Two-drug cocktail reprograms heart cells into an embryonic-like state to repair damage

Adult mammalian hearts have almost no capacity to rebuild themselves after a heart attack. Once cardiomyocytes are lost to ischemia, the damaged muscle is gradually replaced by collagen-rich scar tissue, and the heart slides toward failure. A research team reporting in Cell Regeneration on 11 August 2026 now describes a simplified two-drug cocktail, called 2SM, that pushed adult heart cells back into the cell cycle and improved cardiac function after myocardial infarction in rats and mice. The combination consists of just two compounds drawn from a previously published five-drug mixture: phenylephrine hydrochloride (PE), an alpha-1 adrenergic receptor agonist, and harmine (HM), an inhibitor of the kinase DYRK1A.

The earlier five-compound cocktail, known as 5SM, had already been shown to promote cardiomyocyte proliferation and cardiac regeneration in neonatal and adult rats, but its complexity posed barriers to clinical translation and to mechanistic study. In the new work, the researchers systematically pared the mixture down. Excluding AZD3965, which had shown proliferative potential on its own, they tested the remaining four compounds in six pairwise combinations using a fluorescent cell-cycle indicator system in postnatal day 3 neonatal rat ventricular cardiomyocytes. Three combinations, including PE plus HM, most robustly increased the number of cells in S, G2 or M phase. Crucially, only PE plus HM promoted Aurora kinase B-positive midbody formation, a marker of late-stage mitosis, indicating that this pair uniquely supported completion of cytokinesis.

The team then validated the effect with a battery of proliferation markers. Two days of 2SM treatment increased the numbers of KI67-positive, phospho-histone H3-positive and BrdU-incorporating cardiomyocytes in culture. To confirm actual cell division rather than mere cell-cycle re-entry, the researchers used a mosaic analysis with double markers (MADM) mouse model, in which successful mitotic recombination produces single-colored cardiomyocytes. 2SM significantly increased the number of single-colored cardiomyocytes, whereas neither PE nor HM alone did so. The cocktail also increased the percentage of KI67-positive and phospho-histone H3-positive human induced pluripotent stem cell-derived cardiomyocytes, indicating activity in an in vitro human model system.

Importantly, the proliferative stimulus was not self-sustaining. When 2SM was withdrawn after 48 hours and replaced with vehicle, the numbers of KI67-positive, phospho-histone H3-positive and Aurora kinase B-positive cardiomyocytes fell markedly, while continuous treatment maintained elevated levels. The authors conclude that 2SM stimulates transient cell-cycle activity and does not induce permanent or autonomous proliferation, a property that could matter for safety if the approach is ever moved toward patients.

To test the cocktail in vivo, the researchers applied a GelMA hydrogel containing 2SM directly to the infarcted area of adult mouse hearts immediately after permanent ligation of the left anterior descending coronary artery, followed by daily systemic administration for seven days. Seven days after injury, 2SM-treated mice showed significantly more single-colored cardiomyocytes than vehicle-treated controls, indicating enhanced cytokinesis in the injured heart. In adult rats subjected to myocardial infarction, treatment with the hydrogel plus injections every other day for four weeks and then weekly until eight weeks increased the number of isolated cardiomyocytes, shifted the population toward mononucleated cells and away from polyploid cells, and raised the percentages of KI67-positive and phospho-histone H3-positive cardiomyocytes in the infarct border zone.

Functional outcomes followed. Serial echocardiography from 7 to 56 days after injury showed improved ejection fraction and fractional shortening in 2SM-treated rats compared with controls, and Masson’s trichrome staining at 56 days revealed significantly reduced infarct size. Because pathological hypertrophy can also transiently improve contractile function, the researchers measured cardiomyocyte size: neither cultured cells nor heart tissue showed significant changes in cell area after 2SM treatment, and the team argues that the functional recovery reflects regeneration rather than hypertrophic remodeling. Head-to-head comparison at seven days after injury showed that 2SM and the original 5SM cocktail improved ejection fraction and fractional shortening to comparable degrees, with no significant difference between the two treatments. The researchers also reported no detectable abnormalities in major organs, no changes in organ-to-body weight ratios, and comparable serum alanine aminotransferase and aspartate aminotransferase levels between treated and control animals at 56 days.

Transcriptomic and metabolomic profiling of neonatal rat cardiomyocytes revealed how the cocktail works. RNA sequencing identified 842 upregulated and 777 downregulated genes after 24 hours of 2SM treatment, with upregulated genes enriched for chromosome segregation, DNA replication and cell division, and downregulated genes enriched for lipid and fatty acid metabolism. Metabolomic profiling showed elevated levels of glycolytic and pentose phosphate pathway intermediates, including glucose-6-phosphate, fructose bisphosphate, lactate and ribose-5-phosphate, while several long-chain fatty acids accumulated, consistent with inhibited fatty acid catabolism. This pattern mirrors the metabolic shift that occurs naturally when the neonatal heart moves from a hypoxic, glycolytic fetal state to an oxygenated, oxidative-phosphorylation-dependent adult state, and which prior work has linked to proliferative capacity.

Deconvoluting the two compounds revealed a striking division of labor. PE alone induced a fetal-like and hypertrophy-associated transcriptional priming state, upregulating Nppa, Nppb and Myh7 and raising the Myh7/Myh6 ratio, while suppressing fatty acid oxidation genes. PE activated the mTORC1-S6K pathway, which drove upregulation of cyclin D1 and phosphorylated retinoblastoma, promoting entry into G1/S, but it also increased WEE1 and phosphorylated CDK1, imposing a G2/M checkpoint block. Overexpressing Rheb, a direct activator of mTORC1, raised S6K phosphorylation and upregulated the same cell-cycle regulators yet failed to increase the proportion of cell-cycle-positive cardiomyocytes, and unlike PE it did not induce the fetal genes Nppa and Nppb. The authors conclude that mTORC1 activation is necessary but insufficient, and that PE’s concurrent fetal gene program is what primes the cells for division.

Harmine, by contrast, acted on the other side of the checkpoint. It inhibited DYRK1A, reducing phosphorylation of LIN52, a core MuvB complex subunit whose dephosphorylated form favors assembly of the transcriptionally active MMB and FOXM1-MuvB complexes. HM and 2SM treatment enhanced phosphorylation and total levels of B-MYB and FOXM1, and genetic deletion of Dyrk1a in cardiomyocytes produced the same activation of the B-MYB/FOXM1 axis, providing evidence that the effect is not a pharmacological off-target artifact. The activated complexes upregulated G2/M genes including Mybl2, Foxm1, Cdk1, Ccnb1 and the CDC25 phosphatases, whose protein products dephosphorylate CDK1 and thereby relieve the PE-imposed arrest. Blocking CDC25 with the inhibitor M2N12 increased CDK1 phosphorylation and suppressed the 2SM-induced increases in BrdU-positive cells, phospho-histone H3-positive cells and Aurora kinase B-positive midbodies, with the strongest effect on G2/M markers. Timing proved essential: PE for 24 hours followed by HM for 24 hours significantly increased the proportion of cells in G2/M and matched the proliferative effect of simultaneous 2SM, whereas the reverse sequence did not.

The authors frame the result as a new model of cardiomyocyte proliferation in which cells first enter an embryonic, non-proliferative hypertrophic-like state characterized by fetal gene activation, a metabolic shift from fatty acid oxidation to glycolysis and enhanced protein biosynthesis, and only then complete mitosis and cytokinesis under the direction of a second signal. They also note limitations: the mechanistic experiments relied mainly on neonatal rat ventricular cardiomyocytes, which retain higher intrinsic proliferative capacity than adult cells, and the detailed mechanisms in adult and human cardiomyocytes and in the in vivo cardiac environment require further investigation. Future work using single-cell and spatial transcriptomics is planned to resolve how 2SM affects different cell types and region-specific proliferative responses. The study, published in Cell Regeneration, was led by researchers including Zihao Wang and corresponding author Jing-Wei Xiong, and establishes a simplified, mechanistically defined two-drug regimen that the authors describe as a promising therapeutic approach for ischemic heart disease, pending the further preclinical and translational work they themselves outline.

Subject of Research: A simplified two-drug cocktail that promotes cardiomyocyte proliferation and heart regeneration after myocardial infarction

Article Title: Small-molecule cocktail 2SM promotes heart regeneration via inducing embryonic-like cardiomyocyte state

Article References: Wang, Z., Zheng, L., Ma, X., Gao, P., Chen, Y., Xiao, C., Zhao, X., Xiao, Z., Zhu, X., Zhang, Z., & Xiong, J.-W. (2026). Small-molecule cocktail 2SM promotes heart regeneration via inducing embryonic-like cardiomyocyte state. Cell Regeneration, 15(1), Article 27. https://doi.org/10.1186/s13619-026-00301-0

Image Credits: AI Generated

DOI: 10.1186/s13619-026-00301-0

Keywords: heart regeneration, cardiomyocyte proliferation, myocardial infarction, small molecules, phenylephrine, harmine, cell cycle, mTORC1, DYRK1A, metabolic reprogramming, fetal gene program, preclinical study