Defects in GPI biosynthesis perturb Cripto signaling
during forebrain development in two new mouse
models of holoprosencephaly
David M. McKean and Lee Niswander*
HHMI, Department of Pediatrics, Cell Biology, Stem Cells and Development Graduate Program, and Children’s Hospital Colorado, University of
Colorado Anschutz Medical Campus Aurora, CO 80045, USA
*Author for correspondence (Lee.Niswander@ucdenver.edu)
Biology Open 1, 874–883
Received 16th May 2012
Accepted 6th June 2012
Holoprosencephaly is the most common forebrain defect in
humans. We describe two novel mouse mutants that display a
in the glycerophosphatidyl inositol (GPI) biosynthesis pathway:
gonzo disrupts Pign and beaker disrupts Pgap1. GPI anchors
normally target and anchor a diverse group of proteins to lipid
raft domains. Mechanistically we show that GPI anchored
Disruption of the GPI-anchored protein Cripto (mouse) and
TDGF1 (human ortholog) have been shown to result in
holoprosencephaly, leading to our hypothesis that Cripto is the
key GPI anchored protein whose altered function results in an
HPE-like phenotype. Cripto is an obligate Nodal co-factor
involved in TGFb signaling, and we show that TGFb signaling is
reduced both in vitro and in vivo. This work demonstrates the
importance of the GPI anchor in normal forebrain development
and suggests that GPI biosynthesis genes should be screened for
association with human holoprosencephaly.
? 2012. Published by The Company of Biologists Ltd. This is
an Open Access article distributed under the terms of the
Creative Commons Attribution Non-Commercial Share Alike
Key words: Holoprosencephaly (HPE), forebrain,
glycerophosphatidyl inositol, GPI, Pign, Pgap1, TGFb, Cripto, Nodal
Holoprosencephaly (HPE) is estimated to occur in 1 in 250
pregnancies but due to prenatal lethality, only 1 in 16,000 babies
are born with HPE (Muenke and Beachy, 2000). HPE occurs
when the forebrain fails to septate into the two frontal lobes. The
spectrum of septal defects that result in HPE are classified as
lobar (fully septated, mild), semi-lobar (incompletely septated,
moderate), alobar (not septated, severe) and syntelencephaly
(posterior frontal lobe and parietal lobe fail to septate). Infants
born with HPE, generally representing the mildest spectra, may
present with craniofacial defects, hydrocephaly, feeding issues
and early mortality.
Environmental causes include maternal diabetes and exposure to
alcohol, retinoic acid and cholesterol reducing agents. Large
chromosomal defects, such as trisomy 13 account for the majority
of HPE cases, but single gene disruptions are also linked to HPE.
These single gene mutations segregate into the SHH pathway
(SHH, PTCH, GLI2, ZIC2 and DHCR7) and TGFb pathway
(TGIF, FAST1 and TDGF1), but only account for 15–20% of
human HPE (Ming and Muenke, 2002). SHH pathway mutations
are directly associated with midline defects whereas TGFb
pathway mutants are thought to act upstream of SHH, affecting
SHH expression and/or activity in the prechordal plate (Rohr et al.,
2001). In mouse, additional TGFb pathway genes (Nodal, Gdf1,
Smad2 and ActRIIA) have been linked to HPE (Nomura and Li,
1998; Song et al., 1999; Hoodless et al., 2001; Lowe et al., 2001;
Rohr et al., 2001; Yamamoto et al., 2001; Andersson et al., 2006)
and disruption of both copies of the TGFb genes often leads to
forebrain truncations, rather than classic HPE, due to the
requirement for TGFb signaling in early forebrain initiation steps.
Nodal, a TGFb ligand, and Cripto, Nodal’s obligate co-factor
(Gritsman et al., 1999) are both required for specification and
localization of the distal visceral endoderm (DVE) and anterior
visceral endoderm (AVE) (Varlet et al., 1997; Ding et al., 1998;
Mesnard et al., 2006; D’Andrea et al., 2008; Liguori et al., 2008;
Takaoka et al., 2011). The AVE is a transient organizing center
neuroectoderm. Although DVE progenitors initially require
Nodal signaling (Varlet et al., 1997; Mesnard et al., 2006),
proper migration of the DVE and AVE is only achieved by
antagonism of Nodal and Wnt signaling (Yamamoto et al., 2004;
Kimura-Yoshida et al., 2005). Furthermore, the AVE and future
forebrain organizing centers induce and then maintain forebrain
specification by antagonizing TGFb and Wnt signaling (Perea-
Gomez et al., 2002).
Cripto (the HPE gene TDGF1 in humans) protein is post-
translationally modified with a GPI anchor (Minchiotti et al.,
2000). This sugar-lipid anchor targets Cripto to the plasma
membrane where it binds Nodal to signal in a cell autonomous
function (Yan et al., 2002). Cripto’s GPI anchor may also be
cleaved, releasing Cripto into the extracellular space where it
may bind Nodal and signal non-cell autonomously (Yan et al.,
2002; Chu et al., 2005; Watanabe et al., 2007).
Here, we describe two novel recessive mutations in mouse,
which result in HPE or an anterior truncation phenotype, similar
to phenotypes associated with homozygous mutation of TGFb
genes. These mutations disrupt two different enzymes within the
GPI biosynthesis pathway. We hypothesize that Cripto is a key
GPI-anchored protein, whose lack of a functional GPI anchor
results in an HPE-like phenotype. We show that Nodal/Cripto
signaling is downregulated both in vitro and in vivo in the GPI
Materials and Methods
Mouse strains and genotyping
The gonzo (gnz) line was derived from a forward genetic screen performed in
collaboration with the laboratories of Kathryn Anderson and Elizabeth Lacy and
the beaker (bkr) line was generated from mutagenized males provided by Monica
Justice and the screen performed with the lab of Trevor Williams. Both lines were
generated on a C57BL/6J genetic background and initially out-crossed to either
C3H/HeJ (gnz) or 129SI/Sv1mJ (bkr) strains. Further analysis of gnz and bkr was
performed in 129SI/Sv1mJ and C57BL/6J backgrounds, respectively. Additional
strains used were NodalLacZ(Collignon et al., 1996), GPI-GFP (Rhee et al., 2006)
and Hex-GFP (Rodriguez et al., 2001).
Gonzo was initially mapped between SSLP markers D1MIT136 and D1MIT94.
Additional high-resolution markers were generated from NCBI Mouse SNP
database (http://www.ncbi.nlm.nih.gov/SNP/MouseSNP.cgi) and Mouse Genome
SSR search website (http://danio.mgh.harvard.edu/mouseMarkers/musssr.html).
Ultimately, gnz was genotyped using ARMS primers: TGCTTTCTTGTTACCTC-
CAGCTCACCAG (Pign outer forward), ATGACATCCGTAGGGCCTTTTTC-
CTAGAAA (Pign inner A forward), GGAAGATCTTAACAATCCCAGAGCA-
AAGGA (Pign inner T reverse) and GCACCTGCCATCTCCAAATTTTTGGT
(Pign outer reverse) (Ye et al., 2001). Beaker was initially mapped between
D1MIT123 and D1MIT303 and ultimately genotyped using SNP analysis with
primers: CCGTAGACCATTGCATTCAGCCAT (Pgap1 snpF), GCAATCCCTT-
CCAAATCACAAAGC (Pgap1 snpR), and TCCTTCCCACAAATACTTGGA-
CAGG (Pgap1 snp probe). Pigngnz(renamed Pignm1Nisw/J) and Pgapbkr(renamed
Pgap1m1Nisw/J) have been deposited and are available from Jackson Laboratories.
GFP strains were genotyped using primers oIMR0872 and oIMR1416 and LacZ
identified using oIMR0039 and oIMR0040 (Jackson Labs).
Mouse embryo fibroblasts (MEFs), cell culture
MEFs were prepared from E13.5 embryos. Embryos were dissected in DMEM
(Gibco), eviscerated, decapitated and minced with a sterile blade, followed by
trypsination for up to 30 min. MEFs were maintained in 10% FBS (Gibco) in
DMEM with Penicillin/Streptomycin (Gibco).
For immunostaining, 12,000 MEFS (Pign+/+or Pigngnz/gnz) were plated on
coverslips in 24-well plates (Falcon). MEFs were grown in 10% FBS in DMEM,
fixed in 4% paraformaldehyde then analyzed by immunodetection for bCOP
For in vitro Nodal/Cripto signaling assays, 100,000 MEFs (Pign+/+, Pigngnz/gnz,
Pgap1+/+or Pgap1bkr/bkr) were plated in 6-well plates in 10% FBS in DMEM.
Cells were transfected with pCDNA3-HA-Cripto (Yan et al., 2002) or pCDNA3.1
(Invitrogen) using Lipofectamine 2000 Transfection Reagent (Invitrogen) for
4 hours, then medium changed to DMEM (no serum) overnight. MEFs were
treated with 250 ng/ml recombinant Nodal protein (R&D Systems) or vehicle for
1 hour; protein extracts were prepared and separated by SDS-PAGE.
For detection of Cripto from conditioned medium, 100,000 MEFs (Pign+/+,
Pigngnz/gnz, Pgap1+/+or Pgap1bkr/bkr) were plated in 6-well plates in 10% FBS in
DMEM. Medium was changed to 2 ml DMEM (no serum) overnight. Conditioned
medium was spun down to remove cellular debris, 1.8 ml conditioned medium was
pre-cleared with 50 ml Protein A/G agarose (Santa Cruz) for 4 hours at 4˚C, then
0.4 mg/ml anti-Cripto antibody was added and complexes were rocked 1 hour at
4˚C. 50 ml Protein A/G agarose was added and complexes were rocked overnight
at 4˚C. Complexes were spun down and washed twice with 50 mM Tris (pH 7.5),
150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, and protease inhibitors,
then twice with 50 mM Tris (pH 7.5), 500 mM NaCl, 0.1% NP-40 and 0.05%
sodium deoxycholate and then twice with 50 mM Tris (pH 7.5), 0.1% NP-40 and
0.05% sodium deoxycholate. For each wash, tubes were rotated for 20 minutes at
4˚C followed by brief centrifugation to pellet the Protein A/G agarose and bound
complexes. 16 Laemmli buffer was added and immunoprecipitated conditioned
medium was boiled for 10 min prior to analysis by SDS-PAGE.
Western blots were performed with Smad2, phosphor-Smad2 (Cell Signaling
3122 and 3101) and Cripto (R&D Systems AF1538) antibodies. Blots were
performed in triplicate and quantitated on Bio-Rad gel documentation system.
Real time polymerase chain reaction
Total RNA was extracted from E7.5 embryos using an RNeasy Micro Kit
(QIAgen). Six wildtype littermates and six Pigngnz/gnzor Pgap1bkr/bkrembryos
were used for real time PCR with the Mouse TGFb BMP Signaling Pathway RT2
Profiler PCR Array (SABiosciences) using 200 ng of reverse transcribed RNA as
Characterization of the ENU-derived Gonzo mutant mouse line
To identify novel genes important for normal forebrain
development, we employed an ENU mutagenesis screen in
mice. Briefly, mutagenized C57BL/6J (C57) males were out-
crossed to C3H/HeJ (C3H) females to generate founder males.
Founder males were further out-crossed, and then mated to their
daughters to produce litters that would include homozygous
mutant embryos. From this screen, we identified mutant embryos
with forebrain truncations or an HPE-like phenotype. Due to the
presence of a large proboscis that dominated the craniofacial
region we named this line gonzo (gnz).
In the C3H background, gnz mutant embryos show three
different phenotypes: dysmorphic eyes (n514/39 embryos)
(supplementary material Fig. S1A,B), gastrulation defects (9/
39) (supplementary material Fig. S1I,J), and an HPE-like
phenotype in which mutant embryos either display midline
defects (n52/39) (supplementary material Fig. S1K,L) or ante-
rior truncations (n514/39) (supplementary material Fig. S1E,F;
Fig. 1A,D). When out-crossed into 129S1/Sv1mJ (129S1) back-
ground, gnz mutant embryos largely present with anterior
truncations. While the C3H strain was used for mutation
mapping, the 129S1 strain was used for the majority of results
and figures presented here.
A typical gnz mutant at embryonic day 18.5 (E18.5) shows
midbrain/forebrain truncation and a large proboscis (Fig. 1D). To
identify the rostral-caudal position of the truncation, bone and
cartilage staining was performed (Fig. 1B,C,E,F). In gnz mutant
embryos, bone and cartilage structures anterior to the interparietal
(ip) element remain symmetrical yet are grossly misshapen. The
ip, while recognizable, is tiny and misshapen. The tectum
synoticum (tso) is largely intact, but narrower at the anterior
midline. Occipital elements at the base of the skull are also
dysmorphic. The supraoccipital (so) and exoccipital (eo)
elements are fused and the supraoccipital is present as two
lateral bony elements instead of the normal rod-like shape. The
basioccipital (bo) element appears normal in gnz mutants.
In addition to the anterior truncation, gnz mutant embryos are
smaller than wildtype littermates (supplementary material Fig.
S2A,B). Comparison of the lengths of femurs and humeri from
five mutant and wildtype E18.5 embryos (supplementary material
Fig. S2C,D,I) reveal a statistically significant shortening of the
femur (by 11%, P value50.04; unpaired, 2-tailed Student’s t-test)
but not the humerus (P value50.24) in gnz mutant embryos.
Gnz mutation disrupts Pign, encoding a glycerophosphatidyl
inositol biosynthesis enzyme
We mapped the gnz mutation to a 5 Mb region on mouse
chromosome 1 using Simple Sequence Length Polymorphisms
(SSLPs) and Single Nucleotide Polymorphisms (SNPs) that differ
between the mutagenized C57 DNA and the out-crossed C3H DNA
(Fig. 1G) (between rs3654716 and rs6307336). This gene sparse
region contained seven known or predicted genes. Sequencing of
these seven genes revealed a single base substitution (T to A) in the
splice donor region of intron 23 of the Phosphatidylinositol-glycan
HPE and Cripto disruption in GPI mutants 875
biosynthesis class N (Pign) gene (Fig. 1H). Pign encodes one of 27
endoplasmic reticulum (ER) localized enzymes that are required to
make GPI anchors (Kinoshita et al., 2008). Specifically, Pign adds
phosphoethanolamine to the first mannose in the GPI anchor
(supplementary material Fig. S3) (Gaynor et al., 1999; Hong et al.,
While the T to A transversion in intron 23 of Pign would be
predicted to create a more optimal splice consensus site, splicing
is nevertheless disrupted in gonzo mutant embryos. PCR
amplification of gnz HPE-like mutant cDNA yielded a shorter
amplicon than wildtype, and cDNA from a heterozygous embryo
yielded both products (Fig. 1I). Sequencing of the gnz mutant
amplicon revealed that both exons 23 and 24 were excluded from
the Pign transcript (Fig. 1J). To determine whether nonsense-
mediated decay may contribute to the phenotype, the expression
level and localization of Pign RNA was analyzed by semi-
quantitative PCR and whole mount in situ hybridization of E14.5
embryos. There were no significant changes in expression level
(data not shown) or localization (Fig. 1M,N) of Pign between gnz
mutants and wildtype littermates.
Conceptual translation of the Pigngnztranscript predicts a
frameshift that adds 13 divergent amino acids followed by a stop
codon (Fig. 1K), causing early termination and loss of the
carboxy-terminal 170 amino acids of the 931 amino acid protein
(Fig. 1L). The truncated C-terminus includes part of the
ethanolamine phosphate transferase domain as well as a KKXX
ER retention motif (Vincent et al., 1998), suggesting that the
truncated Pign protein lacks catalytic activity and is mislocalized.
Gnz mutant embryos show mislocalized GPI anchored proteins
Pign mutants have been described in yeast and mouse cell lines
(Gaynor et al., 1999; Hong et al., 1999). These mutated cells
show mislocalization of GPI-anchored proteins (GPI-APs). To
determine whether GPI-APs are mislocalized in gnz mutants, the
gnz line was crossed into a transgenic mouse line in which GFP is
targeted to the membrane by a GPI linkage (Rhee et al., 2006).
While GPI-GFP is highly expressed in wildtype E12.5 embryos,
gnz mutant embryos show a dramatic reduction in GPI-GFP
fluorescence (Fig. 1O). The subcellular distribution of GPI-GFP
in mouse embryonic fibroblasts (MEFs) from E12.5 wildtype
embryos show GFP localization in the plasma membrane,
whereas gnz mutant MEFs show only internal localization of
GFP innumerous apparently
surrounding the nucleus (Fig. 1P,Q). These structures are
consistent with those seen in budding yeast with a mutation in
the Pign ortholog Mcd4 (Gaynor et al., 1999). Immunostaining
with bCOP, a marker of the Golgi compartment, reveals these
structures express Golgi markers (Fig. 1R,S).
Beaker mutant embryos display anterior truncations
A second ENU-derived mouse line that displays an HPE-like
phenotype was isolated from an independent screen. Dependent
on the background strain, homozygous mutant embryos from the
beaker mouse line also display a range of phenotypes – normal
appearance in 129S1, anterior truncations in C57 (Fig. 2B,D),
andholoprosencephaly (Fig. 2A,C)
(supplementary material Fig. S1C,D) in mixed 129S1/C57
strains. Further analysis of the bkr line was performed in the
C57 background, which gave the fully penetrant phenotype
shown in Fig. 2D.
Bone and cartilage staining of wildtype and bkr mutant
embryos at E18.5 (Fig. 2E–H) reveal the interparietal element is
dysmorphic and not ossified and the tectum synoticum is present
as two lateral elements that do not traverse the midline. All
elements rostral to these are largely absent. The supraoccipital
and exoccipital elements are correctly positioned, but not
ossified, whereas the basioccipital element is absent. Bkr
Fig. 1. Mutation of the glycerophosphatidyl inositol biosynthesis enzyme
Pign leads to anterior truncations in Gonzo mutant embryos. Wildtype
(A–C) and gnz (D–F) mutant E18.5 embryos with lateral (B,E) and top (C,F)
views of skull stained for bone (red) and cartilage (blue). Exoccipital (eo),
supraoccipital (so), basioccipital (bo), tectum synoticum (tso) and interparietal
(ip) elements are indicated. Critical 5.14 Mb region of chromosome 1 that the
gnz mutation mapped to by meiotic recombination (G). Recombinants are
shown in parentheses divided by number of recombination opportunities. SNP
and SSLP markers and known or predicted genes (horizontal bars) are
indicated. There is a T to A transversion in the splice donor of Pign’s intron 23
(H) that causes skipping of exons 23 and 24 (J) as shown by PCR amplification
of wildtype (WT), heterozygote (Het) and gnz mutant cDNA (I). Mis-splicing
of exon 22 to 25 results in a frame shift and premature stop codon within 13
amino acids from the mutation (K) leading to truncation and partial loss of the
catalytic domain and KKXX ER-retention motif (L). Pign mRNA detected by
whole mount in situ hybridization in E14.5 wildtype (M) and Pigngnz/gnzmutant
(N) embryos, showing ubiquitous and stable expression in gnz mutants. GFP
fluorescence was detected in E13.5 Pign+/+; TgGPI-GFP(O -left) but was greatly
reduced in Pigngnz/gnz; TgGPI-GFP(O -right) embryos and MEFs (P–S; P and R
are Pign+/+; TgGPI-GFPMEFs and Q and S are Pigngnz/gnz; TgGPI-GFPMEFs).
Panels R and S were immunostained for the Golgi marker bCOP (red). See also
supplementary material Fig. S1 for mutant phenotypes. Ratios in lower right of
panels N,O and subsequent figures indicate number of similar phenotypes
observed out of number of mutant embryos analyzed. Scale bars (A–F)52 mm.
HPE and Cripto disruption in GPI mutants 876
mutant embryos are smaller than wildtype littermates with
significant shortening of the femur (10.4%, P value50.01) and
humerus (5.7%, P value50.02) (supplementary material Fig.
Bkr mutation disrupts Pgap1, encoding a second
glycerophosphatidyl inositol biosynthesis enzyme
We mapped the bkr mutation to a 24 Mb region, containing 145
predicted genes, on mouse chromosome 1 (Fig. 2I). Based on
similarity to the gnz phenotype and discovery of the Pign
mutation, we cross-referenced these genes with known GPI
biosynthesis genes and found that Post-GPI Attachment to
Proteins factor 1 (Pgap1) lies within this region. Pgap1 is an
ER localized deacylase that removes an acyl group from the GPI
(supplementary material Fig. S3). Deacylation of GPI anchored
proteins allows efficient loading into CopII coated vesicles and
transport to the Golgi (Tanaka et al., 2004).
PCR amplification of Pgap1 revealed a size difference
between wildtype and bkr mutant cDNAs in PCR products that
span exon 19 (Fig. 2J). Sequencing of genomic DNA at the exon/
intron borders of exon 19 revealed a base substitution (T to C) in
the splice donor of intron 19 (Fig. 2K) and sequencing of cDNA
showed skipping of exon 19 in bkr mRNA (Fig. 2L). Exon 19 is
39 nucleotides, and its exclusion is predicted to result in an in-
frame deletion of 13 amino acids. While these 13 amino acids do
not correspond to a domain of known function, these residues are
well conserved between human, mouse, chick and zebrafish,
though not in S. cerevisiae (Fig. 2M,N). The Pigngnz(renamed
Pignm1Nisw/J) and Pgapbkr(renamed Pgap1m1Nisw/J) mouse lines
are available from Jackson Laboratories.
Forebrain patterning defect in Pign and Pgap1 mutant
Apart from anterior truncations, gnz (in 129S1 background) and
bkr mutant embryos display relatively normal development,
which is remarkable as there are more than one hundred GPI
anchored proteins. Consistent with normal trunk development,
the trunk mesoderm marker Brachyury is expressed normally in
E6.5 and E7.5 gnz and bkr mutants (Fig. 3A–D).
The dysmorphic occipital elements in both gnz and bkr
mutants prompted investigation of markers of hindbrain
development. Krox20 expression in rhombomeres 3 and 5 (r3
and r5) in headfold-stage embryos (E8.5) is altered in these
mutants (Fig. 3E–H). In gnz mutants, Krox20 expression in r3 is
reduced, and expanded in r5. In bkr mutants, Krox20 expression
is normal in r3, but reduced in r5. This implicates GPI
biosynthesisas being required
development, yet this may be a downstream consequence of the
lack of anterior tissues.
We next investigated the expression of genes involved in
forebrain patterning and/or associated with HPE. In E9.5
Fig. 2. Mutation of the glycerophosphatidyl inositol deacylase Pgap1
results in anterior truncations in Bkr mutant embryos. Wildtype (A) and bkr
mutant (C) E11.5 embryos reveal HPE in bkr mutants. Wildtype (B,E,F) and
bkr mutant (D,G,H) E18.5 embryos with lateral (E,G) and top (F,H) views of
skull stained for bone and cartilage. Abbreviations as in Fig. 1. (I) 23.8 Mb
region of chromosome 1 that the bkr mutation mapped to by meiotic
recombination. The Pgap1 gene (red horizontal bar) is shown but not the other
144 genes in this interval. There is a T to C transition in the splice donor of
intron 19 of Pgap1 (K) that causes skipping of exon 19 (L) as shown by PCR of
wildtype (WT), heterozygote (het) and bkr mutant cDNA (J) and sequencing.
Mis-splicing of exon 18 to 20 results in an in-frame deletion of 13 amino acids
that are conserved (identical amino acids highlighted in yellow, conserved in
pink) among human (HS), chick (GG) and zebrafish (DR) Pgap1 orthologs (N).
These deleted amino acids lie outside of domains of known function (M). Scale
bars (A–H)52 mm.
Fig. 3. Forebrain patterning markers are misexpressed in both Gnz and
Bkr mutants. Gnz mutant embryos (B,F,J,N) and wildtype littermates
(A,E,I,M) and bkr mutant embryos (D,H,L,P) and their wildtype littermates
(C,G,K,O) were analyzed by whole mount RNA in situ hybridization for
Brachyury (A–D), Six3 and Krox20 (E–H), Fgf8 (I–L) and Shh (M–P). Embryos
were at E6.5 (A,B), E7.5 (C,D), E8.5 (E–H,M,N) and E9.5 (I–L,O,P). Black
and white asterisks in E denote Krox20 and Six3 expression, respectively.
Abbreviations: is5isthmus; ba5branchial arches; tb5tailbud. Arrows and
arrowheads denote presence or absence, respectively of Fgf8 expression in the
ANR (I–L). Black bars mark prechordal plate (M,O).
HPE and Cripto disruption in GPI mutants877
embryos, the anterior neural ridge (ANR) acts as a forebrain
‘‘organizing center’’ by secreting factors necessary to maintain
forebrain specification. Fgf8 is expressed in the ANR, isthmus
(midbrain-hindbrain boundary), pharyngeal arches and tail bud.
Fgf8 expression in gnz and bkr mutants is normal, except in the
ANR, where it is not expressed (Fig. 3I–L); lack of Fgf8
expression likely reflects that the ANR fails to develop and
implicates an earlier requirement of GPI biosynthesis in forebrain
Six3 is an earlier anterior forebrain marker. The Six3
expression domain was both reduced in size and level of
expression in E8.5 gnz mutants, and severely reduced or absent in
bkr mutants (Fig. 3E–H). Six3 function is required for forebrain
development (Lagutin et al., 2003) and mutation of human SIX3
causes HPE; hence decreased Six3 expression is consistent with
forebrain defects. The relative decrease in Six3 expression in bkr
versus gnz mutant embryos also correlates with the extent of
Shh is expressed in the prechordal plate underlying the anterior
neural tube and is required for septation of the future forebrain
into left and right hemispheres. Mutations in human SHH and its
signaling pathway are linked to HPE. Shh expression is decreased
in the prechordal plate of both E8.5 gnz and E9.5 bkr mutant
embryos (Fig. 3M–P). The disrupted expression patterns of Fgf8,
Six3 and Shh indicate an early defect in forebrain specification.
Early forebrain organizing centers are disrupted in gnz and bkr
Otx2, which is expressed in the anterior definitive endoderm
(ADE) and underlying anterior neuroectoderm in gastrulation
stage (E7.5) embryos, is required for forebrain specification. The
node-derived ADE migrates anteriorly and acts as an organizing
center to maintain forebrain development in the neuroectoderm.
Otx2 expressing cells are mislocalized at the distal tip of E7.5 gnz
mutant embryos whereas E7.75 bkr mutant embryos display
proper localization of Otx2 expression albeit at a reduced level
(Fig. 4A–D). Alteration in Otx2 localization (gnz) and expression
(bkr) suggests that ADE cellular migration or specification might
be disrupted in GPI biosynthesis mutants.
To examine ADE migration and specification, we crossed gnz
and bkr lines into Hex-GFP transgenic mouse line (Rodriguez et
al., 2001; Stuckey et al., 2011) to visualize Hex expression and
ADE localization. Both gnz and bkr E7.5 embryos showed a
modest displacement of the ADE towards the distal tip compared
can cause anterior truncation phenotypes, examination of Hex-
GFP at an earlier stage allows visualization of both the DVE and
AVE, two molecularly distinct, migratory groups of cells that
contribute to forebrain specification. The DVE migrates towards
the presumptive anterior pole at ,E5.5, ‘paving the road’ for
subsequent AVE migration (Takaoka et al., 2011). At the anterior
pole, the AVE initiates forebrain specification then is displaced
and Hex expression is downregulated (Rodriguez et al., 2001).
Hex-GFP fluorescence (which does not distinguish the DVE from
the AVE) in gnz and bkr E6.5 embryos reveals aberrant DVE/AVE
localization (Fig. 4I–L).
displacement at the distal tip. However, bkr mutant embryos
have promiscuous GFP expression throughout the visceral
endoderm, reminiscent of cultured embryos in which the
extraembryonic ectoderm has been removed (Rodriguez et al.,
2005). Together this indicates that GPI biosynthesis proteins are
required to properly initiate forebrain development. Moreover, this
allowed us to make some predictions as to the identity of the GPI-
AP(s) that may be altered resulting in the HPE-like phenotype
based on involvement in AVE specification and migration, and
perhaps communication with the extraembryonic ectoderm.
The Nodal/Cripto signaling pathway is defective in Pign and
Pgap1 deficient cell lines
Given the migration and specification defects of the AVE in E6.5
gnz and bkr mutant embryos, we hypothesized that Cripto is the
GPI anchored protein whose GPI deficiency leads to the HPE-
like phenotype. We analyzed Nodal/Cripto signaling in wildtype
and mutant MEFs by phosphorylation of Smad2 (Watanabe et al.,
2007). In the absence of exogenous Nodal stimulation, wildtype,
Pigngnzand Pgap1bkrMEFs, which express endogenous Cripto,
have negligible phospho-Smad2 and this is not significantly
increased by Cripto overexpression (transfection of HA-tagged
Cripto expression vector compared to empty vector, mock
stimulated cells) (Fig. 5A–C). Upon Nodal stimulation in
wildtype cells, there was a 16-fold increase of phospho-Smad2
in empty vector transfected cells (P value50.003) and 35-fold
increase with Criptooverexpression
value50.007). In contrast, Nodal stimulation of gnz or bkr
MEFs, in the presence or absence of Cripto overexpression,
showeda significantly lower
(approximately 3-fold lower for both gnz [P value50.03 with
or without Cripto] and bkr [P value50.04 without Cripto, and
0.02 with Cripto]). Thus, GPI biosynthetic activity is required for
efficient Nodal/Cripto signaling.
In Pigngnzand Pgap1bkrmutant MEFs, there was some
response to Nodal stimulation, suggesting that endogenous or
transfected Cripto retained some signaling activity. Cripto can
function non-cell autonomously via enzymatic or genetic
removal of its GPI anchor (Zhang et al., 1998; Yan et al.,
2002; Watanabe et al., 2007), demonstrating that secreted Cripto
retains biological activity. Furthermore, secretion of GPI
Fig. 4. Forebrain organizing centers are mislocalized in both Gnz and Bkr
mutants. E7.5 wildtype (A) and gnz mutant (B) littermates and E7.75 wildtype
(C) and bkr mutant (D) littermates were analyzed by whole mount RNA in situ
hybridization for Otx2. Gnz mutant embryos (F,J) and wildtype littermates
(E,I) and bkr mutant (H,L) and wildtype littermates (G,K), all containing Hex-
GFP transgene were analyzed by GFP fluorescence. Hex-GFP fluorescence
marked the ADE in E7.5 embryos (E–H) and DVE/AVE in E6.5 embryos (I–L).
HPE and Cripto disruption in GPI mutants 878
anchored proteins was detected in Mcd4 (Pign ortholog) mutant
budding yeast (Gaynor et al., 1999). Cripto overexpression, while
resulting in increased Smad2 phosphorylation, has no effect on
intracellular Cripto protein levels (Fig. 5A) suggesting that some
Cripto protein could be secreted and act in a non-autonomous
manner. To test this we immunoprecipitated endogenous secreted
Cripto from conditioned medium from wildtype and mutant
MEFs followed by western blot detection (Fig. 5D). Although
not quantitative, this showed Cripto protein in the medium,
consistent with reduced but not absent Smad2 phosphorylation in
both gnz and bkr MEFs.
TGFb responsive genes are downregulated in gnz and bkr
As another measure of Cripto/Nodal activity we determined the
expression of TGFb target genes by real time PCR using TGFb/
BMP pathway arrays and E7.5 RNA from mutants compared to
wildtype littermates, as the two lines are in different background
strains. Seven of 22 TGFb responsive genes were significantly (P
value,0.05) downregulated in bkr mutants and 3 of 22 TGFb
responsive genes were significantly downregulated in gnz mutants
as compared to wildtype littermates (Table 1). Collagen, type1
alpha-2 (Col1a2) and Interleukin6 (Il6) were downregulated in
both gnz and bkr mutants. BMP signaling is not dependent on
Cripto activity, so BMP responsive genes can serve as a negative
control since these genes are predicted to be unchanged in GPI
biosynthesis mutants. Correspondingly, only the BMP responsive
gene Stat1 showed significantly altered expression, and only in bkr
mutant embryos. The array also includes other genes in the TGFb/
BMP signaling pathway, many of which display significantly
altered expression levels in the GPI biosynthesis mutants.
All TGFb responsive genes that are significantly altered in gnz
and bkr mutant embryos are downregulated. In contrast, 41% of
the other genes in the TGFb/BMP signaling pathway that are
significantly altered are upregulated (9 out of 22). Together these
data indicate that TGFb signaling is defective in E7.5 gnz and bkr
mutant embryos and corroborates the in vitro data showing that
Nodal/Cripto signaling is defective in GPI deficient cell lines.
GPI biosynthesis mutants and HPE
Herein we describe novel mutations in two GPI biosynthesis
genes that result in an HPE-like phenotype in mice. In humans,
mutations in the GPI biosynthesis pathway have not been
implicated in HPE, although mutations are associated with
other diseases, including Paroxysmal Nocturnal Hemoglobinuria
(hypomorphic mutation in the PIGM promoter). A human
PIGN mutation (R709Q) has been linked to an autosomal
recessive syndrome presenting with developmental delay,
dysmorphic facies, seizures and severe neurological impairment
(Maydan et al., 2011). It is interesting to speculate that different
PIGN alleles may yield a range of phenotypes from neurological
defects to structural, HPE-like defects. Our identification of
mutations in two GPI biosynthesis enzymes, which result in
HPE-like phenotypes in mice, greatly expands an understanding
of the genetic causes of HPE and suggests that this entire
enzymatic pathway of 27 genes represent novel candidate genes
for analysis in human HPE or related disorders.
Complete loss of GPI anchors is embryonic lethal in mice, as
demonstrated by knockout of Piga, a critical and early GPI
biosynthesis enzyme (Kawagoe et al., 1996; Tremml et al., 1999).
Conditional knockout of Piga in chondrocytes leads to mice with
shortened bones, similar to that in Pigngnzand Pgap1bkrmutants.
Previous mutant alleles of Pgap1 show a similar forebrain
truncation phenotype as Pigngnzand Pgap1bkrmutant embryos,
though the authors labeled it as otocephaly, as well as growth
retardation, mislocalization of GPI-APs and a sperm activation
defect. (Juriloff et al., 1985; Zoltewicz et al., 1999; Ueda et al.,
2007; Zoltewicz et al., 2009). Both Pigngnzand Pgap1bkrembryos
are growth retarded and GPI-GFP is mislocalized in Pigngnzmutant
embryos. We also observed a genetic interaction between Pign and
harboring a single Nodal null allele (Nodal-LacZ) and a single
6wildtype) we observed a less than expected proportion of Nodal-
value.0.05 using x2test). Moreover, skewing of the expected
ratio only occurred when the father was a trans-heterozygote mated
to a wildtype female (14.8%, n5216, P value50.05). In all other
combinatorial matings, the number of trans-heterozygotes was as
expected (23.1%, n5199). The Pign and Nodal genetic interaction
was statistically significant, although it addressed viability or
perhaps sperm fitness, not forebrain development. The Pign and
Nodal genetic interaction does, however, indicate a genetic link
between GPI biosynthesis and TGFb signaling.
Fig. 5. Cripto/Nodal signaling is defective in MEFs
derived from Gnz and Bkr embryos. Wildtype and
Pigngnz/gnzMEFs (A) or wildtype and Pgap1bkr/bkrMEFs
(B) were transfected with HA-Cripto-pcDNA3 expression
vector (Cripto) or pcDNA3 vector (Empty), serum starved,
then treated with 250 ng/ml recombinant Nodal protein (+)
or vehicle (2) for 1 hour. Whole cell extract was separated
by SDS-PAGE, and phosphorylated (P-Smad2) and total
Smad2 (T-Smad2) and Cripto proteins detected by western
blot. (C) Quantification of at least three independent Cripto/
Nodal signaling experiments. * represents statistical
significance (P value,0.05) by Student’s T-test.
(D) Endogenous Cripto protein was immunoprecipitated
from serum-free conditioned medium overlying Pign+/+,
Pigngnz/gnz, Pgap1+/+and Pgap1bkr/bkrMEFs, separated by
SDS-PAGE and detected by western blot.
HPE and Cripto disruption in GPI mutants 879
GPI anchored proteins and HPE
Here we sought to identify the key GPI-AP(s) that Pign and
Pgap1 act through to regulate normal forebrain development. In a
previous report of a Pgap1 mutant allele, Zoltewicz et al.
demonstrated that Wnts are modified with a GPI-like anchor and
that midbrain Wnt activation is slightly premature (E8.0 vs. E8.2)
(Zoltewicz et al., 2009). The importance of Wnt signaling in
forebrain specification has been shown in mice with mutation of
Apc (Chazaud and Rossant, 2006) or Dikkopf1 (Mukhopadhyay et
al., 2001). Zoltewicz et al. surmised that premature Wnt
activation was responsible for otocephaly (Zoltewicz et al.,
2009). Our data establish a much earlier origin (E6.5) for the
forebrain defect, so it is possible that altered Wnt expression is
downstream of an earlier disruption in forebrain development.
Although we did not detect a change in Wnt signaling using the
Top-Gal Wnt reporter strain, further study is needed to determine
whether Wnt signaling is also affected at this early stage in
Pigngnzand Pgap1bkrmutants, as Wnt signaling, in conjunction
with TGFb is important for AVE formation and migration and for
Gas1 is a GPI-AP that binds Shh and modulates Shh activity,
and has been described as a modulator of HPE. Gas12/2mouse
embryos present with small eyes and reduced body size (Allen et
al., 2007), reminiscent of some gnz mutants in C3H/HeJ
background (supplementary material Fig. S1A,B). It will be of
interest to determine whether GPI disruption targets both the Shh
and TGFb pathways in HPE etiology. It is also interesting to
speculate whether another GPI-AP, Glypican 4, is in part
responsible for gastrulation defects in some gnz C3H/HeJ
mutants. Glypican 4 mediates non-canonical Wnt signaling, and
the zebrafish mutant Knypek fails to gastrulate due to aberrant
cell polarity and disrupted convergent extension (Topczewski et
al., 2001). Cripto nulls also fail to undergo gastrulation (Ding et
al., 1998), so it may be difficult to isolate the contribution of
these two genes in gastrulation.
The GPI-anchored Cripto-related protein Cryptic may also
contribute to the HPE-like phenotype in GPI biosynthesis
mutants. Cryptic expression overlaps Nodal in the DVE, and
these two proteins specify the AVE (Chu and Shen, 2010).
Pgap1bkrmutants might be candidates for altered Cryptic/Nodal
signaling, as Hex-GFP is ectopically expressed throughout the
The following evidence supports our hypothesis that Cripto is
at least one of the key GPI anchored proteins that juxtaposes GPI
biosynthesis and forebrain specification: 1. Pign and Pgap1
mutations result in anterior truncations or HPE-like phenotype;
Cripto hypomorphic mutations in mouse (Chu et al., 2005), Oep
mutations in zebrafish (Gritsman et al., 1999) and TDGF1
mutations in humans (de la Cruz et al., 2002) cause HPE. 2. Pign
and Pgap1 mutations result in mislocalization (secretion and/or
Table 1. Real-time PCR results for TGFb signaling genes indicate defective TGFb responses in Pigngnzand Pgap1bkrembryos.
Shown are statistically significant changes in expression levels of genes on the Mouse TGFb BMP Signaling Pathway RT2Profiler PCR
Array. Relative expression (Relative Expr.) 6 standard error is shown from comparison between E7.5 Pigngnzand Pgap1bkr
homozygous mutants and their respective wildtype littermates (C57BL/6J for bkr line and 129S1/Sv1mJ for gnz). Data were generated
using six embryos of each genotype and discarding the single largest outlier for all genes.
bkr vs. wildtype
gnz vs. wildtype
Gene Symbol Relative Expr.p-valueRelative Expr. p-value
TGFb Responsive Genes
0.5760.15 0.015 0.2660.25
BMP Responsive Genes
Other TGFb Superfamily Genes
HPE and Cripto disruption in GPI mutants 880
internalization) of GPI-anchored proteins; Cripto is a GPI
anchored protein (Minchiotti et al., 2000), and its GPI anchor
is critical for its function (Watanabe et al., 2007). 3. Pign and
Pgap1 mutants show defective AVE migration; Cripto mutations
have AVE migration defects (D’Andrea et al., 2008). 4. Otx2
expression is altered in Pign and Pgap1 mutants; Cripto
genetically interacts with Otx2 to specify the AVE (Kimura et
al., 2001), 5. A percentage of gnz mutant embryos display
gastrulation defects; homozygous Cripto null mutants display
gastrulation defects (Ding et al., 1998). 6. Cripto/Nodal signaling
is reduced in Pign and Pgap1 mutant cells and TGFb responsive
genes are downregulated in Pign and Pgap1 mutant embryos.
These pieces of evidence support the hypothesis that Cripto is a
key GPI-AP whose dysfunction leads to an HPE-like phenotype.
GPI biosynthesis mutant phenotypes are variable
Pigngnzand Pgap1bkrmutant embryos share similar phenotypes,
but are remarkably different in some aspects. First, gnz mutant
embryos display forebrain truncations while bkr mutants show no
phenotype in the 129S1 background. Second, Hex-GFP shows
differential patterns of expression with gnz embryos expressing
GFP at approximately normal levels in the normal number of
cells, albeit those cells are not in their correct position, whereas
bkr mutants have an expanded population of Hex-GFP cells and
those cells overexpress Hex-GFP. Third, TGFb responsive gene
expression is reduced in Pigngnzand Pgap1bkrmutant embryos
but the specific genes that are downregulated show little overlap.
In part these differences may be related to genetic background
differences as the penetrance and severity of the developmental
defects depends on the genetic background. These differences
may also relate to the allele generated by the point mutations. The
Pgap1bkrallele causes an in-frame deletion of 13 amino acids.
morphologically normal, which suggests that Pgap1bkrretains
partial activity. In contrast, Pigngnzis predicted to encode a
truncated protein that both lacks catalytic activity and is not
localized to the ER. Pigngnzphenotypes are influenced by genetic
background but abnormal embryonic development is observed in
all tested backgrounds.
The phenotypic differences may also relate to the different
steps in GPI biosynthesis that Pign and Pgap1 catalyze. Pign
catalyzes an early step in GPI biosynthesis, and disruption of
Pign in cell lines leads to formation of several divergent GPI
anchors, some of which localize to the plasma membrane, while
others do not (Hong et al., 1999). Pgap1 catalyzes inositol
deacylation after the GPI anchor is covalently bound to target
proteins (Tanaka et al., 2004). Acyl-group removal is necessary
for efficient ER export via CopII coated vesicles, and Pgap1
mutant cells accumulate GPI-APs within the ER (Tanaka et al.,
2004). As Pigngnzlikely represents a null allele and Pign
catalyzes a relatively early biosynthetic step in GPI anchor
production, Pigngnzmutants likely represent a more severe class
of GPI biosynthesismutants
Alternatively, although the GPI biosynthesis pathway is largely
thought to follow a linear progression, our in vivo results might
suggest divergent requirements for this pathway in regulating
Cell autonomous vs. non-cell autonomous Cripto signaling
The contribution of cell autonomous versus non-cell autonomous
Cripto/Nodal signaling is poorly understood. Lack of both cell
autonomous and non-autonomous Cripto signaling results in a
‘‘head-without-trunk’’ phenotype (Ding et al., 1998). Gnz and bkr
mutants, which present as a ‘‘trunk-without-head’’, likely retain
some non-cell autonomous Cripto signaling while lacking cell
autonomous Cripto signaling. Further, overexpression of a
soluble form of Cripto can rescue zebrafish Oep mutants
(Zhang et al., 1998). While cell autonomous Cripto signaling is
clearly required for normal development, further studies are
required to determine the role of non-cell autonomous Cripto
signaling during early development.
Downstream effectors of forebrain development
Nodal/Cripto signaling is important for forebrain development,
yet the downstream targets of this pathway are unknown.
Analysis of genes that are altered in gnz and bkr mutants may
shed light upon downstream effectors that mediate this process.
Col1a2 and Il6 are significantly downregulated in gnz and bkr
mutants; however, phenotypes associated with mutation of these
genes in humans are unrelated to forebrain development (Dickson
et al., 1984; Wirtz et al., 1987; Kishimoto, 2005). Three genes
involved in iron homeostasis are either downstream of TGFb
signaling or GPI-APs. Gdf2 and Il6 converge to regulate
expression of Hepcidin (Truksa et al., 2007), a key regulator of
iron homeostasis (Nemeth and Ganz, 2009). The GPI-AP
Hemojuvelin also regulates Hepcidin, so gnz and bkr mutants
may have alterations in the function of genes that regulate iron
levels. Iron levels are important for forebrain development in
mice and mutation of the iron transporter Ferroportin causes
anterior truncations (Mao et al., 2010). It will be interesting to
test whether iron homeostasis is dysregulated in bkr and gnz
mutants and, if so, this could provide a context to begin to
understandthe molecular relationships
pathways that alter early forebrain development.
In humans, there is no current evidence linking mutations in
GPI biosynthesis genes with HPE or forebrain truncations. Our
studies raise the intriguing possibility that the ,thirty GPI
biosynthesis enzymes may represent a new class of genes to test
forlinkage to holoprosencephaly,
syndromes, such as agnathia, dysgnathia, microphthalmia and
or other cranio-facial
We thank Kathryn Anderson, Elizabeth Lacy, Monica Justice and
their labs for ENU mutagenesis; Trevor Williams, Weiguo Feng and
Gartz Hanson for help with the forward genetic screen; Andy
Peterson for sharing unpublished results; Kat Hadjantonakis (GPI-
GFP), Tristan Rodriquez (Hex-GFP) and Michael Shen (Cripto
plasmids) for mice and reagents, and David Clouthier for help
interpreting skull morphologies. This project was supported by
NINDS F31NS060454 and NICHD U01 HD043478. The content is
solely the responsibility of the authors and does not necessarily
represent the official views of the National Institute of Neurological
Disorders And Stroke or the National Institutes of Health. L.N. is an
investigator of the Howard Hughes Medical Institute.
The authors have no competing interests to declare.
Allen, B. L., Tenzen, T. and McMahon, A. P. (2007). The Hedgehog-binding proteins
Gas1 and Cdo cooperate to positively regulate Shh signaling during mouse
development. Genes Dev. 21, 1244-1257.
HPE and Cripto disruption in GPI mutants881
Andersson, O., Reissmann, E., Jo ¨rnvall, H. and Iba ´n ˜ez, C. F. (2006). Synergistic
interaction between Gdf1 and Nodal during anterior axis development. Dev. Biol. 293,
Chazaud, C. and Rossant, J. (2006). Disruption of early proximodistal patterning and
AVE formation in Apc mutants. Development 133, 3379-3387.
Chu, J. and Shen, M. M. (2010). Functional redundancy of EGF-CFC genes in epiblast
and extraembryonic patterning during early mouse embryogenesis. Dev. Biol. 342,
Chu, J., Ding, J., Jeays-Ward, K., Price, S. M., Placzek, M. and Shen, M. M. (2005).
Non-cell-autonomous role for Cripto in axial midline formation during vertebrate
embryogenesis. Development 132, 5539-5551.
Collignon, J., Varlet, I. and Robertson, E. J. (1996). Relationship between asymmetric
nodal expression and the direction of embryonic turning. Nature 381, 155-158.
D’Andrea, D., Liguori, G. L., Le Good, J. A., Lonardo, E., Andersson, O., Constam,
D. B., Persico, M. G. and Minchiotti, G. (2008). Cripto promotes A-P axis
specification independently of its stimulatory effect on Nodal autoinduction. J. Cell
Biol. 180, 597-605.
de la Cruz, J. M., Bamford, R. N., Burdine, R. D., Roessler, E., Barkovich, A. J.,
Donnai, D., Schier, A. F. and Muenke, M. (2002). A loss-of-function mutation in
the CFC domain of TDGF1 is associated with human forebrain defects. Hum. Genet.
Dickson, L. A., Pihlajaniemi, T., Deak, S., Pope, F. M., Nicholls, A., Prockop, D. J.
and Myers, J. C. (1984). Nuclease S1 mapping of a homozygous mutation in the
carboxyl-propeptide-coding region of the pro alpha 2(I) collagen gene in a patient
with osteogenesis imperfecta. Proc. Natl. Acad. Sci. USA 81, 4524-4528.
Ding, J., Yang, L., Yan, Y. T., Chen, A., Desai, N., Wynshaw-Boris, A. and Shen, M.
M. (1998). Cripto is required for correct orientation of the anterior-posterior axis in
the mouse embryo. Nature 395, 702-707.
Gaynor, E. C., Monde ´sert, G., Grimme, S. J., Reed, S. I., Orlean, P. and Emr, S. D.
(1999). MCD4 encodes a conserved endoplasmic reticulum membrane protein
essential for glycosylphosphatidylinositol anchor synthesis in yeast. Mol. Biol. Cell
Gritsman, K., Zhang, J., Cheng, S., Heckscher, E., Talbot, W. S. and Schier, A. F.
(1999). The EGF-CFC protein one-eyed pinhead is essential for nodal signaling. Cell
Hong, Y., Maeda, Y., Watanabe, R., Ohishi, K., Mishkind, M., Riezman, H. and
Kinoshita, T. (1999). Pig-n, a mammalian homologue of yeast Mcd4p, is involved in
transferring phosphoethanolamine to the first mannose of the glycosylphosphatidy-
linositol. J. Biol. Chem. 274, 35099-35106.
Hoodless, P. A., Pye, M., Chazaud, C., Labbe ´, E., Attisano, L., Rossant, J. and
Wrana, J. L. (2001). FoxH1 (Fast) functions to specify the anterior primitive streak
in the mouse. Genes Dev. 15, 1257-1271.
Juriloff, D. M., Sulik, K. K., Roderick, T. H. and Hogan, B. K. (1985). Genetic and
developmental studies of a new mouse mutation that produces otocephaly. J.
Craniofac. Genet. Dev. Biol. 5, 121-145.
Kawagoe, K., Kitamura, D., Okabe, M., Taniuchi, I., Ikawa, M., Watanabe, T.,
Kinoshita, T. and Takeda, J. (1996). Glycosylphosphatidylinositol-anchor-deficient
mice: implications for clonal dominance of mutant cells in paroxysmal nocturnal
hemoglobinuria. Blood 87, 3600-3606.
Kimura, C., Shen, M. M., Takeda, N., Aizawa, S. and Matsuo, I. (2001).
Complementary functions of Otx2 and Cripto in initial patterning of mouse epiblast.
Dev. Biol. 235, 12-32.
Kimura-Yoshida, C., Nakano, H., Okamura, D., Nakao, K., Yonemura, S., Belo, J.
A., Aizawa, S., Matsui, Y. and Matsuo, I. (2005). Canonical Wnt signaling and its
antagonist regulate anterior-posterior axis polarization by guiding cell migration in
mouse visceral endoderm. Dev. Cell 9, 639-650.
Kinoshita, T., Fujita, M. and Maeda, Y. (2008). Biosynthesis, remodelling and
functions of mammalian GPI-anchored proteins: recent progress. J. Biochem. 144,
Kishimoto, T. (2005). Interleukin-6: from basic science to medicine–40 years in
immunology. Annu. Rev. Immunol. 23, 1-21.
Lagutin, O. V., Zhu, C. C., Kobayashi, D., Topczewski, J., Shimamura, K., Puelles,
L., Russell, H. R., McKinnon, P. J., Solnica-Krezel, L. and Oliver, G. (2003). Six3
repression of Wnt signaling in the anterior neuroectoderm is essential for vertebrate
forebrain development. Genes Dev. 17, 368-379.
Liguori, G. L., Borges, A. C., D’Andrea, D., Liguoro, A., Gonc ¸alves, L., Salgueiro, A.
positioning of the A-P axis in the early mouse embryo. Dev. Biol. 315, 280-289.
Lowe, L. A., Yamada, S. and Kuehn, M. R. (2001). Genetic dissection of nodal
function in patterning the mouse embryo. Development 128, 1831-1843.
Mao, J., McKean, D. M., Warrier, S., Corbin, J. G., Niswander, L. and Zohn, I. E.
(2010). The iron exporter ferroportin 1 is essential for development of the mouse
embryo, forebrain patterning and neural tube closure. Development 137, 3079-
Maydan, G., Noyman, I., Har-Zahav, A., Neriah, Z. B., Pasmanik-Chor, M.,
Yeheskel, A., Albin-Kaplanski, A., Maya, I., Magal, N., Birk, E. et al. (2011).
Multiple congenital anomalies-hypotonia-seizures syndrome is caused by a mutation
in PIGN. J. Med. Genet. 48, 383-389.
Mesnard, D., Guzman-Ayala, M. and Constam, D. B. (2006). Nodal specifies
embryonic visceral endoderm and sustains pluripotent cells in the epiblast before
overt axial patterning. Development 133, 2497-2505.
Minchiotti, G., Parisi, S., Liguori, G., Signore, M., Lania, G., Adamson, E. D., Lago,
C. T. and Persico, M. G. (2000). Membrane-anchorage of Cripto protein by
glycosylphosphatidylinositol and its distribution during early mouse development.
Mech. Dev. 90, 133-142.
Ming, J. E. and Muenke, M. (2002). Multiple hits during early embryonic
development: digenic diseases and holoprosencephaly. Am. J. Hum. Genet. 71,
Muenke, M. and Beachy, P. A. (2000). Genetics of ventral forebrain development and
holoprosencephaly. Curr. Opin. Genet. Dev. 10, 262-269.
Mukhopadhyay, M., Shtrom, S., Rodriguez-Esteban, C., Chen, L., Tsukui, T.,
Gomer, L., Dorward, D. W., Glinka, A., Grinberg, A., Huang, S. P. et al. (2001).
Dickkopf1 is required for embryonic head induction and limb morphogenesis in the
mouse. Dev. Cell 1, 423-434.
Nemeth, E. and Ganz, T. (2009). The role of hepcidin in iron metabolism. Acta
Haematol. 122, 78-86.
Nomura, M. and Li, E. (1998). Smad2 role in mesoderm formation, left-right patterning
and craniofacial development. Nature 393, 786-790.
Paulick, M. G. and Bertozzi, C. R. (2008). The glycosylphosphatidylinositol anchor:
a complex membrane-anchoring structure for proteins. Biochemistry 47, 6991-
Perea-Gomez, A., Vella, F. D., Shawlot, W., Oulad-Abdelghani, M., Chazaud, C.,
Meno, C., Pfister, V., Chen, L., Robertson, E., Hamada, H. et al. (2002). Nodal
antagonists in the anterior visceral endoderm prevent the formation of multiple
primitive streaks. Dev. Cell 3, 745-756.
Rhee, J. M., Pirity, M. K., Lackan, C. S., Long, J. Z., Kondoh, G., Takeda, J. and
Hadjantonakis, A. K. (2006). In vivo imaging and differential localization of
lipid-modified GFP-variant fusions in embryonic stem cells and mice. Genesis 44,
Rodriguez, T. A., Casey, E. S., Harland, R. M., Smith, J. C. and Beddington, R. S.
(2001). Distinct enhancer elements control Hex expression during gastrulation and
early organogenesis. Dev. Biol. 234, 304-316.
Rodriguez, T. A., Srinivas, S., Clements, M. P., Smith, J. C. and Beddington, R. S.
(2005). Induction and migration of the anterior visceral endoderm is regulated by the
extra-embryonic ectoderm. Development 132, 2513-2520.
Rohr, K. B., Barth, K. A., Varga, Z. M. and Wilson, S. W. (2001). The nodal pathway
acts upstream of hedgehog signaling to specify ventral telencephalic identity. Neuron
Song, J., Oh, S. P., Schrewe, H., Nomura, M., Lei, H., Okano, M., Gridley, T. and
Li, E. (1999). The type II activin receptors are essential for egg cylinder growth,
gastrulation, and rostral head development in mice. Dev. Biol. 213, 157-169.
Stuckey, D. W., Clements, M., Di-Gregorio, A., Senner, C. E., Le Tissier, P.,
Srinivas, S. and Rodriguez, T. A. (2011). Coordination of cell proliferation and
anterior-posterior axis establishment in the mouse embryo. Development 138, 1521-
Takaoka, K., Yamamoto, M. and Hamada, H. (2011). Origin and role of distal
visceral endoderm, a group of cells that determines anterior-posterior polarity of the
mouse embryo. Nat. Cell Biol. 13, 743-752.
Tanaka, S., Maeda, Y., Tashima, Y. and Kinoshita, T. (2004). Inositol deacylation of
glycosylphosphatidylinositol-anchored proteins is mediated by mammalian PGAP1
and yeast Bst1p. J. Biol. Chem. 279, 14256-14263.
Topczewski, J., Sepich, D. S., Myers, D. C., Walker, C., Amores, A., Lele, Z.,
Hammerschmidt, M., Postlethwait, J. and Solnica-Krezel, L. (2001). The
zebrafish glypican knypek controls cell polarity during gastrulation movements of
convergent extension. Dev. Cell 1, 251-264.
Tremml, G., Dominguez, C., Rosti, V., Zhang, Z., Pandolfi, P. P., Keller, P. and
Bessler, M. (1999). Increased sensitivity to complement and a decreased red
blood cell life span in mice mosaic for a nonfunctional Piga gene. Blood 94, 2945-
Truksa, J., Peng, H., Lee, P. and Beutler, E. (2007). Different regulatory elements are
required for response of hepcidin to interleukin-6 and bone morphogenetic proteins 4
and 9. Br. J. Haematol. 139, 138-147.
Ueda, Y., Yamaguchi, R., Ikawa, M., Okabe, M., Morii, E., Maeda, Y. and
Kinoshita, T. (2007). PGAP1 knock-out mice show otocephaly and male infertility.
J. Biol. Chem. 282, 30373-30380.
Varlet, I., Collignon, J. and Robertson, E. J. (1997). nodal expression in the primitive
endoderm is required for specification of the anterior axis during mouse gastrulation.
Development 124, 1033-1044.
Vincent, M. J., Martin, A. S. and Compans, R. W. (1998). Function of the
KKXX motif in endoplasmic reticulum retrieval of a transmembrane protein depends
on the length and structure of the cytoplasmic domain. J. Biol. Chem. 273, 950-
Watanabe, K., Hamada, S., Bianco, C., Mancino, M., Nagaoka, T., Gonzales, M.,
Bailly, V., Strizzi, L. and Salomon, D. S. (2007). Requirement of glycosylpho-
sphatidylinositol anchor of Cripto-1 for trans activity as a Nodal co-receptor. J. Biol.
Chem. 282, 35772-35786.
Wirtz, M. K., Glanville, R. W., Steinmann, B., Rao, V. H. and Hollister, D. W.
(1987). Ehlers-Danlos syndrome type VIIB. Deletion of 18 amino acids comprising
the N-telopeptide region of a pro-alpha 2(I) chain. J. Biol. Chem. 262, 16376-
Yamamoto, M., Meno, C., Sakai, Y., Shiratori, H., Mochida, K., Ikawa, Y., Saijoh,
Y. and Hamada, H. (2001). The transcription factor FoxH1 (FAST) mediates Nodal
signaling during anterior-posterior patterning and node formation in the mouse. Genes
Dev. 15, 1242-1256.
HPE and Cripto disruption in GPI mutants 882
Yamamoto, M., Saijoh, Y., Perea-Gomez, A., Shawlot, W., Behringer, R. R., Ang, S.
L., Hamada, H. and Meno, C. (2004). Nodal antagonists regulate formation of the
anteroposterior axis of the mouse embryo. Nature 428, 387-392.
Yan, Y.-T., Liu, J.-J., Luo, Y., E, C., Haltiwanger, R. S., Abate-Shen, C. and Shen,
M. M. (2002). Dual roles of Cripto as a ligand and coreceptor in the nodal signaling
pathway. Mol. Cell. Biol. 22, 4439-4449.
Ye, S., Dhillon, S., Ke, X., Collins, A. R. and Day, I. N. (2001). An efficient
procedure for genotyping single nucleotide polymorphisms. Nucleic Acids Res. 29,
Zhang, J., Talbot, W. S. and Schier, A. F. (1998). Positional cloning identifies
zebrafish one-eyed pinhead as a permissive EGF-related ligand required during
gastrulation. Cell 92, 241-251.
Zoltewicz, J. S., Plummer, N. W., Lin, M. I. and Peterson, A. S. (1999). oto is a
homeotic locus with a role in anteroposterior development that is partially redundant
with Lim1. Development 126, 5085-5095.
Zoltewicz, J. S., Ashique, A. M., Choe, Y., Lee, G., Taylor, S., Phamluong, K.,
Solloway, M. and Peterson, A. S. (2009). Wnt signaling is regulated by endoplasmic
reticulum retention. PLoS ONE 4, e6191.
HPE and Cripto disruption in GPI mutants883