A hitherto unidentified N-ethyl-N-nitrosourea (ENU)-induced mutation affects dorsal root ganglia (DRG) formation in ouchless mutant zebrafish larvae. In contrast to previous findings assigning the ouchless phenotypes to downregulated sorbs3 transcript levels, this work re-attributes the phenotypes to an essential splice site mutation affecting adgra2 (gpr124) splicing and function. Accordingly, ouchless mutants fail to complement previously characterized adgra2 mutants and exhibit highly penetrant cerebrovascular defects. The aberrantly spliced adgra2 transcript found in ouchless mutants encodes a receptor lacking a single leucine-rich repeat (LRR) within its N-terminus.

In a research article published in Development, Malmquist et al. (2013) phenotypically characterized the ouchless mutant that was recovered from an F3 forward genetic screen for defective dorsal root ganglion (DRG) neurogenesis. While the initial dorsoventral migration of neural crest-derived cell clusters towards presumptive DRG locales appears unaffected in ouchless mutants, the neurogenic program leading to the generation of neurog1:EGFP+ cells within the ganglion is defective, resulting in a severe reduction of DRG numbers in 72 hours post fertilization (hpf) ouchless mutants. ouchless mutants are viable but exhibit reduced growth rates and interrupted melanophore stripes in the adult skin. The ouchless mutation was mapped by bulk segregation analysis to a 342 kb genomic region of chromosome 8, harboring the sorbs3 gene. No causative mutation could be identified within the coding sequence of sorbs3, but a mutation was suspected to reside within cis-regulatory elements, accounting for the reduced sorbs3 transcript levels observed in ouchless mutants. Antisense sorbs3 morpholino knockdown experiments, as well as BAC and mRNA rescue experiments, further supported the model that sorbs3 regulates DRG neurogenesis and that sorbs3 dysfunction drives the ouchless phenotypes (Malmquist et al., 2013).

The ouchless phenotypes are remarkably analogous to the DRG defects reported in adgra2 (previously known as gpr124) mutants (Vanhollebeke et al., 2015). Adgra2 is a newly discovered Wnt7-specific co-activator of Wnt/β-catenin signaling (Posokhova et al., 2015; Vanhollebeke et al., 2015; Zhou and Nathans, 2014). Along with Reck, it has been shown to control DRG formation by activating Wnt signaling in neural crest-derived sox10:mRFP+ ganglion cells (Vanhollebeke et al., 2015). In addition, Adgra2 and Reck function as essential regulators of brain vascular development by promoting Wnt/β-catenin signaling in cerebrovascular endothelial cells (ECs) (Posokhova et al., 2015; Ulrich et al., 2016; Vanhollebeke et al., 2015; Zhou and Nathans, 2014). While the pivotal role of these proteins in cerebrovascular development is established both in the zebrafish and the mouse model (Anderson et al., 2011; Cullen et al., 2011; de Almeida et al., 2015; Kuhnert et al., 2010; Noda et al., 2016; Posokhova et al., 2015; Ulrich et al., 2016; Vanhollebeke et al., 2015; Zhou and Nathans, 2014), the molecular mechanisms underlying their activation and signal transduction remain to be determined. Given the phenotypic similarities, we therefore set out to test whether adgra2 and ouchless (presumably sorbs3) co-operate during the process of DRG neurogenesis and brain vascularization.

We first tested whether adgra2 and ouchless genetically interact by functional gene dosage experiments. Fish heterozygous for ouchless were crossed with the previously described adgra2 heterozygotes, adgra2s984/+ and adgra2s985/+, and the offspring were assessed at 72 hpf for defects in DRG neurogenesis. From these crosses, ∼25% of the offspring (annotated as ouchless/adgra2s984 and ouchless/adgra2s985) showed an almost complete lack of neurog1:EGFP+ DRG (Fig. 1A,A′). When raised to adulthood, these fish could be distinguished from their siblings by discontinuous dorsal melanophore stripes on their skin (Fig. 1A, brackets). As Adgra2 controls brain angiogenesis, we analyzed the cerebral vasculature of 60 hpf embryos derived from ouchless heterozygotes incrosses and outcrosses to adgra2 heterozygotes. Strikingly, 25% of the offspring of each of the crosses displayed highly penetrant brain vascular defects, characterized by a complete absence of central arteries (CtAs), similar to adgra2 mutants (Fig. 1B,B′). Further assessment using Wnt/β-catenin signaling reporter lines linked this phenotype to defective endothelial Wnt/β-catenin signaling in the perineural primordial hindbrain channel (PHBC) ECs (Fig. 1C). The lack of complementation between ouchless and adgra2, together with the discovery of vascular phenotypes in ouchless mutants that mimic those of adgra2 mutants, raised the possibility that ouchless constitutes a new allele of adgra2. Accordingly, when ouchless mutants were injected at the one-cell stage with mRNA encoding wild-type (WT) Adgra2, significant restoration of neurog1:EGFP+ DRG (Fig. 1D,E) and cerebral blood vessels was observed (Fig. 1D,F).

Fig. 1.

adgra2 and ouchless mutations fail to complement. (A) Dorsal views and (A′) quantification of EGFP+ DRG neurons in 72 hpf Tg(neurog1:EGFP) larvae resulting from crossing ouchless and adgra2 alleles. Red arrowheads indicate DRG neurons. Center and right columns illustrate the corresponding adult pigmentation patterns. The neurog1:EGFP+ DRG were counted on one side of the 72 hpf larvae. Brackets indicate discontinuous dorsal melanophore stripes. (B) Confocal z-stack projections showing the brain vasculature of Tg(kdrl:GFP) WT and mutant embryos at 60 hpf in dorsal views (top) and 3D wire diagram representation of the cranial vessels, with intracerebral vessels (central arteries or CtAs) in red and perineural vessels in white (bottom). (B′) Quantification of the corresponding hindbrain CtAs. (C) Confocal projections in dorsal views of WT and mutant Tg(7xTCF-Xia.Siam:GFP); Tg(kdrl:ras-mCherry) embryos at 32 hpf. Boxes define the areas magnified in the bottom row. (D) Confocal projections of the ouchless mutant Tg(neurog1:EGFP) DRG at 72 hpf (top panels) and Tg(kdrl:GFP) cranial vasculature at 60 hpf (bottom panels) injected, or not, with 100 pg of adgra2 mRNA at the one-cell stage. Red arrowheads indicate DRG neurons. (E) Quantification of neurog1:EGFP+ DRG at 72 hpf and (F) hindbrain CtAs at 60 hpf in WT and ouchless mutant animals injected, or not, with 100 pg of adgra2 mRNA at the one-cell stage. The neurog1:EGFP+ DRG were counted on one side of the 72 hpf larvae. Error bars represent median±interquartile range; *P<0.05, ***P<0.001 (Kruskal–Wallis test). PHBC, primordial hindbrain channel. Scale bars: 50 µm.

Fig. 1.

adgra2 and ouchless mutations fail to complement. (A) Dorsal views and (A′) quantification of EGFP+ DRG neurons in 72 hpf Tg(neurog1:EGFP) larvae resulting from crossing ouchless and adgra2 alleles. Red arrowheads indicate DRG neurons. Center and right columns illustrate the corresponding adult pigmentation patterns. The neurog1:EGFP+ DRG were counted on one side of the 72 hpf larvae. Brackets indicate discontinuous dorsal melanophore stripes. (B) Confocal z-stack projections showing the brain vasculature of Tg(kdrl:GFP) WT and mutant embryos at 60 hpf in dorsal views (top) and 3D wire diagram representation of the cranial vessels, with intracerebral vessels (central arteries or CtAs) in red and perineural vessels in white (bottom). (B′) Quantification of the corresponding hindbrain CtAs. (C) Confocal projections in dorsal views of WT and mutant Tg(7xTCF-Xia.Siam:GFP); Tg(kdrl:ras-mCherry) embryos at 32 hpf. Boxes define the areas magnified in the bottom row. (D) Confocal projections of the ouchless mutant Tg(neurog1:EGFP) DRG at 72 hpf (top panels) and Tg(kdrl:GFP) cranial vasculature at 60 hpf (bottom panels) injected, or not, with 100 pg of adgra2 mRNA at the one-cell stage. Red arrowheads indicate DRG neurons. (E) Quantification of neurog1:EGFP+ DRG at 72 hpf and (F) hindbrain CtAs at 60 hpf in WT and ouchless mutant animals injected, or not, with 100 pg of adgra2 mRNA at the one-cell stage. The neurog1:EGFP+ DRG were counted on one side of the 72 hpf larvae. Error bars represent median±interquartile range; *P<0.05, ***P<0.001 (Kruskal–Wallis test). PHBC, primordial hindbrain channel. Scale bars: 50 µm.

We then re-evaluated the genomic region known to harbor the ouchless mutation and discovered that the adgra2 gene resides within the critical interval, spanning the ca-48 and ca-37 genomic markers (Fig. 2A). This information was missing in the original characterization of the ouchless mutants owing to incomplete genome assembly and annotation at the time of analysis (Malmquist et al., 2013). We cloned the full-length adgra2 coding sequence from ouchless mutants and evaluated the capacity of this allele to rescue DRG and CtA defects in adgra2 morphants by mRNA injection at the one-cell stage. While mRNA encoding the WT receptor (annotated as adgra2) partially suppressed both phenotypes, the ouchless variant (annotated as adgra2ouchless) did not affect either (Fig. 2D,E). Of note, the experiments were performed in morphants rather than mutants in order to increase the number of observations and hence the strength of the statistical analyses. The adgra2 morpholino sequence and dosage used in this study were previously validated to ensure that phenotypic suppression values in morphant and mutant genetic backgrounds do not statistically differ (Vanhollebeke et al., 2015).

Fig. 2.

adgra2 is mutated in ouchless mutants. (A) Representation of the ouchless locus genetic map on chromosome 8. The number of recombinants among 1304 meioses as determined by Malmquist et al. (2013) is indicated above the markers utilized for mapping. (B) Sanger sequencing of the exon 4-intron 4 boundary of adgra2 in WT and ouchless mutant embryos. The G→T change in the ouchless 5′ splice donor sequence appears in red. (C) RT-PCR splicing analysis of adgra2 in 48 hpf WT and ouchless mutant embryos. The amplification primers hybridize to exon 1 and exon 6, as illustrated in the panel on the right. (D) Quantification of 72 hpf neurog1:EGFP+ DRG and (E) 60 hpf hindbrain CtAs in WT and adgra2 morphants after injection of 100 pg of the indicated mRNA at the one-cell stage. Error bars represent median±interquartile range; **P<0.01 (Kruskal–Wallis test). (F) Schematic representation of Adgra2, Adgra2ouchless and Adgra2ΔLRR3 topology and domain organization. The red asterisks indicate the positions of the identified SNPs.

Fig. 2.

adgra2 is mutated in ouchless mutants. (A) Representation of the ouchless locus genetic map on chromosome 8. The number of recombinants among 1304 meioses as determined by Malmquist et al. (2013) is indicated above the markers utilized for mapping. (B) Sanger sequencing of the exon 4-intron 4 boundary of adgra2 in WT and ouchless mutant embryos. The G→T change in the ouchless 5′ splice donor sequence appears in red. (C) RT-PCR splicing analysis of adgra2 in 48 hpf WT and ouchless mutant embryos. The amplification primers hybridize to exon 1 and exon 6, as illustrated in the panel on the right. (D) Quantification of 72 hpf neurog1:EGFP+ DRG and (E) 60 hpf hindbrain CtAs in WT and adgra2 morphants after injection of 100 pg of the indicated mRNA at the one-cell stage. Error bars represent median±interquartile range; **P<0.01 (Kruskal–Wallis test). (F) Schematic representation of Adgra2, Adgra2ouchless and Adgra2ΔLRR3 topology and domain organization. The red asterisks indicate the positions of the identified SNPs.

In order to identify the inactivating mutation, we compared a reference WT adgra2 allele with the adgra2 coding sequence recovered from ouchless mutant embryos. This analysis revealed four non-synonymous single nucleotide polymorphisms (SNPs; M429V, S895P, A1282V and A1302G) as well as an in-frame 72 bp deletion corresponding to exon 4 (Fig. 2C,F). While all four adgra2 SNPs identified in ouchless mutants had been previously identified in functionally validated adgra2 alleles derived from mixed AB/TL genetic backgrounds, alternative splicing resulting in exon 4 skipping is absent from any known zebrafish, mouse or human ADGRA2 isoform. When probed in zebrafish, alternative splicing of the exon 1-exon 6 sequences is undetectable by RT-PCR (Fig. 2C) or by Sanger sequencing of full-length adgra2 coding sequences derived from WT larvae (data not shown). Exon 4 corresponds precisely to the third leucine-rich repeat (LRR) unit of the LRR/CT domain of Adgra2, which comprises an array of four 24-residue-long LRR units followed by a LRR cysteine-rich C-terminal motif (LRR-CT) (Fig. 2F). Exon–protein domain correlations are recurrent amongst eukaryotes and reflect an efficient mechanism for protein modular functionalization through exon shuffling. The genomic structure of ADGRA2 is evolutionary conserved, with LRR units precisely matching exons 2 to 5 in zebrafish, mouse and human (Fig. 2C). Sanger sequencing of genomic regions flanking adgra2ouchless exon 4 identified an essential 5′ splice donor site mutation (GT→TT) at the exon 4–intron 4 boundary, which probably accounts for the exon 4 skipping event (Fig. 2B).

To confirm that the in-frame deletion of exon 4 leads to Adgra2 inactivation, we reproduced this deletion in a WT allele of adgra2 by mutagenesis and evaluated the functionality of the resulting variant, adgra2ΔLRR3, in transient rescue assays. Injections of 100 pg of adgra2ouchless or adgra2ΔLRR3 mRNA into adgra2 morphants at the one-cell stage did not yield detectable rescuing neurogenic or angiogenic activity (Fig. 2D,E). This indicates that Adgra2 lacking LRR3 is functionally null. In light of this evidence, the fact that a limited number of DRG neurons develop in ouchless mutants, but not in the adgra2s984 and adgra2s985 frame-shift mutants, appears puzzling (Fig. 1A,A′). We speculate that the residual ouchless DRG neurons result either from a minor pool of normally spliced transcripts or from a cryptic splice-site activation event restoring a functional allele. However, these events must be very rare and tissue specific, as brain vascular defects of the ouchless and frame-shift mutants are equally penetrant and only the transcript lacking exon 4 could be amplified from 48 hpf ouchless mutants (Fig. 2C).

Altogether, this work reveals that ouchless and adgra2 mutants are allelic and that the ouchless phenotypes result from an essential splice site mutation inactivating Adgra2 through the in-frame deletion of a single LRR in the ectodomain of this adhesion G-protein coupled receptor (GPCR). Further characterization of the ouchless allele suggests that the lack of this LRR motif impairs trafficking of Adgra2 (Bostaille et al., 2016).

Malmquist et al. (2013) previously attributed the ouchless phenotypes to defective sorbs3 function based on convergent evidence from genomic mapping approaches, partial phenotypic rescue after ectopic Sorbs3 expression from cDNA or BAC templates as well as loss-of-function morpholino analyses. sorbs3 was additionally reported to genetically interact with erbb3 signaling in the process of DRG neurogenesis. We note that, while our current work unambiguously re-assigns the ouchless phenotypes to defective adgra2 rather than sorbs3, it does not per se exclude (nor confirm) a role for sorbs3 or erbb3 in Adgra2-controlled DRG neurogenesis. However, in light of the genetic evidence presented here, a careful re-evaluation of their contribution to the ouchless phenotypes is warranted.

Zebrafish strains and cell lines

Zebrafish (Danio rerio) were raised and maintained under standard conditions. The following lines were used: AB, TL, Tg(kdrl:GFP)s843 (Jin et al., 2005), Tg(kdrl:ras-mCherry)s896 (Chi et al., 2008), Tg(7xTCF-Xla.Siam:GFP)ia4 (Moro et al., 2012), Tg(-17.0neurog1:EGFP)w61 (McGraw et al., 2008), ouchless (sorbs3w35) (Malmquist et al., 2013), adgra2s984 and adgra2s985 (Vanhollebeke et al., 2015), as reported previously. All animal experiments were performed in accordance with the rules of the State of Belgium (protocol approval number: CEBEA-IBMM-2012:65).

Cloning strategy, morpholino and RNA expression constructs

The adgra2ΔLRR3 deletion mutant was generated by In-Fusion cloning (Takara, Mountain View, CA) and the deletion corresponds to amino acids 125-148. Capped messenger RNA was synthesized using the mMESSAGE mMACHINE kit (Ambion, Carlsbad, CA). In all panels, one-cell-stage embryos were injected either with 100 pg of the indicated mRNA or 4 ng of a previously validated adgra2 splice-blocking morpholino (Vanhollebeke et al., 2015).

Imaging

Images were acquired on a Zeiss LSM710 confocal microscope. Three-dimensional representations were generated using Imaris FilamentTracer software (BitPlane, Zurich, Switzerland).

Statistical analysis

Statistical analyses were performed using the GraphPad Prism software. Sample size was determined with G*power v.3.1.5 software to reach adequate statistical power. Each dot plot value represents an independent embryo and every experiment was conducted three times independently. P-values were calculated by the Kruskal–Wallis test (post hoc Dunn's test).

We thank the B.V. laboratory members for critical reading of the manuscript and E. Dupont for technical assistance.

Funding

N.B. is supported by a FRIA fellowship from Fonds de la Recherche Scientifique - FNRS. Work in the B.V. laboratory is supported by the FNRS (MIS F.4543.15), a Concerted Research Action (ARC) grant from the Fédération Wallonie-Bruxelles and the Fondation Université libre de Bruxelles. The Center for Microscopy and Molecular Imaging (CMMI) is supported by the European Regional Development Fund (ERDF).

Anderson
,
K. D.
,
Pan
,
L.
,
Yang
,
X.
,
Hughes
,
V. C.
,
Walls
,
J. R.
,
Dominguez
,
M. G.
,
Simmons
,
M. V.
,
Burfeind
,
P.
,
Xue
,
Y.
,
Wei
,
Y.
, et al.
(
2011
).
Angiogenic sprouting into neural tissue requires Gpr124, an orphan G protein-coupled receptor
.
Proc. Natl. Acad. Sci. USA
108
,
2807
-
2812
.
Bostaille
,
N.
,
Gauquier
,
A.
,
Twyffels
,
L.
and
Vanhollebeke
,
B.
(
2016
).
Molecular insights into Adgra2/Gpr124 and Reck intracellular trafficking
.
Biol. Open
5
,
1874
-
1881
.
Chi
,
N. C.
,
Shaw
,
R. M.
,
De Val
,
S.
,
Kang
,
G.
,
Jan
,
L. Y.
,
Black
,
B. L.
and
Stainier
,
D. Y. R.
(
2008
).
Foxn4 directly regulates tbx2b expression and atrioventricular canal formation
.
Genes Dev.
22
,
734
-
739
.
Cullen
,
M.
,
Elzarrad
,
M. K.
,
Seaman
,
S.
,
Zudaire
,
E.
,
Stevens
,
J.
,
Yang
,
M. Y.
,
Li
,
X.
,
Chaudhary
,
A.
,
Xu
,
L.
,
Hilton
,
M. B.
, et al.
(
2011
).
GPR124, an orphan G protein-coupled receptor, is required for CNS-specific vascularization and establishment of the blood-brain barrier
.
Proc. Natl. Acad. Sci. USA
108
,
5759
-
5764
.
de Almeida
,
G. M.
,
Yamamoto
,
M.
,
Morioka
,
Y.
,
Ogawa
,
S.
,
Matsuzaki
,
T.
and
Noda
,
M.
(
2015
).
Critical roles for murine Reck in the regulation of vascular patterning and stabilization
.
Sci. Rep.
5
,
17860
.
Jin
,
S.-W.
,
Beis
,
D.
,
Mitchell
,
T.
,
Chen
,
J.-N.
and
Stainier
,
D. Y. R.
(
2005
).
Cellular and molecular analyses of vascular tube and lumen formation in zebrafish
.
Development
132
,
5199
-
5209
.
Kuhnert
,
F.
,
Mancuso
,
M. R.
,
Shamloo
,
A.
,
Wang
,
H.-T.
,
Choksi
,
V.
,
Florek
,
M.
,
Su
,
H.
,
Fruttiger
,
M.
,
Young
,
W. L.
,
Heilshorn
,
S. C.
, et al.
(
2010
).
Essential regulation of CNS angiogenesis by the orphan G protein-coupled receptor GPR124
.
Science
330
,
985
-
989
.
Malmquist
,
S. J.
,
Abramsson
,
A.
,
McGraw
,
H. F.
,
Linbo
,
T. H.
and
Raible
,
D. W.
(
2013
).
Modulation of dorsal root ganglion development by ErbB signaling and the scaffold protein Sorbs3
.
Development
140
,
3986
-
3996
.
McGraw
,
H. F.
,
Nechiporuk
,
A.
and
Raible
,
D. W.
(
2008
).
Zebrafish dorsal root ganglia neural precursor cells adopt a glial fate in the absence of neurogenin1
.
J. Neurosci.
28
,
12558
-
12569
.
Moro
,
E.
,
Ozhan-Kizil
,
G.
,
Mongera
,
A.
,
Beis
,
D.
,
Wierzbicki
,
C.
,
Young
,
R. M.
,
Bournele
,
D.
,
Domenichini
,
A.
,
Valdivia
,
L. E.
,
Lum
,
L.
, et al.
(
2012
).
In vivo Wnt signaling tracing through a transgenic biosensor fish reveals novel activity domains
.
Dev. Biol.
366
,
327
-
340
.
Noda
,
M.
,
Vallon
,
M.
and
Kuo
,
C. J.
(
2016
).
The Wnt7's Tale: a story of an orphan who finds her tie to a famous family
.
Cancer Sci.
107
,
576
-
582
.
Posokhova
,
E.
,
Shukla
,
A.
,
Seaman
,
S.
,
Volate
,
S.
,
Hilton
,
M. B.
,
Wu
,
B.
,
Morris
,
H.
,
Swing
,
D. A.
,
Zhou
,
M.
,
Zudaire
,
E.
, et al.
(
2015
).
GPR124 functions as a WNT7-specific coactivator of canonical β-catenin signaling
.
Cell Rep.
10
,
123
-
130
.
Ulrich
,
F.
,
Carretero-Ortega
,
J.
,
Menendez
,
J.
,
Narvaez
,
C.
,
Sun
,
B.
,
Lancaster
,
E.
,
Pershad
,
V.
,
Trzaska
,
S.
,
Veliz
,
E.
,
Kamei
,
M.
, et al.
(
2016
).
Reck enables cerebrovascular development by promoting canonical Wnt signaling
.
Development
143
,
147
-
159
.
Vanhollebeke
,
B.
,
Stone
,
O. A.
,
Bostaille
,
N.
,
Cho
,
C.
,
Zhou
,
Y.
,
Maquet
,
E.
,
Gauquier
,
A.
,
Cabochette
,
P.
,
Fukuhara
,
S.
,
Mochizuki
,
N.
, et al.
(
2015
).
Tip cell-specific requirement for an atypical Gpr124- and Reck-dependent Wnt/β-catenin pathway during brain angiogenesis
.
eLife
4
,
e06489
.
Zhou
,
Y.
and
Nathans
,
J.
(
2014
).
Gpr124 controls CNS angiogenesis and blood-brain barrier integrity by promoting ligand-specific canonical Wnt signaling
.
Dev. Cell
31
,
248
-
256
.

Competing interests

The authors declare no competing or financial interests.