Download Two distinct teleost hepatocyte nuclear factor 1 genes, hnf1a/tcf1

Gene 338 (2004) 35 – 46
www.elsevier.com/locate/gene
Two distinct teleost hepatocyte nuclear factor 1 genes, hnf1a/tcf1
and hnf1b/tcf2, abundantly expressed in liver, pancreas,
gut and kidney of zebrafish
Hong-Yi Gong a,b, Cliff Ji-Fan Lin a,b, Mark Hung-Chih Chen b, Meng-Chuen Hu a,b,
Gen-Hwa Lin b, Yi Zhou c, Leonard I. Zon c, Jen-Leih Wu a,b,*
a
Graduate Institute of Life Sciences, National Defense Medical Center, Taipei, Taiwan
Laboratory of Marine Molecular Biology and Biotechnology, Institute of Zoology, Academia Sinica, Nankang 11529, Taipei, Taiwan
c
Department of Pediatrics, Children’s Hospital, Howard Hughes Medical Institute and Division of Hematology/Oncology, Harvard Medical School, Boston,
MA 02115, USA
b
Received 19 January 2004; received in revised form 21 April 2004; accepted 6 May 2004
Received by T. Gojobori
Abstract
Two distinct forms of zebrafish hepatocyte nuclear factor 1 (hnf1) were identified and referred to as hnf1a/tcf1 and hnf1b/tcf2. Both hnf1
genes were shown to be expressed abundantly in liver, pancreas, gut and kidney. Zebrafish HNF1a and HNF1h proteins contain all HNF1
signature domains including the dimerization domain, POU-like domain and atypical homeodomain. Sequence and phylogenetic analysis
reveals that zebrafish hnf1a is closer to tetrapodian hnf1a than to tetrapodian hnf1b and zebrafish hnf1b is highly conserved with tetrapodian
hnf1b. Existences of hnf1a and hnf1b in teleost zebrafish, tilapia and fugu suggest that hnf1 gene duplication might occur before the
divergence of teleost and tetrapod ancestors. Zebrafish hnf1a and hnf1b genes were mapped to linkage group LG8 and LG15 in T51 panel by
RH mapping and are composed of 10 and 9 exons, respectively. Zebrafish hnf1b gene with at least 11 genes in LG15 was identified to
maintain the conserved synteny with those of human in chromosome 17 and those of mouse in chromosome 11. Our results indicate that
distinct hnf1a and hnf1b genes in teleosts had been evolved from the hnf1 ancestor gene of chordate.
D 2004 Elsevier B.V. All rights reserved.
Keywords: hnf1; vhnf1; Homeodomain; Evolution; Gene duplication
1. Introduction
Hepatocyte nuclear factor 1 (HNF1), an atypical homeodomain protein family, is one of the important liverenriched transcription factors and plays crucial roles in
gene expression of liver, pancreas, intestine, stomach and
kidney. There are two hnf1 genes, hnf1a/hnf1/tcf1 and
Abbreviations: HNF1, hepatocyte nuclear factor 1; vHNF1, variant
hepatocyte nuclear factor 1; TCF1, transcription factor 1; TCF2,
transcription factor 2; HNFs, hepatocyte nuclear factors; hpf, hour postfertilization; LG, linkage group.
* Corresponding author. Laboratory of Marine Molecular Biology and
Biotechnology, Institute of Zoology, Academia Sinica, Nankang 11529,
Taipei, Taiwan. Tel.: +886-2-27899534; fax: +886-2-27824595.
E-mail address: [email protected] (J.-L. Wu).
0378-1119/$ - see front matter D 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.gene.2004.05.003
hnf1b/vhnf1/tcf2, in tetrapods (Sourdive and Yaniv, 1997)
and HNF1 could perform function as homodimer or
heterodimer (Rey-Campos et al., 1991). HNF1a is an
essential transcription factor for many hepatic genes including albumin, a1-antitrypsin, h-fibrinogen, liver-type
fatty acid binding protein (L-FABP), etc., which are
involved in detoxification, homeostasis and metabolisms
of glucose, lipid, steroid and amino acid (Shih et al., 2001).
In addition to regulation of liver-specific genes, HNF1a is
also involved in gene expression in pancreas (Okita et al.,
1999), intestine (Mitchelmore et al., 2000) and kidney
(Pontoglio et al., 2000). Although mutations in human
hnf1a gene have been shown to cause maturity-onset
diabetes of the young, i.e., MODY3 (Yamagata et al.,
1996), hnf1a does not seem to play an essential role during
development (Pontoglio et al., 1996). On the other hand,
36
H.-Y. Gong et al. / Gene 338 (2004) 35–46
defects in human hnf1b gene cause not only maturityonset diabetes (MODY5) (Horikawa et al., 1997), but also
renal defects and genital malformation (Ryffel, 2001).
Apart from its functional roles in adult organs, hnf1b is
required for embryonic development including visceral
endoderm, ectoderm differentiation (Coffinier et al.,
1999), mesoderm induction (Vignali et al., 2000) and
organogenesis of kidney, liver and pancreas (Wild et al.,
2000; Sun and Hopkins, 2001). In teleosts, only one hnf1
form was reported in Atlantic salmon (Deryckere et al.,
1995) and two hnf1 partial cDNAs in tilapia were identified by us (Huang et al., 2001). In zebrafish, conserved
vhnf1/hnf1b, which is involved in kidney cysts formation
similar to human symptoms of hnf1b-associated human
familial glomerulocystic kidney disease (GCKD) and in
controlling development of multiple organs through regulating regional specification of organ primordia, was
identified by retroviral insertion mutagenesis (Sun and
Hopkins, 2001). However, the gene numbers and expression of HNF1 family in teleost remain not clear. In this
study, we report the cloning, expression, structure, linkage
mapping and evolution of two distinct hnf1 genes, referred
to as hnf1a/tcf1 and hnf1b/tcf2, from zebrafish (Danio
rerio), a popular model system of vertebrate genetics and
development.
2. Materials and methods
2.1. Fish stocks
Zebrafish (D. rerio) was obtained from Oregon State
University. All adult fishes were maintained in a freshwater
recirculating tank at 28 jC.
2.2. Cloning of zebrafish hnf1a and hnf1b full-length cDNA
Degenerate primers of hnf1a were designed from the
conserved regions of hnf1a cDNAs from human, mouse, rat,
chicken, Xenopus, salmon and tilapia. The sequences of
degenerate primers were as follows. The 5V-primer located at
conserved POU-like domain is 5V-CCCATCT(C/
T)TCCCAGCACCTCAACAAAGGCAC(T/G/C)CCC-3V
and 3V-primer located at 3V-end of conserved homeodomain
is 5V-GCTTC(C/T)TCCTT(G/C/A)CG(C/G)CGGTT(G/A/
C)GCGAACCAGTTG-3V. By using two-step RT-PCR,
397-bp partial cDNA of zebrafish hnf1a was generated from
the zebrafish liver total RNA. Liver first-strand cDNA was
synthesized at 42 jC for 50 min with SUPERSCRIPT II
reverse transcriptase (Invitrogen) and PCR was performed
with Platinum Taq DNA polymerase (Invitrogen). The 397bp zebrafish hnf1a cDNA fragment was used for the probe
preparation by rediPrime II random prime labeling kit
(Amersham Pharmacia Biotech) with [a-32P]dCTP. Approximately 1 106 bacteriophage plaques from a zebrafish 24h post-fertilization (hpf) embryo cDNA library (Stratagene,
USA), which was synthesized by poly(dT) primer, and a 1month-old zebrafish cDNA library (CLONTECH) synthesized by both random primer and poly(dT) primer were
screened with high stringency hybridization and washing
condition, respectively, as mentioned by previous method
(Chen et al., 2001), except hybridization buffer containing
50% formamide and washing to 0.1 SSC –0.1% SDS at
68 jC for 30 min. The partial hnf1b cDNA (500 bp) of
tilapia (Oreochromis mossambicus) (Huang et al., 2001)
was used for probe preparation to screen zebrafish hnf1b
cDNA from the zebrafish 24-hpf embryo cDNA library
under moderate hybridization and wash stringency. Deduced
amino acid sequences of HNF1 family members in zebrafish, salmon, fugu, Xenopus, chicken, mouse, rat, pig and
human were compared by pileup program of Genetics
Computer Group (GCG, Version 7.0) and represented by
GeneDoc software.
2.3. HNF1 transcripts detection by two-step RT-PCR
Total RNAs from various tissues of adult zebrafish and
developmental stages of zebrafish embryo were prepared
by using TRIZOL reagent (Invitrogen). Using 1 Ag of total
RNA, the first-strand cDNAs (20 Al) were synthesized
with oligo(dT) by SUPERSCRIPT III reverse transcriptase
(Invitrogen) according to manufacturer’s protocol. After
reverse transcription, 2-Al first-strand cDNAs were subjected to PCR by using Platinum Taq DNA polymerase
(Invitrogen) with hnf1a and hnf1b specific primers located
at 5V-end and 3V-end of variant activation domain. Genespecific primers are listed as follows: HNF1A-AD-5VP: 5VATGTGCCCTACAGCAGCCAATCAGCTGCCT-3V,
HNF1A-AD-3VP: 5V-TTGTGCAGTGGAGACCATCTGTGCAGGAATG-3V; HNF1B-AD-5VP: 5V-ACAGTGGGCCAGCGCATAGCCTAAACTCCC-3V, HNF1B-AD-3VP: 5VGGACATTGAGCTCAGTGTACTGAGGCTGTT-3. The
thermal cycling condition consisted of 1 cycle of initial
denaturation at 94 jC for 2 min, 35 cycles of threetemperature cycling with denaturation at 94 jC for 30 s,
annealing at 60 jC for 30 s and extension at 72 jC for 1
min and 1 cycle of terminal extension at 72 jC for 7 min.
The RT-PCR products of hnf1a and hnf1b were 869 and
668 bp, respectively, and were confirmed by sequencing.
Elongation factor 1 alpha (ef1a) of zebrafish (GenBank
accession no. X77689) was used as an internal control of
RT-PCR. Zebrafish ef1a5V-primer was 5V-TCCTTCAAGTACGCCTGGGTGTTGG-3Vand 3V-primer was 5V-ACACACTAGGGCTTGCCAGGGACCA-3V. PCR condition
of zebrafish ef1a was the same as hnf1 except for 25
cycles of three-temperature cycling.
2.4. Whole-mount in situ hybridization
Zebrafish embryos of various stages were collected
(Kimmel et al., 1995) and whole-mount in situ hybridization
was performed according to the instructions of The Zebra-
H.-Y. Gong et al. / Gene 338 (2004) 35–46
fish Book (Westerfield, 1995). Variant activation domains of
zebrafish hnf1a and hnf1b were amplified by PCR and
subcloned into pGEM-T Easy (Promega, USA) vector for
riboprobes production. Zebrafish hnf1 gene-specific DIGantisense riboprobes were synthesized with T7 RNA polymerase by DIG RNA labeling kit and detected using DIG
nucleic acid detection kit (NBT/BCIP) (Roche, Germany).
Hybridizations and washes were performed at 68 jC.
Embryos were destained in the 2:1 mixture of benzyl
benzoate/benzyl alcohol and observed with Leica MZ FLIII
stereomicroscope.
2.5. Phylogenetic analysis
Phylogenetic and molecular evolutionary analyses were
conducted using MEGA version 2.1 (Kumar et al., 2001)
with 61-amino-acid peptide of HNF1-specific POU-like
DNA binding domain encoded by exon 2 of HNF1 from
vertebrates and chordates. Tree construction and distance
correction were followed by neighbor-joining and Poisson
correction methods, respectively. GenBank accession
numbers of HNF1 family members used for the analysis
are listed as below: human (Homo sapiens) HNF1a,
M57732; human HNF1h, X58840; pig (Sus scrofa)
HNF1h, X69675; rat (Rattus norvegicus) HNF1a,
J03170; rat HNF1h, X56546; mouse (Mus musculus)
HNF1a, M57966; mouse HNF1h, X55842; chicken (Galllus gallus) HNF1a, X67689; Xenopus (Xenopus laevis)
HNF1a, X64759; Xenopus HNF1h, X76052; Atlantic
salmon (Salmo salar) HNF1, X79486; zebrafish (D. rerio)
HNF1a, AF244934; zebrafish HNF1h, AF244140; tilapia
(Tilapia mossambica) HNF1a, AF348407 and HNF1h,
AF348408; fugu (Fugu rubripes) HNF1a (Ensembl Translation ID: SINFRUP00000135090 from Ensembl Gene ID:
SINFRUG00000127746) and HNF1h (Ensembl translation
ID: SINFRUP00000136747 from Ensembl Gene ID: SINFRUG00000129240); Ciona (Ciona intestinalis) HNF1,
from POU-like domain to homeodomain encoded by nucleotide 49,298 – 51,535 of Scaffold 51 (JGI Ciona v1.0)
(http://genome.jgi-psf.org/ciona4/ciona4.home.html).
2.6. RH mapping zebrafish hnf1a and hnf1b in T51 panel
Primers were designed with the 3V-UTR sequence of
zebrafish hnf1a (GenBank accession no. AF244934) and
hnf1b cDNA (AF244140), respectively. The primers used
for RH mapping of zebrafish hnf1a were forward primer 5VTGCTGCCTAGCTATGCTAATAATG-3Vand reverse primer 5V-GGGTTTTTCTTGTGTTTCAGTCTT-3V. Optimal
annealing temperature of zebrafish hnf1a PCR was 49.4 jC
and target size of PCR was 269 bp. The primers used for RH
mapping of zebrafish hnf1b were forward primer 5VGTGGCGTCAGATAGATAAGAGTCA-3V and reverse
primer 5V-CTGTTTGAAGTGGCGAACTG-3V. Optimal
annealing temperature of zebrafish hnf1b PCR was 53 jC
and target size of PCR was 307 bp.
37
3. Results
3.1. Cloning of zebrafish hnf1a/tcf1 and hnf1b/tcf2 fulllength cDNAs
One 5V-truncated positive cDNA clone #1-1 containing
1708-bp insert composed of 688-bp coding region, 1000bp 3V-UTR and 20-bp poly(A) was obtained from the
zebrafish 24-hpf embryo cDNA library. One cDNA clone
#2-1 containing zebrafish hnf1a cDNA with 276-bp 5VUTR and 5V-end 523-bp coding region was isolated from 1month-old zebrafish cDNA library. According to the sequence of hnf1a cDNA clones #1-1 and #2-1, the 5V- and
3V-primers corresponding to 5V- and 3V-end of coding region
of zebrafish hnf1a cDNA were designed to obtain the fulllength zebrafish hnf1a cDNA (GenBank accession no.
AF442956) from zebrafish gut total RNA by RT-PCR.
Based on the conservation of HNF1h, full-length zebrafish
hnf1b cDNA (GenBank accession no. AF244140) was
obtained by using tilapia hnf1b partial cDNA (Huang et
al., 2001) to screen the 24-hpf-embryo cDNA library.
Zebrafish HNF1a/TCF1 and HNF1h/vHNF1/TCF2 are
composed of 560 and 559 amino acids (a.a.), respectively,
with 61.6% similarity and share dimerization domain,
conserved DNA binding domain including POU-like domain and atypical homeodomain with 21 amino acid (a.a.)
loop between helix 2 and helix 3, but variant transactivation domain (Fig. 1). Zebrafish hnf1a is most similar to
salmon hnf1 closer to tetrapodian hnf1a than to hnf1b.
Compared with other vertebrate HNF1s, HNF1a of teleosts
including zebrafish, salmon, fugu (Ensembl Translation ID:
SINFRUP00000135090) and tilapia all have the most
variant and extra-long dimerization domain (Fig. 1A).
Deduced HNF1a of zebrafish shares 82.6%, 79.2%,
77.6%, 63.9%, 63.1%, 65.4%, 65.9% and 64.5% similarities with that of salmon, tilapia, fugu, Xenopus, chicken,
mouse, rat and human, respectively (Fig. 1A, Table 1).
Interestingly, two conserved teleost-specific cysteines were
found in activation domain of teleost HNF1a (Fig. 1A). It
indicates that the disulfide bond may be involved in the
structure formation of teleost HNF1a and needs to be
further clarified. Deduced HNF1h of zebrafish shares
86.7%, 84.5%, 84.3%, 85.6%, 85.6% and 85.8% high
similarities with that of tilapia, Xenopus, mouse, rat, pig
and human, respectively (Fig. 1B, Table 1). Compared with
published zebrafish vhnf1 (Sun and Hopkins, 2001), zebrafish hnf1b cDNA isolated by us encodes identical protein
but has 6-bp variations in coding region, 13-bp variations
in 3V-UTR and 909-bp-longer 3V-UTR. Although zebrafish
HNF1h has conserved dimerization domain, another translation initiation ATG, 12 bp ahead of conserved translation
initiation site, were found. It indicates that zebrafish
HNF1h may have extra 4 amino acids in the N-terminal
of dimerization domain and this phenomenon also occurs
in other teleost tilapia HNF1h and fugu HNF1h (Ensembl
Translation ID: SINFRUP00000136747).
38
H.-Y. Gong et al. / Gene 338 (2004) 35–46
Fig. 1. Alignment of deduced amino acid sequences of HNF1a (A) and HNF1h (B) in vertebrates. Amino acid residues that are identical in all three proteins are shaded in black, while residues conserved in two
proteins are shaded in gray. Deduced amino acid of HNF1 family members were compared by GCG pileup program and processed using GeneDoc program. Teleost-specific cysteine pair in activation domain of
HNF1a was indicated by arrow.
H.-Y. Gong et al. / Gene 338 (2004) 35–46
39
Table 1
Similarity of zebrafish HNF1a and HNF1h with HNF1 members of other vertebrates
Dr-1A
Dr-1B
Dr-1A
Dr-1B
Dr-1A (%)
Ss-1 (%)
Tm-1A (%)
Fr-1A (%)
Xl-1A (%)
Gg-1A (%)
Mm-1A (%)
Rn-1A (%)
100
61.59
82.58
63.52
79.24
62.84
77.64
63.29
63.94
67.13
63.07
67.12
65.38
66.80
65.93
66.80
Dr-1B (%)
Ss-1B**
Tm-1B (%)
Fr-1B*
Xl-1B (%)
Gg-1B**
Mm-1B (%)
Rn-1B (%)
Sus-1B (%)
Hs-1B (%)
60.46
84.34
61.41
85.61
61.70
85.61
62.24
85.79
61.59
100
–
–
60.84
86.65
–
–
61.00
84.50
–
–
Sus-1A**
–
–
Hs-1A (%)
64.53
66.30
1A, HNF1a; 1B, HNF1h; Dr, zebrafish; Ss, Salmon; Fr, Fugu; Tm, Tilapia; Xl, Xenopus; Gg, chicken; Mm, Mouse; Rn, Rat; Sus, pig; Hs, Human; Fr-1A,
putative protein sequence deduced from gene identified in Fugu genome, although cDNA had not been identified; Fr-1B*, gene had been identified in the Fugu
genome but cDNA has not been identified; Ss-1B**, Gg-1B** and Sus-1A**, both cDNA and gene were not identified; – , not determined.
3.2. Tissue distribution and developmental expression of
zebrafish hnf1a and hnf1b
Zebrafish hnf1a and hnf1b transcripts were detected in
various tissues of adult fish and during embryonic development by RT-PCR with gene-specific primers in activation
domain. Unlike tetrapodian hnf1, zebrafish hnf1a and hnf1b
are detected in all tested tissues, including brain, eye, gill,
gut, heart, kidney, liver, muscle, swim bladder, ovary and
testis (Fig. 2A). As hnf1 of other vertebrates, zebrafish
hnf1a and hnf1b are also abundantly expressed in liver,
gut and kidney of adult fish. Zebrafish hnf1a are weakly
expressed in swim bladder, testis, heart and eye and little
expressed in brain, ovary, muscle and gill. In addition to
kidney, gut and gut-derived liver, zebrafish hnf1b was also
strongly expressed in swim bladder, weakly expressed in
brain, eye, testis, ovary and heart and rarely expressed in
Fig. 2. Expression of zebrafish hnf1a and hnf1b in various tissues (A) and
developmental stages (B) detected by RT-PCR. +, positive control, plasmid
containing zebrafish hnf1 cDNA used as template of PCR;
, negative
control: no template RNA in PCR. B, brain; E, eye; Gi, gill; Gu, gut; H,
heart; K, kidney; L, liver; M, muscle; S, swim bladder; O, ovary; T, testis.
DNA size marker, 1 kb plus ladder (Invitrogen). 1K cells, 1000-cell stage,
mid-blastula transition stage; 50% Eb., 50% epiboly stage, onset of
gastrulation; 90% Eb., 90% epiboly stage, late gastrulation; 11 hpf, early
somitogenesis, 2-somite. hpf, hour post-fertilization. Zebrafish elongation
factor 1 a (ef1a) was used as control to indicate the relative quantities of
RNA template used in RT-PCR.
muscle and gill. The zygotic expression of zebrafish hnf1a
was obviously detected at onset of gastrulation (50% epiboly) and strongly detected from the end of gastrulation (90%
epiboly, 10 hpf). Little expression of zebrafish hnf1a was
detected from 1- to 1000-cell stage. Zebrafish hnf1b was
maternally supplied due to strong detection at early cleavage
stage. Zebrafish hnf1b were significantly expressed from
onset of gastrulation and were detected during whole
developmental process (Fig. 2B). During the embryonic
development, specific and dynamic expression patterns of
zebrafish hnf1b were detected during organogenesis of
hindbrain, kidney, gut, liver and pancreas by whole-mount
in situ hybridization (Fig. 3). Zebrafish hnf1b was obviously
expressed as two patches in the presumptive hindbrain at the
tailbud stage (Fig. 3A). During the early somitogenesis, two
patches of hnf1b expression had fused with strong anterior
border at the posterior hindbrain and extend caudally into
presumptive spinal cord with gradually weaker signal (Fig.
3B – D). On the one hand, hnf1b was expressed in the
intermediate mesoderm, precursor cells of kidney (Fig. 3C
and D) from onset of four-somite stage and strongly
expressed in the pronephric ducts (Fig. 3E – H) extending
to anus during the segmentation stages, and on the other
hand, the expression of hnf1b in the hindbrain disappeared
at eight-somite stage. Zebrafish hnf1b was also detected in
the foregut (Fig. 3I) and in the foregut-derived liver and
pancreas (Fig. 3J), indicating its roles in liver and pancreas
organogenesis and function (Sun and Hopkins, 2001).
Unlike zebrafish hnf1b/tcf2 involved in development of
hindbrain, pronephros and gut, zebrafish hnf1a/tcf1 did
not show obviously specific expression patterns during
embryonic development (Fig. 3K – N). After hatching,
zebrafish hnf1a expression was strongly detected at the liver
and pronephric tubules and weakly detected in the gut,
pancreas and pronephric ducts (Fig. 3O and P).
3.3. Evolution of hepatocyte nuclear factor 1 family
Only one salmon hnf1 cDNA suggested that the duplication event from which hnf1a and hnf1b genes arose
occurred after the divergence of the tetrapod and teleost
ancestors (Deryckere et al., 1995). Sequence and phylogenetic analysis shows that zebrafish HNF1a, most similar to
40
H.-Y. Gong et al. / Gene 338 (2004) 35–46
Fig. 3. Expression of zebrafish hnf1b/tcf2 and hnf1a/tcf1 during embryonic development detected by whole-mount in situ hybridization. (A – J) Zebrafish
hnf1b/tcf2 activation domain antisense riboprobes. (K – P) Zebrafish hnf1a/tcf1 activation domain antisense riboprobes; (A, K) tailbud stage; (B, L) two-somite
stage; (C) dorsal view; (D) lateral view four-somite stage; (E, M) eight-somite stage; (F) 15 hpf; (G) 19 hpf; (N) 22 hpf; (H) 24 hpf; (I) 33 hpf; (J, O, P) 53 hpf.
In (A, B, C, E, J, L, O and P), embryo was shown in dorsal view; in (D, F, G, H, I, K, M and N), embryo was shown in lateral view. Arrows in (A) – (D) indicate
hindbrain expression pattern of zebrafish hnf1b/tcf2 and arrowheads in (C) – (I) indicate pronephros expression of zebrafish hnf1b/tcf2 from intermediate
mesoderm (C, D) to pronephric ducts (E) – (I). Star in (I) indicates foregut. Li, liver; Gu, gut; Pa, pancreas; Pd, pronephric duct; Pt, pronephric tubule.
salmon HNF1, is closer to vertebrate HNF1a than to
HNF1h and conserved zebrafish HNF1h shares high
similarity ( f 85%) with tetrapodian HNF1h (Table 1,
Fig. 4B). Recently, one hnf1 gene was identified based
on HNF1-specific POU-like domain and atypical homeodomain in the tunicate C. intestinalis genome (Wada et al.,
2003). We constructed the phylogenetic tree of vertebrate
HNF1 family members from human, pig, rat, mouse,
chicken, Xenopus HNF1s, teleost salmon, zebrafish, tilapia,
fugu HNF1s and chordate Ciona HNF1 by 61-amino-acid
peptide sequences (Fig. 4A) of the HNF1-specific POUlike domain encoded by exon 2 of HNF1 genes (Fig. 4).
Although conserved hnf1b is still not found in salmon, the
finding of both hnf1a and hnf1b in zebrafish, tilapia and
fugu all supported a new hypothesis of hnf1 evolution and
advanced the hnf1 duplication event to the time before the
divergence of the teleost and tetrapod ancestors. As no
hnf1 homologue gene based on conserved HNF1-specific
POU-like domain and atypical homeodomain was found in
invertebrates including Caenorhabditis elegans and Dro-
H.-Y. Gong et al. / Gene 338 (2004) 35–46
41
Fig. 4. Phylogenetic analysis of HNF1 family members. (A) Alignment of deduced amino acid sequences of POU-like domain of HNF1 family members in
vertebrates and chordate Ciona. Amino acid residues that are identical in all proteins are shaded in black, while residues conserved in proteins are shaded in
gray. Deduced amino acid of functional domains from various HNF1 family members were compared by GCG pileup program and processed using GeneDoc
program. Sixty-one-amino-acid peptide of HNF1-specific POU-like DNA binding domain, which was encoded by exon 2 of HNF1 from vertebrates and
chordates, used in phylogenetic analyses was underlined. (B) Phylogenetic tree of HNF1 genes was constructed by neighbor-joining using Molecular
Evolutionary Genetics Analysis (MEGA2) software (Version 2.1). Ciona HNF1 served as an out-group to root the tree. Bootstrap values >50% from 1000
replicates were listed above branches. The marker of 0.05 is the length that corresponds to a 5% sequence difference.
sophila, Ciona hnf1 seems to be the ancestor of hnf1 genes
of vertebrates. Before divergence of the sarcopterigyans
from the actinopterigyans (about 420 m.y. ago), the duplication event of hnf1 gene had occurred from the ancestor
of primitive chordate.
3.4. Gene structure of zebrafish hnf1a and hnf1b gene
To clarify the gene structure of the zebrafish hnf1a gene
and hnf1b gene, we compared the sequences of zebrafish
hnf1a and hnf1b cDNA with zebrafish hnf1a/tcf1 gene
(Ensembl Gene ID: ENSDARG00000009470) and hnf1b/
tcf2 gene (Ensembl Gene ID: ENSDARG00000003873)
(http://www.ensembl.org/Danio_rerio), respectively. Zebrafish hnf1a/tcf1 gene spans about 14.8 kb and is also
composed of 10 exons and 9 introns as hnf1a gene of
chicken (Hörlein et al., 1993), mouse and rat (http://
www.ncbi.nlm.nih.gov/genome/) (Fig. 5A, Table 2). All
introns of zebrafish hnf1a gene are smaller than those of
mammal hnf1a gene and follow the GT-AG rule except
42
H.-Y. Gong et al. / Gene 338 (2004) 35–46
Fig. 5. Structure of zebrafish hnf1a/tcf1 (A) and hnf1b/tcf2 (B) genes and
functional domains of deduced zebrafish HNF1 proteins. Exons are
represented by open boxes. 5V- and 3V-untranslated region (UTR) are
indicated by oblique lines. Zebrafish hnf1a/tcf1 gene is composed of 10
exons and 9 introns; zebrafish hnf1b/tcf2 gene is composed of 9 exons and
8 introns. DI, dimerization domain; POU, POU-like domain; HOX,
homeodomain; AD, activation domain.
that the 5V-donor site of intron 8 is GC (GC-AG rule),
which is occasionally found at the 5V-end of certain introns
(Table 2). The dimerization domain of zebrafish HNF1a is
encoded by exon 1, POU-like domain encoded by 3V-end
of exon 1 and exon 2, atypical homeodomain encoded by
3V-half of exon 3 and 5V-half of exon 4 and activation domain
encoded by exons 4, 5, 6, 7, 8, 9 and 5V-end of exon 10 (Fig.
5A, Table 2). As hnf1b gene of human, mouse and rat (http://
www.ncbi.nlm.nih.gov/genome/guide/), zebrafish hnf1b
gene is also composed of 9 exons and 8 introns (Fig. 5B).
The dimerization domain of zebrafish HNF1h is encoded by
exon 1, POU-like domain encoded by 3V-end of exon 1 and
exon 2, atypical homeodomain encoded by 3V-half of exon 3
and 5V-half of exon 4 and activation domain encoded by
exons 4, 5, 6, 7, 8 and 5V-end of exon 9 (Fig. 5B). Due to the
smaller introns, the size of zebrafish hnf1b gene is about 15.7
kb, which is only one-fourth of human hnf1b gene in size.
The splicing sites are conserved in zebrafish hnf1b gene and
are all following the GT-AG rule (Table 2). As hnf1 genes of
other vertebrates, zebrafish hnf1a and hnf1b genes both
contain an intron (intron 3) exactly between the second helix
and the 21 amino acid loop of atypical homeodomain. Fugu
genome project also reveals (http://www.ensembl.org/Fugu
rubripes/) the existence of both hnf1a/tcf1 (Ensembl Gene
ID: SINFRUG00000127746) and hnf1b/tcf2 (Ensembl Gene
ID: SINFRUG00000129240) genes. Compared with exon/
intron junction of zebrafish hnf1a gene, fugu hnf1a/tcf1 gene
is also predicted to comprise 10 exons and 9 introns as
zebrafish hnf1a gene. To determine the position in zebrafish
genome, zebrafish hnf1a and hnf1b genes were mapped to
linkage groups 8 (LG8) and 15 (LG15), respectively, in
Goodfellow T51 pannel by RH mapping (http://
zfrhmaps.tch.harvard.edu/ZonRHmapper) (Fig. 6A).
4. Discussion
Liver-enriched transcription factors comprise several
hepatocyte nuclear factors (HNFs) including HNF1,
HNF3, HNF4 and HNF6 and CCAAT/enhancer binding
protein (C/EBP) families and play crucial roles in gene
expression of endoderm-derived liver, pancreas, intestine,
stomach and mesoderm-derived kidney. HNF1 is a crit-
Table 2
Size and junction sequence of exons and introns of hnf1a/tcf1 gene and hnf1b/tcf2 gene in zebrafish
Zebrafish hnf1a/tcf1
Junction sequence
Exon1
Exon2
Exon3
Exon4
Exon5
Exon6
Exon7
Exon8
Exon9
AGCTGCTACAgtaagtcctc . . . . . . tgtgtgttagAGAGGATCCT (2816 bp)
ATCAGTCAGCgtgagtataa . . . . . . atcattacagAATTCACAAA (2714 bp)
AGTGCAACAGgtaagtcaac . . . . . . tgtgtttcagAGCGGAGTGT (1429 bp)
CCTTCACCAGgtattcaaac . . . . . . taaactgaagGTCTGAAGTA (546 bp)
TCACAAACCAgtgagattgt . . . . . .cttctctcagGCATCTGTAG (500 bp)
CTTCTCATTGgtgaaataca . . . . . .tttgctgcagGCTTGACATC (99 bp)
CCATGTCACAgtaagtaatt . . . . . . tgtgttttagTGTACAGCAA (1144 bp)
TGTTCGACAGgcatgagcca . . . . . . tcattatcagATTCTCACCA (360 bp)
CCTGAACCAGgtatacatgc . . . . . . aaacctctagGGTCTTCAGG (2167 bp)
(625 bp) – Intron1 – Exon2
(200 bp) – Intron2 – Exon3
(181 bp) – Intron3 – Exon4
(205 bp) – Intron4 – Exon5
(149 bp) – Intron5 – Exon6
(133 bp) – Intron6 – Exon7
(189 bp) – Intron7 – Exon8
(116 bp) – Intron8 – Exon9
(85 bp) – Intron9 – Exon10 (1134 bp)
Zebrafish hnf1b/tcf2
Junction sequence
Exon1
Exon2
Exon3
Exon4
Exon5
Exon6
Exon7
Exon8
GAATGTTGGCgtgagttttt . . . . . . atcttttcagGGAGGACCCG
ATCTTGCGACgtaagttaat . . . . . . tatgtttcagAATTCAACCA
AGTGCAACCGgtaagatgac . . . . . . ctacattcagGGCTGAGTGC
AAGATGCAAGgtactgatac . . . . . . tttacaacagGTGTCCGGTA
TGCCAAGATGgtaagcactt . . . . . . ttcttcacagATCTCGGTAT
ATTGCACAAAgtgagtgtcc . . . . . . ttgcattcagGTTTGAACAC
CACTCACACAgtcagtctac . . . . . . atctttacagTGTACCCACA
CAGCAAACAGgtcagtaatt . . . . . . tttctcacagTGTCCACTAC
(535
(200
(271
(236
(161
(127
(180
(119
bp) – Intron1 – Exon2
bp) – Intron2 – Exon3
bp) – Intron3 – Exon4
bp) – Intron4 – Exon5
bp) – Intron5 – Exon6
bp) – Intron6 – Exon7
bp) – Intron7 – Exon8
bp) – Intron8 – Exon9 (2199 bp)
(2241 bp)
(2162 bp)
(88 bp)
(3450 bp)
(717 bp)
(81 bp)
(94 bp)
(2870 bp)
H.-Y. Gong et al. / Gene 338 (2004) 35–46
43
Fig. 6. Map of zebrafish hnf1a/tcf1 and hnf1b/tcf2 genes and conserved hnf1b/tcf2 synteny. (A) RH map of zebrafish hnf1a/tcf1 and hnf1b/tcf2 genes in LG8
and LG15 of Goodfellow T51 panel. The maps of relative location of zebrafish hnf1 genes with other ESTs and genes and detailed flanking markers were
modified from the data contained in Zebrafish Genome Resources of Genomic Biology (http://www.ncbi.nih.gov/genome/guide/zebrafish/). (B) Zebrafish
hnf1b/tcf2 gene and at least 11 genes in LG15 maintain conserved synteny with the same set of genes in human chromosome 17 and in mouse chromosome 11.
Dr 15, zebrafish LG15; Hs 17, human chromosome 17; Mm 11, mouse chromosome 11. The relative chromosome locations of human and mouse genes were
deduced from data contained in Genomes (http://www.ncbi.nlm.nih.gov/Genomes/).
44
H.-Y. Gong et al. / Gene 338 (2004) 35–46
ical hepatic transcription factor cooperating with other
liver-enriched transcription factors to regulate many liverspecific genes, e.g., liver-type fatty acid binding protein.
In our laboratory, 2.8-kb promoter of zebrafish liver fatty
acid binding protein gene (L-FABP) could drive GFP
expression specifically in the liver of transgenic zebrafish, and one HNF1 binding site in distal region of
zebrafish L-FABP 2.8 kb-promoter was found to be
responsible for the L-FABP expression in the liver (Her
et al., 2003).
Mice lacking hnf1a are born normally but suffer from
several defects including hyperphenylalaninemia, defective
bile acid and cholesterol metabolism, an insulin secretion
defect and renal Fanconi syndrome (Pontoglio et al., 1996).
HNF1a is an essential regulator of bile acid and plasma
cholesterol metabolism (Shih et al., 2001). HNF1a also
directly controls the low-affinity/high-capacity glucose
cotransporter (SGLT2) gene expression to control renal
glucose reabsorption and maintain glucose homeostasis
(Pontoglio et al., 2000). In addition to the essential role
in epithelium differentiation of visceral endoderm (Coffinier et al., 1999), hnf1b-targeted inactivation in liver
resulted in severe jaundice caused by abnormalities of
the gall bladder and interhepatic bile ducts and revealed
essential function of hnf1b in bile duct morphogenesis and
hepatocyte metabolism (Coffinier et al., 2002). HNF1h
was also shown to play an important role in kidney
formation (Wild et al., 2000) and function (Ryffel, 2001).
Zebrafish vhnf1/hnf1b had been found to be involved in
regional specification of gut, pronephros and hindbrain by
regulating the proper expression of pdx1 and sonic hedgehog (shh) in the gut endoderm, pax2 and wt1 in the
pronephric primordial and valentino (val) in the hindbrain.
Zebrafish vhnf1 mutants generated by retroviral insertion in
vhnf1/hnf1b/tcf2 gene display phenotypes including formation of kidney cysts, underdevelopment of liver and
pancreas and reduction in size of the otic vesicles (Sun
and Hopkins, 2001). Zebrafish vhnf1 functions to subdivide caudal hindbrain domain into individual rhombomeres
by synergizing with FGF signal to activate valentino and
krox20 expression to promote rhombomeres 5 and 6
(r5 + r6) identity and by repressing hoxb1a expression
independent of FGF signal (Wiellette and Sive, 2003).
As hnf1s of other vertebrates, zebrafish hnf1a and hnf1b
are abundantly expressed in liver, pancreas, gut and kidney
(Figs. 2A, 3J, O and P). Hnf1a and hnf1b were expressed
in all tested tissues of zebrafish, more widely expressed
than that of tetrapods. Expression of hnf1s is firstly
demonstrated in the swim bladder and gill of teleost.
Two zebrafish hnf1s are both expressed in testis and ovary,
but only hnf1b is expressed in testis and ovary of tetrapodians. In muscle, heart and brain, zebrafish hnf1a and
hnf1b are both detected, but expression of both hnf1a and
hnf1b was not detected in other vertebrates (Sourdive and
Yaniv, 1997). Overlapping and differential expression of
two zebrafish hnf1 genes in different organs and develop-
mental processes indicate that two zebrafish HNF1 proteins
could also regulate downstream gene expression in more
complicated homodimers and/or heterodimers combination
as mammals.
This is the first report demonstrating the identification
of two hnf1 genes and full-length cDNAs in teleost.
Although there are two hnf1 genes, hnf1a and hnf1b, in
tetrapodians, only one hnf1 closer to tetrapodian hnf1a was
identified in Atlantic salmon (Deryckere et al., 1995). In
zebrafish, not only hnf1a most similar to Atlantic salmon
hnf1, but also hnf1b which was highly conserved among
vertebrates were found. Our results reveal that the duplication of hnf1 gene occurred before the divergence of the
tetrapod and teleost ancestors. Two hnf1 cDNAs, hnf1a/
tcf1 and hnf1b/tcf2, and genes in zebrafish genome and
two hnf1 cDNAs, hnf1a and hnf1b, in tilapia (T. mossambica) were identified by us. Moreover, we also found two
hnf1 genes, hnf1a/tcf1 and hnf1b/tcf2, in fugu (F. rubripes)
genome (http://www.ensembl.org/Fugu_rubripes/). Not only the developmental essential hnf1b gene, but also the
teleost homologue of tetrapodian hnf1a gene is maintained
in teleost zebrafish, tilapia and fugu. The tunicates (Urochordata) are phylogenetically positioned at the base of the
vertebrate tree, so the sea squirt C. intestinalis becomes an
important species for the understanding of vertebrate
evolution. Recently, the draft genome sequence of C.
intestinalis was revealed. A genome-wide survey of developmentally relevant genes including homeobox-containing
transcription factors in C. intestinalis was made, and only
one hnf1 gene in Ciona genome based on conserved
HNF1-specific POU-like domain and atypical homeodomain with 81 amino acids was identified (Wada et al.,
2003). It indicates that in teleosts, two diverse hnf1 genes,
hnf1a and hnf1b, were evolved from the hnf1 ancestor
gene of primitive chordate. The time point of hnf1 gene
duplication still needs to be clarified from the hnf1 gene
number of Cephalochordate amphioxus and primitive vertebrates such as hagfish and lamprey.
Comparative genome maps of zebrafish and human
show that there is not a one-to-one correspondence between
zebrafish and human chromosomes. For example, most
zebrafish orthologs of human chromosome 17 map to four
linkage groups including LG3, LG12, LG15 and LG5
(Postlethwait et al., 2000). To clarify whether zebrafish
hnf1 genes share conserved syntenies with human hnf1a
and hnf1b, we compared zebrafish LG8 and LG15 with
human chromosome Hsa 12 and Hsa 17 and mouse
chromosome Mm 5 and Mm 11. Zebrafish hnf1b gene
with at least 11 genes, including TBP-associated factor 15
(taf15), splicing factor arginine/serine-rich 1 (sfrs1), tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, epsilon polypeptide (ywhae), v-crk sarcoma
virus CT10 oncogene homologue (crk), polymerase delta
interacting protein 38 (pdip38), T-box 2 (tbx2b), LIM
homeobox 1(lhx1/lim1a), replication protein A1 (rpa1),
pre-mRNA processing factor 8 homologue (prpf8), vitro-
H.-Y. Gong et al. / Gene 338 (2004) 35–46
nectin (vtn) and SOCS box-containing WD protein SWiP-1
(wsb1), in LG15 was identified to maintain the conserved
synteny with the same gene set in human chromosome 17
and mouse chromosome 11 (Fig. 6B). Changes in gene
order within this conserved synteny support the frequent
occurrence of inversions and other intrachromosomal rearrangements in these regions since the divergence of teleost
and tetrapod ancestors (Woods et al., 2000). On the
contrary, zebrafish hnf1a lost the synteny with human
hnf1a in chromosome 12 and mouse hnf1a in chromosome
5. Our results indicate that the hnf1 ancestor gene of
chordate had evolved into two distinct hnf1 genes with
differentiated expression and function in teleosts as in
tetrapodians. The duplicated genes evolved from the ancestor gene may share redundant functions, separated
functions or evolve new functions after long time evolution. Zebrafish with partial genome duplication had become
a critical model system to study the functional divergence
of duplicated genes.
Acknowledgements
We thank Dr. Chien-Hsien Kuo, Dr. Chao-Lun Allen
Chen and Dr. Sin-Che Lee for their helpful discussion in
phylogenetic analysis and molecular evolution. We thank
Dr. Thomas T. Chen, Dr. Sheng-Ping Huang, Dr. ChingFong Liao and Dr. Tzy-Wen L. Gong for their suggestions
and supports. We are grateful to Dr. Aseervatham Anusha
Amali, Dr. Joseph Abraham Christopher John and Dr.
Deepa Rekha for critical reading of this manuscript. This
work was supported by grants NSC89-2311-B-001-169,
NSC90-2311-B-001-158 and NSC91-2311-B-001-126 from
the National Science Council of Taiwan and by Academia
Sinica.
References
Chen, M.H.C., Lin, G.H., Gong, H.Y., Weng, C.F., Chang, C.Y., Wu, J.L.,
2001. The characterization of prepro-insulin-like growth factor-1 Ea-2
expression and insulin-like growth factor 1 genes (devoid 81 bp) in
zebrafish (Danio rerio). Gene 268, 67 – 75.
Coffinier, C., Thepot, D., Babinet, C., Yaniv, M., Barra, J., 1999. Essential
role for the homeoprotein vHNF1/HNF1h in visceral endoderm differentiation. Development 126, 4785 – 4794.
Coffinier, C., Gresh, L., Fiette, L., Tronche, F., Schutz, G., Babinet, C.,
Pontoglio, M., Yaniv, M., Barra, J., 2002. Bile system morphogenesis
defects and liver dysfunction upon targeted deletion of HNF1h. Development 129, 1829 – 1838.
Deryckere, F., Byrnes, L., Wagner, A., McMorrow, T., Gannon, F., 1995.
Salmon HNF1: cDNA sequence, evolution, tissue specificity and
binding to the salmon serum albumin promoter. J. Mol. Biol. 247,
1 – 10.
Her, G.M., Yeh, Y.H., Wu, J.L., 2003. 435-bp liver regulatory sequence in
the liver fatty acid binding protein (L-FABP) gene is sufficient to modulate liver regional expression in transgenic zebrafish. Dev. Dyn. 227,
347 – 356.
Horikawa, Y., Iwasaki, N., Hara, M., Furuta, H., Hinokio, Y., Cockburn,
45
B.N., Lindner, T., Yamagata, K., Ogata, M., Tomonaga, O., et al., 1997.
Mutation in hepatocyte nuclear factor-1 beta gene (TCF2) associated
with MODY. Nat. Genet. 17, 384 – 385.
Hörlein, A., Grajer, K.H., Igo-Kemenes, T., 1993. Genomic structure of the
POU-related hepatic transcription factor HNF-1 alpha. Biol. Chem.
Hoppe-Seyler 374, 325 – 419.
Huang, W.T., Gong, H.Y., Lin, C.J.F., Weng, C.F., Chen, M.H.C., Wu, J.L.,
2001. Hepatocyte nuclear factor (HNFs)-1a,-1h and-3h expressed in
the gonad of tilapia (Oreochromis mossambicus). Biochem. Biophys.
Res. Commun. 288, 833 – 840.
Kimmel, C.B., Ballard, W.W., Kimmel, S.R., Ullmann, B., Schilling, T.F.,
1995. Stages of embryonic development of the zebrafish. Dev. Dyn.
203, 253 – 310.
Kumar, S., Tamura, K., Jakobsen, I.B., Nei, M., 2001. MEGA2: molecular evolutionary genetics analysis software. Bioinformatics 17,
1244 – 1245.
Mitchelmore, C., Troelsen, J.T., Spodsberg, N., Sjostrom, H., Noren, O.,
2000. Interaction between the homeodomain proteins Cdx2 and HNF1a
mediates expression of the lactase-phlorizin hydrolase gene. Biochem.
J. 346, 529 – 535.
Okita, K., Yang, Q., Yamagata, K., Hangenfeldt, K.A., Miyagawa, J., Kajimoto, Y., Nakajima, H., Namba, M., Wollheim, C.B., Hanafusa, T.,
Matsuzawa, Y., 1999. Human insulin gene is a target gene of hepatocyte
nuclear factor-1alpha (HNF-1alpha) and HNF-1beta. Biochem. Biophys. Res. Commun. 263, 566 – 569.
Pontoglio, M., Barra, J., Hadchouel, M., Doyen, A., Kress, C., Bach, J.P.,
Babinet, C., Yaniv, M., 1996. Hepatocyte nuclear factor 1 inactivation
results in hepatic dysfunction, phenylketonuria and renal Fanconi syndrome. Cell 84, 575 – 585.
Pontoglio, M., Prie, D., Cheret, C., Doyen, A., Leroy, C., Froguel, P.,
Velho, G., Yaniv, M., Friedlander, G., 2000. HNF1alpha controls
renal glucose reabsorption in mouse and man. EMBO Rep. 1,
359 – 365.
Postlethwait, J.H., Woods, I.G., Ngo-Hazelett, P., Yan, Y.L., Kelly, P.D.,
Chu, F., Huang, H., Hill-Force, A., Talbot, W.S., 2000. Zebrafish comparative genomics and the origins of vertebrate chromosomes. Genome
Res. 10, 1890 – 1902.
Rey-Campos, J., Chouard, T., Yaniv, M., Cereghini, S., 1991. vHNF1 is a
homeoprotein that activates transcription and forms heterodimers with
HNF1. EMBO J. 10, 1445 – 1457.
Ryffel, G.U., 2001. Mutations in the human genes encoding the transcription factors of the hepatocyte nuclear factor (HNF) 1 and HNF4 families: functional and pathological consequences. J. Mol. Endocrinol. 27,
11 – 29.
Shih, D.Q., Bussen, M., Sehayek, E., Ananthanarayanan, M., Shneider,
B.L., Suchy, F.J., Shefer, S., Bollileni, J.S., Gonzalez, F.J., Breslow, J.L.,
Stoffel, M., 2001. Hepatocyte nuclear factor-1alpha is an essential regulator of bile acid and plasma cholesterol metabolism. Nat. Genet. 27,
375 – 382.
Sourdive, D.J.D., Yaniv, M., 1997. The hepatic nuclear factor 1. In: Papavassiliou, A.G. (Ed.), Transcription Factors in Eukaryotes. Landes Bioscience Austin, TX, Chap. 10, pp.189 – 209.
Sun, Z., Hopkins, N., 2001. vhnf1, the MODY5 and familial GCKD-associated gene, regulates regional specification of the zebrafish gut, pronephros, and hindbrain. Genes Dev. 15, 3217 – 3229.
Vignali, R., Poggi, L., Madeddu, F., Barsacchi, G., 2000. HNF1h is required for mesoderm induction in the Xenopus embryo. Development
127, 1455 – 1465.
Wada, S., Tokuoka, M., Shoguchi, E., Kobayashi, K., Di Gregorio, A.,
Spagnuolo, A., Branno, M., Kohara, Y., Rokhsar, D., Levine, M., et
al., 2003. A genomewide survey of developmentally relevant genes in
Ciona intestinalis II. Genes for homeobox transcription factors. Dev.
Genes Evol. 213, 222 – 234.
Westerfield, M., 1995. The Zebrafish Book. A Guide for the Laboratory
Use of Zebrafish (Danio rerio), 3rd ed., University of Oregon Press,
Eugene.
Wiellette, E.L., Sive, H., 2003. vhnf1 and Fgf signals synergize to specify
46
H.-Y. Gong et al. / Gene 338 (2004) 35–46
rhombomere identity in the zebrafish hindbrain. Development 130,
3821 – 3829.
Wild, W., Pogge v. Strandmann, E., Nastos, A., Senkel, S., Lingott-Frieg,
A., Bulman, M., Bingham, C., Ellard, S., Hattersley, A.T., Ryffel, G.U.,
2000. The mutated human gene encoding hepatocyte nuclear factor 1h
inhibits kidney formation in developing Xenopus embryos. Proc. Natl.
Acad. Sci. U. S. A. 97, 4695 – 4700.
Woods, I.G., Kelly, P.D., Chu, F., Ngo-Hazelett, P., Yan, Y.L., Huang, H.,
Postlethwait, J.H., Talbot, W.S., 2000. A comparative map of the zebrafish genome. Genome Res. 10, 1903 – 1914.
Yamagata, K., Oda, N., Kaisaki, P.J., Menzel, S., Furuta, H., Vaxillaire, M.,
Southam, L., Cox, R.D., Lathrop, G.M., Boriraj, V.V., et al., 1996.
Mutations in the hepatocyte nuclear factor-1alpha gene in maturityonset diabetes of the young (MODY3). Nature 384, 455 – 458.