Alzheimer's disease (AD) is a neurodegenerative disease that results in loss of neurons in hippocampus and cerebral cortex. During embryonic development, the most anterior neuroectoderm gives arise to the forebrain. The forebrain then divides into anterior and posterior regions, which are referred to as the telencephalon and diencephalon. The telencephalon is comprised of two major subdivisions, the dorsal and ventral telencephalon (Miyake et al.,2005; Wilson and Houart,2004), and the dorsal telencephalon gives rise to the cerebral cortex. Subsequent to the establishment of the cerebral cortex, additional subdomains are divided (Zilles and Wree,1995).
Early-onset forms of AD are caused by mutation in APP, presenilin1, and presenilin2 (Glenner and Wong,1984; Hardy,1997; Bossy-Wetzel et al.,2004). X11L protein is known to play a role in regulating the amyloidogenic pathway of APP metabolism in neuronal tissues and contributes to the pathogenesis of AD (Lee et al.,2000; Sumioka et al.,2003; Tomita et al.,1999). XB51 (X11L-binding protein of clone number 51) was isolated as a binding protein to the N-terminal of X11L protein. The interaction of XB51 with X11L inhibits the suppression of Aβ40 production (Lee et al.,2000; Sumioka et al.,2003). However, the function of XB51 is unclear in neuronal cell determination or differentiation during early brain development.
The zebrafish, Danio rerio, is a powerful vertebrate model system for the study of neural development due to the availability of a large number of embryos as well as optical clarity of CNS tissues. Therefore, we tried to examine the structure and expression pattern of a zebrafish homologue of XB51 gene. Our data show that the expression of zebrafish xb51 is restricted to the dorsal telencephalon and overlaps with the expression domains of several neuronal markers for differentiation at the early stages of zebrafish embryo development.
RESULTS AND DISCUSSION
Identification of Full-Length Zebrafish xb51
Using the human XB51 sequence (AB039947), we identified a zebrafish xb51 homologue in the zebrafish genomic and expressed sequence tags (EST) databases. The zebrafish xb51 gene (DQ017379) is composed of 14 exons and encodes a protein with 348 amino acids, having 66 and 71% similarity to human and mouse XB51 proteins, respectively (Fig. 1). It contains a Ca2+ binding domain, EF-hand motif, in the N-terminus and the ABM (antibiotic biosynthesis monooxygenase) domain in the C-terminus. The EF-hand domain is highly conserved from human to zebrafish. XB51 is also known as NECAB3, one of the neuronal Ca2+-binding proteins (Sugita et al.,2002). Although many EF-hand type Ca2+-binding proteins contain at least two EF-hands, XB51 contains only a single EF-hand domain.
Expression of xb51 in the Developing Forebrain
To examine the temporal and spatial expression of xb51 during zebrafish development, whole-mount in situ hybridization was employed by using digoxigenin-labeled anti-sense RNA probe. The zebrafish xb51 is expressed relatively late and transcripts were first detected bilaterally in dorsal telencephalon at 22 hpf (Fig. 2A,B). Expression of xb51 in the dorsal telencephalon increased at 24 hpf and persisted to 36 hpf (Fig. 2C–F). At 48 hpf, the expression of xb51 transcripts was expanded to cells of the external layer of the diencephalon (Fig. 2G–I) containing differentiated cells. This expression pattern was maintained in the 72-hpf embryo (Fig. 2J). In retina, xb51 transcripts were expressed in the inner nuclear layer at 72 hpf (Fig. 2K.L).
To identify the xb51-expressing cells, we performed two-color in situ hybridization with several markers. At 24 hpf, the expression domain of xb51 overlapped partially with that of emx1, a dorsal telencephalon marker (Fig. 3A,B), but not with that of dlx2, a ventral telencephalon and diencephalon marker (Fig. 3C,D). The xb51 expression domain partially overlapped with some lim1-expressing cells in the telencephalon (Fig. 3E). However, the xb51-positive cells were clearly excluded from the expression domain of islet1, neurod4/zath3, and neurogenin1, which are the markers for neuronal precursors at 24 hpf (Fig. 3F–H).
To test the possibility that xb51 may be involved in neuronal differentiation rather than determination, we further compared the expression of xb51 with that of elavl3/huC and her4. elavl3/huC encodes a RNA binding protein and serves as an early marker for differentiating neurons (Kim et al.,1996; Park et al.,2000). her4, which is a marker for proliferating neural precursors in the central nervous system, encodes a hairy/Enhancer-of-split-related transcription factor, a target of Notch signaling (Takke et al.,1999; Yeo et al.,2007). xb51-expressing cells were included in the expression domain of elav/3/huC, but did not overlap with the her4-expressing domain (Fig. 3I, J, M, N). To investigate the expression profile in detail, we performed cryostat section after two-color in situ hybridization (Fig. 3K, L, O, P). These results suggest that xb51-expressing cells are differentiated neurons in the forebrain.
xb51 Is Expressed in Neuronal Cells in the Dorsal Telencephalon
To further characterize the neurons that express xb51 in the forebrain, we compared the expression pattern of xb51 with the positions of primary axon tracts using an anti-acetylated α-tubulin antibody. This monoclonal antibody labels primary axon tracts in the developing zebrafish brain (Fig. 4B, E) (Chitnis and Kuwada,1990). The xb51-expressing cells are very close to the region from where the anterior commissural axons originate (Fig. 4C, F). This observation is consistent with the idea that xb51-expressing cells are differentiated neurons in the dorsal telencephalon.
We next examined xb51 expression in the mind bomb (mib) mutant embryos that are defective in Notch signaling. The mib mutant displays a neurogenic phenotype characterized by an excessive number of neurons and depleted progenitors in the CNS (Itoh et al.,2003). To determine whether the number of xb51-expressing cells is altered in mib, we performed whole-mount in situ hybridization at 24 and 36 hpf. Previously, we showed that the number of elavl3/huC-positive cells increased in the mib mutant, while that of her4-positive cells decreased, since progenitor cells have differentiated into neurons prematurely (Itoh et al.,2003). Therefore, we expected that xb51-expressing cells are increased in mib embryos. We found that the bilateral expression domains of xb51 in the mib mutant fused across the midline at 24 hpf (Fig. 4J), compared to the bilateral expression in wild type embryo (Fig. 4I). Furthermore, ectopic xb51-positive cells were detected in the midbrain of the mib mutant at 36 hpf (Fig. 4L). We reasoned that ectopic expression of xb51 may be caused by premature differentiation of the neuronal precursor cells in the mib mutant.
Taken together, our results suggest that xb51 is specifically expressed in a subset of differentiated neurons in the dorsal telencephalon. Since the dorsal telencephalon is known as an evolutionary ancestor of the cerebral cortex and the neurons of the hippocampus and cerebral cortex are selectively lost in AD patients (Bossy-Wetzel et al.,2004), our observations demonstrate that the basic molecular components involved in AD are preserved in evolution between teleosts and humans. This suggests that the zebrafish could be used as a model species for studies of molecular mechanisms involved in AD.
Fish Maintenance and Mutants
Zebrafish embryos were obtained from natural spawning and cultured at 28.5°C in 1/3 Ringer's solution (39 mM NaCl, 0.97 mM KCl, 1.8 mM CaCl2, and 1.7 mM Hepes at pH 7.2). The developmental stages of the embryos were determined by the hours post fertilization (hpf) and by morphological features as described (Kimmel et al.,1995). The wild-type strain came from a local pet shop. The mutant strain used was a mind bombta52b allele (Itoh et al.,2003). Appropriate stages of the embryos were fixed with 4% paraformaldehyde in PBS.
Cloning of Zebrafish XB51 Homologue
Several zebrafish EST clones with a high homology to XB51 were identified in the NCBI database. Total RNA was isolated from wild-type zebrafish embryos at 48 hpf using TRIZOL (MRC, Inc.) and reverse-transcribed. Full-length xb51 cDNA was generated by PCR amplification using the forward primer 5′-GCATTAGAGCAACAC ACAGCAGAA-3′ and the reverse primer 5′-GATAACCTCTACAGTTTCTACTACT-3′, subcloned into the pGEM T-easy vector, and sequenced by an automatic sequencer.
Whole-Mount In Situ Hybridization
To examine the spatiotemporal expression patterns of xb51, in situ hybridization was performed as previously described (Kim et al.,1996). Antisense RNA probe was synthesized from the plasmid containing the full-length cDNA insert. The plasmid DNA was linearized with BamHI before in vitro transcription with T7 RNA polymerase to generate a digoxigenin (DIG)-labeled antisense RNA probe (Roche Applied Sciences). Hybridization and detection with an anti-DIG antibody coupled to alkaline phosphatase (Vector, CA) was performed with fixed zebrafish embryos. To compare expression domains of different genes in single embryos, two-color in situ hybridization was performed with fluorescein- or digoxigenin-labeled probes in parallel. Briefly, antisense probes were synthesized from the linearized plasmids for emx1 (Morita et al.,1995), dlx2 (Akimenko et al.,1994), lim1 (Toyama and Dawid,1997), islet1 (Inoue et al.,1994), neurod4/zath3 (Park et al.,2003), ngn1 (Kim et al.,1997), elavl3/huC (Kim et al.,1996), and her4 (Takke et al.,1999; Yeo et al.,2007) cDNA. Embryos were hybridized with both digoxigenin- and fluorescein-labeled riboprobes. Anti-DIG AP was used to detect the first probe and then the phosphatase was inactivated by incubating in 10 mM EDTA (in PBST, pH 5.5) at 72°C for 20 min. The fluorescein-labeled probe was detected using anti-fluorescein AP and the INT/BCIP substrate (Roche Applied Sciences).
The whole mount antibody staining was performed as previously reported (Yeo et al.,2001) using a monoclonal anti-acetylated alpha-tubulin (Sigma) at 1:1,000 dilutions. Briefly, embryos were incubated with primary antibody in blocking solution overnight at 4°C. Then samples were washed with PBST (phosphate-buffered saline, 0.1% Tween 20) 4 times. Samples were incubated with secondary antibody in blocking solution overnight and washed several times with PBST. Embryos were incubated for 2 hr at room temperature in Vectastain Elite ABC reagent (Vector) prepared by the manufacturer's recommendation and washed with PBST. Embryos were then incubated in 1 ml DAB solution (1% DMSO, 0.5 mg/ml diaminobenzidine, and 0.03% H2O2 in 0.1 M PO4 buffer, pH 7.3) at room temperature. For fluorescent dye labeling, embryos were incubated with rhodamine-conjugated secondary antibody.
After whole-mount in situ hybridization, the stained embryos were dehydrated with a graded ethanol series to 100%. Specimens in xylene were embedded in paraffin and cut at 8-μm thickness. For cryostat section, stained embryos were embedded in an agar-sucrose block. Blocks were cryosectioned at 8-μm thickness using a cryostat microtome (Leica).
We thank Dr. Yong I. Cha for a critical reading of the manuscript.