Two myristoylated alanine-rich C-kinase substrate (MARCKS) paralogs are required for normal development in zebrafish

Abstract

Myristoylated alanine-rich C-kinase substrate (MARCKS) is an actin binding protein substrate of protein kinase C (PKC) and critical for mouse and Xenopus development. Herein two MARCKS paralogs, marcksa and marcksb, are identified in zebrafish and the role of these genes in zebrafish development is evaluated. Morpholino-based targeting of either MARCKS protein resulted in increased mortality and a range of gross phenotypic abnormalities. Phenotypic abnormalities were classified as mild, moderate or severe, which is characterized by a slight curve of a full-length tail, a severe curve or twist of a full-length tail and a truncated tail, respectively. All three phenotypes displayed abnormal neural architecture. Histopathology of Marcks targeted embryos revealed abnormalities in retinal layering, gill formation and skeletal muscle morphology. These results demonstrate that Marcksa and Marcksb are required for normal zebrafish development and suggest that zebrafish are a suitable model to further study MARCKS function.

Keywords: MARCKS; development; zebrafish.

Figure 1

Zebrafish Marcksa and Marcksb are similar to other vertebrate MARCKS sequences. (A) Zebrafish Marcksa and Marcksb protein sequences are aligned with MARCKS sequences from human, mouse, chicken and Xenopus. Black shading indicates identical amino acid alignment; gray shading indicates functionally similar amino acids. The three conserved domains of MARCKS are indicated: the myristoylated amino-terminus, the MH-2 domain and the phospho-site domain (PSD). The three serine residues phosphorylated by PKC are identified by an asterisk (*) and Ser²⁵, which is phosphorylated in neural tissue, is denoted by a caret (^). The pound sign (#) denotes a candidate serine in Marcksb that may also be phosphorylated given its proximity to Ser²⁵ in MARCKS expressed by other species. (B) Phylogenetic comparison of MARCKS and MARCKS-like protein sequences from vertebrate species. Branch lengths are measured in terms of amino acid substitutions with the scale indicated below the tree. Hs = human, Homo sapiens; Mm = mouse, Mus musculus; Dr = zebrafish, Danio rerio; Ss = salmon, Salmo salar; Gg = chicken, Gallus gallus; Xl = frog, Xenopus laevis. (C) Zebrafish marcksa and marcksb loci both share conserved synteny with the MARCKS locus in human and mice. Genes flanking marcksa on chromosome 20 and marcksb on chromosome 17 were determined and the chromosomal location of orthologous genes in humans and mice were determined. Gray shaded areas indicate conserved synteny.

Figure 2

Zebrafish marcksa and marcksb are expressed throughout early development. Reverse transcriptase PCR was performed to detect zebrafish marcksa, marcksb and β-actin transcripts from embryo derived cDNAs. Age of embryos is indicated above each lane in hours post fertilization (hpf). Data is representative of 3 separate experiments and NT indicates no template control.

Figure 3

Phenotypic characterization of Marcksa and Marcksb targeted zebrafish morphants. Representative examples of normal, mild, moderate and severe phenotypes are depicted at 24 hpf (A), 48 hpf (B), 72 hpf (C) and 96 hpf (D).

Figure 4

Phenotypic quantification of zebrafish Marcksa targeted morphants. Embryos were either not injected (WT) or injected at the single cell stage with 0.5, 2 or 4 ng MAT (A–D), 2, 4, or 6 ng MATU (E–H) or respective dose of control MO (CONT). Phenotypic characterization was quantified at 24 hpf (A, E), 48 hpf (B, F), 72 hpf (C, G) and 96 hpf (D, H) and phenotypes were classified into five phenotypes: normal, mild, moderate, severe and dead. The data is pooled from three separate experiments with a total of 120 embryos injected in the MAT and MATU groups and 110 embryos injected in the control injected or WT groups. Data is represented as the percentage of injected embryos.

Figure 5

Phenotypic quantification of zebrafish Marcksb targeted morphants. Embryos were either not injected (WT) or injected at the single cell stage with 2, 4, or 6 ng MBT (A–D), MBS (E–H) or control MO (A–H). Phenotypic characterization was quantified at 24 hpf (A, E), 48 hpf (B, F), 72 hpf (C, G) and 96 hpf (D, H) and phenotypes were classified into five phenotypes: normal, mild, moderate, severe and dead. The data is pooled from three separate experiments with a total of 120 embryos injected in the MBT and MBS groups and 110 embryos injected in the control injected or WT groups. Data is represented as the percentage of injected embryos.

Figure 6

Marcksa and Marcksb are involved in normal zebrafish neural development. Embryos were co-injected in the single cell stage with MAT or MBT MOs along with p53 or control MOs. Embryos were analyzed at 24 hpf for the presence of the neural dead phenotype. Representative embryos are shown; wild type, non-injected (A), 4 ng p53 MO + 2 ng control MO (B), 2 ng MBT + 4 ng p53 MO (C), 2 ng MAT + 4 ng p53 MO (D), 2 ng MBT + 4 ng control MO (E), 2 ng MAT + 4 ng control MO (F).

Figure 7

Marcks is involved in retinal histogenesis of zebrafish. (A) Normal eye from 72 hpf wild type zebrafish embryo. Here, the layers of the retina are relatively well defined. Starting from the outermost layer and working toward the center of the eye: 1) the dark brown/black retinal pigment epithelium; 2) layer of rods and cones; 3) outer nuclear layer (dark blue/purple); 4) outer plexiform layer; 5) inner nuclear layer; 6) inner plexiform layer. The ganglion cell and nerve fiber layers are less distinct in this section. The plexiform layers consist mainly of synapses of the various sensory neural cells. (B) Eye from 72 hpf MAT injected zebrafish embryo (moderate phenotype). The retina consists of a mass of poorly organized neural cell nuclei with no clear separation into normal retinal layers. Only the retinal pigment epithelium forms a distinct layer (1). This morphology is consistent in both the MAT and MBT injected fish, and is most severe in the severe phenotype fish (data not shown).

Figure 8

Marcks targeting zebrafish results in abnormal gill development and muscle cell morphology. (A) Normal gill tissue from 72 hpf wild type zebrafish embryo. The primary lamellae (filaments) of the developing gill are shown in cross section, with normal supporting chondrocytes (arrow heads) at the center and a covering of squamous epithelial cells (arrow). (B) Gill tissue (G) from 72 hpf MAT injected zebrafish embryo (moderate phenotype). In this section, the gill is comprised of a mass of numerous, poorly organized epithelial cells with no clear separation into primary lamellae and no discernable central supporting chrondrocytes. This morphology is consistent in both MAT and MBT injected fish, and is most severe in the severe phenotype fish (data not shown). (C) Skeletal muscle fibers (arrow) from the tail of 72 hpf control MO injected zebrafish are relatively straight, with thin, regularly spaced nuclei. (D) Skeletal muscle from the tail of 72 hpf MAT injected zebrafish (moderate phenotype) contains numerous curved/crescent-shaped fibers (arrow) and plumper, more numerous nuclei as compared to control and wild type fish. This morphology is consistent in both the MAT and MBT fish, and is most severe in the severe phenotype fish (data not shown).