O R I G I N A L A R T I C L E
Yun-Hsin Wang . Yau-Hung Chen . Jun-Hung Lu .
Yu-Ju Lin . Min-Yen Chang . Huai-Jen Tsai
Epidermis-restricted expression of zebrafish cytokeratin II is controlled
by a 141/185 minimal promoter, and cassette 141/ 111 is
essential for driving the tissue specificity
Received November 20, 2005; accepted in revised form January 17, 2006
Abstract
We isolated a 2.3 kb DNA segment from the
upstream region of the zebrafish cytokeratin II (zfCKII)
gene. Transgenic embryos, produced by using a series of
5
0deletions linked to the red fluorescent protein (RFP)
reporter, showed that the
141/185 segment of zfCKII
directed RFP expression in epidermal cells, whereas the
111/185 segment did not. When 141/ 111 was
deleted from
355/185 and microinjected into
one-celled embryos, no fluorescence was observed at later
stages, indicating that the
141/ 111 segment is
re-quired for green fluorescent protein expression in
epider-mal cells. Furthermore, when a putative KLF-binding
site at
119/ 117 was mutated, RFP expression rates
and intensities were reduced dramatically, although still
observed, suggesting that
119/ 117 within 141/
111 is a key cis-element for controlling
epidermis-spe-cific expression of the zfCKII gene. Finally, we generated
a zebrafish transgenic line, Tg(zfCKII(2.3):RFP), which
carries an upstream 2.3 kb regulatory region of the
zfCKII gene fused with RFP. The expression pattern in
the epidermal cells of Tg(zfCKII(2.3):RFP) fish
recapit-ulated that of the endogenous gene. F2 embryos derived
from Tg(zfCKII(2.3):RFP) males crossed with wild-type
females revealed that the earliest onset of RFP expression
was at the sphere stage, indicating that this transgenic
approach can be used for studying zygotic expression of
maternally inherited genes.
Key words
cytokeratin
epidermal cells germline
transmission
red fluorescent protein transgenic
zebrafish
Introduction
There are many families of cytoplasmic filament
pro-teins. Keratin is one such family, having intermediate
size (10 nm) and expression specifically in epithelial cells
as cytoskeletal proteins. At least 30 related members in
the cytokeratin family are known and are encoded by
complex multiple genes. Type I keratins (K9–K20)
are more acidic (pI 5 4–6), whereas type II keratins
(K1-K8) are neutral or basic (pI 5 6–8). Unlike other
intermediate filament (IF) proteins, cytokeratins form
obligatory heteropolymers, consisting of equal numbers
of type I and type II keratins (Fuchs and Weber, 1994;
Klymkowsky, 1995; Rogers et al., 2004) as specific
pairs, and form filaments by the coiled-coil interaction
(Hatzfeld et al., 1987).
There are 19 human keratin isoforms (K1–K19), and
specific pairs of keratin isoforms determine tissue
iden-tity and differentiation state. For example, human
keratins K5 (type II) and K14 (type I) together form the
extensive IF network of mitotically active basal cells in
all stratified epithelia (Byrne et al., 1994). Expression of
human keratin is restricted: E keratins are found in
stratified epithelium (including epidermis), and S
kera-Equal contribution.
Yun-Hsin Wang Jun-Hung Lu Yu-Ju Lin Huai-Jen Tsai ( .*)
Institute of Molecular and Cellular Biology National Taiwan University
No. 1, Sec. 4, Roosevelt Road Taipei 106, Taiwan
Tel: 886-2-3366-2487 Fax: 886-2-2363-8483 E-mail: hjtsai@ntu.edu.tw
Yau-Hung Chen Min-Yen Chang Graduate Institute of Life Sciences Tamkang University,
Tamsui, Taiwan
U.S. Copyright Clearance Center Code Statement:
0301–4681/2006/7404–186 $ 15.00/0
tins are found in simple epithelium. However, in
rabow trout, both E and S keratins are expressed in
in-testinal simple epithelium (Markl and Franke, 1988). In
zebrafish, S keratins are expressed both in stratified
ep-ithelium (including epidermis) and simple epep-ithelium
(Imboden et al., 1997). In addition, fish keratin can be
found in mesenchyme-derived cells and certain nerve
cells (Giordano et al., 1989, 1990), but mesenchymal
tissue usually expresses vimentin in humans. This
evi-dence indicates that fish epidermal keratin diversified
independently from that of mammalian epidermal
ker-atin. Thus, the mechanism of transcriptional regulation
of the keratin in fishes should be studied.
In mammals, the regulation of keratin genes by a
variety of cis-acting elements has been studied. For
example, a keratinocyte-specific enhancer (the 10 bp 5
0-ACCTGCAGGC-3
0) has been identified in the K14 gene
(Vassar et al., 1989; Leask et al., 1990). Nuclear proteins,
such as GKLF/KLF4 and Sp1, which interact with the
upstream region of the human K19 keratin gene, have
also been identified (Brembeck and Rustgi, 2000). Rhodes
and Oshima (1998) reported that the expression of human
and mouse K18 is activated by the transcription factors
c-jun
and c-fos, respectively, by means of an outer AP1 site
in the first intron. Additionally, Popa et al. (2004)
re-ported that the AP-2 transcription factor family is
pre-sumed to play an important role in the regulation of the
keratinocyte squamous differentiation program. There is
extensive knowledge about the transcriptional regulation
of keratin genes in mammals; however, the mechanism of
transcriptional regulation of keratin in fish has not been
reported to date, although fish cytokeratin has been
char-acterized from rainbow trout (Markl and Franke, 1988;
Markl et al., 1989; Schaffeld et al., 2002a, 2002b), goldfish
(Druger et al., 1992), catfish (Ainis et al., 1995), lamprey
(Zaccone et al., 1995), mosquito fish (Arenas et al., 1995),
common carp (Groff et al., 1997), zebrafish (Conrad et al.,
1998; Chua and Lim, 2000; Schaffeld et al., 2003; Wang
et al., 2006b), and shark (Schaffeld et al., 1998).
Previously, a zebrafish type II cytokeratin (zfCKII)
cDNA was isolated and characterized (Chua and Lim,
2000). zfCKII was expressed uniformly in the skin, fins,
scale epidermis, retina, cornea, neurons, and glial cells
of the brain and spinal cord as well as in chondrocytes
of the skull of the adult fish. The tissue-specific
expres-sion of zfCKII is of particular interest because of the
broad diversity of zfCKII-positive epithelial cells. In the
present study, a 2.3 kb zfCKII promoter was isolated
and its function was assayed in both transient and
sta-ble transgenic lines. In stasta-ble transgenic lines, red
flu-orescent protein (RFP) expression faithfully reflected
the expression of the endogenous zfCKII gene.
Micro-injections of serial deletions of the upstream segment of
zfCKII fused with RFP were performed. We found that
a cis-element of the DNA segment in the zfCKII gene
is required for epidermal cell-specific expression of
zfCKII and that a minimal region upstream of the
transcription start site (141 bp) is required for this
ex-pression restricted to epidermal cells.
Materials and methods
Experimental fish
Zebrafish AB strains (wild-type, wt) were kept under a 14 hr light and 10 hr dark photoperiod at approximately 28.51C. After fertilization, the eggs were collected and cultured in an aquarium. Embryonic cleavage and somite formation were observed with a light microscope to determine the developmental stages (Kimmel et al., 1995).
Isolation of the zfCKII upstream regulatory region
The 50-flanking region of the zfCKII gene was isolated according to
the instructions for the Universal GenomeWalker Kit (BD Biosci-ences Palo Alto, CA). Briefly, DNA was extracted from zebrafish embryos (Westerfield, 1995), digested with restriction enzymes, and ligated to a GenomeWalker adaptor to produce GenomeWalker libraries. DNA fragments containing the putative upstream region were isolated after two successive polymerase chain reaction (PCR)-based DNA walking in GenomeWalker libraries. The primary PCR and the outer adapter primer (AP-1) products were obtained using the GeneAmp PCR System with five cycles of 941C for 5 sec and 721C for 3 min, followed by 30 cycles of 941C for 5 sec and 681C for 3 min, and a final extension at 681C for 7 min. For each round of genomic walking, the diluted primary PCR products served as templates for the secondary nested PCR with a nested gene-specific primer (CKIIR) and the nested adaptor primer (AP-2). The sec-ondary PCR products were analyzed and purified from agarose gels and cloned into pGEMT easy Vector (Promega, Madison, WI) to produce pT7CKII. The primers used in the genomic walking and subsequent cloning are listed in Table 1.
50-Rapid amplification of cDNA end (50-RACE)
The protocols of 50-RACE are followed by the manufacturer’s in-structions (FirstChoicet RLM-RACE; Ambion, Austin, TX) with minor revision. Briefly, first-strand cDNA used for 50-RACE was
performed as described before (Chen et al., 2001), and then using 50-RACE outer primer and CKII-271R for the first PCR, and
prim-ers CKII-117R and 50-RACE inner primer for the nesting. Amplified DNA fragments were sub-cloned and sequenced as described above.
Construction of chimeric reporter gene fusions
The upstream segments of zfCKII and of the serial deletion deriv-atives were fused with the RFP reporter gene by cloning the various 50deletion fragments into the polylinker region of the vector
pDs-Red2.1 (Clontech, Palo Alto, CA) upstream of a promoterless RFP gene. To generate pCKII2.3K-RFP, pT7CKII was digested by EcoRI and a 2.3 kb EcoRI-digested fragment was purified and ligated to EcoRI-digested pDsRed2.1 (Clontech). The deletion fragments of 1.7 kb, 1.2 kb, 526 bp, 440 bp, 228 bp, 194 bp, and 164 bp corresponded to the CKII gene from 1709 to 185 ( 1709/185), 1168/185, 441/185, 355/185, 141/185, 110/185, and 80/185, respectively. They were amplified from pT7CKII using forward primers CKII( 1.7K)F, CKII( 1K)F, CKII( 441)F, CKII( 355)F, CKII( 141)F, CKII( 110)F, and CKII( 80)F, respectively, and the common reverse primer CKIIR. These resulting PCR products were cloned into pGEM-T Easy Vector (Promega) to produce pT7CKII1.7K, pT7CKII1K, pT7CKII( 441/185), pT7CKII( 355/185), pT7CKII( 141/ 185), pT7CKII( 110/185), and pT7CKII( 80/185), respective-ly. Then, all plasmids were digested by EcoRI and ligated to EcoRI-digested pDsRed2.1 (Clontech) vectors to generate pCKII1.
7K-RFP, pCKII1K-RFP, pCKII( 441/185)-RFP, pCKII( 355/ 185)-RFP, pCKII( 141/185)-RFP, pCKII( 110/185)-RFP, and pCKII( 80/185)-RFP, respectively. All constructs were con-firmed after sequencing.
To generate pCKII( 560/185)-RFP, a pT7CKII was digested by EcoRI and HindIII, and a 644 bp fragment was ligated to EcoRI–HindIII-digested pDsRed2.1 (Clontech). To generate pCKII( 355/185)D( 141/ 111)-RFP, which contains a frag-ment spanning 355/185 but with 141/ 111 deleted, two PCR products were obtained by amplifying the pT7CKII template using primers of CKIIdel( 141/ 111)F and CKIIR, and CKII( 355)F and CKIIdel( 141/ 111)R. After two PCR prod-ucts were annealed, they served as templates for amplification of primers CKII( 355)F and CKIIR. To generate plasmid pCKII( 141/185)KLFm-RFP, which contains a fragment span-ning 141/185 but containing a mutated KLF-binding site ( 119GGG 117 to 119AAA 117), the two PCR products were annealed after they were amplified from the pT7CKII template us-ing primers of CKII( 141)F and CKII ( 141/ 110)Rm, and primers of CKII ( 141/ 110)Fm and CKIIR. After annealing, the PCR products served as templates for amplification of the mu-tated segment by primers CKII( 141)F and CKIIR. These result-ing PCR products were cloned into pGEM-T Easy Vector (Promega) to produce pT7CKII( 355/185)D( 141/ 111) and pT7CKII( 141/185)KLFm, respectively. Then, pT7CKII( 355/ 185)D( 141/ 111) and pCKII( 141/185)KLFm were digested by EcoRI and ligated to EcoRI-digested pDsRed2.1 (Clontech) vectors to generate pCKII( 355/185)D( 141/ 111)-RFP and pCKII( 141/185)KLFm-RFP, respectively. To construct plasmid pEGFPm( 141/ 111), which contained the 141/ 111 region of CKII fused with a cytomegalovirus (CMV) minimal TATA pro-moter and with the enhanced green fluorescent protein (EGFP) reporter gene, primers of CKII ( 141/ 111)F and CKII ( 141/ 111)R were used and the PCR product was inserted into the EcoRI-digested pEGFPm (Chen et al., 2003).
Preparation of plasmids for microinjection
Plasmids pCKII2.3K-RFP, pCKII1.7K-RFP, pCKII1K-RFP, pCKII( 560/185)-RFP, pCKII( 441/185)-RFP, pCKII( 355/ 185)-RFP, pCKII( 141/185)-RFP, pCKII( 110/185)-RFP,
pCKII( 80/185)-RFP, pCKII( 355/185)D( 141/ 111)-RFP, and pCKII( 141/185)KLFm-RFP were linearized by XhoI and re-covered from a 0.8% low-melting temperature gel (FMC BioProd-ucts, Philadelphia, PA). DNA intensity and gel absorbancy were measured with a GeneQuant II calculator (Pharmacia Biotech, Hong Kong, China). DNA samples were resuspended at a concen-tration of 25 ng/ml in double-distilled water mixed with 0.1% (v/v) phenol red before use.
To get a quantitative view of the relative strength of the promoter constructs, we classified RFP-expressing embryos into three levels based on the expression pattern in the skin epidermis: strong, me-dium, and weak. The strong level included embryos displaying RFP expression in approximately 90% of skin epidermis throughout the body; the medium level included embryos showing RFP expression in 20%–90% of skin epidermis; and the weak level included embryos having RFP-positive signal in less than 20% of the skin epidermis.
Generation of transgenic germline transmitted zebrafish
The 2.3 kb pCKII2.3K-RFP was linearized with XhoI and resus-pended in 0.1 M KCl at a final concentration of 50–100 ng/ml with 0.2% phenol red as a tracer. The DNA solution was microinjected into the cytoplasm of one-cell stage embryos. After microinjection, red fluorescent signal was observed in 24 hr post-fertilization (hpf) embryos. All RFP-positive embryos were raised to adulthood. Pairs of transgenic founder (F0) fish were crossed in a 22 14 13 cm tank. Parental pairs that produced RFP-positive embryos were separated and mated with wt individuals to confirm the putative germline transmission of the parents. At least 200 embryos per cross were examined for the appearance of red fluorescence. After screening, RFP-positive F1 embryos were raised to adulthood and crossed with wt zebrafish to generate a heterozygotic F2 generation.
Embedding and cryosectioning
The embedding and cryosectioning protocols followed were those of Chen and Tsai (2002), with minor modifications. In brief, em-bryos were fixed with 4% paraformaldehyde for 4 hr at 251C, dechorionized, mounted with 5% sucrose containing 1.5% agarose for 1 hr, cut into cubes approximately 5 5 5 mm, and stored in Table 1 Oligonucleotides used in this study
Symbols Nucleotide sequences (50–30)
50RACE outer GCTGATGGCGATGAATGAACACTG
CKII-271R CGAATCCTCCACCACTGCCG
50RACE inner CGCGGATCCGAACACTGCGTTTGCTGGCTTTGATG
CKII-117R GCTGCCCATTGGCACTGCGCTG AP-1 GTAATACGACTCACTATAGGGC AP-2 ACTATAGGGCACGCGTGGT CKIIR CTGTCGTCTACGAGGGGGTGAGG CKII( 2.3K)F GTTCGAACAGTGTATTGTAGTTCCTC CKII( 1.7K)F GACTGTCAGGAACATTAAAAATCGAC CKII( 1K)F GACCAACGGATTAACAATGGGC CKII( 441)F GCACTTAAAGCAAACTGAGGGCCC CKII( 355)F CTGGTTATGTAAATAAGAGGGGC CKII( 141)F GGCGTGTGTATCACTTGGACAGG CKII( 110)F GCAGGACAGAAGCCTGAGGC CKII( 80)F TTGGCCAGGTGAGCCCCTCCC
CKII del ( 141/ 111)F CAGCAAGATGGCAGGACAGAAGCCTGAGGCAGAAAGGG CKII del ( 141/ 111)R CAGGCTTCTGTCCTGCCATCTTGCTGGGCTGGTATGGG CKII ( 141/ 111)Fm AATTGGCGTGTGTATCACTTGGACAAAAAAAAAG CKII ( 141/ 111)Rm AATTCTTTTTTTTTGTCCAAGTGATACACACGCC CKII ( 141/ 111)F AATTGGCGTGTGTATCACTTGGACAGGGGAAAAAG CKII ( 141/ 111)R AATTCTTTTTCCCCTGTCCAAGTGATACACACGCC AP, adaptor primer; 50-RACE, 50-rapid amplification of cDNA end; CK, cytokeratin.
5% sucrose at 41C. All embryos were embedded using Cryomatrix (Shandon, Waltham, MA), and cryosectioning (12 mm thick) was performed with a Microm Cryosector (Heidelberg, Germany).
Microscopy
Transgenic embryos were observed hourly, especially from 1 to 48 hpf, under a stereo dissecting microscope (MZ12, Leica, Wetzlar, Germany) equipped with a fluorescent module having a GFP or DsRed filter cube (Kramer Scientific, Hampton, NJ). Photographs were taken with an S2 Pro digital camera (Fuji, Tokyo, Japan) when embryos developed at specific stages.
Results
Upstream regulatory region of the zebrafish CKII gene
Previous studies have shown that zfCKII mRNA is
ex-pressed uniformly in the skin, fins, scale epidermis,
ret-ina, cornea, neurons, and glial cells of the brain and
spinal cord as well as in chondrocytes of the skull of the
adult fish (Chua and Lim, 2000). To investigate the
molecular mechanism of zfCKII gene expression in
ep-idermal cells, a 2.3 kb DNA segment of the upstream
region of zfCKII was isolated and its partial DNA
se-quence was determined (Fig. 1; the sese-quence
2219/
192, GenBank Accession No. DQ144236). We also
performed a 5
0-RACE experiment to determine the
transcription start site and the length of the 5
0-untrans-lated region of ZFCKII gene. As shown in Fig. 1, the
numbers indicate nucleotide positions with the
tran-scription start site as 11. Using a string-based search
query, the TESS program (http://www.cbil.upenn.edu/
tess/) revealed that the putative binding sites of the
transcription factors within the proximal
441/192
segment were the AP1 site (CATGAATCATA,
409/
399), the Oct-1 site (TGGTTTGCAT, 379/ 370),
two Sp1 sites (GGGTGTGGC,
196/ 188; and
GG-GGATTGGGCC,
85/ 75), and a TATA box
(TA-TAAAA,
46/ 40; Fig. 1).
Transient expression of RFP is driven by the upstream
2.3 kb segment of the zfCKII gene in zebrafish embryos
The tissue-specific expression mediated by the isolated
upstream segment of zfCKII was investigated using a
transient transgenic approach. The earliest RFP
ex-pression in the embryos microinjected with
pCKII2.3K-RFP was in the head region at 19 hpf (data not shown).
At 3 days post-fertilization (dpf), the RFP signals were
extended to the body, including the head (Fig. 2A),
skin epidermis, pectoral fins (Fig. 2B), and caudal fins
(Fig. 2C). At 14 dpf, the red fluorescence was observed
in the anal fins and in the retina (Figs. 2D, 2F). These
transient expression studies indicated that the upstream
segment in the zfCKII2.3K contained regulatory
ele-ments that drove RFP expression in a tissue-specific
manner.
Functional analysis of the zfCKII regulatory sequence
To determine whether the cis-acting element is sufficient
and required for the epidermis-specific expression of
zfCKII, serial deletions of an upstream region of
zfCKII were generated and fused with RFP cDNA.
These constructs were microinjected into zebrafish
fer-tilized eggs. Embryos injected with DNA fragments
containing the
2219, 1709, 1068, 560, 441,
355, and 141 bp upstream sequences [pCKII2.3K-,
pCKII1.7K-, pCKII1K-, pCKII(
560/185)-, pCKII
(
441/185)-, pCKII( 355/185)-, and pCKII( 141/
185)-RFP, respectively] displayed high RFP expression
rates (63%–85%, Table 2) and 100% RFP-positive
Fig. 1 Schematic representation of sequences in the upstream regions of the zebrafish cyto-keratin II (zfCKII) gene. The empty, crossed, and dotted boxes represent the upstream reg-ulatory regions (including 50-untranslated
re-gion, 50-UTR), the 141/ 111 cassette, and
the putative TATA box of the zfCKII gene, respectively. The numbers below the boxes indicate the nucleotide positions, with refer-ence to the transcription start site (11). The detailed sequences of the 441/192 segment are shown on the lower panel. The putative transcription factors or their binding elements are shown below each corresponding se-quence.
signals in the skin epidermis at 3 dpf (Fig. 3A and
Table 2). To obtain a quantitative view of the relative
strength of the promoter constructs, we classified
RFP-expressing embryos into three levels based on the
ex-pression pattern in the skin epidermis: strong, medium,
and weak with reference to the previous studies (Ju
et al., 2003; Lin et al., 2004). The strong level included
embryos displaying RFP expression in approximately
90% of the skin epidermis throughout the body; the
medium level included embryos showing RFP
expres-sion in 20%–90% of skin epidermis; and the weak level
included embryos having RFP-positive signal in less
than 20% of the skin epidermis (Fig. 3C). As shown in
Fig. 3B, embryos displayed relatively high RFP
expres-sion rates in strong (59%–78%) and medium (20%–
38%) levels when the injected upstream sequences were
longer than
141 bp. In addition to red fluorescent
skins, embryos injected with DNA fragments
contain-ing
1709, 1068, 560, and 441 bp upstream
se-quences also displayed high expression rates (49%–
68%) of RFP-positive signals in muscle (Table 2; Fig.
3C).
However,
the
pCKII(
355/185)-RFP- and
pCKII(
141/185)-RFP-injected embryos displayed
red fluorescence in 1.5% and 3.2% of muscle fibers
(Table 2; Fig. 3C).
It was found that the
141 bp upstream sequence
was able to drive skin-restricted expression of zfCKII.
On the other hand, embryos injected with DNA
frag-ments containing less than 110 bp [pCKII(
110/185)-RFP and pCKII(
80/185)-RFP] did not give off
RFP-positive signals (Fig. 3), even when the
concen-tration of injected DNA was increased to 100 ng/ml
(data not shown). On the basis of this evidence, we
suggest that the proximal part of the
141/185
se-quence of zfCKII is a minimal cis-element for
control-ling specific expression.
Cassette
141/ 111 was able to direct epidermal
cell-specific expression
Interestingly, we found that, when a DNA fragment, in
which
141/ 111 was deleted from the zfCKII
355/185 [pCKII( 355/185)D( 141/ 111)-RFP],
was injected into one-celled embryos, only 1% (two of
182, Table 2) of surviving embryos displayed red
fluo-Table 2 Deletion analysis of zfCKII promoter in transient transgenic zebrafish embryos
Constructs Injected Survival1 Expression2 Skin3 Muscle3 pCKII2.3K-RFP 271 204 (75%) 173 (85%) 173 (100%) 0 pCKII1.7K-RFP 303 266 (88%) 217 (81%) 217 (100%) 113 (52%) pCKII1K-RFP 325 218 (67%) 164 (75%) 164 (100%) 105 (64%) pCKII( 560/185)-RFP 288 201 (70%) 158 (78%) 158 (100%) 77 (49%) pCKII( 441/185)-RFP 249 199 (79%) 135 (68%) 135 (100%) 92 (68%) pCKII( 355/185)-RFP 306 211 (69%) 133 (63%) 133 (100%) 2 (1.5%) pCKII( 141/185)-RFP 253 189 (74%) 154 (81%) 154 (100%) 5 (3.2%) pCKII( 110/185)-RFP 204 163 (80%) 0 0 0 pCKII( 80/185)-RFP 198 139 (70%) 0 0 0 pCKII( 355/185) 4 ( 141/ 111)-RFP 221 182 (82%) 2 (1%) 0 2 (100%) pCKII( 141/185)KLFm-RFP 203 143 (70%) 42 (29%) 42 (100%) 2 (5%) pEGFPm( 141/ 111) 190 150 (79%) 67 (45%) 67 (100%) 7 (10%) pEGFPm 156 136 (87%) 3 (2%) 0 3 (100%)
1Percentages indicate survival rate of injected embryos. 2
Percentages indicate RFP-expressing embryos among surviving embryos at 3 dpf.
3
Numbers in parentheses represent percentages of RFP-expressing embryos for a particular tissue. dpf, days post-fertilization; RFP, red fluorescent protein; zfCKII, zebrafish cytokeratin II.
Fig. 2 Transient red fluorescent protein (RFP) expression patterns of the embryos injected with the 2.3 kb upstream region of the zebrafish cytokeratin II gene. (A) Lat-eral view of the head region. (B) Dorsal view of the same embryo in A. (C) Lateral view of the caudal region. (D) Lateral view of the anal fin (darkfield). (E) The same embryo as in D, observed under brightfield. (F) Lateral view of the eye. Developmental stages: (A)– (C), 3 days post-fertilization (dpf); (D)–(F), 14 dpf.
rescence in muscle, and none of the 182 embryos
dis-played RFP-positive skin epidermis (Table 2). To
determine whether cassette
141/ 111 was able to
direct epidermal cell-specific expression of the reporter
gene, we used plasmid pEGFPm(
141/ 111), in
which a mini-promoter of CMV (TATA box only)
was fused with GFP and one copy of the
141/ 111
cassette (Fig. 4A). Only 2% of the 136 embryos injected
with pEGFPm were GFP-positive, and none had signal
detected in the skin (Fig. 4B, left panel; Table 2).
How-ever, the expression rate of epidermal cell-specificity in
the transgenic embryos injected with pEGFPm(
141/
111) was 45% (Fig. 4B, right panel; Table 2).
Furthermore, a mutation plasmid, pCKII(
141/185)
KLFm-RFP, in which
119GGG
117was mutated to
119
AAA
117, was injected into the embryos and
re-vealed a 29% RFP expression rate (Table 2).
Interest-ingly, most of the pCKII(
141/185)KLFm-RFP-injected embryos displayed medium (43%) and weak
(52%) RFP expression, and only 5% displayed strong
RFP expression (Fig. 3B). Based on these observations,
we propose that cassette
141/ 111 was able to direct
epidermal cell-specific expression and that
119GGG
117is an important element within cassette
141/ 111.
RFP expression in germline zfCKII-RFP transgenic
zebrafish
To generate stable transgenic lines for further analysis
of the expression mechanism of zfCKII, embryos
in-jected with pCKII2.3K-RFP and exhibiting RFP
expression were collected and raised to adulthood.
Founder fishes (F0) were crossed with wt or crossed
with each other to generate F1 embryos. Of 63 founder
fish tested, one line, Tg(zfCKII(2.3):GFP), produced
embryos that expressed RFP in the epidermis, but 5%
(11 positive of 234) of embryos expressed RFP at
24 hpf, suggesting that the transgenic founder is a
germ-line mosaic, which is a common finding in transgenic
fish at the founder generation (reviewed by Gong and
Hew, 1995). The F2 inheritance rate of RFP-positive F1
individuals of the Tg(zfCKII(2.3):RFP) line is 48.57%
(102 of 210 total embryos), indicating that there was a
single insertion site of the transgene in the genome.
All F2 embryos (154 of 154 embryos) derived from an
F1 female crossed with a wt male exhibited red
fluo-rescent signal at one-cell (Fig. 5A), blastula (Fig. 5B),
and segmentation (Fig. 5C) stages. Red fluorescence
became stronger and extended to the body at 3 dpf (Fig.
5D). In contrast, none of the F2 embryos derived from
an F1 male crossed with a wt female showed red
flu-orescence at the one-cell (Fig. 5E) and blastula stages
(Fig. 5F), yet 46.9% of 145 embryos began to display
red fluorescence at 14 hpf (data not shown). The red
fluorescence became stronger at 24 hpf (Fig. 5G),
and extended to the trunk in a non-uniform
expres-sion manner (Fig. 5H). This evidence strongly
demon-strates that the zfCKII transcripts are maternally
inherited.
Cryosectioning of 4 dpf embryos derived from the
Tg(zfCKII(2.3):RFP) line revealed RFP signal
distrib-uted in the retina (Fig 5I), head epidermis (Fig. 5I),
trunk epidermis (Fig. 5K), and pectoral fins (Fig. 5K).
In addition, we think the faint red fluorescent signal in
the brain in the 4 dpf larva section (Fig. 5I) represents
real signals reflecting the endogenous gene expression.
In the 1-month-old juveniles, red fluorescence was also
observed on the body surface, on the cornea (Fig. 5M),
and in the epidermal cells of the scales (Fig. 5N). This
red fluorescence was found only in the outer region
of the scale (Fig. 5O). From these observations, we
conclude that Tg(zfCKII(2.3):RFP) fish recapitulate the
endogenous zfCKII expression patterns.
Fig. 3 Red fluorescent protein (RFP) ex-pression levels of transgenic zebrafish em-bryos injected with various upstream regions of the zebrafish cytokeratin II (zfCKII) gene. (A) Schematic representation of DNA constructs containing various sizes of the upstream regulatory sequences of the zfCKII gene. Crossed, blue, and red boxes indicate the cassette 141/ 111, TATA box, and RFP, respectively. (B) The boxes represent the three expression levels of RFP fluorescence (strong, medium, and weak), and the numbers represent the percentage of cells expressing RFP. (The three levels are defined in the Materials and Methods sec-tion.) (C) Embryos injected with the DNA constructs listed in (A) displayed different RFP expression patterns and expression lev-els. The fluorescence was checked and ob-served when the embryos developed at 3 days post-fertilization. Arrows indicate the positions of muscle fibers.
Discussion
In this study, we isolated a 2.3 kb upstream regulatory
sequence of the zfCKII gene. Analyses of transient as
well as stable transgenic lines revealed that this 2.3 kb
segment is able to recapitulate the endogenous zfCKII
expression patterns. Deletion analyses were performed
and showed that a proximal
141/185 sequence of
zfCKII is a minimal cis-element for controlling
expres-sion specificity. Finally, we identified a 31 bp (
141/
111) segment, which is able to drive GFP expression
in the skin epidermis.
Many transcription factors (c-jun, c-fos, C/EBP, and
SP-1) and their binding sites (AP1, CCAAT-, and
GC-box) have been reported to play important roles in
regulating mammal keratin genes at transcription
(Brembeck and Rustgi, 2000; Sterneck et al., 2006;
Wang et al., 2006). As shown in Fig. 1, AP1, CCAAT-,
and GC-box exist in the
441/185 segment of the
zfCKII gene. We examined the putative transcription
factor-binding sites in zebrafish k8 (now renamed as k4)
and k18 genes (Gong et al., 2002; Wang et al., 2006) and
found that AP1, CCAAT-, and GC-box are located in
the upstream regions of k4 and k18. Thus, we propose
that zebrafish keratin genes zfCKII, k4, and k18 shared
some common transcriptional regulatory mechanisms
with those of keratin genes in mammals.
Systematic analyses of mammal keratin revealed that
the tissue specificity is most probably conferred by
se-quences close to the TATA box. But in some cases, the
enhancers for directing tissue specificity are located at
the first intron (Rhodes and Oshima, 1998) and the 3
0-flanking sequence (Hu and Gudas, 1994). A transgenic
analysis of the human k5 promoter revealed that 90 bp
of the 5
0-flanking sequences contain sufficient
informa-tion to direct expression to keratinocytes (Byrne and
Fuchs, 1993). Mahony et al. (2000) reported that a
Fig. 4 Cassette 141/ 111 is able to direct skin-specific expres-sion of the cytomegalovirus (CMV) basal promoter. (A) Plasmids pEGFPm and pEGFPm ( 141/ 111), including a minimal TATA box derived from a CMV promoter fused with the en-hanced green fluorescent protein (EGFP) gene and one copy of cassette 141/ 111 fused with pEGFPm, respectively. (B) Em-bryos were photographed under blue light equipped with a green fluorescent protein (GFP) filter. In pEGFPm- (left panel) and pEGFPm( 141/ 111)-injected (right panel) zebrafish, enhanced GFP (EGFP) signals appeared in the skin epidermis. The uniform green staining seen in the embryo injected with the pEGFPm control construct (left panel) is due to the auto-fluorescence of the embryo. The fluorescence was checked and observed when the embryos developed at 3 days post-fertilization (dpf).
Fig. 5 Red fluorescent protein (RFP) expres-sion in transgenic fish lines. F2 offspring were produced by mating F1 females from the Tg(zfCKII(2.3):RFP) line with wild-type (wt) males (A–D) or by mating the F1 males from the Tg(zfCKII(2.3):RFP) line with wt females (E–H). (A, E) At the one-cell stage. (B, F) At 4 hr post-fertilization (hpf). (C, G) At 24 hpf. (D, H) At 3 days post-fertilization (dpf). (I) Cross-section of the head region at 4 dpf. (J) Brightfield observation of I. (K) Cross-section of the trunk region at 4 dpf. (L) Brightfield observation of K. (M) Eye of a 1-month-old juvenile. (N) A scale of a 1-month-old juve-nile. (O) Darkfield observation. zfCKII, ze-brafish cytokeratin II.
120 bp mouse keratin 6a mini-gene contains sufficient
sequence information to direct uniform and
tissue-specific expression. In zfCKII, we showed that the
proximal
141/185 sequence of zfCKII is a minimal
cis-element and that a 31 bp (
141/ 111) segment
is capable of driving GFP expression in the skin
epi-dermis. Sequence analysis of
141/ 111 segment
revealed that there is one KLF-binding site (
122CA-GGGG
117). KLF has been reported to be involved in
regulating tissue-dependent transcription of the keratin
19 gene (Brembeck and Rustgi, 2000). We have shown
a mutation plasmid, pCKII(
141/185)KLFm-RFP,
in which
119GGG
117has been mutated to
119AAA
117, resulting in an altered KLF-binding site.
Comparing the pCKII(
141/185)-RFP- with the
pCKII(
141/185)KLFm-RFP-injected
groups,
we
found that the RFP expression rates were significantly
reduced (81% versus 29%, respectively; Table 2). Only
5% of pCKII(
141/185)KLFm-RFP-injected
embry-os displayed strong RFP expression, a rate much lower
than that of the pCKII(
141/185)-RFP-injected
group (78%, Fig. 3B). These observations strongly
sug-gest that
122CAGGGG
117might be an enhancer for
the zfCKII gene.
Embryos injected with pCKII2.3K-RFP displayed
nor-mal skin-specific expression (Table 2). However, embryos
injected with pCKII1.7K-, pCKII1K-, pCKII(
560/
185)-, pCKII(
441/185)-, pCKII( 355/185)-, and
pCKII(
141/185)-RFP displayed ectopic expression in
muscle (Table 2). Thus, we propose that there is a muscle
repressor in the
2.3 to 1.7 kb region.
Chua and Lim (2000) showed that zfCKII mRNA
was inherited maternally. The mechanism of
transcrip-tional regulation of maternally inherited genes is
diffi-cult to study because of the interference of maternally
produced mRNA. Here, for the first time, we
demon-strate that an upstream 2.3 kb segment of the zfCKII
gene is able to drive zfCKII expression before
mid-blastula transition and is sufficient to recapitulate
the endogenous zfCKII transcription. We used the
Tg(zfCKII(2.3):RFP) line for studying the regulatory
mechanism. F2 embryos derived from Tg(zfCKII(2.3):
RFP) males crossed with wt females displayed red
flu-orescence without an interfering maternal effect. Thus,
this fish line can be used as an excellent tool for
stud-ying zygotic expression of maternally inherited genes. In
fact, we can knock-down several putative transcription
factors, such as KLF, c-Jun/Fos, or SP1, by injecting the
morpholinos into the F2 embryos that were produced by
mating Tg(zfCKII(2.3):RFP) males with wt females to
study the zygotic regulation of zfCKII. This transgenic
line should provide new insights into zfCKII expression
at the transcription level in early embryogenesis.
Acknowledgments This work was supported by the National Sci-ence Council, Republic of China, under Grant numbers NSC
93-2313-B-032-003 (Y. H. C.), AOC 93-9.2.4-FI(2), and NSC 91-2811-B-002-029 (H. J. T.).
References
Ainis, L., Tagliafierro, G., Mauceri, A., Licata, A., Ricca, M.B. and Fasulo, S. (1995) Cytokeratin type distribution in the skin and gill epithelia of the Indian fresh-water catfish, Heteropenustes fossilisas detected by immunohistochemistry. Folia Histochem Cytobiol 33:77–81.
Arenas, M.I., Fraile, B., de Miguel, M. and Paniagua, R. (1995) Intermediate filaments in the testis of the teleost mosquito fish (Gambusia affinis holbrooki): a light and electron microscopy immunocytochemical study and western blotting analysis. Histochem J 27:329–337.
Brembeck, F.H. and Rustgi, A.K. (2000) The tissue-dependent keratin 19 gene transcription is regulated by GKLF/KLF4 and Sp1. J Biol Chem 275:28230–28239.
Byrne, C. and Fuchs, E. (1993) Probing keratinocyte and differen-tiation specificity of the human K5 promoter in vitro and in transgenic mice. Mol Cell Biol 13:3176–3190.
Byrne, C., Tainsky, M. and Fuchs, E. (1994) Programming gene expression in developing epidermis. Development 120:2369–2383. Chen, Y.H., Lee, H.C., Liu, C.F., Lin, C.Y. and Tsai, H.J. (2003) Novel regulatory sequence 82/ 62 functions as a key element to drive the somite-specificity of zebrafish myf-5. Dev Dyn 228:41–50.
Chen, Y.H., Lee, W.C., Liu, C.F. and Tsai, H.J. (2001) Molecular structure, dynamic expression and promoter analysis of zebrafish (Danio rerio) myf-5 gene. Genesis 29:22–35.
Chen, Y.H. and Tsai, H.J. (2002) Treatment with myf5-morpholino results in somite patterning and brain formation defects in zebrafish. Differentiation 70:447–456.
Chua, K.L. and Lim, T.M. (2000) Type I and type II cytokeratin cDNAs from the zebrafish (Danio rerio) and expression patterns during early development. Differentiation 66:31–41.
Conrad, M., Lemb, K., Schubert, T. and Markl, J. (1998) Biochem-ical identification and tissue-specific expression patterns of keratins in the zebrafish (Danio rerio). Cell Tissue Res 293:195–205. Druger, R.K., Levine, E.M., Glasgow, E., Jones, P.S. and
Schech-ter, N. (1992) Cloning of a type I keratin from goldfish optic nerve: differential expression of keratins during regeneration. Differentiation 52:33–43.
Fuchs, E. and Weber, K. (1994) Intermediate filaments: structure, dynamics, function, and disease. Annu Rev Biochem 63: 345–382.
Giordano, S., Glasgow, E., Tesser, P. and Schechter, N. (1989) A type II keratin is expressed in glial cells of the goldfish visual pathway. Neuron 2:1507–1516.
Giordano, S., Hall, C., Quitschke, W., Glasgow, E. and Schechter, N. (1990) Keratin 8 of simple epithelia is expressed in glia of the goldfish nervous system. Differentiation 44:163–172.
Gong, Z. and Hew, C.L. (1995) Transgenic fish in aquaculture and developmental biology. Curr Top Dev Biol 30:177–214. Gong, Z., Ju, B., Wang, X., He, J., Wan, H., Sudha, P.M. and Yan,
T. (2002) Green fluorescent protein expression in germ-line transmitted transgenic zebrafish under a stratified epithelial pro-moter from keratin 8. Dev Dyn 223:204–215.
Groff, J.M., Naydan, D.K., Higgins, R.J. and Zinkl, J.G. (1997) Cytokeratin-filament expression in epithelial and non-epithelial tissues of the common carp (Cyprinus carpio). Cell Tissue Res 287:375–384.
Hatzfeld, M., Maier, G. and Franke, W.W. (1987) Cytokeratin domains involved in heterotypic complex formation determined by in-vitro binding assays. J Mol Biol 197:237–255.
Hu, L. and Gudas, L.J. (1994) Activation of keratin 19 gene ex-pression by a 30 enhancer containing an AP1 site. J Biol Chem
Imboden, M., Goblet, C., Korn, H. and Vriz, S. (1997) Cytokeratin 8 is a suitable epidermal marker during zebrafish development. C R Acad Sci III 320:689–700.
Ju, B., Chong, S.W., He, J., Wang, X., Xu, Y., Wan, H., Tong, Y., Yan, T., Korzh, V. and Gong, Z. (2003) Recapitulation of fast skeletal muscle development in zebrafish by transgenic ex-pression of GFP under the mylz2 promoter. Dev Dyn 227: 14–26.
Kimmel, C.B., Ballard, W.W., Kimmel, S.R., Ullmann, B. and Schilling, T.F. (1995) Stages of embryonic development of the zebrafish. Dev Dyn 203:253–310.
Klymkowsky, M. (1995) Intermediate filaments: new proteins, some answers, more questions. Curr Opin Cell Biol 7:46–54. Leask, A., Rosenberg, M., Vassar, R. and Fuchs, E. (1990)
Reg-ulation of a human epidermal keratin gene: sequences and nu-clear factors involved in keratinocyte-specific transcription. Genes Dev 4:1985–1998.
Lin, C.Y., Chen, Y.H., Lee, H.C. and Tsai, H.J. (2004) A novel cis-element in intron 1 represses somite expression of zebrafish myf5. Gene 334:63–72.
Mahony, D., Karunaratne, S., Cam, G. and Rothnagel, J.A. (2000) Analysis of mouse keratin 6a regulatory sequences in transgenic mice reveals constitutive, tissue-specific expression by a keratin 6a minigene. J Invest Dermatol 115:795–804.
Markl, J. and Franke, W.W. (1988) Localization of cytokeratins in tissues of the rainbow trout: fundamental differences in expres-sion pattern between fish and higher vertebrates. Differentiation 39:97–122.
Markl, J., Winter, S. and Franke, W.W. (1989) The catalog and the expression complexity of cytokeratins in a teleost fish, the rain-bow trout. Eur J Cell Biol 50:1–16.
Popa, C., Dahler, A.L., Serewko-Auret, M.M., Wong, C.F., Smith, L., Barnes, L.M., Strutton, G.M. and Saunders, A.N. (2004) AP-2 transcription factor family member expression, activity, and regulation in human epidermal keratinocytes in vitro. Differen-tiation 72:185–197.
Rhodes, K. and Oshima, R.G. (1998) A regulatory element of the human keratin 18 gene with AP-1-dependent promoter activity. J Biol Chem 273:26534–26542.
Rogers, A.M., Winter, H., Langbein, L., Bleiler, R. and Schweizer, J. (2004) The human type I keratin gene family: characterization of new hair follicle specific members and evaluation of the chro-mosome 17q21.2 gene domain. Differentiation 72:527–540. Schaffeld, M., Haberkamp, M., Braziulis, E., Lieb, B. and Markl, J.
(2002a) Type II keratin cDNAs from the rainbow trout: impli-cations for keratin evolution. Differentiation 70:292–299. Schaffeld, M., Hoffling, S., Haberkamp, M., Conrad, M. and Markl,
J. (2002b) Type I keratin cDNAs from the rainbow trout: inde-pendent radiation of keratins in fish. Differentiation 70:282–291. Schaffeld, M., Knappe, M., Hunzinger, C. and Markl, J. (2003) cDNA sequences of the authentic keratins 8 and 18 in zebrafish. Differentiation 71:73–82.
Schaffeld, M., Lobbecke, A., Lieb, B. and Markl, J. (1998) Tracing keratin evolution: catalog, expression patterns and primary structure of shark (Scyliorhinus stellaris) keratins. Eur J Cell Biol 77:69–80.
Sterneck, E., Zhu, S., Jorcano, J.L. and Smart, R.C. (2006) Conditional ablation of C/EBPbeta demonstrates its keratin-ocyte-specific requirement for cell survival and mouse skin tumorigenesis. Oncogene 25:1272–1276.
Vassar, R., Rosenberg, M., Ross, S., Tyner, A. and Fuchs, E. (1989) Tissue-specific and differentiation-specific expression of a human K14 keratin gene in transgenic mice. Proc Natl Acad Sci USA 86:1563–1567.
Wang, Y.N., Chen, Y.J. and Chang, W.C. (2006a) Activation of ERK signaling by epidermal growth factor mediates c-Jun ac-tivation and p300 recruitment in keratin 16 gene expression. Mol Pharmacol 69:85–98.
Wang, Y.H., Chen, Y.H., Lin, Y.J. and Tsai, H.J. (2006b) Spatio-temporal expression of zebrafish keratin 18 during early em-bryogenesis and the establishment of a keratin 18:RFP transgenic line. Gene Expr Patterns 6:335–339.
Westerfield, M. (1995) The Zebrafish Book. 3rd ed. University of Oregon Press, Eugene.
Zaccone, G., Howi, A.J., Mauceri, A., Fasulo, S., Lo Casio, P. and Youson, J.H. (1995) Distribution patterns of cytokeratins in ep-idermis and horny teeth of the adult sea lamprey, Petromyzon marinus. Folia Histochem Cytobiol 33:69–75.
