75 with 21 ml -10 StC-IM double-feed 1×T75 with 36 ml 1×T75 with 36 ml -7 Passage 1: TypLE protocol, see below 1×T75 with LMN coating 2×10E6 cells in 15 ml StC-SM 6×T225 with 6×10E6 per T225 in 45 ml StC-SM -5 StC-SM 1×T75 with 15 ml 6×T225 with 45 ml -3 StC-SM 1×T75 with 15 ml 6×T225 with 45 ml 0 Switch to StC-EM+C 1×T75 with 15 ml 6×T225 with 45 ml 2 StC-EM+C 1×T75 with 15 ml 6×T225 with 45 ml 4 Passage 2: Accutase/Versene protocol, see below 1×T75 with VTN coating; 3×10E6 cells in 15 ml StC-EM 12×T225 with 9×10E6 in 45 ml StC-EM 7 StC-EM 1×T75 with 15 ml 12×T225 with 45 ml 9 StC-EM 1×T75 with 15 ml 12×T225 with 45 ml 11 Passage 3: Accutase/Versene protocol, see below 1×T75 with VTN coating; 3×10E6 cells in 15 ml 12×T225 with 9×10E6 in 45 ml 14 StC-EM 1×T75 with 15 ml 12×T225 with 45 ml 16 StC-EM 1×T75 with 15 ml 12×T225 with 45 ml 17 Passage 4 or harvest for freezing: Accutase/Versene protocol, see below 3×T75 with 2×10E6 cells each in 15 ml 36×T225 with 6×10E6 in 45 ml >17 Expand by weekly passages

Materials

[0417] Materials, suppliers, and storage conditions: [0418] iPS Brew Basalmedium, Miltenyi Biotec, Cat No.170-076-317 [0419] iPS Brew Supplement R, Miltenyi Biotec, Cat No.170-076-318 [0420] (CTS) Vitronectin, Thermo Scientific [0421] Rock Inhibitor (Stemolecule Y27632), Stemgent, Cat No. 04-0012-10, -20° C. [0422] Versene (1x), Gibco, Cat No. 15040-033, 100 ml [0423] PBS (1x), Gibco, A1285601 [0424] RPMI 1640 with Glutamax, Invitrogen, Cat No. 61870-010, 4° C. [0425] 100X sodium pyruvate, Invitrogen, Cat No. 11360, 4° C. [0426] L-ascorbic acid 2 phosphate sesquimagnesium salt hydrate (ASC), Sigma, Cat No. A8960-5G, -20° C. [0427] Activin A, CellGenix, 1022-050 GMP or Activin A, R&D, -20° C. [0428] BMP4, Peprotech, AF-120-05ET or BMP4, R&D, -20° C. [0429] bFGF, Peprotech, GMP100-18B or bFGF, Peprotech RUO, -20° C. [0430] VEGF165, GMP100-20 or VEGF165, Peprotech, RUO, -20° C. [0431] CHIR, Stemgent, Cat No. 04-0004, -20° C. [0432] (CTS) B27 supplement, Thermo Scientific, -20° C. [0433] Retinoic acid, Sigma, Cat No. R2625 [0434] (CTS) TrypLE select, Thermo Scientific, 4° C. [0435] (CTS) KO DMEM, Thermo Scientific [0436] (CTS) KO serum replacement, Thermo Scientific [0437] Accutase, Thermo Scientific [0438] LMN-521, Biolamina [0439] Glutamin, Thermo Scientific [0440] CryoSure DMSO, WAK, WAK-DMSO-10, Room Temperature [0441] Sterile filters 0.2 .Math.M (Sartorius Minisart RC25, DMSO-resistant)

Stocks at (-20° C.)

[0442] BMP4 stock at 10 ug/ml [0443] Activin A stock at 10 ug/ml [0444] bFGF stock at 10 ug/ml [0445] VEGF stock at 5 ug/ml [0446] CHIR stock at 10 mM in DMSO [0447] Rock Inhibitor (Stemolecule Y27632), stock at 10 mM in DMSO [0448] ASC-2-P stock at 200 mM in water (ASC) [0449] Retinoic acid (Sigma: R2625): 8 mM stock: Dissolve 50 mg RA in 20.8 ml DMSO, sterile filter using a DMSO-resistant filter and aliquot with 500 .Math.l in 0.5 ml tubes (store at -80° C. for up to 6 months); use at 4 .Math.M > 0.5 .Math.l/ml medium

Laminin (LMN) Coating

[0450] LMN-521 solution was thawed (100 .Math.g/ml) slowly at +2° C. to +8° C. LMN 521 Laminin- (100 .Math.g/ml) was diluted with 1×DPBSplus (Ca ++ / Mg ++), to obtain a final concentration of 0.9 .Math.g/cm.sup.2. For example for a T75cm.sup.2 flask 15.3 ml PBS and 0.706 ml LMN 521 solution was mixed. The coating was allowed to settle overnight at 4° C. The flasks were warmed to room temperature for 1 hour before use. The supernatant was carefully aspirated (without touching the surface) and medium was immediately added to the flask.

Vitronectin (VTN) Coating:

[0451] Vitronectin dilutions were made and sterile filtered (0.2 .Math.m) before use. For example:

TABLE-US-00002 Dilutions Flask Vitronectin-Stock (0.9 mg/ml) 1x PBS Volume per flask Amount per area T-25 25 .Math.l / flask 4975 .Math.l / flask 5000 .Math.l / flask 0.9 .Math.g /cm.sup.2 T-75 75 .Math.l / flask 14925 .Math.l / flask 15000 .Math.l / flask 0.9 .Math.g /cm.sup.2

[0452] The coating was performed for 1 hour at room temperature. The flasks were adjusted to room temperature for 1 h before use.

Media Preparation

iPS Brew Medium

[0453] iPS Brew medium was prepared according to the manufacturer’s instructions (Miltenyi Biotec).

RPMI Wash Medium

[0454] Cells were washed RPMI 1640 with Glutamax medium (Invitrogen).

Stromal Cell Induction Medium (StC-IM), Used in Step (i*) of the Differentiation/amplification protocol

[0455] The final concentration of the ingredients is provided in FIG. 8. For example, 100ml of said StC-IM medium can be obtained according to the following recipe:

TABLE-US-00003 Component Vol. (for 100 ml) Unit RPMI with Glutamax 97 ml Pyruvate 1 ml Asc A 100 .Math.l B27 minus insulin 2 ml CHIR 10 .Math.l BMP4 100 .Math.l ActA 30 .Math.l bFGF 50 .Math.l

Basal Serum Free Medium, One Component During Step (i**) of the Differentiation/Amplification Protocol (BSFM; 4° C.)

[0456] BSFM was prepared and comprised RPMI 1640 with Glutamax, 1% of 100X sodium pyruvate, 2 % B27 supplement, and 200 uM ASC.

Stromal Cell Specification Medium (StC-SM), Used in Step (i**) of the Differentiation/Amplification Protocol

[0457] The final concentration of the ingredients is also provided in FIG. 8. StC-SM was obtained by supplementing BSFM with 50 ng/ml BMP4 (50 .Math.l stock in 10 ml medium), 4 .Math.mol/L retinoic acid (5 .Math.l stock in 10 ml medium), and 1 .Math.M CHIR (1 .Math.l stock in 10 ml medium).

Stromal Cell Amplification Medium With CHIR (StC-EM+C), Used in Step (i) A1) of the Differentiation/Amplification Medium

[0458] StC-EM+C medium was used during step (i) a1)) of the differentiation/amplification protocol. Said medium was obtained by supplementing KO DMEM medium with 1% Glutamine, 10% (CTS) KO Serum Replacement, 50 ng/ml FGF (50 .Math.l stock in 10 ml medium), 25 ng/ml VEGF and 1 .Math.M CHIR (1 .Math.l stock in 10 ml medium).

Stromal Cell Amplification Medium (StC-EM), Used in Step (i) A2) and (ii) of the Differentiation/Amplification Medium

[0459] StC-EM medium was used during the second sub-step of step (i) a2) and during step (ii) of the differentiation/amplification method. Said medium was obtained by supplementing KO DMEM medium with 1% Glutamine, 10% (CTS) KO Serum Replacement, 50 ng/ml FGF (50 .Math.l stock in 10 ml medium), and 25 ng/ml VEGF.

Example 2: Mass Production of Cardiac Stromal Cells; Amplification by at Least 15 Doublings

[0460] In order to show that cardiac stromal cells can not only be differentiated from epicardial cell (or optionally pluripotent stem cells), but can also be amplified, the inventors developed an amplification step (step (ii) of the protocol of FIGS. 1 and 8). During said step, the cardiac stromal cells retain their cellular properties/phenotype but the cell number amplifies to large quantities under serum-free defined conditions. FIG. 2A shows the potential to which cardiac stromal cells can be expanded starting with a regular tissue culture flask with 75 cm.sup.2. On average a culture flask with 75 cm.sup.2 contains about 20×10E6 iPSC (2,7×10E5). When the cells were amplifying during step (ii) for 5 passages (passages 4-9) after the differentiation step (i) (optionally preceded by step (i*) and (i**)), the cells can amplify to a growth area of 515840 cm.sup.2 (i.e. around 52 m.sup.2). The average output per cm.sup.2 was around 1×10E5 cardiac stromal cells. Therefore, this protocol has the potential to achieve 5.2×10E10 cardiac stromal cells when starting from a confluent layer of human pluripotent stem cells in a tissue culture flask with 75 cm.sup.2 (or 2600 cardiac stromal cells per input iPSC).

[0461] FIG. 2B provides experimental evidence on the cardiac stromal cell population doublings by showing the total number of times the cells have doubled. For each passaging step, 1-1.3×10E4 cells per cm.sup.2 were seeded. This data shows that cardiac stromal cells can by expanded by 15 doublings when step (ii) of the protocol was performed for 5 passages (passages 4-9). From this data, it can be appreciated that the cardiac stromal cells show an average doubling time of 2.1±0.4 days during the amplification phase (step (ii) of the protocol). This corresponds to about a tenfold increase in the number of cells per week (per passage). Furthermore, the amplification of the cardiac stromal cells does not plateau, which is also underlining that the cardiac stromal cells expand continuously under the amplification step conditions. In summary, this data shows that the serum-free defined protocol is suitable for large scale production of cardiac stromal cells, mainly due to the serum-free amplification step (ii) of the method.

Example 3: Generation of a Highly Pure and Homogenous Population of Cardiac Stromal Cells

[0462] To verify the directed differentiation and mass amplification into highly pure and homogenous cardiac stromal cells, the inventors analysed the cardiac stromal cells by several independent methods: Firstly, (1) the expression of extracellular and intracellular markers was analysed by flow cytometry and immunofluorescence (FIGS. 3A-C).

[0463] In order to verify the purity and homogeneity of the generated cardiac stromal cells, the cells were analysed by flow cytometry (FIG. 3A). In flow cytometry, as used here, the expression of cardiac stromal cell markers was measured using immunostaining. In particular, the expression of CD90, CD44, CD73, Collagen 1 and Vimentin after step (i) a2) was assessed in order to characterize the generated cardiac stromal cells. CD90, CD44 and CD73 are extracellular markers, while Collagen 1 and Vimentin are intracellular markers. The proportion of CD90 positive cells was 95%; the proportion of CD44 positive cells was 99%; the proportion of CD73 positive cells was 99% measured against respective isotype controls (IgG1, BD Biosciences). The proportion of Collagen 1 positive cells was 97%; the proportion of Vimentin positive cells was 98%. At the same time the negative control (polyclonal rabbit IgG) only showed a background signal of 3%. This shows that the obtained cardiac stromal cells were highly pure and homogenous. In summary, FIG. 3A shows that the cardiac stromal cells obtained after step (i) a2) were a homogenous population of cells characterized by the expression of CD90, CD44, CD73, Collagen 1 and Vimentin in at least 90% of the cells.

[0464] In order to verify the generation of cardiac stromal cells further, the cells were also analysed by fluorescence microscopy (FIG. 3B). In this process, vimentin, anti-human fibroblast specific protein and Collagen1 were stained immunologically and DNA of the cell nuclei was visualized with Hoechst 33342. All three markers show a homogenous distribution and a high purity of cardiac stromal cells after step (i) a2). Furthermore, the morphology of the cells is typical for cardiac stromal cells.

[0465] In order to not only analyze the expression of CD90, CD73 and CD44 in cardiac stromal cells, said markers were specifically assessed by FACS during the differentiation and amplification protocol (FIG. 3C). After completion of step (i) a1) (culture day 4) more than 50% of the cells expressed CD90; less than 50% of the cells expressed CD73; less than 30% of the cells expressed CD44, and more than 80% of the cells expressed vimentin. On culture day 17 (during step (i) a2)) the markers CD90, CD73, and vimentin were expressed in more than 90% of the cells, the marker CD44 was expressed in more than 80% of the cells. After completion of step (i) a2) (culture day 17) the markers CD90, CD73, CD44, and vimentin were expressed in more than 90% of the cells. The expression profile after the optional step (i*) was also assessed (culture day -7; please refer to FIG. 8) and more than 90% of the cells expressed CD90, less than 10% of the cells expressed CD73, less than 10% of the cells expressed CD44, and more than 40% of the cells expressed vimentin. Furthermore, CD73 was successively more expressed over the course of the differentiation. Similarly, CD44 was also successively more expressed over the course of the differentiation and amplification protocol.

Example 4: Mass Amplification and Maintained Expression of Characteristic Markers of Cardiac Stromal Cells

[0466] In order to show that the amplification step (ii) leads to a maintained expression of the characteristic markers of cardiac stromal cells, a FACS analyses of the cardiac stromal cells at passage 9 was performed (FIG. 3D). At passage 9, the cardiac stromal cells were already amplified over 5 passages (see FIG. 2). As demonstrated in FIG. 3D, more than 90% of the amplified cardiac stromal cells expressed CD90, CD44, and CD73. Thus, cardiac stromal cells retained the phenotype after step (i) a2) stably and maintained expression of said characteristic markers. This showed that said amplification step is suitable for mass amplification of cardiac stromal cells under defined serum-free conditions. Furthermore, it is plausible that cardiac stromal cells can be amplified indefinitely under the conditions of step (ii) as long as the characteristic markers CD90, CD44 and CD73 maintain expression.

Example 5: Genetic Analysis and Analysis of Fibroblast Properties of the Obtained Cardiac Stromal Cells

[0467] As a second and third independent method, the genetic characteristics of the cardiac stromal cells were assessed: Specifically, (2) the gene expression of CD140a, vimentin (VIM), collagen-1 (COL1A1), CD90, CD73 and CD44 in iPSCs, iPSC derived cardiomyocytes (CM), (iPSC-derived) cardiac stromal cells (cStC), and heart derived primary cardiac fibroblasts (as control) were compared (FIG. 4); and (3) a transcriptome analysis was performed to compare and contrast the obtained cardiac stromal cells with primary cardiac fibroblasts and other fibroblast types (FIGS. 5A and B);

[0468] To further demonstrate that the gene expression profile of the cardiac stromal cells also mirrored the identity of cardiac stromal cells, the expression of several marker genes was analyzed and compared in induced pluripotent stem cells (iPSC), iPSC derived cardiomyocytes (iPSC CM), (iPSC derived) cardiac stromal cells as disclosed herein (cStC) and human primary cardiac fibroblasts.

TABLE-US-00004 Expression values of FIGS. 4A and 4B Expression values in RPKM iPSCs iPSC-derived cardiomyocytes (iPSC-derived) cardiac stromal cells (cStC) primary cardiac fibroblasts PDGFRA (CD140a) 1 8 70 204 VIM 13 95 1310 2421 COL1A1 4 47 1540 726 THY1 (CD90) 34 1 30 57 NT5E (CD73) 1 2 227 515 CD44 1 2 142 97

[0469] As can be readily seen in FIGS. 4A and 4B, iPSC and iPSC CM did not express CD140a, vimentin, collagen 1, CD73 or CD44 at a level above background. In contrast, cardiac stromal cells as disclosed herein and primary cardiac fibroblasts expressed CD140a, vimentin, collagen 1, CD73 or CD44. The expression of CD90 could be detected in iPSC, cardiac stromal cells as disclosed herein (cStC) and primary cardiac fibroblasts, while cardiomyocytes lost the expression of CD90 over the course of differentiation. In summary, the gene expression analysis showed by an independent method that also the genetic markers or cardiac stromal cells were expressed in the cells as obtained by the method disclosed herein (FIG. 4). Furthermore, the expression profile of primary cardiac fibroblasts were mirrored by the cardiac stromal cells as obtained by the disclosed method. Furthermore, the gene expression profile was different in every tested marker compared to cardiomyocytes, which underlined the clear difference in expression profile between cardiomyocytes and non-myocytes, i.e. the cardiac stromal cells.

[0470] In order to ensure that specifically cardiac stromal cells were generated, the transcriptome of the obtained cardiac stromal cells was compared to primary cardiac fibroblasts, neonatal skin fibroblasts, and gingival fibroblasts (FIGS. 5A and 5B). RNA was extracted from these cells, sequenced, and the population expression pattern was analysed. When comparing the transcriptomes of the obtained cardiac stromal cells, primary cardiac fibroblasts, neonatal skin fibroblasts, and gingival fibroblasts a principal component analysis was performed (FIG. 5A), which indicated that 46% of total variance could be attributed to principle component 1 (PC1). A further detailed analysis of the top 100 variance-causing RNAs of PC1 as well as characteristic, fibroblast-enriched genes from the inventors previous analyses (FIG. 3B in Tiburcy et al. 2017) indicated that the expression patterns of obtained cardiac stromal cells were to a large extent in agreement with the primary cardiac fibroblasts. The indicated genes in FIG. 5B are the following: DDR2, THY1, TIMP2, TBX4, HOXA11, ISL1, MMP1, CD44, NTSE, BMP4, TCF21, SOX17, TBX20, HEY1, HOXAS, TBX1, COL1A1, FN1, DES, HAND1, PECAM1, WT1, TEX, HAND2, GATA4, ALDH2, POSTN.

[0471] However, the obtained cardiac stromal cells showed clear differences when compared to the neonatal skin fibroblasts and gingival fibroblasts. Therefore, the obtained cardiac stromal fibroblasts were clearly more comparable to primary cardiac fibroblasts than skin and gingiva fibroblasts. Consequently, the transcriptome analysis shows that the specific type of stromal cells, i.e. cardiac stromal cells, were obtained by the protocol according to FIGS. 1 and 8.

[0472] To further analyze the fibroblast-properties of the resulting cardiac stromal cells, the inventors also tested whether the cells could be stimulated to enhance collagen secretion and expression of myofibroblast markers. The conversion to myofibroblasts is for example crucial in the context of wound healing processes/scar formation and said conversion can be induced by e.g. TGFb1 and Angiotensin II. FIG. 5C shows that pro-collagen protein and smooth muscle actin (a-SMA) and CTGF were induced by TGFbeta and or/ Angiotensin II, which are all myofibroblast markers. Tubulin served as a loading control. Consistent with an activated primary cardiac fibroblast phenotype in vitro the cardiac stromal cells showed a robust expression of periostin even under non-stimulated control conditions supporting the RNA expression analyses (FIG. 5B). Thus, the cardiac stromal cells produced according to the method disclosed herein were also capable of secreting collagen as well as the conversion to myofibroblasts.

Example 6: Suitability for Engineered Human Tissue

[0473] In addition to the structural analysis of the cardiac stromal cells, the functional characteristics of the obtained cardiac stromal cells was also assessed by the inventors. Thus, as a fourth (4) independent method, the suitability for engineered tissue, exemplified by the engineering of heart muscle and connective tissue, was also tested by the inventors. Said fourth independent method was carried out by performing contraction experiments, which compare primary cardiac fibroblasts and the obtained cardiac stromal cells in engineered human myocardium (EHM; FIG. 6A) and engineered connective tissue (ECT; FIG. 6B).

[0474] For EHM generation, the primary cardiac fibroblasts or the obtained cardiac stromal cells were mixed with cardiomyocytes and extracellular matrix according to the methods disclosed in Tiburcy et al. (2017). Briefly, these contraction experiments were carried out in organ baths and measure the force of contraction (FOC) of the produced engineered heart muscle/engineered myocardium in response to electrical stimulation (see further details in the methods section herein). Specifically, the FOC of engineered human myocardium (EHM) were made with 70% cardiomyocytes and 30% primary cardiac fibroblasts (black bar) or cardiac stromal cells as disclosed herein (gray bar). Extracellular calcium concentration was 2 mmol/L and the EHM was stimulated with an electrical stimulus of 1.5 Hz and the FOC was measured in millinewtons (mN). The FOC of the EHM generated with primary cardiac fibroblasts was 2.0 mN with a standard error or the mean of 0.1 mN, while the FOC of the EHM generated with the cardiac stromal cells as disclosed herein was 1.9 mM with a standard error of the mean of 0.03 mN. Thus, the FOCs were comparable and were nearly identical. Furthermore, the obtained cardiac stromal cells showed an 3-fold smaller standard error, which emphasises the reproducibility being achieved by using serum-free defined cardiac stromal cells obtained by the method as disclosed herein. This very low standard error underlines the high purity and homogeneity of the obtained cardiac stromal cells according to the method of FIGS. 1 and 8.

[0475] Similarly, ECT was generated with the primary cardiac fibroblasts and cardiac stromal cells as disclosed herein and the result in stiffness upon treatment with TGFb1 was compared. A key feature of connective tissue is that the stiffness increases upon treatment with a pro-fibrotic stimulus like TGFbeta1 (TGFb1), as previously described in Dworatzek et al. 2019. Briefly, these stiffness experiments measure stiffness in the presence or absence of TGFb1 (see further details in the methods section herein). Specifically, the stiffness of ECT was measured in kilo Pascals (kPa). In the absence of TGFb1, the stiffness of ECT generated with primary cardiac fibroblasts and cardiac stromal cells as described herein was 6 and 9 kPa, respectively (SEM: 1 kPa for the primary cardiac fibroblasts and 2 kPa for the cardiac stromal cells). In the presence of 5 ng/ml TGFb1, the stiffness of the ECT generated with cardiac stromal cells was 46 kPa (SEM: 1 kPa), while the stiffness of the ECT generated with primary cardiac fibroblasts was only 21 kPa (SEM: 1 kPa). Thus, the stiffness was significantly larger upon induction with TGFb1 indicative of a fibrotic response. Furthermore, the stiffness was significantly larger in ECT generated with cardiac stromal cells as disclosed herein compared to ECT generated with primary cardiac fibroblasts. This again supports underlines the sensitivity of the cardiac stromal cells to pro-fibrotic stimuli and underlines the suitability of the cardiac stromal cells as disclosed herein for fibrosis modeling in engineered human tissue.

Example 7: Amplification of Cardiac Stromal Cells Can Only Be Performed in the Presence of an Extracellular Matrix Protein and Does Not Work on Uncoated Culture Surfaces Under Strictly Serum-Free Conditions

[0476] The method as depicted in FIG. 8 shows that the amplification step (ii) was carried out on vitronectin coated plates. The inventors also tested whether other ECM proteins would also support the amplification step and whether an amplification without coating the plates would also work in serum-free conditions. To test this idea, the cardiac stromal cells were cultivated on vitronectin, laminin LN-521, collagen in the form of gelatine, fibronectin, matrigel coated and on uncoated plates for passage 5 and 6 and the results are shown in FIG. 7. All ECM protein coated plates (vitronectin, LN-521, gelatine, fibronectin) and the matrigel coated plates supported further amplification of the cardiac fibroblasts, while the cardiac stromal cells on the uncoated plates died in passage 5 on the uncoated plates. Thus, the cardiac stromal cells according to FIGS. 1 or 8 cannot be expanded on uncoated plates. For example, US 2018/0094245 A1 discloses in paragraph [0081], that the cardiac fibroblasts were plated on uncoated plates for amplification. At first sight, the data of FIG. 7 seems contradicting to US 2018/0094245 A1. However, paragraph [0081] of said disclosure also states that the cardiac fibroblasts were plated in the presence of 2% fetal bovine serum. Therefore, the plating step is not serum-free in US 2018/0094245 A1. In fact, it is known that FBS contains variable amounts glycoproteins, which may coat the culture substrate and thus may create an undefined surface coating. Thus it is plausible that the serum during the plating step in US 2018/0094245 A1 supported the further growth of the cardiac fibroblasts on initially uncoated plates. The loss of CD90 suggests a transformation of the fibroblasts and thus the fibroblasts according to US 2018/0094245 A1 have a different phenotype compared to the disclosed cardiac stromal cells. As the present differentiation/amplification protocol is fully serum-free, the defined extracellular matrix protein coating of the plates was demonstrated to be essential. The extracellular matrix protein can be selected from vitronectin, LN-521, gelatine, fibronectin or even matrigel, according to FIG. 7. Furthermore, it is plausible that any other extracellular matrix protein will support the amplification step in the same way as vitronectin, LN-521, gelatine, fibronectin supported the amplification step. Of course, a defined extracellular matrix protein such as vitronectin, LN-521, gelatine or fibronectin is preferred.

Example 8: Inhibition of TGF Beta Signalling Stabilizes Amplification of Cardiac Stromal Cells Under Strictly Serum-Free Conditions

[0477] During in vitro culture cardiac stromal cells typically acquire “stress fibres” indicative of a conversion from a fibroblast to a myofibroblast phenotype. TGFbeta is a strong inducer of fibroblast to myofibroblast conversion and negatively impacts cell proliferation in vitro (Driesen et al. 2014). The appearance of stress fibres is associated with the expression of alpha smooth muscle actin (a-SMA, FIG. 5C). When the inventors tested the derivation and amplification of cardiac stromal cells from non-human primate iPSC (Macaca mullata) they found that in step (i) a2) the non-human primate stromal cells quickly acquired a myofibroblast phenotype as indicated by a substantial amount of stress fibres (FIG. 9A). Therefore, the inventors tested if TGFbeta inhibition would stabilize the amplification of cardiac stromal cells. To inhibit TGFbeta signalling, SB431542, a potent and selective transforming growth factor-β (TGF-β) type I receptor/ALK5 inhibitor was added to StC-EM medium (StC-EM+SB) and compared to cultures with StC-EM medium only.

[0478] Following completion of step i a1 (day 4) the cells were passaged onto vitronectin coated plates (Passage 2). Specifically, accutase was warmed to room temperature and versene was warmed to at 37° C. The cells were washed once with PBS (6 ml for a T75 flask), and then incubated with accutase (6 ml for a T75 flask) for 10-20 minutes. Then, versene (6 ml for a T75 flask) was added to the accutase and incubated for 5 min. StC-EM medium (12 ml for a T75 flask) was added to the pool cells, the cell pool was split in two separate tubes and centrifuged at 300×g for 5 min at room temperature. The cell number was determined and the cells were diluted with StC-EM or StC-EM+SB medium in order to obtain a density of 3×10E6 cells in 15 ml so that the cells can be plated. Media concentrations are described in detail in the materials section below. On days 7 and 9, the medium was replaced by fresh StC-EM or StC-EM+SB medium. The cells were passaged again on day 11 on vitronectin plates, as just described (Passage 3). On days 14 and 16 the medium was again replaced by fresh StC-EM or StC-EM+SB medium. On day 17, cells were passaged again on vitronectin plates, as just described (Passage 4). On days 20 and 22 the medium was again replaced by fresh StC-EM or StC-EM+SB medium. On day 24 cells were harvested by enzymatic digestion as described above and counted to determine the cell number and compare proliferation in StC-EM versus StC-EM+SB medium. In the presence of the TGF beta inhibitor, cardiac stromal cells were smaller and showed less stress fibers (FIGS. 9A, B). Flow cytometry of alpha smooth muscle actin showed a lower mean fluorescence of a-SMA in cells treated with a TGFbeta inhibitor confirming a reduced a-SMA protein content per cell after passage 4 (FIG. 9C), indicating an effective inhibition of myofibroblast conversion. In agreement with this observation proliferation was enhanced by addition of SB431542 (FIG. 9D) indicating that inhibition of TGF beta signaling stabilizes the proliferative state of the cardiac stromal cells. The background staining was determined by an IgG2a isotype control. Comparable cell numbers were stained with an antibody concentration titrated for optimal separation of positive and negative cell populations. Mean fluorescence intensity (MFI) of the total cell population was determined after gating for viable and single cells.

[0479] Materials, suppliers, and storage conditions: [0480] (CTS) Vitronectin, Thermo Scientific [0481] Versene (1x), Gibco, Cat No. 15040-033, 100 ml [0482] PBS (1x), Gibco, A1285601° C. [0483] bFGF, Peprotech, GMP100-18B or bFGF, Peprotech RUO, -20° C. [0484] VEGF165, GMP100-20 or VEGF165, Peprotech, RUO, -20° C. [0485] (CTS) KO DMEM, Thermo Scientific [0486] (CTS) KO serum replacement, Thermo Scientific [0487] Accutase, Thermo Scientific [0488] Glutamine, Thermo Scientific [0489] CryoSure DMSO, WAK, WAK-DMSO-10, Room Temperature [0490] Sterile filters 0.2 .Math.M (Sartorius Minisart RC25, DMSO-resistant [0491] SB431542, TOCRIS bioscience” #1614; Lot:15A/254565. [0492] Primary antibody as used in FIG. 9C: Anti-Actin, a-Smooth Muscle antibody, Mouse monoclonal, clone 1A4, #A5228 [0493] Secondary antibody as used in FIG. 9C: Goat anti-Mouse IgG (H+L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor 488, # A-11029

Stocks at (-20° C.)

[0494] SB431542 stock at 10 mM in DMSO

Stromal Cell Amplification Medium With TGFbeta Inhibitor (StC-EM+SB), Used in Step (i) A2) and (ii) of the Differentiation/Amplification Medium of Non-Human Primate Cardiac Stromal Cells

[0495] StC-EM+SB medium was used during the second sub-step of step (i) a2) and during step (ii) of the differentiation/amplification method. Said medium was obtained by supplementing KO DMEM medium with 1% Glutamine, 10% (CTS) KO Serum Replacement, 50 ng/ml FGF (50 .Math.l stock in 10 ml medium), 25 ng/ml VEGF, and 10 .Math.M SB431542.

Methods

Bright Field Microscopy

[0496] Cells were imaged using a Zeiss Axiovert200 microscope with Zeiss Axiocam MRm camera. Photos were acquired using Zeiss Axiovision software.

Flow Cytometry

[0497] Single live cell suspensions were analyzed after staining with CD44-APC H7 (BD Biosciences), CD90-PE, and CD73-APC (all BD Biosciences) for 10 min at 4° C. After washing, cells were incubated for 10 min in Sytox Blue Dead Cell Stain (Life Technologies, 5 nmol/L) to exclude dead cells. Vimentin (abcam) and Collagen 1 (abcam), and alpha smooth muscle actin (Sigma) were stained in 4% formalin fixed cell suspensions for 45 min at 4° C. followed by staining with appropriate secondary antibodies and Hoechst-33342 (Life Technologies, 10 .Math.g/ml) to detect nuclear DNA for 30 min at 4° C. Cells were run on either a LSRII SORP Cytometer (BD Biosciences) or CytoFlex Cytometer (Beckman Coulter). At least 10,000 events were analyzed per sample.

Immunofluorescence Staining

[0498] Cells were fixed in 4% formalin (Histofix, Roth). After 3 washes with PBS, cells were incubated with primary antibodies in PBS, 5% fetal bovine serum (Thermo Scientific), 1% bovine serum albumin, 0.5% Triton-X (both Sigma-Aldrich) for 2 hours at room temperature or overnight at 4° C. The antibodies used were Vimentin (abcam), Anti-human fibroblast (clone TE-7, Millipore), collagen 1 (abcam), and Wilms Tumor Antigen 1 (WT1, abcam). After several washes, appropriate secondary antibodies and/or phalloidin (to label f-actin) and Hoechst-33342 (Life Technologies, 10 .Math.g/ml) to detect nuclear DNA were added for 1 hour at room temperature. Immunostainings were imaged using a Zeiss LSM710 confocal microscope.

RNA Sequencing (Transcriptome Analysis)

[0499] RNA was extracted using the TRIZOL method (Thermo Scientific). Quality and integrity of RNA was assessed with the Fragment Analyzer from Advanced Analytical by using the standard sensitivity RNA Analysis Kit (DNF-471). RNA-seq libraries were prepared using a modified strand-specific, massively-parallel cDNA sequencing (RNA-Seq) protocol from Illumina, the TruSeq Stranded Total RNA (Cat.No. RS-122-2301). Libraries were sequenced on a HiSeq 4000 platform (Illumina) generating 50 bp single-end reads (30-40 Mio reads/sample). Sequence images were transformed with Illumina software BaseCaller to BCL files, which was demultiplexed to fastq files with bcl2fastq v2.17.1.14. The quality check was done using FastQC (version0.11.5, Babraham Bioinformatics). Sequence reads were aligned to the human genome reference assembly (UCSC version hg38) using Star. For each gene, the number of mapped reads was counted using FeatureCounts. Raw counts were normalized and transformed to log2CPM values. Using a Z-scale normalized matrix a heatmap was generated using the heatmap function in R applying a euclidean distance clustering algorithm. Reads Per Kilobase per Million mapped reads (RPKM) were calculated based on Ensembl transcript length using biomaRT (v2.24).

EHM Generation and Analysis

[0500] EHM generation has been described in Tiburcy et al. (2017) in detail. Briefly, for EHM generation single cells suspension of iPSC-derived cardiomyocytes, primary cardiac fibroblasts, and cardiac stromal cells, as obtained by the method described herein, were prepared. Equal volumes of collagen (0.9 mg/ml) and concentrated serum-free medium (2× RPMI, 8% B27 without insulin, 200 U/ml penicillin, and 200 mg/ml streptomycin) were mixed on ice. After pH neutralization by dropwise addition of 0.1 N NaOH either 1×10E6 cardiomyocytes with 0,5×10E6 primary cardiac fibroblasts (Lonza) or 1x10E6 cardiomyocytes with 0,5×10E6 cardiac stromal cells, as obtained by the method described herein, were added to the collagen mix. After 60 min of hydrogel formation at 37° C. EHM culture medium was added: Iscove’s medium with 4% B27 without insulin, 1% non-essential amino acids, 2 mmol/l glutamine, 300 .Math.mol/l ascorbic acid, 100 ng/ml IGF1 (AF-100-11), 10 ng/ml, FGF-2 (AF-100-18B), 5 ng/ml VEGF165 (AF-100-20), 5 ng/ml TGF-β1 (AF-100-21C), and P/S (all growth factors were obtained from PeproTech).

[0501] After 4 weeks of EHM maturation on flexible posts EHM were isometrically suspended in organ baths (Föhr Medical Instruments) filled with Tyrode’s solution (in mmol/L: 120 NaCl, 1 MgCl2, 0.2 CaCl2, 5.4 KCl, 22.6 NaHCO3, 4.2 NaH2PO4, 5.6 glucose, and 0.56 ascorbate) at 37° C. and constant bubbling with 5% CO2 and 95% O2. Force of contraction measurements were performed at 1.5 Hz electrical field stimulation with 5 ms square pulses (150 mA) and 2 mM extracellular calcium concentration.

ECT Generation and Analysis

[0502] In general, the ECT was generated as previously described in Santos et al. 2019 and is described in detail therein. Briefly, to generate hECT, 0.3 mg bovine collagen type I (Collagen solutions LLC) was neutralized with 0.1 N NaOH, buffered with 2× DMEM and then mixed with 0.75×10.sup.6 cardiac stromal cells as described herein per tissue (Santos et al. 2019). The mixture was cast into a polystyrene cell culture plate with 48 wells containing two flexible poles each (produced by TPK-Kunststofftechnik GmbH according to the inventor’s patent 2016060314484000DE) and allowed to gel for 1 h in a humidified incubator at 37° C. and 5% CO.sub.2. During culture ECT condense around the flexible poles and remain suspended between them. ECTs were treated without (control) or with 5 ng/ml TGFb1 for 5 days. Stiffness was measured by destructive tensile stress test using an RSA-G2 rheometer (TA Instruments) and calculated from the slope of the stress-strain curve as Young’s modulus (reported as E in kPa; Santos et al. 2019). Briefly, the engineered tissues were transferred from the flexible poles into an organ bath containing PBS at 37° C. and fixed between two custom-made hooks. Tissues were then stretched at a constant linear rate of 0.03 mm/s for human tissues until the point of rupture.

Western Blot Analysis

[0503] For protein extraction, cells lysis was performed in CytoBuster (Merck Millipore) containing cOmplete protease inhibitor cocktail and PhosStop phosphatase inhibitor cocktail (Merck) for 10 min on ice. After a centrifugation step for 30 min at 12,000 g at 4° C., the supernatant was supplemented with 4× SDS-containing Laemmli buffer. The samples were denatured and subjected to SDS-PAGE on 10% polyacrylamide gels, followed by transfer onto Amersham Protran nitrocellulose membranes (GE Healthcare). The membranes were cut according to the molecular weight of the specific proteins and blocked with 1× Rotiblock (Carl Roth) for 1 h at room temperature, followed by a washing step with TBST (10 mM Tris pH 7.4, 150 mM NaCl, 0.1% Tween 20) before probing the membranes at 4° C. overnight with primary antibodies. The primary antibodies were Pro-collagen-1 (Santa Cruz Biotechnologies), alpha-tubulin (Sigma-Aldrich), smooth muscle actin (Sigma-Aldrich), Periostin (Santa Cruz Biotechnologies), and CTGF (Santa Cruz Biotechnologies). Incubation with appropriate secondary peroxidase-coupled antibodies (Sigma-Aldrich) for 1 h at room temperature was preceded and followed by three washing steps with TBST. Finally, secondary antibodies were detected using SuperSignal West Femto Maximum Sensitivity Substrate (Thermo Scientific) and a ChemiDoc MP Imaging System (Bio-Rad).

TABLES

[0504] TABLE-US-00005 Composition of the serum-free supplement ‘B27 minus insulin’ (50x concentration, liquid) ingredients concentration in B27 final concentration in basal medium .Math.g/ml .Math.g/ml Bovine serum albumin, fraction V IgG free, fatty acid poor 125000 2500 Catalase 125 2.5 Glutathion reduced 50 1 Superoxide Dismutase 125 2.5 Humanes Holo-Transferrin 250 5 T3 (triodo-I-thyronine) 0.1 0.002 L-carnitine-HCl 100 2 Ethanolamine 50 1 D+-galactose 750 15 Putrescine 805 16.1 sodium-Selenite 0.625 0.0125 Corticosterone 1 0.02 linoleic acid 50 1 linolenic acid 50 1 Progesterone 0.315 0.0063 Retinylacetate 5 0.1 DL -alpha tocopherole (Vit E) 50 1 DL-alpha tocopherol Acetate 50 1 Biotin 125 2.5 10 ml of ‘B27 minus insulin’ per 500 ml medium corresponds to 2% B27 minus insulin (v/v); ‘B27 minus insulin’ contains the same ingredients as ‘B27’ but is lacking human insulin.

TABLE-US-00006 Composition of the serum-free supplement B27 (50% concentration, liquid) ingredients concentration in 50% B27 final concentration in basal medium at 2% B27 .Math.g/ml .Math.g/ml Bovine serum albumin, fraction V IgG free, fatty acid poor 125000 2500 Catalase 125 2.5 Glutathion reduced 50 1 Human Insulin 156.25 3.125 Superoxide Dismutase 125 2.5 Humanes Holo-Transferrin 250 5 T3 (triodo-I-thyronine) 0.1 0.002 L-carnitine-HCl 100 2 Ethanolamine 50 1 D+-galactose 750 15 Putrescine 805 16.1 sodium-Selenite 0.625 0.0125 Corticosterone 1 0.02 linoleic acid 50 1 linolenic acid 50 1 Progesterone 0.315 0.0063 Retinylacetate 5 0.1 DL -alpha tocopherole (Vit E) 50 1 DL-alpha tocopherol Acetate 50 1 Biotin 125 2.5 10 ml of (50%-)B27 per 500 ml medium corresponds to 2% B27 (v/v);

TABLE-US-00007 Composition of ECCM supplement (100%) Substance concentration (.Math.g/ml) Substance concentration (.Math.g/ml) Glycine 150 Ag.sup.+ 0.0006 L-Histidine 940 Al.sup.3+ 0.0007 L-Isoleucine 3400 Ba.sup.2+ 0.008 L-Methionine 90 Cd.sup.2+ 0.03 L-Phenylalanine 1800 Cr.sup.3+ 0.003 L-Proline 4000 Ge.sup.4+ 0.003 L-Hydroxyproline 100 Se.sup.4+ 0.02 L-Serine 800 Br.sup.- 0.004 L-Threonine 2200 I.sup.- 0.0007 L-Tryptophane 440 Mn.sup.2+ 0.0004 L-Tyrosine 77 F.sup.- 0.010 L-Valine 2400 Si.sup.4+ 0.01 Thiamine 33 V.sup.5+ 0.003 Glutathione reduced 10 Mp.sup.6+ 0.006 Ascorbic acid -2-PO.sub.4 (Mg-salt) 330 Ni.sup.2+ 0.0002 Transferrin 55 Rb.sup.+ 0.005 Insulin 100 Sn.sup.2+ 0.0002 sodium-Selenite 0.07 Zr.sup.4+ 0.01 AlbuMAX 83 000 AgNO.sub.3 0.0009 KBr 0.0006 AlCl.sub.3 6H.sub.2O 0.006 KI 0.0009 Ba(C.sub.2H.sub.3O.sub.2).sub.2 0.01 MnCl.sub.2 4H.sub.2O 0.002 CdSO.sub.4 8H.sub.2O 0.08 NaF 0.02 CoCl.sub.2 6H.sub.2O 0.01 Na2SiO3 9H.sub.2O 1 Cr.sub.2(SO.sub.4).sub.3 1H.sub.2O 0.003 NaVO.sub.3 0.006 GeO.sub.2 0.003 (NH.sub.4).sub.6Mo.sub.7O.sub.24 0.06 4H.sub.2O Na2SeP.sub.3 0.007 RbCl 0.007 H2SeO3 0.02 SnCl.sub.2 0.0003 ZrOCl.sub.2 8H.sub.2O 0.02

LIST OF REFERENCES

[0505] US 2018/0094245 A1 [0506] WO 98/30679 [0507] WO2015/040142

[0508] Bao X, Lian X, Hacker TA, Schmuck EG, Qian T, Bhute VJ, Han T, Shi M, Drowley L, Plowright A, Wang OD, Goumans MJ, Palecek SP. Long-term self-renewing human epicardial cells generated from pluripotent stem cells under defined xeno-free conditions. Nat Biomed Eng. 2016;1. pii: 0003. doi: 10.1038/s41551-016-0003

[0509] Bao X, Lian X, Qian T, Bhute VJ, Han T, Palecek SP. Directed differentiation and long-term maintenance of epicardial cells derived from human pluripotent stem cells under fully defined conditions. Nat Protoc. 2017 Sep;12(9):1890-1900. doi: 10.1038/nprot.2017.080

[0510] Brewer, G. J., Torricelli, J. R., Evege, E. K. and Price, P. J. (1993), Optimized survival of hippocampal neurons in B27-supplemented neurobasal™, a new serum-free medium combination. J. Neurosci. Res., 35: 567-576.

[0511] Chen VC, Couture SM, Ye J, et al. Scalable GMP compliant suspension culture system for human ES cells. Stem Cell Res. 2012;8(3):388-402.

[0512] Chen VC, Ye J, Shukla P, et al. Development of a scalable suspension culture for cardiac differentiation from human pluripotent stem cells. Stem Cell Res. 2015;15(2):365-375.

[0513] Didié M, Christalla P, Rubart M, Muppala V, Döker S, Unsöld B, El-Armouche A, Rau T, Eschenhagen T, Schwoerer AP, Ehmke H, Schumacher U, Fuchs S, Lange C, Becker A, Tao W, Scherschel JA, Soonpaa MH, Yang T, Lin Q, Zenke M, Han DW, Schöler HR, Rudolph C, Steinemann D, Schlegelberger B, Kattman S, Witty A, Keller G, Field LJ, Zimmermann WH (2013) Parthenogenetic Stem Cells for Tissue Engineered Heart Repair. J Clin Invest 123:1285-1298.

[0514] Driesen RB, Nagaraju CK, Abi-Char J, Coenen T, Lijnen PJ, Fagard RH, Sipido KR, Petrov W. Reversible and irreversible differentiation of cardiac fibroblasts. Cardiovasc Res. 2014 Mar 1;101(3):411-22.

[0515] Dworatzek E, Mahmoodzadeh S, Schriever C, Kusumoto K, Kramer L, Santos G, Fliegner D, Leung YK, Ho SM, Zimmermann WH, Lutz S, Regitz-Zagrosek V (2019) Sex-specific regulation of collagen I and III expression by 17β-Estradiol in cardiac fibroblasts: role of estrogen receptors. Cardiovasc Res. 115:315-327.

[0516] Frank ND, Jones ME, Yang B, Coeshott C (2019) Evaluation of reagents used to coat the hollow-fiber bioreactor membrane of the Quantum® Cell Expansion System for the culture of human mesenchymal stem cells. Materials Science and Engineering 96, 77-85

[0517] Giacomelli E, Meraviglia V, Campostrini G, Cochrane A, Cao X, van Helden RWJ, Krotenberg Garcia A, Mircea M, Kostidis S, Davis RP, van Meer BJ, Jost CR, Koster AJ, Mei H, Míguez DG, Mulder AA, Ledesma-Terrón M, Pompilio G, Sala L, Salvatori DCF, Slieker RC, Sommariva E, de Vries AAF, Giera M, Semrau S, Tertoolen LGJ, Orlova VV, Bellin M, Mummery CL (2020) Human-iPSC-Derived Cardiac Stromal Cells Enhance Maturation in 3D Cardiac Microtissues and Reveal Non-cardiomyocyte Contributions to Heart Disease. Cell Stem Cell. 26:862-879.

[0518] Hynes RO, Naba A. Overview of the matrisome-an inventory of extracellular matrix constituents and functions. Cold Spring Harb Perspect Biol. 2012;4(1):a004903.

[0519] Ieda M, Tsuchihashi T, Ivey KN, Ross RS, Hong TT, Shaw RM, Srivastava D. Cardiac fibroblasts regulate myocardial proliferation through beta1 integrin signaling. Dev Cell. 2009 Feb; 16(2):233-44

[0520] Ivey MJ, Tallquist MD (2016) Defining the Cardiac Fibroblast: A New Hope. Circ J 80(11):2269-2276

[0521] Iyer D, Gambardella L, Bernard WG, Serrano F, Mascetti VL, Pedersen RA, Talasila A, Sinha S. Robust derivation of epicardium and its differentiated smooth muscle cell progeny from human pluripotent stem cells. Development. 2015 Apr 15;142(8):1528-41.

[0522] Kattman SJ, Alec D. Witty, Mark Gagliardi, Nicole C. Dubois, Maryam Niapour, Akitsu Hotta, James Ellis, Gordon Keller (2011) Stage-Specific Optimization of Activin/Nodal and BMP Signaling Promotes Cardiac Differentiation of Mouse and Human Pluripotent Stem Cell Lines, Cell Stem Cell, Volume 8, Issue 2, 228 - 240

[0523] Li Y, Song D, Mao L, Abraham DM, Bursac N, Lack of Thy1 defines a pathogenic fraction of cardiac fibroblasts in heart failure. Biomaterials 2020; 236

[0524] Naito, H. et al. Optimizing engineered heart tissue for therapeutic applications as surrogate heart muscle. Circulation 114, I72-78 (2006).

[0525] Pinto AR, Ilinykh A, Ivey MJ, Kuwabara JT, D′Antoni ML, Debuque R, Chandran A, Wang L, Arora K, Rosenthal NA, Tallquist MD. Revisiting Cardiac Cellular Composition. Circ Res. 2016 Feb 5;118(3):400-9.

[0526] Santos G, Hartmann S, Zimmermann WH, Ridley A, Lutz S (2019) Inhibition of Rho-associated kinases suppresses cardiac myofibroblast function in engineered connective and heart muscle tissues. J Mol Cell Cardiol. 134:13-28.

[0527] Schlick, Susanne Franziska (2018) Fibroblast-Cardiomyocyte Cross-Talk in Heart Muscle Formation and Function, Doctoral Thesis, in partial fulfillment of the requirements for the degree, “Doctor rerum naturalium (Dr. rer. nat.)”, at the Georg-August University Göttingen, November 2018: https://d-nb.info/1177361744/34

[0528] Schlick SF, Spreckelsen F, Tiburcy M, Iyer LM, Meyer T, Zelarayan LC, Luther S, Parlitz U, Zimmermann WH, Rehfeldt F (2019) Agonistic and antagonistic roles of fibroblasts and cardiomyocytes on viscoelastic stiffening of engineered human myocardium. Prog Biophys Mol Biol 144:51-60

[0529] Takahashi, Kazutoshi et al. Induction of Pluripotent Stem Cells from Adult Human Fibroblasts by Defined Factors. Cell. 2007; 131(5), 861 - 872

[0530] Takahashi K, Yamanaka S. Induction of Pluripotent Stem Cells from Mouse Embryonic and Adult Fibroblast Cultures by Defined Factors. Cell. 2006 Aug 25; 126(4):P663-676

[0531] Thomson J, Itskovitz-Eldor, Shapiro et al. Embryonic Stem Cell Lines Derived from Human Blastocysts. Science. 06 Nov. 1998. 282(5391), 1145-1147

[0532] Tiburcy, M., Hudson, J. E., Balfanz, P., Schlick, S., Meyer, T., Chang Liao, M. L., Levent, E., Raad, F., Zeidler, S., Wingender, E., Riegler, J., Wang, M., Gold, J. D., Kehat, I., Wettwer, E., Ravens, U., Dierickx, P., van Laake, L. W., Goumans, M. J., Khadjeh, S., ... Zimmermann,

[0533] W. H. (2017). Defined Engineered Human Myocardium With Advanced Maturation for Applications in Heart Failure Modeling and Repair. Circulation, 135(19), 1832-1847.

[0534] Tran, N. T., Trinh, Q. M., Lee, G. M., & Han, Y. M. (2012). Efficient differentiation of human pluripotent stem cells into mesenchymal stem cells by modulating intracellular signaling pathways in a feeder/serum-free system. Stem cells and development, 21(7), 1165-1175.

[0535] Wehrhan F, Nkenke E, Melnychenko I, Amann K, Schlegel KA, Goerlach C, Zimmermann WH, Schultze-Mosgau S (2010) Skin repair using a porcine collagen I/III membrane--vascularization and epithelization properties. Dermatol Surg. 36:919-930

[0536] Witty AD, Mihic A, Tam RY, Fisher SA, Mikryukov A, Shoichet MS, Li RK, Kattman SJ, Keller G. (2014) Generation of the epicardial lineage from human pluripotent stem cells. Nat Biotechnol. Oct;32(10):1026-35.

[0537] Yanzhen Li, Daniel Song, Lan Mao, Dennis M. Abraham, Nenad Bursac (2020) Lack of Thy1 defines a pathogenic fraction of cardiac fibroblasts in heart failure, Biomaterials, Volume 236, 119824

[0538] Zhang H, Tian L, Shen M, Tu C, Wu H, Gu M, Paik DT, Wu JC. Generation of Quiescent Cardiac Fibroblasts From Human Induced Pluripotent Stem Cells for In Vitro Modeling of Cardiac Fibrosis. Circ Res. 2019 Aug 16;125(5):552-566.

[0539] Zhou P, Pu WT. Recounting Cardiac Cellular Composition. Circ Res. 2016 Feb 5;118(3):368-70.

[0540] Zimmermann WH, Melnychenko I, Wasmeier G, Didie M, Naito H, Nixdorff U, Hess A, Budinsky L, Brune K, Michaelis B, Dhein S, Schwoerer A, Ehmke H, Eschenhagen T (2006) Engineered heart tissue grafts improve systolic and diastolic function in infarcted rat hearts. Nat Med 12:452-458.