Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 146 (13)

Neuropilin 1 Mediates Epicardial Activation and Revascularization in the Regenerating Zebrafish Heart

Affiliations

Neuropilin 1 Mediates Epicardial Activation and Revascularization in the Regenerating Zebrafish Heart

Vanessa Lowe et al. Development.

Abstract

Unlike adult mammals, zebrafish can regenerate their heart. A key mechanism for regeneration is the activation of the epicardium, leading to the establishment of a supporting scaffold for new cardiomyocytes, angiogenesis and cytokine secretion. Neuropilins are co-receptors that mediate signaling of kinase receptors for cytokines with crucial roles in zebrafish heart regeneration. We investigated the role of neuropilins in response to cardiac injury and heart regeneration. All four neuropilin isoforms (nrp1a, nrp1b, nrp2a and nrp2b) were upregulated by the activated epicardium and an nrp1a-knockout mutant showed a significant delay in heart regeneration and displayed persistent collagen deposition. The regenerating hearts of nrp1a mutants were less vascularized, and epicardial-derived cell migration and re-expression of the developmental gene wt1b was impaired. Moreover, cryoinjury-induced activation and migration of epicardial cells in heart explants were reduced in nrp1a mutants. These results identify a key role for Nrp1 in zebrafish heart regeneration, mediated through epicardial activation, migration and revascularization.

Keywords: Epicardium; Heart; Neuropilin; Regeneration; Zebrafish.

Conflict of interest statement

Competing interestsThe authors declare no competing or financial interests.

Figures

Fig. 1.
Fig. 1.
Nrps are upregulated during zebrafish heart regeneration. (A) Absolute qPCR analysis of nrp family genes at 1, 3, 7, 14, 30 and 60 days following cryoinjury or sham surgery. Basal expression was evaluated in uninjured hearts of age-matched wild-type fish. Bars represent normalized copy number per reaction. Data are mean±s.e.m. **P<0.01, ***P<0.005, ****P<0.001 (one-way ANOVA with Sidak's post hoc test for multiple comparisons of n=4 or 5 with each n being a pool of five ventricles). (B) Adult zebrafish ventricle lysates 1, 3, 7, 14 and 30 days following surgery, immunoblotted for Nrp1 and Gapdh (left); western blot quantification of Nrp1 protein in sham and cryoinjured ventricles 1, 3, 7, 14 and 30 days following surgery (right) (n=4 or 5, with each n being a pool of three ventricles). (C) In situ hybridization with digoxigenin-labeled antisense riboprobes were used to detect nrp family isoforms in sham-operated and cryoinjured adult zebrafish hearts 1, 3 and 14 dpci (n≥3). Arrowheads indicate gene expression within the epicardium. CI, cryoinjured; epi, epicardium; IA, injured area. Scale bars: 250 µm.
Fig. 2.
Fig. 2.
Nrp1 is expressed by the endocardium and the epicardium in sham and cryoinjured hearts. (A-C) Immunostaining of 7 days post sham-operated (upper rows) and cryoinjured (lower rows) hearts. Tg(fli1a:EGFP)y1 (A), wild-type fish immunostained for tropomyosin (B) and tg(wt1b:EGFP)li1 (C) zebrafish hearts were used to identify endothelium, myocardium and activated epicardium, respectively (n≥3). The merged images are displayed with DAPI nuclei staining. Dashed boxes denote the location of enlarged images (right panels); dashed lines indicate the injury interface. Scale bars: 100 µm; 20 µm in enlarged images.
Fig. 3.
Fig. 3.
Characterization of nrp1asa1485 mutant fish. (A) Structure of Nrp1a in wild-type (nrp1a+/+) (left) and nrp1asa1485 mutant fish (right). The point mutation results in the generation of a premature stop codon at aa 206, resulting in a truncated Nrp1a fragment. Blue diamonds indicate CUB (a1 and a2) domains, orange circles indicate FA58C (b1 and b2) domains, the green square indicates the MAM domain and the brown square indicates the C-terminal domain. (B) Sequencing chromatograms of wild-type fish, heterozygous nrp1asa1485/+ and homozygous nrp1asa1485/sa1485 mutant fish. An early stop codon (nonsense mutation) TAA, replaces the wild-type TAC codon at aa 206. The genotypes of 14 zebrafish embryos 48 h post fertilization (hpf) were compared against the expected Mendelian ratio after heterozygous fish incross. (C) Absolute RT-qPCR of wild-type (WT; black circles, white bar) or nrp1asa1485 homozygous mutant (black squares, gray bar) uninjured adult zebrafish hearts under basal conditions. nrp1a expression was significantly decreased in nrp1asa1485 samples, suggesting nonsense-mediating decay. ****P<0.0001 (two-tailed t-test; n=7 with each n being a pool of three ventricles). Data are means of normalized copy numbers per reaction±s.e.m. (D) Nrp1a antisense (AS) in situ hybridization of wild-type (upper row) or nrp1asa1485 homozygous mutant (lower row) embryos 24 hpf. Nrp1a expression was clearly decreased in nrp1asa1485 samples. Scale bars: 50 μm. (E) Western blot of wild-type (WT) or nrp1asa1485 homozygous mutant uninjured adult zebrafish ventricle lysates (top). Lysates were immunoblotted with an antibody targeting the Nrp1 cytoplasmic domain and Gapdh. Note the absence of C terminus detection of Nrp1a in the nrp1asa1485 samples. Western blot quantification of Nrp1a (plain bars, circles for wild type, squares for nrp1asa1485) and Nrp1b (striated bars, upward triangles for wild type, downward triangles for nrp1asa1485) normalized to Gapdh (one-way ANOVA with Sidak's post hoc test for multiple comparisons of n=4), confirming the significant reduction in Nrp1a expression (****P<0.0001; plain white bar for wild type versus plain gray bar for nrp1asa1485), whereas Nrp1b was not significantly different between wild type (white striated bar) and nrp1asa1485 (gray striated bar) (P=0.219) (bottom).
Fig. 4.
Fig. 4.
Cardiac regeneration is delayed in nrp1asa1485 mutants following cryoinjury. (A) Heart sections from wild-type (top) and nrp1asa1485 mutant (bottom) fish obtained at 1, 3, 7, 14, 30 and 60 dpci and stained with AFOG to identify the injured region. Dashed lines indicate the interface between cryoinjured and healthy tissue. (B) Cryoinjured areas were measured and represented as mean percentage of total ventricle area±s.e.m. P<0.05 (two-way ANOVA with Sidak's post hoc test for multiple comparisons of n=4-8). (C) A closed wound is one in which compact myocardium recovers following cryoinjury and encases the scar tissue; by contrast, a scar exposed to the surface is defined as an open wound. Wound closure was examined in wild-type (WT) and nrp1asa1485 mutant hearts at 30 and 60 dpci and open versus closed wounds were expressed as a percentage of the total number of hearts (n=4-8). (D,E) AFOG staining of wild-type and nrp1asa1485 mutant hearts at 3 dpci (D) used to evaluate epicardial thickness and injury boundaries to calculate the epicardial area normalized to the length of the injury boundary (continuous line), quantified in E. Dashed line represents outer boundary of epicardial area. *P<0.05 (two-tailed t-test of n=7 for wild-type and n=10 for nrp1asa1485 hearts). A, atrium; ba, bulbus arteriosus; V, ventricle.
Fig. 5.
Fig. 5.
Neovascularization of the cryoinjured area is impaired in nrp1asa1485 mutants. (A,B) Blood vessels in either wild-type (A) or nrp1asa1485 (B) tg(fli1a:EGFP)y1 zebrafish at 1 (upper row) and 3 (lower row) dpci were identified in heart sections using GFP immunofluorescence in vascular structures. Heart sections were also counterstained with DAPI. Smaller images (left) represent DAPI staining (blue) and GFP staining (vessels, green) only; larger images (right) are the merged images. The white dashed line delineates the border of the area of injury. White arrowheads indicate blood vessels. (C,D) GFP-positive vessels were quantified at 1 dpci (C; **P<0.01, two-tailed t-test of n=3 and 4) and 3 dpci (D; *P<0.05, two-tailed t-test of n=4) for wild-type (white bars) versus nrp1asa1485 (gray bars) hearts. Individual data points (circles for wild type and squares for nrp1asa1485) represent individual hearts, each averaged from vessel counts in three to four different sections covering the injury site.
Fig. 6.
Fig. 6.
Epicardial activation is decreased in nrp1asa1485 hearts following cryoinjury. (A-D) Wild-type and nrp1asa1485 mutant cryoinjured fish on the tg(wt1b:EGFP)li1 background were analyzed for epicardial activation at 3 dpci by identification of wt1b:EGFP-positive cells (A,C), and were also stained with DAPI (blue; A,C) and anti-PCNA antibodies (red; C), as indicated. In A, smaller images represent DAPI staining (blue) and GFP staining (activated epicardium, green) only; larger images are the merged images. In C, boxed area in left panel is enlarged in the middle. Smaller images (right) represent PCNA staining (red) and GFP staining (activated epicardium, green) only; larger images in the middle are the merged images. (B) Quantification of percentages of wt1b:EGFP-positive cells adjacent to the area of cryoinjury (indicated by the dashed line in A,C) in wild-type (white bar) and nrp1asa1485 (gray bar) mutant fish. **P<0.01 (two-tailed t-test of n=5). (D) Quantification of percentages of cells positive for wt1b:EGFP and PCNA in the area of cryoinjury. P>0.05 [two-tailed t-test of n=6 and 5 for wild-type (white bar) and nrp1asa1485 (gray bar), respectively]. Individual data points (circles for wild type and squares for nrp1asa1485) represent percentages in individual hearts, each averaged from counts in two to four different sections covering the injury borders. epi, epicardium; IA, injury area; ns, not significant. White-dotted line delineates injury–epicardial border.
Fig. 7.
Fig. 7.
Aldh1a2 expression is unchanged in nrp1asa1485 fish following cryoinjury. (A-D) β-Catenin expression is downregulated in rat epicardial cells. (A) Cryoinjured wild-type and nrp1asa1485 fish were analyzed for Aldh1a2 expression and were also stained with DAPI and anti-PCNA antibody at 3 dpci. (B) Percentage of Aldh1a2-positive cells in the endocardium of the cryoinjury area (left) and percentage of proliferating endocardial cells (PCNA+ and Aldh1a2+ cells) (right). P=0.43 for Aldh1a2+ cells and P=0.59 for PCNA+ Aldh1a2+ cells (two-tailed t-test of n=5). (C) Absolute qPCR analysis of aldh1a2 expression was performed on uninjured and 3 dpci ventricles from wild-type and nrp1asa1485 mutant fish. Bars represent normalized copy number per reaction mean±s.e.m. (one-way ANOVA with Sidak's post hoc test for multiple comparisons of n≥3). (D) Rat epicardial cells were cultured in vitro and infected with control (Ad.LacZ) and shRNA NRP1 (Ad.shNRP1) adenoviral constructs. Cell lysates immunoblotted for NRP1, β-catenin and β-actin (n=2). ns, not significant. Scale bars: 20 µm.
Fig. 8.
Fig. 8.
Epicardial cryoinjury-induced expansion and activation are impaired in nrp1asa1485 mutants. (A-D) The apices of wild-type and nrp1asa1485 zebrafish ventricles were collected 5 days post sham surgery or cryoinjury and cultured on fibrin gels for 7 days. (A) Epicardial cells migrated into the fibrin gels (dotted black lines). (B) Epicardial outgrowths were measured for each condition (sh, sham-operated and CI, cryoinjured hearts) after 7 days culture. Data are mean outgrowth area (mm2)±s.e.m. ***P<0.001 (one-way ANOVA with Sidak's post hoc test for multiple comparisons of n>9). (C) Epicardial explant recovered from wild-type and nrp1asa1485 tg(wt1b:EGFP)li1 cryoinjured fish at 5 dpci were left to grow on fibrin gels for 7 days and stained for GFP. GFP fluorescence was observed at 10× (left column) and 40× magnification at the center and the periphery of the explants (middle and right columns, respectively). (D) Cell size (left) and ploidy (right) were quantified both at the center (top row) and at the edge (bottom row) of the explant. Data are expressed as percentage of cells per field of view ±s.e.m. Each n represents an average of 3 fields of view per explant (two-tailed t-test of n≥5); *P<0.05. ns, not significant. Scale bars: 1 mm in A, 20 µm in C (right); 100 µm in C (left).

Similar articles

See all similar articles

Cited by 1 article

References

    1. Adham S. A. I., Al Harrasi I., Al Haddabi I., Al Rashdi A., Al Sinawi S., Al Maniri A., Ba-Omar T. and Coomber B. L. (2014). Immunohistological insight into the correlation between neuropilin-1 and epithelial-mesenchymal transition markers in epithelial ovarian cancer. J. Histochem. Cytochem. 62, 619-631. 10.1369/0022155414538821 - DOI - PubMed
    1. Ball S. G., Bayley C., Shuttleworth C. A. and Kielty C. M. (2010). Neuropilin-1 regulates platelet-derived growth factor receptor signalling in mesenchymal stem cells. Biochem. J. 427, 29-40. 10.1042/BJ20091512 - DOI - PMC - PubMed
    1. Cahill T. J. and Kharbanda R. K. (2017). Heart failure after myocardial infarction in the era of primary percutaneous coronary intervention: mechanisms, incidence and identification of patients at risk. World J. Cardiol. 9, 407-415. 10.4330/wjc.v9.i5.407 - DOI - PMC - PubMed
    1. Cao Y. and Cao J. (2018). Covering and re-covering the heart: development and regeneration of the epicardium. J. Cardiovasc. Dev. Dis. 6, 3 10.3390/jcdd6010003 - DOI - PMC - PubMed
    1. Cao J. and Poss K. D. (2018). The epicardium as a hub for heart regeneration. Nat. Rev. Cardiol. 15, 631-647. 10.1038/s41569-018-0046-4 - DOI - PMC - PubMed

LinkOut - more resources

Feedback