Neuron Navigator 3 (NAV3) is Required for Heart Development


 Background: As a tightly controlled biological process, cardiogenesis requires the specification and migration of a suite of cell types to form a particular three-dimensional configuration of the heart. Many genetic factors are involved in the formation and maturation of the heart, and any genetic mutations may result in severe cardiac failures. The neuron navigator (NAV) family consists of three vertebrate homologs (NAV1, NAV2, and NAV3) of the neural guidance molecule Uncoordinated-53 (UNC-53) in Caenorhabditis elegans. Although they are recognized as neural regulators, their expressions are also detected in many organs, including the heart, kidney, and liver. However, the functions of NAVs, regardless of neural guidance, remain largely unexplored. In this study, we aim to investigate the role of NAV3 in heart development.Results: The nav3 gene was found to be expressed in the cardiac region of zebrafish embryos from 24 to 48 hours post-fertilization (hpf) by means of in situ hybridization (ISH) assay. A CRISPR/Cas9-based genome editing method was utilized to delete the nav3 gene in zebrafish and loss-of-function of Nav3 resulted in a severe deficiency in its cardiac morphology and structure. The similar phenotypic defects of the knockout mutants could recur by nav3 morpholino injection and be rescued by nav3 mRNA injection. Dual-color fluorescence imaging of ventricle and atrium markers further confirmed the disruption of the heart development in nav3-deleted mutants. Although the heart rate was not affected by the deletion of nav3, the heartbeat intensity was decreased in the mutants. Conclusions: The nav3 gene can be expressed in a variety of tissues or organs although it is initially identified as a neural guiding factor. Gene knockout of nav3 in zebrafish leads to severe development malformation of the heart. Moreover, the defective heart development can be rescued by nav3 mRNA injection. All these findings indicate that Nav3 was required for cardiogenesis in the development of zebrafish embryos.


Background
Cardiogenesis is an indispensable process in vertebrates during embryogenesis, and the heart plays a vital role in the distribution of nutrients and oxygen in the embryos. Any failures during cardiogenesis will lead to cardiac malformations, further resulting in the death of the embryos. It is challenging to study heart development during embryogenesis. Compared with other mammalian models, such as mice, zebra sh offers several unique advantages for studying cardiac development (1,2). First, the zebra sh embryos develop externally after fertilization, allowing direct non-invasive observation of heart development during cardiogenesis. Second, the optical transparency of zebra sh embryos allows high-resolution optical imaging. Third, zebra sh embryos can survive up to 7 days without a functional cardiovascular system to supply the nutrients and oxygen due to their tiny size, offering the possibility to analyze the phenotype of the cardiac defects. Besides, essential genes responsible for cardiac development are also highly conserved throughout vertebrates. Therefore, zebra sh is a valuable animal model for exploring the mechanisms underlying cardiac development.
In zebra sh, the formation of a mature heart includes a series of cellular processes, including cell migration, proliferation, and differentiation (3,4). Numerous genes are involved in embryonic cardiogenesis, such as bmp4 (5), nppa (6), vcana (7), amhc, and vmhc (8,9), as well as transcription factors nkx2.5 (10), hand2 (11), and GATA family members (12)(13)(14). Nav3, the vertebrate homology of C. elegans gene uncoordinated-53 (unc-53), is one of the neural guidance genes, which plays a role in shaping the developing nervous system by modulating/guiding neural cell migration (15,16). Moreover, NAV3 is identi ed as a microtubule-binding protein that regulates actin assembly in extensions of lamellipodia and lopodia (17). Although nav genes are mainly expressed in the brain, their slight expressions are also detected in the heart, kidney, and liver, suggesting potential roles of NAVs in organogenesis (15). In the present study, we focused on the effects of Nav3 on heart development in zebra sh. We found that nav3 was temporally expressed in the cardiac region from 24 h postfertilization (hpf) to 48 hpf through in situ hybridization (ISH) assay, indicating its function in ventricle-atrium differentiation (18).
Deletion of nav3 in zebra sh via CRISPR/Cas9 genome editing technique resulted in severe defects in heart development. Fluorescent imaging of ventricle marker (vmhc) and atrium marker (amhc) further con rmed the malformed morphogenesis of the cardiac structure in nav3-null mutants. In addition to structural alterations, a signi cant decrease in heartbeat intensity (HI) was also detected in nav3 −/− mutants. Moreover, the phenotypic defects caused by lossof-function of Nav3 could be rescued by nav3 mRNA injection. Collectively, our current work revealed a novel role of Nav3 in heart development, which might provide potential strategies for heart disease therapy and cardiac regeneration.
Whole-mount in situ hybridization (WISH) WISH with antisense RNA probes was performed as previously described (24). The probe for the detection of NAV3 (NM_001045143.2) was cloned from cDNA fragments. The nav3 cDNA fragment was ampli ed with a forward primer: 5′-TCACCTTCTGACTCCACCAG-3′ and a reverse primer: 5′-GTTCGATGTTACGCGCTCAC-3′. After hybridization, images of the embryos were acquired with an Olympus stereomicroscope MVX10 equipped with an Olympus DP71 camera.
Generation of nav3-null mutants A CRISPR/Cas9-mediated approach was used to generate nav3-deleted mutants. The target site of CRISPR/Cas9 designed to identify the sequence in the second exon of nav3 was 5′-CATACCCGATGGAGTTCTGCTGG-3′ (underlined was the PAM sequence). The template sequence of sgRNA for in vitro transcription was 5′-TAATACGACTCACTATAGCATACCCGATGGAGTTCTGCGTTTTAGAGCTAGAAATAGCAAGTTAAAATAAGGCTAGTCCGTTATCAACTTGAAAAAGTGGCACCGAGTCGG 3′. For the sgRNA synthesis, a forward primer (5′-TAATACGACTCACTATAGCATACCCGATGGAGTTCTGCGTTTTAGAGCTAGAAATAGC-3′) containing a T7 promoter region and a nav3 gene-targeting region, and a universal reverse primer (5′-AAAAAAAGCACCGACTCGGTGCCAC-3′) were used in the PCR ampli cation with pT7 plasmid as the template to obtain the sgDNA, which was then transcribed into sgRNA in vitro using the T7 mMessage mMachine kit (Ambion). Micro-injection was performed with 1-cell stage zebra sh embryos, and each embryo was co-injected with 100 pg sgRNA and 200 pg Cas9 mRNA. G0 generations were examined by PCR, followed by Sanger sequencing. A mutant with a 55-bp deletion at the second exon was identi ed, leading to the occurrence of a stop codon downstream of the Cas9 cutting site. The heterozygous nav3 +/mutants were incrossed to obtain the homozygous F2 progenies.

Morpholino-mediated gene deletion of nav3 in zebra sh
The nav3 gene-speci c morpholino (Gene Tools, LLC) was used to block the splicing of nav3 pre-mRNA. The sequence of the morpholino antisense oligomer was 5′-CAGCCCATGTGCCATGCTGCTTCTG-3′. In the present study, 2 nL of 0.3 mM morpholino oligo was micro-injected into the embryos at the 1-cell stage.

mRNA rescue experiment
In the rescue experiment, 200 pg nav3 mRNA was co-injected into the embryos with the nav3 antisense morpholino. The cDNA fragment of nav3 was rst subcloned into pCS2+ vector. To make the mRNA, the DNA plasmid was linearized with an appropriate restriction enzyme and transcribed in vitro using the SP6 mMessage mMachine kit (Ambion).

Survival rate quanti cation
Zebra sh embryos of WT and nav3-null mutants were respectively collected after fertilization and divided into 10 tanks (100 embryos in each tank). The counts of surviving shes per tank were determined at 1, 2, 3, 20, and 60 days post-fertilization (dpf). Afterward, the proportions of sh surviving at each time point were assessed.
The heart rate of the embryo was de ned as the beats per minute and counted under a dissecting microscope (Olympus). To examine the HI, the cardiaclabeled transgenic zebra sh lines Tg(mef2a:EGFP) and Tg(nav3 -/ --mef2a:EGFP) were used to visualize the hearts of WT and mutants, respectively. The crosssectional areas of the chambers in GFP-labeled hearts in the dilated state (A d ) and contracted state (A c ) were respectively measured with ImageJ (25). The HI was calculated by the equation as follows, HI = A d -A c . In our present analysis, ve WT and mutant shes were respectively selected, and each sh was measured for 20 successive heartbeats.

Data analysis
All data were analyzed using GraphPad Prism 8.0. A two-tailed, unpaired Student's t-test was used to determine statistical signi cance when comparing independent groups. P<0.05 was considered statistically signi cant. All results were presented as mean ± SD.

Results
Zebra sh nav3 is expressed in the heart during embryogenesis Zebra sh Nav3 has been previously reported as an ortholog of UNC-53 in C. elegans. Compared with other vertebrate NAV3 in humans, mice, and rats, as well as UNC-53 in C. elegans, zebra sh Nav3 was highly conserved among all these species (Supplemental Fig. S1). NAV3 is a huge protein consisting of more than 2,000 amino acids. It contains several conserved domains, including a putative calponin-homology domain (CH-domain), an LKK actin-binding domain, two Src Homology 3 (SH3) domains, and an ATP/GTP-binding AAA domain (19) (Fig. 1A). The CH-domain is essential for the interaction with actin laments, suggesting its role in the formation of lamellipodia and lopodia, and it would further drive cell movements. Phylogenic alignment also revealed the close relationship between zebra sh Nav3 and NAV3 of other species (Fig. 1B).
WISH was performed with zebra sh embryos from 18 hpf to 96 hpf to explore the spatiotemporal expression of zebra sh nav3 during embryogenesis. The nav3 mRNA transcripts could be detected at an early embryonic stage (18 hpf), while its expression was mainly restricted to the brain and somites (Fig. 1C). A slight expression was also presented in cardiac primordium at the same developmental stage (Fig. 1D, E). As nkx2.5 is speci cally expressed in cardiac primordium, the ISH experiment with nkx2.5 probe could further con rm the expression of nav3 in the region of cardiac primordium (Fig. 1F). At 24 hpf, nav3 was more speci cally expressed in the brain, heart, and somites (Fig. 1G-I). However, the expression of nav3 in somites became weaker, and it was rarely detected from 48 hpf. At 48 hpf, its expression was highly regionalized to the brain and heart (Fig. 1J-L). At 96 hpf, there was almost no expression of nav3 in the heart, and its expression was found to be accumulated in other tissues, such as the gill arch, swim bladder, and intestine ( Fig. 1M-O), suggesting that its potential roles were not only restricted to nervous or cardiovascular system. Nav3 loss-of-function in zebra sh exhibits severe phenotypic defects and low survival rates To explore the effects of Nav3 deletion in heart development, a CRISPR/Cas9-based genome editing method was utilized to generate a nav3-deleted mutant ( Fig. 2A). A 55-bp deletion at the second exon of nav3 led to the occurrence of a truncated protein (a 79-amino acid sequence versus the full-length sequence with 2,270 amino acids). The homozygous nav3 −/− mutants were further obtained through genetic selection. An obvious heart malformation phenotype was observed in nav3-null zebra sh embryos. Compared with the wild-type (WT) counterparts, the nav3 −/− mutants displayed severe pericardial edemas not only at the early developmental stage but also in adult zebra shes, although more than 70% of the nav3 −/− mutants died within 24 hpf (Fig. 2B-H). Afterward, only 8% of the mutants could successfully develop into adults (Fig. 2H).
Heart morphogenesis and function are disrupted in nav3 mutant zebra sh embryos To further evaluate the changes in cardiac morphology, structure, and function in nav3-null mutants, the uorescently labeled atrium-speci c marker amhc and ventricle-speci c marker vmhc were respectively adopted. Transgenic lines Tg(vmhc:mCherry::amhc:EGFP) and Tg(nav3 − / − -vmhc:mCherry::amhc:EGFP) were used to compare the heart morphological differences between WT and mutants. Both atrium and ventricle in nav3 −/− mutants at 72 hpf exhibited a long tubular-shaped morphology, and the boundary line between atrium and ventricle was not as clear as the WT (Fig. 3A-F). Next, we compared the heart rate as well as the HI between WT and nav3 −/− mutants to check whether the heart functions were affected upon Nav3 loss-of-function. We found that the heart rate was not affected by the deletion of nav3. The frequency was within a range of 150 to 200 times per minute in both WT and mutant strains (Fig. 3S). However, the HI in WT and mutant strains was different. Here, we introduced a parameter de ned as HI to indicate cardiac capacity. The areas of the GFP-labeled heart were measured in the dilated and contracted states (Fig. 3G-R). The area difference between these two states was regarded as HI. The calculation formula was shown in Methods. The HI was signi cantly lower in the mutants compared with the WT siblings (Fig. 3T). However, there was no obvious difference in heart rate between WT and mutants (Fig. 3U). All these ndings indicated that Nav3 de ciency resulted in heart development defects during zebra sh embryogenesis, including the cardiac structure and functional disruption.
nav3 mRNA injection can rescue the cardiac defects in nav3-deleted embryos To further con rm whether the abnormal phenotype of the mutant was caused by the deletion of nav3, we injected a morpholino against nav3 mRNA into transgenic line Tg(myl7:mCherry) to delete nav3. Tg(myl7:mCherry) line was under a WT background, and the cardiac marker myl7 (myosin light chain 7) was uorescently labeled with mCherry uorescent protein. Upon nav3 expression was down-regulated through morpholino injection, WT zebra sh embryos displayed a similar phenotype to nav3 −/− mutants. Moreover, uorescent imaging of the heart in morpholino-injected embryos mimicked the abnormal cardiac morphology of nav3-null mutants (Fig. 4).
To verify that the phenotype of nav3-morphants was attributed to the loss-of-function of Nav3 rather than non-speci c effects, we performed a rescue experiment by injecting nav3 mRNA together with nav3 morpholino into 1-cell-stage zebra sh embryos of Tg(myl7:mCherry). The nav3 mRNA injection partially rescued the phenotypic defects caused by the down-regulation of nav3. Upon nav3 mRNA injection, the tubular-shaped ventricle and atrium structure of the nav3-morphants were almost recovered to a normal state. However, compared with the WT counterparts, the rescued heart was slightly misshaped in the morphology, such as tubular-shaped ventricle. Additionally, the boundary between the ventricle and atrium became observable after nav3 mRNA injection (Fig. 4).

Discussion
As a complicated but coordinated biological process, the assembly of an intact heart involves the proliferation, differentiation, and migration of cardiomyocytes to form a functional contractile organ (2)(3)(4). A cohort of genes participate in this process and play important roles. Nav3, the vertebrate homology of unc-53 in C.elegans, is rst discovered as a neural guidance gene that is functional in cell migration and outgrowth of axons. Because cell movement is highly relevant to embryonic organogenesis and NAV3 is also involved in liver development (17), we hypothesized that NAV3 played a novel role in cardiogenesis. In our present study, we rst examined the expression of nav3 in zebra sh during embryogenesis. The speci c spatiotemporal expression of nav3 in the zebra sh heart region suggested its potential function in embryonic heart development beyond the nervous system. To con rm our prediction, we generated the nav3-deleted mutants. Loss-of-function of Nav3 in zebra sh embryos displayed severe heart development defects, such as pericardial edemas and chamber malformations. Moreover, the function of the heart was affected by the deletion of Nav3. For instance, the HI was signi cantly decreased in mutants. However, the heart rates were similar in both WT and mutant embryos. The functional analysis revealed that loss of Nav3 resulted in an abnormal dilation and contraction in the heart.
To con rm whether the phenotypic defects of the nav3-null mutants were caused by the loss-of-function of Nav3, a morpholino-mediated gene knockdown experiment and a nav3 mRNA rescue experiment were performed. The pericardial edemas and heart morphogenesis malformations were reproduced in nav3deleted mutants and could also be partially rescued by co-injection of morpholino and nav3 mRNA. Additionally, the mRNA rescue experiment also recapitulated that the severe cardiac edema was not caused by the off-target effects triggered by sgRNA or morpholino injection.
In our present work, we also noticed that the survival rates of nav3-deleted mutants were signi cantly lower compared with their WT counterparts, especially from the very early developmental stage (from 24 hpf). Since zebra sh could survive up to 7 days without a functional cardiovascular system, the high mortality upon Nav3 deletion might not result from the cardiac defects and more probably be attributed to its dysfunction in the nervous system or immunity (20, 21).
As mentioned above, Nav3 loss-of-function could lead to severe defects in zebra sh heart development. However, it remains largely unknown whether some essential cardiac-related genes are altered in nav3-null mutants. In our preliminary experiments, we examined the expressions of nkx2.5, hand2, vmhc, and amhc in nav3 −/− mutants and WT zebra sh embryos using ISH. We found that the expressions of nkx2.5 and hand2 were decreased after the deletion of nav3 (Supplemental Fig. S2). Although they both play roles in embryonic heart development, each of them also owns a distinct and speci c function in regulating this process. Nkx2.5 determines the cardiomyocyte identity and can further maintain the vascular and atrial chamber morphology (10). Hand2 is implicated as a regulator of cardiomyocyte production, which promotes the generation of a proper number of cardiomyocytes during cardiogenesis (11). As key regulators in heart development, the lower expressions of nkx2.5 and hand2 in nav3 −/− mutants supported its de ciency in cardiac development. This nding was consistent with the previous view that nkx2.5 maintains cardiac chamber identity, and hand2 regulates cardiac differentiation and morphogenesis (10,11). Although the structure and morphology of the atrium and ventricle were severely affected by the loss of nav3, the expressions of both their markers remained unchanged (Supplemental Fig. S2). We assumed that the cardiac-related myosin chain genes were not affected by the deletion of Nav3, and the heart morphogenesis defects might result from other factors, such as nkx2.5, or hand2.
Although nav3 is initially discovered as a neural guidance gene in the nervous system, the NAV family members have also been widely accepted as regulators in cell migration among several species (15-17, 19, 22). Moreover, it has been reported to act as a positive modulator in regulating the actin assembly in the extension of lopodia and lamellipodia during zebra sh liver development (17). As a complicated but orchestrated morphogenetic process, heart formation involves speci cation, differentiation, and migration of cardiac progenitor cells. Therefore, NAV3 might also play a role in heart development by facilitating the migration of cardiac progenitor cells or cardiomyocytes. Besides, crosstalk between NAV3 and other essential factors, such as nkx2.5, might be involved in these crucial processes during cardiac development. Further efforts are needed to decipher the mechanism of NAV3-regulated cardiogenesis. Taken together, NAV3 was a potential regulator in cardiac development during embryogenesis.

Conclusions:
The nav3 gene can be expressed in a variety of tissues or organs although it is initially identi ed as a neural guiding factor. Gene knockout of nav3 in zebra sh leads to severe development malformation of the heart. Moreover, the defective heart development can be rescued by nav3 mRNA injection. All these ndings indicate that Nav3 was required for cardiogenesis in the development of zebra sh embryos. All animal-related experiments were carried out following the NIH Guidelines for the care and use of laboratory animals (http://oacu.od.nih.gov/regs/index.htm), and animal protocols were ethically approved by the Administration Committee of Experimental Animals of Nantong University, Jiangsu Province, China (Approval ID: 20180608-Z001). Transgenic zebra sh lines provided by others were approved by the owners with written informed consent. Our study complied with the rules of the Guidelines for the care and use of laboratory animals (https://www.biomedcentral.com/getpublished/editorial-policies#standards+of+reporting). The study was carried out in compliance with the ARRIVE guidelines.

Consent for publication
Not applicable.
Availability of data and materials All data generated or analysed during this study are included in this published article and its supplementary information les.

Figure 4
Nav3 mRNA injection rescues the cardiac defects caused by nav3 deletion. Transgenic zebra sh line Tg(myl7:mCherry) embryos injected with nav3 morpholino was compared with the control. Co-injection of nav3 morpholino and nav3 mRNA partially rescued the defects in cardiac chamber morphogenesis.
The ventricle and atrium are outlined with red and blue dotted lines, respectively.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.