Long-term culture of Engineered Human Myocardium recapitulates hallmarks of cardiac fibrosis and aging

Introduction: Engineered Human Myocardium (EHM) provides a valuable 3D model for investigation of cardiac pathophysiology and drug discovery. EHM resemble key features of the juvenile ventricular myocardium in standard four- to six-week cultures. To test the hypothesis that we can model human heart muscle “aging in the dish”, we investigated EHM over a period of 12 months using contractility assessments and transcriptome analyses. 

Methods and Results: EHM were generated from human iPSC-derived cardiomyocytes and stromal cells in a collagen hydrogel. 313 EHM were initially prepared and maintained for 1, 3, 6 and 12 months (n = 68, 68, 89, 88). Spontaneous tissue function was repeatedly measured using a video-optic recording system (myrImager) and demonstrated stable EHM contractions of all tissues at five weeks of culture, which remained intact until four months of age. After that, we observed loss of spontaneous contractility in 50% of tissues at 6 months and only 2 tissues showed minimal contractions after 12 months of culture. In order to understand underlying cellular and mechanistic processes, we performed endpoint measurements at predefined timepoints including bulk RNA and single-nucleus sequencing (snSeq). We observed molecular markers of cardiomyocyte maturation in 3 vs 1 months old EHM and a marked decline in relative cardiomyocyte proportion in 12 vs 6 months EHM. Stromal cells developed a “myofibroblast signature” with high abundance of transcripts encoding for ACTA2, POSTN and COL1A1 at 3 and 6 months suggesting a fibrotic phenotype. Corresponding to that, measurement of tissue tension as a marker of fibroblast activity showed a clear increase between 3 and 6 months (in mN/mm2: 2.5 ± 0.3 vs 11.4 ± 1.1; p<0.0001). At 12 months, transcriptome signals of autophagy (CTSD, RUBCNL), cellular senescence (CDKN1A, SERPINE1) and inflammation (LIF, NF-κB, IL6R/IL6ST) become apparent, similar to maladaptive processes observed in cardiac aging.

Conclusion: EHM recapitulate cardiac development until 3 months of culture with stable contractility between 1-3 months and hallmarks of fibrosis with related contractile dysfunction at 6 months of culture. Between 3 and 6 months, myofibroblast activation commences, leading to increased tissue tension with evidence for impaired protein homeostasis and inflammation accumulating at 12 months. Our data suggests that modelling of cardiac aging is possible in EHM. Further studies will be required to further define the aging process.