Neural tube defects and neuroepithelial cell death in Tulp3 knockout ...

© 2001 Oxford University Press

Human Molecular Genetics, 2001, Vol. 10, No. 12 1325–1334

Neural tube defects and neuroepithelial cell death in Tulp3 knockout mice Akihiro Ikeda, Sakae Ikeda, Thomas Gridley, Patsy M. Nishina and Jürgen K. Naggert+ The Jackson Laboratory, 600 Main Street, Bar Harbor, ME 04609, USA Received 9 March 2001; Revised and Accepted 4 April 2001

The tubby-like protein 3 (Tulp3) gene has been identified as a member of a small novel gene family which is primarily neuronally expressed. Mutations in two of the family members, tub and tulp1, have been shown to cause neurosensory disorders. To determine the in vivo function of Tulp3, we have generated a germline mutation in the mouse Tulp3 gene by homologous recombination. Embryos homozygous for the Tulp3 mutant allele exhibit failure of neural tube closure, and die by embryonic day 14.5. Failure of cranial neural tube closure coincided with increased neuroepithelial apoptosis specifically in the hindbrain and the caudal neural tube. In addition, the number of βIII-tubulin positive cells is significantly decreased in the hindbrain of Tulp3–/– embryos. These results suggest that disruption of the Tulp3 gene affects the development of a neuronal cell population. Interestingly, some Tulp3 heterozygotes also manifest embryonic lethality with neuroepithelial cell death. Our results demonstrate that the Tulp3 gene is essential for embryonic development in mice. INTRODUCTION Tubby-like protein 3 (Tulp3) is a member of a small gene family which also includes Tub, Tulp1 and Tulp2 (1,2). A mutation in the tubby gene, originally identified as a splicing defect in the C-terminal intron of the tub gene (3,4), leads to progressive retinal and cochlear degeneration and to maturityonset obesity associated with insulin resistance (5–7). The association of retinitis pigmentosa (RP) with mutations in human TULP1 has been reported both in a large pedigree segregating for RP14 (8) as well as in sporadic cases of RP (9,10). Targeted null mutants for the Tulp1 gene also show early-onset retinal degeneration (11,12). Taken together, the mutations/phenotypes observed within this gene family suggest that they are important for the normal function of cell types such as sensory neurons. The N-terminus of the four genes within this family is divergent whereas the C-terminus is highly conserved, suggesting that the C-terminus confers a common function to the family members. Recently, potential functions of the tub gene as an intracellular signaling molecule (13), a transcription factor (14), +To

a molecule associated with intracellular transport (11) or with ribosomal RNA synthesis (15) have been postulated. However, the molecular function of tub and the pathway in which it acts are still not well understood. Although not mutually exclusive, this wide array of suggested functions indicates that we are still far from understanding the function of this gene family. TULP3 is the last member of the tubby gene family to be identified, and is phylogenetically most closely related to the TUB protein (2). Based on sequence similarity and expression patterns, we hypothesized that the Tulp3 gene would have a similar cellular role to that of tub and, therefore, would lead to similar phenotypes when functionally deleted. In the current study, we describe that a disruption of Tulp3 leads to neural tube defects (NTDs) and embryonic lethality, demonstrating that Tulp3 is essential for normal embryonic development. RESULTS Targeted disruption of the Tulp3 gene The mouse Tulp3 gene was disrupted by homologous recombination in embryonic stem (ES) cells. Since mutations in the highly conserved C-terminal region of tub and Tulp1 produced truncated proteins which are loss-of-function mutations (3,4,8), it was expected that the disruption of the C-terminal region of the Tulp3 gene would also cause a loss of Tulp3 function. Therefore, a targeting vector was designed to delete the exons encoding the entire C-terminal and part of the N-terminal region of the TULP3 protein. Homologous recombination results in the deletion of a portion of exon 5 and exons 6–9 and in the insertion of the enhanced green fluorescent protein (EGFP) gene in-frame to Tulp3 exon 5 (Fig. 1A). Correctly targeted clones were identified by PCR and Southern analysis (Fig. 1B and C). Phenotypes in Tulp3–/– mice Initial genotypic analysis of adult F2 intercross progeny showed 27 wild-type and 41 heterozygotes for the targeted allele with no homozygotes detected, suggesting lethality in embryos homozygous for the Tulp3 null mutation. In order to further characterize the mutant phenotype, embryos were examined through embryonic developmental stages. No apparent phenotypic differences were observed among the embryos examined at embryonic day (E) 8.5 (data not shown). The earliest morphological abnormalities in Tulp3–/– embryos were evident at E9.5. The NTD in Tulp3–/– embryos at E9.5 was

whom correspondence should be addressed. Tel: +1 207 288 6384; Fax: +1 207 288 6079; Email: [email protected]

1326 Human Molecular Genetics, 2001, Vol. 10, No. 12

Figure 1. Targeted disruption of the mouse Tulp3 gene. (A) A schematic representation of the genomic structure of the Tulp3 gene is shown at the top. Exons are indicated by black boxes. The targeting vector (middle) contains a portion of exon 5 and ∼4.5 kb 5′-flanking region as well as 2.5 kb 3′ region containing exons 10 and 11. The diphtheria toxin (DT) gene was placed at the 5′ end of the targeting vector for negative selection. Homologous recombination with the targeting vector is predicted to delete parts of exon 5 as well as exons 6–9 (shown at the bottom). B, BglI; E, EcoRI; P, PstI. Primers and probe for the analysis of the targeted allele are indicated. (B) Identification of wild-type (+/+), heterozygous (+/–) and homozygous (–/–) E9.5 embryos by PCR analysis. Primer pairs and the size of the resultant PCR products are shown on the left. (C) Southern analysis of BglI-digested yolk-sac DNA. The Southern blots were hybridized with the indicated probe. Genotypes of progeny are indicated at the top of each lane.

Figure 2. Neurulation defects in Tulp3–/– embryos. Lateral (A and B) and dorsal (C and D) views of wild-type (A and C) and Tulp3–/– (B and D) embryos at E9.5. The Tulp3–/– embryo exhibits an open and everted neural tube. (E and F) Transverse, hematoxylin-stained sections of E9.5 wild-type (E) and Tulp3–/– (F) embryos. Note that the hindbrain (hb) neuroepithelium is not fused and both the hindbrain and the rostral midbrain show abnormal morphology. Scale bar, 100 µm.

Human Molecular Genetics, 2001, Vol. 10, No. 12 1327

Figure 3. (A and B) Lateral view of wild-type (A) and Tulp3–/– (B) embryos isolated at E12.5. Hemorrhages are observed in Tulp3–/– embryos along the caudal neural tube (indicated by arrows). Note that the Tulp3–/– embryo at E12.5 (B) shows exencephaly (indicated by an arrowhead). (C) E11.5 Tulp3–/– embryo with brain hemorrhage (indicated by an arrow). (D) Hemorrhage into the telencephalic vesicle in an E10.5 Tulp3–/– embryo. The telencephalic vesicle contains nucleated red blood cells. (E–H) Para-sagittal sections of E12.5 wild-type (E and G) and Tulp3–/– (F and H). (F) The structure of the vertebrae is disorganized in the Tulp3–/– embryo (indicated by arrowheads). (H) The dorsal root ganglia are irregular in size and shape, and are sometimes not segmented (indicated by an arrowhead). (I and J) SEM analysis of wild-type (I) and Tulp3–/– (J) embryos at E12.5. Facial development is abnormal in Tulp3–/– embryos. The nasal prominences (np) are not completely merged with each other and to the maxillary prominences (mp). Clefting lines are indicated by arrowheads. Scale bars, 1 mm. (K) Spina bifida in an E13.5 Tulp3–/– embryo. Rostral and caudal limits of the lesion are indicated by an arrowhead and an arrow, respectively.

observed in the region from the midbrain to the hindbrain (Fig. 2). Histological analysis showed that both the anterior midbrain and the hindbrain were open, and the hindbrain exhibited a failure to elevate and formed a convex configuration instead of the V-shaped neural tube observed in wild-type embryos (Fig. 2E and F). Approximately 63% of Tulp3–/– embryos exhibited NTDs at E10.5 and at E12.5 77% showed exencephaly resulting from NTDs (Table 1). The remaining embryos did not exhibit NTDs. However, they all exhibited an abnormal brain morphology, similar to that observed in affected heterozygotes (Figure 7; data not shown). At E14.5, all Tulp3–/– embryos were observed to be pale with no circulating red blood cells. Sites of hemorrhage were observed in Tulp3–/– embryos from E10.5 to E13.5, and occurred mainly in the brain and along the caudal neural tube

Table1. Genotypes and phenotypes of litters from heterozygous matings Embryonic day (E)

Numbera

+/+b

+/–b

–/–b

E10.5f

105

26

55 (10)c

24 (15)d

E12.5f

38

9

20 (3)c

9 (7)d

E14.5f

36

13

23 (2)e

0 (11) e

Newborng

68

27

41

0

aTotal

number of mice alive as assessed by heart beat. number of mice for each genotype. cThe number of embryos affected as assessed by brain morphology. dThe number of embryos that showed a neural tube closure defect. The remaining showed craniofacial defects at E12.5. eThe number of embryos that were found dead without red blood cells. fF progeny were derived from matings of 8 F males with 14 F females, a 3 2 2 minimum of one affected Tulp3+/– embryo from each F2 male was observed. gF progeny. 2 bThe


embryos showed spina bifida (6 out of 16 Tulp3–/– embryos at E12.5 and E13.5). The lesion was observed in the region referred to as the posterior neuropore (16) or caudal zone D (17) (Fig. 3K). Expression of the Tulp3 gene during embryonic development To further understand the role of Tulp3 in embryogenesis, we examined its expression during development. The targeting vector included a GFP reporter under the control of the endogenous promoter of the Tulp3 gene. Therefore, normal expression of the Tulp3 gene could be assessed by examining the GFP signal using fluorescent optics in Tulp3+/– embryos. The Tulp3/GFP fusion was ubiquitously expressed at E10.5 (Fig. 4A). Transverse sections of Tulp3+/– embryos showed expression of the Tulp3 gene in the neural tube and the mesenchyme (Fig. 4B and D). Ubiquitous expression of the Tulp3 gene was detected from E8.5 to E14.5 (data not shown). The ubiquitous expression pattern of the Tulp3 gene at E10.5 was confirmed by in situ hybridization (Fig. 4E). Increased cell death in the Tulp3–/– embryo

Figure 4. Expression of the Tulp3 gene assessed by GFP fusion protein expression and by in situ hybridization. (A) Whole-mount GFP expression in a Tulp3+/– E10.5 embryo. (C) Control wild-type embryos do not produce a GFP signal. (B) GFP signals are detected in transverse sections of a Tulp3+/– embryo by confocal microscopy in the neuroepithelium (ne) and mesenchyme in the head. (D) A higher magnification view shows GFP signals in each cell. (E) Expression of Tulp3 in a wild-type E10.5 embryo assessed by in situ hybridization. Tulp3 signal is detected throughout the embryo.

(Fig. 3A–D). The organs such as the heart, liver and kidney were histologically normal, as confirmed by examination of transverse and para-sagittal serial sections. In addition, vascular formation examined at E10.5 using the marker PECAM-1 in whole-mount immunohistochemistry was normal in Tulp3–/– embryos (data not shown). Histological analyses of E12.5 and E13.5 embryos showed malformation of the vertebrae (Fig. 3E and F) and the dorsal root ganglia (DRG) (Fig. 3G and H). Expression patterns of somite lineage markers such as Uncx4.1 and myogenin at E9.5–E10.5 showed no difference between wild-type and Tulp3–/– embryos, suggesting that the abnormalities in the vertebrae and DRG may not be due to a somite patterning defect (data not shown). All Tulp3–/– embryos showed abnormal facial development which became apparent at E11.5–E12.5 (Fig. 3I and J). 37.5% of Tulp3–/–

NTDs can be due to patterning abnormalities or abnormalities in cell proliferation and/or cell death (16,17). Since Tulp3 is ubiquitously expressed, we predicted that the defects caused by Tulp3 deficiency might be due to a defect in basic cellular function rather than abnormalities in specific patterning. Therefore, wild-type and Tulp3–/– E9.5 embryos were assayed to determine the relative degree of cell proliferation (Fig. 5E and F) and cell death (Fig. 5A–D). No significant differences between wild-type and Tulp3 –/– embryos for BrdU incorporation were observed (Figs 5E and F and 6A), suggesting no difference in cell proliferation. Subjecting adjacent sections to terminal dUTP nick-end labeling (TUNEL) analysis revealed an excess of cell death in E8.5 and E9.5 Tulp3–/– embryos (Figs 5 and 6B). In E9.5 Tulp3–/– embryos when compared with wild-type embryos matched by somite numbers (24–26 somites), sagittal sections showed increased apoptosis in the area of the hindbrain and the caudal neural tube (Fig. 5G–L). Transverse sections showed that the excessive cell death in the hindbrain was mainly observed in the ventral half of the neuroepithelium of the hindbrain (Fig. 5D). Quantitation of TUNEL-positive cells revealed a 7-fold increase in the number of these cells in the ventral neuroepithelium of Tulp3–/– embryos (Fig. 6B). Whole-mount TUNEL assays demonstrated that excessive cell death in the hindbrain region was also observed in E8.5 Tulp3–/– embryos (nine somites) (Fig. 5M and N). We did not observe a difference in the degree of cell death in the head mesenchyme. Phenotypes of Tulp3 heterozygous embryos At E10.5, 18% of heterozygous embryos were found to be abnormal (Table 1). The phenotype of affected Tulp3+/– embryos was less severe than that observed in Tulp3–/– embryos. A few Tulp3 heterozygotes displayed an open neural tube (data not shown). The majority of affected Tulp3+/– embryos showed abnormal brain morphogenesis (Fig. 7D and F). The fore-, mid- and hindbrains were underdeveloped and disorganized (Fig. 7B, D and F). As observed in Tulp3–/–


Figure 5. Apoptosis and cell proliferation in Tulp3–/– embryos. Wild-type and Tulp3–/– embryos were isolated and assayed for apoptosis by TUNEL assay (A–D and G–N) and for proliferation by BrdU incorporation (E and F). (A and C) Counterstaining by DAPI in wild-type (A) and Tulp3–/– (C) E9.5 embryos. (B and D) TUNEL-positive cells in wild-type (B) and Tulp3–/– (D) embryos. Apoptotic cells that have incorporated dioxigenin-dNTP were detected by a fluorescent-labeled antibody. (E and F) Transverse sections of BrdU-labeled wild-type (E) and Tulp3–/– (F) E9.5 embryos. Brown nuclei are BrdU-positive cells and indicate active cell proliferation. Sections were counterstained with hematoxylin to detect all nuclei. (G –L) Representative para-sagittal sections of wild-type (G, I and K) and Tulp3–/– (H, J and L) E9.5 embryos (24– 26 somites) counterstained by DAPI. TUNEL signals are light green. Note TUNEL-positive cells are detected from the anterior part of the hindbrain through the caudal neural tube. (M and N) Whole-mount TUNEL assay of wild-type (M) and Tulp3–/– (N) E8.5 embryos (9 somites). An excessive number of TUNEL-positive cells were observed in the hindbrain region in Tulp3–/– embryos. hb, hindbrain; ne, neuroepithelium; ot, otocyst; ht, heart. Scale bar, 100 µm.

embryos, affected Tulp3+/– embryos also showed an increase in the number of apoptotic cells in the hindbrain (Fig. 7G–J). Because we did not observe abnormalities in live-born Tulp3+/– mice and some Tulp3+/– embryos showed sites of hemorrhage at E12.5 and blood loss at E14.5 (data not shown), it is likely that abnormal Tulp3+/– embryos die during embryogenesis. Marker analysis In order to determine the specific cell populations affected in the brain of Tulp3–/– embryos, we examined the expression of markers associated with neuronal differentiation. Class III β-tubulin, a marker of terminal neuronal differentiation

(18,19), and nestin, an intermediate filament protein expressed at high levels in proliferating neuroepithelial stem cells and down-regulated during subsequent differentiation of these precursors into the neuronal or glial lineages (19–21), were used. In E9.5 wild-type embryos, expression of βIII-tubulin was observed throughout the epithelium in the hindbrain (Fig. 8Aa). E9.5 Tulp3 –/– embryos showed a decrease in βIII-tubulin positive cells in the hindbrain (Fig. 8Ae). At E10.5, the difference in the number of βIII-tubulin cells between wild-type and Tulp3–/– embryos was readily apparent (Fig. 8Ac and g). Interestingly, the levels of expression of nestin were comparable in wild-type and Tulp3–/– embryos


(Fig. 8Ba and b). These results suggest that the differentiated cells may undergo programmed cell death in Tulp3–/– embryos. In addition, AP2α was used as a marker for neural crest cells which arise from the neuroepithelium (22). No differences between Tulp3–/– and wild-type embryos were observed with AP2α antibodies at E9.5 (Fig. 8Bc and d). DISCUSSION In this study, we have shown that the Tulp3 gene has an essential role during mammalian development. We provide evidence that Tulp3 is required for cranial development, and specifically for closure of the neural tube. We also demonstrate that heterozygosity for the Tulp3 targeted mutant allele causes a milder defect in brain development in a subset of Tulp3+/– heterozygous mice. Neural tube defect and cell death in Tulp3–/– mice Although more than 50 genes whose mutations lead to NTDs are known in mice (16,17), the relationship among these genes is still unclear. At the tissue level, abnormalities in patterning within neuroepithelial cells are known to be associated with NTDs (16). It has been suggested that NTDs can be caused by aberrant cytoskeletal architecture, cell polarity or cell adhesion (17). In addition, NTDs can also result from an imbalance between cell proliferation and cell death in the neural tube, the paraxial mesoderm, and/or the head mesenchyme (16). Apoptosis is implicated as a critical mechanism in both physiological and pathological states of neural tube closure and of early brain development (16). The increased cell death in neuroepithelial cells of the hindbrain observed from E8.5 to E9.5 in Tulp3–/– embryos might be the direct cause of the failure of neural tube closure. During development, the hindbrain is transiently subdivided into seven or eight compartments called rhombomeres which play pivotal roles in determining subsequent cell fate. For example, the neuroepithelium of the hindbrain gives rise to neural crest cells that populate the branchial arches and cranial mesenchyme (23). The segmentation of the hindbrain is determined by the expression patterns of transcription factors, such as Hox genes, Krox-20 and kreisler, as well as tyrosine kinase receptors such as members of the Eph family (reviewed in ref. 23) which are required for the specification and/or maintenance of specific rhombomeres. In addition, a proper sequence of apoptotic events in the hindbrain is necessary to specify rhombomeres and subsequent patterning of cranial neural crest cell migration (24,25). In normal development, rhombomeres r3 and r5 (which are deficient in the production of migratory neural crest cells) show elevated levels of apoptosis (24). In Hoxa1/Hoxb1 double mutants, misspecification of rhombomeres within the hindbrain leads to defects in craniofacial development through an over-induction of apoptosis (26). Excess cell death was observed through the entire hindbrain in Tulp3–/– embryos, which may alter cell populations within the hindbrain and/or the fate of cells arising from the hindbrain. The defects in craniofacial development observed in Tulp3–/– embryos may also be due, in part, to increased cell death in the neuroepithelium of the hindbrain. In this context, identifying cell populations that undergo cell death in Tulp3–/– embryos would contribute to our understanding of

Figure 6. Cell proliferation (A) and cell death (B) in E9.5 wild-type and Tulp3 –/– embryos (24–26 somites). (A) Mitotic index in the neuroepithelium and the mesenchyme from wild-type and Tulp3–/– embryos. The mitotic index was calculated by dividing the number of BrdU-positive nuclei by the total number of nuclei in each tissue section. (B) Cell death index in the neuroepithelium (ventral and lateral) in the anterior hindbrain and the head mesenchyme from wild-type and Tulp3–/– embryos. The cell death index was calculated by dividing the number of TUNEL-positive nuclei by the total number of nuclei in each tissue section. Data were determined by counting transverse sections from three to four different embryos of each genotype. The mean ± SE is represented by a bar and shown for each genotype and tissue as indicated. *, P < 0.0001.

how specific cell types within the neuroepithelium affect brain morphogenesis and peripheral organ development e.g. facial development. By marker analyses, we have found a reduction in the number of βIII-tubulin positive cells in the hindbrain of Tulp3–/– embryos. Since the area of TUNEL-positive cells overlaps with the cells expressing βIII-tubulin, their reduction might be caused by the death of these cells. Alternatively, since βIII-tubulin is a late marker of neuronal differentiation (18,19), it is possible that precursor cells die by apoptosis prior to differentiating to βIII-tubulin positive cells. Our data demonstrate that there is selective cell death in specific cell types in this targeted mutant model. This idea is supported by the fact that apoptosis is observed specifically in the ventral region of the neuroepithelium in the hindbrain, and the differentiation and development of nestin and AP2α-positive cells that arise from neuroepithelium are intact.


Figure 7. Phenotypic characterization of Tulp3+/– E10.5 embryos. (A–D) Lateral view of wild-type (A and C) and Tulp3+/– (B and D) embryos. Note that the forebrain and midbrain regions of the Tulp3+/– embryos (B) are smaller than the wild-type embryo (A). The hindbrain in the Tulp3+/– embryo shows morphological abnormalities including collapsed roof (D) (indicated by an arrowhead). This morphological abnormality in the hindbrain of Tulp3+/– is confirmed by transverse section with hematoxylin stain (F) compared with a section of wild-type (E). (I and J) Excess TUNEL-positive cells in Tulp3+/– (J) embryos and in wild-type (I). (G and H) Counterstaining by DAPI in wild-type (G) and Tulp3+/– (H) embryos. f, forebrain; m, midbrain; h, hindbrain. Scale bar, 100 µm.

Heterozygous mice are affected by gene dosage or a dominant negative effect Surprisingly, embryonic lethality was observed among some Tulp3+/– F2 mice on a segregating background derived from C57BL/6J and 129 genomes. Although lethality caused by a heterozygous gene disruption is very rare in the literature, a few cases including knockout mice for the p300, VEGF and Tcof1 genes have been reported (27–30). In the p300 knockout, some heterozygous embryos die with NTDs in early development whereas the remaining ones are intact. It was concluded that there was a critical developmental time point that was sensitive to the gene dosage of p300 and the embryonic lethality of heterozygotes was dependent upon genetic background (26). Our observation of lethality in Tulp3+/– mice is very similar to that in p300+/– mice in terms of lethality at an early embryonic stage in a subpopulation of heterozygotes. The lethality of Tulp3+/– may also be due to a combination of gene dosage and genetic background effects. A second possible explanation for the lethality of Tulp3+/– embryos is a dominant-negative effect of the targeted allele. The targeted allele of the Tulp3 gene still encodes 179 amino acids of the N-terminus after the deletion of the C-terminal region, which is conserved to some degree across genes in the tubby family and is predicted to be a functional domain (2,14). It is possible that the product from the targeted allele may be able to bind Tulp3 binding protein(s), and hence antagonize the pathway through which the Tulp3 gene normally acts. If phenotypes in Tulp3+/– embryos are caused by a dominant-negative effect, it is also possible that defects in homozygous Tulp3–/– embryos may not be caused

solely by loss of function but by a dominant-negative effect. Further studies are necessary to determine which of these two possibilities is correct. Affected Tulp3+/– embryos also show excessive apoptosis of neuroepithelial cells in the hindbrain region. This suggests that both the Tulp3–/– and Tulp3+/– embryos might be affected through the same pathway. However, the existence of one copy of the wild-type allele in Tulp3+/– embryos may decrease the extent of apoptosis and/or delay the onset of apoptosis and make the phenotype milder. Summary Unlike mutations for other tubby gene family members, a mutation in the Tulp3 gene causes embryonic lethality. The questions still remaining are: (i) do the tubby gene family members have any common function? and (ii) what is the basic molecular function of the TULP3 protein? A common function among TULPs is likely, because the C-terminal half of TULPs is highly conserved (2). Recently, it has been shown that the C-terminal region of tub and Tulp1 has DNA binding capacity, suggesting that tub and Tulp1 can potentially act as transcription factors (14). Based on sequence similarity, the same may hold true for Tulp3. It is notable that cell death in neuronal cells such as the sensory neuron has been observed as a common feature of mutations of members of the Tulp gene family (11,12,31). The present study suggests that Tulp3 disruption may cause cell death in differentiating neuronal cells. Taken together, these studies indicate that the tubby gene


Figure 8. (A) Expression of βIII-tubulin in the hindbrain neuroepithelium of wild-type and Tulp3–/– E9.5 and E10.5 embryos. (a, c, e and g) Immunostaining for βIII-tubulin and (b, d, f and h) counterstained by DAPI. Cells expressing βIII-tubulin are labeled green by the FITC-secondary antibody in wild-type (a) and Tulp3–/– (e) E9.5 embryos, and wild-type (c) and Tulp3–/– (g) E10.5 embryos. Arrowheads in (b, d, f and h) point to the ventricular surface of the neuroepithelium. Scale bar, 100 µm in (a–h). (B) Expression of nestin and AP2α in the hindbrain region of wild-type and Tulp3–/– E9.5 embryos. In each pair, the Tulp3–/– sample is on the right. (a and b) Immunostaining for nestin in the neuroepithelium and (c and d) AP2α in the mesenchyme. Arrowheads in (A) and (B) point to the ventricular surface of the neuroepithelium. Scale bar, 100 µm.

family members are important in maintaining normal cellular integrity in the neuronal lineage. MATERIALS AND METHODS Targeting vector and generation of Tulp3–/– mutant mice A mouse Tulp3 genomic clone was isolated by screening a 129/Ola P1 library using a fragment spanning nucleotides 50–360 (GenBank accession no. AF045583) of mouse Tulp3 cDNA as a probe. The targeting vector was assembled using a 3′ 2.6 kb EcoRI and a 5′ 5.2 kb partially digested PstI fragment combined with the EGFP gene, which was derived from vector pEGFP-N3 (Clontech) into the positive-negative selection vector p58 (generated at The Jackson Laboratory). The genomic sequences used for the targeting vector flank exons 6–9. The targeting vector was linearized with PvuI and transfected into 129 derived R1 ES cells by electroporation. Following positive selection with G418, ES cell colonies were screened for homologous recombination by Southern blot analysis of DNA digested with BglI. A 3′ external probe was used to

detect a 8.2 kb wild-type band and a 6.3 kb mutant band. Of 400 ES clones analyzed, four were heterozygous for the targeted allele. Chimeras were generated from the ES cell lines. Chimeric mice were crossed to C57BL/6J mice to generate F1 mice. Genotyping of F1 mice showed that one out of four ES lines (6–4D line) was transmitted to the germ line. Therefore, all knockout alleles were derived from one ES cell line. Homozygous embryos used in this study were at F3 or F4 generations which were derived from intercrossing F2 or F3 heterozygous mice, respectively. All heterozygous embryos used in this study were at F3 generation. Knockout alleles were identified by Southern blot analysis as described above. Histological analysis Embryos were isolated and fixed in 4% paraformaldehyde overnight at 4°C, dehydrated in a graded ethanol series, cleared in xylene and embedded in paraffin. Sections were cut at 6 µm, mounted on slides pretreated with Vectabond (Vector Laboratories) and either stained with hematoxylin or processed for immunostaining, in situ hybridization, BrdU or TUNEL analysis.


Immunohistochemistry

REFERENCES

For immunohistochemistry, the sections were rehydrated and, after blocking with 2% goat serum in PBS, incubated with the primary antibody at 4°C overnight. Binding was detected using biotinylated IgG (1:200; Vector) followed by FITC-AvidinD or HRP-Avidin (1:200; Vector). A nuclear counterstain was performed with 4,6 diamidine 2-phenylindoldihydrochloride (DAPI) at a final concentration of 5 µg/ml or with hematoxylin. Images were collected on a Leica DMRXE fluorescent microscope equipped with a SPOT CCD camera using appropriate bandpass filters for each fluorochrome. The following antibodies were used: anti-Ap2α was obtained from Santa Cruz, anti-βIII-tubulin from Sigma and anti-nestin from the Developmental Studies Hybridoma Bank.

1. North, M.A., Naggert, J.K., Yan, Y., Noben-Trauth, K. and Nishina, P.M. (1997) Molecular characterization of TUB, TULP1, and TULP2, members of the novel tubby gene family and their possible relation to ocular diseases. Proc. Natl Acad. Sci. USA, 94, 3128–3133. 2. Nishina, P.M., North, M.A., Ikeda, A., Yan, Y. and Naggert, J.K. (1998) Molecular characterization of a novel tubby gene family member, TULP3, in mouse and human. Genomics, 54, 215–220. 3. Noben-Trauth, K., Naggert, J.K., North, M.A. and Nishina, P.M. (1996) A candidate gene for the mouse mutation tubby. Nature, 380, 534–538. 4. Kleyn, P.W., Fan, W., Kovats, S.G., Lee, J.J., Pulido, J.C., Wu, Y., Berkemeier, L.R., Misumi, D.J., Holmgren, L., Charlat, O. et al. (1996) Identification and characterization of the mouse obesity gene tubby: a member of a novel gene family. Cell, 85, 281–290. 5. Heckenlively, J.R., Chang, B., Erway, L.C., Peng, C., Hawes, N.L., Hageman, G.S. and Roderick, T.H. (1995) Mouse model for Usher syndrome: linkage mapping suggests homology to Usher type I reported at human chromosome 11p15. Proc. Natl Acad. Sci. USA, 92, 11100–11104. 6. Ohlemiller, K.K., Hughes, R.M., Lett, J.M., Ogilvie, J.M., Speck, J.D., Wright, J.S. and Faddis, B.T. (1997) Progression of cochlear and retinal degeneration in the tubby (rd5) mouse. Audiol. Neurootol, 2, 175–185. 7. Coleman, D.L. and Eicher, E.M. (1990) Fat (fat) and Tubby (tub): two autosomal recessive mutations causing obesity syndromes in the mouse. J. Hered., 81, 424–427. 8. Banerjee, P., Kleyn, P.W., Knowls, J.A., Lewis, C.A., Ross, B.M., Parano, E., Kovats, S.G., Lee, J.J., Penchaszadeh, G.K., Ott, J. et al. (1998) TULP1 mutation in two extended Dominican kindreds with autosomal recessive Retinitis pigmentosa. Nat. Genet., 18, 177–179. 9. Hagstrom, S., North, M., Nishina, P.M., Berson, E. and Dryja, T. (1998) Recessive mutations in the gene encoding the tubby-like protein, TULP1, in patients with retinitis pigmentosa. Nat. Genet., 18, 174–176. 10. Gu, S., Lennon, A., Li, Y., Lorenz, B., Fossarello, M., North, M., Gal, A. and Wright, A. (1998) Tubby-like protein-1 mutations in autosomal recessive retinitis pigmentosa. Lancet, 351, 1103–1104. 11. Hagstrom, S.A., Duyao, M., North, M.A. and Li, T. (1999) Retinal degeneration in tulp1–/– mice: vesicular accumulation in the interphotoreceptor matrix. Invest. Ophthalmol. Vis. Sci., 40, 2795–2802. 12. Ikeda, S., Shiva, N., Ikeda, A., Smith, R.S., Nusinowitz, S., Yan, G., Lin, T.R., Chu, S., Heckenlively, J.R., North, M.A. et al. (2000) Retinal degeneration but not obesity is observed in null mutants of the tubby-like protein 1 gene. Hum. Mol. Genet., 22, 155–163. 13. Kapeller, R., Moriarty, A., Strauss, A., Stubdal, H., Theriault, K., Siebert, E., Chickering, T., Morgenstern, J.P., Tartaglia, L.A. and Lillie, J. (1999) Tyrosine phosphorylation of tub and its association with Src homology 2 domain-containing proteins implicate tub in intracellular signaling by insulin. J. Biol. Chem., 274, 24980–24986. 14. Boggon, T.J., Shan, W.S., Santagata, S., Myers, S.C. and Shapiro, L. (1999) Implication of tubby proteins as transcription factors by structurebased functional analysis. Science, 286, 2119–2125. 15. He, W., Ikeda, S., Bronson, R.T., Yan, G., Nishina, P.M., North, M.A. and Naggert, J.K. (2000) GFP-tagged expression and immunohistochemical studies to determine the subcellular localization of the tubby gene family members. Mol. Brain Res., 81, 109–117. 16. Harris, M.J. and Juriloff, D.M. (1999) Mini-review: Toward understanding mechanisms of genetic neural tube defects in mice. Teratology, 60, 292–305. 17. Juriloff, D.M. and Harris, M.J. (2000) Mouse models for neural tube closure defects. Hum. Mol. Genet., 9, 993–1000. 18. Geisert, E.E.,Jr and Frankfurter, A. (1989) The neuronal response to injury as visualized by immunostaining of class III beta-tubulin in the rat. Neurosci. Lett., 102, 137–141. 19. Berk, M., Desai, S.Y., Heyman, H.C. and Colmenares, C. (1997) Mice lacking the ski proto-oncogene have defects in neurulation, craniofacial, patterning, and skeletal muscle development. Genes Dev., 11, 2029–2039. 20. Lendahl, U., Zimmerman, L.B. and McKay, R.D. (1990) CNS stem cells express a new class of intermediate filament protein. Cell, 23, 585–595. 21. Zimmerman, L., Parr, B., Lendahl, U., Cunningham, M., McKay, R., Gavin, B., Mann, J., Vassileva, G. and McMahon, A. (1994) Independent regulatory elements in the nestin gene direct transgene expression to neural stem cells or muscle precursors. Neuron, 12, 11–24. 22. Chen, Z.F. and Behringer, R.R. (1995) twist is required in head mesenchyme for cranial neural tube morphogenesis. Genes Dev., 9, 686–699.

In situ hybridization Sections were deparaffinized in xylene and rehydrated through a graded series of ethanol and PBS. Sections were then treated with proteinase K (10 µg/ml) for 7 min at room temperature, acetylated in a 0.1 M triethenolamine pH 8.0/0.25% acetic anhydride solution, dehydrated, and air dried. [α-33P]UTPlabeled sense and antisense riboprobes (105 cpm/µl) were generated from plasmids containing cDNA fragments of Tulp3 (nucleotides 44–623; GenBank accession no. AF045583). The probe was prepared from a region of the cDNA corresponding to the N-terminal half of the TULP3 protein, where the sequence homology between tubby family members is low (1,2). The specificity of this probe has previously been confirmed by northern hybridization (2). Hybridization was performed as described by Ikeda et al. (31). BrdU incorporation assay E9.5 embryos were labeled with BrdU by injecting BrdU solution (20 mg/ml) to pregnant mice at a dose of 500 µg/g body weight. The mice were killed 2 h later, and embryos were isolated and processed for immunohistochemistry as described above. The primary monoclonal antibody (Klon Bu20a) from DAKO was used. TUNEL assay Sections were prepared as described above. Slides were treated with proteinase K (10 µg/ml) in PBS for 10 min at room temperature and washed four times for 2 min in distilled water. The TUNEL assay was performed using a commercial kit (Apop Tag In situ Apoptosis Detection; Oncor). For a negative control, water was substituted for the terminal transferase (TdT) enzyme in the reaction buffer from the kit. ACKNOWLEDGEMENTS We are grateful to Drs Susan L. Ackerman, Timothy P. O’Brien and Barbara K. Knowles for careful review of the manuscript. This work was supported by grants from NIH DK46977, Foundation for Fighting Blindness and AXYS Pharmaceuticals Inc. Institutional shared services are supported by National Cancer Institute Support grant CA-34196. A.I. is a recipient of an American Heart Association postdoctoral fellowship.


23. Lumsden, A. and Krumlauf, R. (1996) Patterning the vertebrate neuraxis. Science, 274, 1109–1115. 24. Graham, A., Heyman, I. and Lumsden, A. (1993) Even-numbered rhombomeres control the apoptotic elimination of neural crest cells from odd-numbered rhombomeres in the chick hindbrain. Development, 119, 233–245. 25. Graham, A., Francis-West, P., Brickell, P. and Lumsden, A. (1994) The signalling molecule BMP4 mediates apoptosis in the rhombencephalic neural crest. Nature, 372, 684–686. 26. Rossel, M. and Capecchi, M.R. (1999) Mice mutant for both Hoxa1 and Hoxb1 show extensive remodeling of the hindbrain and defects in craniofacial development. Development, 126, 5027–5040. 27. Yao, T.P., Oh, S.P., Fuchs, M., Zhou, N.D., Ch′ng, L.E., Newsome, D., Bronson, R.T., Li, E., Livingston, D.M. and Eckner, R. (1998) Gene dosage-dependent embryonic development and proliferation defects in mice lacking the transcriptional integrator p300. Cell, 93, 361–372.

28. Carmeliet, P., Ferreira, V., Breier, G., Pollefeyt, S., Kieckens, L., Gertsenstein, M., Fahrig, M., Vandenhoeck, A., Harpal, K., Eberhardt, C. et al. (1996) Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. Nature, 380, 435–439. 29. Ferrara, N., Carver-Moore, K., Chen, H., Dowd, M., Lu, L., O’Shea, K.S., Powell-Braxton, L., Hillan, K.J. and Moore, M.W. (1996) Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature, 380, 439–442. 30. Dixon, J., Brakebusch, C., Fassler, R. and Dixon, M.J. (2000) Increased levels of apoptosis in the prefusion neural folds underlie the craniofacial disorder, Treacher–Collins syndrome. Hum. Mol. Genet., 10, 1473–1480. 31. Ikeda, S., He, W., Ikeda, A., Naggert, J.K., North, M.A. and Nishina, P.M. (1999) Cell specific expression of tubby gene family members in the retina. Invest. Ophthalmol. Vis. Sci., 40, 2706–2712.