Generation of a transgenic zebrafish model of Tauopathy using a novel promoter element derived from the zebrafish eno2 gene

Abstract

The aim of this study was to isolate cis-acting regulatory elements for the generation of transgenic zebrafish models of neurodegeneration. Zebrafish enolase-2 (eno2) showed neuronal expression increasing from 24 to 72 h post-fertilization (hpf) and persisting through adulthood. A 12 kb eno2 genomic fragment, extending from 8 kb upstream of exon 1 to exon 2, encompassing intron 1, was sufficient to drive neuronal reporter gene expression in vivo over a similar time course. Five independent lines of stable Tg(eno2 : GFP) zebrafish expressed GFP widely in neurons, including populations with relevance to neurodegeneration, such as cholinergic neurons, dopaminergic neurons and cerebellar Purkinje cells. We replaced the exon 2-GFP fusion gene with a cDNA encoding the 4-repeat isoform of the human microtubule-associated protein Tau. The first intron of eno2 was spliced with high fidelity and efficiency from the chimeric eno2-Tau transcript. Tau was expressed at approximately 8-fold higher levels in Tg(eno2 : Tau) zebrafish brain than normal human brain, and localized to axons, neuropil and ectopic neuronal somatic accumulations resembling neurofibrillary tangles. The 12 kb eno2 promoter drives high-level transgene expression in differentiated neurons throughout the CNS of stable transgenic zebrafish. This regulatory element will be useful for the construction of transgenic zebrafish models of neurodegeneration.

Figure 1.

The zebrafish eno2 orthologue. (A) Four zebrafish enolase genes identified by BLAST search were located on chromosomes 2, 19, 22 and 23. The predicted amino acid sequences encoded by each of these genes were aligned with the human, rat and mouse α-, β- and γ-enolase sequences using the AlignX implementation of the ClustalW algorithm. The dendrogram shown was generated using the neighbour joining method of Saitou and Nei (34). The inferred evolutionary relationships between the zebrafish genes and their mammalian orthologues suggest that zebrafish γ-enolase is encoded by the chromosome 19 gene locus, to which we refer as ‘eno2’ throughout the remainder of the article. (B) An alignment of zebrafish γ-enolase with human γ-enolase is shown. Non-similar amino acid substitutions are shaded black and conservative substitutions are shaded grey.

Figure 2.

The expression pattern of zebrafish eno2. A digoxigenin labelled cRNA probe complementary to the 5′ segment of the eno2 open reading frame and 5′UTR was used to examine the expression pattern of eno2 in adult, embryonic and larval zebrafish by northern blot hybridization and RNA in situ hybridization. (A) Northern blot adult tissues. Total RNA (1.5 µg/lane) derived from adult brain (B), gut and liver (G) and muscle (M) was separated by denaturing gel electrophoresis and the 18S rRNA band photographed (middle panel) to confirm equal sample loading before northern transfer. The blot was probed sequentially with the eno2 cRNA probe (upper panel) and a probe to β-actin (lower panel). Bound probe was detected using a light-emitting substrate and exposure to photographic film. (B and C) RNA in situ hybridization, adult brain. Parasagittal sections of adult zebrafish brains (rostral left, dorsal up) were hybridized with the eno2 antisense cRNA probe (B, upper panel and inset), a control sense eno2 cRNA probe (B, lower panel) or a cRNA probe to an oligodendrocyte marker, mpz (C). Hybridized probe was detected using a colorigenic substrate producing a purple reaction product. The morphology of eno2-expressing cells was examined at high power (B, inset panel). The topographical locations of cells expressing eno2 were identified as illustrated in the accompanying schematic (middle panel). Key: Dl, dorsal telencephalon, lateral zone; Dp, dorsal telencephalon, posterior zone; PGZ, periventricular grey zone of optic tectum; TeV, tectal ventricle; TeO, optic tectum; TS, semicircular torus; Val, valvula cerebelli; G, granule cell layer of cerebellum; P, Purkinje cell layer of cerebellum; CCe, corpus cerebellaris; Lca, caudal lobe of cerebellum; LX, vagal lobe; SRF, IMRF, IRF, superior, intermediate and inferior reticular formation; MLF, medial longitudinal fascicle; LR, lateral recess of diencephalic ventricle; Hd, dorsal zone of periventricular hypothalamus; DIL, diffuse nucleus of the inferior lobe; ON, optic nerve (11). (D) Northern blot, embryo and larval samples. Total RNA (1.5 µg/lane) from whole embryo/larvae lysates at 24–216 hpf was separated by denaturing gel electrophoresis and the 18S rRNA band photographed (lower panel) to confirm equal sample loading before northern transfer. The blot was probed with the eno2 antisense cRNA probe under identical conditions to (A). (E) Whole mount in situ hybridization. AB* zebrafish larvae were fixed and hybridized with the eno2 antisense cRNA probe (right) or a sense control eno2 cRNA probe (left). Hybridized probe was detected using the purple colorigenic substrate shown in (B). Lateral views are shown of zebrafish larvae (rostral left, dorsal up) at 24, 48 and 72 hpf. The arrowheads indicate weak hybridizing signal in the developing nervous system of the 24 hpf embryo.

Figure 3.

The zebrafish eno2 gene and promoter. (A) 5′RACE was used to map the eno2 transcription start site. An ethidium bromide stained 1.8% agarose gel is shown. Lane 1: size marker—annotations show size in base pairs. Lanes 2 and 3: first strand cDNA was derived from reverse transcription of RNA that was either treated with TAP (lane 2, TAP⁺) or untreated (lane 3, TAP⁻), prior to RACE adapter ligation. The cDNA was then amplified by PCR using a RACE adapter 5′ primer and a gene-specific eno2 3′ primer. Lanes 4 and 5: TAP⁺ (lane 4) and TAP⁻ (lane 5) first strand cDNA samples, identical to those used as templates for the reactions shown in lanes 2 and 3, were amplified by PCR using two eno2 gene-specific primers. (B) The genomic sequence encompassing the eno2 promoter. The four transcriptional start sites delineated by 5′RACE are labelled with arrows (+1, +12, +22, +28). Numbering is shown with respect to the most 5′ start site. Exon sequence is shown in upper case; 5′ flanking sequence and intron sequence is shown in lower case. The boundary between exons 1 and 2 at position +73 is shown; the splice donor consensus is underlined. The ATG translational initiation codon in exon 2 is underlined, and the obligate A residue of the Kozak consensus at position −3 with respect to the ATG codon is highlighted in black. (C) The schematic illustrates the genomic organization of eno2 and the DNA constructs used to generate the transgenic zebrafish shown in Figures 4–8.

Figure 4.

Expression of a 12 kb eno2 promoter construct in transgenic zebrafish. The 12 kb eno2:GFP construct shown in Figure 3C was microinjected into single-cell zebrafish embryos and stable transgenic lines were derived as described in the text. (A) Lateral view of the head and rostral trunk region of a microinjected zebrafish larva (rostral left, superior up) at 72 hpf by epifluorescence microscopy, showing punctate mosaic GFP expression. Key: Tel, telencephalon; TeO, optic tectum; MdO, medulla oblongata; SC, spinal cord; R, retina; L, lateral line ganglion. The yolk sac, Y, is autofluorescent. (B) Transgenic founders were identified by transmission of the GFP transgene to progeny embryos by genomic PCR. The picture shows an ethidium bromide stained agarose gel. Lanes 2 and 4, genomic DNA from embryos derived from transgenic parents; lanes 1 and 3, genomic DNA from embryos derived from non-transgenic parents; lane 5 (‘–c’), genomic DNA from wild-type embryos; lane 6 (‘+c’), genomic DNA from wild-type embryos spiked with GFP plasmid. (C) Northern blot analysis of GFP expression in stable transgenic lines. Total RNA (1.5 µg/lane) derived from adult brains of each transgenic line was subjected to denaturing gel electrophoresis and northern transfer and the resulting blot hybridized with a probe to eno2 that is complementary to sequences within both the endogenous eno2 transcript and the eno2-GFP transgene mRNA. The asterisk indicates the unexpected additional hybridizing band detected in line Pt405 and discussed in the text. (D) GFP expression in living stable Tg(eno2:GFP)^Pt401 zebrafish larvae by epifluorescence microscopy. An oblique lateral view of an F3 generation Tg(eno2:GFP)^Pt401 larva is shown at 5 days post-fertilization, compared with a wild-type AB* larva (both larvae are orientated rostral left, dorsal up). The yolk sac, Y, is autofluorescent in both larvae. The major GFP-expressing divisions of the nervous system are labelled; the key is the same as (A). (E) Lateral confocal microscopic image of the spinal cord and trunk of an F3 generation Tg(eno2:GFP)^Pt404 larva at 5 days post-fertilization. GFP expression is seen within the segmental spinal nerve roots (arrows) and in cell bodies within the spinal cord (bracketed). (F) High-power view of spinal cord, showing GFP expression in individual neurons in the spinal grey matter and axons in the ventral funiculus. Occasional GFP-expressing axons (arrow) were seen entering the ventral funiculus from the grey matter.

Figure 5.

GFP expression in the brains of adult Tg(eno2:GFP) zebrafish. (A) Parasagittal sections of Tg(eno2:GFP)^Pt404 (upper panel) and wild-type (lower panel) brain were labelled with an antibody to GFP. Bound antibody was detected using a histochemical reaction with a red product and nuclei were labelled with a blue counterstain. GFP expression was apparent throughout the brain of transgenic adults. (B) High-power view of medullary reticular formation of Tg(eno2:GFP)^Pt404 zebrafish brain, showing GFP expression in cells with neuronal morphology. The arrows show GFP-expressing neurons with axons in the plane of the section. (C–F) Double label confocal images of Tg(eno2:GFP)^Pt404 adult brains. Each panel consists of a set of three images: (i) Upper image; red: a cell type-specific marker of interest was localized using specific antibodies. C: Medulla; ChAT, choline acetyltransferase; cholinergic neurons; D: Olfactory bulb; TH, tyrosine hydroxylase; dopaminergic neurons; E: cerebellum; IP3R1, IP3 receptor 1; cerebellar Purkinje cells; D: thalamus; GABA, γ-amino butyric acid; inhibitory neurons; (ii) Middle image, green: GFP expression was localized in the same sections using a GFP antibody; (iii) Lower image: the merged images show co-localization of GFP and the cell type-specific marker (yellow), and were counterstained with DAPI to show nuclei (blue). The scale bar (10 µm) for each set of images is shown in the lower panel.

Figure 6.

GFP expression outside the central nervous system of Tg(eno2:GFP) zebrafish. (A) Tg(eno2:GFP)^Pt401 larvae at 28 and 72 hpf were subjected to RNA in situ hybridization using a cRNA probe complementary to GFP. The arrowheads indicate GFP mRNA expression in the developing fins in the 28 hpf embryo, and in the distal margins of the pectoral fins in the 96 hpf larva. (B) High-power view of the trunk region of a Tg(eno2:GFP)^Pt404 larva at 96 hpf. Key: SC, spinal cord; G, gut. The arrowheads indicate weak expression of the transgene in the developing intestine. (C) High-power view of the trunk region of a Tg(eno2:GFP)^Pt405 larva at 96 hpf. Key: SC, spinal cord; F, fins. The arrowheads indicate ectopic expression of the transgene in myoseptae. (D) Northern blot analysis of tissues derived from adult Tg(eno2:GFP)^Pt404 and Tg(eno2:GFP)^Pt401 zebrafish. Key: B, brain; M, muscle; G, gut; T, testis. Two micrograms total RNA from each tissue was subjected to denaturing electrophoresis and northern transfer. The blot was probed using a cRNA probe to GFP and exposed for 60 s (upper panel) or 5 min (middle panel). The blot was then re-probed with a cRNA probe to β-actin (lower panel). (E) Epifluorescence microscopy of an adult Tg(eno2:GFP)^Pt404 zebrafish. High-power views are shown of the caudal peduncle region (upper panel) and distal anal fin (lower panel). Arrowheads indicate GFP expression in sensory nerves (upper panel) and distal fin margins (lower panel).

Figure 7.

Splicing of intron 1 from a chimeric eno2:Tau transcript in vivo. Stable transgenic zebrafish were generated using the eno2:Tau transgene shown in Figure 3C. (A) Two micrograms of total RNA derived from Tg(eno2:Tau)^Pt406 (‘Tau’) or wild-type (‘AB*’) adult zebrafish brains was separated by denaturing gel electrophoresis and the 18S rRNA band photographed (lower panel) to confirm equal sample loading before northern transfer. Duplicate blots were probed using digoxigenin-labelled cRNA probes complementary to GFP (left blot) or eno2 (right blot), as shown in Figure 2A. The endogenous eno2 and chimeric eno2:Tau transcripts are labelled. (B) An ethidium stained 1.8% agarose gel is shown; lane 1 contains DNA marker, annotations show size in base pair. Total RNA, derived from Tg(eno2:Tau)^Pt406 (lane 2, ‘Tau’) or wild-type (lane 3, ‘AB*’) adult zebrafish brains, was subjected to reverse transcription followed by PCR amplification using a 5′ primer specific for eno2 exon 1 and a 3′ primer specific for human 4R-Tau (upper panel). Aliquots of the same cDNA samples were then subjected to PCR amplification using two β-actin primers (lower panel). (C) The eno2-Tau PCR product from (B) was cloned and sequenced. The schematic shows the exon 1 splice donor and exon 2 splice acceptor sequences from the endogenous eno2 gene, the chimeric eno2:Tau splice acceptor, and the sequence of the eno2-Tau RT-PCR product across the boundary between the eno2 and human Tau sequences.

Figure 8.

Expression of 4R-Tau in the brains of adult Tg(eno2:Tau) zebrafish. (A) Ten micrograms of protein derived from wild-type AB* zebrafish brain, normal human post-mortem cerebral cortex or Tg(eno2:Tau)^Pt406 zebrafish brain was separated by SDS–PAGE and transferred to a PVDF membrane. The resulting western blot was probed sequentially with an antibody specific to human Tau (upper panel) and an antibody that binds to both human and zebrafish actin (lower panel). (B) Sections from wild-type AB* (upper two panels) or Tg(eno2:Tau)^Pt406 (lower two panels) zebrafish brains were immunolabelled using the human Tau-specific antibody shown in (A). Bound antibody was detected using a peroxidase-conjugated secondary antibody and a histochemical reaction with a red reaction product, as shown in Figure 5A. Nuclei were counterstained blue. The left two panels show low-power views of the optic tectum to demonstrate the regional pattern of Tau expression in axons and neuropil. The right two panels show high-power views of the thalamus to demonstrate expression of Tau in the cell bodies and proximal processes of neurons, which are identified by their large pale-staining nuclei.