Studies on a fish Gene that Relates to Efficient Growth and

博士(理学) Eman Mohamed Mamdouh Mohamed Abbas
平成 23 年 9 月 30 日
Studies on a fish gene that relates to efficient growth and
--- Structure of myostatin gene promoter in Percoidei fishes --(魚類の増殖・分化関連遺伝子に関する研究:スズキ亜目魚類の
加藤 幹男
多田 俊治
八木 孝司
森 展子
論 文 要 旨
Myostatin, also known as growth differentiation factor 8 (GDF-8), is a member of the
transforming growth factor-β (TGF-β) family that functions as a negative regulator of skeletal
muscle development and growth in mammals. A model for the biological activity of myostatin has
been proposed in that myostatin inactivates a specific group of genes, and a negative correlation
between myostatin expression and growth/number of muscle fibers has been observed. It is widely
known that myostatin-knockout mice exhibit a dramatic increase of skeletal muscle mass that
results from a combination of hyperplasia and hypertrophy. Natural mutation of the myostatin
gene has been observed in the double muscle breeds of Belgian Blue and Piedmontese cattle,
which have significantly more muscle mass than normal breeds do. Myostatin is found almost
exclusively in the skeletal muscle in mammals, but appears more ubiquitously in fish, suggesting
more diverse functions in their growth and development. A single encoding gene in mammals
expresses myostatin, whereas teleost fish possess at least two myostatin genes that are
differentially expressed in both muscular and non-muscular tissues. Recently, it was reported that
silencing of the myostatin genes resulted in a giant phenotype in zebrafish, and a
dominant-negative form of myostatin resulted in the doubling of muscle-fiber number in
transgenic medaka.
Myostatin genes have been characterized in several commercially important fishes such as striped
bass, white perch, Mozambique tilapia, white bass, Atlantic salmon, rainbow trout, gilthead sea
bream, shi drum, catfish species, European sea bass, Croceine croaker, orange spotted grouper,
and Japanese sea perch. Worldwide fish consumption has been steadily increasing, and strategies
for enhancing skeletal muscle growth of aquaculture species can help to meet the increasing
demand for this protein source. It is quite important to determine how myostatin gene expression
is regulated and to uncover the protocol that controls the action of myostatin. To gain better
understanding of the mechanisms regulating myostatin gene expression, we have analyzed the
genomic structures of the myostatin gene in a sea bass, Lateolabrax japonicus, spanning primary
DNA sequences to higher-ordered structures (chromatin and DNA methylation). We have also
examined the transient expression of green fluorescent protein in the embryos of medaka (Oryzias
latipes) driven by a myostatin promoter isolated from a sea bream (Acanthopagrus latus). This
trial will assist us in evaluating the environmental effect on myostatin expression in future studies
using transgenic medaka fish.
Results and Discussion
(1) Methylation status of myostatin gene promoter region in Lateolabrax japonicus
DNA methylation at cytosine residues is involved in epigenetic regulation to suppress gene
expression through heterochromatinization. However, the methylation status in fish genomes
remains unclear. To examine the methylation status in myostatin promoter regions of Lateolabrax
japonicus, total genomic DNA was isolated from tissues of the brain, kidney, spleen, liver, heart,
eye, muscle, intestine, and gill. The genomic DNA samples were digested with a restriction
enzyme and subjected to bisulfite modification. The bisulfite treatment introduces specific
changes in the DNA sequence that depend on the methylation status of individual cytosine
residues, yielding information on the DNA segments’ methylation status. DNA fragments were
treated with sodium bisulfite to convert unpaired cytosine residues to uracil under conditions
whereby methylated cytosine remained essentially intact. The modified DNA was used as the
PCR template to specifically amplify the sense and antisense strands of bisulfite-treated DNA
with their respective PCR primer pairs. The PCR products from each reaction were subjected to a
second PCR amplification with nested primers to obtain the DNA fragments that originated from
each strand. They were cloned into a plasmid vector and the nucleotide sequences were
determined. The efficiency of the bisulfite modification (conversion of C to T) was estimated by
counting the numbers of cytosine residues remaining at CpN sites (in the case of N ≠ G), and the
conversion frequency was found to be 99% or higher. The number of cytosine residues remaining
at CpG sites appeared to be slightly higher than that in the background level in sense strands from
the eye and heart, but were identical to that in the background level in complementary strands
from the same tissues. The frequency of methylated cytosine was very low in the tissues examined
(within the error range of random sampling) regardless of the level of myostatin gene expression.
The results may mean that the methylation at CpG is not involved in regulation of the myostatin
gene in L. japonicus.
(2) Probing the chromatin structures in L. japonicus
Gene expression also reflects the status of the chromatin (euchromatin or heterochromatin) in the
region where the gene is situated. The status of the chromatin in particular gene regions is
dynamically modulated to control gene expression and other fundamental cellular processes such
as proliferation and differentiation. Active chromatin can be distinguished from bulk chromatin by
its increased susceptibility to endonuclease. Micrococcal nuclease (MNase) preferentially cleaves
chromatin between nucleosomes (naked DNA regions). To examine tissue-specific chromatin
structures in myostatin gene regions by MNase as an enzymatic probe, nuclei were isolated from
some L. japonicus organs (the liver, eye, kidney, brain, and heart), purified, and treated with
MNase. The mixtures were incubated at 37°C for 15, 30, and 60 min, and one-third of the original
reaction volumes were collected at each time point. The DNA was isolated from each reaction
mixture and subjected to Southern blot hybridization analysis after restriction digestion. A
fragment of the L. japonicus myostatin promoter region was used as a probe to detect an
approximately 2 kb TaqI restriction fragment. Myostatin gene promoter regions in the brain and
eye were highly susceptible to MNase while those in the heart were less susceptible, and those in
the kidney and liver were even more resistant to MNase. The results suggest that the myostatin
gene is compacted into heterochromatin in tissues where it is not expressed, such as in the liver
and kidney, whereas it is susceptible to MNase in the eye and brain, which do express myostatin.
According to the results of methylation analysis and MNase probing, DNA methylation may not
be involved in regulating heterochromatinization and thus, expression of myostatin, in L.
(3) Genomic sequences of myostatin gene promoter regions in Sparidae fishes
The 5′-flanking regions of the myostatin gene were isolated from two sea breams, Acanthopagrus
latus and Acanthopagrus schlegelii, by inverse PCR. Genomic DNA of A. latus and A. schlegelii
were digested with restriction enzymes and the fragments were self-circularized under low DNA
concentration. The closed circular DNA molecules were used as the template for PCR with a set
of DNA primers for the first-round PCR, after which nested PCR was performed with another set
of DNA primers. The PCR products were cloned into a plasmid vector and sequenced by the
primer-walking strategy with internal primers. After the entire sequence was determined, a
genomic DNA fragment was amplified with 5′- and 3′-end primers to confirm the genomic
sequences. The partial-coding regions of cytochrome oxidase subunit I (COI) and 18S rRNA from
the specimens were also amplified and sequenced to confirm identity of species.
Two alleles of the myostatin gene were identified for both A. latus and A. schlegelii. The
nucleotide sequences were aligned with the promoter region of the Sparus aurata myostatin gene,
showing that the conserved regions spanned approximately 1 kb. The potential cis-acting elements
were observed in highly conserved regions between these two Sparidae species. They contain two
putative TATA-boxes, one CAAT box, and seven putative E-boxes. Comparative analysis of
myostatin gene regulatory regions with those of other Percoidei fishes revealed the occurrence of
highly conserved regions approximately 300 bp upstream of the translation start site.
(4) Evaluation of promoter activity in transient expression of GFP in medaka embryos
The genomic fragment was connected to the AcGFP coding sequence to evaluate the activity of
the promoters isolated from A. latus. The recombinant fragments were cloned into a plasmid
vector. Six different myostatin fragments in total, truncated at their 5′ ends, were constructed.
Some were used for microinjection into the cytoplasm of fertilized medaka fish at the one-cell
stage. Embryos were observed under a fluorescence microscope to visualize GFP expression. The
expression of the AcGFP driven by the A. latus myostatin promoter in the medaka embryos was
weak but distinctive, with a series of truncated fragments. Based on the current available results,
we believe that the genomic fragment obtained from A. latus has minimal promoter activity in
medaka embryos. The A. latus promoter might not have worked well in medaka because medaka
(Beloniformes) is distantly related to sea bream (Perciformes) evolutionarily, or the A. latus DNA
fragment used may have been not enough to express full promoter activity.
It is challenging to clarify the regulatory mechanisms of the myostatin gene in fish.
Myostatin expression is suppressed parallel to heterochromatinization but regardless
of DNA methylation in L. japonicus. Sequencing of 5′-flanking DNA fragments of the
myostatin-coding region in A. latus identified several potential cis-elements and
higher similarity to the myostatin gene promoter in other fish species. Weak promoter
activity of DNA fragments isolated from A. latus was observed in medaka embryos in
a transient expression assay. This assay system may allow us to evaluate the role of
promoter elements in vivo. Establishing a transgenic medaka carrying AcGFP driven
by an A. latus myostatin promoter will be extremely valuable to efforts to examine
expression profiles in response to environmental stimuli under culture conditions in
the future.
本研究では、Chapter 1 において、スズキ Lateolabrax japonicus のミオスタチン遺伝子領
域の DNA のメチル化とクロマチン構造について解析した。DNA のメチル化は、哺乳類(CpG
部位の C のメチル化)、および高等植物(CpNpG 部位の C のメチル化)において、遺伝子の
うち近縁種間で高度に保存されている領域のメチル化状態を調べたところ、有意な CpG 部位
のメチル化は検出されなかった。一方、MNase 限定分解によるクロマチン構造解析を行った
ところ、ミオスタチン発現組織において MNase 感受性が高く、非発現組織においては分解
ヘテロクロマチン化による発現制御がなされており、DNA メチル化の関与はないことが示唆
Chapter 2 において。キチヌ Acanthopagrus latus およびクロダイ A. schlegelii のミオス
Sparus aurata における既知のミオスタチン遺伝子構造との比較により、翻訳開始点から上
流約 1kb にわたる領域が保存されており、また、様々な転写因子に対する相互作用部位の存
在が示唆された。続いて、キチヌから得た DNA 断片を緑色蛍光タンパク質遺伝子へとつな