A central function for perlecan in skeletal muscle and cardiovascular development

Abstract

Perlecan's developmental functions are difficult to dissect in placental animals because perlecan disruption is embryonic lethal. In contrast to mammals, cardiovascular function is not essential for early zebrafish development because the embryos obtain adequate oxygen by diffusion. In this study, we use targeted protein depletion coupled with protein-based rescue experiments to investigate the involvement of perlecan and its C-terminal domain V/endorepellin in zebrafish development. The perlecan morphants show a severe myopathy characterized by abnormal actin filament orientation and disorganized sarcomeres, suggesting an involvement of perlecan in myopathies. In the perlecan morphants, primary intersegmental vessel sprouts, which develop through angiogenesis, fail to extend and show reduced protrusive activity. Live videomicroscopy confirms the abnormal swimming pattern caused by the myopathy and anomalous head and trunk vessel circulation. The phenotype is partially rescued by microinjection of human perlecan or endorepellin. These findings indicate that perlecan is essential for the integrity of somitic muscle and developmental angiogenesis and that endorepellin mediates most of these biological activities.

Figure 1.

Analysis of zebrafish perlecan. (A) Schematic of human and zebrafish perlecan protein core. The roman numerals indicate the five domains. A key for the various modules is presented at the bottom. Arrows indicate the perlecan MO-DI, MO-DIII, and MO-DV morpholino targeting sites. See Fig. S1 for additional information (available at http://www.jcb.org/cgi/content/full/jcb.200708022/DC1). (B) Zebrafish perlecan is conserved syntenically. The zebrafish perlecan gene (hspg2) maintains a syntenic relationship with DDOST, PHC2, and SARS of mammalian species. Species and respective chromosome number are depicted; gene symbols are followed by the gene identification numbers from the National Center for Biotechnology Information Entrez Gene database, and distances between genes are displayed in megabases.

Figure 2.

Spatiotemporal expression patterns of zebrafish perlecan mRNA and protein. (A–G) Whole mount ISH with a digoxigenin-labeled perlecan domain V antisense probe for the localization of perlecan mRNA (blue/purple staining). Perlecan mRNA can be detected at the 20-somite stage in regions of the developing brain (A) and along both sides of the notochord (B, arrows). At 1 dpf, perlecan expression is seen in the head region (C), in the developing somites (C, black arrows), and within the developing axial vasculature (D, arrows). Also note perlecan expression localized within the duct of Cuvier (C, red arrows). By 5 dpf (F and G [magnified image of the region boxed in F]), perlecan expression significantly increases throughout the trunk musculature (G, arrows), the major trunk/tail vessels including the DLAV, DA, and PCV, and in the developing gastrointestinal tract and future fin regions. (E) ISH with a digoxigenin-labeled perlecan domain V sense probe. ISH was performed in groups of two to five samples, and representative images are shown. (H–N) Whole mount immunohistochemistry with an affinity-purified anti–mouse perlecan antibody. Perlecan can be detected as early as the 64- and 1,000-cell stage of embryonic development (2 and 3 hpf, respectively) throughout the cell mound (I and J). At 1 dpf, perlecan protein (L and M) is detected throughout the head (L, arrow), trunk, and tail and is specifically localized to the developing muscle myoseptae (M, arrows) and the developing vasculature. By 2 dpf (N), perlecan protein is specifically detected in the trunk vasculature, including the DLAV (N, black arrows) and axial vessels (N, white arrows). (H and K) Immunohistochemistry images in which the primary antibody was omitted. Bars, 500 μm.

Figure 3.

Classification and verification of the perlecan morphant phenotype. (A) All observed perlecan morphant phenotypes can be classified according to the degree of body twisting as presented for MO-DI. Categories include mild (least striking twisted body but noted tail phenotype), moderate (general twisted body usually accompanied by curly tail up), and severe (significant body plan shortening accompanied by twisting of the tail). (B) Mean observed frequencies of the twisted body classes from MO-DI perlecan morphant embryos (n = 232). Error bars represent ±SEM. (C) Phenotypic overview of perlecan splice junction–blocking MO-DIII. (D) RT-PCR verification of domain III splice junction–blocking morpholino effect. Note the band shift in morpholino-injected embryos (lane 5) versus uninjected embryos (lane 4), verifying the splice-blocking/intron retaining effect of MO-DIII. Lane 1, DNA ladder; lane 2, domain III PCR from zebrafish cDNA template; lane 3, domain III PCR from genomic DNA template; lane 4, domain III PCR from uninjected embryos' cDNA template; lane 5, domain III PCR from MO-DIII–injected embryos' cDNA template. The bottom bands in lanes 4 and 5 represent the β-actin control. Template from lanes 4 and 5 were derived from total RNA isolated from 23 embryos. See Fig. 1 A for additional details regarding the targeting positions of the morpholinos. (E–G) Whole mount immunohistochemistry for verification of domain III morpholino-based knockdown of perlecan. (E) Control uninjected embryo (2 dpf) shows perlecan expression throughout the trunk musculature and vasculature. (F and G) MO-DIII embryos (2 dpf) show significantly reduced perlecan protein levels in the trunk, with only minimal staining detected in the head and tail. A and E–G are left-side views with dorsal up and anterior to the left. Bars, 500 μm.

Figure 4.

Ultrastructural analysis of perlecan morphant embryos. (A and B) Parasagittal epon sections stained with toluidine blue from 4 dpf control (A) and MO-DV morphant embryos (B). Note the irregular structure and organization of the morphant's skeletal muscle and associated abnormal u-shaped myoseptal boundaries. The DA appears collapsed in the morphant and devoid of blood cells. NC, notochord; DA, dorsal aorta; PCV, posterior cardinal vein. (C–I) A comparison of morphant and control basement membrane structure. Electron microscopy of the vascular (C–E), epithelial (F and G), and notochord basement membranes (BM; H and I) suggest that morpholino-based perlecan knockdown does not compromise the integrity of the basement membrane. Bars: (A and B) 25 μm; (C–I) 1 μm.

Figure 5.

Perlecan morphants display a complex muscular dystrophy phenotype. (A, D, and F) Ultrastructural analysis of control skeletal muscle from a 5-dpf embryo. Parasagittal (A) and cross sections (D and F) of trunk muscle show typical muscular architecture with alignment of z bands, abundant glycogen (Gly) and mitochondria (Mi), and typical hexagonal arrays of thick and thin filaments (D). High magnification view shows thick filaments surrounded by six thin filaments (circle in F). In contrast, perlecan morphants induced by either translation-blocking (B) or splice-blocking (C, E, and G) morpholinos show disarray of muscular architecture with loss of filaments and irregular filaments traversing the sarcomeres at variable angles (arrows in B and C). Also notice the presence of areas with reduced thin filaments (asterisks in E) adjacent to more normal-appearing hexagonal structures (E). The sarcoplasmic reticulum (SR) appears to be normal (E). High magnification of the area labeled by the asterisk in E shows abnormal arrangement of thin and thick filaments (circle in G). Bars, 1 μm.

Figure 6.

Muscular analysis in perlecan morphant embryos. (A and C) Filamentous actin confocal immunohistochemistry. Arows in C highlight clear spaces between the muscle fibers. (B and D) Corresponding DIC analysis of the trunk musculature from a control and morphant embryo at 5 dpf. (E–J) Birefringence analyses under polarizing light comparing control and morphant trunk muscle at 2–3 dpf. F, H, and J are magnified views of the boxed regions in E, G, and I, respectively. Arrows in H indicate regions of hypobirefringence. (K–N) Binding of fluorescently labeled α-bungarotoxin and fasciculin to 4-dpf control and morphant embryos as a means to examine the distribution of AChR (red; highlighted by arrowheads in K and N) and AChE (green; highlighted by arrows in L and M). Bars, 300 μm.

Figure 7.

Vascular analysis in Tg(fli1:egfp)^y1 perlecan morphant embryos. (A and B) Epifluorescence microscopy with 3D deconvolution comparing the trunk vasculature of 2-dpf control versus MO-DI embryos. Note the correct formation of DLAV, ISV, DA, and PCV in the control (A). In contrast, the perlecan morphant (B) exhibits abnormal ISVs. Often the ISV sprouts fail to completely migrate along the myoseptae and to anastomose (arrows). (C and D) Merged images corresponding to those shown in A and B taken with both fluorescence deconvolution and DIC microscopy. Note that the ISVs barely reach the notochord (NC) and do not properly migrate to the dorsal region, thereby failing to interconnect and properly form the DLAV. (E and F) Lateral and dorsal views of alkaline phosphatase–stained control embryos at 3 dpf. Notice the well-developed SIV. (G–I) Three representative MO-DI perlecan morphants showing the complete absence of SIV and reduced alkaline phosphatase staining throughout the trunk vasculature and head region. Notice that J is a higher magnification of G. The only detectable signal is present along the axial vessels (arrows in G and J). Bars, 500 μm.

Figure 8.

Partial rescue of the perlecan morphants by human perlecan and endorepellin. (A) Coomassie blue–stained 3–8% Tris-borate gradient gel of human perlecan immunopurified from human coronary artery endothelial cells (left). Note the large and broad band corresponding to M _r of >500 kD. The right panel shows a Coomassie blue–stained 10% SDS-PAGE of human endorepellin. (B and C) Composite photographs of control, MO-DI, and MO-DI coinjected with perlecan or endorepellin at the designated concentrations. The rescue phenotype was calculated based on the degree of body twisting as compared with the microinjection of MO-DI alone. Notice the partial rescue of the twisted body phenotype in both cases. Bar, 500 μm.