Characterization of gli2 2b genes in zebrafish hindbrain development 2

Isolation of full length zebrafish gli2b cDNA Using a pair of degenerate primers against the conserved zinc finger region of the mouse Gli genes Hui et al, 1994, the PCR fragment of abo

Chapter III

RESULTS

III RESULTS

3.1 Characterization of zebrafish gli2b gene

3.1.1 Isolation of full length zebrafish gli2b cDNA

Using a pair of degenerate primers against the conserved zinc finger region of the

mouse Gli genes (Hui et al, 1994), the PCR fragment of about 400 bp was previously

amplified using the zebrafish embryonic cDNA as a template Sequencing analysis of the PCR fragment revealed three distinct sequences which were identified to be

zebrafish homologs of gli genes Another pair of degenerate primers, GliF and GliR, was designed based on the sequences of these three gli sequences (Fig 3-1,I) PCR

was carried out using an aliquot of 24 hpf embryonic cDNA library as a template In

this round of amplification, several distinct zebrafish gli-like sequences were isolated

and one fragment was then used to screen a zebrafish embryonic cDNA library and a

cDNA clone (~2Kb) containing incomplete ORF corresponding to a novel gli gene of

zebrafish was obtained (Lim, 2000) However, further attempts to acquire the full length cDNA clone were not successful at that time, probably due to the poor quality

of cDNA library To acquire a full-length cDNA clone, a fresh double stranded cDNA library using SMART TM RACE cDNA Amplification kit (Clontech) was constructed

Both 5’ and 3’ rapid amplification of cDNA ends (RACE) PCR were conducted using the two gene-specific primers based on the partial clone and the universal primers

from the kit and both 5’ and 3’ ends of the novel gli cDNA were obtained using this method (Fig 3-1,II) Finally, the complete novel gli cDNA sequence was amplified

by RT-PCR using a new pair of primers (Gli2bF/Gli2bR) based on the each end of deduced sequence and cloned into pGEMT easy vector (Promega) (Fig 3-1,II)

(I)

Gli#6 AAGGACTGTTCCCGAGAGCAGCGTCCGTTTAAAGCGCAGTACATGCTGGT 100 Gli#5 gt -g c t -a -aag t c g c -t 100 Gli#4 g-c g c a c -aag c c - 100

Gli#6 GGTGCACATGCGCAGACACACCGGAGAAAAACCACACAAGTGCACTTTTG 150 Gli#5 -t-c -ttc -t g g g c t -a 150 Gli#4 -c - 150 Gli#6 AAGGCTGTAATAAAGCGTACTCGCGGCTGGAGAACCTGAAGACGCATCTG 200 Gli#5 t ctcg g t t t a -a t c a t -t 200 Gli#4 - 200 Gli#6 CGCTCACACACCGGAGAGAAACCCTACGTTTGTGAACACGAGGGCTGTAA 250 Gli#5 a-g c -a g -t -a c g - 250 Gli#4 - 250

Gli#6 CAAGGCTTTCTCTAATGCTTCAGATCGAGCAAAGCACCAGAACAGAACAC 300 Gli#5 -gc t c c -c g c -cc-c-c g- 300 Gli#4 - 300

(II)

Fig 3-1 Isolation and cloning of full length zebrafish gli2b cDNA clone (I) A pair

of degenerate primers used for screening the cDNA library (modified from Lim, 2000) (II) Schematic representation of 5’ RACE and 3’ RACE PCR amplification of

both ends of the gli2b cDNA and the isolation of full length gli2b cDNA

3.1.2 The novel gli cDNA encodes a second Gli2 of zebrafish

The complete novel gli clone is 4,730 bp long (GenBank access number

AY928397) and encodes a polypeptide of 1363 amino acids (Fig 3-2) It shares 95.1-96.9% identity in zinc-finger domains and 50.5-62.5% identity for overall amino acid sequence with Gli2 The deduced amino acid sequence shares the highest identity with zebrafish Gli2 (96.9% in zinc-finger domain and 62.5% for overall amino acid sequence) In comparison, it has only 44% identity with Gli3 proteins and 35%

identity with Gli1 proteins Thus, this novel gli gene was named as gli2b

As shown in Fig 3-3, the sequence alignment of mouse Gli2 proteins indicated that the sequence similarity is high in the N-terminal, including the five zinc-finger domains and six putative Protein Kinase A binding domains However, the proposed cyclic-AMP response element binding protein (CBP) binding site (Hughes et al, 1997) and the herpes simplex viral protein 16 (VP-16) -like activator binding domain (Yoon

et al,1998) of Gli2b are relatively divergent comparing to Gli2 suggesting divergence

of functions of these two proteins

3.1.3 Phylogenetic analysis of gli2b

To determine the phylogenetic relationship among members of the Gli family, the amino acid sequences of the zinc finger domain and its surrounding regions were analyzed by the software for phylogenetic analysis, PAUP 3.1 As shown in Fig 3-4, there are three subfamilies of Gli proteins (Gli1, Gli2 and Gli3) in vertebrates A phylogenetic analysis of selected vertebrate Gli protein sequences indicated that zebrafish Gli2b was clustered with all other vertebrate Gli2 proteins and was the closest to zebrafish Gli2 In parallel, it seems that Gli2 and Gli3 subfamilies are more

5’ RACE

/Gli2bF

R1

3’ RACE

Fig 3-2 The complete nucleotide sequence of the zebrafish gli2b cDNA and

deduced amino acid sequence The position of nucleotides and amino acid residues

is indicated on the left The primers (MAF/MAR) used for gene mapping are

indicated by arrows The black box indicates the sequence of gli2b MO used to block

Gli2b translation The five zinc finger domain is boxed R1 is 3’ primer used for

examining efficiency of the gli2b splicing MO (see section 3.4.1) Gli2bF/Gli2bR

and 5’RACE and 3’RACE represent primers indicated in Fig.3-1

Gli2bR

96

Fig 3-3 Alignment of the zebrafish Gli2b with zebrafish (z) and mouse (m) Gli2

proteins Identical amino acid residues are highlighted either in black (identical in all

three sequences) or in grey (identical in two sequences) Dots represent gaps inserted for maximal alignment, and numbers of amino acid residues in each sequence are indicated on the right The specific domains and motifs are indicated by lines or boxes: solid line with Roman numbers, five zinc-finger domains; dashed line, phosphorylation site of protein kinase A; solid box, potential CBP (cyclic AMP response element binding protein) binding site; and dashed box, VP-16 activator-like domain

Fig 3-4 The phylogenetic tree of vertebrate Gli family comprising Gli1, Gli2

and Gli3 The phylogenetic tree of selected vertebrate Gli proteins was constructed

using ClustalW Species are defined by abbreviated prefixes added to to the protein

names: h, human; m, mouse; x, Xenopus; z, zebrafish The sequences were obtained

from the GenBank and their access numbers are: hGli1, P08151; hGli2, P10070; hGli3, P10071; mGli1, BAA85004; mGli2, AAH31171; mGli3, Q61602; xGli1, Q91690; xGli2, AAD28180; xGli3, Q91660; xGli4, Q91661; zGli1, NP_840081;

zGli2, NP_571042; zGli2b, AAX23613; zGli3, NP_991291

3.1.4 Genome mapping of gli2b

Genome mapping was performed by using the Goodfellow T51 zebrafish-hamster somatic cell radiation hybrid DNA panel (Research Genetics, USA) (Kwok et al, 1998; Geisler et al, 1999) Two pairs of PCR mapping primers were used: one for amplification of the 5’UTR (MAF, 5’-CTCACCGCAAACAGTCGCGCTC; MAR: 5’-GAAGAGTGAACCAAAGCTCGC) (as indicated in Fig 3-2); the other targeted one exon and one intron for amplification of a region before the zinc-finger domain

based on gli2b genomic sequence (Contig 9410) obtained from the zebrafish genome

database (http://www.sanger.ac.uk/Projects/D_rerio/WebFPC/zebrafish/) (MBF, 5’- ACCTGAAGGAGCGAGGGTTG; MBR, 5’-CCAGACATAGAGTTCAAGCCA) (Fig 3-5) 94 hybrids were screened by PCR in triplicates and results submitted online

to the Tuebingen Zebrafish Genome Map website at

http://wwwmap.tuebingen.mpg.de gli2b was mapped to position 2,086 cR of the

linkage group (LG) 11, very close to the marker z10727 (Fig 3-6a) Examination of

the zebrafish genetic map revealed that a gli gene (GeneID:30097) was mapped previously to this position and is likely to represent the same gli sequence To date, no

known mutant was mapped to this region

Previously, the zebrafish gli2 has been mapped previously to LG9 (Karlstrom et

al, 1999) The synteny analysis indicated conservation between zebrafish LG9

(gli2)/LG11 (gli2b) and the Gli2 region of mouse chromosome 1 (Fig 3-6b) Thus,

several lines of evidence, including the sequence and phylogenetic analyses as well as

the genome mapping data support the idea that gli2b is a second gli2 ortholog of the zebrafish Our data also indicated that a region of zebrafish LG11 containing gli2b is

100

Fig 3-5 gli2b genomic sequence used to design primers for gene mapping A

partial sequence from ctg9410 was acquired from the genome database The exon

region (blue font) is aligned with gli2b cDNA (black font, from 748 nt to 926 nt)

The following intron region is shown in red Arrows indicate the primers (MBF and

MBR) designed for genome mapping

Fig 3-6 Mapping of zebrafish gli2b and synteny analysis (A) Mapping of

zebrafish gli2b and comparison with the mouse chromosome region containing Gli2

The distances between markers in zebrafish linkage group 11 (LG11) are indicated in centimorgans (cM) or centirays (cR) Gene names are in italic, expressed sequence tags are in gray, and other mapping markers begin with the prefix z Dash lines link conserved syntenic groups on zebrafish LG11 and their mouse counterparts on mouse chromosome 1 (B) Conserved synteny between zebrafish (LG11 and LG9) and

mouse chromosomes (Mmu1) containing gli2 genes

B

A

102

Fig 3-7 Temporal expression of gli2b as defined by RT-PCR A comparison of

temporal expression of gli2b, gli2, and shh Total RNA was prepared from embryos

at selected embryonic stages from one-cell stage to 48 hours postfertilization stage The one-step reverse transcriptase-polymerase chain reaction was performed to 25 cycles to avoid amplification saturation

3.2 Expression analysis of gli2b in zebrafish

3.2.1 Temporal expression of gli2b in zebrafish embryos

Temporal expression of gli2b was first compared with that of gli2 and sonic

hedgehog (shh) by unsaturated reverse transcriptase-polymerase chain reaction

(RT-PCR) analyses (25 cycles; Fig 3-7) Both gli2b and gli2 transcripts are present as maternal mRNAs, but the level of gli2b mRNA at the one-cell stage appeared to be relatively higher than that of gli2 After this initial phase, both gli2b and gli2 mRNAs

decreased during the late gastrulation, followed by a major increase around 10 hpf at

the beginning of neurulation As the expression of shh and tiggy-winkle hedgehog (twhh) starts at around 50–60% epiboly (Ekker et al, 1995), the early gli2b expression

is unlikely to be required within a context of the Hh signaling; thus, Gli2b may have

an early Hh-independent role at the blastula stage Given the fact that Gli2 acts as a repressor in absence of Hh signal, it is possible that during early development Gli2b acts to keep targets of Hh signaling suppressed Whether this is a case is unclear and further analysis is needed for this idea to be proved Alternatively, it could be stored

in preparation for Hh activation at 5-6 hpf Also, the increase of gli2b expression at the beginning of neurulation indicated a requirement for gli2b in early neurogenesis

3.2.2 Comparison of expression patterns of the two gli2s of zebrafish during

early development

The spatial expression of gli2b during zebrafish embryogenesis was examined and compared with that of gli2 by whole-mount in situ hybridization (WISH) (Fig 3-8) At 3 hours post fertilization (hpf), most of gli2b mRNA was detected in cell

nuclei (Fig 3-8A) and later in the cytoplasm (6 hpf, Fig 3-8B), indicating a temporal

regulation of gli2b mRNA nuclear export, which could be important for initiation of the Gli2b early developmental function By the end of gastrulation, gli2b transcripts

were mostly found in the anterior neural keel (Fig 3-8C), excluding the region corresponding to the zona limitans intrathalamica (ZLI) This finding is in a sharp

contrast to expression pattern of gli2, which transcripts were detected mostly in the

lateral mesoderm with only a single patch in the neural keel corresponding to the

midbrain (Fig 3-8E) By 18 hpf, expression of gli2b was intense in the anterior neural

tube except for the ZLI and the midbrain–hindbrain boundary (MHB; Fig 3-8D) At

24 hpf, expression patterns of gli2b and gli2 still differ substantially gli2b transcripts

were present in the ventricular zone of telencephalon, posterior thalamus, optic tectum (Fig 3-8F), and hindbrain, including the cerebellum (Fig 3-8G) Cross-sections at the

hindbrain level showed that gli2b mRNAs were missing from the most ventral neural tube but expressed evenly in its dorsal aspect (Fig 3-8H), similar to that of Gli2 expression in the dorsal neural tube in mice In contrast, gli2 transcripts were found in

the anterior diencephalon, throughout the midbrain and the MHB, but reduced in the

hindbrain (Fig 3-8I, J) At the MHB, expression of gli2b was also different from that

of gli2 gli2b expression was detected in the rhombomere 1 (Fig 3-8G), whereas gli2

104

transcripts were found in the posterior hypothalamus (Fig 3-8K), whereas gli2

transcripts were found in the infundibulum (neurohypophysis; Fig 3-8L)

Fig 3-8 The spatial pattern of gli2b expression during zebrafish embryogenesis

WISH was performed as described in the Experimental Procedures section The

stages and views of embryos as well as probes used for in situ hybridization are indicated in each panel (A) A 3 hours postfertilization (hpf) embryo Note that gli2b mRNA is present mainly in the nuclei (B) A 6 hpf embryo gli2b transcripts are

found in the cytoplasm (C, E) Flat mounted three-somite stage embryos hybridized

with the gli2b (C) and gli2 (E) probes, respectively (D) Dorsal view of the anterior portion of an 18 hpf embryo hybridized with the gli2b probe (F, G) Lateral (F) and dorsal (G) views of 24 hpf embryos hybridized with the gli2b probe (H)

Cross-section of the 24 hpf embryo in (G) at the otic vesicle level (dashed line in G)

(I, J) Expression of gli2 provided for comparison (F, G) (K, L) Expression of gli2b (K) and gli2 (L) mRNA in the ventral diencephalon of 48 hpf embryos

Anterior/posterior (A-P) axis and adenohypophysis (broken line) of embryo are marked ahp, adenohypophysis; di, diencephalon; FMB, forebrain-midbrain boundary; hb, hindbrain; hp, hypothalamus; mb, midbrain; MHB, midbrain-hindbrain boundary; ot, optic tectum; ov, otic vesicle; pt, posterior thalamus; te, telencephalon; ZLI, zona limitans intrathalamica