Download Widespread expression of the bovine Agouti gene results from at

Copyright ª Blackwell Munksgaard 2005
doi: 10.1111/j.1600-0749.2004.00195.x
Widespread expression of the bovine Agouti gene
results from at least three alternative promoters
Michael Girardot, Juliette Martin, Sylvain
Guibert, Hubert Leveziel, Raymond Julien and
Ahmad Oulmouden*
Unite de Génétique Moléculaire Animale, UMR 1061-INRA/
Université de Limoges, Limoges, France
*Address correspondence to Ahmad Oulmouden, Faculté des
Sciences et Techniques UMR 1061, INRA/Université de Limoges,
123 avenue Albert Thomas, 87060 Limoges Cedex, France.
E-mail: [email protected]
Summary
In wild-type mice, it is well known that Agouti is
only expressed in skin where it controls the banded-hair phenotype. As a first step to investigate
the physiological role of Agouti in cattle, we isolated the corresponding gene and studied its
expression pattern. We found no evidence of coding-region sequence variation within and between
eight breeds representing a large panel of coat
colour phenotypes. We detected by northern
hybridization two Agouti mRNA isoforms in brain,
heart, lung, liver, kidney, spleen and a third in
skin. We characterized the full-length Agouti transcript in skin and isolated the 5¢UTR of two
mRNAs expressed in the other tissues. The three
mRNAs have the same coding region but differ by
their 5¢ untranslated regions. Upstream regulatory
sequences display two alternative promoters
involved with the broad expression in tissues
other than skin. Interestingly, these sequences are
highly homologous to upstream sequences of
the orthologous human (76–85% identity) and pig
(82–86% identity) ASIP genes. In addition to its
potential role in pigmentation (as seen in mice),
we suggest that bovine Agouti could be involved
in various physiological functions. Furthermore,
the significant homology between cattle, pig and
human regulatory sequences indicate that these
orthologous genes are regulated alike. Lastly,
since the 5¢UTR of many eukaryotic mRNAs are
physiologically relevant, their impact on bovine
Agouti mRNA performance is discussed.
Key words: Agouti/pigmentation/regulation/mRNA
Introduction
In mice, genetic analysis of several natural dominant
and recessive mutations of the Agouti gene combined
with biochemical and cellular biology studies of the
Agouti protein, have demonstrated that the normal
function of the Agouti gene is to regulate hair-pigment
production by melanocytes. Molecular genetics and
pharmacological studies have shown that mutually
exclusive binding (Ollmann et al., 1998) of the melanocortin 1 receptor (Mc1r) by the Agouti protein or by
a-melanocyte stimulating hormone (a-MSH) signals
hair-bulb melanocytes to synthesize either pheomelanin (yellow–red pigment) or eumelanin (black-brown
pigment), respectively. As expected for a ligand-receptor relationship, Mc1r alleles are epistatic to Agouti
alleles (Silvers, 1979). In this regard, pigmentation is
switched from eumelanogenesis to pheomelanogenesis either in recessive yellow (e/e) mice, which are
homozygous for a loss-of function mutation at the
extension (E) locus, or in lethal yellow mice, which
contain a gain-of-function mutation at the Agouti (A)
locus.
Recessive mutations at the Agouti locus, which
impair either Agouti protein activity or reduce the level
of Agouti mRNA synthesis, result in a darker coat
colour. Conversely, certain Agouti alleles such as Ay
(lethal yellow), which displays constitutive and ubiquitous expression of Agouti, increase yellow pigmentation
of the fur (Miller et al., 1993). In addition to a yellow
coat, mice that carry Ay, display obesity, insulin resistance, premature infertility, increased body length, and
tumour susceptibility (Duhl et al., 1994). Obesity has
been associated with binding of Agouti to Mc4r, inactivating the latter in hypothalamus (Voisey & van Daal,
2002). Although we have some knowledge of the role
that Agouti plays in the mouse, its role in many mammals, including human, is not yet clarified.
In cattle, a wide variety of colours exists. Interestingly, several standardized breeds are homogenous with
respect to coat colour (Guibert et al., 2004) suggesting
that their phenotypes are associated with a specific
locus. On the other hand, many breeds with no deliberate selection for or against any particular colour type
also exist (Adalsteinsson et al., 1995) suggesting that
Received 29 October 2003; in final form 3 August 2004
34
Pigment Cell Res. 18; 34–41
Widespread expression of bovine Agouti gene
their phenotypes result from double, triple or more coat
colour gene mutations. Both types of cattle breeds are
relevant in the field as a model to study the regulation
of melanogenesis and to evaluate interacting functions
of coat colour gene products as described for genetic
strains of laboratory animals.
The impacts of Extension and Agouti on cattle coat
colour have been discussed by several authors (Olson,
1999). We and others (Klungland et al., 1995; Rouzaud
et al., 2000) have identified four Extension alleles including those with loss-of-function (Mc1re) and gain-offunction (Mc1rED) which lead to pheomelanic and
eumelanic coat colour, respectively. Three of them were
suggested by genetics analysis (Wright, 1917). Many
genetics studies have also suggested both loss-of-function and gain-of-function alleles of the Agouti locus in
cattle (Olson, 1982; Lauvergne, 1966; Searle, 1968;
Adalsteinsson et al., 1995). However, the absence of
banded hairs, at least in French cattle breeds homozygous (E/E) at Mc1r, suggests a different expression pattern between bovine and mouse Agouti.
Here, we report that regardless the investigated French
cattle breeds coat colour, the bovine Agouti coding region
is identical. In addition, we show that the Agouti gene is
expressed in all examined bovine tissues. This widespread expression results from at least three promoters.
Results
Gene structure of Agouti
Bovine genomic DNA amplification using Ag1/Ag2 primers displayed a 5139 bp fragment length which contains, in addition to intronic sequences, three exons that
compose the entire coding region of the Agouti gene
(Fig. 1). The exon-intron organization (GenBank accession number: X99691) of the bovine Agouti is similar to
that reported for human and mouse counterparts and
the positions of introns in the three genes are identical.
Coding exons 2, 3 and 4 are separated by 1315 and
3422 bp intronic sequence, respectively. Each exon is
flanked by a consensus splice donor and acceptor sites,
except for the last exon, which, as expected, has only a
splice acceptor site (Fig. 1).
The combined sequence obtained from bovine skin by
5¢ and 3¢ amplification of cDNA ends (RACE) experiments is 732 bp long and is composed of a 402 bp
open reading frame (ORF), a 153 bp 5¢UTR (1C 5¢UTR)
and a 177 bp 3¢UTR regions (Fig. 1). The ORF (GenBank
accession number: X99692) is 84 and 85% identical at
the nucleotide level to mouse and human sequences,
respectively. It encodes a putative 133 amino acid protein, which is 78 and 75% identical to the mouse (131
amino acids) and human (132 amino acids) proteins,
respectively. The amino terminus of the putative bovine
protein contains all the features of a signal peptide (Kyte
& Doolittle, 1982). The highly basic domain and the
cysteine-rich domain near the carboxyl terminus (BultPigment Cell Res. 18; 34–41
man et al., 1992) are also conserved between the predicted bovine, mouse and human proteins.
Polymorphisms screening
Coding exons were screened for polymorphisms to identify potential coding sequence variation which could
explain phenotypic differences among cattle breeds.
Each exon was amplified by PCR with gene-specific
primers located in intronic sequences (GenBank X99691)
and sequenced (data not shown). Ten animals belonging
to 8 breeds (Charolais, Limousine, Salers, Blonde
d’Aquitaine, Monbéliarde, Gasconne, Normande and
Prim’ Holstein) were examined. No polymorphism was
identified in any studied bovine breed individuals suggesting that the Agouti coding region is highly conserved
among cattle breeds.
Agouti is expressed in every studied bovine tissue
Tissue distribution of bovine Agouti mRNA was first
examined by RT-PCR using gene-specific primers Ag1
and Ag2 (Fig. 1) to amplify the entire coding region
(exons 2, 3 and 4). A single 402 bp fragment was amplified from skin samples of each breed and different tissues (brain, heart, kidney, spleen, lung and liver). PCR
fragments were purified and subjected to nucleotide
sequence analysis to verify that they contain bovine
Agouti coding sequences. These results show that the
Agouti gene is expressed in all examined bovine tissues
including skin regardless of their coat colour.
Northern hybridizations on bovine tissues were conducted to detect more accurately Agouti transcripts. A
probe corresponding to the entire coding region of the
cDNA revealed three transcript isoforms. An mRNA of
approximately 800 nucleotides, corresponding to the
full-length cDNA (732 bp) obtained by the RACE-PCR
experiment, was detected in skin. Brain, spleen, lung,
liver and kidney mainly express a 2 kb mRNA isoform. A
1.5 kb mRNA isoform is predominant in heart whereas
the 2 kb transcript seems to be a minor component.
Although lower, the 1.5 kb isoform is also detectable in
spleen and lung (Fig. 2).
Characterization of the 5¢ UTR and promoter
sequences
Since the Agouti transcript isoforms displayed the same
coding sequence and only one Agouti gene was detected by southern hybridization (data not shown), we
hypothesized that these transcripts have different
untranslated regions. 5¢ RACE experiment using lung
mRNA was performed to get more insight regarding
5¢UTRs of the 2 and 1.5 kb transcripts. In this way, nested PCR amplification and sequencing allowed us to
identify a 414 bp (A 5¢UTR) and 714 bp (B 5¢UTR) nucleotides length 5¢UTRs (Fig. 1). Unfortunately, we were
unable to isolate the 3¢UTR sequences from these transcripts probably because of known limitation of the
SMART-RACE protocol because of RNA secondary
35
Girardot et al.
Figure 1. The bovine Agouti gene. Numbers under the gene structure are length of exons, and numbers above are lengths of introns.
Consensus acceptors and donors splice sites are indicated. Boxes show coding exons (2, 3, and 4) and 5¢UTR exons (1A, 2A, 3A, 1B and 1C).
Transcripts A, B and C are represented under the genetic map. Small arrows represent oligonucleotides used and described in Methods. The
two boxes above are promoter sequences A and B with their identified transcription factors underlined. INR, initiator region; DPE,
downstream promoter element.
structures and/or RNA length (Matz et al., 1999). The A
and B 5¢UTRs, corresponding to A and B transcripts,
shared two common sequences (143 and 83 bp) suggesting that they belong to the same genomic
sequence. Subsequently, A transcript upstream region
was obtained using Genome Walker and B transcript
upstream region was obtained using a direct PCR-based
assay with genomic DNA and Ag9 and Ag10 primers
(Fig. 1). Comparison of transcript 5¢ untranslated regions
and genomic sequences revealed that B 5¢UTR corresponding sequences are intron-less whereas the A 5¢UTR
36
results from three exons (1A, 2A and 3A) splicing. They
are separated by 1703 and 489 bp intronic sequences,
respectively. The limits of both introns conformed to
the GT-AG rule (Fig. 1).
The B transcript 5¢end (transcription start site), distinct
from A transcript start site, was experimentally located
20 bp downstream the exon 2A 5¢end using the SMARTRACE protocol which selectively amplifies full length
G-capped mRNA (Matz et al., 1999). Thus, the two transcripts arise from two alternative promoters. The analysis
of the sequences upstream of the transcription start sites
Pigment Cell Res. 18; 34–41
Widespread expression of bovine Agouti gene
Figure 2. Agouti transcripts in bovine tissues. (A) 3 lg of poly(A)+
RNA from brain, spleen, liver, lung, kidney, heart, were used in
Northern hybridization. (B) 1 lg poly(A)+ RNA from bovine skin was
used as described in Materials and methods. The blots were
hybridized with a cDNA probe containing the entire coding
sequence of Agouti (402 bp). Alignment of the autoradiogram with
size markers present on the blot indicated that the bands present
in brain, spleen, lung and kidney were approximately 2 kb, and the
band in heart was approximately 1.5 kb (A). The band present in
skin tissue (B) was approximately 800 nucleotides.
of both transcripts do not reveal consensus TATA or
CAAT boxes but display putative binding sites for transcription such as Sp1 sites known to initiate transcription
in TATA-less promoters (Fig. 1). Indeed, consensus INR
[initiator region: Py ) Py ) A + 1 ) N ) (T/A) ) Py ) Py]
sites are located exactly at the same position in the
experimentally 5¢ end of the A and B transcripts. The
downstream promoter element (DPE) consensus
sequence [DPE: (A/G) ) G ) (A/T) ) (C/T) ) (G/A/C)] is
located 42 nucleotides and 30 nucleotides downstream
the INR for the A and B transcripts, respectively.
Computer analysis of Agouti-regulatory sequences
with comparative genomics
BLASTn software was used to identify potential bovine
Agouti promoters and 5¢UTRs orthologous sequences in
others species. Interestingly, two sequences matched
the submitted upstream non-coding regions of the
bovine Agouti gene with an overall per cent identity of
81 and 72%. The first is the pig genomic clone containing the ASIP gene (gi: 46240693), and the second the
human genomic sequence on chromosome 20q11.1–
11.23 containing the ASIP gene (gi: 6624641). To analyse more accurately these sequences, we used the
LAGAN alignment software and m-Vista for the visualization. We performed a multi-pairwise alignment of the
5¢ non-coding sequences between the bovine, human
and pig genes using the LAGAN alignment software
with a range that allowed us to identify regions which
Pigment Cell Res. 18; 34–41
have at least 75% identity and are 100 bp lengths. Four
highly conserved regions were identified between the
bovine and pig gene located 55 kb upstream the coding
region of the pig ASIP gene, and five between the
bovine and human gene located 40 kb upstream the
coding region of the human ASIP gene (Fig. 3). These
conserved sequences correspond to three main bovine
regions: region I located between )779 and )305 relative to the first transcription start site in bovine 1A exon;
region II located between )52 and +417; and region III
located between +1907 and +2605. Region I corresponding to the A promoter with at least 80% of identity
suggests that this regulatory sequence is also functional
in the two other species. The second region corresponds to the exon 1A and to the beginning of the subsequent intron, which could be a part of the exon in the
other species or an important regulatory sequence. The
third region matches 2A-1B-3A exons.
Discussion
In this paper, we report for the first time the structure
and widespread expression of the bovine Agouti gene.
Its pattern of expression results from at least three promoters. We also report potential regulatory sequences
for pig and human ASIP gene as revealed by a comparative genomic analysis.
Although in mice genotype–phenotype correlations of
several Agouti alleles have been established, the polymorphism study of the bovine Agouti coding region has
been disappointing in this regard. We did not find any
polymorphism in this coding region. Thus, alleles suggested by genetics analysis (Olson, 1999) leave open
the possibility that polymorphisms associated with cattle coat colour exist in non-coding regions of Agouti. It
should be noted that no ASIP polymorphisms have been
identified in humans although a putative regulatory
mutation in the 3¢UTR of its transcript was reported
(Voisey et al., 2001; Kanetsky et al., 2002).
To understand the expression pattern and to look for
possible regulatory mutations of the bovine Agouti
gene, we performed RT-PCR and northern hybridization.
Whereas the wild-type mouse Agouti is expressed only
in skin, we found that its bovine counterpart is
expressed in brain, heart, lung, liver, spleen, kidney and
skin. In addition, while the drafting of this paper was in
progress, Sumida et al. reported the expression of
bovine Agouti in adipocytes (Sumida et al., 2004). The
human ASIP gene is also expressed in several tissues
including heart, liver, kidney, foreskin, ovary (Wilson
et al., 1995), testis, adipocytes (Kwon et al., 1994), leucocytes and placenta (data not shown). Thus, we conclude that the bovine gene expression pattern is similar
to the human gene pattern but different of the mouse
one.
We have identified three mRNA isoforms differing at
least in their 5¢ untranslated regions and exhibiting the
37
Girardot et al.
Figure 3. Multiple alignment between bovine (centre), pig (left) and human (right) genomic sequences. LAGAN results visualized with the
m-Vista program are on the left for bovine vs. pig comparison and on the right for bovine vs. human comparison (graphical representations).
Identity percentages, aligned sequences lengths and genomic coordinates are indicated on columns between genomic sequence maps. Upper
caps roman numbers on the bovine sequence map correspond to the regions I, II and III. Accession numbers for the pig genomic sequence is
AJ427478 and AL035458 for human. The pig and human aligned sequences are drawn to the bovine sequence scale. The pig ASIP gene is on
chromosome 17q21 (Kim et al., 2000), the human ASIP gene on chromosome 20q11.2 (Kwon et al., 1994), and the bovine Agouti gene was
localized on chromosome 13 (Schlapfer et al., 2002).
same coding region. Skin tissue exhibits an mRNA isoform (not shown) with a specific 5¢UTR similar to a
ruminant repetitive DNA sequence described in Bos
Indicus (Kemp & Teale, 1994). The genome walker strategy was unsuccessful to determine the skin transcript
promoter sequences because of the repetitive content
of its 5¢UTR. However, this C promoter sequence that
remains to be isolated must be localized between exon
3A and 2 or upstream A promoter. The two 5¢UTR
sequences corresponding to the 2 and 1.5 kb lung
mRNA transcripts are reported here. The A and B fulllength transcripts cannot be determined yet. The
SMART-RACE protocol known limitation concerning
RNA secondary structures and/or RNA length (Matz
et al., 1999) does not allow us to isolate A and B transcripts 3¢UTRs. It is unlikely that A and B transcripts
use the 3¢UTR isolated from skin because the predicted
mRNA would be different than the average length seen
on Northern blot. This means that these transcripts have
38
different 3¢UTRs, which probably arise from different
splicing events and/or polyadenylation sites usage that
remain to be isolated.
These three transcripts differing in their 5¢UTR part
arise from a single Agouti gene using three different
start sites. Analysis of the putative promoter regions
expressing A and B transcripts did not reveal consensus
TATA or CAAT boxes but displayed putative binding
sites for transcription factors such as Sp1 sites, known
to initiate transcription in TATA-less promoters (Fig. 1).
The lack of CAAT or TATA motifs and the presence of
numerous Sp1 sites preceding transcription initiation
sites are features typical of constitutively expressed
housekeeping genes or proto-oncogenes (Azizkhan
et al., 1993) which is in agreement with the Agouti
widespread expression in bovine tissues. Furthermore,
each promoter exhibits a consensus INR and a DPE
consensus sequence. All together, these data strongly
suggest that A and B transcripts result from distinct proPigment Cell Res. 18; 34–41
Widespread expression of bovine Agouti gene
moters. The third promoter corresponding to the skin
transcript remains to be identified.
Preliminary studies of A and B transcripts with m-fold
using the Zucker algorithm (Zuker, 2003), predicted very
stable secondary structures of the A 5¢UTR (dG ¼
)90 kcal) and B 5¢UTR (dG ¼ )99 kcal). Moreover, we
found four short upstream open reading frames (uORFs)
in the A 5¢UTR, and six uORFs in the B 5¢UTR. Thus,
the alternative utilization of multiple promoters might
serve to fine tune Agouti mRNA performances. This
could occur at the level of transcript production, transcript stability, and/or transcript translation efficiency, as
each variant has a different 5¢UTR. Secondary structures
and/or uORFs in eukaryotic 5¢-UTR mRNAs regulate the
main ORF (open reading frame) translation, as demonstrated experimentally using bicistronic reporter assay
(Negulescu et al., 1998), or mutational analysis of the
uORFs (Zimmer et al., 1994). The three heterogeneous
5¢UTR sequences which belong to three promoters may
be physiologically significant. Accordingly Agouti phenotype of wild-type mouse results from a single coding
sequence with distinct 5¢-untranslated exons regulated
by alternative promoters that control dorsum and ventrum coat colour independently (Vrieling et al., 1994).
However, these preliminary data needs subsequent
investigations to evaluate the potential role of the secondary structures and/or uORFs on the initiation efficiency of protein synthesis from the major ORF of
bovine Agouti transcripts.
A comparative analysis of multi-species sequences
allowed us to identify orthologous regulatory regions in
the pig and human AGOUTI genes. This approach postulates that functional regions are under selective pressure and tend to be more highly conserved than
non-functional regions that are subject to random genetic drift, so local sequence similarity usually indicate
biological functionality (Morgenstern et al., 2002). The
high per cent identity between bovine 5¢UTR, pig and
human non-coding region suggests that some uncharacterized mRNA transcripts containing the orthologous
5¢UTR sequences (region II and III) could also exist in
pigs and humans. Most remarkable is the sequence
conservation of the region I corresponding to the bovine
A promoter. This means that the related A transcripts in
pigs and humans, if they exist, are regulated alike.
Beside this, the bovine B promoter was not included in
the pairwise alignment because of the hidden repetitive
elements (SINE, LINE). Indeed, the LAGAN software
hides repetitive elements before proceeding to do not
distort the pairwise alignment (Brudno et al., 2003). Furthermore, this bovine intronic region between the 1A
and 2A exons seem species specific because the corresponding region is only 315 bp length in pig and does
not exist in human. Thus, if the B transcripts exist in
pigs and humans, their expression must be driven by
another genomic sequence. This region could be the
high identity intronic sequences downstream of the 1A
Pigment Cell Res. 18; 34–41
exon. Although these in silico data need to be supported by experimental evidence, it is now widely accepted
that comparative sequence analysis is a powerful and
universally applicable tool for genome analysis and annotation (Morgenstern et al., 2002; Wasserman & Sandelin, 2004). Moreover, these data are supported by the
fact that the bovine Agouti gene is embedded in a conserved syntenic bloc, which contains HCK, AHCY, ASIP,
GHRH, PLC-II, PPGB, PLTP, present on BTA 13 and
HSA 20 (Schlapfer et al., 2002).
In sum, although the role of Agouti in cattle remains
to be proved, the identification of regulatory regions
from the bovine, pig and human genes is a first step
towards the investigation of regulatory mutations
involved in coat colour phenotypic variations or other
physiological alterations such as adipocyte metabolism.
Materials and methods
Animals
Each animal studied came from standardized breeds homogenous
with respect to coat colour: Charolais (creamy white), Limousine
(red), Salers (reddish brown), Montbéliarde (brown with white
spotting), Gasconne (grey), Normande (brindle brown/black with
white spotting), Blonde d’Aquitaine (Blond) and Prim’ Holstein
(black with white spotting). Serum and skin (routinely 25 cm2)
samples were obtained from Limoges slaughterhouse (France)
and from several UPRA (Union pour la Promotion des Races
Animales).
Long-range PCR
The Expand Long-template PCR (Roche diagnostics, Meylan,
France) was performed according to the manufacturer’s instructions
to amplify the bovine Agouti gene. Genomic DNA was prepared
from blood samples as described previously (Rouzaud et al., 2000).
The consensus DNA sequence deduced from the alignment of murine (Bultman et al., 1992) and human (Kwon et al., 1994) Agouti
genes was used to design the Ag1 (5¢-ATG GAT GTC ACC CGC
YTA CTC CTG GC CA-3¢) and Ag2 (5¢-TCA GCA GTT GRG GYT GAG
YAC KCG RC-3¢) degenerated primers (Fig. 1). Amplification of
bovine genomic DNA with Ag1/Ag2 primers produced a 5139 bp
length PCR product. This product was cloned into Topo XL vector
(Invitrogen SARL, Cergy Pontoise, France) and sequenced (ABI
Prism 310 DNA Genetic Analyzer; Perkin Elmer France, Courtaboeuf, France). The sequence of the amplified DNA fragment was
then used to design gene-specific primers: Ag3 (outer antisense
5¢-CCA CGT TCT TCA TCG GAG CCT TTC T-3¢) and Ag4 (nested antisense 5¢- TCA GCG CCA CGA TAG AGA CTG AAG G-3¢); Ag5 (outer
sense 5¢-TCC TCC TGG CTA CCT TGC TGG TCT G-3¢) and Ag6 (nested sense 5¢-GCT TCC TCA CTG CCT ACA GCC ACC T-3¢) to obtain
5¢ and 3¢ UTRs (UnTranslated Regions), respectively (Fig. 1).
Promoter region sequences
A transcript Promoter sequence was obtained using the Genome
Walker kit (Ozyme France, St Quentin Yvelines, France), following
the protocol recommended by the manufacturer. Briefly, this kit
contains all reagents required to perform four bovine genomic DNA
libraries, each containing adaptor-ligated products from digested
bovine genomic DNA with one of four restriction enzymes (EcoRV,
DraI, PvuII or StuI). Sequences of interest were obtained by one
nested PCR amplification per library. The first PCR amplification
39
Girardot et al.
used the outer adaptor primer (AP1) provided in the kit and an
outer, gene-specific primer. Nested gene-specific primer and nested adaptor primer (AP2) were used in the second PCR to amplify
genomic DNA within the region of interest. Gene specific primer
sequences used to obtain the A transcript promoter region were as
follows: Ag7 (outer primer: 5¢-TTG AAT TGG GCA CAT TCT CCT
TTC CTC-3¢) and Ag8 (nested primer: 5¢-GAG TGA GCA GAT TCC
ACC ACC AAC ATC-3¢). B transcript promoter region was obtained
by direct amplification of the sequence between exons 1A and 2A
using primers Ag9 (5¢-AAA GCA GCA TGA GGA AAG GA-3¢) and
Ag10 (5¢-GGC ACA TAT CAA GCA TCA GC-3¢) (Fig. 1).
Skin RNA preparations
Total skin RNA was prepared as follows. First, skin fatty tissue
was trimmed and hairs shaved. Then, skin samples were cut in
small pieces and subjected to RNA extraction using the RNeasy
Maxi Kit (Qiagen S.A., Courtaboeuf, France), according to the
manufacturer’s instructions. Skin poly (A)+ RNA was purified from
total RNA using NucleoTrappoly (A) RNA kit (Macherey-Nagel,
Hoerdt, France). Poly (A)+ RNA from brain, lung, heart, liver, kidney
and spleen bovine tissues were purchased from Clontech (Ozyme
France).
5¢ and 3¢-Rapid amplification of cDNA end (RACE)
To isolate the full-length cDNA, RACE experiments were performed
on 1 lg total RNA from bovine skin, using the SMARTTM RACE
cDNA Amplification kit (Ozyme France), according to the manufacturer’s instructions. 5¢ and 3¢ UTRs of Agouti skin transcript were
amplified by nested PCR with specific and adaptor primers: Ag3/
UPM (Universal Primer Mix) and Ag4/NUP (Nested Universal Primer) for the first and second 5¢UTR amplifications respectively, and
Ag5/UPM and Ag6/NUP for the first and second 3¢UTR amplifications respectively. First and second PCR amplifications were carried out in a 25 ll reaction volume containing 10 pmol of each
primer and 12.5 ll of 2X working concentration PCR Master Mix
(ABgene France, Yvette Courtaboeuf, France) in the following cycling conditions: initial denaturation at 94C for 2 min followed by 30
cycles (96C for 30 s, 61C for 30 s, 72C for 2 min) and one cycle
(72C for 5 min). PCR products were cloned into the pCR2.1 vector
(Invitrogen SARL) and sequenced (ABI Prism 310 DNA Genetic
Analyzer). The same approach was used on 1 lg lung poly (A)+
RNA to determine 5¢UTR sequences.
Tissue expression of Agouti
Analysis of Agouti expression by Northern hybridization was
achieved using 3 lg mRNA. Purified poly(A)+ RNAs from skin and
various tissues were fractionated on agarose formaldehyde gel
electrophoresis (1.2%) and transferred to a nylon membrane
(Hybon-XL, Amersham Pharmacia Biotech, UK). RNA was crosslinked to the membrane using UVR (120 mJ). The entire coding
region of the bovine Agouti cDNA was labelled with deoxycytidine
5¢[a32 P]triphosphate by the random primer extension method
(Feinberg & Vogelstein, 1983), using a commercially available kit
(GibcoBRL, Cergy Pontoise, France). Blots were hybridized
overnight at 61C with the Agouti coding region probe and stringently washed twice for 5 min each time with 2X SSC (30 mM
sodium citrate, 0.3 M sodium chloride, pH 7) and 0.1% sodium
dodecyl sulphate (SDS), followed by two washes for 10 min each,
with 1X SSC (15 mM sodium citrate, 150 mM sodium chloride,
pH 7) and 0.1% SDS. The blots were then washed twice for
5 min each time, with 0.1X SSC (1.5 mM sodium citrate, 15 mM
sodium chloride) and 0.1% SDS. Finally, the membrane was
exposed to PhosphorimagerTM 445SI (Molecular Dynamics) for
2 h.
40
Bioinformatics analysis
Orthologous sequences were found using the BLASTn software
accessible at: http://www.ncbi.nlm.nih.gov/BLAST/. Genomic
sequence from promoter A to exon 3A (Fig. 1) was analyzed using
M-LAGAN (Brudno et al., 2003), publicly available at http://
lagan.stanford.edu. Alignment results were visualized with the
m-Vista software (Mayor et al., 2000) accessible at http://
gsd.lbl.gov/vista/index.shtml.
Acknowledgements
We would like to thank the bovine Unités pour la sélection et la
promotion des races bovines françaises (UPRAs), France UPRA
Sélection and Labogena (Jouy-en-Josas) for providing blood and skin
samples. We thank Dr V. Amarger and V. Cetica for their helpful
suggestions and critical reading of the manuscript. We thank MariePierre Laforet and Lionel Forestier for their technical assistance. This
work was supported by MJENR, MENRT, and INRA grants.
References
Adalsteinsson, S., Bjarnadottir, S., Vage, D.I., and Jonmundsson,
J.V. (1995). Brown coat color in Icelandic cattle produced by the
loci Extension and Agouti. J Hered. 86, 395–398.
Azizkhan, J.C., Jensen, D.E., Pierce, A.J., and Wade M. (1993).
Transcription from TATA-less promoters: dihydrofolate reductase
as a model. Crit Rev Eukaryot Gene Expr. 3, 229.
Brudno, M., Do, C.B., Cooper, G.M., Kim, M.F., Davydov, E.,
Green, E.D., Sidow, A., and Batzoglou, S. (2003). LAGAN and
Multi-LAGAN: efficient tools for large-scale multiple alignment of
genomic DNA. Genome Res. 13, 721–731.
Bultman, S.J., Michaud, E.J., and Woychik, R.P. (1992). Molecular
characterization of the mouse Agouti locus. Cell. 71, 1195–1204.
Duhl, D.M., Stevens, M.E., Vrieling, H., Saxon, P.J., Miller, M.W.,
Epstein, C.J., and Barsh, G.S. (1994). Pleiotropic effects of the
mouse lethal yellow (Ay) mutation explained by deletion of a
maternally expressed gene and the simultaneous production of
Agouti fusion RNAs. Development. 120, 1695–1708.
Feinberg, A.P., and Vogelstein, B. (1983). A technique for radiolabeling DNA restriction endonuclease fragments to high specific
activity. Anal Biochem. 132, 6–13.
Guibert, S., Girardot, M., Leveziel, H., Julien, R., and Oulmouden,
A. (2004). Pheomelanin coat colour dilution in French cattle
breeds is not correlated with the TYR, TYRP1 and DCT transcription levels. Pigment Cell Res. 17, 337–345.
Kanetsky, P.A., Swoyer, J., Panossian, S., Holmes, R., Guerry, D.,
and Rebbeck, T.R. (2002). A polymorphism in the Agouti signaling protein gene is associated with human pigmentation. Am J
Hum Genet. 70, 770–775.
Kemp, S.J., and Teale, A.J. (1994). Randomly primed PCR amplification of pooled DNA reveals polymorphism in a ruminant repetitive DNA sequence which differentiates Bos indicus and B.
taurus. Anim Genet. 25, 83–88.
Kim, K.S., Mendez, E.A., Marklund, S., Clutter, A.C., Pomp, D., and
Rothschild, M.F. (2000). Rapid communication: linkage mapping
of the porcine Agouti gene. J Anim Sci. 78, 1395–1396.
Klungland, H., Vage, D.I., Gomez-Raya, L., Adalsteinsson, S., and
Lien, S. (1995). The role of melanocyte-stimulating hormone
(MSH) receptor in bovine coat color determination. Mamm Genome. 6, 636–639.
Kwon, H.Y., Bultman, S.J., Loffler, C., Chen, W.J., Furdon, P.J.,
Powell, J.G., Usala, A.L., Wilkison, W., Hansmann, I., and Woychik, R.P. (1994). Molecular structure and chromosomal mapping
Pigment Cell Res. 18; 34–41
Widespread expression of bovine Agouti gene
of the human homolog of the Agouti gene. Proc Natl Acad Sci U
S A. 91, 9760–9764.
Kyte, J., Doolittle, R.F. (1982). A simple method for displaying the
hydropathic character of a protein. J Mol Biol. 157, 105–132.
Lauvergne, J.J. (1966). la couleur du pelage des bovins domestiques. Biblographica Genetica. 20, 1–168.
Matz, M., Shagin, D., Bogdanova, E., Britanova, O., Lukyanov, S.,
Diatchenko, L., and Chenchik, A. (1999) Amplification of cDNA
ends based on template-switching effect and step-out PCR.
Nucleic Acids Res. 27, 1558–1560.
Mayor, C., Brudno, M., Schwartz, J.R., Poliakov, A., Rubin, E.M.,
Frazer, K.A., Pachter, L.S., and Dubchak, I. (2000). VISTA: visualizing global DNA sequence alignments of arbitrary length. Bioinformatics. 16, 1046–1047.
Miller, M.W., Duhl, D.M., Vrieling, H., Cordes, S.P., Ollmann,
M.M., Winkes, B.M., and Barsh, G.S. (1993). Cloning of the
mouse Agouti gene predicts a secreted protein ubiquitously
expressed in mice carrying the lethal yellow mutation. Genes
Dev. 7, 454–467.
Morgenstern, B., Rinner, O., Abdeddaim, S., Haase, D., Mayer,
K.F., Dress, A.W., and Mewes, H.W. (2002). Exon discovery by
genomic sequence alignment. Bioinformatics. 18, 777–787.
Negulescu, D., Leong, L.E., Chandy, K.G., Semler, B.L., and Gutman, G.A. (1998). Translation initiation of a cardiac voltage-gated
potassium channel by internal ribosome entry. J Biol Chem. 273,
20109–20113.
Ollmann, M.M., Lamoreux, M.L., Wilson, B.D., and Barsh G.S.
(1998). Interaction of Agouti protein with the melanocortin 1
receptor in vitro and in vivo. Genes Dev. 12, 316–330.
Olson, T.A. (1982). Inheritance of Coat Colour in Cattle: A Review
(Ames, IA: Iowa State University and Agricultural Experiment
Station Bulletin), p. 595.
Olson, T.A. (1999). The Genetics of Cattle. (Wallingford, Oxon: CAB
International).
Rouzaud, F., Martin, J., Gallet, P.F., Delourme, D., Goulemot-Leger,
V., Amigues, Y., Ménissier, F., Levéziel, H., Julien, R., and
Oulmouden, A. (2000). A first genotyping assay based on a new
Pigment Cell Res. 18; 34–41
allele of the extension gene encoding the melanocortin-1 receptor (Mc1r). Genet Sel Evol. 32, 511–520.
Schlapfer, J., Stahlberger-Saitbekova, N., Comincini, S., Gaillard, C.,
Hills, D., Meyer, R.K., Williams, J.L., Womack, J.E., Zurbriggen,
A., and Dolf, G. (2002). A higher resolution radiation hybrid map
of bovine chromosome 13. Genet Sel Evol. 34, 255–267.
Searle, A.G. (1968). Comparative Genetics of Coat Colour in Mammals. (London: Logos Press).
Silvers, W. (1979). The Coat Colors of Mice. (New York: SpringerVerlag).
Sumida, T., Hino, N., Kawachi, H., Matsui, T., and Yano, H. (2004).
Expression of Agouti gene in bovine adipocytes. Animal Sci J.
75, 49–51.
Voisey, J., and van Daal, A. (2002). Agouti: from mouse to man,
from skin to fat. Pigment Cell Res. 15, 10–18.
Voisey, J., Box, N.F., and van Daal, A. (2001). A polymorphism
study of the human Agouti gene and its association with MC1R.
Pigment Cell Res. 14, 264–267.
Vrieling, H., Duhl, D.M., Millar, S.E., Miller, K.A., and Barsh, G.S.
(1994). Differences in dorsal and ventral pigmentation result from
regional expression of the mouse Agouti gene. Proc Natl Acad
Sci U S A. 91, 5667–5671.
Wasserman, W.W., and Sandelin, A. (2004). Applied bioinformatics
for the identification of regulatory elements. Nat Rev Genet. 5,
276–287.
Wilson, B.D., Ollmann, M.M., Kang, L., Stoffel, M., Bell, G.I., and
Barsh, G.S. (1995). Structure and function of ASP, the human
homolog of the mouse Agouti gene. Hum Mol Genet. 4, 223–
230.
Wright, S. (1917). Color inheritance in mammals. J Hered. 8, 521.
Zimmer, A., Zimmer, A.M., and Reynolds, K. (1994). Tissue specific
expression of the retinoic acid receptor-beta 2: regulation by
short open reading frames in the 5¢-noncoding region. J Cell Biol.
127, 1111–1119.
Zuker, M. (2003). Mfold web server for nucleic acid folding and
hybridization prediction. Nucleic Acids Res. 31, 3406–3415.
41