Zic2 regulates the kinetics of neurulation - PubMed (original) (raw)

Zic2 regulates the kinetics of neurulation

T Nagai et al. Proc Natl Acad Sci U S A. 2000.

Abstract

Mutation in human ZIC2, a zinc finger protein homologous to Drosophila odd-paired, causes holoprosencephaly (HPE), which is a common, severe malformation of the brain in humans. However, the pathogenesis is largely unknown. Here we show that reduced expression (knockdown) of mouse Zic2 causes neurulation delay, resulting in HPE and spina bifida. Differentiation of the most dorsal neural plate, which gives rise to both roof plate and neural crest cells, also was delayed as indicated by the expression lag of a roof plate marker, Wnt3a. In addition the development of neural crest derivatives such as dorsal root ganglion was impaired. These results suggest that the Zic2 expression level is crucial for the timing of neurulation. Because the Zic2 knockdown mouse is the first mutant with HPE and spina bifida to survive to the perinatal period, the mouse will promote analyses of not only the neurulation but also the pathogenesis of human HPE.

PubMed Disclaimer

Figures

Figure 1

Figure 1

The Zic2 knockdown mutation. (a) Mouse Zic2 gene, targeting construct, and mutated Zic2 gene (_Zic2_kd). The targeting vector (Top) contains an 8.0-kb and a 2.3-kb region homologous to the Zic2 gene and a neomycin-resistance gene, respectively, driven by the phosphoglycerate kinase gene (PGK) promoter (NEO). The diphtheria toxin A fragment gene driven by the MC1 promoter (DT) was inserted in the 3′ end of the Zic2 gene. In the Zic2kd allele, a homologous recombination occurred in the 3′ end, whereas a large portion of the 5′ homologous region (red) was deleted and the remaining part was connected to the first intron (green) illegitimately. The connecting point contains the three bases of the overlapping sequence between the 5′ and intron sequence as determined by nucleotide sequencing (data not shown). As a result, PGKneo and 622 bp of the 5′ homologous regions were inserted into the first intron. (b) Southern blots verifying the structure of the mutated allele. Genomic DNA was extracted from Zic2+/+, _Zic2_kd/+, and _Zic2_kd/kd, digested with _Eco_RI and _Bam_HI, and hybridized with four distinct probes. 5′ and 3′ probes, 5′ and 3′ flanking region of the targeting vector, respectively. Ex1 and NEO probes, fragments containing the Zic2 exon1 and neomycin-resistance gene, respectively. The expected sizes of restriction fragments are indicated in a. (c and d) The amount of Zic2 transcript from E11.5 embryonic tissue was measured by Northern blot (c) or quantitative RT-PCR analysis (d). Expression of the mutated allele was 21% that of the wild-type allele. RNA extracted from E11.5 whole embryo (W), head (H), anterior trunk (AT), and posterior trunk (PT) was analyzed. In c, there are additional bands (*) in _Zic2_kd/+ and _Zic2_kd/kd, which correspond to the unspliced mRNA precursor (data not shown), in addition to bands that correspond to the mature transcripts (Zic2). Arrowheads indicates the positions of 28S and 18S ribosomal RNA. The bottom frame shows 28S RNA with similar content and integrity. Analysis of RNA from E17.5 embryos gave similar results (data not shown). (e) Immunoblot using the anti-Zic2 antibody (Right). The density of the bands representing Zic2 protein (55.5 kDa) in the E11.5 Zic2+/+, _Zic2_kd/+, and _Zic2_kd/kd whole embryos was consistent with the amount of Zic2 transcript shown in c and d. (Left) The total protein used for the immunoblot as revealed by Coomassie blue staining.

Figure 2

Figure 2

Neural tube defects found in _Zic2_kd/kd mice. (a) Lateral views of P0 Zic2+/+ (Left) and _Zic2_kd/kd (Right) newborns. Heads of _Zic2_kd/kd animals were significantly smaller (microcephaly). Arrowhead, pes equinus; *, spina bifida. (b) Dorsal close-up view of the P0 _Zic2_kd/kd embryo. Spina bifida was always observed. (c) Lateral close-up view. Tails were irregularly curled. (d_–_g) Transverse sections through thoracic (d and e) and lumbar spinal cord (f and g) of Zic2+/+ (d and f) and _Zic2_kd/kd (e and g) E15.5 embryos. drg, Dorsal root ganglion; sc, spinal cord; osc, open spinal cord; v, vertebral body. (h and i) Scanning electron micrographs show no closure of lumbar spinal cord in _Zic2_kd/kd embryos (i) at E9.5 and successful closure in an E9.5 Zic2+/+ counterpart (h). np, Posterior neuropore. (j_–_s) Holoprosencephaly found in the _Zic2_kd/kd embryo. Coronal (j and k) and parasagittal (l and m) sections through cerebrum of Zic2+/+ (j and l) and _Zic2_kd/kd (k and m) E15.5 brain. Cerebral cortex (cc) of both hemispheres in _Zic2_kd/kd brain is fused at the dorsal midline, with the single ventricle and the juxtaposed ganglionic eminence (ge) (k). Cerebral cortex and olfactory bulb (ob) of _Zic2_kd/kd brain are hypoplastic (m). Magnified frontal views of head (n and o) and horizontal sections through eye (p_–_s) of Zic2+/+ (n, p, and r) and _Zic2_kd/kd (o, q, and s) E10.5 embryo. The telencephalic roof plate region in p and q is magnified in r and s, respectively. Telencephalic roof plate (arrowheads in n, p, and r) is apparently absent in _Zic2_kd/kd animals (arrowheads in o and q), resulting in the loss of discrimination between left and right telencephalic vesicles. (t and u) A typical example of exencephaly found in the E12.5 _Zic2_kd/kd embryo. p, Lateral view; q, dorsal view. Telencephalic (t), diencephalic (d), and mesencephalic (m) neural plate fail to fuse at the midline [indicated by Ex (exencephaly)]. The spinal cord remains open in the lumbosacral region (*). (v) Lateral view of _Zic2_kd/kd embryo (E18.5) possessing anencephaly (arrowheads) and spina bifida (*). (w) Top view of embryo in v. In this particular animal, almost all brain tissue was lost.

Figure 3

Figure 3

Development of the roof plate is disturbed in the _Zic2_kd/kd embryo. Immunohistochemistry (a_–_d), TUNEL staining (e and f), and in situ hybridization (g_–_l) were performed on Zic2+/+ (a, c, e, g, i, and k) and _Zic2_kd/kd (b, d, f, h, j, and l). (a_–_d) Immunohistochemical staining using antiphospho-histone H3 antibody to detect mitotic cells in the sections through E9.5 telencephalon. (a and b) Bright-field view. In the Zic2+/+ animal, the mitotic cells were scarce in the prospective roof plate region with thinning (arrowheads in a and c). Such a scarcity or thinning was not observed in the _Zic2_kd/kd animals (arrowheads in b and d). (e and f) TUNEL staining of the sections through the E10.5 telencephalic roof plate (trp). In the Zic2+/+ animal, staining characteristic of dying cells was observed at the midline (e, arrowheads) whereas no staining in the corresponding region of _Zic2_kd/kd animals was observed (f, arrowheads). (g_–_l) In situ hybridization showing the distribution of Wnt3a (g, h, i, and j, purple), Pax6 (i and j, orange), Pax3 (k and l, purple), and Shh (k and l, orange), in transverse sections through telencephalic roof plate (g and h) and through the lumbar spinal cord of E10.5 embryos (i_–_l). Note that Wnt3a expression is absent in the telencephalic roof plate (h) and reduced in the edges of the open spinal cord (j, arrowheads), which corresponds to the roof plate of the properly closed spinal cord (i, arrowheads). Pax3 staining in the dorsal root ganglia (drg) of _Zic2_kd/kd embryos was hardly visible whereas spinal cord (sc) and myotome (mt) staining remained visibly unchanged (k and l).

Figure 4

Figure 4

The delay of neurulation is concomitant with the delay of Wnt3a expression in _Zic2_kd/kd embryo. Lateral (a_–_c and g_–_i), dorsal (d_–_f), and magnified frontal views (j– l) of the E9.0 (a_–_f) and E10.5 (g_–_l) Zic2+/+ (a, d, g, and j), _Zic2_kd/+ (b, e, h, and k), or _Zic2_kd/kd (c, f, i, and l) littermate embryos stained with Wnt3a probe by whole-mount in situ hybridization. Lines in a_–_c indicate the rostrocaudal extension of the Wnt3a expression in the dorsal neural tube. Arrowheads in d– f indicate the expression in the dorsal midbrain. The anterior neuropore of _Zic2_kd/kd is not closed (f). The arrowheads in g_–_i and j_–_l indicate the lumbosacral expression and telencephalic roof plate expression, respectively. ps; primitive streak region.

Figure 5

Figure 5

Skeletal abnormalities found in the _Zic2_kd/kd mouse. (a and b) Dorsal views of the ossified (red) and cartilaginous (blue) axial skeleton from E17.5 Zic2+/+ (a) or _Zic2_kd/kd (b) mice. Note that the dorsal aspect of the vertebrae (vertebral arches) was irregularly opened and that abnormal fusions between the rostrocaudally adjacent arches had formed (arrowheads). C2, Cervical vertebra 2; T1, thoracic vertebra 1; L1, lumbar vertebra 1. (c and d) Comparison of Zic2+/+ (Upper) and _Zic2_kd/kd (Lower) thoracic vertebra 6 (c, T6) and lumbar vertebra (d, L3). ap, Articular process; tp, transverse process; va, vertebral arch; vb, vertebral body. (e and f) Hand (H) and foot (F) skeletal patterns also were disorganized. The fourth and fifth metacarpal/metatarsal bones were abnormally fused in the _Zic2_kd/kd mouse (f) whereas such fusion was not noted in the Zic2+/+ mouse (e) (arrowheads). Similarly, abnormal fusion was found in the carpal (*) and tarsal bones (arrow).

Similar articles

Cited by

References

    1. Edmonds L D, James L M. Morbid Mortal Wkly Rep. 1990;39:19–23.
    1. Ming J E, Muenke M. Clin Genet. 1998;53:155–163. - PubMed
    1. Brown S A, Warburton D, Brown L Y, Yu C, Roeder E R, Stengel-Rutkowski S, Hennekam R C M, Muenke M. Nat Genet. 1998;20:180–183. - PubMed
    1. Benedyk M J, Mullen J R, DiNardo S. Genes Dev. 1994;8:105–111. - PubMed
    1. Aruga J, Nagai T, Tokuyama T, Hayashizaki Y, Okazaki Y, Chapman V M, Mikoshiba K. J Biol Chem. 1996;271:1043–1047. - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources