ORIGINAL PAPER

Dissection of the role of Pinin in the development of zebrafishposterior pharyngeal cartilages

Shu-Yuan Hsu • Yi-Chuan Cheng • Hung-Yu Shih •

Pin Ouyang

Accepted: 2 April 2012 / Published online: 20 April 2012

� Springer-Verlag 2012

Abstract Pinin (pnn), a nuclear and desmosome-associ-

ated SR-like protein, has been shown to play multiple roles

in cell adhesion, transcriptional regulation, pre-mRNA

splicing and mRNA export. Because of the embryonic

lethality of pnn-deficient mice, here we used the zebrafish

system to investigate the functions of pnn. Injection of

morpholinos into zebrafish to knockdown pnn resulted in

several obvious defective phenotypes, such as short body,

bent tail, and an abnormal pigment distribution pattern.

Moreover, aberrant blood vessels were formed, and most of

the cartilages of pharyngeal arches 3–7 were reduced or

absent in pnn morphants. Because most of the defects

manifested by pnn morphants were reminiscent of those

caused by neural crest-derived malformation, we investi-

gated the effects of pnn deficiency in the development of

neural crest cells. Neural crest induction and specification

were not hindered in pnn morphants, as revealed by normal

expression of early crest gene, sox10. However, the mor-

phants failed to express the pre-chondrogenic gene, sox9a,

in cells populating the posterior pharyngeal arches. The

reduction of chondrogenic precursors resulted from inhibi-

tion of proliferation of neural crest cells, but not from

cellular apoptosis or premature differentiation in pnn mor-

phants. These data demonstrate that pnn is essential for the

maintenance of subsets of neural crest cells, and that in

zebrafish proper cranial neural crest proliferation and dif-

ferentiation are dependent on pnn expression.

Keywords Pinin � Pnn � Neural crest � Zebrafish �Pharyngeal cartilage � Proliferation

Introduction

Pinin (pnn) is a desmosome and nuclear-associated moon-

lighting phosphoprotein (Ouyang 1999) that is involved in

cell–cell adhesion, tumor suppression, pre-mRNA splicing,

transcriptional regulation, and mouse embryogenesis (Shi

et al. 2001; Zimowska et al. 2003; Alpatov et al. 2004; Joo

et al. 2005, 2007). Previous in vitro studies have demon-

strated that overexpression of pnn results in enhancement of

intercellular adhesion and a decrease in cell motility and

proliferation (Ouyang and Sugrue 1996; Shi et al. 2000a),

while downregulation of pnn by shRNA caused cells to

become less adherent and more migratory (Joo et al. 2005).

The tumor-suppressive role of pnn was supported by the

demonstration that expression of pnn is significantly lower

in several carcinoma cell lines, and ectopic expression of

pnn leads to suppression of anchorage-independent growth

of carcinoma-derived tumor cells (Shi et al. 2000b). Con-

sistently, cDNA array analysis following pnn overexpres-

sion in EcR-293 cells revealed alterations in expression of

several cell cycle-related genes, such as p21, CDK4, and

proto-oncogenes c-Jun and c-Myc (Shi et al. 2001).

Electronic supplementary material The online version of thisarticle (doi:10.1007/s00418-012-0950-3) contains supplementarymaterial, which is available to authorized users.

S.-Y. Hsu � P. Ouyang (&)

Transgenic Mouse Core-Lab, Epithelial Biology Laboratory,

Department of Anatomy, Graduate Institute of Biomedical

Sciences, Chang Gung University Medical College,

259 Wen-Hwa 1st Road, Kwei-Shan, Tao-Yuan 333,

Taiwan, ROC

e-mail: ouyang@mail.cgu.edu.tw

Y.-C. Cheng � H.-Y. Shih

School of Medicine, Graduate Institute of Biomedical Sciences,

Chang Gung University, 259 Wen-Hwa 1st Road, Kwei-Shan,

Tao-Yuan 333, Taiwan, ROC

123

Histochem Cell Biol (2012) 138:127–140

DOI 10.1007/s00418-012-0950-3

As well as being located in the desmosome, pnn is also

located within the nucleus where it is distributed in a

speckle pattern (Brandner et al. 1997, 1998; Ouyang 1999).

Because of the RS domain at its c-terminal, pnn is con-

sidered to be an SR-like protein (Ouyang et al. 1997).

In vitro studies demonstrated that pnn interacts with several

SR family proteins that function in splicing regulation,

such as RNPS1, SRp75, SRrp130, SRm300 and SR-

cyclophilin (Li et al. 2003; Zimowska et al. 2003; Lin et al.

2004) and is involved in pre-mRNA splicing and mRNA

export (Li et al. 2003; Wang et al. 2002). Interfering with

pnn expression using siRNA in cultured cells caused

reduced expression of several SR family proteins, includ-

ing SC35, SRm300, SRp55 and SRp40, and also modulated

alternative pre-mRNA splicing in vivo (Chiu and Ouyang

2006). Other studies also indicated that pnn participates in

cell migration and proliferation at the gene regulatory

level, while function of pnn in modulation of E-cadherin

activity was attributed to its association with the CtBP co-

repressor (Alpatov et al. 2004; Shi et al. 2001).

Neural crest cells are an ectoderm-derived cell popula-

tion that originates from the developing neural tube. Fol-

lowing induction, neural crest cells delaminate, migrate

and differentiate into a wide variety of cell types, including

peripheral neurons, glia, melanocytes and craniofacial

skeleton (Le Douarin and Kalcheim 1999). During migra-

tion, neural crest cells divide and begin to express markers

defining the lineage they are destined to form. They are

then guided to their destinations and differentiate into

mature structures. Hypomorphic reduction of pnn resulted

in a wide range of developmental defects in mice (Joo et al.

2007). Disorganized dorsal dermis devoid of brown adi-

pose tissue, and malformation of axial skeleton in pnn

hypomorphic mice suggested that pnn is important for the

development of these somite derivatives. In addition,

defects in the cleft palate, cardiac outflow tract and cra-

niofacial skeleton indicated that pnn also has an essential

role in neural crest development (Joo et al. 2007).

Analysis of pnn expression in mouse embryogenesis

showed that pnn is ubiquitously expressed in mice starting at

the two-cell stage, albeit with differential expression levels

(Leu and Ouyang 2006). Moreover, reduction in pnn

expression led to perinatal mortality (Joo et al. 2007), indi-

cating that pnn is critical for mouse development. However,

pnn hypomorphic mice exhibited a complex phenotype,

making functional studies of pnn difficult. Therefore, in this

study, we chose the zebrafish model system to examine the

effects of depletion of pnn during development. Knockdown

of pnn in zebrafish resulted in cardiac edema, aberrant blood

vessel formation, and defects in pigment distribution and

branchial cartilage formation. Further investigation revealed

neural crest induction, specification and early migration in

pnn morphants appeared undisturbed, but the primordia of

pharyngeal arches failed to form later in development.

Specifically, depletion of pnn resulted in hypoplastic

development of posterior pharyngeal arches 3–7 in zebrafish,

which was mediated by decreasing cell proliferation in cer-

tain regions along the rhombomeres.

Materials and methods

Zebrafish maintenance

Adult fish and embryos were maintained using the methods

described by Westerfield (2000). Development time points

refer to hours (hpf) or days post-fertilization (dpf) and were

staged according to Kimmel et al. (1995). Embryos at

different stages were fixed with 4 % paraformaldehyde in

PBS at 4 �C overnight. After washing with PBST (0.1 %

Tween 20 in PBS) several times, embryos were dehydrated

with a methanol gradient: 25 % methanol in PBST, 50 %

methanol in PBST, 75 % methanol in PBST, 100 %

methanol, for 10 min each. The embryos were then trans-

ferred to -20 �C for storage.

Morpholino anti-sense oligonucleotide and RNA

injections

Translation blocking (MO-1) and splice site blocking

(MO-2) morpholino oligonucleotides of pnn were 50-TCGC

ACGGCCACCGCCATCTTGCAT-30 and 50-GCCAGCCT

AGTACAGACATTTGTAA-30, respectively. Morpholinos

were dissolved in DEPC water and pressure injected into

one- to four-cell zebrafish embryos with approximately 1 nl

MO at concentration indicated. To ensure the specificity of

the morpholinos, rescue of the resulting phenotypes was

performed by co-injecting the synthetic pnn mRNA with

each morpholino. Final concentration of 527 ng/ll of pnn

mRNA was mixed with one half of the MOs and injected into

one- to four-cell embryos.

Plasmid construction

To test the efficiency of translating blocking morpholino

(MO-1) action, a GFP fusion construct was created with

MO-1 target sequence. The sequence was cloned into the

upstream of the coding sequence of EGFP on the pCS2?

EGFP vector. After cloning, the orientation and sequence

of insert was verified by sequencing using vector-specific

primers flanking the insertion site on pCS2? EGFP vector.

Cloning and RNA transcription

RNA extracted from adult fish was used for cDNA syn-

thesis. PCR for pnn was performed using the forward

128 Histochem Cell Biol (2012) 138:127–140

123

primer 50-gAAgTTTgCAAgACCAgCTCgA-30 and the

reverse primer 50-CCATCAACAggCCAAACATgCg-30.The PCR product was cloned into pGEM-T Easy Vector

(Promega). To make sense RNA for injection, the pnn

plasmid was cut with Sac II and in vitro transcription was

driven from the SP6 promoter using Riboprobe in vitro

Transcription System (Promega).

Whole mount cartilage staining

Alcian blue staining was modified from Schilling et al.

(1996). For cartilage detection, 5-dpf zebrafish larvae were

fixed in 4 % paraformaldehyde in PBS at 4 �C overnight.

After brief washing, embryos were bleached in 10 % H2O2

for 1 h. They were then incubated in 0.1 % Alcian Blue

solution in acidic ethanol (70 % ethanol, 5 % concentrated

hydrochloric acid), cleared with acidic ethanol, dehydrated

and stored in 90 % glycerol.

Whole mount in situ hybridization

Whole mount in situ hybridization was performed accord-

ing to Thisse and Thisse (2008) using digoxygenin-labeled

RNA probe (Roche). The following antisense RNA probes

were used: sox9a (cartilage), sox10 (premigratory neural

crest cells), mitfa (melanocyte) and krox20 (rhombomeres 3

and 5). The embryos stored in 100 % methanol were

rehydrated using a reverse methanol gradient. After wash-

ing with PBST several times, embryos were treated with

proteinase K (final concentration 10 lg/ml) at room tem-

perature for the time indicated in Thisse and Thisse (2008).

Proteinase K digestion was stopped by incubating the

embryos in 4 % paraformaldehyde in PBS for 20 min and

then washing with PBST. Embryos were prehybridized in

hybridization mixture (HM) (50 % formamide, 59 SSC,

0.1 % Tween 20, 50 mg/ml heparin and 500 mg/ml tRNA)

for 3–4 h at 70 �C, and then replaced with hybridization

mixture containing antisense DIG-labeled RNA probe

(30–50 ng) at 70 �C overnight. The next day, the embryos

were transferred from the hybridization mixture to 29 SSC

through a series of 10 min, 70 �C washes with 75 % HM,

50 % HM, 25 % HM and 100 % 29 SSC, and then washed

with 0.2 % SSC at 70 �C twice, for 30 min per wash. Then

the 0.2 % SSC was progressively replaced with PBST

through 75 % 0.29 SSC, 50 % 0.29 SSC, 25 % 0.29 SSC

and 100 % PBST, each for 10 min at room temperature.

The embryos were treated in blocking buffer (5 % sheep

serum in PBST) for 3–4 h at room temperature, and then

incubated in anti-DIG antibody solution diluted with

blocking buffer overnight at 4 �C with gentle agitation.

After washing eight times, 1 h per wash, in PBST at room

temperature, the embryos were left in PBST at 4 �C over-

night. Embryos were briefly dried, and then washed with

alkaline Tris buffer (100 nM Tris–HCl, pH 9.5, 50 mM

MgCl2, 100 mM NaCl and 0.1 % Tween 20) 3 times, for

10 min each time. The alkaline Tris buffer was replaced

with staining solution containing 4.5 ll NBT (nitro blue

tatrazolium) and 3.5 ll BCIP (5-bromo 4-chloro 3-indolyl

phosphate) per ml in alkaline Tris buffer. When the desired

staining intensity was reached, the reaction was stopped by

several washes with PBST, and then the embryos were fixed

in 4 % paraformaldehyde in PBS for 20 min. Whole mount

embryos were cleared in 90 % glycerol and examined with

a microscope.

Cell proliferation and cell death assays

Cell proliferation was detected by whole mount immunoflu-

orescence using phospho-histone 3 antibody. The embryos

stored in 100 % methanol were rehydrated using a reverse

methanol gradient. After washing with PBST several times,

fixed embryos were blocked and permeabilized using a

solution containing 5 % sheep serum in PBST for 2 h at

room temperature. Embryos were incubated with mouse

anti-phospho-histone 3 primary antibody (1:500; Millipore)

diluted with 2 % goat serum in PBST at 4 �C overnight.

After extensive washing with PBST, embryos were then

incubated in goat anti-rabbit Alexa Fluor 488 (1:500;

Invitrogen) diluted with 2 % goat serum in PBST at room

temperature for 2 h. Embryos were washed with PBST

several times, post-fixed with 4 % paraformaldehyde for

30 min at room temperature. Then after washing with PBST

several times, embryos were stored in 90 % glycerol at 4 �C

until analysis.

Terminal transferase-mediated dUTP nick end-labeling

(TUNEL) was used to assess apoptosis. For TUNEL assay,

dehydrated embryos, as described above, were rehydrated

with reverse methanol gradient and then washed several

times with PBST. After several washes with PBS, the

embryos were subsequently incubated at 37 �C for 1 h with

a TUNEL apoptosis detection kit (Roche). Embryos were

then extensively washed with PBST and post-fixed with 4 %

paraformaldehyde in PBS at room temperature for 30 min.

Results

Suppression of pnn expression in zebrafish embryos

Before we embarked on morpholino treatment of zebrafish,

we first examined the gene structure of zebrafish pnn and

its expression profiles during fish embryogenesis. Zebrafish

pnn bears much resemblance to its human and higher

vertebrate homologues in terms of gene structure organi-

zation with nine exons spanning about 10.6 kb (Fig. 1A).

Extensive search of genome databases (ensembl and NCBI)

Histochem Cell Biol (2012) 138:127–140 129

123

reveals that there is no other zebrafish pnn homologue,

suggesting only one copy of the gene in the genome. In situ

hybridization of zebrafish embryos showed pnn is ubiqui-

tously expressed in early embryos (Supplementary Fig. 1),

similar to those seen during mouse early embryogenesis

(Leu and Ouyang 2006).

Initial gene knockdown studies were performed using

two antisense morpholinos (MOs), a translation blocking

morpholino (MO-1), and a splice-modifying morpholino

(MO-2) (Fig. 1A). Efficiency of start and splice MOs was

confirmed by co-injection of specific GFP construct con-

taining MO-1 target sequence (Fig. 1B) and RT-PCR

(Fig. 1C), respectively. Little or no fluorescence was

observed in embryos injected with translating blocking

MO-1 after the injection of the GFP mRNA carrying the

corresponding partial sequence. RT-PCR of total RNA

extracted from splice morpholino MO-2-injected embryos

detects the presence of a 560-bp splice-modified pnn tran-

script fragment (Supplementary Fig. 2), which is not

observed in wild-type controls (Fig. 1C). These two assays

therefore demonstrated not only the specificity of mor-

pholinos used but also their efficiency in blocking pnn

transcripts formation in pnn morphants.

Two-day post-fertilization (dpf) pnn morphants showed

several defects, such as cardiac edema and absence of

swim bladder (Fig. 2). We also found that most of the pnn

morphants had bent tails, little swimming activity and were

shorter in length than the control larvae (data not shown).

In addition, the obvious phenotype detectable in pnn-defi-

cient larvae was a severe reduction of pigmentation, clearly

recognizable at 2 dpf (Fig. 2b, c). While wild-type early

larvae exhibited several continuous stripes of pigment

(Fig. 2a), pnn morphants displayed scattered clusters of

spotted pigment (Fig. 2b), or stripes of pigment that were

much thinner than those in the wild type (Fig. 2c). For-

mation of the pigmented epithelium in the retina, however,

was unaffected (data not shown). The specificity of the

phenotype change caused by pnn knockdown was also

tested by co-injection of pnn MO with pnn mRNA. Pnn

mRNA was able to rescue the defects, including the peri-

cardial effusion and the expression pattern of pigments

caused by pnn suppression (Fig. 3) demonstrating the

specificity of the pnn knockdown phenotypes.

Aberrant blood vessel formation was observed in pnn

morphants

Transgenic zebrafish, which express GFP throughout the

vasculature, are a powerful tool for understanding the

dynamic formation of blood vessels. The expression of

Fig. 1 Efficiency of knockdown of pnn in zebrafish by morpholinos.

A Schematic diagram of zebrafish pnn gene organization and location

of start MO (MO-1) and the splice MO (MO-2). Start morpholino

MO-1 and splice morpholino MO-2 oligonucleotides were designed

to bind the transcription start site and the site of exon 4 pnn mRNA,

respectively. Annealing sites are indicated as horizontal bars. Primer

set for RT-PCR is denoted by arrows. B Fluorescent microscopic

images of representative larvae microinjected with MO-1 target

sequence fused with green fluorescence protein (GFP). Bright greenfluorescence was observed in the embryos microinjected with GFP

construct only (a) or co-injected with GFP construct and control

morpholino (b). Little or no GFP fluorescence was observed in larvae

co-injected with GFP construct and pnn morpholino (c). C RT-PCR

analysis of pnn transcripts from pnn morphants treated with MO-2.

Embryos injected with MO-2 exhibited aberrantly spliced products

(560 bp) as compared to that (464 bp) from wild-type (WT) embryos

130 Histochem Cell Biol (2012) 138:127–140

123

EGFP in the vascular endothelium of the zebrafish of Fli

promoter-driven lines makes it possible to perform long-

term observation of angiogenesis in developing living

embryos (Lawson and Weinstein 2002). In order to under-

stand whether pnn plays a role in vascular morphogenesis,

we knocked down pnn in a Fli-transgenic fish line. In normal

embryos, primary sprouts of blood vessels (intersegmental

vessels, ISV) emerged exclusively from the dorsal aorta

(DA) and elongated dorsally adjacent the vertical myotomal

boundaries. These sprouts ramified into rostral and caudal

branches, which extended in a T-shape, along the dorsolat-

eral roof of neural tube and then interconnected to form

continuous dorsal longitudinal anastomotic vessels (DLAV)

(Fig. 4a). In pnn morphants, the dorsal aorta exhibited a

more complex network than wild-type larvae (arrows in

Fig. 4b, c). Formation of the ISVs occurred normally, and

the DLAV was able to form as in the wild type (Fig. 4b, c).

However, instead of formation of the T-junction between

ISVs and DLAV in wild-type embryos, we observed a

network of aberrant branches in the morphant embryos

(arrowheads in Fig. 4b, c). This indicated that pnn affected

the angiogenesis of zebrafish.

Knockdown of pnn resulted in craniofacial defects

in zebrafish

According to the data presented above, in pnn morphants,

several derivatives originated from neural crest had defects,

such as cardiac effusion and abnormal expression patterns of

pigmentation. As well as contributing to pigment cells,

neural crest cells also contribute to most of the cartilage of

the skull, face and pharyngeal skeleton (Minoux and Rijli

2010). To determine the extent of pnn loss of function on

chondrogenesis, we next stained the cartilage patterning

with Alcian blue, which binds proteoglycan of the chon-

drogenic extracellular matrix after the injection of mor-

pholinos. The assay revealed highly specific abnormalities in

branchial cartilage development, with incomplete formation

Fig. 2 Phenotypes of pnn knockdown zebrafish. Lateral views of

2-day post-fertilization (dpf) larvae injected with control (a), pnn

MO-1 (b) and pnn MO-2 (c). Injection of pnn MOs caused mutant

zebrafish with cardiac edema (arrows in b and c) that lacked a swim

bladder (star in a). Defective pigmentation patterns can be observed

in pnn mutant fish. Pnn morphants exhibited not only thinner stripes

of pigment than those in the wild type (double arrows in a and c) but

also a spotted pattern (arrowheads in b) rather than the stripes seen in

the wild type. The insets represent enlarged images of the areas inside

the rectangles

Histochem Cell Biol (2012) 138:127–140 131

123

of the posterior five cartilages (Fig. 5). Closer examination

of Alcian blue-stained pnn-deficient embryos revealed that

the first two cartilages (mandibular and hyoid) were always

present, but were slightly shorter and misshapen (Fig. 5c, e)

compared to the wild-type larvae (Fig. 5a). However, the

cartilages of posterior pharyngeal arches 3–7 were severely

reduced or absent (Fig. 5c–f). Based on their relative loca-

tion with respect to the remaining skeleton, the small rem-

nants observed might represent the incomplete cartilages of

the posterior arches. To test whether the obvious phenotype

changes were not caused by the off-target effects of mor-

pholinos, we co-injected pnn mRNA with either MO-1 or

MO-2 to examine if the pnn RNA could rescue the cranio-

facial defects elicited by pnn depletion. The defects of

pharyngeal cartilages in pnn MO embryos were restored to a

great extent by injection of the pnn mRNA (Fig. 5g, h).

These results indicate that pnn knockdown specifically

affected the neural crest-derived craniofacial skeleton of

pharyngeal arches 3–7, while the first two pharyngeal

arches were only mildly affected. Together, the defective

arrangement of pigments and pharyngeal arches suggests

that pnn may be essential for the proper development of their

common precursors of neural crest origin.

Defects in neural crest migration in pnn morphants

To understand the function of pnn and determine the stage

of neural crest development at which pnn is required, we

analyzed defined steps which are critical for neural crest

cell specification. Neural crest progenitors are induced

during gastrulation and then begin to separate by express-

ing specific genes. One of the earliest genes expressed

during initial stage is sox10, a transcription factor charac-

teristic of neural crest progenitor cells. We studied

expression of this gene to assay neural crest specification in

pnn mutants as compared to wild-type embryos. In pnn-

deficient larvae, there was no change in the expression of

sox10 at the 8-somite stage (ss) (Fig. 6Ag, h). We also

examined this marker at a later stage of development.

At the 10-somite stage, when onset of migration occurs, we

Fig. 3 Phenotypes of zebrafish larvae after rescue of pnn MO with

mRNA. Lateral views of 2 dpf larvae injected with control (a), pnn

MO-2 (b), pnn MO-2 and zebrafish pnn RNA (c). To confirm the

specificity of the phenotype caused by the morpholinos, we carried

out rescue experiments with synthetic mRNA for the pnn gene. We

examined the rescued phenotypes by morphological examination at

2 dpf. Cardiac edema (arrow in b) was not seen in morphants rescued

with pnn RNA. In addition, in rescued morphants, the pattern and

amount of pigment was quite similar to that in the wild type. These

observations indicate that the defects in heart and pigment were

specifically caused by knockdown of pnn

132 Histochem Cell Biol (2012) 138:127–140

123

saw a slight reduction in sox10 expression in pnn mor-

phants (Fig. 6Aa, b, B). Co-injection of pnn mRNA and

MO restored the expression of sox10 to a level comparable

to that in wild-type embryos (Fig. 6Ac, B). During early

development, zebrafish hindbrain is transiently partitioned

into seven rhombomeres. The organization of rhombo-

meres is highly related to the numbers and migration

patterns of neural crest cells (Lumsden et al. 1991; Trainor

and Krumlauf 2000). Thus, anteroposterior patterning could

conceivably alter the number of neural crest cells and their

derivatives. We labeled the embryos at 10 ss for krox20, a

gene that has been shown to localize in rhombomeres 3 and 5.

As shown in Fig. 6A, we did not detect any differences in the

placement, size or compartmentalization of rhombomeres 3

and 5 between wild-type and pnn morphants (Fig. 6Ai, j).

Collectively, these data indicate that the defects in neural

crest derivatives caused by pnn knockdown were not due to

changes in neural crest induction or hindbrain patterning.

As pnn morphants were deprived of chondrogenic

derivatives, while induction of neural crest cells appeared

normally, we next tried to identify whether the chondro-

genic progenitors were eliminated at the posterior pha-

ryngeal arches. The chondrogenic neural crest cells begin

descending toward the pharyngeal arches at the 18-somite

stage. We, thus, investigated the expression pattern and

levels of sox9a, a later chondrocyte differentiation marker.

At 26 hpf, we found that the expression of sox9a in pnn-

deficient embryos was slightly reduced in the cranial crest

Fig. 4 Pattern of distribution of blood vessels in pnn morphants.

Lateral views of fli-EGFP larvae at 2 dpf either uninjected (a) or

injected with pnn MO-1 (b) or pnn MO-2 (c). Wild-type siblings

displayed fully formed intersomitic vessels (ISV), which connected

with the dorsal aorta (DA) and the dorsal longitudinal anastomitic

vessels (DLAV). In pnn morphants, there were more blood vessels in

the ventral region (arrows in b and c) and aberrant blood vessel

branches in the trunk (arrowheads in b and c). The insets showed the

variation in trunk vessels at higher magnification. DA dorsal aorta,

DLAV dorsal longitudinal anastomotic vessels, ISV intersomitic

vessels

Histochem Cell Biol (2012) 138:127–140 133

123

cells populating the first and second streams (Fig. 6Ae, s1,

s2) on route to the mandibular and hyoid arches, respec-

tively. However at this stage, the postotic crest cells sup-

plying the posterior pharyngeal arches 3–7 were almost

undetectable (Fig. 6Ae, arrow). The specificity of this

knockdown effect can be demonstrated again by co-injec-

tion of pnn MO with pnn mRNA, which restored signifi-

cantly the expression level of sox9a at 26 hpf (Fig. 6Af).

This suggests that in contrast with wild-type embryos pnn

MO-injected embryos had few or no migratory neural crest

Fig. 5 Effects of knockdown of pnn on craniofacial cartilages.

Ventral (a, c, e, g) and lateral (b, d, f, h) views of 5-dpf larvae

injected with control (a, b), pnn MO-1 (c, d), pnn MO-2 (e, f) and

both pnn MO and mRNA. The upper two figures are ventral (left) and

lateral (right) views, respectively, of craniofacial skeleton arrange-

ment in wild-type zebrafish. To determine the extent of pnn loss of

function on chondrogenesis, cartilage patterning was assayed by

Alcian blue staining. Examination of pnn morphants showed reduced

and malformed cartilages. The first two pharyngeal cartilages (m and

h) had mild defects; however, the third to seven branchial cartilages

(bc) showed severe defects. Most of the posterior branchial cartilages

were reduced in size (arrows in c–f) or absent. The defects of

pharyngeal cartilages in pnn morphants were rescued by injection

with the pnn mRNA (g, h). Anterior is to the left and posterior is to

the right. bc branchial cartilages, e ethmoid plate, h hyoid, m man-

dibular, t trabeculae cranii

134 Histochem Cell Biol (2012) 138:127–140

123

cells in the postotic region. This was also consistent with

our earlier observation that posterior branchial arch 3–7

cartilages in pnn morphants were severely malformed, but

only minor defects occurred in the first two branchial

cartilages (Fig. 5). Taken together, these results suggest

that pnn might not function in neural crest specification and

patterning, but rather maintain specific cranial neural crest

cell migration and proliferation/survival.

Pnn knockdown reduced cell proliferation

in the hindbrain

Increased cell death and/or reduced proliferation of neural

crest cells may explain the disruption in neural crest

development after pnn knockdown. To evaluate whether

cranial neural crest cell reduction (Fig. 6Aa, b) in pnn

morphants could be due to increased apoptosis, we used

TUNEL assay to visualize the apoptotic cells. To confirm

that the apoptotic cells were of neural crest origin, we

double stained the embryos with sox10 by in situ hybrid-

ization. At the 10-somite stage, we observed many apop-

totic cells in the hindbrain region of wild-type larvae. Some

of these cells were identified as neural crest cells by colo-

calization with the sox10 positive cells (Fig. 7Aa). A sim-

ilar co-localization pattern was also found in pnn morphants

(Fig. 7Ab). After comparing the numbers of double positive

cells in wild-type and pnn morphants, we found there was

little or no statistically significant difference in the number

of dying cells originating from the neural crest between

each group (Fig. 7B, P \ 0.3). Thus, reduced pnn expres-

sion might not function in neural crest cell survival.

Neural crest cells undergo extensive proliferation before

and after migration. Since cellular apoptosis was not

increased in pnn morphants, the decreased number of cells

could be caused by a defect in proliferation. To further

investigate this possibility, we stained proliferating cells in

the larvae by immunolabeling with anti-phospho-histone 3

(pH 3) antibody, and also by double-labeling the neural

crest-originated cells by sox10 in situ hybridization. In

wild-type embryos, we found prominent cell proliferation

occurred at 10-somite stage in hindbrain region (Fig. 7Ac).

However, in pnn morphants, the cell proliferation of neural

crest cells was significantly decreased (Fig. 7Ad, B,

P \ 0.001), and the decreased cell proliferation could be

sustained to the 18-somite stage (ss) (Fig. 7Ae, f, B,

P \ 0.005). Thus, pnn knockdown led to a reduction in the

proliferation of neural crest precursors.

During migration, crest cells continue to proliferate and

also begin to express lineage-defined specific genes. The

reduction of neural crest cell proliferation at 10-somite

stage might thus be due to premature differentiation into

other crest derivatives. To investigate this possibility, we

examined the expression of mitfa, an early marker of the

melanoblast lineage, and sox9a at 10-somite stage. How-

ever, in pnn MO-injected embryos, the expression levels

of both mitfa and sox9a were not different from those in

wild-type embryos (Fig. 7C), suggesting pnn knockdown-

induced decreased cell proliferation is not caused by crest

cell premature differentiation. These findings demonstrate

that in the absence of pnn, neural crest cells maintain their

ability to undergo specification and migrate in an orderly

pattern, but a subset of population, especially those in

postotic streams, was unable to proliferate and proceed to

the later differentiation process.

Discussion

Pnn was first isolated from Madin–Darby canine kidney

(MDCK) cells and characterized as a desmosome-associated

protein (Ouyang and Sugrue 1992, 1996), and later found to

regulate many biological processes such as cell–cell adhe-

sion, tumor suppression and alternative splicing. Targeted

disruption of pnn in mice revealed that loss of function leads

to dramatic phenotypes. Pnn hypomorphic mice exhibit cleft

palate, ventricular septal defects and cardiac outflow tract

defects (Joo et al. 2007); malformations that originate from

neural crest development. However, perinatal embryonic

lethality following knockdown of pnn in mice made it dif-

ficult to identify the specific role of pnn in neural crest

development. Here, we investigated the role of pnn in zeb-

rafish and demonstrated that pnn is required for neural crest

development through promotion of cell proliferation.

After knockdown of pnn by injection with morpholi-

nos, pnn-deficient embryos exhibited smaller hearts with

enlarged pericardial space (Fig. 2). Previous fate-mapping

studies demonstrated that the zebrafish cardiac neural crest

cells, which originate from the first rhombomere in the

hindbrain region extending to the caudal boundary of

the sixth somite, contribute to all segments of the heart,

including the bulbus arteriosus, ventricle, AV junction and

atrium (Sato and Yost 2003). Depletion of pnn in zebrafish

thus resulted in heart defects similar to those seen in pnn

hypomorphic mice (Joo et al. 2007). Collectively, these data

suggest that pnn takes part in the development of the heart,

which is derived from a subpopulation of neural crest cells.

In zebrafish, lateral pre-migratory neural crest cells that

migrate early form neurons, while cartilage and pigment

cells are derived from late-migrating medial crest cells

(Schilling and Kimmel 1994). To examine whether pre-

cursors of the neural crest were affected by pnn knock-

down, embryos were labeled by probes against sox10.

Interestingly, sox10 was normally expressed at the early

stage in the absence of pnn (Fig. 6Ag, h). In addition,

neural crest cells were able to migrate in the absence of

pnn, as seen in the first two pharyngeal arches and other

Histochem Cell Biol (2012) 138:127–140 135

123

136 Histochem Cell Biol (2012) 138:127–140

123

derivatives (Fig. 6Ad, e). We thus concluded that the early

steps of induction and specification of neural crest cells

were not dependent on pnn. However, a sudden change

began at the 10-somite stage when loss of pnn resulted in

cartilage malformation. In the absence of pnn, the number

of sox10 positive neural crest cells was slightly reduced at

the beginning of migration (Fig. 6Aa, b, B). At the later

stage, the migratory neural crest cells failed to express the

cartilage lineage-specific gene sox9a at certain regions along

the hindbrain. Specifically, we observed that neural crest

cells in the hindbrain commenced migration but fewer

reached their final destination at the posterior pharyngeal

arches and differentiated into cartilages. Thus, pnn is

required during a specific time window spanning the interval

between crest cell migration and differentiation.

To our surprise, the neural crest was differentially

affected in pnn morphants. For example, the cartilages of the

first two arches (mandibular and hyoid) were formed in

mutants while posterior pharyngeal arches 3–7 were greatly

reduced or even absent. Likewise, the pattern of sox9a

appeared to be mildly affected in the neural crest contrib-

uting to the mandibular and hyoid arches (first and second

neural crest streams) in pnn-depleted embryos, but strongly

reduced in more posterior neural crest (postotic stream) that

formed arches 3–7. These data suggested that sox9a

expression may be regulated by pnn in posterior pharyngeal

arches but is under the control of different regulatory

pathways in the first two pharyngeal arches.

In addition to defects in chondrogenesis, pnn morphants

exhibited altered pigmentation. In normal zebrafish, speci-

fication of pigment cells from neural crest precursors

occurred before migration. During migration, melanoblasts

continue proliferating and are susceptible to environmental

cues for patterning. After knockdown of pnn, embryos

displayed a reduced number and disorganized distribution

of melanocytes. However, the number of neural crest cells

before the onset of migration remained normal (Fig. 6Ag, h

vs. a, b), hinting that the altered numbers of melanophores

and their dispersal were due to a defect in proliferation/

survival of progenitor cells during migration. In general, we

observed prominent defective development in craniofacial

cartilages and pigments in pnn morphants, which is derived

from late-migrating neural crest cells. Nevertheless, the

neural crest-derived neurons of the dorsal root ganglion

(early migration origin) were barely disrupted by pnn

deficiency (data not shown). Therefore, neural crest-derived

cells are differentially affected in pnn morphants suggesting

that pnn is only required during a short period of neural

crest development, perhaps at the late migration stage.

Neural crest cells arise from the border of the neural

ectoderm and migrate to multiple targets in embryos (Le

Douarin and Kalcheim 1999). Throughout their migration

and differentiation, neural crest cells undergo cell division

(Kalcheim and Burstyn-Cohen 2005). Precise regulation of

neural crest cell proliferation is crucial for the develop-

mental processes of neural crest cells. In this study, we

found that knockdown of pnn caused a reduction in

migratory neural crest cells and defective formation of a

subset of neural crest derivatives. These effects occurred

without any impact on premigratory neural crest cells or on

patterning of the hindbrain. To explore the mechanism by

which pnn deficiency reduced numbers of migratory cells

and caused subsequent defects in neural crest derivatives,

we compared the ratio of apoptosis and proliferating neural

crest cells in embryos with or without morpholino treat-

ment and found that knockdown of pnn resulted in

decreased proliferation of neural crest cells (Fig. 7A, B).

We also excluded the possibility that decreased cell pro-

liferation may be caused by premature differentiation

(Fig. 7C). Thus it appears that pnn plays an important role

in the regulation of cell proliferation in the development of

neural crest derivatives.

In addition to neural crest derivatives which were

affected in pnn morphants, we also examined angiogenesis

using Fli promoter-driven EGFP transgenic fish. In zebra-

fish, new blood vessels are generated by proliferation and

migration of endothelial cells, which form sprouts under

the influence of vascular endothelial growth factor (VEGF)

and many other signals (Coultas et al. 2005). During the

normal course of development, a subset of endothelial cells

becomes highly motile and forms sprouts. Once the con-

nections have been made, the endothelial cells halt their

mobility (Childs et al. 2002). To create a vascular network

with the right density of branches, sprouting behavior must

be strictly controlled. The abnormality of ectopic branches

of blood vessels in pnn-deficient embryos was very similar

to that seen in the vasculature of zebrafish embryos lacking

Fig. 6 Effects of knockdown of pnn on neural crest cell specification

and migration. A In situ hybridization analysis of sox10 (a–c, g, h),

sox9a (d–f) and krox20 (i, j) in zebrafish that were untreated (WT, a, d,

g, i), treated with pnn MO (b, e, h, j), or with both pnn MO and pnn

mRNA (c, f). Sox10 expression patterns were indistinguishable in wild-

type and pnn morphants at the 8-somite stage (ss) (g, h); however,

expression was slightly reduced in pnn knockdown embryos at the

10-somite stage (a, b). Note the comparable expression of krox20 in

hindbrain rhombomeres 3 and 5 in wild-type (i) versus pnn morphants

(j) at the 10-somite stage. In pnn-depleted embryos at 26 hpf,

expression of sox9a was obviously reduced in the hindbrain region

as compared to WT embryos (d, e). Embryos co-injected with pnn MO

and pnn mRNA restored the expression of sox9a at 26 hpf (f). All

panels show dorsal views of flat-mount preparation, and anterior is

toward the superior. The arrows in d–f indicate the relative position of

the postotic neural crest. B Quantification of the number of sox10positive cells in the dorsal central region of the hindbrain with or

without MO treatment or rescue with pnn mRNA at the 10-somite

stage. e eye, hb hindbrain, mb midbrain, ov otic vesicle, r3 rhombomere

3, r5 rhombomere 5, s1 first neural crest stream, s2 s neural crest

stream

b

Histochem Cell Biol (2012) 138:127–140 137

123

Delta-like 4 (Dll4), a Notch signaling ligand (Leslie et al.

2007). In Dll4-morphant embryos, a network of aberrant

interconnected branches was formed. The Dll4-Notch sig-

nals acted as an angiogenic ‘‘off’’ switch by rendering the

endothelial cells unresponsive to VEGF (Leslie et al. 2007).

Supporting this hypothesis, VEGF has been shown to induce

Dll4 expression, which, in turn, inhibits the expression of

VEGF receptors in mammalian cells (Williams et al. 2006).

Analogous to the relationship between VEGF and Dll4,

microarray analysis of VEGF-A-induced human microvas-

cular endothelial cells showed a delayed increase of pnn

expression during vascular network formation (Rennel et al.

2007), implying that pnn maybe activated after VEGF

expression in order to negatively feedback regulate the

angiogenesis by VEGF. These findings suggest that pnn, as

well as the Dll4-Notch signaling, might be involved in

controlling the transition between motility and quiescence in

endothelial cells exposed to VEGF. In addition, previous

studies showed that conditional knockout of pnn in certain

kinds of tissues led to malformation of the organs through

the upregulation of Tcf/Lef reporter activity (Joo et al. 2007,

2010a, b). Several reports implied that the Wnt signaling

Fig. 7 Effects of loss of pnn on

neural crest cell survival and

proliferation. A Flat-mount

dorsal views of 10-somite stage

embryos (a–d) or 18-somite

stage (e, f) injected with control

(a, c, e) and pnn morpholino

(b, d, f). Sox10 in situ

hybridization was carried

out in combination with either

TUNEL assay (a, b) or

phosphohistone 3 (pH 3)

immunostaining (c–f). TUNEL

analysis showed similar

numbers of apoptotic cells in

sox10 positive populations in

wild-type and pnn knockdown

embryos. Phosphohistone 3

immunofluorescence revealed a

marked decrease in proliferative

cells in pnn morphants. Double

positive cells are indicated by

arrows. The insets represent

enlarged images shown by the

arrow. Anterior is toward the

superior. B Histogram

comparing numbers of neural

crest-originated apoptotic/or

proliferative cells in wild-type

larvae (white) and pnn

morphants (black), respectively.

**P \ 0.005, ***P \ 0.001,

t test. C In situ hybridization

analysis of mitfa (a, b) and

sox9a (c, d) expression in

zebrafish embryos that were

untreated (a, c) or treated with

pnn MO (b, d). e eye, hbhindbrain, mb midbrain, ov otic

vesicle

138 Histochem Cell Biol (2012) 138:127–140

123

pathway might promote the sprouting of blood vessels by

upregulating the VEGF-A (Zhang et al. 2001; Franco et al.

2009). These findings raise the further possibility that

knockdown of pnn could stimulate the VEGF signaling to

promote sprouting of endothelial cells by upregulating the

Wnt signaling involving the Tcf/Lef. Nevertheless, further

experiments will be needed to establish the connections

between them.

Previous work has shown that Wnt/Tcf signaling plays

stage-specific roles during neural crest development. Wnt8

is needed during the initial phase of neural crest formation

independent of anterior/posterior patterning (Lewis et al.

2003). In mouse embryos, loss of wnt1 and wnt3a affected

neural crest expansion; however, the proliferation of neural

crest cells was unaffected (Ikeya et al. 1997). Wnt9a is

required for zebrafish palate and lower jaw development

(Curtin et al. 2011). Furthermore, Wnt/Tcf signaling con-

trols the cell cycle in neural crest cells (Burstyn-Cohen

et al. 2004). Conditional knockout of pnn in mice resulted

in upregulated Tcf/Lef reporter activity, as well as misreg-

ulated expression of ß-catenin and Tcf (Joo et al. 2010a, b).

Our results with pnn morphant fish embryos revealed

misshapened cartilage and pigment organization as well as

aberrant vascular formation, developmental processes that

relate to events regulated by diverse Wnt signaling. It

appears to be common for a given signaling pathway to

play different roles in different places and at different times

throughout the development processes. Thus, understand-

ing how pnn regulates the distinct steps in neural crest

development might unravel the distinct mechanism

responsible for Wnt/Tcf signaling during the development

of diverse subsets of neural crest derivatives.

Acknowledgments We are grateful to the Taiwan Zebrafish Core

facility at ZeTH and Zebrafish Core in Academia Sinica for providing

fish. This study was supported by Chang Gung Memorial Hospital

(CMRPD 170073), the Ministry of Education (Tope Center Grant,

EMRPD 190481) and the National Science Council, ROC (NSC-98-

2320-B-182-019-MY3).

References

Alpatov R, Munguba GC, Caton P, Joo JH, Shi Y, Hunt ME, Sugrue

SP (2004) Nuclear speckle-associated protein Pnn/DRS binds to

the transcriptional corepressor CtBP and relieves CtBP-mediated

repression of the E-cadherin gene. Mol Cell Biol 24:10223–

10235

Brandner JM, Reidenbach S, Franke WW (1997) Evidence that

‘‘pnn’’, reportedly a differentiation-specific desmosomal protein,

is actually a widespread nuclear protein. Differentiation

62:119–127

Brandner JM, Reidenbach S, Kuhn C, Franke WW (1998) Identifi-

cation and characterization of a novel kind of nuclear protein

occurring free in the nucleoplasm and in ribonucleoprotein

structures of the ‘‘speckle’’ type. Eur J Cell Biol 75:295–308

Burstyn-Cohen T, Stanleigh J, Sela-Donenfeld D, Kalcheim C (2004)

Canonical Wnt activity regulates trunk neural crest delamination

linking BMP/noggin signaling with G1/S transition. Develop-

ment 131:5327–5339

Childs S, Chen JN, Garrity DM, Fishman MC (2002) Patterning of

angiogenesis in the zebrafish embryo. Development 129:973–982

Chiu Y, Ouyang P (2006) Loss of Pnn expression attenuates

expression levels of SR family splicing factors and modulates

alternative pre-mRNA splicing in vivo. Biochem Biophys Res

Commun 341:663–671

Coultas L, Chawengsaksophak K, Rossant J (2005) Endothelial cells

and VEGF in vascular development. Nature 438:937–945

Curtin E, Hickey G, Kamel G, Davidson AJ, Liao EC (2011)

Zebrafish wnt9a is expressed in pharyngeal ectoderm and is

required for palate and lower jaw development. Mech Dev

128:104–115

Franco CA, Liebner S, Gerhardt H (2009) Vascular morphogenesis: a

Wnt for every vessel? Curr Opin Genet Dev 19:476–483

Ikeya M, Lee SM, Johnson JE, McMahon AP, Takada S (1997) Wnt

signalling required for expansion of neural crest and CNS

progenitors. Nature 389:966–970

Joo JH, Alpatov R, Munguba GC, Jackson MR, Hunt ME, Sugrue S

(2005) Reduction of Pnn by RNAi induces loss of cell–cell

adhesion between human corneal epithelial cells. Mol Vis

11:133–142

Joo JH, Lee YJ, Munguba GC, Park S, Taxter TJ, Elsagga MY,

Jackson MR, Oh SP, Sugre S (2007) Role of Pnn in neural crest,

dorsal dermis, and axial skeleton development and its involve-

ment in the regulation of Tcf/Lef activity in mice. Dev Dyn

236:2147–2158

Joo JH, Kim YH, Munguba GC, Dunn NW, Sugrue S (2010a)

Disruption of mouse corneal epithelial differentiation by condi-

tional inactivation of Pnn. Invest Ophthalmol Vis Sci 51:1927–

1934

Joo JH, Lee YJ, Taxter TJ, Munguba GC, Kim YH, Dhaduvai K,

Dunn NW, Degan WJ, Oh SP, Sugrue S (2010b) Pnn modulates

expression of an intestinal homeobox gene, Cdx2, and plays an

essential role for small intestinal morphology. Dev Biol 345:

191–203

Kalcheim C, Burstyn-Cohen T (2005) Early stages of neural crest

ontology: formation and regulation of cell delamination. Int J

Dev Biol 49:105–116

Kimmel BC, Ballard WW, Kimmel RS, Bonnie U, Schilling FT

(1995) Stages of embryonic development of the zebrafish. Dev

Dyn 203:253–310

Lawson ND, Weinstein BM (2002) In vivo imaging of embryonic

vascular development using transgenic zebrafish. Dev Biol

248:307–318

Le Douarin N, Kalcheim C (1999) The neural crest. Cambridge

University Press, New York

Leslie JD, Ariza-McNaughton L, Bermange AL, McAdow R, Johnson

SL, Lewis J (2007) Endothelial signaling by the Notch ligand

Delta-like 4 restricts angiogenesis. Development 134:839–844

Leu S, Ouyang P (2006) Spatial and temporal expression profiles of

Pnn during mouse development. Gene Exp Pattern 6:620–631

Lewis JL, Bonner J, Modrell M, Ragland JW, Moon RT, Dorsky RI,

Raible DW (2003) Reiterated Wnt signaling during zebrafish

neural crest development. Development 131:1299–1308

Li C, Lin RI, Lai MC, Ouyang P, Tarn WY (2003) Nuclear Pnn/DRS

protein binds to spliced mRNPs and participates in mRNA

processing and export via interaction with RNPS1. Mol Cell Biol

23:7363–7376

Lin CL, Leu S, Lu MC, Ouyang P (2004) Over-expression of SR-

cyclophilin, an interaction partner of nuclear pnn, releases SR

family splicing factors from nuclear speckles. Biochem Biophys

Res Commun 321:638–647

Histochem Cell Biol (2012) 138:127–140 139

123

Lumsden A, Sprawson N, Graham A (1991) Segmental origin and

migration of neural crest cells in the hindbrain region of the

chick embryo. Development 113:1281–1291

Minoux M, Rijli FM (2010) Molecular mechanisms of cranial neural

crest cell migration and patterning in craniofacial development.

Dev Biol 137:2605–2621

Ouyang P (1999) Antibodies differentiate desmosome-form and

nucleus-form pnn: evidence that pnn is a moonlighting protein

with dual location at the desmosome and within the nucleus.

Biochem Biophys Res Commun 263:192–200

Ouyang P, Sugrue SP (1992) Identification of an epithelial protein

related to the desmosome and intermediate filament network.

J Cell Biol 118:1477–1488

Ouyang P, Sugrue SP (1996) Characterization of pnn, a novel protein

associated with the desmosome-intermediate filament complex.

J Cell Biol 135:1027–1042

Ouyang P, Zhen YY, Sugrue SP (1997) Cloning and analysis of

cDNA encoding murine pnn. Gene 197:115–120

Rennel E, Mellberg S, Dimberg A, Petersson L, Botling J, Ameur A,

Westholm JO, Komorowski J, Lassalle P, Cross MJ, Gerwins P

(2007) Endocam as a VEGF-A and PI3K regulated gene with

increased expression in human renal cancer. Exp Cell Res

313:1285–1294

Sato M, Yost HJ (2003) Cardiac neural crest contributes to

cardiomyogenesis in zebrafish. Dev Biol 257:127–139

Schilling TF, Kimmel CB (1994) Segment and cell type lineage

restrictions during pharyngeal arch development in the zebrafish

embryo. Development 120:483–494

Schilling TF, Piotrowski T, Grandel H, Brand M, Heisenberg CP, Jiang

YJ, Beuchle D, Hammerschmidt M, Kane DA, Mullins MC, van

Eeden FJ, Kelsh RN, Furutani-Seiki M, Granato M, Haffter P,

Odenthal J, Warga RM, Trowe T, Nusslein-Volhard C (1996) Jaw

and branchial arch mutants in zebrafish I: branchial arches.

Development 123:329–344

Shi Y, Tabesh M, Sugrue SP (2000a) Role of cell adhesion-associated

protein, pnn (DRS/memA), in corneal epithelial migration.

Invest Ophthalmol Vis Sci 41:1337–1345

Shi Y, Ouyang P, Sugrue SP (2000b) Characterization of the gene

encoding pnn/DRS/memA and evidence for its potential tumor

suppressor function. Oncogene 19:289–297

Shi Y, Simmons MN, Seki T, Oh SP, Sugrue SP (2001) Change in

gene expression subsequent to induction of Pnn/DRS/memA:

increase in p21(cip1/waf1). Oncogene 20:4007–4018

Thisse C, Thisse B (2008) High-resolution in situ hybridization to

whole-mount zebrafish embryos. Nat Protocols 3:59–69

Trainor PA, Krumlauf R (2000) Patterning the cranial neural crest:

hindbrain segmentation and hox gene plasticity. Nat Rev Neurosci

1:116–124

Wang P, Lou PJ, Leu S, Ouyang P (2002) Modulation of alterative

pre-mRNA splicing in vivo by pnn. Biochem Biophys Res

Commun 294:448–455

Westerfield M (2000) The zebrafish book, A guide for the laboratory use

of zebrafish (Danio rerio). University of Oregon Press, Eugene

Williams CK, Li JL, Murga M, Harris AL, Tosato G (2006) Up-

regulation of the Notch ligand Delta-like 4 inhibits VEGF-

induced endothelial cell function. Blood 107:931–939

Zhang X, Gaspard JP, Chung DC (2001) Regulation of vascular

endothelial growth factor by the Wnt and K-ras pathways in

colonic neoplasia. Cancer Res 61:6050–6054

Zimowska G, Shi J, Munguba G, Jackson MR, Alpatov R, Simmons

MN, Shi Y, Sugrue SP (2003) Pnn/DRS/memA interacts with

SRp75, SRm300 and SRrp130 in corneal epithelial cells. Invest

Ophthalmol Vis Sci 44:4715–4723

140 Histochem Cell Biol (2012) 138:127–140

123