Psoralen enhances direct cardiac reprogramming via activating PPARα and driving mitochondrial fission and improves cardiac function after myocardial infarction

Abstract

Background

Direct cardiac reprogramming offers a promising strategy to regenerate cardiomyocytes for heart repair, but its efficiency and maturation remain limited. Psoralen, a natural compound isolated from Psoralea corylifolia L., has been widely studied for its bioactive properties. This study aimed to determine whether psoralen enhances chemical cardiac reprogramming and improves functional maturation, and to evaluate its therapeutic potential in myocardial infarction.

Methods

Mouse embryonic fibroblasts were treated with a chemical reprogramming cocktail containing RepSox and Forskolin, with or without psoralen. Gene expression was assessed by quantitative PCR and Western blotting. Sarcomere ultrastructure was examined by transmission electron microscopy. Mitochondrial content and morphology were analyzed using MitoTracker staining. Calcium handling was evaluated in Fluo-4 loaded induced cardiomyocytes. Metabolic profiles were measured with Seahorse assays. RNA sequencing and protein–protein interaction analysis were used to identify signaling pathways activated by psoralen. To investigate the functional role of mitochondrial fission in psoralen enhanced reprogramming, the DRP1 inhibitor Mdivi-1 was added to the reprogramming cocktail. An immortalized human skin fibroblast cell line was additionally used to evaluate cardiac marker expression and mitochondrial morphology following treatment with RepSox, Forskolin, and psoralen. For in vivo evaluation, a myocardial infarction model was generated in male C57BL/6J mice, which were subsequently treated for 7 days. Cardiac function was measured by echocardiography, and histological changes were assessed by HE and Masson’s trichrome staining.

Results

Psoralen (10 µM) significantly increased the induction efficiency of induced cardiomyocytes, leading to the rapid appearance of beating clusters(day 1 vs. day 6–8) and marked upregulation of cardiac structural genes. Psoralen-treated cells displayed enhanced sarcomere organization, improved calcium transients, and a metabolic shift toward oxidative phosphorylation characterized by higher mitochondrial respiration and reduced glycolysis. Transcriptomic profiling identified activation of the PPAR signaling pathway, with PPARα, RXRG, and UCP1 as central nodes. Psoralen also enhanced mitochondrial fission by upregulating fission-related genes and proteins. Mdivi-1 attenuated psoralen-induced cardiac gene expression and mitochondrial fragmentation. In immortalized human skin fibroblasts, psoralen combined with RepSox and Forskolin increased cardiac marker expression and promoted mitochondrial fragmentation. In vivo, psoralen combined with RepSox and Forskolin enhanced cardiac function and reduced fibrosis after myocardial infarction compared to RepSox and Forskolin alone.

Conclusion

Psoralen enhances chemical direct cardiac reprogramming by activating metabolic and structural maturation programs through mitochondrial fission and PPARα-dependent signaling. The combined treatment provides functional benefits in myocardial infarction, highlighting psoralen as a promising small molecule for cardiac regeneration.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s13287-026-05160-9.

Background

Heart failure remains a major global health burden, largely due to the irreversible loss of cardiomyocytes following myocardial injury [1]. Direct cardiac reprogramming, in which fibroblasts are converted into induced cardiomyocytes (iCMs), has emerged as a promising strategy for myocardial regeneration. Previous studies on direct cardiac reprogramming have primarily focused on transcription factor cocktails, such as Gata4, Mef2c, and Tbx5 (GMT) or their derivatives, which were able to induce partial cardiomyocyte-like phenotypes [2]. Compared with transcription factor based reprogramming, chemically induced reprogramming offers notable advantages, including the avoidance of exogenous genetic material and the potential for scalable and standardized production [3]. Current approaches primarily utilize cocktails of small molecules that modulate key signaling pathways, including transforming growth factor-β (TGF-β), Wnt, BMP, and fibroblast growth factor signaling, to generate iCMs [4, 5]. However, a major limitation of these protocols is that they often requirement for as many as 6–8 molecules (e.g., the CRFVPTZ cocktail comprising CHIR99021, RepSox, Forskolin, VPA, Parnate, TTNPB, and DZnep) and timelines extending to 24 days [5, 6]. This inefficiency presents a major barrier to clinical translation and underscores the need for more robust and simplified chemical formulations. To establish a simplified and tractable chemical platform, we employed a minimal two-molecule combination consisting of RepSox(R) and Forskolin (F). R, a TGF-β receptor inhibitor, has been widely used to suppress fibroblast identity and facilitate lineage conversion, while Forskolin promotes cardiac gene expression and early sarcomere assembly during reprogramming [5, 7]. This minimal RF cocktail provides an effective baseline condition for evaluating additional small molecules while reducing system complexity.

A critical barrier to achieving functional iCM maturation is that the reprogrammed cell fails to recapitulate the metabolic and bioenergetic state of the adult cell [8]. While mature cardiomyocytes rely predominantly on mitochondrial oxidative phosphorylation (OXPHOS) for energy, fibroblasts depend on glycolysis [9]. Successful cardiac reprogramming must therefore involve a bioenergetic switch from glycolysis to OXPHOS, a process intrinsically linked to extensive mitochondrial remodeling [10]. Mitochondria are dynamic organelles whose morphology and function are regulated by fission and fusion processes [11]. Pertinently, mitochondrial fission has been identified as an essential physiological process during perinatal heart development, facilitating the metabolic transition to fatty acid oxidation [12]. Emerging evidence suggests that modulating mitochondrial dynamics can directly influence cell fate; for instance, knockdown of the mitochondrial fission-promoting factor mitochondrial fission regulator1-like protein (Mtfr1l) in mouse cardiac fibroblasts was shown to facilitate reprogramming to iCMs [13]. Much work has investigated the genetic, epigenetic, and translational mechanisms of cardiac reprogramming [14], however, the role of specific small molecules in steering mitochondrial dynamics to overcome this metabolic barrier remains largely unknown.

Psoralen (Psr) is a major active constituent of Psoralea corylifolia L., a traditional Chinese medicine that has been registered as a health-care food by the Ministry of Public Health of China [15]. Psr has been reported to exhibit diverse pharmacological activities, including anti-inflammatory and anti-fibrotic effects, and its safety profile has been evaluated in both experimental and clinical settings [16, 17]. Adverse effects are mainly associated with high doses or light-dependent phototoxicity, whereas lower concentrations used in experimental and clinical studies are generally well tolerated [18]. In our previous work, we demonstrated that Psr enhances direct chemical reprogramming of mouse embryonic fibroblasts into osteoblast-like cells when combined with a Forskolin and Phenamil cocktail [19]. This observation prompted us to explore whether Psr could exert similar effects in other small molecule based reprogramming settings. Based on this rationale, we hypothesized that Psr could act as a synergistic factor in chemical cardiac reprogramming systems. In preliminary experiments, we observed that supplementation of the classical RF cocktail with Psr markedly enhanced the induction of iCMs, accelerating sarcomere assembly and promoting earlier appearance of beating cells. Transcriptomic profiling further suggested that Psr may modulate pathways related to mitochondrial function and metabolic activation, although the underlying mechanisms remain to be elucidated.

Based on these observations, the present study aims to systematically evaluate the role of Psr in chemical cardiac reprogramming. Specifically, we sought to: (1) optimize the RF + Psr combination and assess its ability to promote structural and functional maturation of iCMs, including analyses of sarcomere organization, mitochondrial morphology, calcium transients; (2) identify molecular mechanisms underlying Psr’ s actions, focusing on mitochondrial dynamics and related proteins and markers suggested by transcriptomic profiling; and (3) determine the therapeutic relevance of RF + Psr in vivo using a mouse myocardial infarction model, in which cardiac function, histological remodeling, were comprehensively evaluated. Understanding these processes will not only provide mechanistic insights into chemically induced cardiac reprogramming, but also offer a potential small molecule based strategy for heart regeneration.

Materials and methods

Reagents

The materials used included: DMEM/F12 (1:1) basal medium (1X) (Gibco, 6123073), DMEM basic medium (1X) (Gibco, 6125275), fetal bovine serum (FBS, Procell, 164210-50), MEM non-essential amino acid (Servicebio, G4219), Sodium L-Ascorbyl-2-Phosphate (OriLeaf, S67343), Insulin-Transferrin-Selenium (ITS) G supplement (Servicebio, G4028), Cardiac Troponin T Polyclonal antibody (Protientech, 15513-1-AP), Alpha Actin Polyclonal antibody (Protientech, 23660-1-AP), Beta-actin polyclonal antibody (Protientech, 20536-1-AP), PPARA Monoclonal antibody (Protientech, 66826-1-Ig), RXRG Polyclonal antibody (Protientech, 11129-1-AP), UCP1 Polyclonal antibody (Protientech, 23673-1-AP), DRP1 (C-terminal) Polyclonal antibody (Protientech, 12957-1-AP), FIS1 Polyclonal antibody (Protientech, 10956-1-AP), DRP1 (phospho S616) Monoclonal antibody (abcam, ab314755), Alexa Fluor 594-labeled goat anti-rabbit IgG (Servicebio, GB28301), Alexa Fluor 488-labeled goat anti-rabbit IgG (Servicebio, GB25303), Alexa Fluor 488-labeled goat anti-mouse IgG (Servicebio, GB25301), DAPI staining reagent (Servicebio, G1012), Fluo-4 AM (ThermoFisher, F14201), Mito-Tracker Red CMXRos (Beyotime, C1035), Hoechst 33342 (Solarbio, B8040), First Strand cDNA Synthesis Kit (RNase H minus, Beyotime, D7168M), RepSox (GLPBIO, GC16793), Forskolin (GLPBIO, GC11920), Psr (Beijing Bethealth People Biomedical Technology Co.,Ltd, 23010802), One-Step Western Kit HRP (CWBio, CW2029M), high-strength RIPA buffer (Solarbio, R0010-20 ml), BCA Protein Assay Kit (Solarbio, PC0020), SuperSignal™ West Pico PLUS Chemiluminescent Substrate (Thermo Fisher Scientific, 34577), Cell Counting Kit-8 (Beyotime, C0038), Mdivi-1 (MedChemExpress, HY-15886), and qPCR SYBR Green Master Mix (High Rox Plus, Yeasen, 11203ES08). Primers were synthesized by Sangon Biotech.

Extraction and primary culture of mouse embryonic fibroblasts (MEFs)

MEFs were isolated from embryonic day 13.5 (E13.5) C57BL/6J embryos obtained from 2 timed-pregnant dams. The procedure was performed as described before [20]. Briefly, pregnant dams (8–10 weeks old, 18–20 g) were euthanized under anesthesia and uteri were removed under sterile conditions. Individual embryos were dissected out, and the heads, limbs, and internal organs were removed. The remaining embryonic trunks were rinsed in ice-cold PBS and minced into fragments (< 1 mm³). Tissues were digested in 0.25% trypsin at 37 °C for 15 min. The resultant cell suspension was centrifuged and resuspended in DMEM/F12 medium supplemented with 10% FBS and 1% penicillin/streptomycin, then cultured in T25 flasks. Cells were expanded under standard culture conditions (37 °C, 5% CO₂). MEFs at passage 4 (P4) were used for all reprogramming experiments. For cell culture experiments, each batch of MEFs was derived from a single pregnant dam, representing one independent biological replicate. All experiments were performed using at least two independent biological replicates (from different dams), with at least three technical replicates per biological replicate.

Cell grouping and treatment

Passage-4 MEFs were seeded in 12-well plates at a density of 1 × 10⁵ cells per well. When cultures reached approximately 80% confluency, cells were subjected to cardiac induction using Cardiac Induction Medium (CM), consisting of DMEM supplemented with 10% FBS, 1% penicillin/streptomycin, 50 mg/L Sodium L-Ascorbyl-2-Phosphate, 1% ITS, and 1% MEM non-essential amino acids. All small molecules were dissolved in dimethyl sulfoxide (DMSO) as stock solutions. The final concentration of DMSO in the culture medium was maintained below 0.5% (v/v) in all treatment groups. Different small molecules, including Forskolin (F, 10 µM), RepSox (R, 10 µM), and varying concentrations of Psr (5–50 µM), were added to the CM as required for each experimental group. The culture medium was refreshed every 2 days throughout the induction process. For mitochondrial fission inhibition experiments, Mdivi-1 (10 µM) was added to the RF + Psr treatment group and maintained throughout the 4-day induction period. Medium containing Mdivi-1 was refreshed every 2 days together with the regular medium change.

Immortalized human dermal skin fibroblasts (HSF, QuiCell, QuiCell-H452) were cultured in DMEM supplemented with 10% fetal bovine serum under standard culture conditions (37 °C, 5% CO₂). For cardiac reprogramming, HSF were seeded in 12-well plates at a density of 2.5 × 10⁴ cells per well. When cultures reached approximately 80% confluency, cells were subjected to cardiac induction using CM with the same composition as described for MEFs. F (10 µM), R (10 µM), and Psr (10 µM) were added to the CM as required for each experimental group. The culture medium was refreshed every 2 days throughout the induction process.

Reverse transcription quantitative PCR

Total RNA was extracted using TRIzol reagent following the manufacturer’s instructions. A total of 500 ng RNA was reverse-transcribed into cDNA using the First Strand cDNA Synthesis Kit. RT-qPCR was performed using SYBR Green Master Mix on the CFX96 Real-Time PCR Detection System (Bio-Rad). Reactions were performed in triplicate under the following cycling parameters: initial denaturation at 95 °C for 30 s, then 40 cycles comprising denaturation at 95 °C for 10 s, annealing/extension at 60 °C for 30 s, and melting curve analysis. Gene expression levels of Myh6, Tnnt2, Ryr2, Gata4, β-MHC, Myl7, Nkx2.5, Opa1, Mfn1, Mfn2, Drp1, Fis1, PPARα, Ucp1 were quantified. Gapdh served as the internal reference. Primer sequences for mouse and human genes are provided in Supplementary Tables 1 and 2, respectively.

Immunofluorescence and mitochondrial staining

Cells were fixed with 4% paraformaldehyde for 10 min, washed with PBS, and permeabilized using 0.5% Triton X-100 for 10 min. Following blocking with 5% BSA for 1 h at room temperature, cells were incubated overnight at 4 °C with primary antibodies (Vimentin, 1:400; cTnT, 1:400; cTnI, 1:200; α-Actin, 1:400) diluted in 5% BSA. Subsequently, cells were incubated with Alexa Fluor 594- or 488-conjugated secondary antibodies (1:400) for 1 h at room temperature. Nuclei were counterstained with DAPI. Fluorescence images were captured using a Confocal laser scanning microscope (Olympus FV3000), and mean fluorescence intensity was quantified with ImageJ software.

To assess mitochondrial status, mitochondrial mass was analyzed using MitoTracker Red CMXRos. On day 4 of induction, cells were incubated with 100 nM MitoTracker Red and 1 µg/mL Hoechst for 20 min at room temperature, followed by imaging using confocal microscopy.

RNA-sequencing

The cDNA libraries were constructed for each pooled RNA sample using the NEBNext^® Ultra™ Directional RNA Library Prep Kit for BGI according to the manufacturer’s instructions. In brief, ribosomal RNA was depleted, and the remaining RNA was fragmented into 150–200 bp fragments using divalent cations at 94 ℃ for 8 min. First- and second-strand cDNA synthesis was performed, followed by end repair, A-tailing, and adapter ligation. Library fragments of the desired size were purified using AMPure XP beads (Beckman Coulter), enriched by PCR, and quantified using the Agilent 2200 TapeStation system. Indexed libraries were pooled at equimolar concentrations and subjected to 150 bp paired-end sequencing on the BGI T7 platform.

Raw reads were processed to remove low-quality sequences and adapters. Differential gene expression analysis was conducted using the EBSeq algorithm, and differentially expressed genes (DEGs) were identified based on fold-change, p-value, and false discovery rate (FDR) thresholds according to standard criteria [21]. Gene Ontology (GO) and KEGG pathway enrichment analyses were performed to identify biological processes and pathways associated with psoralen-enhanced reprogramming.

Western blot

Cells were lysed on ice with RIPA buffer supplemented with protease and phosphatase inhibitors, followed by centrifugation at 12,000 g for 10 min at 4 °C. Protein concentration was quantified using the BCA Protein Assay Kit. Equal amounts of protein (30 µg per lane) were separated on 10% SDS-polyacrylamide gels, and then transferred onto PVDF membranes. Membranes were incubated with primary antibodies (PPARα, 1:100; RXRG, 1:1000; UCP1, 1:1000; β-actin, 1:2000; DRP1 (phospho S616), 1:100; DRP1, 1:1000; FIS1, 1:1000) for 45 min at room temperature. After washing, membranes were incubated with HRP-conjugated secondary antibodies provided in the One-Step Western Kit. Protein bands were detected using SuperSignal™ West Pico PLUS Chemiluminescent Substrate and imaged using the ChemiDOCTM XRS+ Imaging System (Bio-Rad).Band intensities were quantified using Image Lab software.

Myocardial ischemia surgery and treatment

Male C57BL/6J mice (8–10 weeks old, 20–25 g) were purchased from SPF (Beijing) Biotechnology Co., Ltd. and housed under specific pathogen-free conditions with controlled temperature (22 ± 2 ℃), humidity (50–60%), and a 12 h light/dark cycle, with ad libitum access to food and water. This study was reported in line with the ARRIVE guidelines 2.0. The murine model of myocardial ischemia was performed as described previously [22]. Briefly, mice were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) and ventilated mechanically. After left thoracotomy, the left anterior descending coronary artery (LAD) was reversibly ligated with a slipknot (6−0 silk sutures). Sham-operated mice were subjected to same surgical procedure without LAD ligation.

A total of 36 mice were used in this study. Animals were first randomly assigned to undergo either sham surgery (Sham group, n = 6) or myocardial ischemia surgery (MI group, n = 30). Randomization was performed using a computer-generated random number sequence. Mice that successfully recovered from MI surgery were subsequently randomized into five treatment subgroups (n = 6 per subgroup): Model + 1% DMSO (vehicle control), Model + Psr, Model + RF, Model + RF + Psr, and Model + Metoprolol (positive control). Treatments were administered daily by intraperitoneal injection or intragastric gavage for 7 days following MI surgery. The sample size of six mice per group was determined based on common practice in similar murine myocardial ischemia studies, while adhering to the principle of reduction in animal use. The primary outcome measure was cardiac function assessed by echocardiography, including ejection fraction (EF) and fractional shortening (FS). Secondary outcome measures included histological assessment of myocardial fibrosis.

Animals were euthanized by cervical dislocation under deep anesthesia induced with 1% sodium pentobarbital (10 ml/kg). Animals that died during surgery or failed to recover from anesthesia were excluded from further analysis. All remaining animals were included in outcome assessment and data analysis. For each experimental group, the exact number of animals analyzed is reported in the corresponding figure legends.

Measurement of cytosolic Ca²⁺ transients

iCMs were loaded with 2 µM Fluo-4 AM (Sigma-Aldrich) for 30 min at 22 ± 2 °C in the dark. After washing twice with PBS, iCMs were placed in a cell chamber. Myocytes were stimulated to contract at a pacing frequency of 1 Hz with 4 ms of electrical stimulation. iCMs were exposed to 340–380 nm excitation wavelengths, and the emitted fluorescent signal was detected at 510 nm. Ca²⁺ transient parameters including amplitude, maximum ascending and descending velocity, and Ca²⁺ decline time constant were also recorded.

Echocardiography

Mouse echocardiography was performed as previously described [23]. Mice were anesthetized with 5% isoflurane for induction and maintained at 0.5% isoflurane during imaging. The anterior chest wall was depilated, and M-mode and two-dimensional echocardiographic recordings were acquired using a high-frequency Vevo 2100 ultrasound system (VisualSonics, Toronto, Canada) with concurrent electrocardiogram monitoring using subcutaneous needle electrodes. During image acquisition, the heart rate was maintained above 500 beats per minute. The heart rate, left ventricular internal dimension during diastole and systole (LVIDd/s), end-diastolic interventricular septal thickness in diastole and left ventricle posterior wall thickness in diastole were recorded. Left ventricular fractional shortening percentage, a measure of cardiac contractility, was calculated as the average of three (LVIDd−LVIDs)/LVIDd measurements per animal.

Histologic examination

Hearts were fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned at 4 μm thickness through the infarct region. Hematoxylin and eosin (H&E) staining was performed to assess myocardial morphology. Collagen deposition and fibrotic remodeling were evaluated using Masson’s trichrome staining. All stained sections were imaged using a stereomicroscope (Olympus SZ61, Tokyo, Japan). Quantification of collagen volume fraction was conducted using ImageJ software.

Statistical analysis

Data are presented as mean ± standard error of the mean (SEM). Statistical analyses were performed using GraphPad Prism 7.0. All datasets were assessed for normality distribution using Shapiro-Wilk tests (for n < 50), and homogeneity of variance was verified using Brown-Forsythe (for ANOVA). Data comparisons were performed using the statistical tests (Student’s t-test, one-way ANOVA, or two-way ANOVA with appropriate post-hoc tests). Sample sizes (n) for each statistical test are noted in the corresponding figure legends. Statistical significance was set at a threshold of P < 0.05. Outcome assessment and data analysis were performed by investigators blinded to group allocation.

Results

Psoralen synergizes with the RF cocktail to potentiate cardiac reprogramming

To determine the optimal concentration of Psr, we first treated MEFs with the RF cocktail supplemented with varying concentrations of Psr (5, 10, 25, and 50 µM). After 4 days of induction, RT-qPCR revealed that 10 µM Psr induced the highest expression of cardiomyocyte maturation markers Ryr2 and Tnnt2 (Fig. 1A), and was therefore used in subsequent experiments. Strikingly, spontaneously beating cells or clusters were observed as early as 24 h after induction with the optimized RF + Psr cocktail, markedly earlier than the previously reported timelines (days 6–8) [5], indicating a substantial acceleration of the cardiac reprogramming process. Representative videos recorded at day 4 further demonstrate robust spontaneous contractions in RF + Psr-treated cells, in contrast to the limited beating observed in the RF group (Video S1 and Video S2). Morphologically, control and CM-treated MEFs maintained a spindle-shaped fibroblast morphology with low refractility. RF and RF + Psr treatments induced a distinct morphological transition, characterized by enlarged, flattened morphology with strong refractility and reduced nuclear size (Fig. 1B). Quantification of beating cells showed that the RF + Psr group produced approximately four-fold more spontaneously beating cells (~ 1.2 × 10⁴ per 10⁵ cells) compared with the RF group (~ 2.5 × 10³ per 10⁵ cells) (Fig. 1C).

Fig. 1.

Psr synergizing with RepSox and Forskolin enhances direct cardiac reprogramming in mouse embryonic fibroblasts. A RT-qPCR analysis of Ryr2 and Tnnt2 expression in mouse embryonic fibroblasts (MEFs) cultured in control medium (control) cardiomyocyte differentiation medium (CM), or treated with RepSox + Forskolin (RF) or RF + Psoralen (RF + Psr) from day 1 to day 6 (n = 3 per group). B Representative bright-field images showing morphological changes in MEFs on day 4 under the indicated conditions. C Quantification of spontaneously beating cells from day 1 to day 6 in RF- and RF + Psr-treated groups (n = 3 per group). D–H mRNA expression levels of Gata4, β-MHC, Myh6, Myl7, and Nkx2.5 in MEFs under the indicated conditions (n = 3 per group). I–P Immunofluorescence staining images and quantification of the fibroblast marker Vimentin and cardiomyocyte markers cardiac troponin I (cTnI), α-actinin, and cardiac troponin T (cTnT) in MEFs cultured in CM or treated with RF or RF + Psr. Scale bars, 50 μm (n = 3 per group). *P < 0.05, **P < 0.01, ***P < 0.001

RT-qPCR further confirmed the enhanced induction of cardiac lineage genes (Fig. 1D–H). CM treatment modestly upregulated Gata4 expression (P < 0.001 vs. Control), whereas RF significantly enhanced all tested genes (Gata4, β-MHC, Myh6, Myl7, Nkx2.5) compared with CM. Importantly, RF + Psr treatment further elevated the expression of all five markers, with significant increases relative to RF alone (P < 0.01 for Myl7; P < 0.001 for all others), indicating a potentiated transcriptional activation of the cardiac program. Consistent with this, immunofluorescence analyses showed loss of fibroblast identity and acquisition of a cardiomyocyte phenotype (Fig. 1I-P). The fibroblast marker Vimentin was highly expressed in cells cultured in CM, reduced in the RF group, and further diminished to ~ 25.48 ± 0.98% of CM levels in the RF + Psr group (Fig. 1I, J). Conversely, RF treatment increased the proportion of cTnI-, α-actinin-, and cTnT-positive cells relative to CM, while RF + Psr induced the highest proportion of all three markers (cTnI: ~31.39 ± 0.84%; α-actinin: ~97.38 ± 1.13%; cTnT: ~70.28 ± 2.3%), significantly exceeding levels in the RF group (Fig. 1K–M). High-magnification immunofluorescence images revealed well-organized sarcomeric structures in a distinct population of RF + Psr-treated cells, which were rarely observed in RF-treated cells alone (Fig. 1N–P).

Together, these results demonstrate that 10 µM Psr markedly enhances the efficiency and speed of cardiac reprogramming by coordinately activating cardiac gene expression and facilitating the assembly of mature contractile units as early as day 4.

Psr enhances ultrastructural and functional maturation of iCMs

Given the marked increase in reprogramming efficiency upon Psr supplementation, we next assessed whether Psr also promotes the structural and functional maturation of iCMs. Transmission electron microscopy (TEM) revealed pronounced ultrastructural differences among groups (Fig. 2A). Representative images revealed that cells in the RF + Psr group acquired well-organized sarcomeric structures, characterized by aligned myofibrils and distinct Z-discs, which were absent in the CM and RF groups. Furthermore, mitochondria in RF + Psr treated cells were more abundant and frequently exhibited dumbbell-shaped morphology, indicative of active mitochondrial fission. Compared with the sparse, elongated mitochondria in RF treated cells, this shift toward a fission-active mitochondrial state suggests that Psr may enhance mitochondrial dynamics during the direct cardiac reprogramming.

Fig. 2.

Psr (10 µM) improves structural maturation, calcium handling, and metabolic function of iCMs. A Representative TEM images of MEFs cultured in CM, or iCMs generated with RF or RF + Psr. Yellow arrows indicate myofibrils and Z-discs; red arrows indicate dumbbell-shaped mitochondria. B Representative calcium transient recordings from Fluo-4-loaded cells in the indicated groups. C–D Quantitation of measured Ratio [(F1-F0)/F0] and decay time constant in iCMs from the RF (n = 161) and RF + Psr (n = 148) groups. E Representative oxygen consumption rate profile analyzed from mito-stress Seahorse assay in the indicated groups (n = 5 per group). F–I Quantification of basal respiration, maximal respiration, spare respiratory capacity in the indicated groups (n = 5 per group) and △OCR in RF and RF + Psr treated group. J, Representative extracellular acidification rate profile analyzed from glucose-stress Seahorse assay in the indicated groups (n = 5 per group). K–M Quantification of glycolysis, glycolytic capacity, and glycolytic reserve in MEFs, MEFs cultured in CM, RF and RF + Psr group (n = 5 per group). *P < 0.05, **P < 0.01, ***P < 0.001. 2-DG indicates 2-deoxy-glucose; ECAR: extracellular acidification rate; FCCP: fluoro-carbonyl cyanide phenylhydrazone; Eto: Etomoxir; OCR: oxygen consumption rate; Z: Z-discs; MF: myofibrils

Consistent with these structural changes, functional maturity was enhanced in RF + Psr iCMs. Spontaneously beating iCMs exhibited robust calcium transients under RF and RF + Psr conditions (Fig. 2B). High-speed video acquisition of Fluo-4-stained cells revealed that RF + Psr significantly increased Ca²⁺ transient amplitude (Fig. 2C) and shortened the decay time constant (τ) (Fig. 2D), reflecting improved Ca²⁺ cycling kinetics. These findings demonstrate that Psr enhances calcium cycling during cardiac direct reprogramming process.

To further characterize cellular metabolism associated with mitochondrial remodeling, we performed Seahorse flux analyses (Fig. 2E). RF + Psr iCMs exhibited significantly elevated basal respiration, maximal respiration, spare respiratory capacity, and overall oxygen consumption rate compared with RF treated cells (Fig. 2F, I), consistent with increased mitochondrial abundance and enhanced oxidative function. Moreover, we measured extracellular acidification rate and observed lower glycolysis, glycolytic capacity, and glycolytic reserve in iCMs after RF + Psr treatment (Fig. 2K-M). Reduced glycolysis and increased oxygen consumption in RF + Psr iCMs indicated that Psr promotes the metabolic switch into adult cardiomyocyte-like mitochondrial respiration.

Collectively, these results demonstrate that the supplementation of Psr significantly enhances the maturation of iCMs, both structurally and functionally, evidenced by the acquisition of mature ultrastructural features, improved functional calcium handling kinetics, and a metabolic shift towards oxidative phosphorylation.

Notably, the increased occurrence of dumbbell-shaped mitochondria suggests that Psr may contribute to cardiac reprogramming and maturation in part through effects on mitochondrial fission.

PPARα signaling serves as a key mediator of Psr action

To elucidate the molecular mechanisms underlying Psr mediated enhancement of cardiac reprogramming, RNA-seq was performed. A total of 251 genes were significantly upregulated and 72 genes downregulated upon Psr treatment (adjusted P < 0.01, fold change > 2). Principal component analysis (PCA) confirmed that the transcriptional profiles of cells treated with RF + Psr formed a distinct cluster separate from those treated with CM or RF alone (Fig. 3A). Notably, many genes with relatively low expression in the RF group were markedly activated in the RF + Psr group (Fig. 3B). Gene Ontology analysis revealed that upregulated genes were predominantly enriched in pathways related to fatty acid metabolism, oxidative phosphorylation, and cardiac structural components such as myosin complexes and Z-discs (Fig. 3C). Gene Set Enrichment Analysis (GSEA) further confirmed significant enrichment of PPAR signaling, tricarboxylic acid (TCA) cycle, fatty acid metabolism, and pyruvate metabolism pathways (Fig. 3D).

Fig. 3.

Psr (10µM) activates the PPARα/RXRG/UCP1 signaling and core metabolic pathways to promote chemical direct cardiac reprogramming. A Principal components analysis (PCA) of the RNA-Seq data (n = 3 per group) from the indicated groups. B Volcano plot displaying differentially expressed genes (DEGs) between RF + Psr and RF groups (adjusted P < 0.01, fold change > 2). C Heatmap of gene expression for upregulated genes in RF + Psr versus RF. The right-side annotations show representative significantly enriched GO terms for adjacent genes. D GSEA of genes upregulated in RF + Psr versus RF. GSEA enrichment plots demonstrate significant activation (NES > 0, FDR < 0.05) of gene sets involved in cellular energy derivation (TCA cycle, pyruvate metabolism) and fatty acid processing (PPAR signaling, fatty acid metabolism). E PPI network identifies key hub genes. Network of upregulated genes shows Pparα, Rxrg and Ucp1 as central hubs, based on their high connectivity, as represented by node size and position. F Bar plot of pathway enrichment analysis for genes upregulated in RF + Psr versus RF. The bar length represents the degree of enrichment [-log10(P-value)]. Pathways are ranked by their statistical significance (FDR < 0.05). G RT-qPCR analysis of Pparα, Rxrg, and Ucp1 mRNA expression across Control, CM, Psr, RF, and RF + Psr groups (n = 3 per group). H Western blot analysis of PPARα, RXRG, and UCP1 protein expression under the indicated treatments (n = 3 per group). β-actin serves as a loading control. *P < 0.05, **P < 0.01, ***P < 0.001

To identify key regulators within this transcriptional network, we constructed a protein-protein interaction (PPI) network using the Psr upregulated genes. The network hubs analysis revealed that PPARα, RXRG, and UCP1 resided at the core of the interactome (Fig. 3E). Concurrently, pathway enrichment of the PPI network identified the PPAR signaling pathway as the most significantly enriched (Fig. 3F), supporting the hypothesis that PPAR activation is a major driver of the Psr response. To experimentally validate these predictions, we quantified expression levels of PPARα, RXRG, and UCP1. Both RT-qPCR (Fig. 3G) and Western blot analysis (Fig. 3H) confirmed that RF + Psr significantly increased the mRNA and protein levels of all three molecules compared with RF alone, indicating activation of the PPARα/RXRG/UCP1 signaling axis.

These results collectively indicate that Psr supplementation enhances direct cardiac reprogramming efficiency in part via activation of the PPARα/RXRG/UCP1 axis, which may contribute to mitochondrial and structural remodeling in iCMs.

Psr promotes cardiac reprogramming by inducing mitochondrial fission

Mitochondria function is critical for cardiomyocyte maturation and provides the major energetic source for contractile activity. Given that transcriptomic analysis revealed enrichment of mitochondrial pathways and activation of the PPARα/RXRG/UCP1 axis in RF + Psr-treated cells (Fig. 3), we next examined whether Psr modulates mitochondrial remodeling during cardiac reprogramming. TEM analysis revealed a high frequency of dumbbell-shaped mitochondria, a morphological feature indicative of active mitochondrial fission, in RF + Psr iCMs, contrasting with the elongated and sparse mitochondrial network in RF iCMs (Fig. 2A). To quantitatively assess these morphological changes, MitoTracker staining was performed and showed increased mitochondrial area and fluorescence intensity in RF + Psr iCMs compared with RF alone (Fig. 4A, B). Morphometric analysis further demonstrated an increased proportion of fragmented and donut-shaped mitochondria, accompanied by a marked reduction in elongated mitochondria (> 5 μm). RF + Psr-treated MEF also exhibited increased mitochondrial circularity and decreased mean branch length, consistent with enhanced mitochondrial fragmentation (Fig. 4C). RT-qPCR analysis showed decreased expression of the fusion-associated genes Opa1, Mfn1, and Mfn2, together with increased expression of the fission-related genes Drp1 and Fis1 following RF + Psr treatment (Fig. 4D, E). Western blot analysis further confirmed increased expression of DRP1 and FIS1 in the RF + Psr group (Fig. 4F and I).

Fig. 4.

Psr (10µM) promotes mitochondrial fission to enhance chemical cardiac reprogramming. A, B Representative images and quantification of mitochondria content labeled with MitoTracker in Control, CM, Psr, RF and RF + Psr group (n = 30 per group). C Quantitative analysis of mitochondrial morphology in Control, CM, RF and RF + Psr group (n = 30 per group). D, E RT-qPCR analysis showing mRNA expression of mitochondrial dynamics regulators Opa1, Mfn1, Mfn2 (fusion) and Drp1, Fis1 (fission) in Control, CM, Psr, RF and RF + Psr group (n = 3 per group). F–I Western blotting of mitochondrial fission markers p-DRP1, DRP1 and FIS1 in Control, CM, Psr, RF and RF + Psr group (n = 3 per group). β-Actin was used as the loading control for total protein (DRP1 and FIS1). Total DRP1 protein levels served as the internal control for phospho-DRP1 (p-DRP1). *P < 0.05, **P < 0.01, ***P < 0.001

To investigate the functional contribution of mitochondrial fission to RF + Psr-enhanced cardiac reprogramming, MEFs were treated with RF + Psr in the presence of the DRP1 inhibitor Mdivi-1 (10 µM) for 4 days. CCK-8 assays showed that Mdivi-1 treatment did not significantly affect MEFs viability (Supplementary Figure S1). RT-qPCR analysis demonstrated that the RF + Psr+Mdivi-1 group exhibited significantly reduced expression of Drp1 and Fis1 compared with RF + Psr group (Fig. 5A, B). Consistently, Western blot analysis showed decreased levels of total DRP1 and phosphorylated DRP1 in the RF + Psr+Mdivi-1 group (Fig. 5C and E).

Fig. 5.

Mdivi-1 (10 µM) attenuates RF + Psr-induced cardiac reprogramming in MEFs. A, B RT-qPCR analysis of Drp1 and Fis1 expression in MEFs treated with RF, RF + Psr or RF + Psr + Mdivi-1 (10µM) for 4 days (n = 3 per group). C–E Western blotting of mitochondrial fission markers DRP1 and p-DRP1 in RF, RF + Psr and RF + Psr + Mdivi-1 (10µM) group (n = 3 per group). β-Actin was used as the loading control for total DRP1 protein levels. Total DRP1 protein levels served as the internal control for p-DRP1. F Representative images of mitochondria content labeled with MitoTracker under the indicated conditions. G–I Quantitative analysis of mitochondrial morphology under the indicated conditions (n = 30 per group). J Quantification of mitochondria content labeled with MitoTracker under the indicated conditions (n = 30 per group). K–N mRNA expression levels of Gata4, Myh6, Ryr2 and Tnnt2 in MEFs under the indicated conditions (n = 3 per group). O–T Immunofluorescence staining images and quantification of cardiomyocyte markers cTnI, α-actinin, and cTnT in MEFs under the indicated conditions (n = 3 per group). Scale bars, 50 μm (n = 3 per group). ***P < 0.001

MitoTracker staining revealed that RF + Psr-induced mitochondrial fragmentation was markedly attenuated by Mdivi-1 (Fig. 5F). Morphometric analysis showed that the RF + Psr+Mdivi-1 group exhibited an increased proportion of elongated mitochondria (> 5 μm), decreased mitochondrial circularity, and increased mean branch length compared with RF + Psr alone (Fig. 5G and I), indicating a shift toward a more fused mitochondrial network state. In addition, MitoTracker fluorescence intensity was reduced in the RF + Psr+Mdivi-1 group compared with RF + Psr treatment (Fig. 5J), suggesting altered mitochondrial content or functional status. To determine whether inhibition of mitochondrial fission influences cardiac reprogramming, we next evaluated cardiac lineage marker expression. RT-qPCR analysis showed that the RF + Psr+Mdivi-1 group exhibited significantly lower mRNA levels of Gata4, Myh6, Ryr2, and Tnnt2 compared with the RF + Psr group (Fig. 5K and N). Immunofluorescence analysis further demonstrated reduced proportions of cTnI⁺, α-actinin⁺, and cTnT⁺ cells (cTnI: ~12.45 ± 0.40%; α-actinin: ~36.24 ± 0.39%; cTnT: ~34.54 ± 0.48%) in the RF + Psr+Mdivi-1 group relative to RF + Psr alone (Fig. 5O and T). Collectively, these data support a functional contribution of mitochondrial fission to the enhanced cardiac reprogramming induced by RF + Psr.

RF + Psr promotes cardiomyocyte-like gene expression and mitochondrial remodeling in human fibroblasts

To validate the applicability of the RF + Psr cocktail in human cells, we used an immortalized human dermal skin fibroblast (HSF) cultured in CM, as described above. RT-qPCR analysis revealed low baseline expression of cardiac marker genes under CM conditions. Treatment with RF or RF + Psr increased the expression of Gata4, Myh6, Nkx2.5, Ryr2, and Tnnt2 compared with CM conditions, with significantly higher expression observed in the RF + Psr group relative to RF treatment (Fig. 6A and E). Immunofluorescence analysis further confirmed these findings (Fig. 6F and K). Specifically, RF + Psr treatment significantly increased the proportion of cells positive for cardiac markers cTnI, α-actinin, and cTnT compared with RF group, indicating enhanced induction of cardiomyocyte-associated protein expression.

Fig. 6.

RF + Psr enhances direct cardiac reprogramming in HSF. A–E RT-qPCR analysis of Gata4, Myh6, Nkx2.5, Ryr2 and Tnnt2 expression in HSF cultured in control medium (control), cardiomyocyte differentiation medium (CM), or treated with RF or RF + Psr for 4 days (n = 3 per group). F–K Immunofluorescence staining images and quantification of the cardiomyocyte markers cTnI, α-actinin, and cTnT in HSF cultured in CM or treated with RF or RF + Psr. Scale bars, 50 μm (n = 3 per group). L Representative images of mitochondria content labeled with MitoTracker under the indicated conditions. M–O Quantitative analysis of mitochondrial morphology under the indicated conditions (n = 30 per group). P Quantification of mitochondria content labeled with MitoTracker under the indicated conditions (n = 30 per group). ***P < 0.001

We next assessed mitochondrial morphology in HSFs using MitoTracker staining (Fig. 6L). Similar to observations in MEFs, RF + Psr treatment induced a shift toward a fission-dominant mitochondrial architecture. Morphometric analysis revealed an increased proportion of fragmented and donut-shaped mitochondria, along with a reduction in elongated mitochondria (> 5 μm). In addition, RF + Psr-treated HSFs exhibited increased mitochondrial circularity and decreased mean branch length compared with the RF alone group (Fig. 6M and O). MitoTracker fluorescence intensity showed a modest increase in the RF group relative to CM alone, whereas RF + Psr treatment further significantly increased MitoTracker intensity compared with both CM and RF groups (Fig. 6P). Collectively, these findings extend our observations to a human fibroblast and support the applicability of RF + Psr in promoting cardiomyocyte-like gene expression accompanied by mitochondrial remodeling.

RF + Psr synergistically attenuates MI-induced cardiac dysfunction and fibrosis in vivo

To investigate whether the synergistic effects of RF and Psr observed in vitro translate into therapeutic benefits in vivo, we established a mouse myocardial infarction (MI) model and administered treatment for 7 days. Echocardiographic assessment at day 7 post-MI revealed severe impairment of cardiac function in the Model group, as evidenced by significantly reduced ejection fraction (EF%: Model 38.47 ± 2.78 vs. Sham 71.00 ± 1.78, P < 0.001), fractional shortening (FS%: Model 16.55 ± 1.86 vs. Sham 33.52 ± 1.99, P < 0.001), left ventricular posterior wall systolic thickness (LVPWs: Model 0.94 ± 0.11 mm vs. Sham 1.81 ± 0.02 mm, P < 0.001), and stroke volume (SV: Model 12.81 ± 2.38 µL vs. Sham 36.15 ± 0.47 µL, P < 0.001) compared to Sham-operated animals (Fig. 7A-E). Among all the therapeutic groups, RF + Psr exhibited the most substantial functional improvement, significantly surpassing both RF and Psr (Fig. 7B-E). RF + Psr nearly doubled EF and SV relative to the untreated MI group (P < 0.001), demonstrating clear therapeutic synergy. Notably, the functional improvement achieved with RF + Psr was comparable to that observed with metoprolol, a standard β‑blocker used in the clinical management of heart failure (Fig. 7B–E). In contrast, Psr alone showed a modest improvement in FS (22.97 ± 2.47 vs. Model 16.55 ± 1.86, P < 0.01), but did not significantly enhance EF, LVPWs, or SV, emphasizing the importance of combining Psr with RF to achieve a comprehensive therapeutic benefit.

Fig. 7.

A 7-day treatment with RF + Psr improves cardiac function and ameliorates myocardial remodeling in mice with myocardial infarction. A Representative M-mode echocardiographic images of mice from different groups. B–E Quantitative analysis of cardiac function parameters: ejection fraction (EF, %), fractional shortening (FS, %), left ventricular posterior wall thickness at end-systole (LVPWs, mm), and stroke volume (SV, µL) (n = 6 per group). F Quantification of myocardial fibrosis. Collagen volume fraction was assessed by Masson’s trichrome staining and quantified using ImageJ (n = 3 per group). G Representative histological images of heart sections. HE staining shows myocardial cell morphology and structure. Masson’s trichrome staining visualizes collagen deposition (blue) and myocardial fibers (red). Sections are from the experimental groups of Sham, Model (Myocardial infarction), Myocardial infarction + Metoprolol (10 mg/kg, p.o.), Myocardial infarction + Psr (10 mg/kg, i.p.), Myocardial infarction + RepSox (10 mg/kg, i.p.)+Forskolin (10 mg/kg, i.p.), Myocardial infarction+RepSox (10 mg/kg, i.p.)+Forskolin (10 mg/kg, i.p.) + Psr (10 mg/kg, i.p.). *P < 0.05, **P < 0.01, ***P < 0.001

Histological analysis further supported the functional improvements (Fig. 7F-G), revealing that Metoprolol, RF, and RF + Psr treatments all significantly reduced cardiac injury. This was evidenced by a decrease in infarct size and myocardial fibrosis, as shown by H&E and Masson’s trichrome staining (Fig. 7F). Among these, RF + Psr treatment provided the most pronounced attenuation of fibrosis and infarct area, consistent with the functional recovery observed in the echocardiographic assessment. Masson’s trichrome staining revealed extensive myocardial fibrosis in the Model group, with a collagen volume fraction of 20.95 ± 0.4%. RF treatment significantly reduced fibrosis (~ 9.72 ± 0.25%, P < 0.001), while the combination of RF + Psr produced an even greater reduction (6.43 ± 0.4%, P < 0.001 vs. Model; P < 0.001 vs. RF). Psr alone showed minimal effect on fibrosis reduction (19.22 ± 0.46%, P < 0.001 vs. Model). These findings are consistent with the echocardiographic assessment and suggest that Psr synergistically enhances the anti-fibrotic activity of the RF cocktail.

Together, these findings demonstrate that RF + Psr therapy not only enhances the efficiency of cardiac reprogramming in vitro, but also offers significant protection against MI-induced cardiac dysfunction and remodeling in vivo. The synergistic effects of Psr with RF underscore its potential as a promising therapeutic approach for heart regeneration.

Discussion and conclusion

In this study, we identify the natural compound Psr as a potent synergistic molecule that markedly accelerates and enhances the efficiency of direct cardiac reprogramming. The RF + Psr cocktail generated spontaneously beating iCMs within 24 h, a speed substantially faster than previously reported chemical or transcription factor based protocols, which typically require 6–8 days or longer to induce visible contractile activity [5]. Moreover, RF + Psr generates functionally mature iCMs within only four days, compared with conventional approaches requiring up to 24 days [6]. RF + Psr iCMs displayed robust upregulation of cardiomyocyte structural genes and more defined sarcomeric organization than RF alone, indicating that Psr acts early during lineage conversion to accelerate progression toward a contractile phenotype.

Mechanistically, our transcriptomic, metabolic, and functional analyses converge on a regulatory axis centered on PPARα, RXRG, and UCP1, suggesting that psoralen enhances iCM induction by promoting oxidative metabolic maturation. PPARα is a canonical regulator of FAO and mitochondrial biogenesis in cardiomyocytes [24, 25], and our findings are consistent with reports showing that enhancing FAO through PPAR pathways improves both efficiency and functional maturation of induced cardiomyocytes [26]. RXRG, an obligate heterodimerization partner of PPAR family members, acts as a transcriptional regulator of fatty acid oxidation genes, promoting mitochondrial biogenesis and enhancing oxidative ATP production [27]. The strong induction of UCP1, a key mediator of mitochondrial uncoupling [28], provides a plausible mechanistic link between Psr induced PPARα activation and the observed mitochondrial remodeling. Together, the coordinated upregulation of PPARα, RXRG, and UCP1 suggests that Psr drives metabolic specialization toward an oxidative, FAO-dependent state that is highly permissive for cardiomyocyte lineage acquisition.

Metabolic remodeling is a major barrier in direct cardiac reprogramming, as fibroblasts rely primarily on glycolysis while mature cardiomyocytes depend on FAO-driven oxidative phosphorylation [29]. Consistently, the transcriptomic enrichment of FAO, TCA cycle, and OXPHOS pathways, combined with Seahorse flux analyses demonstrating increased oxygen consumption and diminished glycolytic reliance, support the conclusion that Psr establishes an oxidative metabolic environment conducive to iCM maturation. These metabolic features closely resemble those of adult cardiomyocytes and likely contribute to the rapid acquisition of contractile function.

In addition to metabolic remodeling, our data highlight a prominent role for mitochondrial fission in the Psr-enhanced cardiac reprogramming process. RF + Psr iCMs exhibited a high abundance of dumbbell-shaped mitochondria, increased DRP1 and FIS1 expression, and enhanced mitochondrial fragmentation. Mitochondrial dynamics are critical for cell fate determination [30]. Although excessive fission can be detrimental, controlled fission is required for developmental cardiogenesis and supports the removal of immature mitochondria, redistribution of mitochondrial mass to nascent sarcomeres, and transitions to oxidative metabolism [31]. In our accelerated direct reprogramming system, functional features such as sarcomeric assembly, calcium transients, and spontaneous contractions appear within four days, representing an accelerated early phase of fibroblast-to-cardiomyocyte conversion. The increased mitochondrial fission observed at this stage likely reflects an adaptive response to reprogramming stimuli that facilitates mitochondrial turnover, metabolic remodeling, and redistribution of mitochondria to nascent sarcomeres, thereby supporting the rapid acquisition of contractile function. While terminal maturation of cardiomyocytes is generally associated with increased mitochondrial fusion [32], the early fission observed here occurs in the context of the RF + Psr cocktail, which ultimately promotes enhanced functional properties of iCMs.

The mechanisms linking PPARα activation to DRP1-mediated mitochondrial fission remain to be fully defined. Accumulating evidence suggests that PPARα plays an important role in mitochondrial remodeling during cardiomyocyte maturation. For example, PPARα activation, in combination with HIF-1α inhibition, has been reported to increase mitochondrial content, improve mitochondrial distribution, and enhance mitochondrial respiratory function in human iPSC-derived cardiomyocytes [33]. PPARα-driven metabolic remodeling can also elevate mitochondrial reactive oxygen species production, which facilitates DRP1 recruitment to the outer mitochondrial membrane and promotes mitochondrial fission [34]. In addition, PPAR-associated signaling pathways can modulate DRP1 phosphorylation at regulatory residues such as Ser616 via downstream kinases including AMPK and PKA, thereby directly influencing its GTPase activity and fission capacity [35]. Consistent with these observations, UCP1 was markedly induced by RF + Psr in our system, potentially creating a bioenergetic environment that favors the selective removal of immature or dysfunctional mitochondria during early lineage conversion [36]. Together, these observations support a model in which PPARα-driven FAO activation increases oxidative demand, DRP1-mediated fission enhances mitochondrial turnover and redistribution, and UCP1-mediated mild uncoupling maintains a permissive bioenergetic state. These processes coordinate metabolic remodeling with mitochondrial dynamics, creating a mitochondrial network optimized for oxidative metabolism and contractile maturation, thereby contributing to the enhancement of iCM induction by Psr. Although direct causal interactions among PPARα activation, UCP1 induction, and DRP1-mediated fission remain to be experimentally established, their temporal and transcriptional convergence provides strong support for this integrative model.

The in vivo relevance of these findings was supported by results from a mouse myocardial infarction model, in which short-term administration of RF+psoralen improved cardiac function, enhanced stroke volume, and attenuated post-infarction remodeling compared with RF alone. Although the in vivo assessment was limited to echocardiographic and histological outcomes, these results suggest that the metabolic and structural improvements seen in vitro translate into functional benefits in injured myocardium. These observations highlight the potential of psoralen as an adjunct to chemical or hybrid reprogramming strategies aimed at cardiac repair.

The rapid in vivo functional improvement observed within seven days is likely attributable to a combination of direct reprogramming and indirect cardioprotective effects. In the mouse myocardial infarction model, significant improvements in cardiac function were observed after seven days of intraperitoneal administration, whereas functional recovery in comparable models typically requires several weeks. This raises the possibility that, in addition to promoting direct fibroblast-to-cardiomyocyte reprogramming, the RF + Psr combination may exert early cardioprotective effects. Psr has been reported to possess anti-inflammatory and anti-fibrotic properties [37], while R and F have also been implicated in modulating inflammation and reducing fibrosis in injured myocardium [38]. RF alone produced moderate functional benefits, whereas Psr alone had minimal effects, indicating that Psr acts synergistically with RF to enhance both reprogramming efficiency and cardioprotective outcomes. Nevertheless, without lineage-tracing evidence, the relative contribution of direct reprogramming versus indirect effects remains speculative.

Nevertheless, several limitations should be considered. First, the in vivo analyses demonstrated improved myocardial repair, but the evaluation window remained relatively short. Longer observation periods, particularly those assessing structural integration and functional persistence, will help define the durability of Psr-enhanced reprogramming. Second, although our in vivo data demonstrate functional improvement and reduced fibrosis following RF + Psr treatment, the current study does not include lineage tracing to definitively confirm that these benefits are directly attributable to fibroblast-to-cardiomyocyte conversion. Future studies employing fibroblast-specific reporter models (e.g., Tcf21-CreER; Rosa26-tdTomato) will be critical to distinguish the relative contributions of direct reprogramming versus indirect cardioprotective effects. Finally, although HSF experiments demonstrated induction of cardiomyocyte-like gene expression and mitochondrial remodeling, these cells did not exhibit full functional maturation. Further validation in primary adult human fibroblasts will help strengthen the translational relevance of Psr-assisted cardiac reprogramming.

In summary, this study identifies Psr as an effective enhancer of chemical cardiac reprogramming that promotes structural and metabolic maturation through coordinated activation of mitochondrial fission and PPARα-associated metabolic pathways. These findings highlight the therapeutic potential of integrating natural small molecules into chemical reprogramming strategies and provide a tractable approach for improving the efficiency and maturity of iCMs, with potential applications in cardiac repair and regeneration.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

iCMs

Induced cardiomyocytes

GMT

Gata4, Mef2c, Tbx5

MEFs

Mouse embryonic fibroblasts

RF

RepSox + Forskolin

Psr

Psoralen

RF + Psr

RepSox + Forskolin + Psoralen

CM

Cardiomyocyte differentiation medium

Ryr2

Ryanodine receptor 2

Tnnt2

Troponin T2, cardiac type

Gata4

GATA binding protein 4

β-MHC

Beta-myosin heavy chain

Myh6

Myosin heavy chain 6

Myl7

Myosin light chain 7

Nkx2.5

NK2 homeobox 5

cTnI

Cardiac troponin I

cTnT

Cardiac troponin T

α-actinin

Alpha-actinin

Vimentin

Intermediate filament protein marker of fibroblasts

TEM

Transmission electron microscopy

OCR

Oxygen consumption rate

ECAR

Extracellular acidification rate

2-DG

2-Deoxy-glucose

FCCP

Carbonyl cyanide-4-(trifluoromethoxy)phenylhydrazone

Eto

Etomoxir

PPAR

Peroxisome proliferator-activated receptor

PPARα

Peroxisome proliferator-activated receptor alpha

RXRG

Retinoid X receptor gamma

UCP1

Uncoupling protein 1

Drp1

Dynamin-related protein 1

Fis1

Fission 1

Opa1

Optic atrophy 1

Mfn1/Mfn2

Mitofusin 1/Mitofusin 2

MFI

Mean fluorescence intensity

EF

Ejection fraction

FS

Fractional shortening

LVIDd/s

Left ventricular internal dimension at diastole/systole

LVPWs

Left ventricular posterior wall thickness at end-systole

SV

Stroke volume

MI

Myocardial infarction

LVADs

Left ventricle assist devices

Author contributions

Zhiguo Zhang and Ruyuan Zhu secured the funding that supported this study. Ruyuan Zhu designed and supervised the research. Wenjie Li performed the majority of the experiments and drafted the manuscript. Xinyu Wan and Ding Cheng contributed to experimental execution and data acquisition. Zhiguo Zhang, Haixia Liu, and Ruyuan Zhu critically revised the manuscript for important intellectual content. Ruyuan Zhu and Zhiguo Zhang contributed equally to this work and should be regarded as co-corresponding authors. All authors have read and approved the final manuscript.

Funding

This study was financially supported by the Fundamental Research Funds for the Central public welfare research institutes (ZZ16-YQ-033, YZX-202410), Basic Scientific Research Fund of Chinese Academy of Traditional Chinese Medicine (grant number: CYZ-202507), National Natural Science Foundation of China (grant number: 82304919), National Natural Science Foundation of China (grant number: 82074297).

Data availability

The raw RNA sequencing data generated during this study have been deposited in the National Center for Biotechnology Information (NCBI) BioProject database under accession number PRJNA1393497. The processed gene expression count matrix is provided as Supplementary Data File 1(All_samples_gene_counts.xlsx). Any other datasets used or analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

All experimental procedures involving animals were conducted in strict accordance with the Guidelines for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee of the Laboratory Animal Center, Institute of Basic Theory for Chinese Medicine, China Academy of Chinese Medical Sciences (Approval number: IBTCMS27-2409-06; Date of approval: 2024-09-04). The approved project title is: Optimization and Mechanistic Study of a Novel Psoralen-Containing Formulation for Direct Cardiac Reprogramming. Every effort was made to minimize animal suffering and to reduce the number of animals used while maintaining scientific validity.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.