Expression of popdc2 during zebrafish development. (A) RT-PCR analysis of popdc2 expression during development. (B–R) Whole mount in situ hybridization analysis using a full-length popdc2 probe of zebrafish embryos at (B,C) 16-somites stage, (D,E) 22 hpf, (F–H) 24 hpf, (K–M) 48 hpf, and (P–Q) 72 hpf. Stained embryos are viewed from (B,D,F,G,K,L,P,Q) left lateral, (C,E) dorsal, and (H,M) ventrofrontal. Panel (I,J,N,R) depicts transversal and (O) sagittal sections through stained embryos. (S) RT-PCR analysis of popdc2 expression in adult tissues. Abbreviations: a—atrium; cc—cardiac conus; dpw1–5—dorsal pharyngeal wall muscles 1–5; eom—extraocular muscles; fm—facial muscles; GI tract—gastro-intestinal tract; h—heart; ht—tubular heart; m—myotome; pf—pectoral fin bud; s—somites; skm—skeletal muscle; tv1–5—transversus ventralis muscles; v—ventricle. Scale bar = 100 μm.

Aberrant skeletal muscle development in popdc2 morphants. Skeletal muscle tissue formation in the trunk of (A–C) 24 hpf and (D–F) 72 hpf embryos, which were (A,D) wild type (WT), or injected with (B,E) MO1-popdc2, and (C,F) MO1/MO2-popdc2 morpholinos, respectively. Skeletal muscle development was studied by confocal microscopy using rhodamine-phalloidin staining. (A′–F′) depict the morphology of single muscle segments at higher magnification, which are labeled in (A–F) with a white square. The trunk musculature is highly aberrant in the morphants. Segment borders are not well developed and variable in size (square brackets in B). Moreover, muscle fibers are missing in some areas (arrows in E, F) or are detached (arrowheads in C). Scale bar = 100 μm.

Fast and slow muscle formation in popdc2 morphants. Confocal imaging of transverse sections (A,D,G,J) through the 27 hpf trunk and lateral flatmounts of (B,E,H,K) 1 dpf and (C,F,I,L) 3 dpf embryos, which were injected with (A–C,G–I) MO-control (MO-CTR) or (D–F,J–L) MO1/MO2-popdc2 morpholinos, respectively and stained (A–F) with F310 antibody to visualize fast muscle fibers, and (G–L) with F59 antibody to detect slow muscle fibers. Sections stained for slow muscle fibers (G,J) were also stained with phalloidin detecting F-actin to visualize the fast muscle compartment (green). Asterisks in (K,L) pointing to gaps in the fiber arrangement. Arrows in (F) indicate ruptured fast muscle fibers and in (L) point to slow muscle fibers with a wavy morphology. Scale bars = 50 μm.

Myotendinous junctions are altered in popdc2 morphants. Confocal imaging of the trunk musculature of 24 hpf embryos in a lateral view, anterior to the left. (A,B) FAK and (C,D) vinculin expression in (A,C) MO-control and (B,D) MO1-popdc2 injected embryos. In the popdc2 morphants the shape of the myotomal boundaries is disturbed and the expression domains of FAK and vinculin are discontinuous (arrow heads). (A′–D′) Beta-catenin staining. (A′′–D′′) Merge of panel (A–D) and (A′–D′). Scale bar = 50 μm.

Murine Popdc2 is able to rescue the morphant phenotype. (A–H) Lateral view of the (A,E) uninjected wild-type embryos (WT) as controls and embryos injected with either (B,F) 100 pg Popdc2 cRNA, (C,G) 1 ng of MO1-popdc2, or (D,H) 100 pg Popdc2 cRNA and MO1-popdc2 morpholino, which were photographed at (A,C,E,G) 24 hpf, or (B,D,F,H) 48 hpf. (I–L) Lateral confocal stacks of the zebrafish trunk labeled with phalloidin rhodamine at 48 hpf. (I) control, (J) Popdc2 cRNA, (K) MO1-popdc2, (L) Popdc2 cRNA and MO1-popdc2 morpholino. (M) Graphic depiction of the percentage of embryos with wildtype appearance (WT) (blue), or weak (pink), and severe (red) pathological phenotypes. Significant difference (p < 0.05) is marked with an asterisk. Scale bar = 100 μm.

The craniofacial skeletal muscles in popdc2 morphants exhibit an abnormal morphology. Confocal analysis of heads of 3 dpf Tg(acta:GFP) embryos, which were injected with (A,C) MO-control (MO-CTR) or MO1/MO2-popdc2 morpholinos, respectively. In the morphants craniofacial muscles are malformed and in many cases significantly reduced in size. Abbreviations: ah, adductor hyoideus; am, adductor mandibulae; ao, adductor operculi; do, dilator operculi; hh, hyohyoideus; ih, interhyoideus; ima, intermandibularis anterior; imp, intermandibularis posterior; io, inferior oblique; ir, interior rectus, lap, levator arcus palatini; mr, medial rectus; sh, sternohyoideus; so, superior oblique; tv 1–5, transversus ventralis 1–5.Scale bar = 50 μm.

Cardiac dysmorphogenesis in the popdc2 morphant. Zebrafish embryos were injected with 2 ng/embryo of (A,C,E) control morpholino (MO-CTR) or (B,D,F) MO1/MO2 popdc2 morpholinos. (A,B) Confocal analysis of 3 dpf transgenic Tg(cmlc2:GFP) hearts. The heart of the popdc2 morphants is dysmorphic and lacks trabeculation. (C,D) Anti-Alcam antibody staining reveals normal cell shape in the popdc2 morphants. (E,F) Anti-tropomyosin antibody staining shows a reduction in myofibrillar content in cardiac myocytes of the popdc2 morphants. Abbreviations: A, atrium; V, ventricle. Scale bar = 10 μm.

Popdc2 morphants display cardiac conduction defects. (A) Luminance periodogram of hearts injected with control morpholino (CTR), or MO1-popdc2 morpholino at 5 dpf. Black bars indicate 3:1 and light gray bars 2:1 atrio/ventricular rhythm. (B) Beating frequencies of popdc2 morphants and control animals. The number of animals that were analyzed at each time point is given at the base of each column. The difference in beating frequency of control and popdc2 morphants reached statistical significance (p ≤ 0.05) at 5 dpf and 7 dpf.

The hearts of popdc2 morphants display a variety of cardiac conduction defects. SPIM videos of the heart of zebrafish embryos injected with (A) control (CTR) and (B–D) MO1-popdc2 morpholino in the Tg(cmlc2:gCaMP)s878 background. The individual movies (see SI Movie 5, Movie 6, Movie 7 and Movie 8) were processed to determine the fluorescence intensity of selected regions of the heart over time. Each selected region has a corresponding number plotted below. (A) In a control (CTR), fluorescence intensity varies with time in atrial and ventricular regions as the depolarization wave propagates through the heart. (B) Heart displaying a single incidence (asterisks) of a 4:1 AV block. (C) Heart displaying alternating phases of normal conduction and 2:1 AV block (asterisks). Both hearts depicted in (B) and (C) also display irregular atrial conduction waves. (D) Example of a 6 dpf heart with sinoatrial block (asterisk).The hearts of popdc2 morphants display a variety of cardiac conduction defects. SPIM videos of the heart of zebrafish embryos injected with (A) control (CTR) and (B–D) MO1-popdc2 morpholino in the Tg(cmlc2:gCaMP)s878 background. The individual movies (see SI Movies 5–8) were processed to determine the fluorescence intensity of selected regions of the heart over time. Each selected region has a corresponding number plotted below. (A) In a control (CTR), fluorescence intensity varies with time in atrial and ventricular regions as the depolarization wave propagates through the heart. (B) Heart displaying a single incidence (asterisks) of a 4:1 AV block. (C) Heart displaying alternating phases of normal conduction and 2:1 AV block (asterisks). Both hearts depicted in (B) and (C) also display irregular atrial conduction waves. (D) Example of a 6 dpf heart with sinoatrial block (asterisk).

Morpholino-mediated knockdown of popdc2 affects muscle and heart development. (A) Genomic organization of the popdc2 gene in zebrafish. The gene consists of four exons of which exon 1–3 are coding (black box), whereas exon 4 is non-coding. The morpholino-modified oligonucleotides were directed against the splice donor (MO2) and acceptor sequences (MO1) of intron 1, respectively. Arrows depict the location of the two primer sets to amplify exon (E) and intron (I) sequences, which were utilized in the RT-PCR analysis. (B–D) RT-PCR amplification of total RNA isolated of (B) MO1-popdc2 and control morpholino-injected embryos, (CTR) respectively at 4 dpf, (C) MO2-popdc2 and control morpholino-injected embryos, respectively, at 3 dpf, and (D) MO1/MO2-popdc2 and control morpholino-injected embryos, respectively, at 2 dpf. In each case expression analysis of β-actin served as a control. The morphants displayed a significant decrease of popdc2 transcripts, and an increase in aberrantly spliced transcripts as revealed by the presence of intron 1. (E–H) Lateral view of (E) control (CTR), (F) MO1-popdc2 morphants, (G) MO2-popdc2 morphants, and (H) MO1/MO2-popdc2 morphants at 24 hpf (left) and 48 hpf (right), respectively. All morphants develop similar morphological phenotypes displaying aberrant tail morphology (arrows in F, G and H) and signs of embryonic heart failure as evidenced by blood retention and pericardial edema (arrowheads in F,G,H).

Whole mount in situ hybridization analysis of myoD expression in (A, A′) control (MO-CTR) and (B, B′) MO1-popdc2 morphants at the 16-somite stage. The orientation of the embryos depicted is lateral in panel A, B and dorsal in A2, B2. No differences in the segmentation pattern are seen between control and MO1-popdc2 morphants.

Muscle fiber alignment defect in popdc2 morphants. Confocal imaging of lateral flat mounts of the trunk musculature of 3 dpf old (A) MO-CTR and (B) MO1/MO2-popdc2 injected embryos, which were stained with an α-actinin antibody. The morphant displays severe myofiber alignment defects, which are likely to cause uncoordinated swimming behavior as has been observed in the morphants. Scale bars = 200 μm.

The early histogenesis of trunk skeletal muscle is not affected in the popdc2 morphant. (A, E, C, G) MO-control and (B, F, D, H) MO1-popdc2 morpholino injected embryos subjected to confocal immunostaining (anterior to the left, lateral view) with (A–D) vinculin and (E–H) focal adhesion kinase (FAK), both labeled in green and beta-catenin (red label). Scale bar = 50 μm.

Aberrant craniofacial skeletal muscle development in popdc2 morphants. Confocal microscopy of whole mount immunostained embryos using MF20 antibody. Embryos were injected with (A, C, E, G) control (MO-CTR), or (B, D, F, H) MO1-popdc2 morpholinos, respectively. Specimens are viewed from (A–D) the left lateral side, or are (E–H) ventral views. Many of the craniofacial muscles were malformed and reduced in size in the popdc2 morphants. Abbreviations: ah, adductor hyoideus; am, adductor mandibulae; ao, adductor operculi; do, dilator operculi; dpw1–5, dorsal pharyngeal wall 1–5; hh, hyohyoideus; ih, interhyoideus; ima, intermandibularis anterior; imp, intermandibularis posterior; io, inferior oblique; ir, interior rectus; lap, levator arcus palatini; lr, lateral rectus; mr, medial rectus; sh, sternohyoideus; so, superior oblique; sr, superior rectus; tv 1–5, transversus ventralis 1–5; v, ventricle.

Craniofacial bone and cartilage formation is not affected in popdc2 morphants. (A,B,C,D) Left lateral and (A2,B2,C2,D2) ventral views of (A,A′,B,B′) MO-CTR and (C,C′,D,D′) MO1-popdc2 injected embryos, which were stained for bone (alizarin red) and cartilage (alcian blue) at (A,A′,C,C′) 4 dpf and (B,B′,D,D′) 5 dpf respectively. At 4 dpf, popdc2 morphants (n = 12/13) display normal pattern of cartilage and bone formation as the control (n = 17). At 5 dpf, most of the morphants (32/39) showed a normal distribution of bone and cartilage elements of MO1-popdc2 when compared to the controls (n = 24).

Cardiac chamber dimensions and morphology of MO1-popdc2 morphants. Confocal analysis of 5 dpf (A–H) Tg(cmlc2:eGFP-ras)^s883, and (I–L) Tg(flk1:eGFP)^s843 hearts, which were injected with 1 ng (A,B,E,F,I,J) control (CTR), or (C,D,G,H,K,L) MO1-popdc2 morpholinos (MO1), respectively. (A,C) 3D visualization of atrial and ventricular chamber dimensions using confocal stacks. (B,D) Single images through the ventricular wall revealing normal trabeculation in popdc2 morphants displaying arrhythmia. The frames in panels (E,G,I,K) demarcate the area of magnification depicted in (F,H,J,L), respectively. No differences in morphology of the (E–H) AV myocardium or the (I–L) developing valve were observed between control and MO1-popdc2 morphants with rhythm defects. Abbreviations: a—atrium, v—ventricle. Scale bar = 100 μm.

Expression of marker genes for cardiac chamber and atrioventricular canal development are unchanged in popdc2 morphants. Whole mount in situ hybridization analysis of (A,B) cmlc2, (C,D) vmhc, (E,F) bmp4, and (G,H) notch1b expression in (A,C,E,G) control (CTR) and (B,D,F,H) MO1-popdc2 (MO1) morphants at (A–D) 72 hpf and (E–H) 54 hpf, respectively. (A–D) No obvious differences in chamber dimensions were observed between control and MO1-popdc2 morphants. Note the lack of vmhc-stained facial muscles (fm) in the MO1-popdc2 morphant (arrowheads in D). (E–H) Expression of the myocardial atrioventricular canal (AVC) marker gene bmp4 is observed in the inflow tract, AVC and outflow tract and no difference was observed between control embryos and popdc2 morphants. (G,H) Likewise, the endocardial marker notch1b is expressed in the outflow tract and in the AVC with similar intensity in MO-control and MO1-popdc2 injected embryos, respectively. Abbreviations: a—atrium; v—ventricle.

Morphological analysis of the heart in the popdc2 morphant. (A, B) Methylene blue/Azur II-stained transversal semi-thin sections and (C, D) transmission electron micrographs of the heart of wild type (WT) and MO1-popdc2 morphants (MO1). No differences in the overall histology and in the ultrastructure were observed. Frames in panels (A) and (B) demarcate approximately the area depicted in panels (C) and (D).

Efficient knockdown activity of MO1/MO2-popdc2 morpholinos at 4 and 5 dpf. RT-PCR analysis of total RNA isolated of MO1/MO2-popdc2 and wild type embryos (WT) at 4 and 5 dpf. At both time points, MO1/MO2-popdc2 morphants display a significant decrease in popdc2 transcripts. These transcripts are aberrantly spliced and do not code for a functional protein as revealed by intron 1-specific PCR amplification. In each case expression analysis of β-actin served as a control.

Unillustrated author statements