Reactivation of developmental genes to promote neovascularisation and cardiac regeneration after myocardial infarction

Abstract

Myocardial infarction (MI) remains a leading cause of heart failure (HF), compounded by the limited regenerative capacity of the adult heart. Despite advances in MI treatments, HF remains common, highlighting the urgent need for novel strategies to prevent HF progression and promote cardiac regeneration. Effective cardiac repair requires rapid reconstruction of a functional vascular network, yet the mechanisms driving neovascularisation remain poorly understood. Increasing evidence suggests that genes deployed during embryonic development can be reactivated in the adult heart to support regeneration pathways. Owing to its exceptional regenerative ability, the zebrafish represents a powerful model to investigate myocardial-vascular regeneration pathways. I hypothesised that reactivation of foetal genes in adult coronary endothelial cells (ECs) promotes neovascularisation and cardiac repair post- MI.

Multimodal omics analyses identified Annexin A2 (ANXA2) and Poly(A) binding protein cytoplasmic 1 (PABPC1) as endothelial genes expressed during development and reactivated in the adult coronary vasculature post-MI. Protein level upregulation of both targets was confirmed in cardiac ECs from MI patients versus healthy controls (n = 7; %ANXA2⁺CD31⁺ ECs = 77.1± 12.1 versus. 56.9 ± 9.2, P = 0.002; %PABPC1⁺CD31⁺ ECs = 68.0 ± 15.6 versus. 37.9 ± 18.6, P = 0.013). The ANXA2 fibrinolytic co-regulator, Serpin Family E Member 1 (SERPINE1), was also significantly increased in MI (%SERPINE1⁺CD31⁺ ECs = 32.7 ± 11.7 versus. 16.1 ± 7.3, P = 0.001). These findings suggest that a balanced interplay between ANXA2 and SERPINE1 may influence coronary neovasculogenic responses post-MI, in addition to their known roles in fibrinolysis.

ANXA2 was prioritised as a lead candidate for functional analysis. CRISPR/Cas9- mediated knockdown of orthologues anxa2a/b was generated in Tg(fli1:eGFP)y1tg / Tg(myl7:DsRed2-NLS)f2) zebrafish to investigate the role of ANXA2 in cardiovascular development and regeneration. At 3 days post-fertilisation (dpf), crispants (CRISPR- Cas9 engineered animals) exhibited developmental abnormalities, including smaller eyes (n = 45 – 80; eye diameter in mm = 0.68 ± 0.16 versus. 0.91 ± 0.07, P < 0.0001), shorter body length (body length in mm = 9.75 ± 0.99 versus. 10.82 ± 0.41, P < 0.0001), and disrupted intersegmental vessels (n = 25-48; number of complete intersegmental vessels = 23.31 ± 3.05 versus. 27.83 ± 1.36, P < 0.0001). By 5 dpf, 45.3% of crispants displayed mild phenotypes, and 45.0% showed moderate or severe developmental defects, such as curled tails, pericardial oedema, failure to hatch, and full body malformations. I developed and optimised a cardiac laser injury and regeneration model in <5 dpf larvae. Laser injury significantly reduced heart rate, EC number, and ventricular volume at 2 hours post-injury (hpi) (n = 20-38; heart rate in bpm = 142.63 ± 67.44 versus. 200.74 ± 66.15, P = 0.0012; GFP+ EC counts = 62.00 ± 28.27 versus. 107.09 ± 29.46, P=0.0018; diastolic endocardial volume of ventricle (µm3) = 3.02e6 ± 2.37e6 versus. 7.29e6 ± 5.22e6, P = 0.0014), all of which had shown complete recovery by 48 hpi in wild type (WT) siblings. Compared with injured WT siblings, mild anxa2ab crispants showed partial regeneration at 48 hpi, characterised by reduced EC repopulation (n = 24; GFP+ EC counts = 61.70 ± 21.25 vs 89.00 ± 20.11, P < 0.0001), diminished cardiomyocytes numbers (mCherry+ cardiomyocyte counts = 56.61 ± 17.03 vs 83.25 ± 14.06, P < 0.0001), and smaller ventricular volume (diastolic endocardial volume of ventricle (µm3) = 6.19e6 ± 2.00e6 vs 1.03e7 ± 2.62e6; P < 0.0001). Moderate anxa2ab crispants exhibited severely impaired regeneration, with persistently low EC and cardiomyocyte counts, reduced ventricular volume, and distorted ventricular morphology (n = 21; GFP+ EC counts = 34.71 ± 19.21 vs 76.79 ± 13.01; mCherry+ cardiomyocyte count = 25.29 ± 17.01 vs 72.47 ± 13.34; diastolic endocardial volume of ventricle (µm3) = 4.14e6 ± 2.49e6 vs 8.88e6 ± 2.00e6; P < 0.0001).

Quantitative reverse transcription polymerase chain reaction (qRT-PCR) at 48 hpi revealed transcriptional alterations in targets associated with ANXA2 function in crispants, including S100 calcium binding protein A10ab (s100a10ab), vascular endothelial growth factor Ab (vegfab), C-X-C motif chemokine ligand 12b (cxcl12b), Notch receptor 1a/3 (notch1a/3), and pabpc1ab, but not serpine1. This suggests ANXA2 may act upstream of a network of endothelial and regenerative signalling genes, and that its loss compromises repair responses. Overall, this study identifies PABPC1 and ANXA2 as developmental genes reactivated upon cardiac injury, with ANXA2 deficiency leading to impaired cardiac regeneration.

The phenotypic defects observed in anxa2ab knockdown zebrafish highlight the essential role of ANXA2 in embryonic development, neovascularisation, and cardiac repair. ANXA2 and SERPINE1 act antagonistically to regulate plasmin generation and fibrin turnover; an imbalance between ANXA2 and SERPINE1 may contribute to the impaired neovascularisation, pathological cardiac remodelling and fibrosis that drives the progression of heart failure. Therefore, elucidating ANXA2/SERPINE1-mediated pathways may provide strategies to enhance neovascularisation and heart regeneration. Future work should focus on understanding the mechanisms by which ANXA2, PABPC1 or other related genes regulate endothelial responses to injury, to inform the development of novel regenerative therapies for cardiovascular disease and the prevention of heart failure, for which there is currently no cure.