asb5a/asb5b Double Knockout Affects Zebrafish Cardiac Contractile Function

Abstract

Ankyrin repeat and suppression-of-cytokine-signaling box (Asb) proteins, a subset of ubiquitin ligase E3, include Asb5 with six ankyrin-repeat domains. Zebrafish harbor two asb5 gene isoforms, asb5a and asb5b. Currently, the effects of asb5 gene inactivation on zebrafish embryonic development and heart function are unknown. Using CRISPR/Cas9, we generated asb5a-knockout zebrafish, revealing no abnormal phenotypes at 48 h post-fertilization (hpf). In situ hybridization showed similar asb5a and asb5b expression patterns, indicating the functional redundancy of these isoforms. Morpholino interference was used to target asb5b in wild-type and asb5a-knockout zebrafish. Knocking down asb5b in the wild-type had no phenotypic impact, but simultaneous asb5b knockdown in asb5a-knockout homozygotes led to severe pericardial cavity enlargement and atrial dilation. RNA-seq and cluster analyses identified significantly enriched cardiac muscle contraction genes in the double-knockout at 48 hpf. Moreover, semi-automatic heartbeat analysis demonstrated significant changes in various heart function indicators. STRING database/Cytoscape analyses confirmed that 11 cardiac-contraction-related hub genes exhibited disrupted expression, with three modules containing these genes potentially regulating cardiac contractile function through calcium ion channels. This study reveals functional redundancy in asb5a and asb5b, with simultaneous knockout significantly impacting zebrafish early heart development and contraction, providing key insights into asb5’s mechanism.

1. Introduction

Ankyrin repeat and suppressor-of-cytokine-signaling (SOCS) box (Asb) proteins are part of the ubiquitin ligase E3 protein family. The Asb protein subfamily, featuring six ankyrin repeat domains (Asb5, Asb9, Asb11, and Asb13), draws interest owing to unusually strong evolutionary conservation [1]. The asb5 gene is highly homologous in higher vertebrates, such as humans, mice, and rabbits; it is a single-copy gene and is not typed [2]. However, in zebrafish, the asb5 gene is divided into two subtypes: asb5a and asb5b. In 2003, Asb5 was identified as a novel protein linked to arteriogenesis [2]. Additionally, omics analysis revealed the close association of the Asb gene family, including asb5, with skeletal muscle development [3]. Furthermore, porcine skeletal muscle gene expression analysis indicated upregulated expression of the asb5 gene, which is associated with muscle cell proliferation and differentiation [4]. Knockdown of asb11 in zebrafish has prompted the formation of an enlarged pericardial cavity. All of these preliminary studies provide some clues that asb5 may play a role in heart development. However, limited studies exist on the asb5 gene’s role in heart development and heart function, with a 2005 study proposing its relationship with cardiovascular formation due to it as a target gene of serum response factor [5].

Zebrafish is considered important as a crucial model organism currently used to study cardiovascular development. It shares similarities of cardiovascular and hematopoietic processes with humans. During the early embryonic developmental stages of zebrafish, its body color is transparent, and the shape and function of the heart can be visually observed in living embryos. However, so far, there have been, despite this advantage, to the best of our knowledge, no reports of any existing studies on asb5 function in zebrafish. By consulting the database, we found that asb5 is divided into two subtypes, asb5a and asb5b, in zebrafish. In this article, using CRISPR/Cas9 genome editing technology [6,7,8], we successfully established three asb5a knockout zebrafish lines. The results showed that asb5a knockout homozygotes had, and subsequently found, no obvious abnormal phenotypes during early embryonic development. Subsequently, whole-body embryonic in situ hybridization [9] was performed, and the results showed highly similar asb5a and asb5b expression patterns in early zebrafish development. We speculated that these two genes may have potential functional redundancy, resulting in an absence of malformation when knocking out asb5a alone. Using morpholino interference technology [10], we successfully obtained the asb5a^−/−:asb5b-ATGmo double-knockout zebrafish strain. RNA sequencing [11,12] was used to analyze the transcriptome differences between the asb5a^−/−:asb5b-ATGmo double-knockout group and the wild-type (WT) group, as well as the asb5a^−/− single-knockout group, respectively. We found that the double-knockout group was significantly enriched in functional items related to cardiac muscle contraction. Semi-automatic heartbeat analysis [13] highlighted significant changes in cardiac function indicators, including notable decreases in heart rate and cardiac fractional shortening, in the asb5a^−/−:asb5b-ATGmo double-knockout group, which were verified via quantitative polymerase chain reaction (qPCR) results.

Using the STRING database, Cytoscape, and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses, all genes under the cardiac muscle contraction entry in the asb5a^−/−:asb5b-ATGmo double-knockout group were compared with genes in the other two groups, leading to the production of an accurate protein–protein interaction (PPI) regulatory network. The top 20 hub genes from the two groups were screened, and 11 hub genes were identified which were closely linked to cardiac contraction in the asb5a^−/−:asb5b-ATGmo double-knockout group: acta1a, actc1a, actc1c, cacng2a, cacna2d1a, cacnb2a, cox7c, cox8b, uqcrc1, tnnc1a, and myl4. Based on the combination of PPI network analysis and literature data, we hypothesized that these genes form three modules regulating early heart development in zebrafish, potentially influencing cardiac contractile function by regulating calcium ion channels. Our results suggest functional redundancy between asb5a and asb5b, as their simultaneous knockout significantly affected early zebrafish heart development and contraction. Hence, our findings provide a crucial foundation for investigating the mechanism of the asb5 gene in zebrafish early development.

2. Results

2.1. Construction of asb5a-Knockout Zebrafish Line

To investigate the role of the asb5a gene in early zebrafish development, we developed an asb5a-knockout zebrafish strain. A pair of small guide RNAs (sgRNAs), namely Target 1 and Target 2 (Figure 1A), were designed for exon 1 of asb5a, resulting in threezebrafish lines (Lines 1–3) with frameshift mutant alleles. Sanger sequencing revealed specific mutations: Line 1, 8 bp insertion and 174 bp deletion; Line 2, 172 bp deletion; and Line 3, 175 bp deletion (Figure 1B,C). F2 homozygotes, derived from self-crossed heterozygotes, exhibited significantly downregulated asb5a mRNA expression (Figure S1) but no apparent phenotypic differences compared with the WT. This led us to explore the potential functional redundancy between asb5a and asb5b. We hypothesized that asb5b gene expression would compensate for the deficiency caused by complete asb5a gene knockout.

2.2. Whole-Embryo In Situ Hybridization Reveals the Similar Early Expression Patterns of asb5a and asb5b

As no observable embryonic malformations occurred in asb5a homozygous knockout lines, we hypothesized that this was due to functional compensation by asb5b in vivo. Whole-embryo in situ hybridization(WISH) experiments revealed the consistent expression patterns of asb5a (Figure 2A–E) and asb5b (Figure 3A–E) from early development to 4 days post-fertilization (dpf). Both genes exhibited similar expression patterns during early stages, with ubiquitous expression starting at 48 hours post-fertilization (hpf), and expression particularly concentrated in the eyes and head areas. Expression gradually decreased as the embryo developed, up to 4 dpf.

2.3. Construction of asb5a^−/−:asb5b-ATGmo Double-Knockout Zebrafish Strain

To confirm that asb5a knockout alone did not induce malformation due to asb5b functional compensation in vivo, we performed specific knockdown of asb5b in asb5a^−/− embryos and WT embryos, respectively. Morpholino oligos specifically targeting asb5b were produced in GeneTools to block transcription (asb5b-ATGmo: 5′-TAACCACAATGTGTGCCTACTTACT-3′; Figure S2A). Injecting WT one-cell stage embryos with 0.25Mm morpholino oligos effectively reduced asb5b mRNA expression (Figure S2B–D). An asb5b-ATGmo concentration gradient was used to determine the optimal injection concentration (Figure 4B–E). Injecting asb5b-ATGmo at the one-cell stage in the WT led to no obvious phenotype malformation or embryonic lethality at 48 hpf, regardless of low or high morpholino concentrations (Figure S3; Table S2). However, injecting asb5b-ATGmo at the one-cell stage in asb5a^−/− homozygous embryos(hereinafter, asb5a^−/−:asb5b-ATGmo double-knockout zebrafish) resulted in pericardial cavity enlargement and atrial dilation at 24 hpf, progressing to 48 hpf (Figure 4G–G″). Malformation severity increased with higher asb5b-ATGmo concentrations, with 0.75 mM considered the optimal concentration for subsequent experiments.

2.4. Transcriptome Sequencing Data Analysis

As the asb5a^−/−:asb5b-ATGmo double-knockout zebrafish displayed severe cardiac malformations by 48 hpf, high-throughput transcriptome sequencing was used to compare the WT, asb5a^−/− homozygous, and asb5a^−/−:asb5b-ATGmo double-knockout groups at 48 hpf. Relative to the WT, the asb5a^−/− group exhibited 879 upregulated and 1164 downregulated differentially expressed genes (DEGs). The asb5a^−/−:asb5b-ATGmo double-knockout group had 5766 upregulated and 4469 downregulated DEGs. Compared with the asb5a^−/− group, the double-knockout group had 6354 upregulated and 4233 downregulated DEGs (Figure 5A–C). All DEGs were shown in heatmaps in Figure 5D–F. Gene Ontology(GO) and KEGG pathway enrichment analyses revealed significant biological functions and pathways affected by the asb5a/asb5b deletion. Figure 6A–C showed the 20 most important GO entries, which included biological regulation (GO:0065007; biological process), catalytic activity (GO:0003824; molecular function), and cell part (GO:0044464, cellular component). In addition, the top 20 KEGG pathways included the MAPK signaling pathway, cardiac muscle contraction, and human papillomavirus infection (Figure 7A,B). Notably, asb5a knockout alone did not significantly impact any signaling pathways compared with the WT, further highlighting the potential limited effect of asb5a gene deletion in isolation.

2.5. Early Cardiac Contraction Abnormalities Induced by asb5a^−/−:asb5b-ATGmo Double-Knockout in Zebrafish

A detailed RNA-seq analysis revealed an abundance of cardiac muscle contraction pathways in KEGG pathway enrichment analysis regarding the asb5a^−/−:asb5b-ATGmo double-knockout group. This enrichment persisted compared with both WT and asb5a^−/− groups (illustrated in the purple box in Figure 7A,B). Thus, we hypothesized that concurrent knockout of the asb5a and asb5b genes in zebrafish impacts early cardiac contractile function. Using a high-speed electron-multiplying charge-coupled device (EM-CCD) camera, we recorded 10 s videos at 130 frames s⁻¹. Subsequently, semi-automatic heartbeat analysis software (Version 3.4.0.0) was employed to analyze the recorded heartbeat movies, generating various indicators related to cardiac contraction, including heart rate, heart period (HP), diastolic interval (DI), systolic interval (SI), diastolic diameter (DD), systolic diameter (SD), and fractional shortening(FS). All data were generated in M-mode. Approximately 30 zebrafish embryos at 48 hpf in each group were randomly selected and imaged under a 20× lens. Focusing on the atrial tissue, given the previously observed atrial dilation in the asb5a^−/−:asb5b-ATGmo double-knockout group in the myl7:EGFP strain, we photographed the atrium in M-mode to assess various aspects of this tissue. Cross-sectional images from the 10 s heartbeat movie enabled intuitive analysis of DD, SD, SI, HP, and DI for each group (Figure 8A). Compared with the WT group and the single-gene-knockout group, the asb5a^−/−:asb5b-ATGmo double-knockout group exhibited a significantly lower heart rate (Figure 8B) and markedly prolonged HP, SI, and DI (Figure 8C–E). The asb5a^−/− group and asb5b-ATGmo group, in contrast to the WT group, showed an upward trend in heart rate (Figure 8B), along with significantly lower HP and DI (Figure 8C,E), whereas SI showed no apparent change (Figure 8D). No significant differences in various indicators were observed between the asb5a^−/− group and the asb5b-ATGmo single-gene–knockdown group (See the Supplementary Material Excel Table S1 for details).

By capturing individual frames of the video (image sizes:1300 × 500), we observed a notable increase in the DD/SD of the atrium in the asb5a^−/−:asb5b-ATGmo double-knockout group compared with the other groups (Figure 9A–C). This suggests an evident trend toward atrial expansion. In contrast, the single-gene-knockout group showed no significant difference in DD/SD compared with the WT group (Figure 9A–C). No significant variations in any indicators were detected between the asb5a^−/− group and the asb5b-ATGmo single-gene-knockdown group. We calculated the relative value of cardiac fractional shortening using the following formula: FS = (DD − SD)/DD × 100%. FS represents the ejection volume at each contraction of the heart. Reduced FS indicates the inability of the heart to contract completely during systole, corresponding to systolic heart failure in the human adult heart. Statistical results revealed a significantly lower cardiac fractional shortening in the asb5a^−/−:asb5b-ATGmo double-knockout group compared with the other groups (Figure 9D) (See the Supplementary Material Excel Table S1 for details).

2.6. PPI Network Construction and Hub Gene Screening

In a more in-depth examination of RNA-seq data, using the KEGG pathway enrichment analysis, it was evident that the asb5a^−/−:asb5b-ATGmo double-knockout group exhibited pronounced enrichment of cardiac muscle contraction entries compared with other groups (Figure 7A,B; highlighted in purple). Under these entries, the asb5a^−/−:asb5b-ATGmo double-knockout group showed enrichment in 87 genes compared with the WT group, and 91 genes compared with the asb5a^−/− group. Initially, the STRING database was employed to analyze the network diagram of unified regulatory genes in these two sets of comparative experiments. Subsequently, using Cytoscape software (Version 3.9.1), the connectivity (degree) was organized from high to low, producing a precise PPI regulatory network diagram (Figure 10A,C). By sorting from high to low according to connectivity (degree), the top 20 hub genes were identified, and their expression patterns were visualized in a heatmap based on RNA-seq data (Figure 10B,D).

2.7. The qPCR Validation of 20 Abnormally Regulated Hub Genes

To corroborate whether the expression levels of the top 20 hub genes between the two groups aligned with the transcriptome data, qPCR experiments were conducted. The results revealed that, in comparison to the WT group, six genes (acta1a, actc1a, actc1c, cacng2a, cacna2d1a, and cox7c) in the asb5a^−/−:asb5b-ATGmo double-knockout group were significantly upregulated, and cacnb2a was significantly downregulated. In comparison to the asb5a^−/− group, five genes (acta1a, cox8b, uqcrc1, tnnc1a, and myl4) in the asb5a^−/−:asb5b-ATGmo double-knockout group were significantly upregulated. These results were consistent with the aforementioned transcriptome findings (Figure 11A,B). Through qPCR validation, overlapping genes were identified between groups, which were all differentially expressed in the asb5a^−/−:asb5b-ATGmo double-knockout group compared with the control group. Subsequently, the STRING database and Cytoscape software (Version 3.9.1) were employed to construct a novel PPI network diagram, indicating the unified regulation of these 11 DEGs (acta1a, actc1a, actc1c, cacng2a, cacna2d1a, cacnb2a, cox7c, cox8b, uqcrc1, tnnc1a,and myl4). The PPI network contained three modules with 11 nodes and 13 edges (Figure 11C).

3. Discussion

In this study, zebrafish served as the animal model, and we successfully established an asb5a^−/−:asb5b-ATGmo double-knockout line. This research elucidated, for the first time, the crucial role of the asb5 gene in early cardiac morphogenesis and function in zebrafish, suggesting potential functional redundancy between asb5a and asb5b.

Initially, we employed CRISPR/Cas9 technology for successful asb5a gene knockout in zebrafish. However, no apparent phenotype was observed in the homozygous mutant. We hypothesized that the lack of observable effects stemmed from the division of the asb5 gene into two subtypes, asb5a and asb5b, with the latter compensating for defects caused by asb5a homozygosity. Subsequently, using morpholino interference technology, we further knocked down the asb5b gene in the asb5a homozygous mutant. This revealed significant atrial amplification malformations in embryos with simultaneous knockout of these two genes. Through RNA-seq cluster analysis, we targeted cardiac muscle contraction-related functions. Semi-automatic heartbeat analysis software (Version 3.4.0.0) revealed disordered cardiac function indicators in the double-knockout strain, impacting early zebrafish cardiac contractility. Further analysis of the genes enriched in KEGG under the cardiac muscle contraction entry, conducted through the STRING database, Cytoscape software (Version 3.9.1), and qPCR experiments, led to the identification of 11 hub genes (acta1a, actc1a, actc1c, cacng2a, cacna2d1a, cacnb2a, cox7c, cox8b, uqcrc1, tnnc1a, and myl4). Their expression disorders may directly or indirectly regulate early cardiac contractile function in zebrafish.

Myosin, as the primary component of thick myofilaments, mediates force generation during muscle contraction by interacting with actin filaments [14]. The N-terminus of the cardiac myosin light chain can regulate cardiac contractile function by binding to the C-terminus of actin [15]. Additionally, troponin may interact with actin, myosin, Ca²⁺-binding proteins, and tropomyosin, inhibiting actomyosin ATPase and the movement of actin filaments on myosin in vitro [16]. The troponin complex serves as a molecular switch linking changes in intracellular calcium concentration to actin and myosin association and dissociation, allowing excitation–contraction coupling in striated muscle [17]. Troponin may act as a mediator between calcium ion channels and myosin/actin, effectively regulating cardiac muscle contraction. Cardiomyocyte contraction is triggered by the entry of extracellular calcium through specific calcium channels [18]. L-type calcium channels (LTCCs) are abundant in the myocardium. Voltage-gated LTCCs are located on the cardiomyocyte surface and contribute to cardiac action potentials by mediating calcium entry into the cell and triggering calcium-induced calcium release. Calcium voltage-gated channel auxiliary subunit β 2 (CACNB2) is a membrane-associated guanylate kinase protein, and the CACNB2 LTCC auxiliary subunit transports pore-forming CACNA subunits to the membrane and regulates channel dynamics [19]. Voltage-gated calcium ion channels consist of pore-forming α(1) and auxiliary α(2)δ, β, and γ subunits [20]. CACNG2 (gamma 2) may regulate calcium channel homeostasis by interacting with CACNA and CACNB. Finally, cytochrome c oxidase (COX), the rate-limiting enzyme of mitochondrial respiration, is regulated by various mechanisms [21]. COX is the terminal enzyme of the electron transport chain and catalyzes the transfer of electrons from cytochrome c to oxygen. Cox7c, Cox7b, ATP synthase, complex III subunit X, and myosin heavy chain 7 may be involved in cardiovascular metabolic pathways [22]. Cox7c and Cox8b potentially regulate oxygen consumption and ATP production. Ubiquitin cytochrome c reductase core protein 1 (UQCRC1) is an integral component of mitochondrial complex III, playing a key role in cardiac protection and maintaining mitochondrial function [23].

Through analysis of the preliminary PPI network shown in Figure 11C and a review of existing literature, we constructed a probable molecular regulatory network affecting early cardiac contractile function in zebrafish in the asb5a^−/−:asb5b-ATGmo double-knockout line (Figure 12). We speculate that it may impact early cardiac contractile function in zebrafish through three pathways: first, inactivation of asb5a and asb5b genes leads to the upregulation of myosin light chain 4 (myl4) [24], actc1a, actc1c [25] and acta1a [26]. These myl4-associated genes are significantly upregulated, ultimately leading to increased transcription of the cardiac troponin C regulatory gene tnnc1a [27], potentially influencing calcium ion channels. Second, knockout of asb5a and asb5b genes results in disorders in the mRNA levels of the calcium-channel-related genes cacng2a [28], cacnb2a [19], and cacna2d1a [29], impacting calcium channel homeostasis. Third, deletion of asb5a and asb5b genes leads to the upregulation of the COX family genes cox7c [30] and cox8b [31], followed by increased expression of uqcrc1 [23], potentially affecting calcium ion channel stability and leading to cardiac contractile dysfunction.

In summary, our study has preliminarily affirmed the crucial role of the asb5 gene in early heart development in zebrafish. Its deletion disrupts the expression of several early cardiac contraction-related genes, affecting the morphogenesis and functional integrity of the early heart, providing vital insights for further investigation into the molecular regulatory mechanisms of this gene in heart development, and expanding the range of novel potential targets for understanding early congenital heart disease.

4. Materials and Methods

4.1. Zebrafish Strain Rearing and Breeding

The AB zebrafish strain and the myl7:EGFP strain (myl7 is a heart-specific gene, which is especially expressed in the heart, and myl7:EGFP transgenic zebrafish is used as a cardiac-specific expression marker) were maintained at a constant temperature (28.5 °C) in a water circulation system. Zebrafish breeding involved placing two females and three males into hybrid bars, with overnight separation using clear plastic partitions. Partitions were removed around 09:00 the next day to facilitate mating, egg laying, and subsequent embryo collection.

4.2. CRISPR/Cas9-Mediated asb5a Gene Knockout

The CRISPR/Cas9 gene editing system was employed for zebrafish asb5a knockout. PCR was performed using the P42250 plasmid as the sgRNA template to synthesize the sgRNA transcript. The reverse primer was sgRNA-R (a universal R-terminal primer for the target site; Table S1), and the two forward primers containing the T7 promoter and target sequence were sgRNA-primer-F1 and sgRNA-primer-F2, respectively (Table S1). The sgRNA was synthesized via in vitro transcription using the Riboprobe^® System-T7 Transcription Kit (P1440, Promega, Madison, WI, USA). The sgRNAs (final concentration: 20 ng/µL each) were co-injected with Invitrogen TrueCut Cas9 v2 (A36499, Thermo Fisher Scientific, Waltham, MA, USA; final concentration: 300 ng/µL) into one-cell-stage fertilized eggs. The sgRNAs were designed to use CRISPR/Cas9 technology to introduce frameshift mutations into the coding sequence of asb5a through non-homologous end joining [6,32,33], thereby causing the asb5a frameshift mutation to produce a premature stop codon for protein translation. Surviving embryos with chimeras were mated with WT zebrafish, and offspring with partial deletion mutant alleles of asb5a were sequenced. Genotype identification for the asb5a-knockout lines was conducted using the forward and reverse primers asb5a-F and asb5a-R, respectively (Table S1).

4.3. RNA Probe Synthesis and WISH

For mRNA antisense probe preparation, RT-PCR was used to amplify a gene’s mRNA sequence using primers with T7 promoter sequences (asb5a and asb5b forward and reverse primers, as detailed in Table S3). Digoxigenin-labeled antisense RNA probes were synthesized through in vitro transcription using the Riboprobe^® System-T7 Transcription Kit (P1440, Promega, n vitro transcription using the Riboprobe^® System-T7 Transcription Kit (P1440, Promega, Madison, WI, USA) and ROCHE DIG RNA Labeling Mix (REF 11277073910, Roche, Basel, Switzerland). Zebrafish embryos at specific stages were fixed in 4% paraformaldehyde and stored in 100% methanol. WISH was used to detect asb5a and asb5b mRNA expression, and images were captured using a fluorescence microscope (Leica M205FA, Wetzlar, Germany) with Leica Application Kit imaging software (version 3.2.0).

4.4. Morpholino-Mediated Specific Knockdown of the asb5b Gene

A morpholino oligo sequence targeting asb5b was obtained using GeneTools (https://www.gene-tools.com, Philomath, OR, USA) to inhibit transcription (asb5b-ATGmo: 5′-TAACCACAATGTGTGCCTACTTACT-3′). Injection of a low asb5b-ATGmo concentration (0.25 mM) was performed at the one-cell stage in WT embryos. Four tubes of embryos from each group (WT and experimental groups) were collected (10 embryos per tube) as biological replicates, and RNA was extracted and reverse-transcribed into a cDNA library. Grayscale analysis performed via RT-PCR confirmed effective knockdown of asb5b mRNA levels at low concentrations; qPCR yielded the same results (Figure S2).

4.5. RNA-Seq-Based Transcriptome Analysis

Zebrafish embryos at 48 hpf were collected in batches of 50 embryos per tube, constituting a biological replicate; three tubes were randomly collected and stored at −80 °C. Each RNA-seq dataset was derived from a biological replicate of 50 embryos. Subsequent RNA-seq was conducted at Majorbio (Shanghai, China), with data analysis performed on the Majorbio Cloud platform. DEGs were analyzed using DESeq with screening conditions set at a significance threshold of |log2FC| > 2.0 and p < 0.05. RNA-seq results for the WT group, asb5a^−/− group (Group A in Supplementary Excel Table S2) and asb5a^−/−: asb5b-ATGmo double-knockout group(Group B in Supplementary Excel Table S2) are shown in Supplementary File S2.

The volcanic map of the DEGs was drawn using three software packages, DESeq2 (version 1.57.1) [34], DEGseq (version 1.56.0) [35], and edgeR (version 4.0.1) [36]. Genes/transcripts with similar expression patterns may have similar functions or participate in the same metabolic process or cellular pathway. Based on the expression level of genes/transcripts in different samples, the distances between genes/transcripts or samples were calculated, and then iterative methods were used to cluster genes or samples. GO enrichment analysis was performed using Goatools software (https://github.com/tanghaibao/GOatools, version 0.6.9, accessed on 25 July 2023) [12], and when corrected p values (FDR) < 0.05, they were considered to be significantly enriched, and sufficient for GO function. KEGG PATHWAY enrichment analysis was performed using R language. The principle of calculation was similar to that of enrichment analysis of GO function, with p < 0.05 as the cut-off criterion.

4.6. Imaging Technique and Image Analysis

Zebrafish embryos from the myl7:EGFP strain were imaged using a fluorescence microscope (Leica M205FA, Wetzlar, Germany). In situ hybridization images were captured with Leica Application Kit imaging software (version 3.2.0). Live embryos, incubated in Petri dishes with fresh water at 28.5 °C, were immobilized using 6% methylcellulose for imaging. An Axiocam (Zeiss Company, Jena, Germany) was used to photograph and analyze larval morphology at 24 and 48 hpf. The pericardial cavities and cardiac phenotypes of juvenile fish at these two stages were analyzed as previously described. Cardiac atria labeled with myl7:EGFP were visible against a green fluorescence background, revealing a propensity for severe atrium dilation in the double-knockout line. Images were processed using Zeiss Axio Vision 3.0.6 software and Adobe Photoshop. A high-speed EM-CCD camera captured zebrafish heart activity at 48 hpf for 10 s at 130 frames s⁻¹ under a 20× microscope. Semi-automatic heartbeat analysis software was then used to analyze the recorded heartbeat movie, generating indicators related to cardiac contraction: heart rate, HP, DI, SI, DD, SD, and fractional shortening (FS). All data were also reflected in the generated M-mode (See the supplementary material Excel Table S1 for details).

4.7. PPI Network Construction and Module Selection

PPI networks were constructed using STRING version 11.0b (https://string-db.org/; accessed on 2 June 2022) and Cytoscape (version 3.9.1). DEGs were sorted based on degree values, and the top 20 DEGs were designated as hub genes.

4.8. The qPCR detection

To detect gene expression, qPCR was employed. Each group comprised three biological replicates, with each replicate consisting of ten zebrafish embryos at 48 hpf and four technical replicates. Total RNA extraction involved the use of the Novozyme Universal RNA Extraction Kit (RC112-01, Vazyme, Düsseldorf, Germany), and cDNA was synthesized using the PrimeScript RT Kit (RR036A-1, Takara, Japan). Target gene expression was normalized using the internal control actb1 (also named β-actin, an internal reference gene, which has a relatively constant expression in various tissues and cells, and is often used as a reference when detecting gene expression level changes). The QuantStudio 5 Real-Time PCR system (Thermo Fisher Scientific, Waltham, MA, USA) and QuantStudio design and analysis software (version 1.5.2) were employed for qPCR. Primers for target genes, including those for acta1a, actc1a, actc1c, cacng2a, cacna2d1a, cacnb2a, cox7c, cox8b, uqcrc1, tnnc1a, and myl4, are detailed in Table S4. Relative gene expression values were calculated using the ΔΔCT method.

4.9. Statistical Analysis

The statistical significance of qPCR and heartbeat test data was assessed using an unpaired t-test with the significance level set at p < 0.05.