TFAP2 INHIBITION FOR TREATING CARDIAC DISEASE INVOLVING FIBRO-FATTY REPLACEMENT

Abstract

The present invention relates to novel treatments for treating cardiac disease involving fibro-fatty replacement, such as arrhythmogenic cardiomyopathy, atrial fibrillation, myocardial infarction and dilated cardiomyopathy. Such cardiac diseases can e.g. be caused by a mutation in a desmosomal protein such as plakophilin-2 (PKP2). The invention provides for agents for use in the prevention or treatment of such cardiac diseases, wherein the agent is at least one of: a) an agent that causes a reduction in expression in at least one TFAP2 subtype; and, b) an agent that causes a reduction in TFAP2-induced transcription. More preferably the agent is at least one of: a) an inhibitor of TFAP2; and, b) an agent that causes an increase in expression of PKP2.

Claims

1.-15. (canceled)

16. A method for treating or preventing cardiac disease wherein fibro-fatty replacement is part of the disease etiology, wherein the method comprises administering to a subject in need thereof at least one of: a) an agent that causes a reduction in expression in at least one TFAP2 subtype; and, b) an agent that causes a reduction in TFAP2-induced transcription.

17. The method according to claim 16, wherein the agent is at least one of: a) an inhibitor of TFAP2; and b) an agent that causes an increase in expression of PKP2.

18. The method according to claim 16, wherein the agent is (a source of) at least one of a genome editing complex, an antibody, a compound, preferably the agent is a nucleic acid molecule, a siRNA, miRNA, more preferably, the agent is at least one of: a genome editing complex that restores PKP2 deficiency or inactivates TFAP2, a neutralizing antibody against TFAP2 and an siRNA complementary to TFAP2 mRNA.

19. The method according to claim 17, wherein the agent that causes an increase in expression of PKP2 is an inhibitor of Wnt3a, preferably the inhibitor of Wnt3a is selected from an anti-Wnt3a antibody, Tricostatin A, hexachlorophene and niclosamide.

20. The method according to claim 16, wherein the agent is administered to a subject intermittently or continuously.

21. The method according to claim 16, wherein the agent is administered locally to the epicardium/pericardial sac region.

22. The method according to claim 16, wherein TFAP2 expression is reduced by at least 10%, 20%, 30%, 40%, 60%, 80%, or more or wherein TFAP2-induced transcription is reduced by at least 10%, 20%, 30%, 40%, 60%, 80%, or more, preferably the reduction of the reduction of TFAP2 expression or the reduction of TFAP2-induced transcription is determined by PCR or immunostaining.

23. The method according to claim 16, wherein the cardiac disease is at least one of arrhythmogenic cardiomyopathy, atrial fibrillation, myocardial infarction and dilated cardiomyopathy.

24. The method according to claim 16, wherein the cardiac disease is caused by a mutation in a desmosomal protein, preferably wherein the desmosomal protein is plakophilin-2 (PKP2), more preferably wherein the protein is PKP2 and the mutation is c.2013delC.

25. An in vivo, in vitro, or ex vivo method for reducing TFAP2 expression, the method comprising the step of contacting a cell with: a) an agent that causes a reduction in expression in at least one TFAP2 subtype; or, b) an agent that causes a reduction in TFAP2-induced transcription

26. A method for identifying an agent that causes at least one of a reduction in TFAP2 expression, and a reduction in TFAP2-induced transcription, the method comprising the steps of: a) contacting a PKP2-deficient epicardial cell with a candidate agent; b) determining in the cell in a) at least one of the level of TFAP2 expression and the level of TFAP2-induced transcription; c) identifying the agent as an agent that causes a reduction in TFAP2 expression if the level of TFAP2 expression as in determined in b) is less that the level in a corresponding control cell in the absence of the agent, and, d) identifying the agent as an agent that causes a reduction in TFAP2-induced transcription if the level of TFAP2 expression as in determined in b) is less that the level in a corresponding control cell in the absence of the agent.

Description

DESCRIPTION OF THE FIGURES

[0080] FIG. 1: Differentiation of hiPSCs into epicardial cells. (A) Schematic of CRISPR-Cas9-mediated targeting strategy to revert PKP2 c.2013delC mutation. A guide RNA and DNA repair template containing the WT PKP2 sequence were designed to target the PKP2 locus. Blocking mutations were introduced to prevent target re-cutting. The middle panel shows sequencing results in mutant and reverted cells. Mutation site correction is indicated by an arrow. PKP2 c.2013delC and reverted hiPSCs were subsequently differentiated into epicardial cells. (B) Schematic of hiPSC to epicardial cell differentiation protocol using WNT pathway modulators. (C,D) Analysis of WT1 protein expression by (C) Western blot and (D) flow cytometry in PKP2 c.2013delC and reverted hiPSC-epicardial cells. (E) Immunofluorescent staining of epicardial markers in PKP2 c.2013delC and reverted hiPSCs differentiated into epicardial cells.

[0081] FIG. 2: Enhanced fibro-fatty gene expression in PKP2 c.2013delC hiPSC-epicardial cells. (A) Volcano plot of RNAseq data showing differentially expressed genes in in PKP2 c.2013delC versus reverted hiPSC-epicardial cells (n=4). (B-C) ClueGO pathway analysis of (B) downregulated and (C) upregulated genes on RNAseq. (D) qPCR, flow cytometry and western blot analysis of WT1 expression in PKP2 c.2013delC and reverted day 80 hiPSC-epicardial cells. (E) qPCR analysis of the indicated fat and fibroblast markers in PKP2 c.2013delC and reverted hiPSC-epicardial cells. Samples were relatively compared to day 0 c.2013delC hiPSC-epicardial cells (n=3). (F, G) Immunofluorescent staining of WT1, PPARG and ACTA2 in PKP2 c.2013delC and reverted hiPSC-epicardial cells. (H) Oil red-O and hematoxylin staining PKP2 c.2013delC and reverted hiPSC-epicardial cells. Data are represented as mean±SEM. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001. A Student's t-test was used for statistical analysis.

[0082] FIG. 3: AP2 factors drive fibro-fatty signaling in mutant hiPSC-epicardial cells. (A) qPCR analysis of AP2A, AP2B and AP2C in hiPSC-epicardial cells at different time points. Samples were relatively compared to day 0 PKP2 c.2013delC hiPSC-epicardial cells (n=3). (B) Immunofluorescent staining of AP2A and AP2C in day 80 PKP2 c.2013delC and reverted hiPSC-epicardial cells. (C-E) qPCR analysis of the indicated genes following siRNA-mediated knock down of (C) AP2A, AP2B and AP2C in PKP2 c.2013delC hiPSC-epicardial cells (n=3) (D) PKP2 in healthy iPSC-epicardial cells (n=6) (E) PKP2 in healthy hiPSC-cardiomyocytes (n=3). Values were relatively compared to si-scramble-treated samples indicated by the dashed lines. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, ND=not detected. A Student's t-test was used for statistical analysis.

[0083] FIG. 4. JUP knock down induces AP2 and fibro-fatty gene expression. qPCR analysis of the indicated genes following siRNA-mediated knock down of JUP in healthy iPSC-epicardial cells (n=6). Values were relatively compared to si-scramble-treated samples indicated by the dashed lines. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, ND=not detected. A Student's t-test was used for statistical analysis.

[0084] FIG. 5. Single cell RNA sequencing (scRNAseq) reveals TFAP2A-expressing clusters in PKP2 c.2013delC hiPSC-epicardial cells. (A) Schematic of the scRNAseq protocol. Day 70 PKP2 c.2013delC and reverted hiPSC-epicardial cells were dissociated and sorted into single cells for subsequent sequencing and in silico analysis. (B) Bioanalyzer plots of sorted cells showing intact RNA. (C-D) Monocle 2-generated tSNE maps of scRNAseq displaying 5 distinct cellular clusters in (D) and the origin of each cellular. (E) tSNE maps showing expression levels of different epicardial, fibroblast and fat markers among cellular clutsers. (F-G) Pseudotime trajectory-reconstruction of the different cellular clusters. (H-I) TFAP2A expression presented in (I) pseudotime and (J) tSNE plot.

[0085] FIG. 6. TFAP2A mediates epicardial to fibro-fatty signaling through EMT. (A) Western blot analysis of TFAP2A protein levels in PKP2 reverted, PKP2 c.2013delC and TFAP2A siRNA-treated PKP2 c.2013delC hiPSC-epicardial cells. (B) qPCR analysis of the indicated genes following siRNA-mediated knock down of TFAP2A in PKP2 c.2013delC hiPSC-epicardial cells (n=5). Values were relatively compared to si-scramble-treated samples indicated by the dashed lines using Student's t test. (C) qPCR analysis of CDH1/CDH2 expression ratio in the culture timeline and of PKP2 c.2013delC and reverted hiPSC-epicardial cells. (D) Control hiPSC-epicardial cells were triggered to undergo EMT and subsequently transfected with TFAP2A-targeting siRNAs. qPCR analysis of the indicated genes after transfection are shown below. (*P<0.05, **P<0.01, ***P<0.001, ****P<0.0001).

[0086] FIG. 7. Cardiac explants from ACM patients show epicardial activation and TFAP2A induction. Upper images in each panel show Masson Trichome stainings of healthy and diseased hearts depicting myocardium in red, fibrosis in blue and adipose tissue in white. Zoomed in images below were stained with WT1 or TFAP2A. Epi indicates epicardium and Myo indicates myocardium. Scale bars in each panel: Upper=10 mm, middle=1 mm, lower=50 um.

DESCRIPTION OF THE SEQUENCES

[0087]

TABLE-US-00001 TABLE 1 Sequences SEQ ID NO: Name  1 TFAP2A variant 1 (Nucleotides)  2 TFAP2A variant 2 (Nucleotides)  3 TFAP2A variant 3 (Nucleotides)  4 TFAP2B (Nucleotides)  5 TFAP2C (Nucleotides)  6 TFAP2D (Nucleotides)  7 TFAP2E (Nucleotides)  8 Guide 1  9 Guide 2 10 Repair template sequence 45 TFAP2A variant 1 (Amino acid) 46 TFAP2A variant 2 (Amino acid) 47 TFAP2A variant 3 (Amino acid) 48 TFAP2B (Amino acid) 49 TFAP2C (Amino acid) 50 TFAP2D (Amino acid) 51 TFAP2E (Amino acid)

Examples

Example 1

Material and Methods

CRISPR-Cas9 Targeting

[0088] PKP2 mutant (c.2013deIC) iPSCs were kindly provided by Huei-Sheng Vincent Chen. c.2013delC mutation was reverted into wild type (introducing C nucleotide) using CRISPR-Cas9. Two guide RNAs were used to target the PKP2 genomic locus (Guide 1: ATACCAGGACGTGCCGATGC (SEQ ID NO: 8), Guide 2: CCTCCGGCATCGGCACGTCC (SEQ ID NO:9) and a repair template was used to revert the deletion. A mutation in the PAM sequence was included to prevent re-cutting by the guide RNA. Repair template sequence:

TABLE-US-00002 (SEQ ID NO: 10) 5′ACACTTTTGGCGATCAAGGACAGATACATCCTTATAACAATTGAATGC CACAGCCACTCCACGCCCTTGGGGTTGCTCTTTTCCTCGGGCATCGGCAC GTCCTGGTATTGCTGACCACACACAAAAG3′

Cell Culture

[0089] Human iPSCs were maintained in Essential 8 Medium (Thermofisher Scientific, #A1517001) on Geltrex (Thermofisher Scientific, A1413302)-coated plates. iPSC-epicardial cell differentiation protocol was adapted from Bao et al. 2016.

Immunofluorescence

[0090] Cells were fixed with 4% PFA, blocked with goat serum and incubated for 1 hour with the primary antibodies WT1 (abcam, ab89901), K18 (Life technologies, MS-142-PO), TBX18 (Sigma-Aldrich SAB1412362), ZO1 (Life technologies, 40-2200), PPARG (Santa-cruz sc-7273), ACTA2 (Sigma-Aldrich A5228), AP2A (abcam, ab52222), AP2C (santa-cruz, sc-12762). Cells were subsequently stained with the corresponding Alexa Fluor antibodies (Invitrogen) for 1 hour. Images were taken using the Leica TCS SPE confocal microscope.

Western Blot

[0091] Proteins were isolated from cells using RIPA buffer (50 mM Tris-HCl pH 7.5, 150 mM NaCl, 0.1% SDS, 0.5% sodium deoxycholate (Sigma Aldrich), 1% Triton X-100 (Sigma), protease inhibitor (Roche)) and protein concentration was determined using Bradford assay (BioRad). SDS-PAGE and Western Blot were performed using Mini-PROTEAN Tetra Vertical Electrophoresis Cell with Mini Trans-Blot (Bio-Rad). Membranes were blocked in 3% BSA and subsequently incubated with the primary antibodies WT1 (abcam, ab89901) and GAPDH (Millipore, MAB374). Blots were incubated with Peroxidase-conjugated AffiniPure Rabbit Anti-mouse IgG (IR 315-035-003) and Goat Anti-rabbit IgG secondary antibodies (IR 111-035-003) for 45 min and proteins were visualized using ECL solution (BioRad, #170-5061) on the LAS4000 software program. Western bots were quantified by ImageJ.

Oil Red O Staining

[0092] Cells were stained using the Lipid Oil red 0 staining kit (Sigma, MAK194) according to the manufacturer's instructions.

Flow Cytometry

[0093] Cells were fixed with 70% ethanol and incubated in blocking buffer (PBS, 5% FBS, 1% BSA, 0,5% Triton X-100). Cells were subsequently incubated with the primary antibody WT1 (abcam, ab89901) and the secondary antibody Alexa Fluor 488 donkey anti-rabbit IgG (Thermo Fisher, A21206). WT1+cells were quantified using FACS Aria SORP (BD bioscience).

RNA Isolation and qPCR

[0094] Total RNA was isolated using the RNeasy Mini Kit (Qiagen, #74104) according to the manufacturer's instructions. RNA was reverse transcribed into cDNA using iScript cDNA Synthesis Kit (Bio-Rad, #1708891) and used for qPCR using iQ SYBR Green Supermix (Bio-Rad, #170-8885). Transcript levels were normalized for endogenous loading. Primer sequences used are provided in the table 2 below.

TABLE-US-00003 TABLE 2 qPCR primers SEQ SEQ ID NO Gene name Forward primer ID NO Reverse primer 11 TFAP2A CTCGATCCACTCCTTACCTCAC 12 ATTGCTGTTGGACTTGGACAG 13 TFAP2B TAACAGCGGCATGAATCTATTG 14 CAGGAAGCCGTCTTTATTCATC 15 TFAP2C GATCAGACAGTCATTCGCAAAG 16 GTAGAGCTGAGGAGCGACAATC 17 PKP2 TGCTAAAGGCTGGCACAA 18 TAATCGCTGTGCGTGTAGTG 19 WT1 CATGACCTGGAATCAGATGAAC 20 CGTGCGTGTGTATTCTGTATTG 21 UCP1 AGTACAAAAGTGTGCCCAACTG 22 TCGTTTCAGTTGTTCAAAGCAC 23 PPARG GTACTGTCGGTTTCAGAAATGC 24 ATTCAGCTGGTCGATATCACTG 25 PPARGC1A GGTGCAGTGACCAATCAGAA 26 AATCCGTCTTCATCCACAGG 27 COL2A1 CCTGGTGTCATGGGTTTCC 28 GTCCTGCAGCACCTGTCTC 29 POSTN TGCCCTTCAACAGATTTTGG 30 GCAGCCTTTCATTCCTTCC 31 FN1 TCTCATTCAACAAGAAACCACTG 32 TTCACGTCTGTCACTTCCACA 33 CEBPA ACGATCAGTCCATCCCAGAG 34 TTCACATTGCACAAGGCACT 35 EBF3 AACAGAGCAAGATCTGTATGTTCG 36 CCACAACTTTTCTTGTCACAGC 37 RORA CTTTCCCTACTGTTCGTTCACC 38 ACGTTATCTGCTGGAGCTCTTC 39 ACTA2 CCAGCCATGTATGTGGCTATC 40 CCTCATAGATGGGGACATTGTG 41 COL1A2 TGATGGAAAAGGAGTTGGACTT 42 CAGGTCCTTGGAAACCTTGA 43 DLK1 CACGGACTCTGTGGAGAACC 44 GCATTCATAGAGGCCATCGT
siRNA Transfection

[0095] siRNA trilencers were purchased from Origene to target TFAP2A (SR304787), TFAP2B (SR304788), TFAP2C (SR304789), PKP2 (SR303544) and JUP (SR302502). 10 nM siRNA trilencer was used for transfection using Lipofectamine 3000 (Life Technologies, L3000008) according to the manufacturer's instructions.

Results

[0096] Generation of PKP2 c.2013deIC and Reverted hiPSC-Epicardial Cells.

[0097] We made use of CRISPR-Cas9 to repair the PKP2 c.2013delC mutation in human induced pluripotent stem cells (hiPSCs) derived from a patient diagnosed with arrhythmogenic cardiomyopathy (ACM) (FIG. 1a). PKP2 c.2013delC and reverted hiPSCs were subsequently differentiated into epicardial cells (hiPSC-epicardial cells) using WNT pathway modulators. (FIG. 1b). This was confirmed by the time-dependent enhanced expression of the epicardial markers WT1, K18, TBX18 and ZO1 (FIG. 1c-e).

PKP2 c.2013delC hiPSC-Epicardial Cells Display Enhanced Fibro-Fatty Gene Signaling.

[0098] In order to identify the molecular differences between the 2 lines, day 27 PKP2 c.2013delC and reverted hiPSC-epicardial cells were subjected to RNAseq. We identified 567 downregulated (less than −Log 2 fold change) and 939 upregulated (more than Log 2 fold change) genes (FIG. 2a). Gene ontology analysis showed that upregulated genes were mainly involved in cardiomyopathies, cell adhesion pathways and interestingly adipogenesis (FIG. 2c). Having identified molecular differences in the 2 lines despite of their phenotypical similarities, we subjected those cells to long term culturing. When analyzed, day 80 PKP2 c.2013delC, and not the reverted cells, showed a dramatic loss of the expression of the epicardial marker WT1 (FIG. 2d). Those cells profoundly expressed higher levels of the major fat markers PPARG, PPARGC1A and UCP1 as compared to their reverted isogenic controls. A similar expression pattern was also observed in fibroblast markers such as COL2A1, POSTN, FN1 and ACTA2 (FIG. 2e, f). In addition, only PKP2 c.2013delC hiPSC-epicardial cells progressively accumulated lipid droplets as revealed by Oil Red O staining (FIG. 2f). These data indicate that the PKP2 c.2013delC mutation potentially drives a fibro-fatty transition in epicardial cells.

hiPSC-Epicardial Cells Transition into Fibro-Fatty Cells in an AP2-Dependent Mechanism.

[0099] In order to identify the mechanism behind this process, we further analyzed our RNAseq data, by which we identified the Activating Enhancer Binding Protein 2 (AP2) family of transcription factors to be highly induced in PKP2 c.2013delC hiPSC-epicardial cells (FIG. 2c). AP2 factors have been previously identified as master regulators of adipogenesis which indicated their potential in mediating this effect in mutant epicardial cells. Expression of the AP2 factors was further validated by qPCR and immunofluorescent staining (FIG. 3a, b). To investigate whether AP2 factors drive this adipogenic signaling, we set out to use a mixture of targeting siRNAs to knock down AP2A, AP2B and AP2C in PKP2 c.2013delC hiPSC-epicardial cells. After 48 h of transfection, we could detect a significant reduction in AP2 expression as well as in fat and fibroblast markers suggesting an AP2-mediated fibro-fatty signaling in epicardial cells (FIG. 3c). The PKP2-dependence of these findings was further validated in healthy hiPSC-derived epicardial cells treated with siRNAs targeting PKP2, which recapitulated the observations made in the mutant cells (FIG. 3d). To identify whether PKP2 knock down in other cardiac cells would trigger the same effect, we differentiated cardiomyocytes from healthy hiPSCs and performed the same PKP2 knock down experiment. We did not observe any changes in fibro-fatty gene expression in those cells indicating that this effect is epicardial cells-specific (FIG. 3e). To validate whether the induction of AP2 factors and fibo-fatty genes is triggered by a dysregulation of other desmosomal proteins, we knocked down JUP in healthy hiPSC-derived epicardial cells. We observed a significant increase of AP2 and fibro-fatty genes in si-JUP treated samples (FIG. 4) indicating that this is a common mechanism in ACM.

CONCLUSION

[0100] We generated epicardial cells from induced pluripotent stem cells (iPSCs) of an ACM patient carrying a mutation in the desmosomal gene Plakophilin-2 (PKP2) (c.2013delC) as well as their isogenic control in which the mutation was reverted into wild type using CRISPR-Cas9.

[0101] Both lines exhibited a similar epicardial state when cultured up to 1 month as assessed by qPCR, western blotting, flow cytometry and immunofluorescent staining of epicardial markers. However, when subjected to RNA sequencing, only the mutant line highly expressed genes enriched in adipogenic signaling, which suggested the potential of those cells towards adipocytic differentiation. Indeed, when subjected to long term culturing conditions, mutant PKP2 cells and not their isogenic controls displayed a high expression of adipogenic and fibroblast markers and accumulated lipid droplets as seen by Oil-Red-O staining.

[0102] Follow up experiments showed that AP2 family of genes showed an increase in the PKP2 mutant cells. AP2 genes are known for their roles in in development, cell growth and differentiation, but have not been studied in the heart. Interestingly, a recent study identified those factors to act as regulators of lipid droplet biogenesis.

[0103] To test whether AP2 might be regulating the fibro-fatty phenotype seen in the PKP2 mutant epicardial cells, using siRNAs against AP2 we observed a reduced expression of fibroblast and lipid markers and an increased expression of epicardial markers in the siRNA-transfected cells, indicating that PKP2 mutant epicardial cells are undergoing differentiation into fibroblasts and adipocytes in an AP2-dependent mechanism.

[0104] To show whether the induction of AP2 is a direct effect of a decreased level of PKP2 in the patient cells we used an siRNA against PKP2 in healthy epicardial cells. In doing so we were able to show that PKP2 inhibition leads to an induction in AP2 and fibro-fatty markers.

Example 2

[0105] TFAP2A Mediates Fibro-Fatty Transition in PKP2 c.2013deIC hiPSC-Epicardial Cells.

[0106] In order to gain more insights into the subcellular identities generated in long term cultures of PKP2 c.2013delC and reverted hiPSC-epicardial cells, we subjected day 70 cultures to single cell RNA sequencing (scRNA-seq) (FIG. 5A). After sorting, we obtained >96% live cells and intact RNA as checked on Bioanalyzer (FIG. 5B). Single cell RNA sequencing analysis showed that all cells clustered into 5 distinct cellular clusters as shown in the representative t-Distributed Stochastic Neighbor Embedding (t-SNE) maps (FIG. 5C-D). The differential expression of epicardial, fibroblast and fat markers between the 2 lines was further validated between cellular clusters indicating an epicardial to fibro-fatty switch in the mutant cells (FIG. 5E). We then used Monocle's pseudotime trajectory-reconstruction algorithm to create a branched tree structure representing directions of cellular differentiation (FIG. 5F-G). This analysis could give an indication of the cellular events triggering the epicardial to fibro-fatty cell transition. We generated a list of the top 10 transcription factors which showed a significant differential expression between the branches, among which Transcription factor TFAP2A showed a clear induction in the mutant clusters, suggesting a potential role in driving this phenotype early during differentiation (FIG. 51, J). Furthermore, we used HOMER (Hypergeometric Optimization of Motif EnRichment) to identify transcription factors with enriched sequence motifs in differentially expressed genes between the two lines. In doing so, we observed a significant enrichment for motifs recognized by TFAP2A in cluster 5 which mainly contains mutant cells (FIG. 5H). Altogether, these data further suggested a role for TFAP2A in fibro-fatty signaling in epicardial cells.

TFAP2A Mediates Epicardial to Fibro-Fatty Transition Through EMT.

[0107] To investigate whether TFAP2A is sufficient to induce to this fibro-fatty gene signaling, we set out to use a mixture of targeting siRNAs (as described here above) to knock down TFAP2A in PKP2 c.2013delC hiPSC-epicardial cells. After 48 h of transfection, we could detect a significant reduction in TFAP2A expression (FIG. 6A) as well as in several fibroblast and fat markers suggesting a potential TFAP2A-mediated fibro-fatty signaling in epicardial cells (FIG. 6B). Epicardial cells reside in an epithelial state which upon activation can undergo epithelial-to-mesenchymal transition (EMT) and start cellular differentiation. Therefore, we postulated that desmosomal suppression might act as an EMT-driving trigger. We analyzed the expression ratio of E-cadherin versus N-cadherin (CDH1/CDH2) in PKP2 c.2013delC hiPSC-epicardial cells versus the reverted controls, which is a measure of the epithelial versus mesenchymal state of the cells. We observed a similar expression ratio among our culture timeline up to 1 month. However, by day 80 only mutant cells displayed a dramatic reduction in this ratio indicating an EMT switch (FIG. 6C). To investigate whether artificially-induced EMT would promote a similar fibro-fatty phenotype, we cultured control hiPSC-epicardial cells in the presence of TGFβ1 and bFGF, well known inducers of epicardial EMT. As expected, TGFβ1+bFGF-treated cells showed a significant induction of TFAP2A as well as fibroblast, smooth muscle and fat cell markers. However, addition of siRNAs targeting TFAP2A abolished this effect (FIG. 6D). These results suggest that loss of desmosomal proteins can affect the epicardial integrity leading to its activation, EMT and cellular differentiation in an TFAP2A-dependent mechanism.

Human Explanted ACM Hearts Express WT1 and TFAP2A

[0108] In line with our observations in the in vitro hiPSC-epicardial cell cultures, we performed immunohistological stainings on heart sections explanted from healthy and ACM patients. In the healthy heart, the epicardium resided as a thin membranous layer surrounding the myocardium with very little fat and fibrosis underneath. However, hearts of ACM patients displayed a thicker sub-epicardium covering massive fibro-fatty infiltrates. In addition, those epicardial/subepicardial cells expressed WT1, a phenomenon also observed in animal models after cardiac injury, further suggesting epicardial activation in diseased hearts. WT1 induction was associated with TFAP2A expression in the sub-epicardial areas and among fibro-fatty infiltrates further supporting our findings revealing TFAP2A as a key factor in epicardial EMT and fibro-fatty tissue transition (FIG. 7).

REFERENCES

[0109] Gemayel C, Pelliccia A, Thompson P D, Arrhythmogenic right ventricular cardiomyopathy. J Am Coll Cardiol. 2001 December; 38(7):1773-81 [0110] Saffitz J E. The pathobiology of arrhythmogenic cardiomyopathy. Annual Review of Pathology: Mechanisms of Disease. 2011 Feb. 28; 6:299-321. [0111] Gorman, L., Suter, D., Emerick, V., Schumperli, D., and Kole, R. (1998). Stable alteration of pre-mRNA splicing patterns by modified U7 small nuclear RNAs. Proceedings of the National Academy of Sciences of the United States of America 95, 4929-4934 [0112] Scott C C, Vossio S, Rougemont J, Gruenberg J. TFAP2 transcription factors are regulators of lipid droplet biogenesis. eLife. 2018 Sep. 26; 7:e36330. [0113] Bao X, Lian X, Hacker T A, Schmuck E G, Qian T, Bhute V J, Han T, Shi M, Drowley L, Plowright A, Wang Q D, Goumans M J, Palecek S P. Long-term self-renewing human epicardial cells generated from pluripotent stem cells under defined xeno-free conditions. Nat Biomed Eng. 2016; 1. pii: 0003. [0114] Corrado D, Basso C, Judge D P. Arrhythmogenic Cardiomyopathy. Circulation Research. 2017 Sep. 15; 121(7):784-802.