Báo cáo khoa học: Cellular retinol-binding protein type II (CRBPII) in adult zebrafish (Danio rerio) cDNA sequence, tissue-specific expression and gene linkage analysis pptx

Cellular retinol-binding protein type II (CRBPII) in adult zebrafish
(
Danio rerio
)
cDNA sequence, tissue-specific expression and gene linkage analysis
Marianne C. Cameron
1
, Eileen M. Denovan-Wright
2
, Mukesh K. Sharma
1
and Jonathan M. Wright
1
1
Department of Biology, and
2
Department of Pharmacology, Dalhousie University, Halifax, Nova Scotia, Canada
We have determined the nucleotide sequence of a zebrafish
cDNA clone that codes for a cellular retinol-binding protein
type II (CRBPII). Radiation hybrid mapping revealed that
the zebrafish and human CRBPII genes are located in
syntenic groups. In situ hybridization and emulsion autora-
diography localized the CRBPII mRNA to the intestine and
the liver of adult zebrafish. CRBPII and intestinal fatty acid
binding protein (I-FABP) mRNA was colocalized to the
same regions along the anterior-posterior gradient of the
zebrafish intestine. Similarly, CRBPII and I-FABP mRNA
are colocalized in mammalian and chicken intestine.
CRBPII mRNA, but not I-FABP mRNA, was detected in
adult zebrafish liver which is in contrast to mammals where
liver CRBPII mRNA levels are high during development but

rapidly decrease to very low or undetectable levels following
birth. CRBPII and I-FABP gene expression appears there-
fore to be co-ordinately regulated in the zebrafish intestine as
has been suggested for mammals and chicken, but CRBPII
gene expression is markedly different in the liver of adult
zebrafish compared to the livers of mammals. As such,
retinol metabolism in zebrafish may differ from that of
mammals and require continued production of CRBPII in
adult liver. The primary sequence of the coding regions of
fish and mammalian CRBPII genes, their relative chromo-
somal location in syntenic groups and possibly portions of
the control regions involved in regulation of CRBPII gene
expression in the intestine appear therefore to have been
conserved for more than 400 million years.
Keywords: Danio rerio; fatty acid binding protein; cellular
retinol binding protein; tissue-specific expression; retinol
metabolism.
Cellular retinol-binding proteins (CRBPs) are members of
the intracellular lipid-binding protein family which includes
the retinoic acid (CRABP) and fatty acid (FABP) binding
proteins. This family consists of low molecular mass ( 14–
16 kDa) polypeptides that bind and transport retinoids,
fatty acids, and bile salts [1,2]. Members of this protein
superfamily have a common three dimensional shape
described as a clamshell structure composed of two
orthogonal b-sheets, each consisting of five antiparallel
b-strands and two a-helices [3]. Hydrophobic ligands are
held in the central cavity of the bivalve-like polypeptide.
The three CRBPs, type I, II, and III, are named according
to the order in which they were discovered in mammals.

Their putative role in cell physiology is in the metabolism of
retinol (vitamin A). Retinol and its derivates are important
for vision, reproduction, metabolism, cellular differentiation
and pattern formation during embryogenesis [4]. After
absorption in the mammalian intestine, the enzyme
b-carotene dioxygenase catalyzes the oxidative cleavage of
b-carotene to retinal. Retinal is reduced to retinol by the
enzyme retinal dehydrogenase. Retinol is then esterified by
the microsomal enzyme lecithin:retinal acyltransferase
(LRAT) to retinoic acid and packaged into chylomicrons
for subsequent uptake by the liver. CRBPI and II bind
retinal and retinol whereas CRBPIII binds retinol, but not
retinal [4–7]. CRABPs bind and transport retinoic acid
[4,5,7]. The CRBPs are thought to participate in ligand
binding and regulate the metabolism of retinal and retinol
while protecting the CRBP-bound ligands from nonspecific
reactions [8]. Biochemical studies have shown that CRBPII-
bound retinol serves as a substrate for the enzyme, LRAT,
implicating CRBPII as a component in directing and
channeling dietary retinol to nascent chylomicroms. Direct
evidence for the role of CRBPI in vitamin A metabolism has
been provided by transgenic knockout studies which
demonstrated that CRBPI knockout mice are phenotypi-
cally normal when fed a vitamin A-enriched diet, but when
the diet is deficient in vitamin A, stores of retinyl esters are
depleted over 5 months and the mice develop abnormalities
consistent with postnatal hypovitaminosis A [9]. To date,
there are no reports of CRBPII and CRBPIII gene
knockouts and therefore direct evidence for their function
in vitamin A metabolism remains speculative. CRBPII is

restricted to the small intestine of adult rat [10], human [11],
chicken [12,13] and pig [14]. Studies have shown that the
small intestine has the highest levels of CRBPII mRNA in
Correspondence to J. M. Wright, Department of Biology,
Dalhousie University, Halifax, Nova Scotia, Canada B3H.
Fax: + 902 494 3736, Tel.: + 902 494 6468,
E-mail:
Abbreviations: FABP, fatty acid binding protein; cDNA, comple-
mentary DNA; CRBP, cellular retinol-binding protein; EST,
expressed sequence tag; LG, linkage group; LRAT, lecithin:retinal
acyltransferase; rbp2, zebrafish retinol binding protein 2 gene;
RBP2, human retinol binding protein2 cellular gene.
Note: web site available at />(Received 8 April 2002, revised 5 July 2002, accepted 5 August 2002)
Eur. J. Biochem. 269, 4685–4692 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03170.x
adult rats, with undetectable or low levels of CRBPII
mRNA in other tissues [4,5]. Presumably, CRBPII is
directly involved in the intestinal uptake and binding of
retinol based on calculated rates of retinol uptake in a
human intestinal cell culture model [15].
Members of the intracellular lipid-binding protein super-
family are derived from at least 14 gene duplications [16].
Prior to the vertebrate/invertebrate split, the liver/intestinal/
ileal FABP and heart/adipose/mylein FABP clades diverged
approximately 700 million years ago. It has been suggested
that the CRBP genes diverged from the liver/intestinal/ileal
FABP clade about 500 million years ago. The mammalian
CRBPI and CRBPII genes presumably arose by gene
duplication sometime after the divergence of amphibia from
mammals as a single copy of the CRBP gene is found in
Xenopus [17]. An alternative explaination, however, is that

one of the duplicated copies of the CRBP gene may have
been lost in the amphibian lineage.
The structure and function of fatty acid and retinoid-
binding proteins have been studied extensively in mammals,
but only superficially in other taxa such as the teleost fishes.
Vitamin A and its derivates are clearly important mediators
of normal vertebrate development [4,8,9]. As zebrafish is
promoted as a model experimental system for study of
vertebrate development, an understanding of the function of
CRBPs in vitamin A metabolism during zebrafish embryo-
genesis would be of interest to developmental biologists.
Moreover, comparative studies of CRBP gene expression in
fishes and mammals may provide insight into the role(s) of
these intracellular retinol- and retinal-binding proteins in
vitamin A metabolism. As part of ongoing studies in our
laboratories on the evolution, tissue-specific expression and
gene regulation of members of the intracellular lipid-binding
protein family in zebrafish [18–20], we have determined the
nucleotide sequence of a cDNA clone and deduced the
amino-acid sequence for a zebrafish CRBPII. Furthermore,
we report the tissue-specific distribution of the CRBPII
mRNA in adult zebrafish and assignment of the CRBPII
gene to linkage group 15 in the zebrafish genome.
MATERIALS AND METHODS
Searches of the zebrafish EST database in GenBank
identified a cDNA clone (GenBank accession number
AI544932) that was similar to the 5¢ end of the rat cellular
retinol-binding protein type II. This clone (fb69e02.y1) was
purchased from Incyte Genomics Inc. and the complete
nucleotide sequence was determined [18]. The deduced

amino-acid sequence of the cDNA sequence was aligned
with other intracellular lipid-binding protein sequences in
GenBank using
CLUSTALW
[21] and an output of percentage
sequence identity generated.
DNA from the LN54 radiation hybrid panel [22]
(zebrafish DNA in a mouse background) was used
as template in PCR reactions to assign the linkage group
for the CRBPII gene in the zebrafish genome (see Fig. 1 for
primer location). PCR reactions contained 1X PCR buffer
(MBI Fermentas), 1.5 m
M
MgCl
2
,0.4l
M
sense primer
(5¢-TTCGCCACCCGTAAGATC-3¢), 0.4 l
M
antisense
primer (5¢-AAACTCCTCTCCAATGACG-3¢), 0.2 m
M
Fig. 1. Nucleotide sequence of a cDNA clone coding for a zebrafish CRBPII. The complete nucleotide sequence of the EST clone fb69e02.y1 was
determined (GenBank accession number AF363957). The 549 bp sequence contained an open-reading frame of 405 nucleotides coding for a protein
of 135 amino acids. The predicted amino-acid sequence of the zebrafish cDNA clone was most similar to mammalian CRBPIIs (see Fig. 2). The
sequence complementary to the antisense oligonucleotide probe used for in situ hybridization analysis is underlined and in bold font. The position of
the nucleotides corresponding to the 5¢ or complimentary to the 3¢ primers used for PCR amplification of the CRBPII cDNA probe used for
Northern blot and radiation hybrid linkage mapping analysis are boxed and numbered Ô1Õ and Ô2Õ, respectively. The polyadenylation signal
sequence, AATAAA, is italicized and in bold font.

4686 M. C. Cameron et al. (Eur. J. Biochem. 269) Ó FEBS 2002
dNTPs, 1.25 U Taq DNA polymerase and 100 ng of hybrid
cell DNA. Control reactions contained 100 ng of either
zebrafish or mouse parental cell line DNA or a 1 : 10
mixture of zebrafish and mouse parental cell line DNA.
PCR conditions were 94 °C for 4 min followed by 32 cycles
of 94 °Cfor30s,60°Cfor30s,72°C for 30 s and a final
extension at 72 °C for 7 min Products were separated by
agarose gel electrophoresis and the radiation hybrid panel
was scored and then analyzed according to the directions at
(:8000/zfrh/beta.cgi).
PCR primers (5¢-CCAGCACATCCAGCTTC-3¢)and
(5¢-GCCTGTTTGGAGCATTAG-3¢) (see Fig. 1 for pri-
mer location) were used to amplify a 442-bp product from
DNA of clone fb69e02.y1. This product was used as a
hybridization probe for Northern blot analysis [23]. The size
of the hybridizing mRNA was determined by comparing its
electrophoretic mobility with molecular mass markers
(0.24–9.5 kB RNA ladder, Gibco BRL).
In situ hybridization was performed using an antisense
oligonucleotide probe (see Fig. 1) to determine the pattern
of CRBPII expression in adult zebrafish. Based on
BLASTN
searches of GenBank, the in situ hybridization probe did
not exhibit significant sequence similarity to any other
DNA sequence currently available in GenBank. Fourteen
micrometer transverse, sagittal, and coronal sections of
adult zebrafish were hybridized to DNA probes using
previously described methods [24]. Following hybridization
and post-hybridization washes, the sections were exposed

to autoradiographic film. Emulsion autoradiography of the
tissue sections that hybridized to the CRBPII antisense
probe was performed to localize the in situ hybridization
signal to the cellular level [24]. A hybridization probe
corresponding to the sense strand of a portion of a
zebrafish I-FABP mRNA, shown previously not to
hybridize to any transcript in total zebrafish RNA [18],
was used as a negative control for in situ hybridization
studies. A probe complementaty to the coding strand of
the zebrafish I-FABP mRNA [18] was used as a positive
control for in situ hybridization and emulsion autoradio-
graphy. Following emulsion autoradiography, the sections
were stained with cresyl violet and viewed under bright-
field and dark-field illumination [24].
RESULTS AND DISCUSSION
The nucleotide sequence of a zebrafish EST clone reported
in GenBank to have sequence similarity to the 5¢ end of
CRBPII cDNAs from mammals and chicken was deter-
mined (Fig. 1). Sequence from forward and reverse sequen-
cing reactions was aligned and discrepancies were resolved
by examination of the primary sequence data. The complete
sequence (GenBank accession number AF363957), deter-
mined from both strands, differed from that of the partial
sequence of the EST clone reported in GenBank (accession
number AI544932) at several positions. The cDNA
sequence contained an open-reading frame of 405 bp
encoding a protein of 135 amino acids. The percentage
amino-acid sequence similarity between the open-reading
encoded by the cDNA clone and the amino-acid sequences
of intracellular lipid-binding proteins from zebrafish and

other species indicate that the cDNA clone codes for the
zebrafish CRBPII (Fig. 2). The zebrafish CRBPII protein
was one amino acid longer than mammalian CRBPII and
equal in length to chicken CRBPII. The molecular mass of
the CRBPII protein in zebrafish, based on the predicted
amino-acid sequence, is 15.8 kDa. The molecular mass of
this zebrafish CRBPII is comparable to other members of
the intracellular lipid-binding protein family which are all
between 14 and 16 kDa [1,2]. The zebrafish CRBPII amino-
acid sequence is most similar to the chicken CRBPII (76%
identity). The zebrafish CRBPII amino-acid sequence
exhibits 73–75% sequence similarity to mammalian
CRBPIIs and less than 40% amino-acid sequence identity
to other intracellular lipid-binding proteins.
Cheng et al. [25] proposed that Arg106 and Arg126
present in some members of the lipid-binding protein family
correspond to Gln109 and Gln129 in CRBPI and CRBPII.
While all FABPs and CRBPs studied to date have the same
tertiary structure, the amino-acid residues at positions 109
and 129 may determine ligand-binding specificity. Both Gln
residues are found in the zebrafish CRBPII sequence at the
comparable positions within the amino-acid alignment of
other intracellular lipid-binding proteins (Fig. 2). Gln109 is
not strictly conserved in all CRBPs, however, as the chicken
CRBPII and mouse CRBPIII have a histidine residue at this
position.
Phylogenetic analyses of 51 intracellular lipid-binding
proteins, from vertebrates and invertebrates, indicate that at
least 14 gene duplications have occurred during the
evolution of this multigene family [16]. As the amino-acid

sequence of the zebrafish CRBPII reported here is most
similar to CRBPIIs from other species, and not to CRBPI
or other intracellular lipid-binding proteins from zebrafish
(Fig. 2), the duplication of the ancestral CRBP gene that
gave rise to the CRBPI and CRBPII genes most likely
occurred before the divergence of the teleost fishes and
mammals, approximately 400 mya. We have shown that the
CRBPI gene exists in the zebrafish genome (M.K. Sharma
& J.M. Wright, unpublished data). The mammalian CRBPI
and CRBPII genes are linked on chromosome 9 in mouse
and 3 in humans and share sequence similarity including the
conserved Gln residues at positions 109 and 129 [25,26].
Radiation hybrid mapping studies [22] assigned the CRBPII
gene to linkage group 15 (LOD score 19.8) in the zebrafish
genome. (Primary data and the RH vector for linkage
analysis is available upon request to the corresponding
author). The CRBPII gene is flanked by the growth
associated protein 43 (GAP 43) gene on one side and the
chordin (CHRD) gene on the other in both zebrafish and in
human (Table 1). This synteny suggests that a common
linkage group was inherited from the ancestor of fishes and
mammals. In mouse, however, the synteny has not been
maintained as CRBPII is located on chromosome 9, while
GAP 43 and CHRD are located on chromosome 16
(Table 1). This suggests a translocation/rearrangement of
this region of the mouse genome after the divergence of
fishes and mammals. The conservation of amino-acid
sequence among all CRBPIIs and the evidence that
zebrafish and human CRBPII genes are in the same
syntenic group suggest that fish and mammals share a

common ancestral CRBPII gene.
Northern blot-hybridization of the zebrafish CRBPII
cDNA to total RNA extracted from whole adult zebrafish
detected a single mRNA transcript of  720 nucleotides
(Fig. 3A). The difference in size between the mRNA
transcript detected by Northern blot ( 720 nucleotides)
Ó FEBS 2002 Expression of CRBPII in zebrafish (Eur. J. Biochem. 269) 4687
and the size of the cDNA sequence (549 nucleotides) suggests
that the cDNA clone is probably lacking the complete poly A
tail or part of the 5¢ untranslated region, or both.
In situ hybridization analysis of adult zebrafish tissue
sections revealed that the hybridization signal resulting from
the specific annealing of the CRBPII antisense probe was
confined to the intestine and, at relatively lower levels, to the
zebrafish liver (Fig. 3B,C). Hybridization of the CRBPII-
specific probe to the intestine is most evident in the
transverse (Fig. 3B) and coronal (Fig. 3C) sections while
hybridization to the liver is more clearly seen in the coronal
sections (Fig. 3C). The radiolabel associated with the layer
beneath the external skin appears to be non–specific
interaction of the probe with this tissue as it is seen in all
autoradiograms regardless of the hybridization probe
employed, i.e. the CRBPII or I-FABP antisense probes or
the I-FABP negative control sense probe (Figs 3B,C). The
hybridization signal resulting from the specific annealing of
the I-FABP antisense probe was confined to the intestine as
previously reported [18].
As CRBPII and I-FABP mRNA have been colocalized in
the mammalian and chicken proximal portion of the small
intestine [27–29], we examined the distribution of CRBPII

and I-FABP mRNA in adjacent tissue sections of adult
zebrafish. Emulsion autoradiography of tissue sections that
hybridized to the CRBPII and I-FABP antisense and
negative control I-FABP sense probes was performed to
localize the hybridization signal at the cellular level. The
CRBPII mRNA was localized to the enterocytes in the
microvilli of the intestine and to the hepatocytes of the liver
(Fig. 4A,B). The I-FABP mRNA was similarly localized to
the enterocytes of the intestine, but was not detected in the
Fig. 2. Amino-acid sequence alignment of zebrafish CRBPII with other CRBPs, CRABPs, and FABPs. The amino-acid sequences of zebrafish
CRBPII (ZbfshCRBPII; GenBank Accession number AF363957), chicken CRBPII (ChickCRBPII [43]), pig CRBPII (PigCRBPII; P50121),
human CRBPII (HumanCRBPII; AAC50162), rat CRBPII (RatCRBPII; P06768), mouse CRBPIII (MouseCRBPIII; AA466092), rat CRBPI
(RatCRBPI; P02696), rat CRABPII (RatCRABPII; P51673), human CRABPI (HumanCRABPI; P29762), zebrafish brain FABP (ZbfBFABP;
Af237712), and zebrafish intestinal FABP (ZbfIFABP; AF180921) were aligned using
CLUSTALW
[21]. Dashes, indicating addition/deletion dif-
ferences between the zebrafish CRBPII and other amino-acid sequences, were added to maximize alignment. Dots indicate identity between the
amino-acid sequence of zebrafish CRBPII and other CRBPs, CRABPs and FABPs. The percentage amino-acid sequence similarity relative to the
zebrafish CRBPII is indicated at the end of each sequence.
Table 1. Zebrafish-human conserved syntenies.
Zebrafish
a
Human
b
Mouse
b
LG Locus
Accession
Number Gene
Chromosomal

Position Gene
Chromosomal
location
15 gap43 L27645 GAP43 3q13.1-q13.2 Gap43 16 29.5 cM
15 rbp2 (CRBPII) AF363957 RBP2 3q23 Rbp2 9 57.0 cM
15 chd AF034606 CHRD 3q27 Chrd 16 14.0 cM
a
Wood et al. [45];
b
LocusLink ( NCBI.
4688 M. C. Cameron et al. (Eur. J. Biochem. 269) Ó FEBS 2002
liver (Fig. 4A,B) [18]. In the transverse sections, the
positional difference of CRBPII mRNA demarcates the
anterior and posterior parts of the intestine (Fig. 4A). In
adjacent sections, the distribution of I-FABP-specific
hybridization signal was the same as that observed for
CRBPII (Fig. 4A). Due to the entwined positioning of the
intestine within the zebrafish, the coronal sections display
three cross-sections corresponding to different regions of the
intestine (Fig. 4B). There was hybridization to only two of
the three cross sections of the intestine for the CRBPII
antisense and I-FABP antisense probes. The radiolabel
associated with the centre of the intestine is present in all
sections labeled with antisense and sense oligonucleotides
indicating nonspecific binding of the probes to the contents
of the gut. Evidence of I-FABP mRNA restricted to just the
anterior of the zebrafish intestine was not previously
observed by us possibly owing to the limited sections
assayed [18]. It is possible that subtle differences in the
amount of CRBPII and I-FABP mRNA exist along the

anterior-posterior of the intestine that are not detectable
using in situ hybridization and emulsion autoradiography.
Moreover, using whole-mount in situ hybridization to
zebrafish embryos, it was previously determined that I-
FABP mRNA is first expressed in the intestinal tube 3 days
postfertilization and, by 5 days postfertilization and on-
ward, I-FABP mRNA is abundant in the anterior intestine
but is not detectable in the posterior intestine [27].
In adult mammals, CRBPII mRNA levels gradually
decrease along the anterior-posterior axis of the intestine
[28,29]. The proximal intestine of mammals has a higher
capacity to absorb retinol than does the distal portion
[30,31]. Levin et al. [15] demonstrated that CRBPs are
directly involved in retinol absorption in a human intestinal
cell line and that the amount of CRBPI and CRBPII
mRNA and protein is directly related to the rate of retinol
absorption. It is believed that CRBPII directs retinal and
retinol to the enzymes, retinal dehydrogenase and LRAT,
respectively, in the intestine. CRBPII, microsomal retinal
reductase and LRAT are colocalized in the mammalian
intestine [32–34]. The CRBPII gradient in the intestine
parallels the change in enzyme activity of LRAT and retinal
reductase which is greater in the anterior than in the
posterior of the intestine [34]. Thus, these findings are
consistent with the proposed role of CRBPII in retinol
metabolism. Similarly, previous studies in mammals have
shown that there is a gradient of I-FABP expression along
the horizontal axis as concentrations of I-FABP mRNA
and proteins gradually decrease from high levels in the
jejunum to negligible levels in the colon [35–38]. CRBPII

and I-FABP expression is similar along the anterior-
posterior axis in the intestine of mammals. Therefore, a
corresponding trend for CRBPII and I-FABP expression in
zebrafish is consistent with their expression in mammals.
The abrupt termination of CRBPII expression along the
anterior-posterior axis of the zebrafish intestine, however,
contrasts with the gradual decrease in CRBPII and I-FABP
expression pattern in mammals.
In the 5¢-control regions of the mammalian CRBPII [39]
and I-FABP [35] genes, a closely related cis-element that
consists of nearly perfect tandem repeats, termed retinoid x
response element (RXRE) [40] has been found. It is
conceivable therefore that these two genes may both be
regulated by the action of retinoid x receptor (RXR)
binding to RXRE [41]. The similar distribution of CRBPII
and I-FABP mRNA in the zebrafish intestine reported here
may reflect the co-ordinate regulation of these genes by
common intestinal transcriptional factors in zebrafish.
In addition to being abundant in intestine, CRBPII is
found in neonatal liver hepatocytes of the rat and chick
[10,42]. In rat, however, the levels of CRBPII mRNA in the
Fig. 3. CRBPII mRNA expression in the adult zebrafish. The complete
coding sequence of the zebrafish CRBPII cDNA clone was amplified
by PCR and used as a hybridization probe in Northern blot analysis of
total cellular RNA isolated from adult zebrafish (A). The zebrafish
CRBPII-specific probe hybridized to a transcript of approximately 720
nucleotides. The1.35 kb (upper line) and 0.24 kb (lower line) RNA
molecular mass markers are shown on the left of the panel. In situ
hybridization analysis was performed using a 3¢ end-labelled oligo-
nucleotide complementary to an internal portion of the zebrafish

CRBPII coding region (see Fig. 1). In situ hybridization of adjacent
transverse (B) and coronal (C) sections of adult zebrafish to the the
zebrafish CRBPII-specific and I-FABP-specific antisense probes are
shown. An oligonucleotide corresponding to the sense strand of the
I-FABP coding region was included as a negative control. This probe
was previously shown to bind nonspecifically to adult zebrafish sec-
tions in in situ hybridization analysis [18]. Labelled arrows point to the
intestine (I) and liver (L) of the zebrafish. The hybridization signal
resulting from the annealing of the CRBPII- and I-FABP-specific
probes are intense in the intestine. A lower intensity of hybridization of
the CRBPII-, but not the I-FABP-specific probes, was seen in the
zebrafish liver.
Ó FEBS 2002 Expression of CRBPII in zebrafish (Eur. J. Biochem. 269) 4689
liver decrease after birth and are undetectable in the adult
liver [10]. In adult chicken, CRBPII mRNA is abundant in
the intestine but is not detected in liver [43]. As stated above,
it is thought that CRBPII directs retinal to the enzyme
retinal dehydrogenase in the mammalian perinatal liver [42].
Takase et al. [42] have suggested that following birth, the
absorptive cells of the intestine may be functionally
immature and unable to convert b-carotene to retinal.
b-carotene is transferred to the liver in neonatal mammals
and converted to retinal by the enzyme b-carotene dioxyg-
enase in hepatocytes. Later, the intestine matures and can
convert b-carotene to retinal. High levels of hepatic CRBPII
are then no longer necessary for the production of retinol
and CRBPII levels decrease in adult liver. Chicken and rat
are known to convert most ingested b-carotene to retinal
and then to retinol in enterocytes such that lower levels of
b-carotene and retinal are found in their circulation

compared to the levels observed in humans [44]. These
findings suggest that hepatic CRBPII may play a role in
metabolizing hepatic b-carotene to retinal and the subse-
quent esterification of the converted retinol only during the
perinatal period in mammals [42]. The in situ hybridization
and autoradiographic emulsion studies show that CRBPII
mRNA is abundant in the liver of adult zebrafish (Fig. 4).
This pattern of CRBPII expression therefore differs mark-
edly from that observed in rat and chicken [10,42]. Retinol
metabolism of fishes may differ from that of mammals and
chicken in that large amounts of b-carotene continue to be
transported to the adult liver of teleost fishes resulting in the
need for high levels of CRBPII mRNA observed in the liver
of adult zebrafish.
CRBPII and I-FABP mRNA are colocalized in the fish
and mammalian intestine and may be co-ordinately regu-
lated by RXR acting at RXRE within the control regions of
these genes. The differential expression of CRBPII and
I-FABP in the adult zebrafish liver, however, suggests that
other transcription factors may regulate CRBPII gene
expression in the livers of adult zebrafish.
In summary, the zebrafish CRBPII cDNA reported here
has sequence similarity to CRBPs isolated from mammals.
The patterns of gene expression for CRBPII and I-FABP in
fishes and mammals suggest that there is co-ordinate
regulation of these genes in the intestine, but not in the
liver. This may reflect differences in retinol metabolism
between adult teleost fishes and mammals.
ACKNOWLEDGEMENTS
We wish to thank Dr Marc Ekker for providing DNA samples from the

LN54 collection of radiation hybrids. This work was supported by a
grant from the Natural Sciences and Engineering Research Council of
Canada to J. M. W. and a grant from the Canadian Institutes of
Health Research to E. M. D-W. M. C. C. was the recipient of a
Natural Sciences and Engineering Research Council Undergraduate
Student Research Award and M. K. S. was the recipient of an Izaak
Walton Killiam Memorial scholarship.
REFERENCES
1. Banaszak, L., Winter, N., Xu, Z., Bernlohr, D.A., Cowan, S. &
Jones, T.A. (1994) Lipid-binding proteins: a family of fatty acid
and retinoid transport proteins. In Advances in Protein Chemistry
Fig. 4. Autoradiograhpic emulsion of zebrafish
sections hybridized to the CRBPII- and
I-FABP-specific antisense and negative control
I-FABP sense probes. Transverse (A) and
coronal (B) zebrafish sections were hybridized
to the I-FABP, CRBPII and negative control
probes and directly exposed to autoradio-
graphic emulsion. Following development of
the emulsion, the sections were counterstained
with cresyl violet and viewed under brightfield
(BF) and darkfield (DF) illumination at 10·
magnification. The intestine (I), liver (L) and
pancreas (P) are indicated. The white granules
viewed under darkfield illumination corres-
ponded to the location of either the CRBPII
or I-FABP mRNA as indicated.
4690 M. C. Cameron et al. (Eur. J. Biochem. 269) Ó FEBS 2002
(Anfinsen, C.B., Edsall, J.T., Richards, F.M. & Eisenberg, D.S.,
eds), Vol. 45, pp. 89–151. Academic Press, San Diego.

2. Veerkamp, J.H., van Kuppevelt, T.H.M.S.M., Maatman,
R.G.H.J. & Prinsen, C.F.M. (1993) Structural and functional
aspects of cytoplasmic fatty-acid binding proteins. Prostaglandins
Leukot. Essent. Fatty Acids 49, 887–906.
3. Sacchettini, J.C., Gordon, J.I. & Banaszak, L.J. (1989) Crystal
structure of rat intestinal fatty acid-binding protein. J. Mol. Biol.
208, 327–339.
4. Sporn, M.B., Roberts, A.B. & Goodman, D.S. (1994) The
Retinoids: Biology, Chemistry and Medicine, 2nd edn. Raven Press,
New York, New York.
5. Li, E., Demmer, L.A., Sweetser, D.A., Ong, D.E. & Gordon, J.I.
(1986) Rat cellular retinol-binding protein type II: Use of a cloned
cDNA to define its primary structure, tissue-specific expression,
and developmental regulation. Proc. Natl Acad. Sci. USA 83,
5779–5783.
6. MacDonald, P.N. & Ong, D.E. (1987) Binding specificites of
cellular retinal-binding protein and cellular retinol-binding pro-
tein, type II. J. Biol. Chem. 262, 10550–10556.
7. Vogel, S., Mendelsohn, C.L., Mertz, J.R., Piantedosi, R., Wald-
burger, C., Gottesman, M.E. & Blaner, W.S. (2001) Character-
ization of a new member of the fatty acid-binding protein family
that binds all-trans retinol. J. Biol. Chem. 276, 1353–1360.
8. Troen,G.,Eskild,W.,Fromm,S.H.,DeLuca,L.M.,Ong,D.E.,
Wardlaw, S.A., Reppe, S. & Blomhoff, R. (1999) Vitamin
A-sensitive tissues in transgenic mice expressing high levels of
human cellular retionol-binding protein type I are not altered
phenotypically. J. Nutr. 129, 1621–1627.
9. Ghyselinck, N.B., Bavik, C., Sapin, V., Mark, M., Bonnier, D.,
Hindelang,C.,Dierich,A.,Nilsson,C.B.,Hakansson,H.,Sau-
vant, P., Azais-Braesco, V., Frasson, M., Picaud, S. & Chambon,

P. (1999) Cellular retinol-binding protein I is essential for vitamin
Ahomeostasis.EMBO J. 18, 4903–4914.
10. Ong, D.E. (1984) A novel retinol-binding protein from rat. J. Biol.
Chem. 259, 1476–1482.
11. Ong, D.E. & Page, D.L. (1987) Cellular retinol-binding protein
(type two) is abundant in human small intestine. J. Lipid Res. 28,
739–745.
12. Finlay, J.A. & DeLuca, H.F. (1988) Purification and properties of
an 18-kilodalton, 1,25-dihydroxyvitamin D
3
-modulated protein
from embryonic chick intestine. Biochemistry 27, 3381–3387.
13. Goda, T. & Takase, S. (1989) Purification, properties, and devel-
opmental changes of cellular retinol-binding protein, type II, in
chicken intestine. J. Nutr. Sci. Vitaminol. 35, 545–557.
14. Perozzi, G., Barlia, D., Murgia, C., Kelly, D., Begbie, R. & King,
T.P. (1993) Expression of differentiated functions in the develop-
ing porcine small intestine. J. Nutr. Biochem. 4, 669–705.
15. Leven, M.S. (1993) Cellular retinol-binding proteins are determi-
nants of retinol uptake and metabolism in stably transfected caco-
2cells.J. Biol. Chem. 268, 8267–8276.
16. Schleicher, C.H., Cordoba, O.L., Santome, J.A. & Dell’Angelica,
E.C. (1995) Molecular evolution of the multigene family of
intracellular lipid-binding proteins. Biochem. Mol. Biol. Int. 36,
1117–1125.
17. Matarese, V., Stone, R.L., Waggoner, D.W. & Bernlohr, D.A.
(1989) Intracellular fatty acid trafficking and the role of cytosolic
lipid binding proteins. Prog. Lipid Res. 28, 245–272.
18. Pierce, M., Wang, Y., Denovan-Wright, E.M. & Wright, J.M.
(2000) Nucleotide sequence of a cDNA clone coding for an

intestinal-type fatty acid binding protein and its tissue-specific
expression in zebrafish (Danio rerio). Biochem. Biophys. Acta.
1490, 175–183.
19. Denovan-Wright, E.M., Pierce, M., Sharma, M.K. & Wright,
J.M. (2000) cDNA sequence and tissue-specific expression of a
basic liver-type fatty acid binding protein in adult zebrafish (Danio
rerio). Biochim. Biophys. Acta. 1492, 227–232.
20. Denovan-Wright, E.M., Pierce, M. & Wright, J.M. (2000)
Nucleotide sequence of cDNA clones coding for a brain-type fatty
acid binding protein and its tissue-specific expression in adult
zebrafish (Danio rerio). Biochim. Biophys. Acta. 1492, 221–226.
21. Thompson, J.D., Higgins, D.G. & Gibson, T.J. (1994) ClustalW:
improving the sensitivity of progressive multiple sequence align-
ment through sequence weighting, position-specific gap penalties
and weight matrix choice. Nucleic Acids Res. 22, 4673–4680.
22. Hukriede, N.A., Joly, L., Tsang, M., Miles, J., Tellis, P., Epstein,
J.A., Barbazuk, W.B., Li, F.N., Paw, B., Postlethwait, J.H.,
Hudson,T.J.,Zon,L.I.,McPherson,J.D.,Chevrette,M.,
Dawid, I.B., Johnson, S.L. & Ekker, M. (1999) Radiation hybrid
mapping of the zebrafish genome. Proc. Natl Acad. Sci. USA 96,
9745–9750.
23. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular
Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor
Laboratory Press, Cold Spring Harbor, NY.
24. Denovan-Wright, E.M., Newton, R.A., Armstrong, J.N., Babity,
J. & Robertson, H.A. (1998) Acute administration of cocaine, but
not amphetamine, increase the level of synaptotagmin IV mRNA
in the dorsal striatum of rat. Mol. Brain Res. 55, 350–354.
25. Cheng, L., Qian, S., Rothschield, C., Avignon, A., Lefkowith, J.,
Gordon, J. & Li, E. (1991) Alteration of the binding specificity of

cellular retinol-binding protein II by site-directed mutagenesis.
J. Biol. Chem. 266, 24404–24412.
26. Nilsson, M.H., Spurr, M.K., Lundvall, J., Rask, L. & Peterson,
P.A. (1988) Human cellular retinol-binding protein gene organi-
zation and chromosomal location. Eur. J. Biochem. 173, 35–44.
27. Andre, M., Ando, S., Ballagny, C., Durliat, M., Poupard, G.,
Briancon, C. & Babon, P.J. (2000) Intestinal fatty acid binding
protein gene expression reveals the cephalocaudal patterning
during zebrafish gut morphogenesis. Int. J. Dev. Biol. 44, 249–252.
28. Ong, D.E. & Page, D.L. (1987) Cellular retionol-binding protein
(type two) is abundant in human small intestine. J. Lipid Res.
28, 739–745.
29. Inagami, S. & Ong, D.E. (1992) Purification and partial char-
acterization of cellular retinol-binding protein, type two, from
human small intestine. J. Nutr. 122, 450–456.
30. Hollander, D. & Muralidhara, K.S. (1977) Vitamin A intestinal
absorption in vivo: influence of luminal factors on transport.
Am.J.Phsiol.232, E471–E477.
31. Said, H.M., Ong, D.E. & Redha, R. (1988) Intestinal uptake of
retinol in suckling rats: characteristics and ontogeny. Pediatr. Res.
24, 481–485.
32. Kaddad, B.P. & Ong, D.E. (1988) Reduction of retinaldehyde
bound to cellular retinol-binding protein (type II) by microsomes
from rat small intestine. J. Biol. Chem. 263, 12916–12919.
33. Ong, D.E. (1993) Retinoid metabolism during intestinal absorp-
tion. J. Nutr. 123, 351–355.
34. Ong, D.E., Kakkad, B. & MacDonald, P.N. (1987) Acyl-CoA-
independent esterification of retinol bound to cellular retinol-
binding protein (type II) by microsome from rat intestine. J. Biol.
Chem. 262, 2729–2736.

35. Sweetser, D.A., Birkenmeier, E.H., Klisak, I.J., Zollman, S.,
Sparkes, R.S., Mohandas, T., Lusis, A.J. & Gordon, J.I. (1987)
The human and rodent intestinal fatty acid-binding protein genes:
a comparative analysis of their structure, expression, and linkage
relationships. J. Biol. Chem. 262, 16060–16071.
36. Green, R.P., Cohn, S.M., Sacchettini, J.C., Jackson, K.E. &
Gordon, J.I. (1992) The mouse intestinal fatty acid binding protein
gene: nucleotide sequence, pattern of developmental and regional
expression, and proposed structure of its protein product. DNA
Cell Biol. 11, 31–41.
37. Sheilds, H.M., Bates, M.L., Bass, N.M., Best, C.J., Alpers, D.H. &
Ockner, R.K. (1986) Light microscope immunocytochemical
localization of hepatic and intestinal types of fatty acid binding
proteins in rat small intestine. J. Lipid Res. 27, 549–557.
Ó FEBS 2002 Expression of CRBPII in zebrafish (Eur. J. Biochem. 269) 4691
38. Iseki, S., Amano, O., Kanda, T., Fujii, H. & Ono, T. (1993)
Expression and localization of intestinal 15 kDa protein in the rat.
Mol. Cell. Biochem. 123, 113–120.
39. Demmer, L.A., Birkenmeier, E.H., Sweetser, D.A., Levin, M.S.,
Zollman, S., Sparkes, R.S., Mohandas, T., Lusis, A.J. & Gordon,
J.I. (1987) The cellular retinol-binding protein II gene. J. Biol.
Chem. 262, 2458–2467.
40. Mangelsdorf, D.J., Umesono, K., Kliewer, S.A., Borgmeyer, U.,
Ong, E.S. & Evans, R.M. (1991) A direct repeat in the cellular
retinol-binding protein type II gene confers differential regulation
by RXR and RAR. Cell 66, 555–561.
41. Goda, T., Yasutake, H. & Takase, S. (1994) Dietary fat regulates
cellular retinol-binding protein II gene expression in rat jejunum.
Biochim. Biophys. Acta. 1200, 34–40.
42. Takase, S., Suruga, K., Suzuki, R. & Goda, T. (1996) Relationship

between perinatal appearance of cellular retinol-binding protein,
type II and retinal reductase activity in chick liver. Life Sci. 58,
135–144.
43. Suruga, K., Goda, T., Igarashi, M., Kato, S., Masushige, S. &
Takase, S. (1997) Cloning of chick cellular retinol-binding pro-
tein, type II and comparison to that of some mammals: expression
of the gene at different developmental stages, and possible
involvement of RXRs and PPAR. Comp. Biochem. Physiol. 118,
859–869.
44. Nierenberg, D.W. & Stukel, T. (1987) Diurnal variation in plasma
levels of retinol, tocopherol, and beta-carotene. Am.J.Med.Sci.
294, 187–190.
45. 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.
4692 M. C. Cameron et al. (Eur. J. Biochem. 269) Ó FEBS 2002