Adrenergic-thyroid hormone interactions drive postnatal thermogenesis and loss of mammalian heart regenerative capacity

Newborn mice possess a robust but transient capacity for heart regeneration that is lost when cardiomyocytes (CMs) permanently exit the cell cycle and binucleate during the first week after birth¹. We discovered that increasing levels of circulating thyroid hormones (THs) observed during the acquisition of endothermy promote postnatal CM cell cycle arrest and limits heart regenerative potential². In this study, we describe novel interactions between adrenergic receptor (AR) and TH signaling regulating thermogenesis, CM proliferation, and heart regeneration.

Diploid CM abundance, a proxy for heart regenerative potential, is inversely correlated with Standard Metabolic Rate (SMR), which is defined as Basal Metabolic Rate (BMR) divided by animal body mass to the ¾ power². BMR is directly proportional to the blood volume flow rate required for oxygen and nutrient transport³. Thus, the >10-fold increase in SMR during the acquisition of endothermy² would require similar elevations in blood flow and cardiac function. These cardiac adaptations may impede CM proliferative and regenerative capacity.

We hypothesize that elevations in cardiac function necessary to support mammalian thermogenic pathways drive postnatal CM cell cycle arrest, polyploidization, and loss of heart regenerative potential. Sympathetic nerves signal through α- and β-ARs, and are critical regulators of thermogenesis⁴. Indeed, mice administrated with both αAR and βAR inhibitors – phenoxybenzamine (αARi) and propranolol (βARi) – have reduced body temperatures (Figure 1A), similarly to those treated with propylthiouracil (PTU), which inhibits thyroid hormone synthesis (Figure 1A). Importantly, treatment with both AR inhibitors and PTU together has the most profound effect in reducing body temperature (Figure 1A). These results suggest that AR and TH signaling interactions drive postnatal thermogenesis.

Figure 1. Inhibition of adrenergic receptor (AR) and thyroid hormone (TH) signaling delays postnatal body temperature elevation, cardiomyocyte (CM) cell-cycle exit, and the loss of cardiac regenerative potential.

(A) Body temperatures of neonatal mice treated with saline, αARi (phenoxybenzamine; 10 μg/g/day), βARi (propranolol; 20 μg/g/day), and propylthiouracil (PTU; 0.15% in iodine deficient diet; ad libitum) individually and in combination. PTU was given to pregnant and nursing females through food chow starting at E13.5 and maintained thereafter. αARi and βARi was subcutaneously injected from P1 to P14. Representative infrared thermography images are shown on the right. (B) Bodyweights, heart weights, and heart weight-to-bodyweight (HW:BW) ratios after treatment with αARi, βARi, and/or PTU at P14. (C-E) Analyses of total cardiomyocyte (CM) number at P14 determined by design-based stereology (as described in Hirose et al., 2019) (C), proliferating CMs determined by colocalization of PCM1 (CM perinuclear marker) and Ki67 (pan-cell cycle marker) (D), and the percentage of diploid mononucleated CMs determined by quantification of CM-specific nuclear DAPI fluorescence intensity (as described in Hirose et al., 2019) (E) in animals with indicated treatments. Arrowheads denote cycling CMs in (D) and diploid CMs in (E), respectively. (F-J) Assessment of cardiac regenerative potential after P14 myocardial infarction (MI). (F) Schematic presentation of experimental design. (G) Analysis of CM proliferative activity determined by colocalization of PCM1 with either Ki67 or phospho-histone H3 (pHH3) (mitosis marker). (H) EdU incorporation and CM ploidy analysis in dissociated CMs 28 days after MI (as described in Hirose et al., 2019). (I) Analysis of cardiac ejection fraction by echocardiography. (J) Cardiac fibrosis analysis. Representative images of heart sections (left) at 300 μm (for the first three groups) or 250 μm (for the fourth group treated with αARi, βARi and PTU) intervals and quantification of fibrotic area% (right). Fibrotic tissues are stained by Sirius red and viable myocardial cells are stained by Fast green. (K) Differential gene expression analysis after combined treatment with αARi, βARi, and PTU in whole P14 hearts (as described Hirose et al., 2019). False discovery rate (FDR) < 0.1 and fold-change > 2 were applied as cutoffs. K-means clustering and gene ontology (GO) classification of differentially expressed genes. Top upregulated and downregulated pathways are shown. The RNA-seq dataset has been deposited in Gene Expression Omnibus (GEO) under accession number GSE174511. All values are reported as Mean ± SEM. Number of animals (n) analyzed is identified in each figure. Statistical analyses were performed in GraphPad Prism. Two-way repeated measures ANOVA was used in (A and I). Only the analysis results at P14 are presented in (A) due to the space constraint. Two-way ANOVA was used in (C, G, H, J) and three-way ANOVA was used in (B, D, E). Pair-wise comparisons between all pairs of means were performed following ANOVA in multiple comparisons. In all analyses, Tukey test was used to correct for multiple comparison. *p<0.05, **p<0.01, ***p<0.001. ****p<0.0001. Scale bars, 10 μm (D), 50 μm (E), 4 mm (J).

We next determined if AR and TH signaling regulate CM number and proliferation. Certain chemical combinations decrease animal bodyweight and heart weight but not heart weight-to-bodyweight ratio (Figure 1B). We quantified total CM number in postnatal day 14 (P14) hearts using stereology. Combined treatment with α/β-AR inhibitors and PTU increases total CM numbers (Figure 1C), and CM cell cycle activity (Figure 1D). Additionally, treatment with the inhibitor cocktail increases retention of mononucleated diploid CMs (59%), substantially higher than that observed in control animals (5.3%) and those treated with just α/β-AR blockers (16%) or PTU (21%) (Figure 1E). Consistent with a recent report that βAR signaling promotes CM cytokinesis failure⁵, we observed that inhibition of βAR signaling alone increases diploid CM abundance (Figure 1E). These results suggest that AR and TH signaling interactions also promote postnatal CM cell-cycle arrest and polyploidization.

To test if combined AR and TH signaling inhibition enhances CM regeneration after injury, we treated mice with pathway inhibitors and then occluded the coronary artery to induce myocardial infarction (MI) at P14 (Figure 1F). Combined pathway inhibition increases CM cell cycle activity in the border zone, overall CM EdU incorporation, and diploid EdU-positive CMs indicating successful CM cell division (Figures 1G and 1H). CM-specific inactivation of TH signaling alone improved cardiac function after ischemia-reperfusion (IR) injury², but not after MI (data not shown). MI may cause more CM loss than IR injury and likely requires robust CM proliferation to support regeneration. Thus, we investigated cardiac regenerative capacity in mice treated with the inhibitor cocktail. At 28 days post-MI, we observed a significant increase in cardiac ejection fraction and reduction in cardiac fibrosis in these animals (Figures 1I and 1J). Additionally, RNA-seq and differential gene expression analysis revealed that genes involved in cell cycle regulation were significantly upregulated in whole P14 hearts after combined inhibition of AR and TH signaling (Figure 1K), suggesting these pathways are upstream regulators of CM cell cycle. These data suggest that combined α/β-AR and TH inhibition extends postnatal cardiac regenerative capacity in part by promoting CM cell division. It is possible that combined pathway inhibition also affects inflammation, angiogenesis, revascularization after injury, cardiac metabolism, and heart size which may contribute to enhanced cardiac regeneration.

Collectively, our results demonstrate that postnatal AR and TH signaling interactions promote mammalian thermogenesis, inhibit CM cell division, and limit cardiac regenerative capacity. It is unlikely that reducing body temperature directly increases CM cell cycle entry as culturing E18 rat ventricular CMs at lower temperatures reduces CM proliferation based on EdU incorporation assays (37°C: 29.26±0.025%; 31°C: 13.14±0.023%; 25°C: 0.14±0.002%, n=3). Future studies will be necessary to determine if manipulation of AR and TH signaling beyond postnatal development will improve cardiac regenerative capacity in adult animals. Understanding how mammalian thermogenesis and CM cell cycle withdrawal are linked are likely to yield evolutionary insights into why adult mammals cannot regenerate the heart.

Animal procedures were conducted in accordance with UCSF Institutional Animal Care and Use Committee. The data, analytical methods, and study materials are available on request.