The bmp10 gene was knocked out in zebrafish using CRISPR–Cas9 technology, generating bmp10^+⁣/- and bmp10^-⁣/- mutants. kdrl:EGFP transgenic zebrafish, including bmp10 wild type (bmp10^+⁣/+, WT) and bmp10^-⁣/- variants, were derived from the AB wild-type strain (Danio rerio), obtained from the National Zebrafish Resource Center (Wuhan, China). These gene-edited zebrafish served as parental fish for generating the bmp10^-⁣/- offspring. For convenience in selected figure legends, bmp10^-⁣/- zebrafish are termed HO, and bmp10^+⁣/+ zebrafish as WT. In this study, 15 °C is defined as hypothermic conditions, whereas 28 °C represents normothermic conditions.

The bmp10 gene is located on chromosome 5. Gene knockout was achieved using the CRISPR–Cas9 system targeting the sequence GGGTTGGATGATGGACATGG, resulting in a 72-bp deletion. gRNA was synthesized using the MAXIscript™ T7 in vitro transcription kit (AM1314, Ambion [Invitrogen], Waltham, MA, USA), and Cas9 protein (C120010, Sigma-Aldrich, St. Louis, MO, USA) (800 pg) with gRNA (100 pg) was co-injected into one-cell stage embryos. Genotyping of bmp10 was performed by PCR using the following primers: forward, ATTTTCATCTGCCACGTCTGC; reverse: TAGAGTTCCAGCATATACTCAG. PCR amplification was conducted using Takara rTaq DNA polymerase (RR901A, Takara Bio Inc., Kusatsu, Shiga, Japan). bmp10^-⁣/- mutants were identified by agarose gel electrophoresis using molecular biology-grade agarose (75510-019, Invitrogen, Carlsbad, CA, USA).

Offspring from bmp10^+⁣/- self-crosses were raised under normothermic conditions (28 °C) until 6 weeks post fertilization (wpf). Juvenile zebrafish (bmp10^+⁣/+, bmp10^-⁣/-, and bmp10^+⁣/-) were then assigned to either a hypothermic group 15 °C or a normothermic control group 28 °C.

The hypothermic exposure protocol consisted of three stages:

Stage 1: Two-week hypothermic exposure

At 6 wpf, zebrafish were exposed to hypothermic or normothermic conditions for 2 weeks (until 8 wpf; followed by analyses of body weight, cardiac morphology, gene expression, apoptosis, histology, and blood cell counts.

Stage 2: Temperature-switching experiments

(1) Zebrafish exposed to hypothermia from 6–8 wpf were transferred to normothermic conditions until 10 wpf.

(2) Zebrafish maintained under normothermic conditions from 6–8 wpf were transferred to hypothermic conditions until 10 wpf.

Stage 3: Long-term hypothermic exposure

Beginning at 6 wpf, zebrafish were continuously maintained under hypothermic conditions for 8 weeks (until 14 wpf).

Angiotensin-converting enzyme inhibitors (ACEIs), such as benazepril hydrochloride (HY-B0093A, MedChemExpress, Monmouth Junction, NJ, USA), reduce blood pressure and alleviate cardiac load by inhibiting the renin–angiotensin–aldosterone system. The experiment utilized bmp10^+⁣/- zebrafish offspring, which were raised at 28 °C until they reached 6 wpf. At the start of the experiment, equal numbers of juveniles (bmp10^+⁣/+; bmp10^-⁣/-) were assigned to either the benazepril hydrochloride treatment group or the control group (which received only the solvent without the drug) for a 2-week treatment period (until 8 wpf). Before the experiment, juveniles in both groups were allocated in equal numbers, with each tank containing no more than 15 juveniles (ensuring a sample size of n $\ge$5 for each group). Benazepril hydrochloride treatment was administered once daily for 1 hour, at a concentration of 50 µM. The working solution was prepared by adding 100 µL of a 25.0 mg/mL dimethyl sulfoxide (DMSO) (HY-Y0320, MedChemExpress, Monmouth Junction, NJ, USA) stock solution to 400 µL of polyethylene glycol 300 (PEG300) (HY-Y0873, MedChemExpress, Monmouth Junction, NJ, USA), thoroughly mixing. Then, 50 µL of tween-80 (HY-Y1891, MedChemExpress, Monmouth Junction, NJ, USA) was added, followed by mixing, and 450 µL of physiological saline was added to bring the total volume to 1 mL. During the experiment, water quality parameters (including oxygen levels), feeding frequency, and timing were kept consistent between the two groups. Juvenile survival was monitored daily, and a simulated day–night cycle was maintained. At the end of the 2-week treatment (at 8 wpf), the zebrafish from both groups were sampled, and body morphology and cardiac changes in the bmp10 mutant zebrafish were recorded.

Zebrafish were anesthetized, and hearts were rapidly dissected and fixed in 4% paraformaldehyde (E672002-0500, BBI Life Sciences, Shanghai, China) at 4 °C for 24 h. Following fixation, samples were subjected to graded ethanol (A500737-0500, BBI Life Sciences, Shanghai, China) dehydration, xylene (X103350, Sinopharm Chemical Reagent Co., Ltd., Beijing, China) clearing, and paraffin embedding. Paraffin-embedded heart tissues were sectioned into 4 µm-thick serial sections using Leica low-profile disposable blades 819 (14035838925, Leica Biosystems, Wetzlar, Germany) for subsequent histological analyses. Heart sections were stained with hematoxylin and eosin staining kit (E607318-0200, BBI Life Sciences, Shanghai, China) and masson’s trichrome staining kit (G1340, Solarbio, Beijing, China) in accordance with the manufacturers’ instructions to assess overall cardiac morphology and histological features. Images were acquired using a stemi 2000 stereomicroscope (Zeiss, Oberkochen, Germany).

Myocardial cell apoptosis in cardiac sections (4 µm) was examined using the TUNEL BrightGreen apoptosis detection kit (A112-02, Vazyme Biotech Co, Nanjing, China). The specimens were mounted with anti-fade mounting medium (C02-04003, Bioss Antibodies, Beijing, China) medium for fluorescence and sealed. Apoptotic signals were captured using a LSM710 confocal laser scanning microscope (Carl Zeiss AG, Oberkochen, Germany), and quantitative analysis of the apoptotic signals was performed using ImageJ software (version 1.53, National Institutes of Health, Bethesda, MD, USA).

Zebrafish were anesthetized with 0.02% tricaine (MS-222) (A5040, Sigma-Aldrich, St. Louis, MO, USA). Approximately 0.5 µL of peripheral blood was collected via tail vein transection and immediately diluted with 2 µL of ultrapure water. The diluted blood was dropped onto one end of a glass slide, and blood smears were prepared using the push–slide method. After air-drying, smears were fixed with absolute methanol (A601617, BBI Life Sciences, Shanghai, China) for 5 min. Blood smears were stained using a Wright–Giemsa staining kit (E607314, BBI Life Sciences, Shanghai, China), and blood cell morphology was analyzed under a light microscope.

Paraffin-embedded zebrafish heart sections were deparaffinized prior to staining. Cell membranes were stained with wheat germ agglutinin, alexa fluor 488 conjugate (G1730, Invitrogen, Carlsbad, CA, USA) for 15 min at room temperature in the dark, followed by three washes with phosphate-buffered saline (PBS) (E607008-0500, BBI Life Sciences, Shanghai, China). Nuclei were counterstained with Hoechst 33258 dye (33258, Solarbio Life Sciences, Beijing, China) for 5 min at room temperature in the dark, followed by PBS washes. After staining, sections were mounted using an anti-fade mounting medium (C02-04003, Bioss Antibodies, Beijing, China) and imaged using a confocal microscope.

Gene sequences for bmp10, bone morphogenetic protein 10-like (bmp10l), and Hypoxia-inducible factor 3 (egln3) were downloaded from the Ensembl website, and primers were designed based on the CDSs of these genes. The primer sequences for bmp10 were F: ATTTTCATCTGCCACGTCTGC and R: TAGAGTTCCAGCATATACTCAG. The primer sequences for bmp10l were F: AGCGGAGATGTACGTGGATT and R: CTGGACGATGGCGTGTTTAG. The primer sequences for egln3 were F: GCTTTGTTGGATCAGGGCTT and R: GATGTTCCTCCGGCAAACTC. First, RNA extraction was performed using the RNeasy Mini Kit (74104, Qiagen N.V, Hilden, Germany), followed by cDNA synthesis using TransScript Uni All-in-One First-Strand cDNA Synthesis SuperMix for qPCR (AU341-02-V2, TransGen Biotech Co, Beijing, China) with the extracted RNA. Finally, quantitative fluorescence analysis was performed using TB Green Premix Ex Taq II (Tli RNaseH Plus) (RR420A, Takara Bio Inc, Kusatsu, Shiga, Japan) on a CFX Opus 96 Real-Time PCR System (Bio-Rad Laboratories, Inc., Hercules, CA, USA) instrument.

All statistical analyses were performed using GraphPad Prism software (version 8.0.1, GraphPad Software Inc., San Diego, CA, USA). Data are presented as mean $\pm$ standard error of the mean (SEM). Differences between two groups were analyzed using a two-tailed Student’s t-test. Statistical significance to be defined as p 0.05 (*), p 0.01 (**), p 0.001 (***), and p 0.0001 (****); ns indicates no statistically significant difference.

Zebrafish larvae were immersed in 200 mg/L tricaine (MS-222; Sigma–Aldrich) for approximately 5 minutes until all movement and responses completely ceased. For euthanasia, zebrafish were exposed to an overdose of 400 mg/L tricaine for 15 minutes until they were completely unresponsive and cardiac activity had stopped. To ensure complete death, the fish were subsequently subjected to an ice-water bath at approximately 0 °C for 5 minutes following anesthesia. This procedure was conducted in accordance with the AVMA Guidelines for the Euthanasia of Animals (2020 edition). The tricaine stock was prepared in deionized water and diluted with system water to working concentrations of 200 or 400 mg/L. The pH of the working solutions was adjusted to 7.0–7.5 using sodium bicarbonate (NaHCO₃) (A100865-0500, BBI Life Sciences, Shanghai, China).

Chronic hypothermic exposure markedly reversed cardiac hypertrophy in bmp10^-⁣/- zebrafish (Fig. 1A,B). Under normothermic conditions (28 °C), bmp10^-⁣/- zebrafish exhibited pronounced cardiac hypertrophy by 8 wpf (Fig. 1A,B). Following chronic hypothermic exposure (15 °C), heart size was reduced in both wild-type and bmp10^-⁣/- zebrafish, and the difference between the two groups was no longer detectable. These results indicate that hypothermic exposure attenuates myocardial hypertrophy induced by bmp10 deficiency, rather than producing a mutant-specific effect. Consistently, under normothermic conditions, body weight in 8-week-old bmp10^-⁣/- zebrafish differed significantly from that of control fish, and this difference was more pronounced than that observed at the early stage (6 weeks of age) (Fig. 1C). Notably, following hypothermic exposure, this difference was markedly abolished (Fig. 1D). Based on these findings, bmp10^-⁣/- zebrafish were selected as an appropriate experimental model for subsequent mechanistic investigations under hypothermic conditions. Downregulation of egln3 (Fig. 1G–I) was accompanied by activation of the hif-1 signaling pathway involved in cardiomyocyte hypoxic responses, together with significant downregulation of the pro-apoptotic genes bnip3 and bcl2l10 (Fig. 2F). This molecular profile is consistent with reduced cardiomyocyte size (Fig. 1B) and attenuation of myocardial injury. Although bmp10l exhibits partial functional redundancy with bmp10 during cardiac development, its expression did not fully compensate for the structural defects caused by bmp10 deficiency (Fig. 1E,F). In addition, hypothermic exposure significantly increased the expression of Kruppel-like factor 1 (klf1) in bmp10^-⁣/- zebrafish hearts (Fig. 2F), a gene implicated in myocardial injury repair and cardiomyocyte proliferation.

Fig. 1.

Comparison of zebrafish lacking bone morphogenetic protein 10 (bmp10^-⁣/-) treated at different temperatures. (A) Calculation of cardiac area based on heart chambers (ventricle, atrium, and bulbus arteriosus). Scale bar = 200 µm. (B) The cardiac area of bmp10^-⁣/- zebrafish maintained under normothermic conditions (28 °C) was significantly larger than that of bmp10^-⁣/- zebrafish subjected to chronic cold exposure under hypothermic conditions (15 °C). (C) Body weight of bmp10^-⁣/- zebrafish at 6 weeks of age (4 weeks of age followed by 2 weeks of hypothermic treatment) under normothermic and hypothermic conditions. (D) Body weight of bmp10^-⁣/- zebrafish at 8 weeks of age (6 weeks of age followed by 2 weeks of hypothermic treatment) under normothermic and hypothermic conditions. (E) Heart rate measurements of bmp10^-⁣/- zebrafish at 72–84 hours post-fertilization (hpf) under normothermic conditions (28 °C). (F) Heart rate measurements of bmp10^-⁣/- zebrafish at 96–108 hpf under normothermic conditions (28 °C). (G) Gene expression comparisons under different temperature conditions: 28 °C–15 °C HO (gene expression differences in bmp10^-⁣/- zebrafish between 28 °C and 15 °C), 28 °C–15 °C WT (gene expression differences in bmp10^+⁣/+ zebrafish between 28 °C and 15 °C), 15 °C HO–15 °C WT (gene expression differences between bmp10^-⁣/- and bmp10^+⁣/+ zebrafish at 15 °C), and 28 °C HO–28 °C WT (gene expression differences between bmp10^-⁣/- and bmp10^+⁣/+ zebrafish at 28 °C). (H) Analysis of Hypoxia-inducible factor-1 (hif-1) pathway factor Egl-9 family hypoxia-inducible factor 3 (egln3) expression in bmp10^+⁣/+ zebrafish under hypothermic (15 °C) and normothermic (28 °C) conditions. (I) Analysis of hif-1 pathway factor egln3 expression in bmp10^-⁣/- zebrafish under hypothermic (15 °C) and normothermic (28 °C) conditions. Data are presented as mean $\pm$ SEM. p 0.05 (*), p 0.01 (**); ns indicates no statistically significant difference.

Fig. 2.

Chronic cold exposure alleviates cardiomyocyte hypertrophy. (A) Morphological and fibrotic staining analysis of bmp10^-⁣/- zebrafish hearts after 2 weeks of exposure to hypothermic (15 °C) or normothermic (28 °C) conditions (from 6–8 wpf). Scale bar = 250 µm. (B) Red blood cell density analysis of bmp10^-⁣/- zebrafish hearts after 2 weeks of exposure to hypothermic (15 °C) or normothermic (28 °C) conditions (from 6–8 wpf). Scale bar = 500 µm (estimated). (C) Cardiomyocyte apoptosis comparison between bmp10^-⁣/- and bmp10^+⁣/+ zebrafish under different temperature conditions. Scale bar = 10 µm. (D) Quantitative analysis of cardiomyocyte apoptosis in panel C. (E) Relative mRNA expression levels of egln3, bone morphogenetic protein 10-like (bmp10l), and hepcidin antimicrobial peptide (hamp) in bmp10^-⁣/- zebrafish hearts after 2 weeks of exposure to hypothermic (15 °C) or normothermic (28 °C) conditions (by 8 wpf). (F) Relative mRNA expression levels of egln3, Bone morphogenetic protein 10-like (bmp10l), Kruppel-like factor 1 (klf1), BCL2 interacting protein 3 (bnip3), BCL2-like 10 (bcl2l10), BCL2-like 11 (bcl2l11), forkhead box O (foxo3b), Uncoupling protein 1 (ucp1), Uncoupling protein 3 (ucp3), Unc-5 netrin receptor b (unc5b), and Unc-5 netrin receptor da (unc5da) in bmp10^-⁣/- zebrafish cardiomyocytes under hypothermic (15 °C) and normothermic (28 °C) conditions. For individual gene expression levels, refer to Supplementary Fig. 1. (G) Imaging of major arteries in bmp10^-⁣/- and bmp10^+⁣/+ zebrafish at 48 and 72 hpf under normothermic conditions (28 °C). Scale bar = 200 µm (estimated). Data are presented as mean $\pm$ SEM. p 0.05 (*), p 0.001 (***), and p 0.0001 (****).

To ameliorate cardiac hypertrophy that had already by 8 wpf, bmp10^-⁣/- zebrafish were subjected to hypothermic exposure for 2 weeks starting at 6 wpf (Fig. 3Aa). Following hypothermic intervention (15 °C), bmp10^-⁣/- zebrafish exhibited substantial alterations in overall cardiac morphology (Fig. 2A). Moreover, erythrocyte density within the cardiac chambers showed substantial differences compared with those maintained under normothermic conditions (28 °C) (Fig. 2B). Hypothermic exposure significantly reduced cardiomyocyte apoptosis, as demonstrated by TUNEL staining and quantitative analysis (Fig. 2C,D). Gene expression analysis indicates that egln3 expression was significantly lower under hypothermic conditions than under normothermic conditions (Fig. 1G–I). These data indicate that egln3 downregulation may contribute to reduced cardiomyocyte apoptosis in bmp10^-⁣/- zebrafish.

Fig. 3.

Cardiac and overall health status of bmp10^-⁣/- zebrafish following temperature transitions. (A) Low temperature: Normothermic group (Control group)—bmp10^-⁣/- zebrafish were maintained under normothermic conditions (28 °C) until 8 wpf. Hypothermic group (Experimental group)—bmp10^-⁣/- zebrafish were maintained under normothermic conditions until 6 wpf and then transferred to hypothermic conditions (15 °C) for 2 weeks (until 8 wpf, green box). Temperature adaptation protocol: (i) bmp10^-⁣/- zebrafish were maintained under normothermic conditions (28 °C) until 6 wpf, followed by an additional 2 weeks at 28 °C (until 8 wpf; yellow box), and subsequently transferred to hypothermic conditions (15 °C) for 2 weeks (until 10 wpf); (ii) bmp10^-⁣/- zebrafish were maintained at 28 °C until 6 wpf, then exposed to 15 °C for 2 weeks (green box), and finally returned to 28 °C for an additional 2 weeks (until 10 wpf). (B) Cardiac area comparison between bmp10^-⁣/- and bmp10^+⁣/+ zebrafish following 8 weeks of chronic cold exposure (until 14 wpf). (C) Cardiac area comparison between bmp10^-⁣/- and bmp10^+⁣/+ zebrafish after transitioning from hypothermic (15 °C) back to normothermic (28 °C) conditions (until 10 wpf). (D) Cardiac area comparison between bmp10^-⁣/- and bmp10^+⁣/+ zebrafish after transitioning from normothermic (28 °C) to hypothermic (15 °C) conditions (until 10 wpf). (E) Body weight comparison between bmp10^-/– and bmp10^+⁣/+ zebrafish following temperature transition from normothermic to hypothermic conditions (until 10 wpf). (F) Body weight comparison between bmp10^-⁣/- and bmp10^+⁣/+ zebrafish after transitioning from hypothermic back to normothermic conditions (until 10 wpf). (G) Body weight comparison between bmp10^-⁣/- and bmp10^+⁣/+ zebrafish following 8 weeks of chronic cold exposure (until 14 wpf). (H) Survival rates of bmp10^-⁣/- zebrafish maintained under normothermic conditions (28 °C) or exposed to three different hypothermic treatment (until 10 wpf). For individual survival curves at each stage, refer to Supplementary Fig. 2. Data are presented as mean $\pm$ SEM. p 0.01 (**); ns indicates no statistically significant difference.

Forkhead box O (foxo) family members (including foxo3b, foxo1a, and foxo1b) can modulate the stability and activity of hif-1$\alpha$ by regulating the expression or activity of egln genes, thereby influencing cellular responses to hypoxic conditions. The present findings reveal that foxo3b increases hif-1$\alpha$ activity by suppressing egln3 expression (Fig. 2F), thereby attenuating the hydroxylation of hif-1$\alpha$. Downregulation of egln3 through activation of the hif-1 signaling pathway inhibited excessive apoptosis in bmp10^-⁣/- cardiac cells, alleviating damage to bmp10^-⁣/- cardiomyocytes and ultimately protecting cardiac structure and function (Fig. 2E,F). Specifically, reduced egln3 expression facilitated activation of the anti-apoptotic hif-1$\alpha$ signaling pathway, resulting in increased expression of hif-1$\alpha$ and its downstream target gene BCL2-like 11 (bcl2l11) (Fig. 2F). In parallel, the expression levels of pro-apoptotic genes bnip3 and bcl2l10 were correspondingly reduced (Fig. 2F), thereby suppressing mitochondrial outer membrane–mediated apoptotic signaling and attenuating apoptotic signal transmission. Furthermore, exposure to hypothermic conditions increased the expression of uncoupling proteins, including uncoupling protein 1 (ucp1) and ucp3, in the heart (Fig. 2F); leading to reduced reactive oxygen species production and attenuation of oxidative stress.

Fertilized embryos transferred to 15 °C within 3 hours post-fertilization (hpf) resulted in embryonic lethality by 48 hpf. Comparative analysis of cardiac vascular morphology in bmp10^-⁣/- zebrafish maintained at 28 °C demonstrated substantial differences at 48 and 72 hpf relative to control embryos (Fig. 2G), indicating that bmp10 plays an essential role in vascular development. Differential gene expression analysis demonstrated significant upregulation of unc-5 netrin receptor b (unc5b) and unc-5 netrin receptor da (unc5da) in bmp10^-⁣/- zebrafish hearts under hypothermic conditions (Fig. 2F). These genes are involved in endothelial cell migration and proliferation and may influence vascular network formation and cardiac contractile function.

Under normothermic conditions, bmp10 deficiency significantly reduced survival after 6 wpf compared with wild-type controls (Fig. 3H). Continuous hypothermic exposure from 6–14 wpf significantly increased survival, reaching 72.7% compared with 45% under normothermic conditions (Fig. 3H). These data demonstrate that chronic hypothermic exposure improves both survival and cardiovascular health in bmp10^-⁣/- zebrafish (Fig. 4A–F; Fig. 3).

Continuous treatment of bmp10^-⁣/- zebrafish with benazepril hydrochloride (50 µM, 1 hour daily for two weeks), an ACEI, significantly ameliorated myocardial hypertrophy and reduced cardiomyocyte apoptosis (Fig. 5A–C). Concomitantly, erythrocyte density increased and ventricular wall thickness approached normal levels (Fig. 5D,E). These phenotypic improvements were accompanied by downregulation of egln3 expression. The molecular and phenotypic alterations induced by ACEI treatment closely resembled those observed under chronic hypothermic exposure (Fig. 5F).

Fig. 5.

Treatment of bmp10^-⁣/- zebrafish with ACEIs. (A) Cardiomyocyte apoptosis in bmp10^-⁣/- and bmp10^+⁣/+ zebrafish after 2 weeks of ACEI treatment under normothermic conditions (28 °C) (until 10 wpf). Scale bar = 10 µm. (B) Relative expression levels of egln3 and bmp10l following 2 weeks of ACEI treatment (until 10 wpf). (C) Quantitative analysis of body weight, heart area, and cardiomyocyte apoptosis in bmp10^-⁣/- zebrafish following 2 weeks of ACEI treatment (until 10 wpf). (D) Erythrocyte density in bmp10^-⁣/- zebrafish following 2 weeks of ACEI treatment (until 10 wpf). Scale bar = 500 µm (estimated). (E) Morphological analysis of myocardial structure in bmp10^-⁣/- zebrafish following 2 weeks of ACEI treatment (until 10 wpf). Scale bar = 200 µm. The area at the edge of the tissue indicated by the white arrow is the ventricular wall of a zebrafish. (F) Expression levels of egln3, bmp10l, klf1, bnip3, bcl2l10, bcl2l11, foxo3b, ucp1, ucp3, unc5b, and unc5da following 2 weeks of ACEI treatment. For individual gene expression data, refer to Supplementary Fig. 5. Data are presented as mean $\pm$ SEM. p 0.05 (*), p 0.01 (**), p 0.001 (***); ns indicates no statistically significant difference.