Skip Navigation


Human Molecular Genetics Advance Access originally published online on November 13, 2007
Human Molecular Genetics 2008 17(4):587-601; doi:10.1093/hmg/ddm333
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
17/4/587    most recent
ddm333v2
ddm333v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Wong, R. L. Y.
Right arrow Articles by Finnell, R. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wong, R. L. Y.
Right arrow Articles by Finnell, R. H.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Mouse Fkbp8 activity is required to inhibit cell death and establish dorso-ventral patterning in the posterior neural tube

Rebecca Lee Yean Wong1,*, Bogdan J. Wlodarczyk1, Kyung Soo Min1, Melissa L. Scott1, Susan Kartiko1, Wei Yu2, Michelle Y. Merriweather1, Peter Vogel4, Brian P. Zambrowicz4,5,6 and Richard H. Finnell1,3

1 Center for Environmental and Genetic Medicine 2 Center for Molecular Development and Disease, Institute of Biosciences and Technology, The Texas A&M University System Health Science Center, 2121 W. Holcombe Blvd, Houston, TX 77030, USA 3 Texas Institute for Genomic Medicine, 2121 W. Holcombe Blvd, Houston, TX 77030, USA 4 Department of Pharmaceutical Biology 5 Department of Biotherapeutics 6 Department of Genetics, Lexicon Pharmaceuticals, Inc., 8800 Technology Forest Place, The Woodlands, TX 77381, USA

* To whom correspondence should be addressed at: Center for Environmental and Genetic Medicine, Institute of Biosciences and Technology, The Texas A&M University System Health Science Center, 2121 W. Holcombe Blvd, Houston, TX 77030, USA. Tel: +1 7136777667; Fax: +1 7136777784; Email: lwong{at}ibt.tamhsc.edu

Received October 4, 2007; Accepted November 9, 2007


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Neural tube defects (NTDs) are birth defects that can be disabling or lethal and are second in their prevalence after cardiac defects among major human congenital malformations. Spina bifida is a NTD where the spinal cord is dysplastic, and the overlying spinal column is absent. At present, the molecular mechanisms underlying the spinal bifida development are largely unknown. In this study, we present a Fkbp8 mouse mutant that has an isolated and completely penetrant spina bifida, which is folate- and inositol-resistant. Fkbp8 mutants are not embryo lethal, but they display striking features of human spina bifida, including a dysplastic spinal cord, open neural canal and disability. The loss of Fkbp8 leads to increased apoptosis in the posterior neural tube, demonstrating that in vivo FKBP8 inhibits cell death. Gene expression analysis of Fkbp8 mutants revealed a perturbation of expression of neural tube patterning genes, suggesting that endogenous FKBP8 activity establishes dorso-ventral patterning of the neural tube. These studies demonstrate that Fkbp8 is not important for embryo survival, but is essential for spinal neural tube patterning, and to block apoptosis, in the developing neural tube. The mutant Fkbp8 allele is a new experimental model which will be useful in dissecting the pathogenesis of spinal NTDs, and enhance our understanding of the etiology of human NTDs.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Neural tube defects (NTDs) are common birth defects which can be associated with serious disability, whose prevalence is second among major human congenital malformations after cardiac defects (1). NTDs involving the spinal cord and/or vertebral arches in the thoracico–lumbo–sacral region are termed spina bifida. NTDs affect ~324 000 births worldwide (2) and 3000 pregnancies annually in the United States (3). Depending on the type of spina bifida, affected infants may have significant damage to their spinal cord, spinal nerves and vertebrae, resulting in paralysis of lower limbs and other associated defects such as hydrocephalus. Infants with spina bifida who survive are likely to have severe, life-long disabilities, and are at risk of psychosocial maladjustment (4). Spina bifida and other NTDs pose a serious public health problem in the United States and worldwide. Therefore, understanding the cellular and molecular mechanisms underlying the etiology of these malformations is crucial as this intention could lead to novel preventive strategies and improved counseling for high-risk individuals.

The etiology of spina bifida, like many NTDs, is believed to be multi-factorial, involving both genetic and environmental factors. For decades, researchers have relied upon spontaneous, teratogenic and transgenic animal models in order to better understand the mechanisms underlying spina bifida development. Unfortunately, many of these models suffer from either incomplete phenotypic penetrance or poor recapitulation of either the complex genetics or the phenotype observed in human spina bifida. Furthermore, the spina bifida lesion seen in these models usually occurs in association with various organ defects (5) whereas in human condition is often isolated. Recently, we identified an Fkbp8 knockout mouse that recapitulates most of the clinical features of human spina bifida. This gene belongs to a family of immunophilin proteins functioning in immunoregulation, and protein folding and trafficking (6,7). FKBP8 is a 44 kDa mitochondrial protein which has three tetratricopeptide repeat (TPR) motifs, which are protein–protein interaction domains that mediate physical associations between FKBP8 and its protein interactors (8–10). The N-terminus of FKBP8 contains a peptidylprolyl cis–trans isomerase (PPIase) domain that is conserved throughout the FKBP family, and is important for FK506 binding and PPIase activity (11,12).

The Fkbp8 gene has been previously targeted in the mouse where a neo cassette was inserted to disrupt part of the PPIase and TPR domains (13). Homozygous mutants displayed posterior neural tube, dorsal root ganglion and optic defects, and were embryo lethal by E13.5. In this study, we describe a new null Fkbp8 mouse mutant that is not embryo lethal. More importantly, this mutant allele displays similar clinical features of human spina bifida, including cystic protrusion at the thoracico–lumbo–scaral region with a dysplastic spinal cord and vertebrae arches which failed to fuse at the dorsal midline, as well as lower body paralysis. Phenotypic analysis revealed that Fkbp8 mutants have a dilated neural tube. In addition, these mutants exhibit patterning defects and increased apoptosis in the neural tube. Gene expression analysis revealed primarily the perturbation of expression of several neural tube patterning genes in the mutants. Overall, these studies demonstrate that the Fkbp8 gene is not critical for embryo survival, but it is essential for spinal neural tube patterning, and to inhibit cell death in the developing neural tube. The new Fkbp8 mutant allele is a valuable experimental mouse model which will be useful in dissecting the etiology and pathogenesis of spinal NTDs in humans.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Identification of the new Fkbp8Gt(neo) null allele in mouse
The previously targeted Fkbp8 mutant has a severely dilated posterior neural tube and the homozygous mutants are embryo lethal by E13.5 (13). We discovered a gene-trapped Fkbp8 allele that produced a milder phenotype, resulting in mice that showed posterior NTD typical of spina bifida, but survived up to 5 months of age. Sequence comparison with genomic Fkbp8 revealed that the gene trap had inserted into the first intron at nucleotide position 1085, which is 666 bp upstream of the ATG (Fig. 1A). A PCR genotyping assay was performed to identify the wild-type and gene-trapped Fkbp8 alleles, which were 442 and 457 bp, respectively (Fig. 1B).


Figure 1
View larger version (27K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Gene structure, genotyping and transcript analysis of wild-type (WT) and gene-trapped (Gt) Fkbp8 (Fkbp8Gt(neo)) alleles. (A) Upper panel: mouse Fkbp8 gene has 10 exons and 3 transcriptional start sites as marked with black arrows. The mRNA isoforms are labeled as 1a-WT, 1b-WT and 1c-WT, and are translated from the ATG in exon 3. An in-frame ATG within the second intervening intron/exon (asterisk) produces a longer 1c-WT. Bottom panel: the gene trap inserted into intron 1, between the second and third transcriptional start site. The primers for RT–PCR and PCR-genotyping are indicated with smaller black arrows. (B) Example of Fkbp8 PCR-genotyping; WT allele is 442 bp and the Gt Fkbp8Gt(neo) allele is 457 bp. (C) RT–PCR analysis of transcripts from WT, heterozygous (WT/Gt) and homozygous mutant (Gt/Gt) embryos at E9.5. 1a-, 1b- and 1c-WT isoforms are barely detectable in the Gt/Gt samples. Fkbp8 coding regions like exons 3–6 (E3–E6) and exons 6–10 (E6–E10) are also absent. Arrow points to a non-specific product. 1a- and 1b-NEO transcripts are present only in WT/Gt and Gt/Gt samples. The BTK-Fkbp8 transcript is not produced in Gt-containing embryos because the PGK promoter is only active in ES cells. The GAPDH PCR product is shown as an internal RT control.

 
The mouse Fkbp8 gene has three alternate first exons designated as 1a, 1b and 1c, and four transcript variants generated by alternative promoter usage and mRNA splicing (14). We re-mapped the gene trap integration site, which was found to be between the transcriptional start sites 1b and 1c (Fig. 1A), suggesting that transcripts 1a and 1b are trapped, whereas 1c is not. Thus, wild-type Fkbp8 mRNA is produced in the gene-trapped (Gt) mutants. To verify this and also to characterize the different transcripts in the wild-type (WT), heterozygous (WT/Gt) and homozygous mutant (Gt/Gt) Fkbp8 embryos, RT–PCR was performed at E9.5 (Fig. 1C). Transcript-specific PCR of 1a-, 1b-, and 1c-WT amplicons spanned mostly the 5'-UTR and part of the ORF. 1a- and 1b-WT were detected in the WT and WT/Gt embryos, but not in the Gt/Gt mutants (Fig. 1C). Furthermore, the splice acceptor in the neomycin gene (NEO) enabled fusion to the first exon of 1a and 1b, generating 1a- and 1b-NEO transcripts, which were detected only in the WT/Gt and Gt/Gt Fkbp8 mutants, but not in the WT samples (Fig. 1C). Because the retroviral integration occurred upstream of the transcriptional start site of 1c, we asked whether 1c-WT transcripts were present in Gt-containing embryos. Surprisingly, 1c-WT was present in the WT and WT/Gt embryos, but was barely detectable in the Gt/Gt mutants (Fig. 1C). Examination of Fkbp8 mRNA including exons 3–6 and 6–10 revealed the same result, suggesting that either no or very little wild-type Fkbp8 mRNA was present in the Gt/Gt Fkbp8 mutants (Fig. 1C). This was substantiated by the absence of Fkbp8 mRNA in Gt/Gt Fkbp8 mutants as determined by whole-mount in situ hybridization (ISH) at E8.5 to E11 (Fig. 2C–E and data not shown).


Figure 2
View larger version (59K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Fkbp8Gt(neo) is a null allele of the mouse Fkbp8 gene. (A) Quantitative real-time PCR analysis at E9.5 and E10.5 revealed background levels of Fkbp8 mRNA in Gt/Gt Fkbp8Gt(neo) mutant embryos. The difference in expression level between WT and WT/Gt or Gt/Gt Fkbp8Gt(neo) embryos is significant as determined by Student’s t-test; n > 3 per genotype, ***P<0.0001. (B) Immunoblotting with anti-FKBP8 antibody detected FKBP8 protein expression in WT and WT/Gt, but not in Gt/Gt Fkbp8Gt(neo) embryos at E9.5. As an internal control, the blot was also reacted with anti-β-tubulin antibody. (C–E) In situ hybridization using full length Fkbp8 ORF probe revealed ubiquitous Fkbp8 expression in WT embryos at E8.5, E10 and E11. Fkbp8 mRNA expression was absent in Gt/Gt Fkbp8Gt(neo) embryos at these stages.

 


View this table:
[in this window]
[in a new window]

  		Table 1. Summary of Fkbp8Gt(neo) heterozygous intercross

 
We next validated our RT–PCR results with quantitative real-time PCR using E9.5 and E10.5 total RNA (Fig. 2A). While Fkbp8 mRNA was clearly detected in the WT, and to a lesser extent in the WT/Gt embryos, it was reduced to background levels in the Gt/Gt mutants (Fig. 2A; n > 3 embryos per genotype, ***P < 0.0001 by Student’s t-test). To examine whether wild-type FKBP8 protein levels were also reduced, we probed E9.5 protein extracts with an anti-FKBP8 antibody. As shown in Fig. 2B, while FKBP8 protein was clearly detected in the WT and WT/Gt embryos, it was absent in the Gt/Gt mutants. Taken together, the fact that neither Fkbp8 mRNA nor protein could be detected in the Gt/Gt Fkbp8 mutant clearly indicates that this mutant represents a null allele of the mouse Fkbp8 gene, and is designated as Fkbp8Gt(neo) allele.

Mouse Fkbp8 gene is not required for embryo survival but has an essential role for posterior neural tube development
The Gt/Gt Fkbp8Gt(neo) mutants are live born, and survive through postnatal stages. Morphological examination at E18.5 revealed that all mutant fetuses present with an isolated spina bifida, which spans the thoracic and lumbar region (Table 1; arrow in Fig. 3E). The defect is characterized by a skin-covered cystic protrusion containing a dysplastic spinal cord and cleft spine (insert in Fig. 3E). At birth, the mutant mice are not significantly different from their littermates in terms of body weight. As postnatal life progresses, the Fkbp8Gt(neo) mutants develop splayed hind limbs (arrows in Fig. 3F) and lower body paralysis. These mutants have great difficulty ambulating, which might hamper their ability to be adequately nursed. Possibly because of lack of nutrition, the mutants are smaller in size and the size difference relative to the unaffected littermates becomes increasingly evident with continued postnatal development. As shown in Fig. 3F, the mutant is one-third the size of its wild-type littermate.


Figure 3
View larger version (126K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3. Mouse Fkbp8 gene is essential for posterior neural tube development. (A) At E9.5, the mutant embryo is morphologically similar to WT. (B) In the Gt/Gt Fkbp8Gt(neo) mutants, the dilated neural tube (see arrow), which starts immediately caudal to the forelimb bud, is prominent at E10.5. (C) The edema in the spinal cord increases in volume and can be found at the level of the forelimb in the mutants. (D) In E15.5 Gt/Gt Fkbp8Gt(neo) mutants, protrusion of the spinal cord tissue is seen caudal to the forelimb (arrow). The spinal cord is irregular and appears folded (see inset). (E) At E18.5, the lesion spans the thoracic and lumbar region of the Gt/Gt Fkbp8Gt(neo) fetus (see arrow). The cystic protrusion is skin covered and contains a dysplastic spinal cord and cleft spine (see inset). (F) An example of a Gt/Gt Fkbp8Gt(neo) mutant surviving till 3 weeks postnatally, and is about one-third the size of its WT littermate, and has splayed hind limbs (arrows).

 
The spinal cord abnormality in the Gt/Gt Fkbp8Gt(neo) mice is evident at E10.5, with an edema posterior to the forelimb bud and extending to the caudal region (arrow in Fig. 3B). At E9.5, the posterior neural tube in the mutant appears normal (Fig. 3A). Between the stages of E11.5 and E13.5, the posterior neural tube dilates even more (Fig. 3C). From E14.5 onwards, the dilation subsides and the skin-covered defect contains protruded neural tissue that can be seen from the thoracic to the lumbosacral region (arrow in Fig. 3D and data not shown).

The analysis of tissue sections of E11.5 Gt/Gt Fkbp8Gt(neo) and wild-type fetuses revealed that the mutant neural tube at the lumbar region was only half the thickness of that in wild-type (Fig. 4C). Moreover, the dorsal root ganglia were smaller and disorganized in the mutant (arrows in Fig. 4C and D). At the cervical and thoracic level, the overall neural tube morphology did not differ significantly (Fig. 4A and data not shown). At E15.5, the mutant spinal cord at the thoracic level is significantly larger than that of the wild type, and it lacks the ‘butterfly’ characteristics of a normally differentiated spinal cord (compare Fig. 4E to F). There are expansion and folding of the spinal cord which worsen toward the caudal axis (inset in Fig. 3D). At the lumbar region, the mutant spinal cord shows no distinct distribution of gray and white matter. There is also discontinuity in the spinal cord thickness; the ventral aspect being thinner compared with the lateral and dorsal regions (Fig. 4G).


Figure 4
View larger version (94K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4. Homozygous Gt/Gt Fkbp8Gt(neo) mutants have spinal cord, dorsal root ganglion and vertebrae defects. Tissue sections of Gt/Gt Fkbp8Gt(neo) mutant and WT fetuses were examined at the thoracic and lumbar regions at E11.5, E15.5 and E18.5. All adjacent panels, e.g. (A) and (B), were taken at the same magnification to allow comparison. (A, B) The spinal cord and dorsal root ganglion (arrows) appear normal between E11.5 WT and mutant fetuses. (E, F) At E15.5, the mutant spinal cord is larger than that of WT, and the gray matter lacks the ‘butterfly’ pattern. (I) At E18.5, the mutant has a stretched spinal cord with no distinct marginal layer, ventral and dorsal grey horns. The cartilage primordia of the vertebral arches are not closed (asterisk), and the spinal cord is covered only by the skin. (J) The WT spinal cord is protected by the over and underlying vertebrae (asterisks). (C, D) In comparison to WT, the E11.5 mutants had smaller dorsal root ganglia (arrows) and 50% thinner spinal cords. (G, H) The E15.5 mutant spinal cord is thinner at the ventral aspect and lacks distinct distribution of gray and white matter tissue, compared with that in WT. (K, L) The skin is reduced to a thin layer, overlying very little spinal tissue in the E18.5 mutant (arrows). While normal vertebral structure protects the WT spinal cord (asterisks), the underlying vertebrae in the mutant is abnormally flattened (asterisk, and the dotted line separating spinal cord and underlying vertebrae).

 
At E18.5, the Gt/Gt Fkbp8Gt(neo) mutants displayed a drastically stretched spinal cord at the thoracic level (Fig. 4I). Unlike the wild-type, the mutant spinal cord per se has no distinct marginal layer, and lacks ventral and dorsal grey horns (compare Fig. 4I to J). The cartilage primordia of the vertebral arches are not fused (asterisk in Fig. 4I), and the spinal cord is covered only by the skin. By comparison, the spinal cord in wild-type fetuses is completely protected by the over and underlying vertebrae (asterisk in Fig. 4J). At the lumbar region of the mutants, the skin is reduced to a thin layer and overlying very little spinal tissue (arrows in Fig. 4K). In contrast to the vertebral structure developing around and protecting the wild-type spinal cord, the underlying vertebra in the mutant is flattened (asterisks in Fig. 4K and L). Taken together, these results clearly indicate that the spinal cord at the lumbar region of the Fkbp8Gt(neo) mutants is most significantly affected. At the thoracic level, the enlarged but less well-defined spinal cords at E15.5 and E18.5 also suggest that there could be proliferation and differentiation defects that might exacerbate toward the caudal axis of the affected mutants.

Fkbp8Gt(neo) mutants display severe bone and cartilage abnormalities
The analysis of E18.5 Gt/Gt Fkbp8Gt(neo) skeletal system with Alizarin Red/Alcian Blue staining revealed severe bone and cartilage defects, which were limited to the axial skeleton. All mutant axial skeletons had a normal number of vertebrae, but the total length of the spinal column was slightly shorter as a result of vertebrae malformations/fusions (Fig. 5A and B). Detailed skeletal analysis showed that the spinal dysraphism in affected fetuses ranged from thoracic (T2–5) to sacral (S4) vertebrae. The neural canal from T2–5 to S4 in all mutants remained open; the cartilaginous, dorsal ends of vertebral arches from both lateral sides did not fuse to form a neural arch (Fig. 5A, magnified view in C). In contrast, the wild-type fetuses have well-shaped cartilage and ossified vertebrae centers, and lateral vertebral arches with dorsal parts that are bent toward each other (Fig. 5B, magnified view in D).


Figure 5
View larger version (98K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5. Homozygous Fkbp8Gt(neo) mutants display severe bone and cartilage abnormalities. Skeletal staninig of spinal column of WT and Gt/Gt Fkbp8Gt(neo) E18.5 fetuses. (A, B) Dorsal view of cervical to sacral spinal column. Note the misshapen, fused vertebral centrums in thoracic, lumbar and sacral region of the Gt/Gt mutants. Vertebraes from T2 to S4 are dorsally open and fused to one another. Vertebral arches are shortened and directed laterally, demonstrating spina bifida in the thoracico-lumbo-sacral region of Gt/Gt Fkbp8Gt(neo) mutants. (C) Magnified thoracic region in the Gt/Gt Fkbp8Gt(neo) mutants showing deformed vertebral arches which are fused between consecutive vertebrae (see asterisks). (D) Magnified thoracic region in WT fetuses showing well-shaped cartilage, ossified vertebrae centers and fused vertebral arches at the dorsal midline. (E) Magnified lumbar region in the Gt/Gt Fkbp8Gt(neo) mutants. All cartilaginous and some ossified centers are fused (see asterisk and dotted line), and are irregular in shape. The neural canal is wider, and their cartilaginous parts are fused between consecutive vertebrae (see arrows). The vertebral arches are shortened, causing a more severe spina bifida in this region. (F) Magnified lumbar region in the WT fetus showing well-separated vertebrae and ossified vertebrae centers (see asterisk and dotted line). C, cervical; T, thoracic; L, lumbar; S, sacral.

 
The open neural canal in the Gt/Gt Fkbp8Gt(neo) widens toward the caudal axis (Fig. 5A and E). The vertebral arches are directed laterally, and their cartilaginous portions are fused between consecutive vertebrae (arrows in Fig. 5E). In contrast to the well-separated vertebral centers in the wild-type fetuses, the vertebral centers in the Gt/Gt Fkbp8Gt(neo) mutants are irregular, and are fused between consecutive vertebrae (asterisks in Fig. 5E and F). In addition, the cartilaginous part of the lumbar and sacral vertebrae is completely fused, forming a lumbo-sacral cartilage (dotted line in Fig. 5E, and compare to Fig. 5F). In the Gt/Gt Fkbp8Gt(neo) mutants, the last thoracic and all lumbar vertebrae are the most affected region of the spinal column, and these vertebral abnormalities are completely absent in the cervical region.

Endogenous FKBP8 activity prevents apoptosis in the posterior neural tube
As shown in HeLa and Huh7 cells, siRNA knockdown of Fkbp8 led to an increased apoptosis (15,16). FKBP8 inhibits apoptosis through its interaction with anti-apoptotic proteins such as BCL2 and BCL-XL, and targeting these proteins to the mitochondria (15,17). We questioned whether in vivo FKBP8 possesses anti-apoptotic functions, and if its deficiency leads to increased cell death in Gt/Gt Fkbp8Gt(neo) mutants. To this end, we performed TUNEL staining on tissue sections prepared from wild-type and mutant embryos, and quantified the number of TUNEL-positive cells. At E9.5, when the neural tube lumen was slightly enlarged in the mutant embryo, the number of TUNEL-positive cells (arrows in Fig. 6A and B) in the neural tube was comparable between wild-type and mutant embryos when analyzed at the lumbar levels (Fig. 6G). At E10, the number of TUNEL-positive cells increased dramatically in the mutant neural tube, and was over 2-fold higher than that in the wild-type (Fig. 6C, D and G; n > 3 embryos per genotype, P < 0.0001 by Student’s t-test). Cell death in the mutant neural tube was even more pronounced at E10.5. In the latter stages, TUNEL-positive cells were restricted to the ventral neural tube. In all examined stages, there was no significant difference in cell death at the thoracic level or other embryonic regions (data not shown). Similarly, cell proliferation in the wild-type and mutant neural tubes was identical, based on anti-phospho-Histone H3 immunostaining (data not shown), suggesting that increased cell death in the mutants was not because of overproliferation.


Figure 6
View larger version (69K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 6. Endogenous FKBP8 inhibits apoptosis in the posterior neural tube. (A–F) TUNEL staining of neural tube sections of lumbar region of WT and mutant embryos. All adjacent panels, e.g. (A) and (B), were taken at the same magnification to allow comparison. (A and B) At E9.5, the number of TUNEL-positive cells (green fluorescent-labeled, red arrows) was identical between WT and mutant. (C and D) Increased cell death in E10 mutant, compared with that in WT. (E and F) Cell death in the mutant was even more pronounced at E10.5. At both stages, cell death is restricted to the ventral aspect. (G) TUNEL-positive cells were counted at E9.5 and E10, and was over two-fold higher in E10 mutant compared with that in WT (***P < 0.0001, Student’s t-test). NS, not significant. (H–P) Cell death pattern was verified with nile blue sulfate (NBS) staining. (H) No detectable difference between E9.5 WT and mutant embryos in NBS staining patterns or intensities. (I and J) Intense NBS staining was seen in E10 and E10.5 mutant embryos in the caudal region. (K and L) Dorsal view of NBS-stained E10 WT and mutant embryo. Arrow points to neural tube, and the mutant had stronger staining. (M and N) NBS staining revealed an irregular mutant neural tube (arrowheads), which was absent in the WT. (O) NBS labeling was strongest in E10.5 mutant embryo, which intensified immediately caudal to the forelimb bud (see dotted line). (P) NBS staining in WT embryo showing cell death in the limb buds, neural tube and somites.

 
To verify that cell death occurs only at the lumbar region of the Fkbp8Gt(neo) mutants, we performed nile blue sulfate (NBS) staining, a vital stain that marks dying cells (18). At E9.5, there was no detectable difference in either NBS staining pattern or intensity (Fig. 6H). However, E10 or E10.5 mutant embryos displayed intense staining in the caudal region (Fig. 6I and J). The staining was stronger than that in the wild-type (arrows in Fig. 6K and L) and was restricted to the ventral aspect of the neural tube (data not shown). Consistent with our TUNEL data, cell death in E10.5 mutant embryo was most pronounced of all the examined stages (Fig. 6J; compare intensities between O and P). The NBS staining also revealed an irregular neural tube at the level of and caudal to the forelimb bud (arrowheads in Fig. 6M), which was absent in the wild-type (Fig. 6N). The staining intensity at this region was, however, similar to that in the wild-type, but was weaker than that in the caudal axis of the mutants (see dotted red line indicating boundary in Fig. 6M). The results from our cell death analysis clearly demonstrate that endogenous FKBP8 exhibits anti-apoptotic functions, specifically in the posterior neural tube.

Endogenous FKBP8 activity is required for posterior neural tube patterning
Finally, to identify genes or pathways of gene expression that are affected by the loss of Fkbp8, we performed microarray analysis using the posterior embryonic region isolated from E9.5 Gt/Gt Fkbp8Gt(neo) mutant and wild-type littermates. The dissected tissue included the region caudal to the forelimb bud and extending to the tail region. We chose E9.5 because it is the stage where cell death is not significantly different between both genotypes. This was to eliminate any non-specific gene expression changes as a result of apoptosis. The microarray analysis indicated a total of 110 genes with a differential gene expression in the Gt/Gt Fkbp8Gt(neo) mutants; 71 genes were downregulated, whereas 39 genes were upregulated, and some of these genes were further sub-categorized into three Gene Ontology (GO) terms, which included regulation of transcription, signal transduction activity and cell communication. As shown, most of these genes were expressed in the spinal cord (Table 2). Fkbp8 was the most downregulated gene (Table 2), which validated the usefulness of the microarray experiment. The microarray analysis did not reveal dynamic gene expression perturbations, and we chose to concentrate on transcriptional regulators because of the higher z-score (2.61). Subsets of transcriptional regulator genes identified to be significantly up- or downregulated were further validated by quantitative real-time PCR (qRT-PCR). Different collections of wild-type and mutant posterior embryonic tissues were again harvested and their total RNA was processed for qRT-PCR using Taqman® gene expression assays. Except for Hsf2, the direction (positive/negative regulation) was consistent between the microarray data and qRT-PCR results for the remaining 15 genes, indicating an excellent fidelity of the microarray analysis to identify gene expression changes (Table 3).


View this table:
[in this window]
[in a new window]

  		Table 2. Genes differentially expressed in homozygous Fkbp8Gt(neo) mutants at E9.5

 


View this table:
[in this window]
[in a new window]

  		Table 3. Validation by quantitative Real-time PCR (qRT-PCR) of differentially expressed genes in homozygous Fkbp8Gt(neo) mutants

 
We have previously observed ventralized neural tubes in the Gt/Gt Fkbp8Gt(neo) mutants, as evidenced by expanded Shh and Ben expression domains at E10.5 (R.L.Y. Wong, unpublished data). Consistently, in our microarray analysis, most of the ventral neural tube markers including Nkx2-9, Myt1, FoxA2 and Nkx6-1 were over-expressed in the mutants (Tables 2 and 3). In parallel, dorsal neural tube markers such as Zic1 and Scube2 were underexpressed (Tables 2 and 3). In situ expression analysis at E10 or E10.5 indicated an upregulation of the Shh gene in the mutant neural tube (data not shown), but it was only slightly over-expressed at E9.5, as indicated by microarray analysis (Table 3). Further examination using KEGG pathway analysis revealed a moderate increase in the expression of Shh pathway genes such as Ptch1, Smo and Glis, in the mutants (data not shown). Gene expression changes of other transcriptional regulators in the mutants were also noted, which included upregulation of pro-neuronal differentiation genes such as Neurod4, Lhx2 and Myt1 (Tables 2 and 3) (19–21). In parallel, the mouse Hes3 gene, which has been shown to be an antagonist for neuronal differentiation (22), was downregulated (Tables 2 and 3). To verify further that the gene expression changes could indeed be detected in vivo, we collected E9.5 wild-type and mutant embryos and performed ISH with FoxA2, Nkx2-9 and Hes3 probes. In the mouse, FoxA2 is expressed in the floor plate and notochord at E9.5 (23) (Fig. 7A). Nkx2-9, on the other hand, is expressed in the ventral domain of neuronal progenitors that give rise to V3 interneurons (24) (Fig. 7A). Hes-3 is also expressed in the spinal cord, but only at E8.5 and E9.5 (25) (Fig. 7A). Expression analysis of these genes revealed a correct pattern of up- or downregulation, consistent with the microarray and qRT-PCR data, and the gene expression change was restricted only to the neural tube (Fig. 7A). In the mutant neural tube, the FoxA2 and Nkx2-9 expression domains were expanded and the upregulation appeared caudal to the forelimb bud (Fig. 7A). In contrast, Hes3 expression in the mutant neural tube was only minimally maintained in the most posterior region (Fig. 7A). These results further substantiate the notion that the Gt/Gt Fkbp8Gt(neo) mutants have primarily neural tube patterning defects, which might contribute to the differentiation defects observed in the mutant spinal cords at later stages.


Figure 7
View larger version (88K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 7. Endogenous FKBP8 activity establishes dorso-ventral characteristics in the posterior neural tube (A) In the mutant neural tube, FoxA2 and Nkx2-9 expression domains were expanded and the upregulation was noted after the forelimb bud. Hes3 expression in the mutant neural tube at the caudal region was mostly lost and expression was only maintained at the most posterior region. (B) Zic1 expression in the E9.5 mutant was downregulated in the neural tube and somites (see arrowheads). Examination of E10 embryos revealed similar results in the Gt/Gt Fkbp8Gt(neo) mutants, though underexpression in the somites (see arrowheads) was not as significant compared with that at E9.5.

 
Next, we questioned whether any of the known spina bifida genes were underexpressed in our microarray data. Fgfr1, Itgb1, Lrp6, Msx1, Rab23 and Zic2 were only mildly downregulated, whereas Zic1 was the most underexpressed amongst this panel of genes. In the mouse Zic1 knockout, the dorsal horn mass was decreased in the thoracic spinal cord at E15.5 (26). We re-examined Gt/Gt Fkbp8Gt(neo) mutant spinal cords at E15.5 and found a smaller dorsal horn mass at the thoracic region; a phenotype similar to that of Zic1 mutant phenotype (Fig. 4E). In both zebrafish and mouse Zic1 loss-of-function mutants, the vertebral arches did not fuse along the dorsal midline. Additionally, the spinous processes, which arise from the most dorsal part of the vertebral arches, were also missing (27,28). These skeletal deformities were highly similar to that seen in our Gt/Gt Fkbp8Gt(neo) mutants (Fig. 5A). In our Gt/Gt Fkbp8Gt(neo) mutants, Zic1 was downregulated by –4.49-fold (***P < 0.0001) as determined by qRT-PCR analysis (Table 3). We examined endogenous Zic1 expression in WT and Gt/Gt Fkbp8Gt(neo) mutant embryos at E9.5 and found that in the mutants, Zic1 expression was indeed downregulated not only in the neural tube, but also in the somites (arrowheads in Fig. 7B). The analysis of E10 embryos revealed similar results in the Gt/Gt Fkbp8Gt(neo) mutants, although underexpression in the somites was not as significant compared with that at E9.5 (Fig. 7B). It is possible that the spinal cord and vertebral arch abnormalities seen in the Gt/Gt Fkbp8Gt(neo) mutants could be due, in part, to the diminished expression of Zic1 in the dorsal neural tube and somites.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
In this study, we present the phenotypic and molecular characterization of a new Fkbp8Gt(neo) mutant allele. We provide evidence that this is a null allele because neither wild-type Fkbp8 mRNA nor FKBP8 protein is detected in the homozygous mutants. We also show in the Fkbp8 loss-of-function mutants, that the posterior neural tube displays increased cell death and altered dorso-ventral characteristics, which most likely contribute to the differentiation defects in the mutant spinal cord. Mouse Fkbp8 deficiency is not embryo lethal, and its phenotypes are reminiscent of human spina bifida cases, which include a cystic protrusion containing a dysplastic spinal cord, the unfused vertebral arches at dorsal midline and disability. Our studies demonstrate that mouse Fkbp8 gene is not required for embryo survival, but is important for normal spinal neural tube patterning, and to inhibit cell death, in the developing neural tube.

Gene-trapped mutants can sometimes have tissue-specific skipping, or splicing around gene trap insertions, which will produce wild-type transcripts (29–31). We verified that the Fkbp8Gt(neo) allele is not a hypomorph, and as demonstrated in our homozygous mutants, wild-type mRNA was undetectable from early to late-gestation stages. Even though the retroviral integration occurred upstream of the transcriptional start site of transcript 1c-WT, we failed to detect any 1c-WT in the Gt/Gt mutants. This might be because of the presence of the 5.2 kb gene trapping cassette, which potentially hindered transcriptional initiation of 1c-WT.

Mice carrying homozygous Fkbp8Gt(neo) alleles do not die in utero, which is in stark contrast to the targeted Fkbp8 mutants which died at E13.5 (13). It is unlikely that the latter mutants died of posterior NTDs because such defects do not cause embryonic lethality. Both the targeted and Gt Fkbp8 mutants were analyzed on a mixed 129/C57 genetic background (T. Li, personal communication). As demonstrated in our studies using Fkbp8Gt(neo) mutants, mouse Fkbp8 gene is dispensable for embryo survival. The mid-gestational lethality in the targeted Fkbp8 allele is likely because of the intact neo cassette which could have impact on neighboring gene expression. One gene that could be impacted is elongation factor RNA polymerase II (Ell), which lies ~4 kb downstream of the Fkbp8 gene. Homozygous Ell null mutants die prior to E6.5 (32). Thus, it is possible that Ell expression was compromised in the targeted Fkbp8 mutants, which led to lethality at E13.5. One could also argue that Ell expression is dependent upon cis-regulatory elements in the TPR and PPIase domains, and that Ell expression was affected by the loss of these elements in the targeted Fkbp8 mutants. Taking this notion into consideration, the Fkbp8Gt(neo) allele is thus a more favorable loss-of-function model because all cis-regulatory elements are intact. The fact that the Fkbp8Gt(neo) allele is not embryo lethal, and yet present with identical knockout phenotypes to those of the targeted Fkbp8 mutants raises the possibility that the Fkbp8Gt(neo) allele is the bona fide Fkbp8 null allele.

In the targeted Fkbp8 mutants, the most striking phenotype is the dilated neural tube (13). We have observed an identical neural tube phenotype in our Fkbp8Gt(neo) model in terms of the onset, extent and morphological characteristics of the defect. The earliest visual indication of the spinal cord abnormality in our Fkbp8Gt(neo) mutants is suggested by an edematous region restricted to the posterior neural tube. At later stages, these mutants also develop dorsal root ganglion and skeletal abnormalities. It is possible that the latter defects are secondary to the spinal cord defect; the increasing pressure from within the spinal cord could exert a physical obstruction and affect the development of immediately adjacent tissues, such as the dorsal root ganglia and somites. However, we cannot exclude the possibility that intrinsic gene expression change could in addition or by itself, contribute to these abnormally developed structures. For example, Zic1 expression is downregulated in the somites at E9.5, coinciding with the time when the neural tube lumen is only slightly enlarged, and the neural tube per se, is morphologically normal. Zic1 is one of the 26 spina bifida associated genes in the mouse (5), and its mutant phenotypes include spinal cord and vertebral arch defects (26–28), which are similar to those observed in our mutants. In a properly developed spinal cord, the nerve fibers in the spinal cord transmit sensory information toward the brain and motor signals to the appropriate parts of the body. In our Fkbp8Gt(neo) mutants, the spinal cord is profoundly malformed, and are most likely to suffer from significant spinal nerve damage, particularly at the lumbo-sacral level, and as a result, display all the features of lower body paralysis.

The mouse Fkbp8 gene is ubiquitously and uniformly expressed during the period of neural tube closure and development; therefore, it is not clear why the loss of Fkbp8 would impact specifically the caudal neural tube. One possible explanation is functional redundancy among the FKBP gene family. In mouse, the spinal cord development is thought to integrate both FGF and retinoic acid (RA) signals. FGF signaling is required for the continued growth for the spinal cord whereas RA signaling induces neural differentiation throughout the spinal cord (33). Interestingly, spina bifida was observed in mice which had excessive RA signaling (34,35). We questioned whether the loss of Fkbp8 might increase RA signaling in the Gt/Gt Fkbp8Gt(neo) embryos and in so doing, contribute to the development of spina bifida. The analysis of expression of the Raldh2 gene, which is a major RA synthesizing enzyme (36), showed no significant difference between the mutant and wild-type embryos, indicating that spina bifida in the Fkbp8Gt(neo) mutants was caused by RA signaling-independent mechanisms (K.S. Min, unpublished data).

In the neural tubes of the Fkbp8Gt(neo) mutants, cell death is always localized at the ventral aspect. Consistently, the mutant spinal cord tissues at later stages are also comparatively thinner, suggesting that most of the ventral neural tube cells were eliminated via apoptosis. Interestingly, the pattern of cell death at E10 and E10.5 is similar to the expanded Shh expression domain (R.L.Y. Wong, unpublished data). Shh has been shown to induce the death of ventral neuronal precursors and floor plate cells, resulting in a loss of cells in the chick neural tube (37). Thus, it is possible that the observed cell death in the ventral neural tube of the Fkbp8Gt(neo) mutants might similarly be induced by Shh overexpression. As shown by Bulgakov et al. (13), cellular apoptosis in the targeted Fkbp8 mutants was found to be normal. In their studies, activated caspase 3 was used as a cell death marker, whereas we used TUNEL and NBS staining, which gave consistent results in all examined developmental stages. Both activated caspase 3 and TUNEL staining are frequently used to assess cellular apoptosis and have been shown to co-localize in apoptotic cells (38,39). Judging from the morphology of the targeted Fkbp8 mutant neural tube, the discrepancy in the cell death results between the two Fkbp8 mutants is most likely because of the gestational age of the embryos used for the analysis. More importantly, our studies have highlighted the critically important function of FKBP8, which is to inhibit cell death, specifically in the posterior neural tube. Our data is in good agreement with previously published in vitro studies, which showed FKBP8’s anti-apoptotic function via siRNA knockdown (15,16). In the Splotch mutant mouse model, excessive cell death was observed in the neuroepithelial cells at the region of the spina bifida (40). Similarly, excessive cell death was detected in the neural tube of an RA-induced spina bifida mouse model (35). It is possible that the increased cell death in the Fkbp8Gt(neo) mutant neural tube could, in part, explain the spina bifida phenotype in these mice.

Microarray analysis of the posterior embryonic region revealed primarily gene expression perturbations of transcriptional regulators in the E9.5 Fkbp8Gt(neo) mutants. In these mutant neural tubes, disruption of the regional dorso-ventral characteristics would inevitably account for the spinal cord differentiation defects observed at later developmental stages. It has been suggested that Fkbp8 negatively regulates the Shh pathway (13). Even though the Shh pathway, per se, is not overtly upregulated at E9.5 in the Fkbp8Gt(neo) mutants, correct overexpression or repression of various Shh-responsive genes such as FoxA2, Nkx2-9, Nkx6-1, Pax6 and Dbx1 indicates that there is still increased Shh signaling. From our observations, Shh overexpression is more evident at E10 and E10.5 (R.L.Y. Wong, unpublished data). Excessive Shh signaling has been observed in open brain (opb) and Protein Kinase A-deficient mouse mutants, which had closed, but deformed neural tubes (41–43). Similar to the Gt/Gt Fkbp8Gt(neo) mutants, these mutants have a dilated posterior neural tube and neural tube patterning defects. While it is likely that excessive Shh signaling is responsible for the cell death and patterning defects in the Gt/Gt Fkbp8Gt(neo) mutants, it remains unresolved whether Shh overexpression alone is sufficient to induce the dilated posterior NTD.

The conventional rescue regime for mouse NTDs is through maternal folic acid or inositol supplementation (44,45). We questioned whether the spina bifida phenotype in the Gt/Gt mutants could be rescued through maternal supplementation with these agents. Unfortunately, neither folic acid nor inositol had any effect on lowering either the occurrence, or severity of spina bifida in the mutant fetuses (Supplementary Material, Table S5), suggesting that the loss of Fkbp8 affects molecular pathways that are resistant to folic acid- and inositol-mediated rescue effects. This Gt Fkbp8Gt(neo) allele could thus be a useful model to study folate- and inositol-resistant NTDs, which is an important sub-population comprising up to 50% of all NTDs.

From a clinical perspective, the Gt/Gt Fkbp8Gt(neo) mutant mouse model, being 100% penetrant and providing liveborn affected offspring, will be a valuable experimental model to understand the cellular and molecular pathogenesis of spinal NTDs from conception, through birth and into early adulthood. Because this model does not exhibit the disadvantages seen in current spina bifida animal models and more importantly, as it recapitulates many clinical features of human spina bifida, it should represent a relevant model system with a great potential to study these malformations. In closing, with the aid of the Fkbp8Gt(neo) mouse model, we have demonstrated the important roles of FKBP8 in neural tube patterning, and blocking apoptosis in the developing posterior neural tube. Understanding abnormal gene function at the cellular and molecular levels provides an entry point in efforts to find appropriate intervention strategies to prevent the occurrence of these devastating birth defects.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Generation and genotyping of mutant Fkbp8Gt(neo) mice and drug rescue
The ES cell clone OST287985 containing a gene trap insertion in the Fkbp8 gene was identified from the Omnibank database from Lexicon Pharmaceuticals Inc. (The Woodlands, TX, USA), and microinjected into C57BL/6 blastocysts to generate germline chimera. Chimeric males were bred to C57BL/6 females for germline transmission. Animal experimentation was approved by the IBT Animal Care Committee. Mice used in this study were of mixed 129/C57 background. Either tail or yolk sac genomic DNA was extracted and used for genotyping (primer sequences in Table S4, Supplementary Material). For rescue experiments, at least six pregnant Fkbp8Gt(neo) dams were daily gavaged from E0.5 to E18.5 with either 40 mg/kg folic acid (Merck Eprova AG, Schaffhausen, Switzerland) or 400 mg/kg inositol (Sigma-Aldrich, St. Louis, MO, USA). Spina bifida in the E18.5 fetuses was scored via external morphological examination and skeletal staining.

Analysis of mRNA and protein expression level in wild-type and mutant Fkbp8Gt(neo) mice
Total RNA was isolated from WT, heterozygous (WT/Gt) and homozygous (Gt/Gt) Fkbp8Gt(neo) embryos at E9.5 and purified using PureLink Total RNA Purification System (Invitrogen, Carlsbad, CA, USA). First-strand synthesis was oligo-d(T)-primed and different Fkbp8 cDNA regions, NEO and BTK fusion transcripts, and GAPDH were then amplified. For real-time PCR, Fkbp8 mRNA levels were determined using a TaqMan® probe Mm.141864 (ABI, Foster City, CA, USA). Total RNA samples were purified and reverse-transcribed as described previously. The assays were performed according to manufacturer’s protocol on an ABI PRISM® 7900HT Sequence Detection System (ABI). Mouse GAPDH TaqMan® probe (4352339E) (ABI) was used as RNA input control and to normalize Fkbp8 mRNA levels. Fold changes of Fkbp8 expression were calculated based on the cycle differences between WT and WT/Gt or Gt/Gt Fkbp8Gt(neo) samples using the {Delta}{Delta}CT method. Student's t-test was used to assess statistical significance. Real-time PCR was also used to validate a subset of microarray clones using gene-specific TaqMan® probes and following the same procedures to determine expression level changes between E9.5 WT and Gt/Gt Fkbp8Gt(neo) samples. All RT–PCR and real-time experiments were repeated using different RNA batches, and also E10.5 and E16.5 RNA samples, and identical results were obtained.

To detect FKBP8 protein by western blotting, E9.5 WT, WT/Gt and Gt/Gt Fkbp8Gt(neo) embryos were homogenized in lysis buffer (1% SDS, 1 mM sodium ortho-vanadate, 10 mM Tris, pH 7.4) containing 1x protease inhibitor cocktail (Sigma-Aldrich). An amount of 5 µg of protein lysate was loaded and immunoblotted with a rabbit polyclonal antibody to FKBP8 [provided by Dr. Frank Edlich, Max-Planck Research Unit for Enzymology of Protein Folding, Halle (Saale), Germany], and a mouse monoclonal antibody to β-tubulin (Abcam Inc., Cambridge, MA, USA), which was used as a protein loading control. Immunoblotting was repeated at least twice with different batches of protein samples and identical results were obtained using E10.5 samples.

Histological and skeletal analysis, ISH and cell death analysis
Wild-type and Gt/Gt Fkbp8Gt(neo) fetuses from various stages were harvested, fixed and processed for standard histological analysis. E18.5 WT and Gt/Gt Fkbp8Gt(neo) fetuses were subjected to bone and cartilage staining, which was performed according to previously described methods but with minor modifications (46). Double-stained skeletons were examined under a Leica MZ 95 stereomicroscope and digital pictures were taken (Leica Microsystems Digital Camera DFC480, Wetzlar, Germany). Skeletal abnormalities were scored according to the methods published earlier (47).

To detect Fkbp8 mRNA, an image clone 3601511 that contains the full length mouse Fkbp8 ORF was used to synthesize the antisense RNA probe. Whole-mount ISH was performed using InsituPro robot (INTAVIS Bioanalytical Instruments AG, Koeln, Germany). ISH was also used to validate expression changes of microarray clones. Part of the cDNAs was PCR amplified and used for RNA probe synthesis. TUNEL staining was performed according to the manufacturer’s protocol (In Situ Cell Death Detection kit, Roche, Indianapolis, IN, USA). TUNEL-positive cells from the anterior and posterior regions were counted from at least five sections from each embryo at E9.5 and E10 (n > 3 embryos per genotype). Student's t-test was used to assess statistical significance of cell death between WT and Gt/Gt mutants. For whole-mount NBS staining, embryos were dissected in cold PBS and incubated at 37°C in 10 mg/ml NBS solution (Sigma-Aldrich) in PBS containing 0.1% Tween 20 (PBT). The embryos were then briefly rinsed in PBT and photographed immediately.

Microarray hybridization, signal detection and microarray statistical analysis
Total RNA was isolated from the posterior region of E9.5 WT and Gt/Gt Fkbp8Gt(neo) mutant embryos (three embryos per genotype) using methods described previously. The quality and concentration of total RNA samples were analyzed on the Agilent Bioanalyzer 2100 (Agilent, Palo Alto, CA, USA). RNA synthesis and preparation of biotin-labeled cRNA were performed according to the manufacturer’s protocol (GE Healthcare, Piscataway, NJ, USA). Biotinylated cRNA samples were then hybridized to CodeLink Mouse Whole Genome Bioarray (GE Healthcare), which contains ~36 000 mouse gene targets. Six bioarrays were hybridized following the conditions as outlined by the manufacturer (GE Healthcare). The microarray slides were scanned using a GenePix 4000B Array Scanner (Molecular Devices, Palo Alto, CA, USA) and the image from each bioarray was processed using the CodeLink Expression Analysis Software version 4.1 (GE Healthcare). Briefly, the signal intensities of spots were extracted from the scanned images and the expression values were globally normalized to the median expression value of the whole array spots. The normalized expression data were then imported into GeneSifter (VizX Labs, Seattle, WA, USA) for further data analysis, where a standard pairwise t-test analysis with Benjamin Hochberg correction was applied, and following selection criteria of fold change ≥1.7 and P < 0.05 to filter out significant changes. Z-scores were calculated to indicate the statistical significance of detecting the differentially expressed genes with GO terms, and a score of ≥2 or ≤–2 was considered statistically significant. The gene expression data were deposited on the NCBI Gene Expression Omnibus website (http://www.ncbi.nlm.nih.gov/projects/geo/) and can be accessed via accession number GSE9604 [NCBI GEO] .


   SUPPLEMENTARY MATERIAL
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
FUNDING
 REFERENCES
 
Supplementary material is available at HMG Online.


   FUNDING
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
REFERENCES
 
This work is supported by NIH grant to RHF (DE016315).


   ACKNOWLEDGEMENTS
 
We would like to thank Drs Brad Amendt and Laura Mitchell (IBT, The Texas A&M University System Health Science Center) for their helpful suggestions on this manuscript. We are also grateful to Dr Frank Edlich [Max-Planck Research Unit for Enzymology of Protein Folding, Halle (Saale), Germany] for providing us with the anti-FKBP8 antibody and to Dr. Candace Chi (Renovis Corporation, South San Francisco, CA) for her excellent technical advice on NBS staining. We would also like to thank the IBT core facility for their technical assistance with the in situ hybridization experiments.

Conflict of Interest statement. None declared.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 

  1. Manning N., Archer N. Treatment and outcome of serious structural congenital heart disease. Semin. Neonatol. (2001) 6:37–47.[CrossRef][Medline]

  2. Christianson A.L., Howson C.P., Modell B. March of Dimes Global Report on Birth Defects: the Hidden Toll of Dying and Disabled Children. (2006) New York, USA: March of Dimes Birth Defects Foundation.

  3. Mersereau P., Kilker K., Carter H., Fassett E., Williams G.J., Flores A., Prue C., Williams L., Mai C., Mulinare J. Spina bifida and anencephaly before and after folic acid mandate – United States, 1995–1996 and 1999–2000. In: Morbidity and Mortality Weekly Report (2004) GA, USA: Centers for Disease Control and Prevention.

  4. Botto L.D., Moore C.A., Khoury M.J., Erickson J.D. Neural-tube defects. N. Engl. J. Med. (1999) 341:1509–1519.[Free Full Text]

  5. Harris M.J., Juriloff D.M. Mouse mutants with neural tube closure defects and their role in understanding human neural tube defects. Birth Defects Res. A Clin. Mol. Teratol. (2007) 79:187–210.[CrossRef][Web of Science][Medline]

  6. Snyder S.H., Sabatini D.M., Lai M.M., Steiner J.P., Hamilton G.S., Suzdak P.D. Neural actions of immunophilin ligands. Trends Pharmacol. Sci. (1998) 19:21–26.[CrossRef][Medline]

  7. Göthel S.F., Marahiel M.A. Peptidyl-prolyl cis–trans isomerases, a superfamily of ubiquitous folding catalysts. Cell Mol. Life. Sci. (1999) 55:423–436.[CrossRef][Web of Science][Medline]

  8. Wang H.Q., Nakaya Y., Du Z., Yamane T., Shirane M., Kudo T., Takeda M., Takebayashi K., Noda Y., Nakayama K.I., Nishimura M. Interaction of presenilins with FKBP38 promotes apoptosis by reducing mitochondrial Bcl-2. Hum. Mol. Genet. (2005) 14:1889–1902.[Abstract/Free Full Text]

  9. Barth S., Nesper J., Hasgall P.A., Wirthner R., Nytko K.J., Edlich F., Katschinski D.M., Stiehl D.P., Wenger R.H., Camenisch G. The peptidyl prolyl cis/trans isomerase FKBP38 determines hypoxia-inducible transcription factor prolyl-4-hydroxylase PHD2 protein stability. Mol. Cell. Biol. (2007) 27:3758–3768.[Abstract/Free Full Text]

  10. Edlich F., Erdmann F., Jarczowski F., Moutty M.C., Weiwad M., Fischer G. The Bcl-2 regulator FKBP38-calmodulin-Ca2+ Is inhibited by Hsp90. J. Biol. Chem. (2007) 282:15341–15348.[Abstract/Free Full Text]

  11. Moore J.M., Peattie D.A., Fitzgibbon M.J., Thomson J.A. Solution structure of the major binding protein for the immunosuppressant FK506. Nature (1991) 351:248–250.[CrossRef][Medline]

  12. Wiederrecht G., Hung S., Chan H.K., Marcy A., Martin M., Calaycay J., Boulton D., Sigal N., Kincaid R.L., Siekierka J.J. Characterization of high molecular weight FK-506 binding activities reveals a novel FK-506-binding protein as well as a protein complex. J. Biol. Chem. (1992) 267:21753–21760.[Abstract/Free Full Text]

  13. Bulgakov O.V., Eggenschwiler J.T., Hong D.H., Anderson K.V., Li T. FKBP8 is a negative regulator of mouse sonic hedgehog signaling in neural tissues. Development (2004) 131:2149–2159.[Abstract/Free Full Text]

  14. Nielsen J.V., Mitchelmore C., Pedersen K.M., Kjaerulff K.M., Finsen B., Jensen N.A. Fkbp8: novel isoforms, genomic organization, and characterization of a forebrain promoter in transgenic mice. Genomics (2004) 83:181–192.[CrossRef][Web of Science][Medline]

  15. Shirane M., Nakayama K.I. Inherent calcineurin inhibitor FKBP38 targets Bcl-2 to mitochondria and inhibits apoptosis. Nat. Cell Biol. (2003) 5:28–37.[CrossRef][Web of Science][Medline]

  16. Wang J., Tong W., Zhang X., Chen L., Yi Z., Pan T., Hu Y., Xiang L., Yuan Z. Hepatitis C virus non-structural protein NS5A interacts with FKBP38 and inhibits apoptosis in Huh7 hepatoma cells. FEBS Lett. (2006) 580:4392–4400.[CrossRef][Web of Science][Medline]

  17. Massaad C.A., Portier B.P., Taglialatela G. Inhibition of transcription factor activity by nuclear compartment-associated Bcl-2. J. Biol. Chem. (2004) 279:54470–54478.[Abstract/Free Full Text]

  18. Tone S., Kanaka S., Kato Y. The inhibitory effect of 5-bromodeoxyuridine on the programmed cell death in the chick limb. Dev. Growth Differ. (1983) 25:381–391.[CrossRef][Web of Science]

  19. Takebayashi K., Takahashi S., Yokota C., Tsuda H., Nakanishi S., Asashima M., Kageyama R. Conversion of ectoderm into a neural fate by ATH-3, a vertebrate basic helix-loop-helix gene homologous to Drosophila proneural gene atonal. EMBO J (1997) 16:384–395.[CrossRef][Web of Science][Medline]

  20. Kolterud A., Alenius M., Carlsson L., Bohm S. The Lim homeobox gene Lhx2 is required for olfactory sensory neuron identity. Development (2004) 131:5319–5326.[Abstract/Free Full Text]

  21. Perron M., Opdecamp K., Butler K., Harris W.A., Bellefroid E.J. X-ngnr-1 and Xath3 promote ectopic expression of sensory neuron markers in the neurula ectoderm and have distinct inducing properties in the retina. Proc. Natl. Acad. Sci. USA (1999) 96:14996–145001.[Abstract/Free Full Text]

  22. Hirata H., Ohtsuka T., Bessho Y., Kageyama R. Generation of structurally and functionally distinct factors from the basic helix-loop-helix gene Hes3 by alternative first exons. J. Biol. Chem. (2000) 275:19083–19089.[Abstract/Free Full Text]

  23. Echelard Y., Epstein D.J., St-Jacques B., Shen L., Mohler J., McMahon J.A., McMahon A.P. Sonic hedgehog, a member of a family of putative signaling molecules, is implicated in the regulation of CNS polarity. Cell (1993) 75:1417–1430.[CrossRef][Web of Science][Medline]

  24. Pabst O., Herbrand H., Arnold H.H. Nkx2-9 is a novel homeobox transcription factor which demarcates ventral domains in the developing mouse CNS. Mech. Dev. (1998) 73:85–93.[CrossRef][Web of Science][Medline]

  25. Hirata H., Tomita K., Bessho Y., Kageyama R. Hes1 and Hes3 regulate maintenance of the isthmic organizer and development of the mid/hindbrain. EMBO J. (2001) 20:4454–4466.[CrossRef][Web of Science][Medline]

  26. Aruga J., Tohmonda T., Homma S., Mikoshiba K. Zic1 promotes the expansion of dorsal neural progenitors in spinal cord by inhibiting neuronal differentiation. Dev. Biol. (2002) 244:329–341.[CrossRef][Web of Science][Medline]

  27. Aruga J., Mizugishi K., Koseki H., Imai K., Balling R., Noda T., Mikoshiba K. Zic1 regulates the patterning of vertebral arches in cooperation with Gli3. Mech. Dev. (1999) 89:141–150.[CrossRef][Web of Science][Medline]

  28. Ohtsuka M., Kikuchi N., Yokoi H., Kinoshita M., Wakamatsu Y., Ozato K., Takeda H., Inoko H., Kimura M. Possible roles of zic1 and zic4, identified within the medaka Double anal fin (Da) locus, in dorsoventral patterning of the trunk-tail region (related to phenotypes of the Da mutant). Mech. Dev. (2004) 121:873–882.[CrossRef][Web of Science][Medline]

  29. Elmore C.L., Wu X., Leclerc D., Watson E.D., Bottiglieri T., Krupenko N.I., Krupenko S.A., Cross J.C., Rozen R., Gravel R.A., Matthews R.G. Metabolic derangement of methionine and folate metabolism in mice deficient in methionine synthase reductase. Mol. Genet. Metab. (2007) 91:85–97.[CrossRef][Web of Science][Medline]

  30. Voss A.K., Thomas T., Gruss P. Efficiency assessment of the gene trap approach. Dev. Dyn. (1998) 212:171–180.[CrossRef][Web of Science][Medline]

  31. Carroll E.A., Gerrelli D., Gasca S., Berg E., Beier D.R., Copp A.J., Klingensmith J. Cordon-bleu is a conserved gene involved in neural tube formation. Dev. Biol. (2003) 262:16–31.[CrossRef][Web of Science][Medline]

  32. Mitani K., Yamagata T., Iida C., Oda H., Maki K., Ichikawa M., Asai T., Honda H., Kurokawa M., Hirai H. Nonredundant roles of the elongation factor MEN in postimplantation development. Biochem. Biophys. Res. Commun. (2000) 279:563–567.[CrossRef][Web of Science][Medline]

  33. Maden M. Retinoids and spinal cord development. J. Neurobiol. (2006) 66:726–738.[CrossRef][Web of Science][Medline]

  34. Abu-Abed S., Dolle P., Metzger D., Beckett B., Chambon P., Petkovich M. The retinoic acid-metabolizing enzyme, CYP26A1, is essential for normal hindbrain patterning, vertebral identity, and development of posterior structures. Genes Dev. (2001) 15:226–240.[Abstract/Free Full Text]

  35. Alles A.J., Sulik K.K. Retinoic acid-induced spina bifida: evidence for a pathogenetic mechanism. Development (1990) 108:73–81.[Abstract]

  36. Niederreither K., McCaffery P., Dräger U.C., Chambon P., Dollé P. Restricted expression and retinoic acid-induced downregulation of the retinaldehyde dehydrogenase type 2 (RALDH-2) gene during mouse development. Mech. Dev. (1997) 62:67–78.[CrossRef][Web of Science][Medline]

  37. Oppenheim R.W., Homma S., Marti E., Prevette D., Wang S., Yaginuma H., McMahon A.P. Modulation of early but not later stages of programmed cell death in embryonic avian spinal cord by Sonic hedgehog. Mol. Cell. Neurosci. (1999) 13:348–361.[CrossRef][Web of Science][Medline]

  38. Mirkes P.E., Little S.A., Umpierre C.C. Co-localization of active caspase-3 and DNA fragmentation (TUNEL) in normal and hyperthermia-induced abnormal mouse development. Teratology (2001) 63:134–143.[CrossRef][Web of Science][Medline]

  39. Wang C.Z., Johnson K.M. The role of caspase-3 activation in phencyclidine-induced neuronal death in postnatal rats. Neuropsychopharmacology (2007) 32:1178–1194.[CrossRef][Web of Science][Medline]

  40. Pani L., Horal M., Loeken M.R. Rescue of neural tube defects in Pax-3-deficient embryos by p53 loss of function: implications for Pax-3-dependent development and tumorigenesis. Genes Dev. (2002) 16:676–680.[Abstract/Free Full Text]

  41. Eggenschwiler J.T., Espinoza E., Anderson K.V. Rab23 is an essential negative regulator of the mouse Sonic hedgehog signalling pathway. Nature (2001) 412:194–198.[CrossRef][Medline]

  42. Günther T., Struwe M., Aguzzi A., Schughart K. Open brain, a new mouse mutant with severe neural tube defects, shows altered gene expression patterns in the developing spinal cord. Development (1994) 120:3119–3130.[Abstract]

  43. Huang Y., Roelink H., McKnight G.S. Protein kinase A deficiency causes axially localized neural tube defects in mice. J. Biol. Chem. (2002) 277:19889–19896.[Abstract/Free Full Text]

  44. Greene N.D., Copp A.J. Inositol prevents folate-resistant neural tube defects in the mouse. Nat. Med. (1997) 3:60–66.[CrossRef][Web of Science][Medline]

  45. Carter M., Ulrich S., Oofuji Y., Williams D.A., Ross M.E. Crooked tail (Cd) models human folate-responsive neural tube defects. Hum. Mol. Genet. (1999) 8:2199–2204.[Abstract/Free Full Text]

  46. Kimmel C.A., Trammell C. A rapid procedure for routine double staining of cartilage and bone in fetal and adult animals. Stain Technol. (1981) 56:271–273.[Web of Science][Medline]

  47. Wise L.D., Beck S.L., Beltrame D., Beyer B.K., Chahoud I., Clark R.L., Clark R., Druga A.M., Feuston M.H., Guittin P., et al. Terminology of developmental abnormalities in common laboratory mammals (Version 1). Teratology (1997) 55:249–292.[CrossRef][Web of Science][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Hum Mol GenetHome page
R. Y. Walder, B. Yang, J. B. Stokes, P. A. Kirby, X. Cao, P. Shi, C. C. Searby, R. F. Husted, and V. C. Sheffield
Mice defective in Trpm6 show embryonic mortality and neural tube defects
Hum. Mol. Genet., November 15, 2009; 18(22): 4367 - 4375.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
V. L. Patterson, C. Damrau, A. Paudyal, B. Reeve, D. T. Grimes, M. E. Stewart, D. J. Williams, P. Siggers, A. Greenfield, and J. N. Murdoch
Mouse hitchhiker mutants have spina bifida, dorso-ventral patterning defects and polydactyly: identification of Tulp3 as a novel negative regulator of the Sonic hedgehog pathway
Hum. Mol. Genet., May 15, 2009; 18(10): 1719 - 1739.
[Abstract] [Full Text] [PDF]

Home page
 Sci SignalHome page
I. F. Tsigelny, V. L. Kouznetsova, D. E. Sweeney, W. Wu, K. T. Bush, and S. K. Nigam
Analysis of Metagene Portraits Reveals Distinct Transitions During Kidney Organogenesis
Sci. Signal., December 9, 2008; 1(49): ra16 - ra16.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
17/4/587    most recent
ddm333v2
ddm333v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Wong, R. L. Y.
Right arrow Articles by Finnell, R. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wong, R. L. Y.
Right arrow Articles by Finnell, R. H.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?