spacer gif spacer gif spacer gif spacer gif spacer gif
 QUICK SEARCH:   [advanced]


spacer gif
     Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents    

First published online July 21, 2003
doi: 10.1242/10.1242/dev.00600


This Article
Right arrow Summary Freely available
Right arrow Full Text
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Mayer, A. N.
Right arrow Articles by Fishman, M. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mayer, A. N.
Right arrow Articles by Fishman, M. C.

nil per os encodes a conserved RNA recognition motif protein required for morphogenesis and cytodifferentiation of digestive organs in zebrafish

Alan N. Mayer1,2,* and Mark C. Fishman1,{dagger}

1 Cardiovascular Research Center, Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Charlestown, MA 02129, USA
2 Pediatric Gastroenterology Unit, Deptartment of Pediatrics, Massachusetts General Hospital and Harvard Medical School, Charlestown, MA 02129, USA



View larger version (121K):

[in a new window]
 
Fig. 1. npo-mutant phenotype. (A,B) Live larvae at 4.5 dpf, showing that the jaw, branchial arches and gut tube are markedly underdeveloped at 96 hpf, but that dorsal structures (somites, notocord) appear unaffected by the mutation. (C,D) Alcian Blue staining of wild-type and mutant larvae at 4.5 dpf demonstrates failure of branchial arch growth beyond the initial primordium. (E-G) Cross sections of embryos stained with Hematoxylin and Eosin at the level of the pancreatic islet. (E,F) 72 hpf embryos (genotyped prior to embedding) showing the first morphologically detectable differences between mutant and wild type, with hypoplastic gut and liver in the mutant. (G,H) 96 hpf embryos show a stark contrast between wild type and mutant. The mutant gut tube is substantially smaller and the epithelial architecture is less organized, with no villi formation. The exocrine pancreas is not recognizable in the mutant, yet the islet appears relatively normal. The liver is not seen because it is much smaller in the AP dimension and does not extend to this level. g, gut; p, pancreas; l, liver; i, pancreatic islet; ep, exocrine pancreas; gb, gall bladder; sb, swim bladder. Scale bars: E,F, 20 µm; G,H, 30 µm.

 


View larger version (115K):

[in a new window]
 
Fig. 2. npo is required for endoderm-intestine transition. (A,B) foxa3 expression pattern at 48 hpf showing formation of liver, gut and pancreatic anlage in both wild-type and npo mutant embryos. (C,D) shh expression at 48 hpf reveals normal foregut patterning, with label exclusion at the prepancreatic endoderm. (E,F) ifabp expression detected in the wild-type intestine but not in the mutant. (G,H) Histochemical staining for alkaline phosphatase. Robust staining of apical aspect of epithelium in the wild type, but staining of mutant epithelial cells appears basal, suggesting mislocalization. (I,J), Na/K ATPase immunofluorescence demonstrates basolateral localization of fluorescence in wild type, but no clear localization in the mutant. (K,L) Zo1 immunofluorescence demonstrates formation of zona occludens in both wild-type and mutant embryos. (M,N) Transmission electron micrograph of intestine. Junctional complexes are present in both mutant and wild-type epithelia. Microvilli in the wild type are smooth and uniform, but in the mutant they are fewer and pleiomorphic. +/+, wild type; -/-, mutant; g, gut; L, liver; p, pancreas; mv, microvilli; TJ, tight junction. Scale bars: A-F, 150 µm; G,H, 20 µm; I-L, 10 µm; M,N, 1 µm; insets, 30 µm.

 


View larger version (60K):

[in a new window]
 
Fig. 3. npo mediates exocrine pancreas and liver cytodifferentiation. (A-F) Whole-mount in situ hybridizations show formation of pancreatic primordia in both mutant and wild-type embryos at 48 hpf (A,B), and subsequent failure to form differentiated exocrine pancreas in mutant embryos by 96 hpf (C,D). By contrast, islet formation is independent of npo. (G,H), Double immunofluorescence staining for insulin (green) and carboxypeptidase (red) shows selective failure of exocrine pancreas formation in the mutant, with sparing of the pancreatic islet. (I,J) Immunofluorescence detection of cytokeratin (monoclonals AE1/AE3), demonstrating staining of wild-type liver and gut epilthelia, but absent specific staining of mutant epithelia. White outline defines organ boundaries identified by phase contrast. g, gut; p, pancreas; l, liver. Scale bars: A,B, 80 µm; C-E, 150 µm; G-J, 25 µm.

 


View larger version (26K):

[in a new window]
 
Fig. 4. Positional cloning of npo. (A) Integrated genetic/physical map depicting chromosome walk across the npo locus, initiated at microsatellite marker z8532. Two overlapping BACs covered the npo interval, and these were subjected to complete sequencing, resulting in a 129 kb contig. Within this interval, we identified four genes and new microsatellite markers. These were used to genotype the flanking recombinants and to exclude the two outer genes. Thus only epha2 and a hypothetical RRM protein encoding gene fine-mapped within the genetic locus. We isolated full-length cDNAs corresponding to both genes from homozygous-npo mutant and wild-type embryos, and six independent clones from each were completely sequenced in both directions. No mutations were found in the epha2 coding sequence, whereas a T to A (encoding a Y to stop) mutation was identified in the RRM-encoding gene, which we designated npo. (B) Sequence tracing of one of 12 independent clones derived from PCR of genomic DNA from mutant and wild-type embryos, showing TAT (Y) to TAA (stop) mutation. (C) cDNA from wild-type and mutant embryos subjected to in vitro transcription/translation produces a truncation of mutant Npo, in agreement with the conceptual translation. The doublet most likely arises from an alternate translation start site.

 


View larger version (80K):

[in a new window]
 
Fig. 5. Rescue and phenocopy of npo mutation. (A,C,E) 96 hpf npo-mutant embryos that had been injected with NotI-digested BAC 37b12 at the 1-cell stage. (B,D,F) Mutants injected with BAC 37b12 pre-cut with restriction endonucleases NotI and SnaBI, which cuts only between the first and second exon of the npo gene. (G,H) Wild-type embryos injected with morpholino oligonucleotides encoding antisense and sense sequences, respectively, from the npo transcript. (A,B) Representative embryos stained by whole-mount in situ hybridization for ifabp show rescue of this marker by BAC 37b12 injection in the expected mosaic pattern, and undetectable rescue observed using the cleaved BAC. By comparing panel A with Fig. 2E, we note that the BAC-injected mutant embryos have a broader ifabp expression pattern than the wild-type embryos. (C,D) Staining for trypsin demonstrates mosaic rescue of the npo phenotype, also with a broader expression pattern than seen in the wild-type (compare with Fig. 3C). The control-injected embryos shows a few cells stained for trypsin expression, which we also note occasionally in control-uninjected embryos. (E,F) Embryos shown in A and B, respectively, were sectioned and stained with Hematoxylin and Eosin. This revealed an over-expanded intestinal and exocrine pancreatic epithelium in the embryos injected with intact BAC 37b12, but no histological effect on the mutant phenotype in the control (arrows). (G,H) Histological sections of embryo injected with npo-specific morpholino oligonucleotides, sense or antisense as indicated, stained for insulin by whole-mount in situ hybridization, then sectioned and counterstained with nuclear Fast Red. Antisense injection leads to hypoplasia, dysmorphogenesis and abrogation of epithelial cytodifferentiation similar to that seen in the npo-homozygous mutant. insulin expression is not affected, as seen in the homozygous mutant. Normal digestive organs are seen in the sense control. Note, the effects on the digestive organs due to manipulation of npo expression often result in uncoupling of the AP axes of the neural tube from the digestive organs. For this reason, we chose internal landmarks in the digestive tract (i.e. islet or gall bladder) rather than neural tube or somite landmarks. Arrows indicate intestine. Scale bars: A-D, 200 µm; E-H, 20 µm.

 


View larger version (69K):

[in a new window]
 
Fig. 6. npo encodes a conserved, eukaryotic RNA-binding protein. (A) CLUSTALW alignment of Npo-related proteins from divergent eukaryotes reveals a high degree of conservation, particularly within RRM regions. Dark shading indicates amino acid identity and light shading indicates similarity. Red line indicates canonical RNA recognition motifs. (B) Conservation of RRM-domain configuration between Npo-related proteins. We show a manual alignment of RRM domains identified by RPS blast search of the GenBank conserved domain database (CDD). Domain numbering system was guided by CLUSTALW alignments as shown in panel A. Asterisk indicates site of termination codon encoded by npofW07-g. GenBank Accession Numbers are as follows: Human, NP_055667; Drosophila, AAM29657; C. elegans, NP502432; S. cerevisiae, NP_015437; Arabidopsis, CAB40378. (C) Npo binding of homopolymeric RNA. [35S]-labeled Npo protein produced by in vitro transcription/translation was incubated with solid support-bound homopolymeric RNA of the indicated base composition. After several washes, the radioactivity from bound and input protein was quantitated by autoradiography and digital densitometry. The percent bound was highest for poly G and poly U. Npo926 indicates full-length wild-type protein, Npo220 indicates the truncated mutant. Error bars indicate±s.e.m. from three independent measurements.

 


View larger version (107K):

[in a new window]
 
Fig. 7. Developmental expression pattern of npo. (A-F) Whole-mount in situ hybridization to npo at the indicated developmental stages. (A,B,D) Dorsal view demonstrates dynamic expression in the digestive organ primordia. Expression begins at about 48 hpf, peaks at 72 hpf, then recedes in an anterior to posterior wave around 84 hpf. (C) Right oblique view showing expression of npo in the exocrine pancreas at 72 hpf, but not in the pancreatic islet. (E,F) Lateral view showing brain and endoderm expression at 24 hpf, then branchial arch and endoderm expression at 60 hpf. l, liver; g, gut; p, pancreas; i, islet; ep, exocrine pancreas; en, endoderm; ba, branchial arches.

 


View larger version (113K):

[in a new window]
 
Fig. 8. Expression and function of npo in the endoderm. (A) Mid-sagittal section of 72 hpf embryo labeled for npo by in situ hybridization, showing punctate, heterogeneous staining in epithelial cells. (B) Double in situ hybridization to npo (dark blue; arrowheads) and foxa3 (brown), showing npo expression in a subset of endoderm-derived epithelial cells. (C) Anti-Npo immunofluorescence (red) at 72 hpf shows expression primarily in gut epithelium, and also in scattered subepithelial mesenchymal cells. The yolk signal is caused by autofluorescence. (D) Mosaic analysis: wild-type cell (arrow) in npo-/- host shows cell autonomous rescue of ifabp expression. Panels show nuclei (DAPI; blue), lineage tracer (rhodamine dextran; red) and labeling for ifabp (dark blue). e, epithelium; m, mesenchyme; y, yolk. Scale bars: A,B, 10 µm; C, 15 µm; D, 7 µm.

 





© The Company of Biologists Ltd 2003