δEF1 is a DNA binding protein containing a homeodomain and two zinc finger clusters, and is regarded as a vertebrate homologue of zfh-1 (zinc finger homeodomain-containing factor-1) in Drosophila. In the developing embryo, δEF1 is expressed in the notochord, somites, limb, neural crest derivatives and a few restricted sites of the brain and spinal cord. To elucidate the regulatory function of δEF1 in mouse embryogenesis, we generated δEF1 null mutant (δEF1null(lacZ)) mice. The δEF1null(lacZ) homozygotes developed to term, but never survived postnatally. In addition to severe T cell deficiency of the thymus, the δEF1null(lacZ) homozygotes exhibited skeletal defects of various lineages. (1) Craniofacial abnormalities of neural crest origin: cleft palate, hyperplasia of Meckel’s cartilage, dysplasia of nasal septum and shortened mandible. (2) Limb defects: shortening and broadening of long bones, fusion of carpal/tarsal bone and fusion of joints. (3) Fusion of ribs. (4) Sternum defects: split and asymmetric ossification pattern of the sternebrae associated with irregular sternocostal junctions. (5) Hypoplasia of intervertebral discs. These results indicate that δEF1 has an essential role in regulating development of these skeletal structures. Since the skeletal defects were not observed in δEF1ΔC727 mice, δEF1 bears distinct regulatory activities which are dependent on different domains of the molecule.

Skeletons of vertebrates develop from three distinct lineages (Erlebacher et al., 1995). The neural crest gives rise to the branchial arch derivatives of the craniofacial skeletons. The sclerotome generates most of the axial skeleton including vertebrae and ribs. The lateral plate mesoderm contributes to the limb and sternum elements. Correct patterning and development of these skeletal elements is fundamental to the diverse morphology of vertebrate skeletons, and is a model system for understanding the pattern formation during animal development. Hox genes, the vertebrate counterpart of Drosophila HOM/C genes, have been considered to be master regulators of body plan in vertebrate development. Their essential role in development and patterning of skeletal elements has been demonstrated (Duboule, 1994). Recent studies employing the gene targeting method, however, have also shown equally important involvement of other transcription factor genes in skeletal development (Erlebacher et al., 1995). As presented in this report, δEF1 is one such new member participating in skeleton patterning.

δEF1 was originally identified as an enhancer binding factor of the chicken δ1-crystallin gene, but was found to be expressed in a variety of tissues (Funahashi et al., 1991, 1993). δEF1 contains two Krüppel type-C2H2 zinc finger clusters located close to N- and C-termini with a homeodomain in between (Funahashi et al., 1993). The homeodomain shows highest similarity to those in LIM proteins (Bürglin, 1994). δEF1 is regarded as a vertebrate homologue of the Drosophila zinc finger homeodomain factor, zfh-1 (Fortini et al., 1991; Lai et al., 1991), based on the similarity of the sequence and organization of zinc-finger clusters and homeodomain, and of exon-intron organization of the genes (Funahashi et al., 1993). Other vertebrate homologues have been reported, which show highly conserved amino acid sequences; nil-2-a (a partial sequence; Williams et al., 1991), AREB6 (Watanabe et al., 1993) and ZEB (Genetta et al., 1994) of human, MEB1 of mouse (Genetta and Kadesch, 1996) identical to mouse δEF1 (Sekido et al., 1996), BZP (Franklin et al., 1994) of hamster, and zfhep (Cabanillas and Darling, 1996) of rat.

The δEF1 homologues have been shown to have repressive activities on transcription (Franklin et al., 1994; Genetta et al., 1994; Kamachi and Kondoh, 1993; Postigo et al., 1997; Sekido et al., 1994, 1997; Watanabe et al., 1993; Williams et al., 1991). δEF1 preferentially binds to the CACCT sequence. The C- proximal zinc finger clusters are essential for DNA binding activity (Sekido et al., 1994), while the homeodomain by itself does not show an appreciable DNA binding activity (Ikeda and Kawakami, 1995; Sekido et al., 1997). One possible mechanism in which δEF1 functions as a repressor is to compete with various bHLH activators for binding to the E2 box (CACCTG) sequence which overlaps with the δEF1 DNA binding sequence (Sekido et al., 1994). Overexpression of δEF1 in the 10T1/2 cells actually inhibited the MyoD-induced muscle differentiation (Postigo and Dean, 1997; Sekido et al., 1994).

δEF1 is expressed in various anlages of developing tissues besides lens cells. In chick embryos, δEF1 is first detected during the postgastrulation period in the mesodermal tissues: initially in the notochord, followed by somites, heart, limb bud and other components. In addition, δEF1 is also expressed in the nervous system and neural crest derivatives such as cephalic mesenchyme (Funahashi et al., 1993). A similar expression pattern was observed in the mouse (this paper).

To investigate the function of δEF1 during mouse embryogenesis, we took advantage of gene targeting. We previously generated δEF1 mutant mice (δEF1ΔC727) lacking the C-terminal portion downstream of the 727th amino acid residue, which included the C-proximal zinc-finger clusters. δEF1ΔC727 mutant mice showed severe hypocellularity in the thymus, resulting from the depletion of c-kit+ early T cell precursors, while many other tissues were normal in spite of expression of δEF1 (Higashi et al., 1997).

To approach the entire function of δEF1, and to gain insight into the possible domain-specific regulatory functions, we generated δEF1 null mutant (δEF1null(lacZ)) mice which lacked almost the entire δEF1 coding sequence. As presented in this report, skeletal defects of craniofacial, limb, and vertebral column development were observed in these mice as well as the same thymus defect as found in the δEF1ΔC727 mutant. The results demonstrate the essential role of δEF1 in skeleton development, and suggest differential usage of its functional domains in different tissues.

Construction of targeting vector

Using mouse δEF1 cDNA (Sekido et al., 1996) as a probe, a mouse genomic DNA library of 129/Sv strain was screened to obtain the mouse δEF1 gene sequence. The DNA fragment containing exon 1 was used for construction of the targeting vector (Fig. 1).

Fig. 1.

Generation of δEF1null(lacZ) mutant mice. (A) Structure of the δEF1 gene and targeting vector. Open boxes represent exons. Exon 2 is not shown since its exact location has not been determined. Vertical bars above the line indicate EcoRI sites while those below the line NcoI sites. EH and BN are the 5′ and 3′ flanking sequence probes, respectively, used for identification of the homologous recombinants and for the subsequent genotyping. (B) Confirmation of homologous recombination and genotyping of mutant mice by Southern analysis. The blot of EcoRI-digested DNAs from the ES (R1) cells, the targeted ES clones #292 and #355, and the embryos of each genotype were hybridized with the EH probe. (C) Northern blot analysis of δEF1 expression in δEF1null(lacZ) mutant embryos at E13.5. 5 μg of whole embryo RNA was hybridized with 2.5 kb δEF1 cDNA as probe. The same membrane was rehybridized with G3PDH. Heterozygotes had half the amount of δEF1 RNA compared with wild-type, and there was no mRNA in the homozygotes.

Fig. 1.

Generation of δEF1null(lacZ) mutant mice. (A) Structure of the δEF1 gene and targeting vector. Open boxes represent exons. Exon 2 is not shown since its exact location has not been determined. Vertical bars above the line indicate EcoRI sites while those below the line NcoI sites. EH and BN are the 5′ and 3′ flanking sequence probes, respectively, used for identification of the homologous recombinants and for the subsequent genotyping. (B) Confirmation of homologous recombination and genotyping of mutant mice by Southern analysis. The blot of EcoRI-digested DNAs from the ES (R1) cells, the targeted ES clones #292 and #355, and the embryos of each genotype were hybridized with the EH probe. (C) Northern blot analysis of δEF1 expression in δEF1null(lacZ) mutant embryos at E13.5. 5 μg of whole embryo RNA was hybridized with 2.5 kb δEF1 cDNA as probe. The same membrane was rehybridized with G3PDH. Heterozygotes had half the amount of δEF1 RNA compared with wild-type, and there was no mRNA in the homozygotes.

A 2.5 kb genomic fragment spanning the region from upstream to the middle of the exon 1 of δEF1, the 3.8 kb BamHI fragment of pMoZtk (Ueno et al., 1987) coding for β-galactosidase, the XhoI- BamHI fragment of pSTneoB (Kato et al., 1987) containing the neomycin-resistance gene in reverse orientation, and a 7.8 kb genomic fragment of intron I were joined using appropriate linker sequences of pBluescrjpt II SK+. The combined fragment was inserted into the DT-A (Diphtheria Toxin A-fragment) cassette vector (Yagi et al., 1993), generating the targeting vector pTVδEF1null(lacZ)-DT. The lacZ fusion protein coded by this vector contains 15 amino acids of the exon 1 and 12 amino acids from the linker sequence followed by the β-galactosidase coding sequence. The resulting targeting vector contains a 2.5 (5′) and a 7.8 kb (3′) sequence homologous to the genomic sequence (Fig. 1A).

Gene targeting

ES cells (R1; Nagy et al., 1993) were electroporated with the targeting vector and selected for G418 resistance (200 μg/ml) as described by Sawai et al. (1991). The targeted clones were identified by Southern hybridization (Fig. 1B). Germ-line chimera mice were obtained by blastocyst injection of the targeted ES cells. Heterozygotes were back- crossed to C57BL/6 for five generations before analysis. Genotypes were determined by Southern hybridization or PCR (polymerase chain reaction) analysis of DNAs from ear, tail or yolk sac. PCR was done at 94°C for 30 seconds, 65°C for 1 minute and 72°C for 1 minute with 35 cycles, using: primer J, TAGGTGTTAGGAAGGTGATGTCG; primer I, AACCGTGCATCTGCCAGTTTGAG.

In situ hybridization

Whole-mount in situ hybridization was done according to Wilkinson (1992) using digoxigenin-labeled RNA probes. δEF1, 0.8 kb SalI- ApaI cDNA fragment; Hoxd-13, a 253 bp cDNA fragment provided by Dr D. Duboule.

Skeleton staining

Bone and cartilage were stained according to a standard protocol (Hogan et al., 1994), except for E12.5-15.5 embryos which were first fixed in Bouin’s fixative for 3 hours, and then washed in 70% ethanol with frequent change of the solution for 2 days or until the yellowish color disappeared.

Histology

Specimens were fixed in Bouin’s fixative, embedded in paraffin, sectioned at a thickness of 7 μm, and stained with hematoxylin and eosin (HE).

Expression pattern of mouse δEF1 during early embryogenesis

The expression pattern of mouse δEF1 was examined at various developmental stages by in situ hybridization. In embryos at E8.5, strong δEF1 expression was first detected in the headfold and moderate expression in the presomitic and lateral plate mesoderm (Fig. 2A).

Fig. 2.

δEF1 expression in early embryos. δEF1 expression in embryo was examined by in situ hybridization of the δEF1 probe (A-E,G,H,I-M) with whole embryos (A-G, I-K, N), with a transverse section at the trunk level (H) and with sections of forelimb (L,M). The Hoxd-13 expression was examined for comparison with the δEF1 expression pattern by in situ hybridization of the Hoxd-13 (F,N) probe. The δEF1 expression pattern was also examined by X-gal staining of δEF1null(lacZ) heterozygous embryos (O,P). Notable differences in the tone of staining between X-gal staining and in-situ hybridization were partly ascribable to the fact that the β-galactosidase had the δEF1 N-terminal 15 amino acids mostly localized in the nucleus (data not shown), while in situ hybridization signals were in the cytoplasm. cg, cranial ganglia; d, dorsal root ganglia; fl, forelimb bud; h, headfold; hl, hindlimb bud; lm, lateral myotome; mm, medial myotome; my, migrated myotome; n, neural tube; nc, neural crest cells; ps, presomitic mesoderm; tb, tail bud; v, ventricular zone of neural tube. A and P indicate the direction of anterior-posterior axis. Asterisks in B and G indicate non-specific signals produced by probes trapped in the brain ventricles and observed with sense probes (data not shown). Scale bars: (A) 160 μm, (B,E,O) 320 μm, (C,D,F,I) 250 μm, (G,N,P) 500 μm, (H) 250 μm, (J,K) 400 μm, (L,M) 320 μm.

Fig. 2.

δEF1 expression in early embryos. δEF1 expression in embryo was examined by in situ hybridization of the δEF1 probe (A-E,G,H,I-M) with whole embryos (A-G, I-K, N), with a transverse section at the trunk level (H) and with sections of forelimb (L,M). The Hoxd-13 expression was examined for comparison with the δEF1 expression pattern by in situ hybridization of the Hoxd-13 (F,N) probe. The δEF1 expression pattern was also examined by X-gal staining of δEF1null(lacZ) heterozygous embryos (O,P). Notable differences in the tone of staining between X-gal staining and in-situ hybridization were partly ascribable to the fact that the β-galactosidase had the δEF1 N-terminal 15 amino acids mostly localized in the nucleus (data not shown), while in situ hybridization signals were in the cytoplasm. cg, cranial ganglia; d, dorsal root ganglia; fl, forelimb bud; h, headfold; hl, hindlimb bud; lm, lateral myotome; mm, medial myotome; my, migrated myotome; n, neural tube; nc, neural crest cells; ps, presomitic mesoderm; tb, tail bud; v, ventricular zone of neural tube. A and P indicate the direction of anterior-posterior axis. Asterisks in B and G indicate non-specific signals produced by probes trapped in the brain ventricles and observed with sense probes (data not shown). Scale bars: (A) 160 μm, (B,E,O) 320 μm, (C,D,F,I) 250 μm, (G,N,P) 500 μm, (H) 250 μm, (J,K) 400 μm, (L,M) 320 μm.

At E9.5, δEF1 was expressed in the derivatives of the cranial neural crest, such as branchial arches and cranial ganglia (Fig. 2B). In individual somites δEF1 expression was homogenous and at moderate levels. Strong δEF1 expression also occurred in the limb buds (Fig. 2B). In hindlimbs of the E9.5 embryo, which was just at the beginning of limb outgrowth, δEF1 was expressed evenly in the entire limb bud (Fig. 2C), while in the forelimb of the same embryo, where the limb bud was already prominent, δEF1 expression was confined to the posterior portion (Fig. 2D). δEF1 expression included the region of the Hoxd-13 expression (Fig. 2F) or other genes which are known to be expressed posteriorly (e.g. Shh, Bmp2, Fgf4). Later at E10.5, δEF1 expression in the limb was restricted to the distal- posterior region (Fig. 2E).

At E11.5, δEF1 expression in the neural tube was clearly detected (Fig. 2E,G,H), being particularly prominent in the ventricular zone (Fig. 2H). Dorsal root ganglia derived from the neural crest also expressed δEF1 (Fig. 2H). In the somites of the trunk, δEF1 was confined to the myotome. Expression was particularly high in the lateral myotome (Fig. 2G,H). This myotomal δEF1 expression was maintained in those cells that migrated in the limb (Fig. 2G,I). In limb, δEF1 expression was lost from the condensation of mesenchyme which gives rise to cartilaginous skeleton (Fig. 2I), then expressed transiently in the interdigit mesenchyme at E12.0 (Fig. 2J), and finally in the perichondrium of the cartilage when the skeletal pattern was already formed (Fig. 2K,L). At this stage of limb development also, overlap of expression of δEF1 and Hoxd-13 was noted (Fig. 2K,N).

The results described above in general agreed with the immunohistological observations of chicken δEF1 (Funahashi et al., 1993). In contrast to the chicken, however, δEF1 expression was not detected in lens cells of the mouse embryos (data not shown)

Generation of δEF1 null mutant mice

A targeting vector was designed so that the STneoB (Kato et al., 1987) and lacZ gene, each carrying a poly(A) addition and termination signals, were inserted into exon 1 of the δEF1 gene (Fig 1A). The resulting mutated allele was expected to lack the δEF1 activity but encoded a fusion protein of the first 15 amino acids of δEF1 with a β-galactosidase. (Fig. 1A).

A total of 10 clones of homologous recombinants were identified among 324 G418 resistant clones by Southern blot analysis using 5′ and 3′ sequences as probes (Fig. 1B and data not shown). Four clones were used to generate chimeric mice by blastocyst injection. Three of them gave rise to male chimeras which transmitted the mutation to their offspring.

Heterozygous mutants were fertile and healthy without appreciable abnormalities in morphology and behavior. The X-gal staining pattern of δEF1null(lacZ) heterozygotes at E9.5 (Fig. 2O) and E11.5 (Fig. 2P) mimicked the in situ hybridization data. One large deviation was the δEF1 expression in the notochord which was clearly indicated by the X-gal staining of the δEF1null(lacZ) heterozygotes (Fig. 2O), but was barely detectable by in situ hybridization (Fig. 2B,H). There might be some specific mechanisms regulating completion and stability of δEF1 transcript in the notochord.

Northern analysis revealed that δEF1 transcripts in heterozygotes were reduced to half the level of wild-type, while they were undetectable in the homozygotes (Fig. 1C). Thus, the homozygote was regarded as null, and the allele was termed δEF1null(lacZ).

Morphology of δEF1null(lacZ) mutant mice

δEF1null(lacZ) homozygous mutant embryos developed to term, but died shortly after birth (data not shown). Upon Caesarean section at E18.5, the homozygote did not start breathing, and died of cyanosis. The cause(s) of the respiratory failure has not been clarified. Earlier in embryogenesis, the homozygotes could be distinguished from other genotypes by their morphology (Fig. 3): growth retardation which became evident from the stage around E15.0; short and dumpy limbs (E14.5-18.5, 100%=67/67); shortened distal maxilla and mandible; curled tails (E12.5-18.5, 89%=73/82); frequent edema at midgestation (E13.5-16.5 52%=34/65). Also observed, though less frequently, was exencephaly and internal bleeding in the nasal region (E13.5-18.5, 3%=3/80 and 5%=4/80, respectively). Failure of spinal cord closure was observed in cranial and caudal ends, 1% (1/80) and 4% (3/80), of the homozygotes, respectively, while no such abnormalities were observed in sibling heterozygotes or wild types.

Fig. 3.

External morphology of δEF1null(lacZ) homozygous mutant embryo at E15.5. Control (A) and mutant (B,C) embryos of the same litter were compared. Note that the mutant embryos were stooped and were shorter in the body. The embryo in B had short limbs and mandible (arrowheads). The phenotype of the embryo in C was more severe and included edema and a curled tail (additional arrowheads). Growth and development were apparently retarded in this embryo. Scale bar, 1.6 mm.

Fig. 3.

External morphology of δEF1null(lacZ) homozygous mutant embryo at E15.5. Control (A) and mutant (B,C) embryos of the same litter were compared. Note that the mutant embryos were stooped and were shorter in the body. The embryo in B had short limbs and mandible (arrowheads). The phenotype of the embryo in C was more severe and included edema and a curled tail (additional arrowheads). Growth and development were apparently retarded in this embryo. Scale bar, 1.6 mm.

A series of sections of homozygotes at E14.5 and E16.5 showed no gross abnormalities other than those in skeletal elements which are discussed below, and the infrequent neural tube closure defect same phenotype in thymus as seen in δEF1ΔC727 mutant mice (data not shown).

Skeletal analysis of the δEF1null(lacZ) mutants by staining with alcian blue for cartilage and with alizarin red S for ossified bones revealed abnormalities in various skeletal elements. The mutant phenotype, including these abnormalities of skeletal and exencephaly. Sections of E18.5 homozygotes, however, revealed small hypocellular thymi with no distinction of medulla and cortex, as observed previously in the δEF1ΔC727 mutant mice (Higashi et al., 1997). The total number of thymocytes in E18.5 embryos was reduced to 1/10 that of the control. FACS analysis showed significant reduction of c-kit+ cells in the CD4CD8 thymocyte cell population, indicating that δEF1null(lacZ) mutation caused exactly the elements, are tabulated and compared with those of other mutant mice (Table 1).

Table 1.

Summary of δEF1 expression sites and phenotype of δEF1null(lacZ) mutant mice

Summary of δEF1 expression sites and phenotype of δEF1null(lacZ) mutant mice
Summary of δEF1 expression sites and phenotype of δEF1null(lacZ) mutant mice

Craniofacial skeletons

When examined at E18.5, δEF1null(lacZ) mutants had a cleft in the secondary palate (Fig. 4A-D). Hypoplasia of the palate bones as well as deformation of basisphenoid and pterygoid bones were noted (Fig. 4A,B). Dysplasia of the nasal septum was also observed in the mutants (56%=5/9; Fig. 4A,B). Sections revealed formation of an ectopic cavity in the nasal septum (Fig. 4E,F).

Fig. 4.

Craniofacial defects in the δEF1null(lacZ) mutant. (A-D) δEF1null(lacZ) mutant mice had a cleft in the secondary palate. (A,B) Ventral view of the base of the skull of control (A) and mutant embryos at E18.5. (C,D) Frontal sections of the heads of control (C) and mutant (D) embryos at E18.5 were stained with HE. (E,F) Transverse sections of the head of control (E) and mutant (F) embryos at E18.5 stained with HE. Ectopic cavity structure (arrow) was formed in the nasal septum of the mutant. The number of serous glands (sg) was less in the mutant. (G,H) Skeletal preparation of control (G) and mutant (H) mandibles at E18.5. (I,J) Cartilage staining of control (G) and mutant (J) embryos at stage E14.5. a, angular process; bs, basisphenoid bone; c, coronoid process; co, condylar process; d, dental process; m, Meckel’s cartilage; ns, nasal septum; sg, serous gland; pl, palatal bone; ps, palatal shelf; pt, pterygoid bone. Scale bars: (A,B) 570 μm, (C,D) 320 μm, (E,F) 320 μm, (G,H) 640 μm, (I,J) 250 μm.

Fig. 4.

Craniofacial defects in the δEF1null(lacZ) mutant. (A-D) δEF1null(lacZ) mutant mice had a cleft in the secondary palate. (A,B) Ventral view of the base of the skull of control (A) and mutant embryos at E18.5. (C,D) Frontal sections of the heads of control (C) and mutant (D) embryos at E18.5 were stained with HE. (E,F) Transverse sections of the head of control (E) and mutant (F) embryos at E18.5 stained with HE. Ectopic cavity structure (arrow) was formed in the nasal septum of the mutant. The number of serous glands (sg) was less in the mutant. (G,H) Skeletal preparation of control (G) and mutant (H) mandibles at E18.5. (I,J) Cartilage staining of control (G) and mutant (J) embryos at stage E14.5. a, angular process; bs, basisphenoid bone; c, coronoid process; co, condylar process; d, dental process; m, Meckel’s cartilage; ns, nasal septum; sg, serous gland; pl, palatal bone; ps, palatal shelf; pt, pterygoid bone. Scale bars: (A,B) 570 μm, (C,D) 320 μm, (E,F) 320 μm, (G,H) 640 μm, (I,J) 250 μm.

The mandible and premaxilla were shortened in the mutants (Fig. 4G,H and data not shown). The coronoid, condylar, and angular processes in the mutant mandible were hypoplastic (Fig. 4G,H), while Meckel’s cartilage was hyperplastic (Fig. 4G,H), which was already noticeable at early chondrogenic stage, E14.5 (Fig. 4I,J).

Limb skeletons

Long bones in the stylopod and zeugopod were shorter in the mutants, and appeared to be slightly thicker than the control (Fig. 5A-D). The radius and tibia in the mutants were curved (Fig. 5A-D). The scapula was also shortened (Fig. 5A,B). In the autopod, growth of phalanges (P1, P2, P3) in fore- and hindlimbs was severely attenuated: endochondral ossification of P1 and ossification at the distal end of P3 did not occur at E18.5 (Fig. 5E-H). Fusion of P1 and P2 in digits III and IV of forelimb was detected in some cases (Fig. 5F).

Fig. 5.

Skeletal defects in the limbs of δEF1null(lacZ) mutant mice. (A-D) Forelimb (A,B) and hindlimb (C,D) of control (A,C) and mutant (B,D) embryos at E18.5. Arrowheads indicate the curved radius (r) and tibia (t) in the mutant. (E-H) Autopod of forelimb (E,F) and hindlimb (G,H) of control (E,G) and mutant (F,H) embryos at E18.5. Arrowheads in F indicate absence of ossification of distal end of P3 and P1. Arrowheads in H indicate a fusion of metatarsal (mt) and cuneiform 1 (c1) in the mutant. (I-L) Magnification of distal part of carpal/tarsal bones of forelimb (I,J) and hindlimb (K,L) of control (I,K) and mutant (J,L) embryos. Arrowheads in J and L indicate the malformed carpal and tarsal bones, respectively, in the mutant. See text for details. The navicular (n) and talus (t) were also fused in the mutant (L). (M-P) Autopod of forelimb (M,N) and hindlimb (O,P) of control (M,O) and mutant (N,P) embryos at E14.5. Malformation of carpal/tarsal bones was caused by mislocation of chondrogenic precursors in the mutant. Scale bars: (A-D) 800 μm, (E-H) 270 μm, (I-L) 160 μm, (M-P) 160 μm.

Fig. 5.

Skeletal defects in the limbs of δEF1null(lacZ) mutant mice. (A-D) Forelimb (A,B) and hindlimb (C,D) of control (A,C) and mutant (B,D) embryos at E18.5. Arrowheads indicate the curved radius (r) and tibia (t) in the mutant. (E-H) Autopod of forelimb (E,F) and hindlimb (G,H) of control (E,G) and mutant (F,H) embryos at E18.5. Arrowheads in F indicate absence of ossification of distal end of P3 and P1. Arrowheads in H indicate a fusion of metatarsal (mt) and cuneiform 1 (c1) in the mutant. (I-L) Magnification of distal part of carpal/tarsal bones of forelimb (I,J) and hindlimb (K,L) of control (I,K) and mutant (J,L) embryos. Arrowheads in J and L indicate the malformed carpal and tarsal bones, respectively, in the mutant. See text for details. The navicular (n) and talus (t) were also fused in the mutant (L). (M-P) Autopod of forelimb (M,N) and hindlimb (O,P) of control (M,O) and mutant (N,P) embryos at E14.5. Malformation of carpal/tarsal bones was caused by mislocation of chondrogenic precursors in the mutant. Scale bars: (A-D) 800 μm, (E-H) 270 μm, (I-L) 160 μm, (M-P) 160 μm.

In normal embryos, there are 5 distal carpal bones called d1, d2, d3, d4 and a central carpal bone (c). d1, d2 and d3 are located proximal to the metacarpi, I, II, and III, respectively. d4 is generated by fusion of d4a and d4b which are made proximal to metacarpal IV and V, respectively. c is produced by a split of d2. In most of the δEF1null(lacZ) mutant embryos, three of the carpal bones (d2, d3 and c) did not exist as separate structures but as a single large fused bone (stage E15.5-18.5, 94%=16/17; Fig. 5I,J). d4a and d4b, which normally fuse to form d4 at stage E15.0, were often separated in the mutants (at stage E15.5-18.5 56%=9/16; Fig. 5I,J).

There are also 5 distal tarsal bones, cuneiform 1 to 3, cuboideum and navicular. In all the δEF1null(lacZ) mutants examined, cuneiform 2 and 3 were replaced by a large bone (stage E15.5-18.5 100%=16/16; Fig. 5K,L). Fusion of navicular and talus, and that of metatarsal I and cuneiform 1 in digit I were also observed in all the mutants examined at E18.5 (100%=18/18; Fig. 5G,H,K,L).

The distal carpal/tarsal bones were already affected at the chondrogenic precursor stage of E14.5 embryos. In the mutant forelimbs, the central carpal bone precursor was reduced while d4 was hyperplastic. d2-c was fused with d3 (Fig. 5M,N). In the hindlimb, cuneiform 2 was hypoplastic and already fused with cuneiform 3. The navicular was also hypoplastic, and dislocated near the talus (Fig. 5O,P). The navicular became fused to the talus at later stages (data not shown). Comparison of the mutant carpal/tarsal bones of different stages indicated that the morphological alterations in the mutant bones were caused mainly by the fusion of primary cartilaginous precursors at an early stage, rather than by deletion of the precursors.

Joints

The elbow is a compound joint of three bone components, derived from radius, ulna and humerus. In the mutant elbow, the characteristic joint structure did not develop, but instead the humerus was fused to the ulna or radius (100%=18/18; Fig. 6B, D). Fusion of joint components was also observed in the hip joint where femur and os coxae were fused (100%=18/18, data not shown).

Fig. 6.

Joint defects of the δEF1null(lacZ) mutant. (A,B) Elbow joint of the control (A) and mutant (B) embryos at E18.5. The humerus (h) and ulna (u) were fused in the mutant (arrowhead). (C,D) Sections of the elbow of control (C) and mutant (D) embryos at E18.5. The radius (r) and humerus (h) were fused in this mutant (D). (E,F) The position of the patella (p) was shifted more medially in the mutant (F) as compared to the control (E). t, tibia; f, fibula. Scale bars: (A,B) 230 μm, (C,D) 320 μm, (E,F) 640 μm.

Fig. 6.

Joint defects of the δEF1null(lacZ) mutant. (A,B) Elbow joint of the control (A) and mutant (B) embryos at E18.5. The humerus (h) and ulna (u) were fused in the mutant (arrowhead). (C,D) Sections of the elbow of control (C) and mutant (D) embryos at E18.5. The radius (r) and humerus (h) were fused in this mutant (D). (E,F) The position of the patella (p) was shifted more medially in the mutant (F) as compared to the control (E). t, tibia; f, fibula. Scale bars: (A,B) 230 μm, (C,D) 320 μm, (E,F) 640 μm.

The knee joint was affected in the mutants in a different way: although femur and tibia were separated in normal fashion, the position of the patella was shifted more medially in the mutants (81%=13/16; Fig. 6E,F). The shoulder joints appeared morphologically normal (data not shown).

Rib and sternum elements

Skeletal analysis revealed anomalous processes in the distal region of ribs which sometimes resulted in fusion between the ribs in δEF1null(lacZ) at E18.5 (78%=14/18; Fig. 7A-D). This kind of rib fusion occurred between T5 to T8 levels without bilateral symmetry. Examination at the earlier stage of E15.0 indicated that irregular expansion in the distal portion of rib cartilage precursor was apparently the cause of the rib fusion (Fig. 7C,D).

Fig. 7.

Rib and sternum defects of the δEF1null(lacZ) mutant. (A,B) Ribs and sternum of the control (A) and mutant (B) embryos at E18.5. Arrows indicate anomalous fusion of ribs in the mutant. Arrowheads indicate the split and asymmetric ossification of sternebrae accompanied by irregular sternocostal junctions and absence of xiphoid process in the mutants. (C,D) Ribs of early chondrogenic stage at E15.0 in the control (C) and mutant (D). The distal portion of the ribs in the mutant (arrowhead) was thicker and shorter than in the control. Scale bars: (A,B) 1.38 mm, (C,D) 400 μm.

Fig. 7.

Rib and sternum defects of the δEF1null(lacZ) mutant. (A,B) Ribs and sternum of the control (A) and mutant (B) embryos at E18.5. Arrows indicate anomalous fusion of ribs in the mutant. Arrowheads indicate the split and asymmetric ossification of sternebrae accompanied by irregular sternocostal junctions and absence of xiphoid process in the mutants. (C,D) Ribs of early chondrogenic stage at E15.0 in the control (C) and mutant (D). The distal portion of the ribs in the mutant (arrowhead) was thicker and shorter than in the control. Scale bars: (A,B) 1.38 mm, (C,D) 400 μm.

The δEF1null(lacZ) mutants had unexpected defects in the sternum. In E18.5 mutants, the sternum was always shorter and sternabrae displayed split ossification centers (100%=15/15; Fig. 7A,B). Irregular sternocostal junctions were also observed in the mutant accompanied by an asymmetric ossification pattern of the sternebrae (67%=10/15; Fig. 7A,B). In addition, the xiphoid processes were almost lost (Fig. 7A,B).

Intervertebral disc

Severe hypoplasia of intervertebral disc was evident in the δEF1null(lacZ) mutants (Fig. 8A,B,D,E). An intervertebral disc is composed of two tissues, nucleus pulposus, which is derived from notochord, and annulus fibrosus, the cartilaginous cells derived from intervertebral sclerotomal cells (Fig. 8B,C). Histological analysis of the mutant intervertebral discs revealed that the cell number of nucleus pulposus was significantly reduced (Fig. 8C,F), while the surrounding cartilaginous cells of annulus fibrosus did not appear much affected (Fig. 8F).

Fig. 8.

Hypoplastic intervertebral disc of the δEF1null(lacZ) mutant. Lateral (A,D) and frontal (B,E) view of vertebral columns of the control (A,B) and mutant (D,E) embryos at E18.5. Note that the intervertebral disc (iv) was severely hypoplastic in the mutant (D,E arrowheads). (C,F) Sagittal sections of the control (C) and mutant (F) embryos at E18.5 were stained with HE. Nucleus pulposus (np) in the intervertebral disc was clearly hypoplastic in the mutant (arrows). Arrowheads indicate the annulus fibrosus (af) surrounding the nucleus pulposus. Scale bars: (A,D) 250 μm, (B,E) 250 μm, (C,F) 160 μm.

Fig. 8.

Hypoplastic intervertebral disc of the δEF1null(lacZ) mutant. Lateral (A,D) and frontal (B,E) view of vertebral columns of the control (A,B) and mutant (D,E) embryos at E18.5. Note that the intervertebral disc (iv) was severely hypoplastic in the mutant (D,E arrowheads). (C,F) Sagittal sections of the control (C) and mutant (F) embryos at E18.5 were stained with HE. Nucleus pulposus (np) in the intervertebral disc was clearly hypoplastic in the mutant (arrows). Arrowheads indicate the annulus fibrosus (af) surrounding the nucleus pulposus. Scale bars: (A,D) 250 μm, (B,E) 250 μm, (C,F) 160 μm.

Deviation from normal morphology of the disc had already been apparent in the mutant at E14.5 (data not shown). Expression of collagen type II (Cheah et al., 1991), Pax-1 (Wallin et al., 1994) and Ndr-1 (Shimono et. al., unpublished data) in the intervertebral discs was not affected in the mutants analyzed at E15.5 by in situ hybridization (data not shown).

Concordance between δEF1 expression and the mutant phenotype

We generated δEF1 null mutant mice to investigate the function of δEF1 in vivo. The δEF1null(lacZ) mutant mice exhibited skeletal defects in craniofacial bones, ribs, intervertebral disc, sternum and limb skeletons. These skeletal elements are derived from different primordia, the cranial neural crest, somite, notochord and lateral plate mesoderm; all these tissue primordia express δEF1 (Fig. 2A, B). Thus, the results indicate a good concordance between the tissues affected by the mutation and earlier expression of δEF1 in those tissues (Table 1).

The case of rib abnormality in δEF1null(lacZ) mutants, however, is not straightforward. It is intriguing that δEF1 is expressed in the entire early somite but not in the sclerotome, the direct precursor of the ribs. An analogous situation is found in the mice mutant for myogenic factors, myogenin, myf5 and MRF4, all having certain rib defects, although the genes are not appreciably expressed in the sclerotome (Braun et al., 1992; Hasty et al., 1993; Zhang et al., 1995). Expression of these myogenic factors and δEF1 in an early somite stage or in myotome of a later stage seems essential for proper development of the ribs. Considering the possible interaction of these myogenic factors with δEF1 as discussed below, a rib phenotype may be developed when there is an imbalance among the interacting factors in the early somite or the myotome.

Major deviation from the concordance between the δEF1 expression site and mutant phenotypic manifestation are noted in the muscular and neuronal cell lineages. In muscle cells, possible interaction of δEF1 with myogenic factors has been suggested, since δEF1 can counteract the function of bHLH proteins, including myoD, by competitive binding to common cis-element E2-box in transfected cultured cells (Postigo and Dean, 1997; Sekido et al., 1994). δEF1 is indeed highly expressed in tissues where myogenic bHLH proteins are expressed (Fig. 2; Funahashi et al., 1993); however, no obvious abnormality in the muscular system was detected in the mutants. Overall levels and distribution of mRNAs, as examined by northern blot at E13.5 and whole mount in situ hybridization at E10-11, respectively, of myoD, myogenin, muscle creatine kinase and desmin were not significantly affected in δEF1null(lacZ) mutant embryos (data not shown).

α4 integrin is also activated during muscle cell differentiation (Rosen et al., 1992). In addition, α4 integrin promoter is reported to be repressed in cultured cells by the δEF1 homologue, ZEB (Postigo et al., 1997). We reported recently that α4 integrin expression in immature thymocytes was significantly increased in the δEF1ΔC727 mutant mice (Higashi et al., 1997). We, therefore, examined the overall level of α4 integrin mRNA at E13.5 and E15.5, but no obvious change was detected by northern blot analysis (data not shown).

It is possible that some other factors which are expressed in myogenic lineage cells and have a function similar to δEF1 could compensate for the effects of deficiency of δEF1. The mouse Sna is such a candidate gene which could compensate for the loss of δEF1 function (see Discussion in Higashi et al., 1997). An analogous situation may exist in neural tissues where no morphological defect was found in spite of the high expression of δEF1.

Analogies between skeletal phenotypes of the δEF1null(lacZ) and Hox mutant mice

It is interesting to note that many of the skeletal defects observed in the δEF1null(lacZ) mutants recapitulated those in mice mutant for some of the Hox genes (Table 1): cleft palate in Hoxa-2 mutant (Fig. 4A-D; Gendron-Maguire et al., 1993; Rijli et al., 1993), irregularities in the sternocostal junctions in Hoxa-4, Hoxa-5 and Hoxd-3 mutants (Fig. 7A,B; Condie and Capecchi, 1993; Horan et al., 1994; Jeannotte et al., 1993) and fusion of ribs in Hoxc-9 mutants (Fig. 7A,B; Suemori et al., 1995). The shared mutant phenotypes between δEF1 and Hox genes were particularly remarkable in the limb. The radius of the δEF1null(lacZ) mutant was curved, showing a close resemblance to the Hoxa-11−/− mutant (Davis et al., 1995). A characteristic of the δEF1null(lacZ) mutant limb was fusion of the three (d2, d3 and c) carpal bones (Fig. 5J). Variations of the carpal bone phenotype similar to those of the δEF1null(lacZ) mutants were observed in single or compound mutants of some of the Hox genes: fusion of d1 and d2 in heterozygotes of spontaneous Hoxa-13 mutation, hypodactyly (Mortlock et al., 1996); fusion of c and d3 in Hoxd-11−/− mutants (Favier et al., 1996); fusion of d3 and d4, and indentation of d4 in compound heterozygotes, [Hoxd-11+/−; Hoxd-13+/−] and [Hoxd-12+/−; Hoxd-13+/−] (Davis and Capecchi, 1996); split of d4 in Hoxd- 13−/− mutant and compound heterozygote [Hoxa-13+/− ; Hoxd- 13+/−] (Dolle et al., 1993; Fromental-Ramain et al., 1996).

Major expression domains of δEF1 and 5′ proximal Hoxd genes are superimposed in the limbs (Fig. 2C-F,I-N): in early limb bud, δEF1 is expressed uniformly throughout the limb bud as is Hoxd-9 and 10 (Dolle et al., 1989). As the limb bud grows, δEF1 became localized posteriorly, and then confined to the distal-posterior region (Fig. 2D,E). Similarly, Hoxd-11, 12 and 13 are known to be expressed in the posterior region (Fig. 2F; Dolle et al., 1989). When the skeletal pattern is once formed (Fig. 2J, K), δEF1 expression is observed in the perichondrium of the cartilage as are the Hoxd-12 and 13 genes (Fig. 2N; Dolle et al., 1989, 1993). Thus, the spacio-temporal pattern of δEF1 expression during the limb development significantly overlaps that of 5′ proximal Hoxd genes.

The shared mutant phenotype and expression pattern between the δEF1 and Hox genes, particularly in the limb, led us to postulate possible models for their genetic interactions. Since expression of the 5′ proximal Hoxd and Hoxa genes in limb were not significantly affected in δEF1null(lacZ) mutants (data not shown), the Hox genes are probably not downstream of δEF1, but epistatic to the δEF1 gene in a stage of limb development. It would thus be interesting to examine whether δEF1 expression is affected in the Hox mutants. Another intriguing possibility is direct interaction at the protein level.

δEF1 contains a homeodomain (Funahashi et al., 1993), which itself does not show DNA binding activity (Ikeda and Kawakami, 1995; Sekido et al., unpublished data). It has been indicated that Hox proteins heterodimerize with other homeodomain-containing proteins and function in vivo (Mann and Chan, 1996). It is also to be noted that the δEF1ΔC727 mutants did not exhibit skeletal defects (Higashi et al., 1997). Considering the fact that the homeodomain is retained in the δEF1ΔC727 mutant protein, direct interaction of δEF1 with Hox proteins is an attracting model.

Other mouse mutants which show some similarities to δEF1null(lacZ) mice

Recently reported GDF5 (growth and differentiation factor 5) mutant mice also showed the skeletal phenotype observed in δEF1null(lacZ) mutant mice: the fusion of some carpal and tarsal bones; shortening of the long bones in fore- and hindlimb; and medial displacement of the patella in the knee joint (Storm et al., 1994; Storm and Kingsley, 1996). GDF5 and its related molecules, GDF6 and 7, belong to the TGF-β superfamily, and are expressed specifically in the regions which give rise to joints of the limb (Hatterley et al., 1995; Storm and Kingsley, 1996). Hence, these molecules are believed to be important in development of proper joint structure. δEF1 is also expressed in developing elbow and carpals (Fig. 2L), and δEF1null(lacZ) mutant mice actually showed defects of joint formation resulting in the fused elbow and carpal bone malformation. It would be interesting to see whether there is any functional interaction between GDFs and δEF1.

Mhox, a paired-like homeobox-containing gene, is widely expressed in the mesenchymal cells including the various skeletal precursors (Kuratani et al., 1994). Mhox mutant mice also showed defects of skeletogenesis in multiple lineages, some of which are shared with the δEF1null(lacZ) mutant (Martin et al., 1995), i.e., the morphology of the maxillary and mandibular processes, and radius and tibia.

Mo et al. (1997) recently reported that Gli mutant mice also showed multiple skeletal abnormalities including cleft palate, tooth defects, absence of vertebral body and intervertebral discs, shortened limb and sternum. Among these, large reduction of intervertebral discs and delayed ossification of digit bones observed in Gli2/ mutants are reminiscent of the affected morphology in these tissues of δEF1null(lacZ) mutant mice. The similarity among δEF1null(lacZ), Mhox and Gli mutants may suggest genetic interactions among these genes.

Domain-specific regulatory function of δEF1

Previously, we generated δEF1ΔC727 mutant mice (Higashi et al., 1997). Comparison of the phenotype of δEF1null(lacZ) and δEF1ΔC727 mutants revealed the unique function of the internal domain of δEF1 protein. Impairment of thymus development is shared by mutants of both alleles (Higashi et al., 1997 and data not shown), but other phenotypes described in this paper are unique to the δEF1null(lacZ) allele. These results clearly show that the C-proximal portion of δEF1 protein including a zinc finger cluster is essential for gene regulation leading to proper T cell development, while function of the remaining portion of δEF1 is required for normal skeleton development (Fig. 9). Of particular interest is the homeodomain in the middle of the δEF1 protein since genetic interaction between δEF1 and Hox genes is suggested. Other δEF1 alleles with a mutation in the homeodomain or the N-proximal zinc finger cluster are being introduced into the mouse. Phenotypes of these mutant mice will further distinguish them from one another, which will further clarify a specific functional domain of the δEF1 protein.

Fig. 9.

A model for domain-specific regulatory function of δEF1. Protein encoded by each allele of δEF1 (wild-type, δEF1ΔC727 and δEF1null(lacZ)) is δEF1null(lacZ) Thymus and skeletal elements schematically shown, and correspondence with the phenotype is indicated. N- and C-fin indicate the zinc finger clusters proximal to N- and C-termini, respectively. HD indicates the homeodomain sequence.

Fig. 9.

A model for domain-specific regulatory function of δEF1. Protein encoded by each allele of δEF1 (wild-type, δEF1ΔC727 and δEF1null(lacZ)) is δEF1null(lacZ) Thymus and skeletal elements schematically shown, and correspondence with the phenotype is indicated. N- and C-fin indicate the zinc finger clusters proximal to N- and C-termini, respectively. HD indicates the homeodomain sequence.

Evolutional conservation of the δEF1 function

δEF1 is thought to be a vertebrate homologue of Drosophila zfh-1 (Fortini et al., 1991; Funahashi et al., 1993) because of the similarity of zinc-finger clusters and homeodomain, and of their gene organization, although conservation of sequence other than the putative DNA binding domains is low. Not only is there similarity in the protein structure between δEF1 and zfh-1, but there is also similarity in their expression patterns. In the Drosophila embryo, expression of zfh-1 is first observed in the presumptive procephalic mesoderm, then in the mesodermal anlages and later in the CNS including motor neurons and in mesoderm-derived structures including adult muscle precursors (Lai et al., 1991). δEF1 expression is first detected in the headfold, then at an early stage in mesodermal derivatives (e.g., somite, notochord, limb bud mesenchyme), and at a later stage strongly in CNS, PNS and muscle cells (Fig. 2).

The phenotype of the zfh-1 loss-of-function mutation (Lai et al., 1993) shows similarity to that of δEF1null(lacZ) mutant mice: early histogenesis of primitive mesodermal tissues and the nervous system proceeds normally, but in later stages mispositioning and cell fate alteration of mesodermal tissues become evident. These observations are consistent with the idea that δEF1 is a structural and functional counterpart of Drosophila zfh-1.

We thank Dr A. Nagy for ES R1, and Dr D. Duboule for Hoxd sequences. This work was supported by grants from the Ministry of Education, Science and Culture of Japan to T. T., Y. H. and H. K., and from the Science and Technology Agency of Japan to Y. H. T. T. was the recipient of a fellowship from the Japan Society for the Promotion of Science for Japanese Junior Scientists.

Braun
,
T.
,
Rudnicki
,
M. A.
,
Arnold
,
H.-H.
and
Jaenisch
,
R.
(
1992
).
Targeted inactivation of the muscle regulatory gene Myf-5 results in abnormal rib development and perinatal death
.
Cell
71
,
369
382
.
Bürglin
,
T. R.
(
1994
).
A comprehensive classification of homeobox genes
. In
Guidebook to the Homeobox Genes
(ed.
D.
Duboule
), pp.
25
71
.
Oxford
:
Oxford University Press
Cabanillas
,
A. M.
and
Darling
,
D. S.
(
1996
).
Alternative splicing gives rise to two isoforms of Zfhep, a zinc finger/homeodomain protein that binds T3-response element
.
DNA Cell Biol
.
15
,
643
651
.
Cheah
,
K. S.
,
Lau
,
E. T.
,
Au
,
P. K.
and
Tam
,
P. P.
(
1991
).
Expression of the mouse α1(II) collagen gene is not restricted to cartilage during development
.
Development
111
,
945
953
.
Condie
,
B. G.
and
Capecchi
,
M. R.
(
1993
).
Mice homozygous for a targeted disruption of Hoxd-3 (Hox-4.1) exhibit anterior transformations of the first and second cervical vertebrae, the atlas and the axis
.
Development
119
,
579
595
.
Davis
,
A. P.
and
Capecchi
,
M. R.
(
1996
).
A mutational analysis of the 5′ HoxD genes: dissection of genetic interactions during limb development in the mouse
.
Development
122
,
1175
1185
.
Davis
,
A. P.
,
Witte
,
D. P.
,
Hsieh-Li
,
H. M.
,
Potter
,
S. S.
and
Capecchi
,
M. R.
(
1995
).
Absence of radius and ulna in mice lacking hoxa-11 and hoxd-11
.
Nature
375
,
791
795
.
Dolle
,
P.
,
Dierich
,
A.
,
LeMeur
,
M.
,
Schimmang
,
T.
,
Schuhbaur
,
B.
,
Chambon
,
P.
and
Duboule
,
D.
(
1993
).
Disruption of the Hoxd-13 gene induces localized heterochrony leading to mice with neotenic limbs
.
Cell
75
,
431
441
.
Dolle
,
P.
,
Izpisúa-Belmonte
,
J. C.
,
Falkenstein
,
H.
,
Renucci
,
A.
and
Duboule
,
D.
(
1989
).
Coordinate expression of the murine Hox-5 complex homeobox-containing genes during limb pattern formation
.
Nature
342
,
767
772
.
Duboule
,
D.
(
1994
).
Temporal coliniarity and the phylotypic progression: a basis for the stability of a vertebrate Bauplan and the evolution of morphologies through heterochrony. Development
120
, Supplement
135
142
.
Epstein
,
D. J.
,
Vekemans
,
M.
and
Gros
,
P.
(
1991
).
splotch (Sp2H), a mutation affecting development of the mouse neural tube, shows a deletion within the paired homeodomain of Pax3
.
Cell
67
,
767
774
.
Erlebacher
,
A.
,
Filvaroff
,
E. H.
,
Gitelman
,
S. E.
and
Derynck
,
R.
(
1995
).
Toward a molecular understanding of skeletal development
.
Cell
80
,
371
378
.
Favier
,
B.
,
Rijli
,
F. M.
,
Fromental-Ramain
,
C.
,
Fraulob
,
V.
,
Chambon
,
P.
and
Pascal
,
D.
(
1996
).
Functional cooperation between the non-paralogous genes Hoxa-10 and Hoxd-11 in the developing forelimb and axial skeleton
.
Development
122
,
449
460
.
Fortini
,
M. E.
,
Lai
,
Z.
and
Rubin
,
G. M.
(
1991
).
The Drosophila zfh-1 and zfh-2 genes encode novel proteins containing both zinc-finger and homeodomain motifs
.
Mech. Dev
.
34
,
113
122
.
Franklin
,
A. J.
,
Jelton
,
T. L.
,
Shelton
,
K. D.
and
Magnuson
,
M. A.
(
1994
).
BZP, a novel serum-responsive zinc finger protein that inhibits gene transcription
.
Mol. Cell. Biol
.
14
,
6773
6788
.
Fromental-Ramain
,
C.
,
Warot
,
X.
,
Messadecq
,
N.
,
LeMeur
,
M.
,
Pascal
,
D.
and
Chambon
,
P.
(
1996
).
Hoxa-13 and Hoxd-13 play a crucial role in the patterning of the limb autopod
.
Development
122
,
2997
3011
.
Funahashi
,
J.-I.
,
Kamachi
,
Y.
,
Goto
,
K.
and
Kondoh
,
H.
(
1991
).
Identification of nuclear factor δEF1 and its binding site essential for lens-specific activity of the δ1-crystallin enhancer
.
Nucl. Acids Res
.
19
,
3543
3547
.
Funahashi
,
J.-I.
,
Sekido
,
R.
,
Murai
,
K.
,
Kamachi
,
Y.
and
Kondoh
,
H.
(
1993
).
δ-crystallin enhancer binding protein δEF1 is a zinc finger homeodomain protein implicated in postgastrulation embryogenesis
.
Development
119
,
433
446
.
Gendron-Maguire
,
M.
,
Mallo
,
M.
,
Zhang
,
M.
and
Gridley
,
T.
(
1993
).
Hoxa-2 mutant mice exhibit homeotic transformation of skeletal elements derived from cranial neural crest
.
Cell
75
,
1317
1331
.
Genetta
,
T.
and
Kadesch
,
T.
(
1996
).
Cloning of a cDNA encoding a mouse transcriptional repressor displaying striking conservation across vertebrates
.
Gene
169
,
289
290
.
Genetta
,
T.
,
Ruezinsky
,
D.
and
Kadesch
,
T.
(
1994
).
Displacement of an E-box-binding repressor by basic helix-loop-helix proteins: Implication for B-cell specificity of the immunoglobulin heavy-chain enhancer
.
Mol. Cell. Biol
.
14
,
6153
6163
.
Hasty
,
P.
,
Bradley
,
A.
,
Morris
,
J. H.
,
Edmondson
,
D. G.
,
Venuti
,
J. M.
,
Olson
,
E. N.
and
Klein
,
W. H.
(
1993
).
Muscle deficiency and neonatal death in mice with a targeted mutation in the myogenin gene
.
Nature
364
,
501
506
.
Hatterley
,
G.
,
Hewick
,
R.
and
Rosen
,
V.
(
1995
).
In situ localization and in vitro activity of BMP-13
.
J. Bone Min. Res
.
10
,
S163
.
Higashi
,
Y.
,
Moribe
,
H.
,
Takagi
,
T.
,
Sekido
,
R.
,
Kawakami
,
K.
,
Kikutani
,
H.
and
Kondoh
,
H.
(
1997
).
Impairment of T cell development in δEF1 mutant mice
.
J. Exp. Med
.
185
,
1467
1480
.
Hogan
,
B.
,
Beddington
,
R.
,
Constantini
,
F.
and
Lacy
,
E.
(
1994
).
Staining embryos for cartilage and bone
. In
Manipulating the Mouse Embryo: a Laboratory Manual, pp.325-379
.
Cold Spring Harbor, New York
:
Cold Spring Harbor Laboratory Press
.
Horan
,
G. S.
,
Wu
,
K.
,
Wolgemuth
,
D. J.
and
Behringer
,
R. R.
(
1994
).
Homeotic transformation of cervical vertebrae in Hoxa-4 mutant mice
.
Proc. Natl. Acad. Sci. USA
91
,
12644
12648
.
Ikeda
,
K.
and
Kawakami
,
K.
(
1995
).
DNA binding through distinct domains of zinc-finger-homeodomain protein AREB6 has different effects on gene transcription
.
Eur. J. Biochem
.
233
,
73
82
.
Jeannotte
,
L.
,
Lemieux
,
M.
,
Charron
,
J.
,
Poirier
,
F.
and
Robertson
,
E. J.
(
1993
).
Specification of axial identity in the mouse: role of the Hoxa-5 (Hox1.3) gene
.
Genes Dev
.
7
,
2085
2096
.
Kamachi
,
Y.
and
Kondoh
,
H.
(
1993
).
Overlapping positive and negative regulatory elements determine lens-specific activity of the δ1-crystallin enhancer
.
Mol. Cell. Biol
.
13
,
5206
5215
.
Kato
,
K.
,
Takahashi
,
Y.
,
Hayashi
,
S.
and
Kondoh
,
H.
(
1987
).
Improved mammalian vectors for high expression of G418 resistance
.
Cell Struct. Func
.
12
,
575
580
.
Kingsley
,
D. M.
,
Bland
,
A. E.
,
Grubber
,
J. M.
,
Marker
,
P. C.
,
Russell
,
L. B.
,
Copeland
,
N. C.
and
Jenkins
,
N. A.
(
1992
).
The mouse short ear skeletal morphogeneis locus is associated with defects in a bone morphogenetic member of the TGFβ superfamily
.
Cell
71
,
399
410
.
Kuratani
,
S.
,
Martin
,
J. F.
,
Wawersik
,
S.
,
Lilly
,
B.
,
Eichele
,
G.
and
Olson
,
E. N.
(
1994
).
The expression pattern of the chick homeobox gene gMhox suggests a role in patterning of the limbs and face and in compartmentalization of somites
.
Dev. Biol
.
161
,
357
369
.
Lai
,
Z.
,
Fortini
,
M. E.
and
Rubin
,
G. M.
(
1991
).
The embryonic expression patterns of zfh-1 and zfh-2, two Drosophila genes encoding novel zinc-finger homeodomain proteins
.
Mech. Dev
.
34
,
123
134
.
Lai
,
Z.-C.
,
Rushton
,
E.
,
Bate
,
M.
and
Rubin
,
G. M.
(
1993
).
Loss of function of the Drosophila zfh-1 gene results in abnormal development of mesodermally derived tissues
.
Proc. Natl. Acad. Sci. USA
90
,
4122
4126
.
Mann
,
R. S.
and
Chan
,
S.-k.
(
1996
).
Extra specificity from extradenticle; the partnership between HOX and PBX/EXD homeodomain proteins
.
Trends Genet
.
12
,
258
262
.
Martin
,
J. F.
,
Bradley
,
A.
and
Olson
,
E. N.
(
1995
).
The paired-like homeo box gene Mhox is required for early events of skeletogenesis in multiple lineages
.
Genes Dev
.
9
,
1237
1249
.
Mo
,
R.
,
Freer
,
A. M.
,
Zinyk
,
D. L.
,
Crackower
,
M. A.
,
Michaud
,
J.
,
Heng
,
H. H.-Q.
,
Chik
,
X. W.
,
Shi
,
X.-M.
,
Tsui
,
L.-C.
,
Cheng
,
S. H.
,
Joyner
,
A. L.
and
Hui
,
C.-c.
(
1997
).
Specific and redundant function of Gli2 and Gli3 zinc finger genes in skeletal patterning and development
.
Development
124
,
113
123
.
Mortlock
,
D. P.
,
Post
,
L. C.
and
Innis
,
J. W.
(
1996
).
The molecular basis of hypodactyly (Hd): a deletion in Hoxa13 leads to arrest of digital arch formation
.
Nature Genet
.
13
,
284
289
.
Nagy
,
A.
,
Rossant
,
J.
,
Nagy
,
R.
,
Abramow-Newerly
,
W.
and
Roder
,
J. C.
(
1993
).
Derivation of completely cell culture-derived mice from early-passage embryonic stem cells
.
Proc. Natl. Acad. Sci. USA
90
,
8424
8428
.
Postigo
,
A. A.
and
Dean
,
D. C.
(
1997
).
ZEB, a vertebrate homologue of Drosophila Zfh-1, is a negative regulator of muscle differentiation
.
EMBO J
.
16
,
3935
3943
.
Postigo
,
A. A.
,
Sheppard
,
A. M.
,
Mucenski
,
M. L.
and
Dean
,
D. C.
(
1997
).
C-Myb and Ets proteins synergize to overcome transcriptional repression by ZEB
.
EMBO J
.
16
,
3924
3934
.
Rijli
,
F. M.
,
Mark
,
M.
,
Lakkaraju
,
S.
,
Dierich
,
A.
,
Pascal
,
D.
and
Chambon
,
P.
(
1993
).
A homeotic transformation is generated in the rostral branchial region of the head by disruption of Hoxa-2, which acts as a selector gene
.
Cell
75
,
1333
1349
.
Rosen
,
G. D.
,
Sanes
,
J. R.
,
LaChance
,
R.
,
Cunningham
,
J. M.
,
Roman
,
J.
and
Dean
,
D. C.
(
1992
).
Roles for the integrin VLA-4 and its counter receptor VCAM-1 in myogenesis
.
Cell
69
,
1107
19
.
Satokata
,
I.
and
Maas
,
R.
(
1994
).
Msx1 deficient mice exhibit cleft palate and abnormalities of craniofacial and tooth development
.
Nature Genet
6
,
349
355
.
Sawai
,
S.
,
Shimono
,
A.
,
Hanaoka
,
K.
and
Kondoh
,
H.
(
1991
).
Embryonic lethality resulting from disruption of both N-myc alleles in mouse zygotes
.
The New Biologist
3
,
861
869
.
Sekido
,
R.
,
Murai
,
K.
,
Funahashi
,
J.
,
Kamachi
,
Y.
,
Fujisawa-Sehara
,
A.
,
Nabeshima
,
Y.
and
Kondoh
,
H.
(
1994
).
The δ-crystallin enhancer-binding protein δEF1 is a repressor of E2-box-mediated gene activation
.
Mol. Cell. Biol
.
14
,
5692
5700
.
Sekido
,
R.
,
Takagi
,
T.
,
Okanami
,
M.
,
Moribe
,
H.
,
Yamamura
,
M.
,
Higashi
,
Y.
and
Kondoh
,
H.
(
1996
).
Organization of the gene encoding transcriptional repressor δEF1 and cross-species conservation of its domains
.
Gene
173
,
227
232
.
Sekido
,
R.
,
Murai
,
K.
,
Kamachi
,
Y.
and
Kondoh
,
H.
(
1997
).
Two mechanisms in the action of repressor δEF1: binding site competition with an activator and active repression
.
Genes to Cells
2
(in press).
Storm
,
E. E.
,
Huyuh
,
T. Y.
,
Copeland
,
N. G.
,
Jenkins
,
N. A.
,
Kingsley
,
D. M.
and
Lee
,
S.-L.
(
1994
).
Limb alterations in brachypodism mice due to mutations in a new menber of the TGF-beta super family
.
Nature
368
,
639
643
.
Storm
,
E. E.
and
Kingsley
,
D. M.
(
1996
).
Joint patterning defects caused by single and double mutations in members of the bone morphogenetic protein (BMP) family
.
Development
122
,
3969
3979
.
Suemori
,
H.
,
Takahashi
,
N.
and
Noguchi
,
S.
(
1995
).
Hoxc-9 mutant mice show anterior transformation of the vertebrae and malformation of the sternum and ribs. Mech. Dev
.
51
,
265
273
.
Ueno
,
K.
,
Hiramoto
,
Y.
,
Hayashi
,
S.
and
Kondoh
,
H.
(
1987
).
Introduction and expression of recombinant β-galactosidase genes in cleavage stage mouse embryos
.
Dev. Growth Differ
.
30
,
61
73
.
Wallin
,
J.
,
Wilting
,
J.
,
Koseki
,
H.
,
Fritsch
,
R.
,
Christ
,
B.
and
Balling
,
R.
(
1994
).
The role of Pax-1 in axial skeleton development
.
EMBO J
.
120
,
1109
1121
.
Watanabe
,
Y.
,
Kawakami
,
K.
,
Hirayama
,
Y.
and
Nagano
,
K.
(
1993
).
Transcription factors positively and negatively regulating the Na, K-ATPase α1 subunit gene
.
J. Biochem
.
114
,
849
855
.
Wilkinson
,
D. G.
(
1992
).
Whole mount in situ hybridization of vertebrate embryos
. In
In Situ Hybridization: A Practical Approach
, (ed.
D. G.
Wilkinson
), pp.
75
84
.
Oxford
:
IRL Press
.
Williams
,
T. M.
,
Moolten
,
D.
,
Burlein
,
J.
,
Romano
,
J.
,
Bhaerman
,
R.
,
Godillot
,
A.
,
Mellon
,
M.
,
Rauscher III
,
F. J.
and
Kant
,
J. A.
(
1991
).
Identification of a zinc finger protein that inhibits IL-2 gene expression
.
Science
254
,
1791
1794
.
Yagi
,
T.
,
Nada
,
S.
,
Watanabe
,
N.
,
Tamemoto
,
H.
,
Kohmura
,
N.
,
Ikawa
,
Y.
and
Aizawa
,
S.
(
1993
).
A novel negative selection for homologous recombination using diphtheria toxin A fragment gene
.
Anal. Biochem
.
214
,
77
86
.
Zhang
,
W.
,
Behringer
,
R. R.
and
Olson
,
E. N.
(
1995
).
Inactivation of the myogenic bHLH gene MRF4 results in up-regulation of myogenin and rib anomalies
.
Genes Dev
.
9
,
1388
1399
.