Synergistic and independent roles for Nodal and FGF in zebrafish CPC migration and asymmetric heart morphogenesis

20 Asymmetric vertebrate heart development is driven by an intricate sequence of morphogenetic 21 cell movements, the coordination of which requires precise interpretation of signaling cues by heart 22 primordia. Here we show that Nodal functions cooperatively with FGF during heart tube formation and 23 asymmetric placement. Both pathways act as migratory stimuli for cardiac progenitor cells (CPCs), but 24 FGF is dispensable for directing heart tube asymmetry, which is governed by Nodal. We further find 25 that Nodal controls CPC migration by inducing left-right asymmetries in the formation of actin-based 26 protrusions in CPCs. Additionally, we define a developmental window in which FGF signals are 27 required for proper heart looping and show cooperativity between FGF and Nodal in this process. We 28 present evidence FGF may promote heart looping through addition of the secondary heart field. Finally, 29 we demonstrate that loss of FGF signaling affects proper development of the atrioventricular canal 30 (AVC), which likely contributes to abnormal chamber morphologies in FGF-deficient hearts. Together, 31 our data shed insight into how the spatiotemporal dynamics of signaling cues regulate the cellular 32 behaviors underlying organ morphogenesis. 33 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint

34 Vertebrate heart development requires the faithful execution of intricate and successive cell 35 movements. The coupling of these movements to morphogenesis is mediated by the precise 36 interpretation of signaling cues by heart primordia. Perturbations in these signaling pathways can 37 underlie cellular processes that go awry in the roughly 40,000 infants born with a congenital heart 38 defect (CHD) each year in the United States (Tsao et al., 2022). Therefore, a comprehensive analysis 39 of signaling cues that direct cardiac cell behavior is critical for understanding the pathogenic 40 mechanisms of CHDs. 41 The presence of conserved molecular mechanisms, the amenability to genetic manipulation and 42 live imaging, and the ability for continued embryonic development in the presence of heart defects 43 make zebrafish particularly advantageous for studying the cellular behaviors underlying cardiac 44 morphogenesis (Bakkers, 2011; Genge et al., 2016; Grant et al., 2017; Nguyen et al., 2008; Smith and 45 Uribe, 2021). During the initial stages of zebrafish heart development, cardiac progenitor cells (CPCs) 46 are organized bilaterally within the anterior lateral plate mesoderm (LPM) as epithelial sheets, with 47 ventricular precursors positioned medial to atrial cells. These sheets converge on the embryonic 48 midline, where they form the cardiac cone, a disc-shaped structure comprised of atrial cells at its base 49 and ventricular cells at its apex (19 hours post fertilization, hpf; Bakkers, 2011; Grant et al., 2017; Smith 50 and Uribe, 2021). 51 During a process known as cardiac jogging, CPCs migrate in a leftward and anterior direction, 52 with cells on the left half of the cardiac cone migrating more rapidly than the right-sided cells (20-24 53 hpf). This asymmetry in migration velocities results in clockwise rotation of the cardiac cone and 54 displacement of atrial cells to the left and anterior of ventricular cells as the heart tube elongates. 55 Cardiac jogging culminates in the asymmetric placement of the heart tube to the left of the embryonic 56 midline, establishing the first asymmetry in the heart (Baker et al., 2008; de Campos-Baptista et al., 57 2008; Kidokoro et al., 2022; Lenhart et al., 2013; Rohr et al., 2008; Smith et al., 2008; Veerkamp et al., 58 2013). Between 30-48 hpf, the heart undergoes an evolutionarily conserved process known as cardiac 59 looping. During this process, the heart tube bends rightward to begin defining and aligning the atrial and 60 ventricular chambers of the heart, with actin polymerization being critical for dictating looping chirality 61 (Desgrange et al., 2018; Noel et al., 2013; Smith and Uribe, 2021). Along with cardiac looping, the 62 cardiac chambers undergo expansion and become morphologically discernible from one another via 63 cardiac ballooning. Regionalized differences in cell morphology bring about chamber formation, and 64 blood flow and contractility are critical for regulating these morphologies (Auman et al., 2007; Dietrich et 65 al., 2014; Smith and Uribe, 2021). 66 Left-sided Nodal signaling in the LPM governs asymmetric behaviors during jogging and dextral 67 looping of the heart. Nodal directs the asymmetric migration of CPCs during cardiac jogging by 68 increasing the velocities of left-sided cells in the cardiac cone and properly positioning the heart to the 69 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint left of the embryonic midline (Baker et al., 2008; de Campos-Baptista et al., 2008; Kidokoro et al., 2022; 70 Lenhart et al., 2013; Rohr et al., 2008; Veerkamp et al., 2013). While looping of the jogged heart tube 71 involves intrinsic mechanisms (Noel et al., 2013), this process is influenced by earlier Nodal signaling to 72 produce robustness in dextral looping morphogenesis (Grimes et al., 2020). 73 Although Nodal signaling is the dominant signal for sidedness, it is well-established that Nodal 74 signals integrate with other signaling pathways to orchestrate events controlling organogenesis (Hill, 75 2016; Massague, 2003). For example, during cardiac cone rotation and subsequent jogging, Nodal and 76 BMP signals act synergistically to influence cell migration (Lenhart et al., 2013; Veerkamp et al., 2013). 77 How other signaling pathways potentially cooperate with Nodal signaling to direct asymmetric heart 78 morphogenesis remains uncharacterized. 79 Here, we characterize the cooperative and independent functions of Nodal and FGF signaling 80 during asymmetric morphogenesis of the zebrafish heart. We find that FGF signaling is required during 81 cardiac jogging as a migratory stimulus for CPCs but is dispensable for determining heart tube 82 laterality. We use confocal live cell imaging to analyze metrics of cell migration and find that aberrant 83 FGF and Nodal signaling alters the velocity dynamics of CPC migration during cone rotation. We also 84 demonstrate that Nodal signals generate sidedness by directing left-right asymmetric protrusive F-actin 85 activity in CPCs. We reveal roles for FGF signaling in promoting proper chamber placement during 86 cardiac looping and show FGF and Nodal signaling cooperate in this process. We provide evidence 87 that the role of FGF in cardiac looping may be through promoting secondary heart field (SHF) addition 88 to the ventricle, suggesting SHF promotes proper looping in vivo. Finally, we provide evidence that FGF 89 signaling prior to looping is required for proper development of the atrioventricular canal (AVC). 90 Together, these findings shed insight into how Nodal and FGF signals cooperate to ensure proper 91 migration and morphogenesis in the heart at multiple stages. 92 93

94 FGF and Nodal signals function synergistically during heart tube formation 95 Previous studies have revealed that the left-sided activation of Nodal within the anterior LPM 96 produces directional migration of CPCs and thus determines the direction of heart tube displacement 97 from the embryonic midline during cardiac jogging (Baker et al., 2008; de Campos-Baptista et al., 2008; 98 Kidokoro et al., 2022; Lenhart et al., 2013; Rohr et al., 2008; Veerkamp et al., 2013). To elucidate 99 mechanisms by which Nodal induces left-right asymmetries in CPC migration rates, we mined our 100 previous microarray studies identifying cardiac-specific transcriptional targets of Nodal. We found a 101 number of components of the FGF signaling pathway as being upregulated by Nodal signals, including 102 FGF ligands, transcriptional targets, and three of four tyrosine kinase receptors (FGFRs; Williams, 103 2015). Interestingly, various genes involved in FGF signaling have been previously identified as Nodal 104 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint targets in zebrafish, including fgf17b, fgf3, and fgf8 (Bennett et al., 2007), and fgf8 is expressed in the 105 cardiac cone prior to cardiac jogging (Fig. 1A; Reifers et al., 1998; Reifers et al., 2000). 106 To explore the function of FGF signals during heart development, we administered the pan-107 FGFR inhibitor SU5402 (Mohammadi et al., 1997). Because inhibition of FGF signaling can affect 108 cardiac precursor development (Felker et al., 2018; Marques et al., 2008; Reifers et al., 2000), the use 109 of SU5402 enables precise temporal control over the inhibition of FGF signaling after CPC 110 specification. The addition of SU5402 at 24 hpf resulted in various cardiac defects, including severe 111 pericardial edema and aberrant chamber placement, implicating a role for FGF in asymmetric cardiac 112 morphogenesis (Fig. 1B). 113 To confirm that treatment with SU5402 did not affect the expression of left-right patterning 114 genes, we analyzed the effect of inhibiting FGF signaling at 18.5 hpf on expression of the Nodal target 115 gene lefty2 in the cardiac cone. lefty2 expression serves as both a readout of Nodal responsiveness 116 and a direct readout of the sidedness of Nodal in the embryo (Fig. 1C). In WT embryos, lefty2 117 expression was restricted to the left in the cardiac cone, as expected (Fig. 1D). spaw mutants lacked 118 lefty2 expression (Grimes et al., 2020; Noel et al., 2013), while the majority of ntl mutants expressed 119 lefty2 bilaterally (Amack and Yost, 2004), as expected (Fig. 1D). Loss of FGF signaling in SU5402-120 treated WT embryos did not alter the left-sidedness of Nodal target lefty2, nor did loss of FGF change 121 the expression profile of lefty2 in spaw and ntl mutants (Fig. 1D), suggesting that any effect of FGF on 122 asymmetric chamber placement was downstream of left-right axis establishment. 123 To explore the function of FGF signals during cardiac jogging, the first morphological break in 124 symmetry in the embryo, we administered SU5402 at 18.5 hpf, just before the formation of the cardiac 125 cone, and continued treatment until assessment at 24 hpf (Fig. 2A), a time at which most wild-type 126 (WT) embryos have completed extension of the cardiac cone into a linear heart tube (Fig. 2B). As 127 expected, most WT embryos had completed migration and jogging (Fig. 2C). no tail (ntl/ta) mutants, 128 which have bilateral Nodal expression in the LPM (Amack and Yost, 2004), exhibited only a slight delay 129 in cone-to-tube transition, with most embryos having completed jogging at this time point (Fig. 2C). By 130 contrast, southpaw (spaw) mutants, which lack Nodal expression in the LPM (Grimes et al., 2020; Noel 131 et al., 2013), displayed a significant delay in tube formation with the majority of hearts remaining at 132 cone stage or starting to extend (Fig. 2C). SU5402-treated WT embryos also experienced a significant 133 delay in tube formation, suggesting a role for FGF signaling in promoting CPC motility in this process 134 (Fig. 2C). Interestingly, treating ntl and spaw mutants with SU5402 significantly increased delays in 135 tube formation, with most hearts remaining in the cardiac cone stage (Fig. 2C). The strongest effect 136 was observed in spaw mutants treated with SU5402, in which all hearts of embryos analyzed remained 137 at the cardiac cone stage at 24 hpf (Fig. 2C). While heart tube formation was delayed at 24 hpf in 138 SU5402-treated and mutant embryos, heart tubes eventually formed in all groups after several 139 additional hours (Fig. 2D,F). Heart tubes were significantly shorter in SU5402-treated embryos, likely 140 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint due to inhibition of SHF addition as previously reported (Fig. 2D; de Pater et al., 2009; Felker et al., 141 2018; Lazic and Scott, 2011). Heart tubes were also shorter in ntl and spaw mutants, suggesting that 142 elongation is impacted in these mutants or that elongating in a midline position may be inhibited by 143 surrounding tissues (Fig. 2D). 144 To determine if inhibiting FGF signaling causes defects in cardiac laterality, we examined the 145 direction of heart jogging in embryos with extended heart tubes at 26.5 hpf (Fig. 2E). In WT embryos, 146 heart tubes were displaced to the left of the embryonic midline, producing the expected “left jog” (Fig. 147 2F). spaw mutants and ntl mutants exhibited primarily “midline jog”, in which the heart tube fails to be 148 displaced to either side of the midline, as expected (Fig. 2F). Though inhibiting FGF signaling affected 149 the dynamics of heart tube formation, hearts consistently jogged correctly to the left (Fig. 2F), 150 suggesting that FGF signaling is dispensable for instructing cardiac laterality at this point. Overall, this 151 suggests that Nodal and FGF signaling cooperate to promote cell migration during jogging, while only 152 Nodal signaling is required to induce lateralized migration during this process. 153 154 Nodal and FGF signals regulate the velocity dynamics and directionality of CPCs during heart 155 tube formation 156 Perturbation of the Nodal and FGF pathways resulted in defects in heart tube morphogenesis 157 that we hypothesized were due to defects in cell migration. To explore this possibility, we conducted 158 confocal live imaging of embryos that express EGFP under the cardiac-specific myl7 promoter (Huang 159 et al., 2003) to quantify metrics of cell migration (Fig. 3A). Starting from the formation of the cardiac 160 cone, cell movements were analyzed for three hours in each condition. WT embryos completed jogging 161 within the three-hour time frame and formed a left-lateralized heart tube (Fig. 3B; Movie S1). ntl 162 mutants completed jogging in the same time frame but produce midline jogged heart tubes due to 163 bilateral Nodal expression (Fig. 3B; Movie S2). spaw mutants also display midline jogging; however, 164 these CPCs migrated more slowly compared to WT or ntl mutants, often resulting in incomplete tube 165 extension during the three-hour time frame (Fig. 3B; Movie S3). Similarly, CPCs in SU5402-treated WT 166 embryos migrated more slowly compared to WT or ntl mutants, resulting in incomplete tube extension 167 during the imaging time frame (Fig. 3B; Movie S4). Intriguingly, CPCs in SU5402-treated spaw mutants 168 scarcely migrated during the same time frame and exhibited a severe delay in heart tube formation 169 (Fig. 3B; Movie S5). These results indicate that FGF signaling acts as a CPC migration stimulus during 170 jogging and acts cooperatively with Nodal in this process. 171 To better characterize asymmetric migration during jogging, we quantified the trajectories of 172 CPCs by tracking them throughout the duration of the time-lapse movies (Fig. 3C). As the outermost 173 atrial cells of the cardiac cone display the largest disparity in left-right migration rates, we analyzed the 174 movements of this population; the brighter, inner ventricular cells were not included in our analysis. 175 Analysis of the direction and magnitude of CPC trajectories revealed that CPCs in WT embryos 176 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint exhibited a marked anterior and leftward displacement, while CPCs in ntl and spaw mutants failed to 177 undergo strong left or right displacement (Fig. 3D). Furthermore, in ntl and spaw mutants, overall 178 anterior displacement of CPCs was decreased (Fig. 3D). SU5402-treated embryos displayed a leftward 179 bias, though the overall displacement was not as marked as in WT embryos, and anterior migration 180 was reduced in these embryos (Fig. 3D). Lateral and anterior migrations were most severely disrupted 181 in SU5402-treated spaw mutants, further confirming that Nodal and FGF signals work cooperatively to 182 promote CPC migration during jogging (Fig. 3D). The CPC trajectories further suggest that Nodal and 183 FGF signals contribute to the general anterior migration of the heart during jogging, while asymmetric 184 Nodal signals promote lateral migration. 185 To better understand how Nodal and FGF signals contribute to cardiac laterality, we analyzed 186 the migration velocities of CPCs. Cardiac laterality is known to be directed by left-right biases in cell 187 migration velocity driven by Nodal signaling, with left CPCs displaying a higher velocity that right CPCs 188 in WT embryos (de Campos-Baptista et al., 2008; Kidokoro et al., 2022; Lenhart et al., 2013; Veerkamp 189 et al., 2013). We found that only in WT embryos did strong asymmetric migration velocities exist, with 190 left-sided cardiac cone cells exhibiting significantly higher velocities than right-sided cells (Fig. 3E). 191 Given the bilateral exposure to Nodal of CPCs in ntl mutants, we expected migration to be increased on 192 both sides of the cardiac cone as we observed in ntl morphants (Lenhart K. F. et al., 2013). However, 193 we found that ntl mutants did not display increased migration velocities compared to WT; nevertheless, 194 the expected symmetry in migration velocities was observed (Fig. 3E). Consistent with previous 195 findings (de Campos-Baptista et al., 2008; Kidokoro et al., 2022; Lenhart et al., 2013; Veerkamp et al., 196 2013), migration velocities of CPCs in spaw mutants are symmetric and reduced compared to WT (Fig. 197 3E). Treatment with SU5402 in WT embryos produced a pronounced reduction of migration velocity 198 similar to that of ntl and spaw mutants (Fig. 3E). We did observe weak asymmetry in left versus right 199 migration velocities, as is expected given these heart tubes jog correctly to the left and Nodal 200 expression is unaffected in these embryos (Fig. 1D, 2F, 3E). However, the differences in left versus 201 right velocities were not statistically significant in this analysis. Intriguingly, SU5402-treated spaw 202 mutants had a more severe reduction in migration velocity compared to spaw mutants or SU5402-203 treated embryos, further suggesting cooperation between Nodal and FGF signals in this process (Fig. 204 3E). Taken together, our findings on CPC migration, velocity, and trajectories all support Nodal acting 205 to bias left migration velocities, while FGF acts with Nodal to promote overall CPC velocities during 206 heart tube morphogenesis. 207 208 Nodal elicits left-right asymmetries in migration by inducing left-right asymmetries in actin 209 polymerization in CPCs 210 Though previous studies have elucidated a role for the Nodal pathway in governing cardiac 211 laterality, the molecular mechanisms by which the pathway does so remain unclear. A promising 212 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint possibility is that Nodal promotes cell migration by asymmetrically influencing the actin cytoskeleton 213 and its regulators during cardiac jogging. It is known that actin can be regulated by Nodal signaling 214 during cardiac looping and in other developmental contexts (Noel et al., 2013; Woo et al., 2012). 215 Furthermore, FGF signaling is also known to regulate cytoskeletal dynamics, including actin 216 polarization (Sai and Ladher, 2008; Yang et al., 2020). To assess this possibility, we generated 217 Tg(myl7:Lifeact-EGFP) fish, in which EGFP-labeled filamentous actin (F-actin) is restricted to CPCs 218 (Huang et al., 2003; Reischauer et al., 2014; Riedl et al., 2008), permitting us to observe F-actin levels 219 and cellular protrusive activity (Fig. 4A). Two-photon imaging of the cardiac cone in WT Lifeact-EGFP 220 transgenics revealed left-sided atrial cells have higher levels of F-actin than the corresponding right-221 sided cells, and this asymmetry in F-actin persisted throughout jogging (Fig. 4B; Movie S6). Imaging 222 injected embryos mosaically expressing Lifeact-EGFP at the cellular level revealed numerous Lifeact-223 EGFP-positive protrusions in left-sided atrial cells where motility rates are high (Fig. 4C; Movie S7). 224 To determine if the higher F-actin levels we observe in left atrial cells are in response to Nodal 225 signaling, we examined actin dynamics in ntl morphants, in spaw morphants, and in embryos treated 226 with SB-505124, a pharmacological inhibitor of Nodal signaling that works well in zebrafish (DaCosta 227 Byfield et al., 2004; Hagos and Dougan, 2007). In WT embryos, left-sided CPCs exhibited high levels of 228 F-actin (Fig. 4D). No increase in F-actin levels in atrial cells was observed when Nodal signaling was 229 lost in spaw morphants or with administration of SB-505124 (Fig. 4D; Movies S8-9). Conversely, we 230 observed a symmetrical increase in F-actin level in ntl morphants, in which Nodal is expressed 231 bilaterally in the LPM (Fig. 4D; Movie S10). To quantify these results, we measured the ratio of 232 fluorescence intensity values between left-sided and right-sided cells (Fig. 4E). An intensity ratio of left-233 to-right cells greater than 1 is indicative of left-biased levels, while an intensity ratio equal to 1 indicates 234 symmetric levels. We found the fluorescence intensity ratio in WT embryos to be left-biased, which is 235 indicative of asymmetric F-actin levels (Fig. 4F). However, ntl morphants, spaw morphants, and 236 SB505124-treated embryos had ratios indicative of symmetric F-actin levels (Fig. 4F). Taken together, 237 these findings support our hypothesis that Nodal induces left-right asymmetries in migration velocity by 238 promoting actin cytoskeleton dynamics. 239 240 FGF signaling is necessary for cardiac looping 241 FGF signaling is known to influence various aspects of zebrafish heart morphogenesis, such as 242 CPC differentiation and proliferation, as well as chamber size and identity establishment (de Pater et 243 al., 2009; Felker et al., 2018; Lazic and Scott, 2011; Marques et al., 2008; Pradhan et al., 2017; Reifers 244 et al., 1998; Zeng and Yelon, 2014). In embryos treated with SU5402 from 24-48 hpf (continuous 245 treatment; Fig. 5A), differences in heart tube morphology between treated and vehicle-treated embryos 246 are apparent. SU5402-treated embryos exhibit defects in heart looping, chamber shape, and ventricle 247 size, as reported previously (Fig. 5B). Given the ostensible importance of FGF signaling in heart 248 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint looping, the second and more evolutionarily conserved asymmetry in heart development, we wanted to 249 determine if there was a more specific developmental window in which FGF signaling influences this 250 process. 251 To further investigate the role of FGF in cardiac looping, we treated embryos with SU5402 252 during more specific developmental windows: during the heart tube elongation phase (24-30 hpf, prior 253 to the onset of looping) and the heart tube bending phase (30-36 hpf, during looping; Fig. 5A). In 254 vehicle-treated embryos, the heart has completed looping and the atrium and ventricle are positioned 255 side-by-side at 48 hpf (Fig. 5B). Treatment during both of our defined developmental windows resulted 256 in a vertically “stacked” chamber phenotype indicative of looping defects (Fig. 5B). Quantification of 257 these chamber placement defects was conducted by measuring the looping angle: the angle formed by 258 the intersection of a line along the anterior-posterior axis and a line through the atrioventricular canal 259 (AVC) at 48 hpf, with a properly developed heart having a looping angle of ~25.0° (Chernyavskaya et 260 al., 2012). We found that SU5402 treatment during the elongation phase resulted in significantly larger 261 looping angles (mean looping angle = 41.2°), suggesting impaired looping, while treatment during the 262 bending phase was not significantly different than WT (Fig. 5C). This suggests FGF signaling between 263 24 to 30 hpf is important for proper heart looping. Heart looping is compromised in spaw mutants (Fig. 264 5C; mean looping angle = 65.1°), and we find loss of FGF signaling during the elongation phase in 265 spaw mutants exacerbates heart looping defects (Fig. 5C; mean looping angle 74.6°). Our results 266 suggest that Nodal and FGF cooperate to promote heart looping. 267 268 SHF addition is required for heart looping 269 Others have shown that loss of FGF signaling post 24 hpf prevents addition of the secondary 270 heart field (SHF) to the arterial pole of the heart tube. To determine if our treatment with SU5402 271 affected SHF addition during the elongation phase, we treated embryos starting at 24 hpf and imaged 272 the heart tube at both 26.5 and 28 hpf. Measuring heart tube length revealed significantly shorter heart 273 tubes in SU5402-treated embryos (Fig. 5D), indicative of loss of the SHF. This suggested to us that the 274 addition of SHF cells is critical for proper heart looping. To address this possibility, we injected embryos 275 at the 1-cell stage with a morpholino against ltbp3, a canonical SHF marker that is necessary for SHF 276 addition to the heart tube (Zhou et al., 2011), and assessed the heart at 48 hpf. ltbp3 morphants 277 exhibited looping and chamber placement defects similar to SU5402-treated embryos (Fig. 6A). 278 Quantification revealed the looping angle in morphants was significantly higher (mean looping angle = 279 51.1°) than in vehicle-injected embryos (mean looping angle = 25.8°; Fig. 6B). This indicates that 280 addition of SHF cells to the heart tube is needed for proper cardiac looping in vivo. 281 282 Loss of FGF signaling during elongation and bending phases compromises AVC development 283 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint SU5402 treatment during the elongation and bending phases produced aberrant chamber 284 morphologies with smaller ventricles and elongated atria (Fig. 5B). Intriguingly, it is known that 285 nppa/nppb double mutants exhibit abnormal atrial morphologies due to errors in AVC development; 286 AVC defects result in improper blood flow, which in turn cause defects in chamber ballooning (Auman 287 et al., 2007; Dietrich et al., 2014; Grassini et al., 2018). To determine if FGF inhibition influences AVC 288 development, we analyzed expression of nppa. nppa is typically expressed in the outer curvatures of 289 the heart chambers but is excluded from the AVC, a phenotype observed in vehicle-treated embryos 290 (Fig. 6C-D). In SU5402-treated embryos, nppa was expressed in the AVC, which is an abnormal nppa 291 expression pattern (Fig. 6C-D). Overall, these findings suggest that FGF signaling may influence 292 chamber morphology by promoting proper formation of the AVC. 293 294

295 Nodal and FGF signals function synergistically to promote migration underlying cardiac jogging 296 and tube extension 297 Our results show that Nodal and FGF signals cooperatively promote overall cell migration during 298 heart tube formation and asymmetric placement. Losing either signal reduces cell velocities in the cone 299 and delays the cone-to-tube transition compared to WT, while losing both signals more severely 300 perturbs CPC migration and tube formation. Though anterior movement of CPCs is reduced in hearts 301 lacking Nodal or FGF signals, it is not completely abolished, raising the question as to what additional 302 signals govern the anterior migration of CPCs. During jogging, our data suggests that FGF signals act 303 permissively to promote motility, while Nodal signals act instructively to direct CPC laterality. FGF 304 signaling is known to cooperate with Nodal signaling in coupling cell migration in various morphogenetic 305 events, including in the Ciona neural tube and the zebrafish brain (Navarrete and Levine, 2016; Regan 306 et al., 2009; Roussigne et al., 2018). Intriguingly, in the zebrafish brain, FGF and Nodal signaling 307 function together to control left-right asymmetry by mediating collective cell migration; Nodal restricts 308 and biases the activation of FGF to leading parapineal cells, resulting in a leftward, asymmetric 309 migration (Regan et al., 2009; Roussigne et al., 2018). Similar to what we report in this paper, while 310 FGF is critical for parapineal cell migration, Nodal signaling directs the laterality of this movement. 311 We also find that the heart tubes of spaw and ntl mutants are significantly shorter than those of 312 WT embryos. Whether this defect is only the consequence of compromised motility, or whether there 313 are other forces that restrict elongation, remains to be determined. For example, the defects we 314 observe are highly reminiscent of heart and mind (had) and heart and soul (has) mutants, which have 315 reduced heart tube extension (Rohr et al., 2006; Shu et al., 2003). Interestingly, had and has encode 316 regulators of epithelial polarity, the establishment and maintenance of which is crucial for proper 317 cardiac morphogenesis. Interrogating epithelial polarity and cell shape changes under the conditions in 318 our study may well prove informative in understanding heart tube extension. Alternatively, the decrease 319 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint in heart tube length we observe in our study conditions could be due to tissue constraints on the tube 320 as it attempts to elongate closer to the midline. Future studies are needed to determine what drives the 321 overall anterior displacement of the heart cone/tube and the forces that drive tube extension during 322 development. 323 324 Nodal signals induce left-right asymmetries in CPCs by impinging on the actin cytoskeleton 325 To determine the mechanisms by which Nodal regulates asymmetric CPC migration, we imaged 326 actin filament formation during cardiac jogging using a Lifeact reporter. Previous studies have 327 suggested roles for Nodal in inducing cytoskeletal rearrangements in migrating cells (Noel et al., 2013; 328 Schier, 2009; Woo et al., 2012). Here, we link its asymmetric impingement on the actin cytoskeleton to 329 tissue morphogenesis in vivo. In wild-type embryos we observe an increase in F-actin in migrating left 330 atrial cells that is not observed in right atrial cells. This increase in F-actin is Nodal dependent, as it is 331 lost in embryos injected with spaw morpholino or treated with Nodal inhibitor. In agreement, in embryos 332 injected with ntl morpholino, where Nodal signals to both sides of the cardiac cone, both left and right 333 atrial cells show increases in F-actin. This suggests that Nodal induces asymmetric migration by 334 increasing actin dynamics within left-atrial cells, which migrate faster than their counterparts on the 335 right, driving clockwise rotation of the cone. The cells in the cone are epithelial and migrate collectively 336 during jogging. Intriguingly, using mosaic labeling with our Lifeact reporter, we observed highly 337 dynamic, protrusive F-actin activity in these left-sided migrating cells. It is intriguing to speculate that 338 these are “cryptic” basal extensions that are driving migration similar to what has been observed in 339 cultured MDCK kidney cells during wound healing (Farooqui and Fenteany, 2005). Future studies can 340 determine if these protrusions are basal-to-junctional complexes coming only from edge cells or if 341 additional cells deeper in this epithelium are helping drive migration. 342 343 FGF signaling is necessary for proper cardiac looping and AVC development 344 Extending our studies with the FGF inhibitor SU5402, we find that treatment with SU5402 from 345 24 to 30 hpf (elongation) affected both looping and chamber expansion, while treatment from 30 to 36 346 hpf (bending) primarily affected chamber expansion. This suggests that FGF continues to play 347 important roles in heart morphogenesis after jogging is complete. Cardiac looping is a highly conserved 348 event in vertebrate heart development, but much remains to be discovered regarding its underlying 349 mechanisms. We previously reported that jogging correctly increases the correct dextral loop in the 350 heart, suggesting that the first asymmetric migration at 20 hpf still provides input to heart-intrinsic 351 chirality mechanisms at later stages (Grimes et al., 2020). Our work here suggests that at least one 352 additional input into looping in vivo involves FGF. We provide evidence that this input may be through 353 SHF addition given the similarity in phenotypes between ltbp3 morphants and SU5402-treated 354 embryos, both of which lack SHF structures. This is further supported by work showing heart looping 355 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint defects in isl2a and isl2b mutant embryos, which produce defects in the SHF (Witzel et al., 2017). 356 Determining how the SHF contributes to proper heart looping in zebrafish will be an important area for 357 future studies. Intriguingly, we show that Nodal and FGF cooperate to promote heart looping, as losing 358 both pathways produces more severe defects in this process than losing either alone. 359 A requirement for proper blood flow in chamber expansion is clear (Auman et al., 2007; Dietrich 360 et al., 2014); however, the physical forces governing chamber shape remain to be fully determined. We 361 find that inhibiting FGF at elongation or bending phases affects AVC development. In treated embryos, 362 nppa expression expands into the AVC region, suggesting that there are defects in specification or 363 differentiation of this structure. Such an issue could lead to defective functioning of the AVC, resulting in 364 aberrant blood flow, improper chamber expansion, and the abnormal chamber shapes we observe. 365 Determining what aspects of cell movements or cell specification are influenced by FGF in the AVC is 366 an area for future studies. 367 368 Nodal and FGF signaling are critical for heart morphogenesis 369 Here, we highlight molecular mechanisms through which the spatiotemporal dynamics of 370 signaling cues influence cardiac precursor cell behaviors and heart morphogenesis. We further show 371 that the differential effects exerted by interacting signals on CPCs manifest as dramatic asymmetries in 372 heart tube morphogenesis, highlighting the fact that earlier patterning events regulate morphogenetic 373 processes occurring later in development. Heart tube malformations are among the most commonly 374 diagnosed congenital heart defects, thus understanding how they arise is crucial for reducing the 375 mortality and morbidity associated with CHDs. Ultimately, further interrogation of the roles of Nodal and 376 FGF signaling throughout heart morphogenesis will prove informative in understanding how signaling 377 and cellular aberrancies manifest as morphological defects. 378 379

380 Zebrafish strains 381 All experimental procedures in this study were conducted in accordance with the Princeton University 382 Institutional Animal Care and Use Committee. Zebrafish (Danio rerio) embryos were maintained at 383 28°C and grown in E3 embryo medium (5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl2, 0.33 mM MgSO4). 384 We used the zebrafish mutants ntlb160 and spawsa177 (Grimes et al., 2020; Halpern et al., 1993). We 385 used the transgenic strain Tg(myl7:EGFP) (Huang et al., 2003). 386 387 Generation of transgenic line 388 We generated the Tg(myl7:Lifeact-EGFP) transgenic strain by using the Tol2 transposase system 389 (Kawakami, 2007). Briefly, 1-cell stage embryos were injected with Tol2 mRNA and pTol2005b-390 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint myl7:Lifeact-EGFP plasmid. Only correctly developing embryos with EGFP-positive hearts were raised. 391 Injected F0 fish were outcrossed to generate a stable transgenic line. 392 393 Drug treatments and morpholino injections 394 A 10mM stock of the FGFR inhibitor SU5402 (Sigma-Aldrich) in DMSO was diluted to a working 395 concentration of 6μ M or 10μ M in a 1% DMSO/1X E3 solution. 6μ M was used for inhibition during 396 bending and elongation phases and 10μ M was used for inhibition during jogging phase. Embryos were 397 incubated at 28°C in the dark for varying durations of time: 18.5-24 hpf, 24-30 hpf, and 30-36 hpf. Upon 398 completion of SU5402 incubation, embryos underwent live imaging or were fixed in 4% 399 paraformaldehyde (PFA; Electron Microscopy Sciences) for RNA in situ analysis. Due to the light-400 sensitivity of SU5402, treated embryos were incubated in the dark. For SB-505124 (Sigma-Aldrich) 401 treatments, a 10mM stock in DMSO was diluted to a working concentration of 40 μ M and administered 402 to embryos at the tailbud stage; embryos then underwent live imaging during jogging as described 403 (Lenhart et al., 2013). Note that DMSO-treated vehicles did not differ from WT embryos in the 404 processes studied. Morpholino injections were performed as previously described using established 405 morpholinos for spaw, ntl, and ltbp3 (Lenhart et al., 2013; Long et al., 2003; Nasevicius and Ekker, 406 2000; Zhou et al., 2011). 407 408 RNA in situ hybridization 409 Chromogenic whole-mount RNA in situ hybridization was performed using digoxygenin-labelled RNA 410 probes for myl7, lefty2, and nppa (Berdougo et al., 2003; Bisgrove et al., 1999; Thisse and Thisse, 411 2014; Yelon et al., 1999). Images were acquired using the Leica DMRA microscope. 412 Fluorescent whole-mount in situ hybridization was performed using the Molecular Instruments HCR kit 413 as described (Choi et al., 2018) for probes against myl7 and fgf8. Images were acquired using the 414 Nikon A1 confocal microscope. 415 416 Live imaging and analysis 417 To analyze cell migration during cardiac jogging, dechorionated embryos were mounted in 0.8% 418 low-melt agarose as previously described (Lenhart et al., 2013). The dish was covered with a 0.5% 419 DMSO/1X E3 solution containing 0.13mM tricaine (MS-222; Sigma-Aldrich). For SU5402 treatments, a 420 5 μ M dilution in 0.8% agarose was used for mounting embryos, which were incubated in inhibitor at 421 least 30 minutes before mounting and covered with 10μ M inhibitor in 0.5% DMSO/1X E3 containing 422 0.13mM tricaine. 423 For analysis of cardiac cell movements, embryos were imaged from 18-24 hpf using a Leica 424 SP5 confocal microscope. The Leica Mark & Find feature was used to image up to 4 embryos per 425 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint imaging session for higher throughput. A heated stage set to 28.5°C was used, which permitted 426 imaging for several hours without perturbing development. 427 We acquired confocal z-stacks at 4- or 5-minute intervals with a spacing of 2 μ m. To track 428 individual cells during jogging, we compressed the z-stacks into a single dimension and then utilized 429 the ImageJ program TrackMate to track and quantify metrics of each cell during its migration. To 430 ensure consistent quantification of velocity over time between different genotypes, we tracked CPCs 431 over the same developmental window: we began tracking cells upon anterior fusion of CPCs at the 432 midline and ended tracking 3 hours thereafter. Note that only outer, atrial cells were chosen for tracking 433 as they undergo the most migration during jogging; the bright ventricular cells around the lumen are 434 excluded from analysis (Lenhart et al., 2013). Average velocity and angle of displacement for each cell 435 was calculated using raw position, displacement, and time measurements generated by TrackMate. 436 Resulting cell velocities, angles, and displacements were input into Prism or RStudio for analysis. Note 437 that distinct microscopes, acquisition parameters, and analysis software were utilized and may account 438 for differences in velocity values, compared to our previous publications. 439 We note that the decreased velocity we observe here in ntl mutants conflicts with our previously 440 published data using ntl morphants, in which we observed higher velocities in CPCs compared to WT 441 (Lenhart et al., 2013). It is possible that there may be an unexpected difference in ntl mutants versus 442 morphants that produces this discrepancy (Stainier et al., 2017). However, distinct microscopes, 443 acquisition parameters, and analysis software were utilized in this study and could also account for 444 differences in velocity values, compared to our previous publication. 445 For two-photon analysis of protrusive F-actin activity, embryos were mounted and handled as 446 described above. Embryos were imaged from 19-23 hpf using a Prairie Ultima two-photon microscope. 447 Z-stacks were acquired at 15-minute intervals with a spacing of 4 μ m. Fluorescence intensity analysis 448 was performed using ImageJ. 449 For observing heart tube length and cardiac looping, embryos were mounted and handled as 450 described above. Images were acquired using the Leica M205 FA fluorescent stereoscope. Analysis 451 was performed using ImageJ as described previously or below. 452 453 Heart measurements 454 For fixed embryos processed via RNA in situ hybridization, heart tube length was measured using the 455 ruler tool in Photoshop or ImageJ. For observing heart tube length and cardiac looping in live embryos, 456 embryos were mounted and handled as described above. Images were acquired using the Leica M205 457 FA fluorescent stereoscope. Analysis was performed using ImageJ ruler and angle tools. The looping 458 angle is defined as the angle between the plane of the anterior-posterior axis and the AVC 459 (Chernyavskaya et al., 2012). 460 461 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint Statistical analysis 462 All bar graphs were generated using GraphPad Prism 9.4.0. Statistical tests of significance (Student’s t-463 test, Chi-square analysis) were calculated using GraphPad Prism analysis software. 464 465

466 We would like to thank Dr. Gary Laevsky and the Confocal Imaging Facility, a Nikon Center of 467 Excellence, in the Department of Molecular Biology for their assistance with imaging. We also thank 468 Phillip Johnson and his team for their assistance with fish husbandry. 469 470 Competing Interests 471 The authors declare that they have no conflict of interest. 472 473 Funding 474 M.S. was a visiting scientist supported by the Strategic International Research Exchange Program 475 between Princeton University and National Institutes of Natural Sciences, Japan. Research reported in 476 this manuscript was supported by a National Science Foundation under grant IOS-1147123 to RDB and 477 a predoctoral fellowship to JRW; a Janssen Scholars of Oncology Diversity Engagement Fellowship to 478 VG; grant R01HD048584 from the National Institutes of Child Health and Human Development to RDB, 479 and grant T32GM007388 from the National Institute of General Medical Sciences to JRW. The content 480 is solely the responsibility of the authors and does not necessarily represent the official views of the 481 National Institutes of Health. 482 483 Data availability 484 The data that support the findings of this study are available from the corresponding author RDB, upon 485 reasonable request. 486 487 Diversity and inclusion statement 488 One or more of the authors of this paper self-identifies as an underrepresented ethnic minority in 489 science. One or more of the authors of this paper self-identifies as living with a disability. One or more 490 authors of this paper received support from a program designed to increase minority representation in 491 science. Five of the six authors on this paper are women who are often underrepresented in science. 492 493

494 Amack, J. D. and Yost, H. J. (2004). The T box transcription factor no tail in ciliated cells controls 495 zebrafish left-right asymmetry. Curr Biol 14, 685-690. 496 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint Auman, H. J., Coleman, H., Riley, H. E., Olale, F., Tsai, H. J. and Yelon, D. (2007). Functional 497 modulation of cardiac form through regionally confined cell shape changes. PLoS Biol 5, e53. 498 Baker, K., Holtzman, N. G. and Burdine, R. D. (2008). Direct and indirect roles for Nodal signaling in 499 two axis conversions during asymmetric morphogenesis of the zebrafish heart. Proc Natl Acad 500 Sci U S A 105, 13924-13929. 501 Bakkers, J. (2011). Zebrafish as a model to study cardiac development and human cardiac disease. 502 Cardiovasc Res 91, 279-288. 503 Bennett, J. T., Joubin, K., Cheng, S., Aanstad, P., Herwig, R., Clark, M., Lehrach, H. and Schier, 504 A. F. (2007). Nodal signaling activates differentiation genes during zebrafish gastrulation. Dev 505 Biol 304, 525-540. 506 Berdougo, E., Coleman, H., Lee, D. H., Stainier, D. Y. and Yelon, D. (2003). Mutation of weak 507 atrium/atrial myosin heavy chain disrupts atrial function and influences ventricular 508 morphogenesis in zebrafish. Development 130, 6121-6129. 509 Bisgrove, B. W., Essner, J. J. and Yost, H. J. (1999). Regulation of midline development by 510 antagonism of lefty and nodal signaling. Development 126, 3253-3262. 511 Chernyavskaya, Y., Ebert, A. M., Milligan, E. and Garrity, D. M. (2012). Voltage-gated calcium 512 channel CACNB2 (beta2.1) protein is required in the heart for control of cell proliferation and 513 heart tube integrity. Dev Dyn 241, 648-662. 514 Choi, H. M. T., Schwarzkopf, M., Fornace, M. E., Acharya, A., Artavanis, G., Stegmaier, J., Cunha, 515 A. and Pierce, N. A. (2018). Third-generation in situ hybridization chain reaction: multiplexed, 516 quantitative, sensitive, versatile, robust. Development 145. 517 DaCosta Byfield, S., Major, C., Laping, N. J. and Roberts, A. B. (2004). SB-505124 is a selective 518 inhibitor of transforming growth factor-beta type I receptors ALK4, ALK5, and ALK7. Mol 519 Pharmacol 65, 744-752. 520 de Campos-Baptista, M. I., Holtzman, N. G., Yelon, D. and Schier, A. F. (2008). Nodal signaling 521 promotes the speed and directional movement of cardiomyocytes in zebrafish. Dev Dyn 237, 522 3624-3633. 523 de Pater, E., Clijsters, L., Marques, S. R., Lin, Y. F., Garavito-Aguilar, Z. V., Yelon, D. and 524 Bakkers, J. (2009). Distinct phases of cardiomyocyte differentiation regulate growth of the 525 zebrafish heart. Development 136, 1633-1641. 526 Desgrange, A., Le Garrec, J. F. and Meilhac, S. M. (2018). Left-right asymmetry in heart 527 development and disease: forming the right loop. Development 145. 528 Dietrich, A. C., Lombardo, V. A., Veerkamp, J., Priller, F. and Abdelilah-Seyfried, S. (2014). Blood 529 flow and Bmp signaling control endocardial chamber morphogenesis. Dev Cell 30, 367-377. 530 Farooqui, R. and Fenteany, G. (2005). Multiple rows of cells behind an epithelial wound edge extend 531 cryptic lamellipodia to collectively drive cell-sheet movement. J Cell Sci 118, 51-63. 532 Felker, A., Prummel, K. D., Merks, A. M., Mickoleit, M., Brombacher, E. C., Huisken, J., Panakova, 533 D. and Mosimann, C. (2018). Continuous addition of progenitors forms the cardiac ventricle in 534 zebrafish. Nat Commun 9, 2001. 535 Genge, C. E., Lin, E., Lee, L., Sheng, X., Rayani, K., Gunawan, M., Stevens, C. M., Li, A. Y., Talab, 536 S. S., Claydon, T. W., et al. (2016). The Zebrafish Heart as a Model of Mammalian Cardiac 537 Function. Rev Physiol Biochem Pharmacol 171, 99-136. 538 Grant, M. G., Patterson, V. L., Grimes, D. T. and Burdine, R. D. (2017). Modeling Syndromic 539 Congenital Heart Defects in Zebrafish. Curr Top Dev Biol 124, 1-40. 540 Grassini, D. R., Lagendijk, A. K., De Angelis, J. E., Da Silva, J., Jeanes, A., Zettler, N., Bower, N. 541 I., Hogan, B. M. and Smith, K. A. (2018). Nppa and Nppb act redundantly during zebrafish 542 cardiac development to confine AVC marker expression and reduce cardiac jelly volume. 543 Development 145. 544 Grimes, D. T., Patterson, V. L., Luna-Arvizu, G., Schottenfeld-Roames, J., Irons, Z. H. and 545 Burdine, R. D. (2020). Left-right asymmetric heart jogging increases the robustness of dextral 546 heart looping in zebrafish. Dev Biol 459, 79-86. 547 Hagos, E. G. and Dougan, S. T. (2007). Time-dependent patterning of the mesoderm and endoderm 548 by Nodal signals in zebrafish. BMC Dev Biol 7, 22. 549 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint Halpern, M. E., Ho, R. K., Walker, C. and Kimmel, C. B. (1993). Induction of muscle pioneers and 550 floor plate is distinguished by the zebrafish no tail mutation. Cell 75, 99-111. 551 Hill, C. S. (2016). Transcriptional Control by the SMADs. Cold Spring Harb Perspect Biol 8. 552 Huang, C. J., Tu, C. T., Hsiao, C. D., Hsieh, F. J. and Tsai, H. J. (2003). Germ-line transmission of a 553 myocardium-specific GFP transgene reveals critical regulatory elements in the cardiac myosin 554 light chain 2 promoter of zebrafish. Dev Dyn 228, 30-40. 555 Kawakami, K. (2007). Tol2: a versatile gene transfer vector in vertebrates. Genome Biol 8 Suppl 1, 556 S7. 557 Kidokoro, H., Saijoh, Y. and Schoenwolf, G. C. (2022). Nodal signaling regulates asymmetric cellular 558 behaviors, driving clockwise rotation of the heart tube in zebrafish. Commun Biol 5, 996. 559 Lazic, S. and Scott, I. C. (2011). Mef2cb regulates late myocardial cell addition from a second heart 560 field-like population of progenitors in zebrafish. Dev Biol 354, 123-133. 561 Lenhart, K. F., Holtzman, N. G., Williams, J. R. and Burdine, R. D. (2013). Integration of nodal and 562 BMP signals in the heart requires FoxH1 to create left-right differences in cell migration rates 563 that direct cardiac asymmetry. PLoS Genet 9, e1003109. 564 Long, S., Ahmad, N. and Rebagliati, M. (2003). The zebrafish nodal-related gene southpaw is 565 required for visceral and diencephalic left-right asymmetry. Development 130, 2303-2316. 566 Marques, S. R., Lee, Y., Poss, K. D. and Yelon, D. (2008). Reiterative roles for FGF signaling in the 567 establishment of size and proportion of the zebrafish heart. Dev Biol 321, 397-406. 568 Massague, J. (2003). Integration of Smad and MAPK pathways: a link and a linker revisited. Genes 569 Dev 17, 2993-2997. 570 Mohammadi, M., McMahon, G., Sun, L., Tang, C., Hirth, P., Yeh, B. K., Hubbard, S. R. and 571 Schlessinger, J. (1997). Structures of the tyrosine kinase domain of fibroblast growth factor 572 receptor in complex with inhibitors. Science 276, 955-960. 573 Nasevicius, A. and Ekker, S. C. (2000). Effective targeted gene 'knockdown' in zebrafish. Nat Genet 574 26, 216-220. 575 Navarrete, I. A. and Levine, M. (2016). Nodal and FGF coordinate ascidian neural tube 576 morphogenesis. Development 143, 4665-4675. 577 Nguyen, C. T., Lu, Q., Wang, Y. and Chen, J. N. (2008). Zebrafish as a model for cardiovascular 578 development and disease. Drug Discov Today Dis Models 5, 135-140. 579 Noel, E. S., Verhoeven, M., Lagendijk, A. K., Tessadori, F., Smith, K., Choorapoikayil, S., den 580 Hertog, J. and Bakkers, J. (2013). A Nodal-independent and tissue-intrinsic mechanism 581 controls heart-looping chirality. Nat Commun 4, 2754. 582 Pradhan, A., Zeng, X. I., Sidhwani, P., Marques, S. R., George, V., Targoff, K. L., Chi, N. C. and 583 Yelon, D. (2017). FGF signaling enforces cardiac chamber identity in the developing ventricle. 584 Development 144, 1328-1338. 585 Regan, J. C., Concha, M. L., Roussigne, M., Russell, C. and Wilson, S. W. (2009). An Fgf8-586 dependent bistable cell migratory event establishes CNS asymmetry. Neuron 61, 27-34. 587 Reifers, F., Bohli, H., Walsh, E. C., Crossley, P. H., Stainier, D. Y. and Brand, M. (1998). Fgf8 is 588 mutated in zebrafish acerebellar (ace) mutants and is required for maintenance of midbrain-589 hindbrain boundary development and somitogenesis. Development 125, 2381-2395. 590 Reifers, F., Walsh, E. C., Leger, S., Stainier, D. Y. and Brand, M. (2000). Induction and differentiation 591 of the zebrafish heart requires fibroblast growth factor 8 (fgf8/acerebellar). Development 127, 592 225-235. 593 Reischauer, S., Arnaout, R., Ramadass, R. and Stainier, D. Y. (2014). Actin binding GFP allows 4D 594 in vivo imaging of myofilament dynamics in the zebrafish heart and the identification of Erbb2 595 signaling as a remodeling factor of myofibril architecture. Circ Res 115, 845-856. 596 Riedl, J., Crevenna, A. H., Kessenbrock, K., Yu, J. H., Neukirchen, D., Bista, M., Bradke, F., 597 Jenne, D., Holak, T. A., Werb, Z., et al. (2008). Lifeact: a versatile marker to visualize F-actin. 598 Nat Methods 5, 605-607. 599 Rohr, S., Bit-Avragim, N. and Abdelilah-Seyfried, S. (2006). Heart and soul/PRKCi and nagie 600 oko/Mpp5 regulate myocardial coherence and remodeling during cardiac morphogenesis. 601 Development 133, 107-115. 602 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint Rohr, S., Otten, C. and Abdelilah-Seyfried, S. (2008). Asymmetric involution of the myocardial field 603 drives heart tube formation in zebrafish. Circ Res 102, e12-19. 604 Roussigne, M., Wei, L., Tsingos, E., Kuchling, F., Alkobtawi, M., Tsalavouta, M., Wittbrodt, J., 605 Carl, M., Blader, P. and Wilson, S. W. (2018). Left/right asymmetric collective migration of 606 parapineal cells is mediated by focal FGF signaling activity in leading cells. Proc Natl Acad Sci 607 U S A 115, E9812-E9821. 608 Sai, X. and Ladher, R. K. (2008). FGF signaling regulates cytoskeletal remodeling during epithelial 609 morphogenesis. Curr Biol 18, 976-981. 610 Schier, A. F. (2009). Nodal morphogens. Cold Spring Harb Perspect Biol 1, a003459. 611 Shu, X., Cheng, K., Patel, N., Chen, F., Joseph, E., Tsai, H. J. and Chen, J. N. (2003). Na,K-ATPase 612 is essential for embryonic heart development in the zebrafish. Development 130, 6165-6173. 613 Smith, K. A., Chocron, S., von der Hardt, S., de Pater, E., Soufan, A., Bussmann, J., Schulte-614 Merker, S., Hammerschmidt, M. and Bakkers, J. (2008). Rotation and asymmetric 615 development of the zebrafish heart requires directed migration of cardiac progenitor cells. Dev 616 Cell 14, 287-297. 617 Smith, K. A. and Uribe, V. (2021). Getting to the Heart of Left-Right Asymmetry: Contributions from 618 the Zebrafish Model. J Cardiovasc Dev Dis 8. 619 Stainier, D. Y. R., Raz, E., Lawson, N. D., Ekker, S. C., Burdine, R. D., Eisen, J. S., Ingham, P. W., 620 Schulte-Merker, S., Yelon, D., Weinstein, B. M., et al. (2017). Guidelines for morpholino use 621 in zebrafish. PLoS Genet 13, e1007000. 622 Thisse, B. and Thisse, C. (2014). In situ hybridization on whole-mount zebrafish embryos and young 623 larvae. Methods Mol Biol 1211, 53-67. 624 Tsao, C. W., Aday, A. W., Almarzooq, Z. I., Alonso, A., Beaton, A. Z., Bittencourt, M. S., Boehme, 625 A. K., Buxton, A. E., Carson, A. P., Commodore-Mensah, Y., et al. (2022). Heart Disease 626 and Stroke Statistics-2022 Update: A Report From the American Heart Association. Circulation 627 145, e153-e639. 628 Veerkamp, J., Rudolph, F., Cseresnyes, Z., Priller, F., Otten, C., Renz, M., Schaefer, L. and 629 Abdelilah-Seyfried, S. (2013). Unilateral dampening of Bmp activity by nodal generates cardiac 630 left-right asymmetry. Dev Cell 24, 660-667. 631 Williams, J. R. (2015). Nodal Directs Asymmetric Cardiac Morphogenesis Through Regulation of Actin 632 Cytoskeleton. In Molecular Biology. Princeton, NJ: Princeton University. 633 Witzel, H. R., Cheedipudi, S., Gao, R., Stainier, D. Y. and Dobreva, G. D. (2017). Isl2b regulates 634 anterior second heart field development in zebrafish. Sci Rep 7, 41043. 635 Woo, S., Housley, M. P., Weiner, O. D. and Stainier, D. Y. (2012). Nodal signaling regulates 636 endodermal cell motility and actin dynamics via Rac1 and Prex1. J Cell Biol 198, 941-952. 637 Yang, S., Weske, A., Du, Y., Valera, J. M., Jones, K. L. and Johnson, A. N. (2020). FGF signaling 638 directs myotube guidance by regulating Rac activity. Development 147. 639 Yelon, D., Horne, S. A. and Stainier, D. Y. (1999). Restricted expression of cardiac myosin genes 640 reveals regulated aspects of heart tube assembly in zebrafish. Dev Biol 214, 23-37. 641 Zeng, X. X. and Yelon, D. (2014). Cadm4 restricts the production of cardiac outflow tract progenitor 642 cells. Cell Rep 7, 951-960. 643 Zhou, Y., Cashman, T. J., Nevis, K. R., Obregon, P., Carney, S. A., Liu, Y., Gu, A., Mosimann, C., 644 Sondalle, S., Peterson, R. E., et al. (2011). Latent TGF-beta binding protein 3 identifies a 645 second heart field in zebrafish. Nature 474, 645-648. 646 647 Figure legends 648 Figure 1. FGF signaling is active in the developing heart and necessary for proper heart 649 development. (A) RNA in situ hybridization (ISH) by hybridization chain reaction (HCR) for fgf8 650 (magenta) and myl7 (green) in the cardiac cone of a 20 hours post fertilization (hpf) wild-type (WT) 651 embryo. Scale = 25 µm. (B) Representative images of 5 days post fertilization (dpf) embryos treated 652 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint with vehicle or FGF inhibitor SU5402 from 24-30 hpf (ventricle in magenta, atrium in cyan). SU5402-653 treated embryos display aberrant cardiac looping, blood pooling at the sinus venosus, and pericardial 654 edema (lateral view) as well as the atrium failing to migrate behind the ventricle (ventral view). (C) ISH 655 for Nodal target gene lefty2 in the cardiac cone of 20 hpf embryos shows correct left-sided lefty2 and 656 incorrect bilateral and absent lefty2 expression. (D) Quantification of lefty2 expression sidedness. n is 657 WT = 74, ntl-/- = 50, spaw-/- = 36, SU5402 = 41, ntl-/- + SU5402 = 177, spaw-/- + SU5402 = 63. Statistical 658 significance was determined by Chi square analysis. (A, C) images are dorsal views, with anterior to 659 the top and left to the reader’s left. ns = not significant, *** = p<0.0001. 660 661 Figure 2. FGF signaling is necessary for proper CPC migration but not cardiac laterality during 662 jogging. (A) Schematization depicting the 18.5-24 hpf SU5402 treatment window during the “Cone 663 Formation and Jogging phase”. (B) RNA ISH for myl7 at 24 hpf displaying the stages from cone 664 through extending to correctly jogged heart tubes we see in our treated embryos. (C) Quantification of 665 cardiac migration phenotypes. Number of embryos (n) is WT = 686, ntl-/- = 152, spaw-/- = 171, SU5402 666 = 531, ntl-/- + SU5402 = 117, spaw-/- + SU5402 = 146. Statistical significance determined by Chi-square 667 analysis. (D) Quantification of heart tube length at 24 hpf. n is WT = 39, ntl-/- = 51, spaw-/- = 32, SU5402 668 = 16. Statistical significance determined by Student’s t-test. (E) RNA ISH for myl7 at 26.5 hpf 669 visualizing correct left-sided positioning and incorrect midline and right-sided positioning of the heart 670 tube. (F) Quantification of cardiac laterality phenotypes. n is WT = 534, ntl-/- = 113, spaw-/- = 138, 671 SU5402 = 112, ntl-/- + SU5402 = 12, spaw-/- + SU5402 = 30. Statistical significance determined by Chi-672 square analysis. (B, E) images are dorsal views, with anterior to the top and left to the reader’s left. ns 673 = not significant, * = p<0.05, *** = p<0.0001. 674 675 Figure 3. FGF signaling promotes CPC migration while Nodal promotes CPC migration and 676 laterality during jogging. (A) Schematization of imaging strategy for time-lapse imaging of the cardiac 677 cone throughout jogging, with outer atrial cells in light green and inner ventricular cells in dark green. 678 Left-sided (blue) and right-sided (orange) cells that underwent tracking for analysis are shown. (B) 679 Representative time-lapse images depicting cardiac progenitor cell (CPC) migration from the formation 680 of the cardiac cone through the next 3 hours of development. WT and ntl-/- embryos typically complete 681 jogging and tube formation in this timeframe. Arrows indicate heart tube direction. Scale = 50 µm. (C) 682 Trajectories of left-sided (blue) and right-sided (orange) CPCs during 3 hours of jogging. Scale = 50 683 µm. (D) Rose plots demonstrating the angle of displacement of each tracked left-sided (blue) and right-684 sided (orange) CPC after 3 hours of jogging. (E) Average velocity of left-sided and right-sided CPCs 685 during 3 hours of jogging. Statistical significance determined by Student's t-test. (C-E) n = 3 embryos (5 686 left-sided and 5 right-sided cells tracked per embryo) per condition. (B-C) images are dorsal views, with 687 anterior to the top and left to the reader’s left. ns = not significant, * = p<0.01. 688 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint 689 Figure 4. F-actin dynamics are asymmetric and Nodal-dependent in CPCs during jogging. (A) 690 Tg(myl7:Lifeact-EGFP) embryo at 24 hpf demonstrating the fluorescence of the CPCs composing the 691 heart tube. (B) F-actin protrusive activity in left (blue box) compared to right (orange box) CPCs in the 692 cardiac cone of a Tg(myl7:Lifeact-EGFP) embryo over the course of jogging from 19-23 hpf. (C) F-actin 693 protrusions (pink asterisks) in a singular CPC of a mosaic Tg(myl7:Lifeact-EGFP) embryo at 21 hpf. (D) 694 Cardiac cones of Tg(myl7:Lifeact-EGFP) embryos, with left-sided (blue oval) and right-sided (orange 695 oval) CPCs marked. (E) Schematization of strategy for quantification of fluorescence intensity ratios in 696 the cardiac cone, with the outer atrial cells in light green and the inner ventricular cells in dark green. 697 (F) Fluorescence intensity ratio between left-sided and right-sided CPCs. n is WT = 4, ntl MO = 4, spaw 698 MO = 7, SB-505124 = 5. Statistical significance determined by Student's t-test. (A-D) images are dorsal 699 views, with anterior to the top and left to the reader’s left. ns = not significant, * = p<0.05. 700 701 Figure 5. FGF signaling is necessary for cardiac looping and heart tube extension. (A) 702 Schematization depicting the SU5402 treatment windows during the “Elongation phase” from 24-30 hpf, 703 the “Bending phase” from 30-36 hpf, and the “Continuous treatment” window from 24-48 hpf. (B) 704 Representative images of vehicle and SU5402-treated Tg(myl7:EGFP) embryo hearts at 48 hpf after 705 various SU5402 treatments, depicting how the looping angle is measured (pink lines; AP = 706 anterior/poster axis line and AVC = line through the AVC). (C) Quantification of looping angle in vehicle 707 and SU5402-treated hearts at 48 hpf. n is vehicle = 35, SU5402 elongation phase = 30, SU5402 708 bending phase = 37, spaw-/- = 53, spaw-/- + SU5402 = 22. Student's t-test. (D) Quantification of heart 709 tube length in vehicle and SU5402-treated Tg(nkx2.5:ZsYellow) embryo heart tubes at 26.5 or 28 hpf. n 710 is vehicle 26.5 hpf = 24, SU5402 26.5 hpf = 7, vehicle 28 hpf = 13, SU5402 28 hpf = 19. Student's t-711 test. (B) images are ventral views, with anterior to the top and left to the reader’s right. ns = not 712 significant, * = p<0.05, ** = p<.001, *** = p<0.0001. 713 714 Figure 6. FGF signaling is necessary for proper looping via SHF and necessary for proper AVC 715 development. 716 (A) Representative images of vehicle and ltbp3 morphants at 48 hpf. (B) Quantification of looping 717 angle in vehicle and ltbp3 morphant hearts at 48 hpf. n is vehicle = 29, MO = 50. Student's t-test. (C) 718 RNA ISH for nppa in the heart of 48 hpf embryos. WT vehicle-injected embryos have no nppa 719 expression in the AVC, while SU5402-injected embryos do (black arrows). (D) Quantification of nppa 720 expression pattern of 48 hpf hearts. n is vehicle = 22, SU5402 elongation phase = 7, SU5402 bending 721 phase = 18, spaw-/- = 15, spaw-/- + SU5402 = 13. Chi-square analysis. (A, C) images are ventral views, 722 with anterior to the top and left to the reader’s right. ns = not significant, *** = p<0.0001. 723 724 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint Supplemental Movies 725 Movie S1. Live imaging of CPCs jogging in a WT transgenic Tg(myl7:EGFP) embryo from the onset of 726 cardiac cone formation until three hours after. 727 728 Movie S2. Live imaging of CPCs jogging in a ntl mutant Tg(myl7:EGFP) embryo from the onset of 729 cardiac cone formation until three hours after. 730 731 Movie S3. Live imaging of CPCs jogging in a spaw mutant Tg(myl7:EGFP) embryo from the onset of 732 cardiac cone formation until three hours after. 733 734 Movie S4. Live imaging of CPCs jogging in a SU5402-treated Tg(myl7:EGFP) embryo from the onset 735 of cardiac cone formation until three hours after. 736 737 Movie S5. Live imaging of CPCs jogging in a spaw mutant and SU5402-treated Tg(myl7:EGFP) 738 embryo from the onset of cardiac cone formation until three hours after. 739 740 Movie S6. Live imaging of F-actin in a WT transgenic Tg(myl7:Lifeact-EGFP) embryo from 19-23 hpf, 741 following the cardiac cone as it undergoes jogging. 742 743 Movie S7. Live imaging of F-actin in a singular cardiac precursor cell (CPC) from a WT mosaic 744 Tg(myl7:Lifeact-EGFP) embryo at 21 hours post fertilization (hpf). Movie is a total of 5 minutes. 745 746 Movie S8. Live imaging of F-actin in a spaw morphant Tg(myl7:Lifeact-EGFP) embryo from 19-23 hpf, 747 following the cardiac cone as it undergoes jogging. 748 749 Movie S9. Live imaging of F-actin in a SB-505124-treated Tg(myl7:Lifeact-EGFP) embryo from 19-23 750 hpf, following the cardiac cone as it undergoes jogging. 751 752 Movie S10. Live imaging of F-actin in a ntl morphant Tg(myl7:Lifeact-EGFP) embryo from 19-23 hpf, 753 following the cardiac cone as it undergoes jogging. 754 755 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint 756 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint 757 758 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint 759 760 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint 761 762 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint 763 764 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint 765 766 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint 767 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 6, 2024. ; https://doi.org/10.1101/2024.01.05.574380doi: bioRxiv preprint