The zebrafish reveals dependence of the mast cell lineage on Notch signaling in vivo

Abstract

We used the opportunities afforded by the zebrafish to determine upstream pathways regulating mast cell development in vivo and identify their cellular origin. Colocalization studies demonstrated zebrafish notch receptor expression in cells expressing carboxypeptidase A5 (cpa5), a zebrafish mast cell-specific marker. Inhibition of the Notch pathway resulted in decreased cpa5 expression in mindbomb mutants and wild-type embryos treated with the γ-secretase inhibitor, Compound E. A series of morpholino knockdown studies specifically identified notch1b and gata2 as the critical factors regulating mast cell fate. Moreover, hsp70::GAL4;UAS::nicd1a transgenic embryos overexpressing an activated form of notch1, nicd1a, displayed increased cpa5, gata2, and pu.1 expression. This increase in cpa5 expression could be reversed and reduced below baseline levels in a dose-dependent manner using Compound E. Finally, evidence that cpa5 expression colocalizes with lmo2 in the absence of hematopoietic stem cells revealed that definitive mast cells initially delineate from erythromyeloid progenitors. These studies identify a master role for Notch signaling in vertebrate mast cell development and establish developmental origins of this lineage. Moreover, these findings postulate targeting the Notch pathway as a therapeutic strategy in mast cell diseases.

Figure 1

The zebrafish notch signaling mutant mindbomb (mib^ta52b) displays decreased cpa5 expression. WISH using a digoxigenin-labeled RNA antisense probe to zebrafish cpa5 was performed at 28 and 48 hpf. Bar graphs represent the percent of embryos categorized by WISH expression levels. At 28 hpf, the mib^ta52b mutant (n = 107) shows a dramatic decrease in cpa5 expression compared with wild-type embryos (n = 96), whereas at 48 hpf, the mib^ta52b mutant (n = 19) displays reduced cpa5 expression in the ALM and absent cpa5 expression in the CHT compared with wild-type (n = 12). Insets: Magnified view of the area of interest (5× objective, MZ6 microscope; Leica).

Figure 2

Notch pathway inhibition of wild-type embryos by the γ-secretase inhibitor CpdE similarly displays decreased cpa5 expression. WISH using a digoxigenin-labeled RNA antisense probe to zebrafish cpa5 was performed at 30 and 48 hpf. Bar graphs represent the percent of embryos categorized by WISH expression levels. (A) Treatment with 50μM CpdE results in a decrease in cpa5 expression at 30 hpf (n = 35), whereas treatment with 75μM CpdE shows absent cpa5 expression (n = 10) compared with untreated controls (n = 44). Similarly, at 48 hpf, treatment with CpdE shows a dose-dependent decrease in cpa5 expression (n = 25 for 50μM CpdE, n = 30 for 75μM CpdE, n = 24 for wild-type embryos). DMSO 0.5% was used as a control, and from this group 24 and 23 embryos were analyzed at 30 and 48 hpf, respectively. (B) Treatment with 25 or 50μM CpdE results in a dose-dependent decrease of cpa5 expression in mib^ta52b mutants at 48 hpf, both anteriorly and posteriorly (n = 12 for 25μM, n = 11 for 50μM CpdE). Insets: Magnified view of the area of interest (5× objective, MZ6 microscope; Leica).

Figure 3

Morpholino knockdown reveals a specific requirement of notch1b for embryonic mast cell development in zebrafish. Morpholinos to zebrafish notch gene homologs were injected in 1 to 4 cell stage wild-type embryos followed by WISH for cpa5, gata2, pu.1, and gata1. notch1b morphants demonstrate a dramatic decrease in the level of cpa5 expression and reduced gata2 expression at 28 hpf, implicating a key role for notch1b in mast cell development mediated through gata2. Bar graphs represent the percent of embryos categorized by WISH expression levels. The table summarizes the predominant phenotype observed for each probe with each morpholino. The numbers of wild-type embryos, notch1a, notch1b, notch2, and notch3 morphants analyzed for each probe were n = 37, 17, 33, 19, and 11 for cpa5; n = 13, 9, 5, 10, and 9 for pu.1; n = 36, 9, 8, 23, and 24 for gata1; and n = 19, 12, 10, 22, and 16 for gata2. MO indicates morpholino; N1a, notch1a; N1b, notch1b; N2, notch2; and N3, notch3. 5× objective, MZ6 microscope (Leica).

Figure 4

Rescue of embryonic mast cells completely in notch1b morphants and partially in pu.1 morphants by injecting gata2 mRNA. Zebrafish notch1b or pu.1 morpholino and gata2 mRNA were injected in 1 to 4 cell stage wild-type embryos followed by WISH for cpa5 expression at 28 hpf. (A) notch1b morphants demonstrate decreased cpa5 expression (n = 33) and gata2 expression (n = 8) compared with wild-type embryos (n = 34 and 16, respectively) that is completely rescued by injecting gata2 mRNA (n = 25). (B) pu.1 morphants demonstrate a dramatic decrease in cpa5-positive mast cells (n = 24) compared with wild-type embryos (n = 34) that is partially rescued by injecting gata2 mRNA (n = 29; Leica MZ6 microscope, 5× objective). Bar graphs represent the percent of embryos categorized by WISH expression levels.

Figure 5

Zebrafish nicd1a overexpression results in an increase of cpa5-positive embryonic mast cells, which can be reversed by Notch inhibition using CpdE. WISH using a digoxigenin-labeled RNA antisense probe to zebrafish cpa5 was performed at different time points in hsp70::GAL4;UAS::nicd1a embryos heat shocked at 13 to 15 hpf. Bar graphs represent the percent of embryos categorized by WISH expression levels. (A) Overexpression of nicd1a results in an increase in the number of cpa5-positive mast cells at 24 hpf (n = 22) and 28 hpf (n = 49) compared with wild-type embryos (n = 20 and 52, respectively). (B) Treatment of hsp70::GAL4;UAS::nicd1a transgenic embryos (n = 43) with the Notch inhibitor CpdE for 8 hours (22-30 hpf) restores wild-type cpa5 expression levels at 25μM (n = 39) with further reduced expression at 50μM (n = 41) compared with wild-type embryos (n = 47). Insets: Magnified view of the area of interest (5× objective, MZ6 microscope, Leica).

Figure 6

Mast cells originate from EMP at 28 to 48 hpf. Double WISH using digoxigenin-labeled RNA antisense probe to zebrafish cpa5 stained with Fast Red and fluorescein-labeled RNA antisense probe to zebrafish lmo2 demonstrate coexpression of the zebrafish EMP marker lmo2 with the mast cell specific marker cpa5 at the ALM and posterior blood island (PBI) at 28 hpf, and at both the ALM and the CHT at 48 hpf (40× objective, Z1 inverted stereo-microscope; Carl Zeiss).

Figure 7

Mast cells arise from HSCs from 48 to 72 hpf. Two morpholinos targeting the zebrafish runx1 gene were injected in 1 to 4 cell stage wild-type embryos followed by WISH for cpa5. runx1 morphants demonstrate a mild decrease in the number of cpa5-positive mast cells at 48 hpf (n = 28) and a progressive decrease at later time points, 60 hpf (n = 14) and 72 hpf (n = 41), compared with wild-type embryos (n = 31, 16, and 17, respectively), implicating that HSCs contribute to the production of the definitive mast cell lineage (5× objective, MZ6 microscope; Leica). Insets: Higher magnification views of the anterior of the embryos. Bar graphs represent the percent of embryos categorized by WISH expression levels. MO indicates morpholino.