Spatiotemporal expression pattern of Myt/NZF family zinc finger transcription factors during mouse nervous system development

Authors

  • Fumio Matsushita,

    1. Department of Biology, School of Health Science, Fujita Health University, Toyoake, Aichi, Japan
    2. Division of Cell Biology, Institute for Comprehensive Medical Science (ICMS), Fujita Health University, Toyoake, Aichi, Japan
    Search for more papers by this author
  • Toshiki Kameyama,

    Corresponding author
    1. Division of Cell Biology, Institute for Comprehensive Medical Science (ICMS), Fujita Health University, Toyoake, Aichi, Japan
    2. Division of Gene Expression Mechanism, Institute for Comprehensive Medical Science (ICMS), Fujita Health University, Toyoake, Aichi, Japan
    • Correspondence to: Toshiki Kameyama, Division of Gene Expression Mechanism, Institute for Comprehensive Medical Science, Fujita Health University, Toyoake, Aichi 470–1192, Japan. E-mail: tkame@fujita-hu.ac.jp

    Search for more papers by this author
  • Yuzo Kadokawa,

    1. Division of Cell Biology, Institute for Comprehensive Medical Science (ICMS), Fujita Health University, Toyoake, Aichi, Japan
    2. Department of Biology, School of Medicine, Fujita Health University, Toyoake, Aichi, Japan
    Search for more papers by this author
  • Tohru Marunouchi

    1. Division of Cell Biology, Institute for Comprehensive Medical Science (ICMS), Fujita Health University, Toyoake, Aichi, Japan
    Search for more papers by this author

Abstract

Background:

Three members of the Myt/NZF family of transcription factors are involved in many processes of vertebrate development. Several studies have reported that Myt1/NZF-2 has a regulatory function in the development of cultured oligodendrocyte progenitors or in neuronal differentiation during Xenopus primary neurogenesis. However, little is known about the proper function of Myt/NZF family proteins during mammalian nervous system development. To assess the possible function of Myt/NZF transcription factors in mammalian neuronal differentiation, we determined the comparative spatial and temporal expression patterns of all three types of Myt/NZF family genes in the embryonic mouse nervous system using quantitative reverse transcriptase polymerase chain reaction and in situ hybridization. Results: All three Myt/NZF family genes were extensively expressed in developing mouse nervous tissues, and their expression was transient. NZF-1 was expressed later in post-mitotic neurons. NZF-2 was initially expressed in neuronal cells a little earlier than NZF-3. NZF-3 was initially expressed in neuronal cells, just after proliferation was complete. Conclusion: These expression patterns suggest that the expression of NZF family genes is spatially and temporally regulated, and each Myt/NZF family gene may have a regulatory function in a specific phase during neuronal differentiation. Developmental Dynamics 243:588–600, 2014. © 2013 Wiley Periodicals, Inc.

INTRODUCTION

Myelin transcription factor (Myt)/Neural zinc finger (NZF) proteins comprise a unique family of transcription factors that play a key role in differentiation of a wide array of cell types. Myt/NZF family proteins have characteristic CCHHC-type zinc finger motifs as DNA-binding domains and are predominantly expressed in the developing nervous system. Myt1 was identified as a binding protein for the promoter region of myelin proteolipid protein (Kim and Hudson, 1992). Three types of Myt/NZF family proteins have been reported in mammals: Myt1 (NZF-2 and NZF-2b) (Matsushita et al., 2002), Myt1l (NZF-1 or png-1) (Jiang et al., 1996; Weiner and Chun, 1997), and St18 (NZF-3 or Myt3) (Yee and Yu, 1998; Jandrig et al., 2004). Myt/NZF family proteins act as transcriptional regulators by recruiting a transcriptional co-repressor, Sin3B, and histone deacetylases (Romm et al., 2005). A series of studies on the function of Myt/NZF family proteins was done, focusing on oligodendrocyte differentiation, neuronal differentiation, pancreas islet cell differentiation, and oncogenesis (Bellefroid et al., 1996; Nielsen et al., 2004; Wang et al., 2008, 2007; Yang et al., 2008; Kameyama et al., 2011). These studies reported that all of the Myt/NZF family genes are expressed in the developing mammalian central nervous system (CNS). However, the description of the expression of each Myt/NZF family gene in these previous studies was not comprehensive in terms of tissue or temporal expression. In this report, we performed a detailed and comparative analysis of the expression patterns of Myt/NZF family genes during development of the embryonic mouse nervous system. Interestingly, all three Myt/NZF family genes were expressed over a wide area and in almost all regions of the developing nervous system. However, temporal expression of each Myt/NZF family gene was regulated during the neurogenic phases.

RESULTS

We previously isolated three types of known mouse Myt/NZF family cDNA clones (Kameyama et al., 2011). Myt/NZF family genes have many different names. In this study, to simplify the comparison of each Myt/NZF family gene, we used NZF-1 (Myt1l, png-1), NZF-2 (Myt1), and NZF-3 (St18, Myt3) as the nomenclature. Moreover, two forms of NZF-2 have been reported that have or do not have an additional CCHHC zinc finger motif at the N-terminus (Kim et al., 1997; Matsushita et al., 2002). In this study, we used primers and probes common to both forms of NZF-2. Because NZF-2b is the predominant form of NZF-2, the observed results for NZF-2 were almost identical to the expression levels and the distribution of NZF-2b transcripts (Matsushita et al., 2002).

Temporal Changes in Expression Levels of NZF Family Molecules During Brain Development

We analyzed expression levels of the NZF family genes in the developing mouse brain with quantitative reverse transcriptase polymerase chain reaction (RT-PCR) (Fig. 1). Expression levels of NZF-1 and NZF-2 were up-regulated from 10.5 days post coitum (dpc) to 15.5 dpc and then down-regulated. Expression of NZF-3 reached the first peak around 12.5 dpc, the second peak around postnatal day (P) 7, and the third peak in the adult. Compared with the rapid down-regulation of NZF-2 and NZF-3 transcripts in embryonic stages, NZF-1 was down-regulated more slowly throughout postnatal stages. At 15.5 dpc, NZF-1 transcripts were 5- to 10-fold more abundant than other NZF family genes. These developmental changes in expression levels indicate that all NZF family genes are likely involved in neuronal and glial differentiation and maturation. To investigate the functional importance of NZF family genes in neuronal differentiation during prenatal nervous system development, we next analyzed the expression pattern of transcripts of NZF family genes in mouse embryos.

Figure 1.

Quantitative real-time reverse transcriptase polymerase chain reaction analyses of the expression of NZF (neural zinc finger) family genes. Copy numbers of transcripts of NZF family genes are shown by bars for sets of primers specific to NZF-1 (gray bars), NZF-2 (open bars), and NZF-3 (black bars). The sources of total RNAs were heads at 10.5 days post coitum (dpc) and whole brains at 12.5, 15.5, and 18.5 dpc, postnatal day (P) 2, P7, P14, P21, and adult mice. The final copy number of transcripts shown in the vertical axis was normalized to 2.0 × 105 molecules obtained with a set of primers specific to glyceraldehyde-3-phosphate dehydrogenase (GAPDH), which is a housekeeping gene. This normalized copy number corresponds to that in approximately 10 ng of total RNA.

Histological Analysis of Expression Patterns of NZF Family Genes

Comparison of the structure of NZF family proteins shows that the characteristic zinc finger domains are highly conserved among them and that an acidic region consisting of a stretch of glutamic acids and aspartic acids is present in NZF-1 and NZF-2. Therefore, as probes for in situ hybridization (ISH), we chose nonhomologous regions of cDNA lacking the zinc fingers and acidic domains (Fig. 2).

Figure 2.

Structural comparison of mouse NZF (neural zinc finger) proteins. NZF-1 (1188 amino acids), NZF-2 (1127 amino acids), and NZF-3 (1045 amino acids) proteins are schematically showed by boxes. The black boxes and the shaded boxes indicate CCHC-type zinc finger units and the acidic domains, respectively. Percentages of the corresponding nucleotide identities of the zinc finger clusters and of the other portions (set apart by straight lines) between NZF-2 and each other members are shown. The positions of in situ probes are indicated below each protein by blue horizontal bars. Portions of the bars outside the N-terminal and the C-terminal of these proteins show 5′ untranslated regions (UTRs) and the 3′-UTR of the corresponding mRNAs, respectively.

Expression patterns of NZF family genes in the forebrain region

Overview

At 9.5 dpc, NZF-2 and NZF-3 transcripts were detected in the caudo-ventral region of the forebrain vesicle, where the earliest differentiated neurons are observed with TuJ1 immunostaining (Fig. 3B–D, arrowheads). NZF-2 was first detected in the optic vesicle at 9.5 dpc (Fig. 3B, arrows). At 10.5 dpc, NZF-2 and NZF-3 transcripts were detected in the subventricular zone (SVZ) and TuJ1-positive differentiated neurons of optic vesicles, the future regions of the cerebral cortex (dorsal telencephalon; DTe), basal ganglia (ventral telencephalon; VTe), and hypothalamus (Fig. 3F–H). NZF-2 transcripts were more widely distributed in these areas than NZF-3 transcripts. Up to 10.5 dpc, NZF-1 transcripts were not detected in the forebrain region. At 11.5 dpc, transcripts of NZF family genes were detected in regions where neurogenesis occurs, including the cerebral cortex and basal ganglia (Fig. 3I–L). Among the three NZF family genes, the expression pattern of NZF-1 transcripts overlapped with the TuJ1 immunoreactive region, which clearly showed that NZF-1 was expressed in post-mitotic neurons. The expression of NZF-3 was the most restricted in the SVZ, with an increasing ventromedial gradient. The expression of NZF-2 overlapped with expression of NZF-1 and NZF-3, with higher expression levels in the SVZ. At 12.5 dpc and 13.5 dpc, NZF family genes were expressed in a remarkably widespread area including the cerebral cortex, archipallium, basal ganglia, thalamus, and hypothalamus (Fig. 3M–T). After 12.5 dpc, the expression levels of NZF-1 were maintained throughout the embryonic stages. However, those of NZF-2 were gradually decreased after 14.5 dpc, and those of NZF-3 were also decreased after 13.5 dpc, except for expression in the basal ganglia and hippocampus (Fig. 3Q–Z). These data were consistent with the quantitative RT-PCR analysis.

Figure 3.

A–Z: Expression of NZF (neural zinc finger) family genes in the forebrain. Distribution of NZF-1 mRNA (A,E,I,M,Q,U,X), NZF-2 mRNA (B,F,J,N,R,V,Y), NZF-3 mRNA (C,G,K,O,S,W,Z), and TuJ1 immunoreactivity (D,H,L,P,T) in coronal sections of 9.5 days post coitum (dpc) (A–D), 10.5 dpc (E–H), 11.5 dpc (I–L), 12.5 dpc (M–P), 14.5 dpc (U–W), 15.5 dpc (X–Z), and on near-horizontal sections of 13.5 (dpc) (Q–T) forebrains. Arrowheads in B–D indicate the earliest differentiated neurons in diencephalic regions. Arrows in B indicate NZF-2 in optic vesicles. APa, archipallium; BG, basal ganglia; CTX, cortex; DTe, dorsal telencephalon; fIC, fibers of the internal capsule; HC, hippocampus; HT, hypothalamus; IC, internal capsule; LGE, lateral ganglionic eminence; MGE, medial ganglionic eminence; OpV, optic vesicle; Pal, pallidum; POA, preoptic area; Str, striatum; TH, thalamus; Vg, trigeminal ganglion; VTe, ventral telencephalon. Scale bars = 0.1 mm in A–D; 0.2 mm in E–H; 0.5 mm in I–W; 1 mm in X–Z.

Cerebral cortex

Next, we analyzed the expression patterns of NZF family genes in the cerebral cortex compared with Tbr1 and TuJ1 as a marker of cerebral cortical histogenesis (Hevner et al., 2001; Englund et al., 2005). NZF-2 and NZF-3 were expressed in the future cerebral cortex region at 10.5 dpc, and all NZF family transcripts were expressed at 11.5 dpc. At 12.5 dpc, all NZF family genes were expressed at the preplate (PP) where Tbr1 was expressed (Fig. 4A–E). After 13.5 dpc, the expression patterns of NZF family genes in the cerebral cortex became different from each other (Fig. 4F–J). At 13.5 dpc, NZF-1 was expressed in TuJ1-positive post-mitotic neurons, with higher levels at the upper limit of the cortical plate (CP, Fig. 4F compared with 4I and 4J). NZF-2 was detected in all layers including the ventricular zone (VZ), with higher levels around the boundary of the CP and VZ (Fig. 4G). The expression of NZF-3 was distinct in the upper cortical plate, and low-level expression was observed in the lower boundary adjacent to the VZ (Fig. 4H). At 14.5 dpc, the expression of NZF-1 was increased in post-mitotic neurons (Fig. 4K). NZF-2 was also expressed in all layers (Fig. 4L). On the other hand, the expression of NZF-3 was remarkably reduced in the dorsal and lateral part of the cerebral cortex (Fig. 4M).

Figure 4.

A–N: Expression of NZF (neural zinc finger) family genes in the cerebral cortex. Coronal sections of 12.5 days post coitum (dpc) (A–E), 14.5 dpc (K–N), and near-horizontal sections of 13.5 dpc (F–J) cerebral cortices hybridized with NZF-1 (A,F,K), NZF-2 (B,G,L), NZF-3 (C,H,M) and Tbr-1 (D,I,N) cRNA probes, or immunostained with TuJ1 antibody (E,J) are shown. CP, cortical plate; IZ, intermediate zone; PP, preplate; SVZ, subventricular zone; VZ, ventricular zone. Scale bar = 0.1 mm.

Basal ganglia

At 10.5 dpc in the future basal ganglia (VTe), NZF-2 transcripts were present with distinct level, whereas NZF-3 transcripts were faintly detected. After 11.5 dpc, all NZF family genes were intensely expressed in the basal ganglia. At 12.5 dpc, similar to the other regions of the forebrain, NZF-1 was expressed in TuJ1-positive post-mitotic neurons (Fig. 5A,E,D,H). NZF-2 was also expressed in post-mitotic neurons, in SVZ cells, and in some VZ cells (Fig. 5B,F,D,H). NZF-3 exhibited the most restricted expression pattern among the three NZF family genes. In the lateral ganglionic eminence (LGE), distinct expression of NZF-3 was mainly detected in the SVZ, whereas in the medial ganglionic eminence (MGE), expression of NZF-3 was detected in the SVZ, some VZ cells, and post-mitotic neurons (Fig. 5C and D compared with Fig. 5G and H). The prominent expression of NZF-3 continued in the basal ganglia after 14.5 dpc.

Figure 5.

A–K: Expression of NZF (neural zinc finger) family genes in the basal ganglia and hippocampus. Coronal sections of 12.5 days post coitum (dpc) lateral ganglionic eminences (A–D), 12.5 dpc medial ganglionic eminences (E–H), and 18.5 dpc hippocampus (I–K) were hybridized with NZF-1 (A,E,I), NZF-2 (B,F,J), and NZF-3 (C,G,K) cRNA probes, or immunostained with TuJ1 antibody (D,H). CA1, CA3, cornu ammonis fields; DG, dentate gyrus; LGE, lateral ganglionic eminence; MGE, medial ganglionic eminence; SVZ, subventricular zone; VZ, ventricular zone. Scale bar = 0.1 mm.

Hippocampus

The transcripts of NZF family genes were observed in the hippocampus (HC) anlage from 12.5 dpc. At 18.5 dpc, NZF-1 transcripts were detected in the pyramidal cells of the cornu ammonis (CA) and granule cells of the dentate gyrus (DG, Fig. 5I). The expression of NZF-2 at 18.5 dpc was remarkably low, whereas NZF-2 was expressed in CA cells at 14.5 dpc and 15.5 dpc (Fig. 5J compared with Fig. 3R,V,Y). NZF-3 transcripts were detected specifically in the pyramidal cells of the CA (Fig. 5K).

Expression patterns of NZF family genes in the midbrain and hindbrain regions

At 9.5 dpc, NZF-2 was detected in the basal plate of the hindbrain, which is the future pons and medulla (Fig. 6B). NZF-3 showed more a restricted expression pattern than NZF-2 in the ventral basal plate just adjacent to the floor plate of the hindbrain (Fig. 6C). Until 10.5 dpc, NZF-1 could not be detected in the midbrain and hindbrain regions (Fig. 6A,E). At 10.5 dpc, both NZF-2 and NZF-3 were expressed in the pons and medulla (Fig. 6F,G). At 12.5 dpc and 13.5 dpc, NZF family genes were detected in the midbrain and hindbrain, including the superior colliculus (sCo), tegmentum (Teg), inferior colliculus (iCo), isthmus (Is), pons (Ps), cerebellum (CBL), and medulla (Me). NZF-1 was expressed in TuJ1-positive post-mitotic neurons, the same as in other regions (Fig. 6I,L,M,P). NZF-2 was expressed very weakly in the VZ, at high levels in the SVZ, and moderately in post-mitotic neurons (Fig. 6J,N). NZF-3 showed a restricted expression pattern only in the SVZ (Fig. 6K,O). At 13.5 dpc and later, the external granular layer (EGL) of the cerebellum was the major region expressing NZF-3. At 18.5 dpc, NZF-2 was also detected in the EGL at a significant level, whereas NZF-1 was not detected in the EGL (Fig. 6Q,R). NZF-3 was only detected in the EGL of the cerebellum within the midbrain and hindbrain (Fig. 6S).

Figure 6.

A–S: Expression of NZF (neural zinc finger) family genes in the midbrain and hindbrain. Coronal sections of 9.5 days post coitum (dpc) (A–D), 10.5 dpc (E–H), 12.5 dpc (I–L), 13.5 dpc (M–P), and 18.5 dpc (Q–S) brain stems were hybridized with NZF-1 (A,E,I,M,Q), NZF-2 (B,F,J,N,R), and NZF-3 (C,G,K,O,S) cRNA probes, or immunostained for TuJ1 (D,H,L,P). In all figures, dorsal is to the top, and in A–D, the bottom regions outside the figures are the forebrains with optic vesicles. Arrows in S indicate ISH signals for NZF-3 transcripts in the external germinal layer of the cerebellum. CBL, cerebellum; cSp, cervical spinal cord; EGL, external germinal layer of cerebellum; iCo, inferior colliculus; Is, isthmus; LDN, lateral deep nucleus; Me, medulla; Ps, pons; OtV, otic vesicle; sCo, superior colliculus; Teg, tegmentum; VII–VIII, facio-acoustic neural crest complex. Scale bars = 0.1 mm in A–D; 0.2 mm in E–H, Q–S; 0.5 mm in I–P.

Expression of NZF family genes in the neural retina

In optic vesicles, NZF-2 was first detected at 9.5 dpc, and NZF-3 was detected at 10.5 dpc. At 12.5 dpc, transcripts of all the NZF family genes were observed, and the expression level of NZF-3 was much weaker than the others (Fig. 7A–C). NZF-2 transcripts were detected at high intensity in the inner part of the neural layer of the retina and at low intensity in dispersed cells of the outer part of the neural layer of the retina at 12.5 dpc (Fig. 7B). At 13.5 dpc, the expression of NZF-1 was restricted to the inner part of the developing neural retina (Fig. 7E). Prominent expression of NZF-2 was observed in the inner half of the developing neural retina at 13.5 dpc (Fig. 7F). The expression levels of NZF-3 were very low at 13.5 dpc (Fig. 7G). At 15.5 dpc, NZF-1 expression was found in the innermost part of the inner neuroblastic layer of the neural retina (iNBL) (Fig. 7I), which is post-mitotic and the future ganglion cell layer. In addition to expression in the iNBL, obvious expression of NZF-2 was detected in the outer neuroblastic layer of the neural retina at 15.5 dpc (oNBL; Fig. 7J). Moderate expression of NZF-3 was detected in some oNBL cells, and weak expression was observed in dispersed cells throughout the retina (Fig. 7K).

Figure 7.

A–N: Expression of NZF (neural zinc finger) family genes in the neural retina and olfactory epithelium. Coronal sections of 12.5 days post coitum (dpc) (A–D), 13.5 dpc (E–H), 15.5 dpc (I–K) eyes, and 13.5 dpc olfactory epithelium were hybridized with NZF-1 (A,E,I,L), NZF-2 (B,F,J,M), or NZF-3 (C,G,K,N) cRNA probes, or immunostained for TuJ1 (D,H). iNBL, inner neuroblastic layer of the neural retina; oNBL, outer neuroblastic layer of the neural retina. Scale bar = 0.2 mm in A–K; 0.1 mm in L–N.

Expression of NZF family genes in the olfactory epithelium

In the olfactory epithelium, NZF family genes were expressed at 12.5 dpc and later. High-level expression of NZF-1 transcripts was detected in the middle layer of the epithelium but not in the luminal and basal layer at 13.5 dpc (Fig. 7L). NZF-2 transcripts were detected in scattered cells spanning the epithelium at 12.5 dpc and in the basal half of the epithelium at 13.5 dpc with a much weaker level of expression than that of NZF-1 (Fig. 7M and data not shown). Weak expression of NZF-3 was detected at 12.5 dpc in scattered cells that were more densely populated in the basal half of the epithelium, and moderate expression at 13.5 dpc was restricted to the basal layer of the epithelium (Fig. 7N and data not shown).

Expression of NZF family genes in the spinal cord, dorsal root ganglia, and cranial ganglia

At 9.5 dpc, NZF-1 transcripts were detected at the most ventrolateral margin of the spinal cord (Fig. 8A, arrowhead). At the same stage (9.5 dpc), NZF-2 transcripts were clearly detected in the earliest born neurons of the spinal cord (Fig. 8B compared with 8D). They were detected in the motor neurons of the ventrolateral margin of the spinal cord and in the dispersed cells of alar plate interneurons (Fig. 8B arrows) (Pfaff et al., 1996; Matsushita et al., 2002). At 9.5 dpc, the expression region of NZF-3 overlapped with that of NZF-2, but was more restricted in the SVZ (Fig. 8C). To visualize the motor neuron pools, we performed double immunofluorescence staining at 10.5 dpc. NZF-2 protein was strongly expressed in Isl-1-positive motor neurons in addition to the SVZ (Fig. 8F–H). At 12.5 dpc, NZF-1 was expressed in TuJ1-positive post-mitotic neurons with an increasing ventrolateral gradient (Fig. 8I, compared with L). NZF-2 was expressed mostly in post-mitotic neurons at higher levels in the SVZ and at lower levels in differentiated neurons in the mantle layer (ML) with a decreasing ventrolateral gradient, in contrast to NZF-1 expression (Fig. 8J). NZF-3 was expressed exclusively in the SVZ (Fig. 8K). At 13.5 dpc, NZF-1 showed overall, constant expression in post-mitotic neurons and no expression in the white matter (WM, Fig. 8M). NZF-2 and NZF-3 were expressed in the SVZ and the ML and at higher levels in the SVZ of the alar plate (Fig. 8N,O). The expression of NZF-3 in the SVZ extended more ventrally than that of NZF-2 at this stage (black arrowheads in Fig. 8O compared with white arrowheads in N and O). The expression level of NZF-3 in differentiated neurons in the ML was lower than that of NZF-2.

Figure 8.

A–T: Expression of NZF (neural zinc finger) family genes in the spinal cord and cranial ganglia. Distribution of NZF-1 mRNA (A,I,M,Q), NZF-2 mRNA (B,J,N,R), NZF-3 mRNA (C,K,O,S), and neuron-specific class III β-tubulin (D,L,P,T) detected with the monoclonal antibody TuJ1 at 9.5 days post coitum (dpc) (A–D), 12.5 dpc (I–L), and 13.5 dpc (M–T) are shown. Double immunofluorescence staining for Isl-1 (F) and NZF-2 (G) and counter-staining with DAPI (E) in 10.5 dpc spinal cord. Panel H shows a merged image of Isl-1, NZF-2, and DAPI. In B and C, arrows indicate alar plate interneurons, and arrowheads point to differentiated motor neurons. In N and O, white arrowheads indicate the SVZ of the alar plate, and black arrowheads indicate the SVZ of the ventral region in which NZF-3 transcripts were detected at a higher level. DRG, dorsal root ganglion; ML, mantle layer; MN, motor neuron; SVZ, subventricular zone; sXg, superior ganglion of the vagus nerve; Vg, trigeminal ganglion; VIIg, facial ganglion; VIIIg, vestibulocochlear ganglion; VZ, ventricular zone; WM, white matter. Scale bars = 0.1 mm in A–P; 0.2 mm in Q–T.

In dorsal root ganglia (DRG), NZF-2 was clearly detected at 9.5 dpc and 10.5 dpc, whereas NZF-1 and NZF-3 were faintly detected at 9.5 dpc (Fig. 8A–H). At 11.5 dpc and 12.5 dpc, all three members were clearly detected in DRG. Most cells in DRG expressed NZF-2, but only restricted cells expressed NZF-1 or NZF-3 (Fig. 8I–K and data not shown). At 13.5 dpc, the expression of NZF family genes was down-regulated, and NZF-3 in particular was faintly detected in DRG cells (Fig. 8M–O).

In cranial neural crest tissues, NZF-2 and NZF-3 were first detected at 9.5 dpc in the facio-acoustic (VII-VIII) neural crest complex and the trigeminal neural crest tissues, with much higher expression of NZF-3 (Fig. 6B,C, and data not shown). At 11.5 dpc, transcripts of all NZF family genes were observed in trigeminal ganglia (Vg, Fig. 3I–K). At 12.5 dpc, transcripts of all NZF family genes were observed in all cranial ganglia (data not shown). At 13.5 dpc, NZF-1 transcripts showed a moderate level of expression in all ganglia (Fig. 8Q). At the same stage, among the NZF family genes, the expression of NZF-2 was most prominent in the trigeminal ganglia (Vg), facial ganglia (VIIg), vestibulocochlear ganglia (VIIIg), and glossopharyngeal ganglia (IXg) (Fig. 8R). In the VIIIg, strong ISH signals for NZF-2 covered almost all cells, whereas in the Vg and VIIg, ISH signals for NZF-2 were only detected in scattered cells. The expression of NZF-3 in cranial ganglia was low and was restricted to a portion of the population of cells in the Vg, and in other ganglia at 13.5 dpc, whereas the expression of NZF-3 were detected at high intensity in the vagus ganglia (Xg) and part of the VIIIg at 12.5 dpc (Vg, VIIg, VIIIg, sXg; Fig. 8S and data not shown).

NZF family genes were expressed in other neural crest-derived neurons during embryonic development, including the autonomic nervous system of sympathetic cervical ganglia and the enteric nervous system, with especially higher expression levels of NZF-2 (Matsushita et al., 2002).

Regulation of the Timing of Expression of Each NZF Family Gene

NZF family genes were expressed in remarkably widespread areas of embryonic nervous tissues. Most cells of the central and peripheral nervous tissues seemed to express NZF family genes. However, temporally, expression of each NZF family gene was transient and different from each other. In general, NZF-1 expression was detected only in post-mitotic neurons in the TuJ1-positive ML. NZF-2 was expressed in some VZ cells, SVZ cells, and post-mitotic neurons. NZF-3 expression was restricted to SVZ cells. To confirm the precise timing of expression of NZF family genes, we labeled mitotic cells with bromodeoxyuridine (BrdU) pulse-labeling for 2 hr at 12.5 dpc. Then, we performed double staining for ISH of NZF family genes and anti-BrdU immunohistochemistry in the spinal cord. NZF-1-expressing cells barely overlapped with BrdU-labeled cells (Fig. 9A), indicating that they were completely post-mitotic neurons. Many NZF-2-expressing cells were located in or near the VZ and overlapped with BrdU-labeled cells (Fig. 9B, arrowheads). Several NZF-3-expressing cells overlapped with BrdU-labeled SVZ cells (Fig. 9C, arrowheads). These data indicate that NZF-2 and NZF-3 were expressed in post-mitotic neurons and in some mitotic cells.

Figure 9.

A–C: Bromodeoxyuridine (BrdU) pulse labeling in the spinal cord. Distribution of immunoreactivity against BrdU (brown) and in situ hybridization (ISH) signals for NZF-1 (A), NZF-2 (B), and NZF-3 (C) transcripts (blue) on transverse sections of 12.5 days post coitum (dpc) spinal cords. BrdU was injected intraperitoneally 2 hr before killing. Arrowheads indicate BrdU-positive cells that are double positive for ISH signals for the NZF-2 probe (B) and the NZF-3 probe (C). Scale bar = 0.1 mm.

DISCUSSION

Previously, we reported that all three NZF family genes were expressed after neuronal induction and had a potential positive role in neuronal differentiation in P19 embryonal carcinoma cells (Kameyama et al., 2011). In this study, we compared the spatiotemporal expression pattern of three NZF family genes in the developing mouse central and peripheral nervous systems. One of the unique features of the expression patterns of NZF family genes is the widespread expression. NZF-1, NZF-2, and NZF-3 were all expressed in almost all regions of embryonic nervous tissues. NZF family transcription factors are unique, ubiquitously expressed, positive regulators of neuronal differentiation. Well-known neurogenic transcription factors are expressed in specific regions, such that neurog1/2 is expressed in progenitor cells in the developing cortex and ascl1 is expressed in progenitor cells in the LGE (Porteus et al., 1994; Sommer et al., 1996; Ma et al., 1997; Toresson et al., 2000; Schuurmans and Guillemot, 2002). Our previous study showed that expression of NZF-3 was positively regulated by neurog1/2 (Kameyama et al., 2011). On the other hand, the expression region of NZF-3 in the developing forebrain was wider than that of neurog1/2. These results suggest that NZF-3 may regulate neuronal differentiation and is generally under the control of neurogenic transcription factors including neurog1/2. Hes 1 and Hes 5 are also expressed in the VZ throughout the CNS (Akazawa et al., 1992; Sasai et al., 1992). Members of the Hes family of basic helix-loop-helix factors are known as negative regulators of neuronal differentiation.

Another unique feature was the difference in timing of expression among each NZF family gene. The expression of NZF-3 was strictly transient and present mainly in the SVZ. The expression of NZF-2 began in the VZ and SVZ and was wider than that of NZF-3. During cortical development, neural precursor cells sequentially pass through phases of expansion, neurogenesis, and gliogenesis (Miller and Gauthier, 2007; Miyata et al., 2010). During the neurogenic stage, neural precursor cells produce two different types of daughter cells (Farkas and Huttner, 2008; Shioi et al., 2009). These asymmetric cell divisions generate a neural precursor cell and a post-mitotic immature neuron or an intermediate neural progenitor cell committed to the neuronal lineage. This neural progenitor cell produces neurons in the SVZ. The position and timing of expression of NZF-2 and NZF-3 coincided with the location of these immature neurons and neural progenitors.

The expression of NZF-1 followed the expression of NZF-2 and NZF-3 and was maintained thereafter. Moreover, the expression level of NZF-1 was much higher than that of NZF-2 and NZF-3. Recently, NZF-1 (Myt1l) was shown to be a member of Brn2, Ascl1, and Myt1l (BAM; BAMN: Brn2, Ascl1, Myt1l, and NeuroD1) that induces fibroblast cells to differentiate directly into neurons (Vierbuchen et al., 2010; Ambasudhan et al., 2011; Pang et al., 2011; Pfisterer et al., 2011; Yoo et al., 2011). The widespread expression of NZF-1, which was observed in almost all regions of the marginal zone in the developing nervous system, may be suitable for the general neurogenic ability of molecules such as BAM (BAMN) factors. During Xenopus leavis development, X-MyT1 is essential for neuronal differentiation (Bellefroid et al., 1996). All NZF family genes have a strong ability to induce neuronal differentiation in embryonal carcinoma stem cells (Kameyama et al., 2011). Together with these previous functional studies, the widespread expression in almost all regions of the CNS of all NZF family genes and the timing of expression around neuronal differentiation demonstrated in this study suggest that NZF family genes are important for neuronal differentiation. Performing in vivo functional studies will now be important for elucidating the physiological role of NZF family genes in CNS development and function.

EXPERIMENTAL PROCEDURES

Animals

Timed pregnant ddY strain mice were purchased from Nihon-SLC (Shizuoka, Japan). Noon of the day at which a vaginal plug was detected was designated as 0.5 dpc.

Real-Time RT-PCR

Quantitative real-time RT-PCR analyses were performed as described (Matsushita et al., 2002). Sets of primers for NZF-1 (5′-AAGCGGTACTGCAAGAATGC-3′, 5′-TTGCTACACGTGCTACTGGC-3′), NZF-2 (5′-ATACCTCTGTCCAGAAGGCG-3′, 5′-TGTCATCATCAGAGCGAACC-3′, common primers to NZF-2a and NZF-2b), NZF-3 (5′-AGTCCGTGCCAGCTCTTATG-3′, 5′-AGAGGATGTCTGTGGCTTCC-3′), and GAPDH (5′-ATTGTGGAAGGGC TCATGAC-3′, 5′-ATGCAGGGATGAT GTTCTGG-3′) give specific RT-PCR products of 116 bp, 102 bp, 113 bp, and 121 bp, respectively.

Isolation of Mouse NZF-1 cDNA

We isolated an NZF-1/Myt1l/png-1 cDNA (clone 5-2-1; Genbank accession no. AB212898) encoding the entire open reading frame of mouse NZF-1 from an embryonic mouse brain cDNA library (ICR, 17.5 dpc; Kameyama et al., 1996), using an RT-PCR product (Genbank accession no. U86338; nucleotides 211–4305) as a probe. Our clone 5-2-1 and two already obtained clones (png-1; strain BALB/c; Genbank accession no. U86338 and Myt1l; strain not specified; Genbank accession no. NM_008666) are almost identical but slightly different from each other, probably due to strain differences or deletion of some nucleotides. This causes some differences in the deduced products (png-1; Genbank accession no. AAC53157.1, Myt1l; Genbank accession no. AAC53457.1, and NZF-1_521; deduced product of our clone 5-2-1). NZF-1_521 is completely identical to png-1 except that amino acid 127 (aspartic acid) in NZF-1_521 is substituted with amino acids 127-8 (alanine-serine) in png-1 due to insertion of nucleotides. Compared with Myt1l, NZF-1_521, and png-1 are characterized by insertion of two amino acids (glycine-lysine) (caused by selection of another splice acceptor site) into 10 amino acids upstream of the second zinc finger. They are also completely different from Myt1l in the region (amino acids 633–646 of NZF-1_521) between the third and fourth zinc fingers due to deletions.

Tissue Preparation

Mouse embryos were fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS; 25 ml for the embryos of one pregnant mouse) with continuous rotation for 2 hr (8.5 dpc), overnight (9.5 dpc–11.5 dpc), 3 days (12.5 dpc–13.5 dpc), or 6 days (older than 14.5 dpc). Fixed embryos were washed twice with PBS and dehydrated stepwise with ethanol. These embryos were stored at −20°C in 100% ethanol until use. The fixed and dehydrated embryos were paraffin embedded using methyl benzoate and benzene as mediators between ethanol and paraffin. Serial paraffin sections (10 μm) were processed for histological analyses. Histological assignments were done with atlases of mouse embryos (Kaufman, 1992) and rat brain (Altman and Bayer, 1995).

ISH

ISH was performed as described with minor modifications (Nagai et al., 2000, 1997). Briefly, rehydrated paraffin sections were treated with 6% H2O2 in PBS, followed by proteinase K solution (1 μg/ml proteinase K [Roche, Penzberg, Germany], 50 mM Tris-HCl, and 5 mM EDTA) at 37°C, acetylated, and air dried. The sections were incubated with prehybridization buffer (5× SSC, 50% formamide, 0.1% Tween 20) at 60°C for 1 hr, and then hybridized with 1 μg/ml digoxigenin (DIG)-labeled cRNA probes in prehybridization solution at 60°C overnight. Unhybridized probes were washed out with several washes and treated with RNase A (20 μg/ml RNase A, 10 mM Tris-HCl (pH 8.0), 1 mM EDTA, 0.5 M NaCl, 0.1% Tween 20, 37°C, for 30 min). After incubation with blocking reagent (Roche), the slides were incubated with alkaline phosphatase-conjugated anti-DIG antibody (Roche), and finally DIG-labeled probes were detected using NBT/BCIP solution (Roche) as chromogenic substrates with 10% polyvinylalcohol 500 (Wako, Osaka, Japan) as an inhibitor of diffusion of resultant dye precipitates.

cRNA probes were transcribed with T7 or SP6 RNA polymerase (Roche) with DIG-RNA labeling mix (Roche). The following DNAs were used as templates: NZF-1 (Myt1l/png-1; a mixture of a 715-bp cDNA fragment corresponding to the 5′ untranslated region (UTR) and the N-terminal region, a 996-bp fragment between the first and second zinc fingers, and a 978-bp fragment between the two zinc finger clusters; PCR amplified; DDBJ/Genbank accession number U86338; nucleotides 1-715, 1148-2143, and 2366-3343, respectively), NZF-2 (Myt1) (NZF-2com probe [Matsushita et al., 2002]; a mixed probe, an 855-bp fragment corresponding to the region between the two zinc finger clusters, and a 1721-bp fragment corresponding to the 3′-UTR; DDBJ/Genbank accession number AB082378; nucleotides 1947-2,801, and 3790-5510, respectively), NZF-3 (St18/Myt3; a 1350-bp cDNA fragment corresponding to a 5′-UTR and the N-terminal region, which is upstream of the zinc finger domains, and a 792-bp HincII-BglII cDNA fragment between the two zinc finger clusters; Genbank accession number AB097467; nucleotides 1-1350 and 1606-2397), and Tbr1 (a 3550-bp fragment containing the entire coding region and approximately 1.5 kb of 3′-UTR; PCR-amplified; Genbank accession number U49251; nucleotides 76-3625).

BrdU Incorporation

BrdU incorporation was carried out as described (Miller and Nowakowski, 1988). BrdU solution (10 mg/ml in 1× PBS) was intraperitoneally injected into pregnant mice at 12.5 dpc at a dose of 50 μg/g body weight. Two hours after injection, the mice were killed by cervical dislocation. Immunodetection of the incorporated BrdU was carried out after ISH with the following pretreatments: The sections were fixed with 4% paraformaldehyde in PBS and then treated with 2 N HCl for 20 min at room temperature. The other immunostaining procedure is described below.

Immunohistochemistry and Immunofluorescence

Rehydrated paraffin sections were treated at room temperature with 0.3% H2O2 in methanol for 30 min, followed by blocking in blocking solution (2% skim milk (Nacalai Tesque, Kyoto, Japan), 0.3% Triton X-100 in PBS) for 1 hr, and then incubated with primary antibody for 1 hr or overnight at 4°C. After several washes, the sections were incubated with biotin-labeled affinity-purified anti-mouse IgG secondary antibody (Kirkegaard and Perry Laboratories, MD) in blocking solution for 1 hr at room temperature. Immunoreactive signals were detected with a Vectastain Elite ABC kit (Vector, CA) using diaminobenzidine and H2O2 as substrates for horseradish peroxidase. Primary antibodies used were mouse monoclonal anti-neuron-specific class III β-tubulin antibody (2 μg/ml, clone TuJ1; Covance Research Products, CA) and mouse monoclonal anti-BrdU antibody (10 μg/ml, MBL, Nagoya, Japan).

For immunofluorescence staining, 10.5 dpc mouse embryos were fixed in 4% paraformaldehyde in PBS. For cryosections, after incubation in a sucrose gradient (10, 20, and 30%; for 1 hr at each concentration), the embryos were embedded in Cryo Mount (Muto Pure Chemicals, Tokyo, Japan), frozen in liquid nitrogen, and stored at −80°C. Specimens were sectioned at a thickness of 12 μm using a cryostat (CM1800; Leica, Heerbrugg, Switzerland). The hydrated samples were treated with Dako REAL Target Retrieval Solution (DakoCytomation, Glostrup, Denmark) for 45 min at 95°C to liberate antibody binding. After blocking nonspecific reactions with 4% skim milk (Nacalai Tesque) in 10 mM Tris-HCl (pH 7.6)-buffered saline containing 0.1% Tween 20 (TBST) for 1 hr, sections were incubated with anti-MYT1 (affinity purified goat polyclonal antibody, 1:50 dilution, Santa Cruz Biotechnology, Dallas, TX) and with 20 μg/ml mouse anti-rat Islet-1 (40.2D6; Developmental Studies Hybridoma Bank, Iowa City, IA) in 4% skim milk/TBST overnight at 4°C. The sections were washed with TBST and then incubated with biotin-conjugated anti-goat IgG (1:200, Jackson ImmunoResearch, West Grove, PA) in 4% skim milk/TBST for 1 hr at room temperature. After several washes with TBST, sections were incubated with Alexa Fluor 488-conjugated donkey anti-mouse IgG (Life Technologies Corporation, Carlsbad, CA) and Alexa Fluor 594-conjugated streptavidin (Life Technologies Corporation) in 4% skim milk/TBST for 1 hr at room temperature. After several final washes, samples were mounted using ProLong Gold antifade reagent with DAPI (Life Technologies Corporation).

ACKNOWLEDGMENTS

We thank Ms. K. Shimura (Fujita Health University; FHU) for histological techniques, Dr. K. Hashimoto (FHU) and Dr. N. Hayashi (Tokyo Institute of Technology) for the real-time RT-PCR analysis, and Ms. Y. Kawakami for preparation of probes for ISH. T.K. was funded by the Japan Society for the Promotion of Science (JSPS): Challenging Exploratory Research.

ABBREVIATIONS APa archipallium BAM Brn2

Ascl1, and Myt1l BAMN Brn2, Ascl1, Myt1l, and NeuroD1 BG basal ganglia BrdU bromodeoxyuridine CA1, CA3 cornu ammonis fields CBL cerebellum CNS central nervous system CP cortical plate cSp cervical spinal cord CTX cortex DG dentate gyrus DIG digoxigenin dpc days post coitum DRG dorsal root ganglion DTe dorsal telencephalon EGL external germinal layer of the cerebellum fIC fibers of the internal capsule HC hippocampus HT hypothalamus IC internal capsule iCo inferior colliculus iNBL inner neuroblastic layer of the neural retina Is isthmus ISH in situ hybridization IXg glossopharyngeal ganglia IZ intermediate zone LDN lateral deep nucleus LGE lateral ganglionic eminence Me medulla MGE medial ganglionic eminence ML mantle layer MN motor neuron Myt myelin transcription factor NZF neural zinc finger oNBL outer neuroblastic layer of the neural retina OpV optic vesicle OtV otic vesicle Pal pallidum PBS phosphate-buffered saline POA preoptic area PP preplate Ps pons sCo superior colliculus SVZ subventricular zone Str striatum sXg superior ganglion of the vagus nerve Teg tegmentum TH thalamus UTR untranslated region Vg trigeminal ganglion VIIg facial ganglion VIIIg vestibulocochlear ganglion VII–VIII facio-acoustic neural crest complex VTe ventral telencephalon VZ ventricular zone WM white matter Xg vagus ganglia.

Ancillary