Kobe University Repository : Thesis 学位論文題目 Title Elucidation of the mechanism leading to the onset of chicken muscular dystrophy(ニワトリ筋ジストロフィー発症機序の解明) 氏名 Author Matsumoto, Hirokazu 専攻分野 Degree 博士(農学) 学位授与の日付 Date of Degree 2010-03-25 資源タイプ Resource Type Thesis or Dissertation / 学位論文 報告番号 Report Number 甲4944 権利 Rights JaLCDOI URL http://www.lib.kobe-u.ac.jp/handle_kernel/D1004944 ※当コンテンツは神戸大学の学術成果です。無断複製・不正使用等を禁じます。著作権法で認められている範囲内で、適切にご利用ください。 PDF issue: 2020-10-12
103
Embed
Elucidation of the mechanism leading to the onset of ...muscular dystrophies include progressive muscle wasting, limited range of movement, walking difficulty, respiratory difficulty
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Kobe University Repository : Thesis
学位論文題目Tit le
Elucidat ion of the mechanism leading to the onset of chicken musculardystrophy(ニワトリ筋ジストロフィー発症機序の解明)
G ENE RA L I NT ROD UC T ION -----------------------------------------------------------------------1
CHAPTER I Pinpointing the candidate region for muscular dystrophy in chickens with an abnormal muscle gene I n trod u cti 0 n ----------------------------------------------------------------------------5 Mate rials & Meth od s -------------------------------------------------------------7
Res u Its & 0 iscuss io n ------------------------------------------------------------12
CHAPTER II The ubiquitin ligase gene (WWP1) is responsible for the chicken muscular dystrophy I ntrod u ctio n --------------------------------------------------------------------------19
M ate ria Is & Method s -------------------------------------------------------------21
Res u Its ---------------------------------------------------------------------------------27
Dis cu ss ion ----------------------------------------------------------------------------3 2
CHAPTER III Expression pattern of WWP1 in muscular dystrophic and normal chickens I nt rod u ct ion --------------------------------------------------------------------------3 7
Materia Is & Method s -------------------------------------------------------------39
Resu Its & 0 iscussion ------------------------------------------------------------42
CHAPTER IV Mutated WWP1 induces an aberrant expression of myosin heavy chain gene in C2C12 skeletal muscle cells I n trod u cti 0 n --------------------------------------------------------------------------4 7
Mate ri a Is & Meth od s -------------------------------------------------------------49
Res u Its & 0 iscu ssio n ------------------------------------------------------------54
CHAPTER V Overexpression of caveolin-3 protein is limited in damaged muscle in chicken muscular dystrophy I nt rod u cti a n --------------------------------------------------------------------------60
Materia Is & Meth ad s -------------------------------------------------------------63
Res u Its ---------------------------------------------------------------------------------67
Dis cuss ian ----------------------------------------------------------------------------73
G ENE RA L 0 I S CESS ION ----------------------------------------------------------------------------77
A C KN OW L E DG MEN T S ------------------------------------------------------------------------------82
REF ERE N C E -------------------------------------------------------------------------------------------------83
GENERAL INTRODUCTION
Muscular dystrophies are defined as the group of inherited diseases with
progressive weakness and degeneration of skeletal muscle, and there exist
plural genetical backgrounds (Partridge, 1991). Main symptoms of the
muscular dystrophies include progressive muscle wasting, limited range of
movement, walking difficulty, respiratory difficulty and so on. Some types of
muscular dystrophies can also affect the heart, causing cardiomyopathy or
arrhythmias (Nonaka, 1987a). The prognosis for the patients varies widely.
Some patients of muscular dystrophies die in infancy, while others are
affected only moderately and live into adulthood. In addition, the muscles
affected vary according to the type of the disease. Many types of muscular
dystrophies tend to occur in early childhood, though some forms can affect
adults.
In many cases, abnormalities of muscle proteins to compose of the
linkage between sarcolemma and basal lamina can lead the onset of
muscular dystrophies (Imamura et a1., 2000). The most famous and earliest
identified protein responsible for muscular dystrophies is dystrophin.
Mutations in dystrophin gene lead loss or lack of dystrophin protein from
sarcolemma, which results in two types of muscular dystrophies, Duchenne
and Becker muscular dystrophy, accounting for most of the diseases (Nonaka,
1987a). Duchenne muscular dystrophy (dystrophin-deficient) is among the
most severe forms of muscular dystrophies; hence the function of dystrophin
- 1 -
and its related proteins has been intensively studied for these years.
However, numbers of muscular dystrophies and related diseases need
elucidating the responsible genes or the mechanisms (Terri & Kunkel, 2000;
Lisi & Cohn, 2007).
Animal models are very useful tools to examine a pathophysiologic
role of certain molecule, to elucidate a mechanism of a disease and to
promote a therapeutic research. From this perspective, various animal
models for muscular dystrophies have been established and utilized.
Representative models in Japan are mdxmouse (dystrophin-deficient), grmd
dog (dystrophin-deficient), dy mouse (merosin-deficient), muscular
dystrophic hamster (O-sarcoglycan-deficient) and muscular dystrophic
chicken (Nonaka, 1987b). These models are ones with abnormalities of
dystrophin or other related proteins, except for muscular dystrophic chicken.
Thanks to these animal models, the studies about muscular dystrophies
lacking dystrophin and/or its related proteins have been intensively
progressed, while those have not about other types of muscular dystrophies.
Chicken muscular dystrophy with abnormal muscle (Alld) has been
known since 1956 (Asmundson & Julian, 1956), and several strains for this
disease have been established. One of them is New Hampshire 413-strain
(NH-413). This strain was introduced from University of California, Davis to
Japan in 1976 (Kondo et a1., 1982). The phenotype of AM dystrophic chickens
is relatively mild and they are able to breed, which makes it easy to obtain
- 2-
fertilized eggs with AM gene and AlJ1lAMhomozygous chicks. Therefore, this
animal model is regarded as the suitable for a diachronic study to examine
when and how muscular dystrophies start to affect (Nonaka, 1987b). The
disease in NH-413 is transmitted co dominantly by a single gene, and the
phenotype is modified by other backcross genes (Asmundson & Julian, 1956;
Asmundson et ai., 1966; Kikuchi et ai., 1981; Wagner & Peterson, 1970). The
responsible gene has not been identified yet, and dystrophin and/or other
known causative proteins seem not to be responsible gene for this disease
(Saito et a1., 2005), suggesting that chicken muscular dystrophy is the one
triggered by a different mechanism from known muscular dystrophies.
Elucidating the responsible gene and the mechanism leading the onset of
chicken muscular dystrophy would provide new insights to understand
muscular dystrophies and related diseases whose causes are unknown.
In prevlOUS study applying linkage analysis, the AM locus was
mapped to chicken chromosome 2q (Lee et ai., 2002) and the candidate
region was narrowed down to 3.6 Mbp range (Yoshizawa et ai., 2004). This
region included 34 functional genes, some of which functions were unknown.
Therefore, further analyses were required to pinpoint the AM candidate
region, identify the responsible gene and understand the mechanism of
chicken muscular dystrophy.
The aim of this study was to identify the responsible gene for chicken
muscular dystrophy and to elucidate the mechanism leading its onset. By
- 3-
haplotype analysis with F2 resource family for chicken muscular dystrophy,
AM candidate region was reduced to approximately 1 Mbp, including seven
functional genes as candidate genes. Subsequent sequence comparison and
expression analysis of seven candidate genes in normal and dystrophic
chickens revealed that one of candidate genes, WWP1, has the mutation
specific to muscular dystrophic chickens, suggesting that WWPl is the
causative gene of chicken muscular dystrophy. Transfecting mouse WWPl
with the mutation homologous to the responsible mutation for chicken
muscular dystrophy into C2C12 mouse myoblasts resulted in the disturbed
expressions of muscle differentiation markers. Caveolin-3 expression was
analyzed as a candidate protein interacting with WWPl, which suggested
that the amount of caveolin-3 protein is controlled by WWPI and its
aberrant regulation causes the onset of chicken muscular dystrophy.
- 4-
CHAPTER I
Pinpointing the candidate region for muscular dystrophy
in chickens with an abnormal muscle gene
Introduction
Chicken muscular dystrophy with abnormal muscle (.AA1) has been known
for over 50 years (Asmundson & Julian, 1956), but the responsible gene has
not been identified yet. It has been well studied that abnormalities of muscle
proteins to compose of the linkage between sarcolemma and basal lamina,
such as dystrophin, can lead muscular dystrophies in many cases (Imamura
et a1., 2000). However, the expressions of dystrophin and/or its related
proteins seem unaffected in this disease (Saito et a1., 2005). Elucidating the
responsible gene of chicken muscular dystrophy would provide new insights
to understand muscular dystrophies and related diseases whose causes are
unknown.
Previously, the AM locus was mapped to chicken chromosome 2q
using a linkage map constructed with the Kobe University resource family
(Lee et a1., 2002). Chicken consensus map by Schmid et a1. (2000) revealed
that this region was syntenic to the human chromosome 8ql1-24.3,
indicating the genes on this region were candidates for the disease, although
approximately 600 functional genes exist on the human chromosome
8ql1-24.3. Subsequently, the candidate region was narrowed down to 3.6
- 5-
Mbp range on GGA2q, including 34 functional genes (Yoshizawa et ai., 2004).
There did not exist any known causative genes in this region. Therefore,
further analyses are required to identify the responsible gene.
This study attempted to reduce the candidate region of the AMlocus.
The F2 resource family for chicken muscular dystrophy was newly
established. Total of 487 F2 chickens were used for haplotype analysis with
22 genetic markers developed in this study. The candidate region was
successfully reduced to approximately 1 Mbp, including seven functional
genes as candidate genes.
. 6·
Materials & Methods
Genetic resource_ Chicken muscular dystrophy F2 resource family was
established in this study_ F1 generation (AlI17wt), which gained by crossing a
NH-413 male (AlI17.AAd) to a GSP female (wt/wt), was intercrossed to produce
F2 offspring. NH-413 is a New Hampshire muscular dystrophy strain
introduced from California University in 1976 (Kondo et ai., 1982), and GSP
has been established as an inbred strain from Fayoumi breed by Nippon
Institute for Biological Science in 1971. Total of 487 F2 chickens and the
parents as controls were used for haplotype analysis in this study.
To judge the phenotypes among AM homozygous, heterozygous and
normal homozygous chickens, the pectoral muscles, which are selectively
affected by chicken muscular dystrophy, were analyzed histologically. The
pectoral muscles fixed in a commercial rapid fixative (U fix; Sakura Finetek
Japan, Tokyo, Japan) were embedded in paraffin. For pathological
observation (Kikuchi et a1., 1981), paraffin sections were stained with
hematoxylin-eosin (H-E) method (Fig. I-la, b, c) and periodic acid-Schiff
(PAS) reaction (Fig. 1-ld, e, D. The degree of pathological abnormality was
evaluated by vacuolation, PAS-positive deposits and increase of nuclear in a
number within muscle fiber, hypertrophy and necrosis of muscle fiber, and
adipose replacement of muscle. FigurE' 1-la and d are the representative
images of AMlAM chickens' sections, Figure I-Ib and e the representative
ones of AMl wt chickens' and FigurE' I -lc and f the representative ones of
wt/ wt chickens'. In addition to histological judgment, enzymological analyses
- 7-
c Figure 1-1 Histopathological features of pectoral muscles of abnormal muscle (AM) homozygous chicken (a,d), heterozygous chicken (b,e) and normal homozygous chicken (c,f). (a-c) hematoxylin-eosin staining, (d-f) periodic acid-Schiff (PAS) reaction . The inset in e shows a higher magnification of a muscle fiber in e. Though vacuoles (-) were observed in the muscle fibers both .of the abnormal muscle (AM) homozygous and heterozygous chickens, the number of muscle fibers containing vacuoles in the heterozygous chicken is less than that in the AM homozygous chicken. Fatty replacement (*) was observed in the muscle tissue of the AM homozygous chicken. The PAS reaction revealed the presence of granules (arrowhead) in the sarcoplasma of both AM homozygous and heterozygous chickens. Some muscle fibers of the AM homozygous chicken contained many granules in the whole fiber. Scale bar = 120 f.J. m.
- 8-
were carried out measuring concentrations of creatine kinase and pyruvate
kinase using CK-L kit (Serotec, Sapporo, Japan) and PK test-UV kit
(Dainippon Sumitomo Pharma, Osaka, Japan).
Polymorphism searching. Based on the Gallus gallus data of Map Viewer
Database at NCBI (National Center for Biotechnology Information;
http://www.ncbi.nlm.nih.gov/), new polymorphic markers were established in
two ways, micro satellite (MS) and SNPs markers for expressed genes.
The MS region between IMPAl and CBFA2Tl were searched using
chicken genome database. Primers for MS markers were designed to amplify
using OLIGO 4.0 program (Molecular Biology Insights, Cascade, CO, USA).
The polymorphisms detected between the parents were applied to
genotyping for the resource family.
Sequencing for LOC420205, LOC420206, CGI-77 and FLJ20530
using parents' DNA samples was performed with GENE Mate Gelpure DNA
purification Kit (Isc BIO EXPRESS, Kaysville, UT, USA), BigDye®
Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA,
USA), and ABI PRISM® 3100 Genetic Analyzer (Applied Biosystems). The
primers used in this procedure were as follows:
5'-GCTAGGTCCATTGCTTTGCTCAGGTG-3' (LOC420205-F)
5'-GGAACCAACAATCTTAAACAAACGTC-3' (LOC420205-R)
5'-ATGTGTAGGGAAAAGTGAAAAGCAGC-3' (LOC420206-F)
5'-AGTGAGTGTGAGTGATAGGTAGAAAG-3' (LOC420206-R)
5'-GAATTTTGAGCTTGAAGGACGGGAAC-3' (CGI -77-F)
- 9-
5'-AGATGCCTTGCTCCTGTTAAATTTTC-3' (CGI -77-R)
5'-GAGTGTGTTGCCTTCCTGTTAAACTG-3' (FLJ20530- F)
5'-CCCAGAGATCTTTAGCTGTCCTTCTG-3' (FLJ20530-F).
For LOC420205 and LOC420206, long and accurate PCR was applied using
the kit of TaKaRa LA Taq with GC buffer (Takara, Tokyo, Japan). The
condition of long and accurate polymerase chain reaction (PCR) was
following: 14 cycles at 94°C for 20 sec, 68°C for 10 min and 16 cycles at 94 °c
for 20 sec, 68°C for 10 min (+ 15 sec/cycle). For others, PCR was done for 30
cycles at 94°C for 30 sec, 60°C temperatures for 30 sec, 72°C for 1 min
with TaKaRa Ex Taq® Hot Start Version (Takara). The sequences were
analyzed by Chromas 1.45 in Griffith University (http://www.tec
hnelysium.com.au/chromas.htmD, BLAST search in NCBI GenBank
Database, and CLUSTALW of DDBJ homology search system
(http://www.ddbj.nig.ac.jp/search/clustalw-j.htmD. For substitution or
insertion-deletion GndeD events, new primers detecting the polymorphisms
were designed and applied to genotyping for the resource family.
Genotyping. The PCR was done for 30 cycles at 94°C for 30 sec, annealing
temperatures for 30 sec, 72°C for 1 min using TaKaRa Ex Taq® Hot Start
Version (Takara) and TaKaRa LA Taq® with GC buffer (Takara). The primer
sequences of informative gene markers and their annealing temperatures
were shown in Table 1 and 2. PCR products were subjected in electrophoresis
in either an ethidium bromide-added 1 % agarose gel or a 5 %
polyacrylamide gel. For the latter cases, SILVER SEQUENCETM DNA
- 10-
straining reagents (Promega, Madison, WI, USA) was used for DNA to be
visualized.
With the genotypes obtained, haplotypes were constructed manually,
by minimizing the number of recombinants and assuming no mutation of
marker alleles. The candidate region was defined on the basis of observed
ancestral recombination events.
·11·
Results & Discussion
The AMlocus has been mapped to chicken chromosome 2q by previous study
using resource family with 110 backcross offspring (Yoshizawa et ai., 2004).
The result indicated the candidate region located between IMPAl and
CBFA2Tl genes within 3.6 Mbp range on GGA2q. It was difficult to identify
the responsible gene using this family because of the limited number of
offspring. Therefore, we established new chicken muscular dystrophy
resource family with 487 F2 offspring in current study.
At the first step, total of 77 MS markers were developed usmg
information of chicken genome sequences and tested their polymorphisms on
the parents of new resource family. Twenty-five MS markers of them were
polymorphic between the parents and 14 markers were possible to use for
haplotype analysis. In addition, polymorphism searches for two functional
genes, CGI-77 and FLJ20530, were performed to add genetic markers in the
region where MS markers were absent. By sequencing analyses, two indel
polymorphisms were observed in intron regions of both two genes and
utilized as genetic markers. Using these 16 informative markers (Table I-I),
haplotype analysis was carried out for 240 F2 individuals. As a result, the
region between KCAM-006 and KCAM-010 (approximately 1.8 Mbp) was
suggested as AM candidate region and included 21 functional genes.
At second step, further analysis was carried out to reduce the
- 12-
"""" CI:l
Table 1-1 Sixteen informative primer sets used in first step
Figure 1-2 The abnormal muscle (AM) candidate region and locations of genetic markers used in this study. The region between IMPA 1 and CBFA2T1 is the AM candidate region by Yoshizawa et a/. (2004). The upper figure shows the region of 16 informative gene markers on first step. The lower figure shows reduced AM candidate region by 11 informative gene markers on the second step. The genes underlined are the AM candidate genes.
Figure 1-3 The haplotype analysis showing the phenotype of abnormal muscle (AM) chickens that are AM: AM homozygous. N, normal homozygous; and X, heterozygous chickens. (A, 8) show the alleles assumed from a NH-413 male and (C, D) the alleles assumed from a GSP female.
- 15 -
1699
x
B
A D
Table 1-3 All F2 individuals'haplotypes on the region between KCAM-104 and KCAM-106
Haplotype
AlB AID CID
Number AM 97 o o
Phenotype x o
259 o
N o o
131 Phenotype of abnormal muscle (AM) chickens; AM, AM homozygous chicken; N, normal homozygous chicken; X, heterozygous chicken. A and B show the alleles from a NH-413 male and C and 0 the alleles from a GSP female.
dystrophy Ie and its allelic limb-girdle muscular dystrophy 21 (Balci et ai.,
2005; Muntoni et ai., 2002), congenital muscular dystrophy ID (Matsumoto
et ai., 2005) and hereditary inclusion body myopathy (Huizing et ai., 2004),
and now they are termed "a -dystroglycanopathy." The
a-dystroglycanopathies are assumed to have common pathological
mechanism that a mutant glycosyltransferase for each disease glycosylates
a -dystroglycan insufficiency, and this imperfect glycosylation disrupts the
connection between a-dystroglycan and its ligands, and triggers onset of the
disease.
This study indicated that seven expressed functional genes are now
strong candidates as a responsible gene of chicken muscular dystrophy.
Although a kind of glycosyltransferase gene had been expected as a
responsible gene for this disease according to a previous study (Saito et ai.,
2005), there are no known glycosyltransferases within reduced candidate
region. Of seven candidates, functions of the two genes, LOC420214 and
LOC428367, have not yet been understood well, but these are unlikely to be
. 16·
glycosyltransferases according to prevIOUS studies (Zhang et a1., 1997;
Tomsig & Creutz, 2002). Functions of LOC420211 and LOC420213 are
utterly unknown. ATP6VOD2is less likely to be the causative gene since the
expression of this gene is severely limited in kidney, osteoclast and lung, and
the expression in muscle was not observed (Smith et a1., 2002).
One of the most likely candidates is MMP16which belongs to Matrix
metalloproteinase (MMP) family. Lattanzi et a1. (2000) reported new model
of human muscular dystrophy that an abnormal MMP breaks a basal lamina
and triggers the disorder. It has been proposed that the processing against
fl'dystroglycan by MMP be disrupted and influence the interactions between
the extracellular matrix and the cytomembrane (Yamada et a1., 2001).
The WWPl is also one of the most likely candidates and belongs to a
member ofE3s (Wood et a1., 1998). Recent studies have reported that some of
ubiquitin ligases (E3) are associated with muscular dystrophies
(Pallares-Trujillo et a1., 1997; Acharyya et ai., 2005; Kudryashova et a1.,
2005). The members of E3s are divided into some classes structurally:
HECT-type E3s, RING-type E3s, PHD-type E3s, U-box-type E3s and others
(Pickarta & Eddins, 2004). The WWPl is classified into HECT-type E3s,
although E3s assumed to have a relationship with muscular dystrophies,
such as Murfl, Trim32 and Cbi-l, are all into RING-type E3s.
In this chapter, we could reduce the candidate regIOn to
- 17-
approximately 1 Mbp on GGA2q and indicate seven functional genes as the
most likely AM candidates. In order to identify a responsible gene of the
chicken muscular dystrophy out of seven candidates, further genetic and
biochemical research was required.
·18·
CHAPTER II
The ubiquitin ligase gene (WWP1) is responsible for
the chicken muscular dystrophy
Introduction
In chapter I, the AM candidate region was narrowed down to approximately
1 Mbp on chicken chromosome 2q. Seven functional genes, A TP6VOD2,
LOC420211, WWP1, LOC420213, LOC420214, LOC428367 and MMPl6, in
this region were the candidate genes for chicken muscular dystrophy
(Matsumoto et aI., 2007). None of the candidates have been determined to be
genes for other muscular dystrophies so far.
Aberrant glycosylation of a-dystroglycan in muscular dystrophic
chicken had been reported (Saito et aI., 2005), suggesting a kind of
glycosyltransferase gene might be a responsible gene for this disease. But,
quite intriguingly, there are no known glycosyltransferases within reduced
candidate region. The functions of four genes among the candidates are not
fully understood, which still leaves the possibility that they belong to some
forms of glycosyltransferases or that they interact with certain
glycosyltransferase for a-dystroglycan. At the same time, it occurred that
aberrant glycosylation of a-dystroglycan in the disease is just a secondary
effect of the pathological change.
- 19·
Accumulating data proposed that MMP16 or WWPl be the most
likely responsible gene for chicken muscular dystrophy, since similar
proteins as them have been reported as the causative of muscular
dystrophies (Lattanzi et a1., 2000; Kudryashova et a1., 2005). However, the
data directly proving one of them is the responsible for chicken muscular
dystrophy is lacking. This chapter attempted to identify the responsible gene
and mutation by sequence comparison and expression analysis in normal
and dystrophic chickens. The mutation was identified in WWP1, detected
only in dystrophic chickens within several tetrapods, suggesting WWP 1 is
the responsible for chicken muscular dystrophy.
- 20-
Materials & Methods
Genetic resource. For sequencing and expression analyses, NH-413 and OPN
strains were tested as dystrophic chickens (Kondo et a1., 1982), and White
Leghorn-F (WL-F) and GSP strains were as normal chickens.
Genomic DNA samples were obtained from total 111 individuals
consisting of 16 strains: 20 White Cornishes, 19 White Plymouth Rocks, 20
broilers (White Cornish x White Plymouth Rock), 20 White Leghorns, and 20
Brown Leghorns, maintained at the Tokushima Forest and Forestry
Research Institute, Japan, an Onagadori, an Ukokkei, a White Leghorn, a
White Plymouth Rock, a Black Minorca, a Fayoumi, a Vietnam native, a
Laos native, an Echigo native, a Tosa native, and two Red Jungle fowls from
Laos and Indonesia. Muscle cDNA libraries of pigeon, alligator, lizard, turtle
and frog, and liver cDNA library of snake, were used to determine partial
sequences of WWPl gene in this study (Mannen et a1., 1996, 1997).
Sequencing for seven candidate genes. Pectoral muscle cDNAs were used as
template DNA to sequence for WWP1, LOC420213, LOC420214 and MMPl6,
while genomic DNA for A TP6VOD2, LOC420211 and LOC428367. In using
cDNA as template, primers were designed to amplify each coding region, and
in using genomic DNA, to amplify each exon. The primer sequences and the
PCR conditions were summarized in Table II-la. Sequencing was performed
with BigDye® Terminator v3.1 Cycle Sequencing Kit and ABI PRISM® 3100
Genetic Analyzer (Applied Biosystems, Foster City, CA). For polymerase
- 21-
chain reaction (PCR), TaKaRa Ex Taq® Hot Start Version or TaKaRa LA
Taq® with GC buffer (Takara, Tokyo, Japan) was used as polymerase.
Genotyping chickens for WfVPl. Genotyping of 111 individuals from 16
strains for WWPl was carried out by PCR-restriction fragment length
polymorphism (RFLP) method. Primer set was established to amplify the
region with a single nucleotide polymorphism (SNP) on WWPl. One of them
was mismatched primer and designed so that the chicken with wild-type
WWPl yielded three DNA fragments and the one with mutated WWPl
yielded two through digestion by a MboI restricted enzyme (Fermentas
International Inc., Ontario, Canada). The primer sequences and the PCR
condition applied were shown in Table II -lb.
WWPl homology search among species_ To obtain WWPl partial sequences
of pigeon, snake, alligator, lizard, turtle and frog, degenerate primers were
designed based on those of chicken (NM_OOI012554), human (NM_007013),
mouse (NM_177327) and rat (NM_OOI024757) published in NCB!. To
amplify each WWPl of snake, alligator and pigeon, WWPld-Fl and
WWPld-R were used as primers, and WWPld-F2 and WWPld-R were used
for other species. The primer sequences and the PCR condition were shown
in Table II-Ie. The amplified bands corresponding to WWPl cDNAs were
extracted from agarose gel and purified with Ultra-Clean™ 15 DNA
Purification Kit (MO BIO Laboratories Inc., Carlsbad, CA). The purified
cDNA fragments were ligated into pGEM-T vectors (Promega Corporation,
- 22-
Madison, wI) and transformed into Escherichia coli JM109 competent cells.
To confirm the sequence integrity, eight independent clones were sequenced.
Expression analysis. For Northern blot analysis, pectorals mRNAs of a
NH -413 and a WL-F strain female were used. 2 pg of mRNAs were resolved
by 1.2 % agarose gel electrophoresis in the presence of formaldehyde and
blotted on to Hybond-N+ membrane (GE Healthcare Bio-Sciences AB,
Uppsala, Sweden). The mRNAs were visualized using digoxigenin (DIG)
reagents (Roche Diagnostics, Basel, Switzerland). The DIG-labeled DNA
probes were prepared by PCR with DIG-dUTP using pectoral cDNA. The
primers used in this procedure were shown in Table II old.
For genes whose expression was not detected by Northern blotting,
reverse transcription (RT)-PCR method was applied with same condition in
Table H-ld.
- 23-
Table 11-1 Sequences of primers and peR conditions for sequencing, genotyping and expression analysis
Name Primer1) Sequence (5'-3')
(a) Primers used for sequencing to produce template DNA ATP6VOD2 ATP6VOD2s-F1 TCTAAAACTATGTGAGCCTGGAGTAG
lEach columns upper primers are forward primers and lower ones reverse primers.
2Each column shows PCR method (polymerase): annealing temperature, extension time, cyclic number applied by each primer sets. Normal PCR was performed as the following: cyclic number at 94 OC for 30 s, annealing temperature for 30 s, 72 OC for extension time, SH PCR: cyclic number at 98 OC for 10 s, 68 OC for extension time, and DG PCR: 3 cycles at 95 OC for 30 s, 50 OC for 30 s, 72 OC for 30 sand 32 cycles at 95 OC for 30 s, 50 OC for 30 s, 72 OC for 30 s.
3To analyze their expressions, RT-PCR was performed.
4Expression ofthis gene was not confirmed.
·26·
Results
Sequence comparison of seven candidate genes in normal and dystrophic
chickens revealed three synonymous mutations and one missense mutation
specific to the phenotype of muscular dystrophy. Two of the synonymous
mutations were detected in LOC420214 (C660T and CI009A), and the other
in LOC428367 (C954T). The WWPl missense mutation (G1321A) caused
amino acid substitution from arginine to glutamine, leading to the molecular
alteration from a basic side chain to an uncharged polar side chain. We
focused further studies on this missense mutation since it was predicted to
influence the function of the WWPI protein.
The domain structure of human WWPI was determined previously
(Flasza et ai., 2002). It consists of three types of unique domains: one C2
domain, four WW domains and one HECT domain. The amino acid
sequences of human and chicken WWPI share 83% identity, suggesting that
the deduced structure of chicken WWPI was quite similar to that of human's.
According to NCBI database, chicken WWPI possesses one C2 domain, three
WW domains and one HECT domain. The detected mutation lay between
WW domains 1 and 2 (Fig. II-I).
In order to exclude the possibility of strain specificity of WWPl
mutation, this SNP was genotyped in OPN strain to determine if the WWPl
gene has this mutation in any strain with a different genetic background.
- 27·
R441Q
1 y
ww Figure 11-1 The domain structure of chicken WWP1 and the site of missense mutation. Chicken VVWP1 protein is composed of 922 amino acids. Here are shown VVWP1 functional domains: C2 domain, three VVW domains and HECT domain. C in HECT domain indicates an active cysteine residue. The arrow indicates the site of missense mutation detected in this study. VVW domains bind proline-rich region.
PCR-RFLP analysis indicated that OPN dystrophic chickens have the
identical mutation in the WWPl gene. Genotyping was also applied to 111
normal birds from 16 strains with genetically varied backgrounds. None of
the normal birds exhibited this type of substitution. The region of WWPl
including the SNP specific to chicken muscular dystrophy was highly
conserved among normal birds. Additional homology research among
tetrapods was conducted to estimate amino acid conservation in this region.
The sequences of chicken, human, chimpanzee (XP _519843), monkey
(XP _001083173), mouse, rat, dog (XP _535119) and cattle (NP _001032540)
were available in the NCBI database. To obtain further information from
other tetrapod species, we sequenced partial WWPl gene of pigeon, snake,
alligator, lizard, turtle and frog (DDBJ Accession Nos. AB385863 to
AB385868). The amino acid sequence around the region was highly
conserved among these tetrapods (Fig. II-2), suggesting that the region was
· ............ O ............................................. .
...................... A ..... Y ........ K .... T ................ .
...................... A ..... Y ........ K .... T ................ .
...................... A ..... Y ........ K .... T ................ .
...................... A ..... Y ........ K .... T ................ .
...................... A ..... Y ........ K .... T ................ .
...................... A ..... Y ........ K .... T ................ .
AD .................... A ..... Y ........ K .... T ................ .
· ..................... T .................................... .
...................... T .................................... .
· ..................... T ................... T ................ .
Figure 11-2 Homology study ofWWP1 among tetrapods. Chicken Nand A show amino acid sequences of normal and dystrophic chickens, respectively. R441Q WWP1 mutation is specific to chicken muscular dystrophy, and the amino acid sequence on this region is highly conserved among tetrapods. Dots indicate the same amino acids with above sequences. Asterisk indicates the residue that the substitution was detected in dystrophic chicken. (1) Amino acid sequence published in NCBI.
important for the function ofWWPl.
Expressions of candidate genes were analyzed by Northern blotting.
As depicted in Fig. n·3a, the expression of WWP1, LOC420213 and
LOC420214 in the pectoral muscle of both genotypes derived from both
·29·
a WWPc At N LOC4202t3 LOC420214 CAN CAN
GAPDH CAN
b LOC420211
25 30 35(CXC1es)
A NAN A N
-- .... -MM~
A N 30
A N 35(cyc1e5)
A N
... --
G~
A N 30
A N
Figure 11-3 Expression analysis of candidate genes. Extreme alteration of expression level was not observed in any genes. (a) Northern blotting analysis toward ATP6VOD2, LOC420211, WWP1, LOC420213, LOC420214, MMP16 using chicken pectoral mRNA (2 fJ. g). In any of these genes, there was no difference in size of mRNA between affected and normal birds. The C indicates loaded PCR product which has the same sequence as probe, the A indicates mRNA of dystrophic chickens and the N indicates mRNA of normal chickens, respectively. The arrows indicate the band detected. (b) RT-PCR analysis to'v'Vard ATP6VOD2, LOC420211, MMP16 using chicken pectoral cDNA.
normal and dystrophic chicken could be detected by Northern blotting. There
was no difference in size of mRNA between affected and normal birds in any
of these genes. LOC420213was highly expressed in affected individuals. Two
bands were detected in WWPl and LOC420214. WWPl was expressed
slightly higher in normal than in affected chickens, while a slightly higher
level of expression was exhibited for LOC420214 from affected chickens.
The expression of other genes (ATP6VOD2, MMPl6, LOC420211 and
LOC42836'/) was not detected by Northern blotting. RT-peR analysis in
dystrophic chickens (Fig. II-3b) revealed higher expression of A TP6 VOD2 and
- 30-
35(cycle5)
A N
MMP16 than in normal birds. The pectoral muscles from both genotypes
expressed LOC420211 to identical level. The expression of LOC428367 was
not confirmed in either genotype.
·31 -
Discussion
In chapter I, we narrowed down the AM candidate region to approximately 1
Mbp on GGA2q. Seven functional genes, A TP6VOD2, LOC420211, WWP1,
LOC420213, LOC420214, LOC428367 and MMPl6, in this region were the
candidate genes for chicken muscular dystrophy (Matsumoto et ai., 2007),
but none of them have been determined to be genes for other muscular
dystrophies so far. In this study, sequence comparison of normal and
dystrophic chickens was conducted to detect a mutation responsible for the
disease.
We detected a mutation site specific for the AM phenotype in
G 1321A of the WWPl gene that caused amino acid replacement, from
arginine with a basic side chain (a basic amino acid) to glutamine with an
uncharged polar side chain (a neutral amino acid), which can affect the
function of the WWPI protein. WWPl mutated in the coding region of the
protein, providing the most likely candidate responsible for causing this
disease. This type of mutation was only observed in dystrophic chickens. No
mutation was detected in any normal chickens analyzed. Furthermore, the
amino acid sequence on this region is highly conserved among tetrapods (Fig.
II·2). The region is thus probably critical for the function of WWPl.
The expression patterns of the candidate genes were analyzed in Fig.
II-3. WWP1, LOC420213, A TP6VOD2 and MMP16 exhibited some difference
- 32-
III expreSSlOn level between normal and dystrophic individuals, none of
which were drastic alterations. No difference in mRNA size was observed in
any gene. Since no extreme alteration of expression level and abnormal
splicing were observed, the onset of chicken muscular dystrophy might not
be attributed to aberrant expression of these genes.
The WW domain containing E3 ubiquitin protein ligase 1 (WWPl),
the most likely AM candidate gene found by this study, is classified as a
ubiquitin ligase (E3) that plays an important role in ubiquitin-proteasome
pathway (UPP). Ubiquitination, addition of ubiquitin (Ub) chains to a target
protein, is one of the most common forms of posttranslational modification,
and it controls important aspects of cell functions (Passmore et a1., 2004;
Pickart et a1., 2004). In UPP, at least three types of enzymes are required,
namely, El Ub-activating enzyme, E2 Ub-conjugating enzyme and E3
(Hershko et a1., 1983). E3 recognizes and catalyzes Ub conjugation to specific
protein substrates (Liu, 2004). E3s are structurally divided into several
classes: HECT-type E3s, RING-type E3s and others (Pickarta & Eddins,
2004).
Accumulating data indicate that some E3s are related to muscular
dystrophies (Acharyya et a1., 2005; Kudryashova et a1., 2005; Trujillo et a1.,
1997). E3s that are assumed to be related to muscular dystrophies, such as
Trim32 (Jackson et a1., 2000), are all classified as RING-type E3s, while
WWPI is classified as a HECT-type E3. RING-type E3s facilitate
·33·
ubiquitination indirectly, while HECT-type E3s transfer Db directly to
substrates bound to a non-catalytic domain of themselves (Jackson et a1.,
2000). Though they are structurally and mechanically distinct, their basic
role in DPP is common. Therefore, it is possible for some abnormal
HECT-type E3s to cause muscular dystrophies. Actually, it has been reported
that E6AP, another HECT-type E3, contributes to a severe neurological
disorder, Angelman syndrome (Kishino et a1., 1997; Cooper et a1., 2004)
which exhibits muscular hypotonia (Gillessen-Kaesbach et a1., 1999).
WWPl is expressed ubiquitously but more strongly in liver, bone
marrow, testis, and muscle (Flasza et a1., 2002). Flasza et a1. (2002) also
mentioned at least six splice variants of WWPl. Two of the six products are
commonly observed among multiple tissues, and their expression levels are
higher than the others. The bands detected by Northern blotting (Fig. II-3a)
may correspond to these two major products, but little is known about the
function of these transcript variants.
E6AP is known as the protein responsible for Angelman syndrome
(Kishino et a1., 1997). Cooper et al. reported that two mutations lying in the
non-catalytic amino-terminal portion of E6AP cause Angelman syndrome
(Cooper et a1., 2004): one of them affects E6AP enzymatic activity, and the
other may disturb substrate binding, subcellular localization or protein
stability. The mutation of WWPI detected in this study is located between
WW domains 1 and 2 (Fig. II-I). Though the mutation is outside all domains,
- 34-
it is predicted that the region near the mutation is functionally important
because of its high homology (Fig. 1I-2). The R441Q mutation probably
influences the WWP1 function like Angelman syndrome causative mutations,
so that it triggers the onset. Since the mutation lies between two domains
both of which recognize target proteins (Sudol et a1., 1995), it could alter the
conformation of WWP1 (Verdecia et a1., 2003), the physical relationship
among domains and the preference for substrates.
Aberrant regulation of membrane protein may lead to the onset of
chicken muscular dystrophy. The HECT-type E3s with WW domains
generally regulate membrane proteins (Chen & Matesic, 2007). Though
much remains unclear about WWP1's substrates, it has been demonstrated
that WWP1 could interact with B-dystroglycan, an important muscle protein
consisting of membrane (Pirozzi et a1., 1997). Abnormal glycosylation of
a-dystroglycan in chicken muscular dystrophy was also reported (Saito et a1.,
2005). Some E3s are known to recognize sugar chain (Lederkremer &
Glickman, 2005; Yoshida et a1., 2002, 2003), leading to the hypothesis that
WWP1 might be able to recognize the sugar chain of a-dystroglycan and
regulate the glycosylated molecules, and that insufficiently glycosylated
a-dystroglycan, which originally requires degradation, accumulates and
causes the disease because altered WWP1 can not recognize and degrade it.
However, all known E3s to recognize sugar chain are divided into F-box-type
E3s, therefore it is doubtful that WWP1 should degrade insufficiently
glycosylated a-dystroglycan.
- 35-
The study in this chapter identified R441Q WWPI mutation as being
specific to chicken muscular dystrophy. This is the most likely candidate
mutation responsible for chicken muscular dystrophy because the region
nearby is highly conserved among species and no drastic alteration of
expression patterns of the candidate genes was observed. In order to clarify
the mechanism by which mutated WWPI triggers the onset of chicken
muscular dystrophy, further biochemical research is required .
. 36·
CHAPTER III
Expression pattern of WWP1
in muscular dystrophic and normal chickens
Introduction
The WW domain containing E3 ubiquitin protein ligase 1 (WWP1) IS
classified into a ubiquitin ligase (E3) which plays an important role III
ubiquitin-proteasome pathway (UPp) to degrade unneeded or damaged
proteins (Scheffner & Staub, 2007). E3 recognizes and catalyzes ubiquitin
(Ub) conjugation to specific protein substrates (Liu, 2004). Comparative
genome analysis reveals few genes encoding E 1, tens of E2 encoding genes
and hundreds of E3 encoding genes (Semple et ai., 2003).
The WWPl gene is classified into a HECT (homologous to the E6-AP
~arboxyl terminus)-type E3 which possesses one C2 domain, multiple WW
domains and one HECT domain (Pirozzi et ai., 1997; Flasza et ai., 2002). The
C2 domain binds to the cellular membranes in a Ca2+-dependent manner
(Plant et ai., 1997) and mediates interactions with other proteins (Plant et
ai., 2000; von Poser et ai., 2000; Augustine, 2001). The WW domain has two
conserved tryptophan residues and binds proline-rich region (Sudol et ai.,
1985). HECT domain, similar to E2s structurally, has a cysteine residue as
an active center that transfers the activated Ub from E2 onto first itself, and
then onto its substrates (Jackson et ai., 2000).
- 37-
The WWPl gene was identified as a candidate responsible for the
chicken muscular dystrophy by the linkage analysis in chapter I (Matsumoto
et al., 2007) and the sequence comparison between normal and dystrophic
chickens in chapter II (Matsumoto et al., 2008). The R441Q missense
mutation was found in the WWPl gene to cause the phenotype of muscular
dystrophy.
The WWPls of human (Flasza et al., 2002; Komuro et al., 2004),
mouse (Dallas et al., 2006) and C. elegans (Huang et al., 2000) were
intensively studied and known that the WWP1· gene is expressed
ubiquitously, but strongly in liver, bone marrow, testis and skeletal muscles
(Flasza et al., 2002; Komuro et al., 2004). In chicken, however, the WWPl
expression has not been studied. The expression analysis of WWP 1 gene is
important since it was reported that altered expression of known responsible
gene could lead dystrophic phenotype (Smythe & Rando, 2006).
In this chapter, the mRNA expression of WWP 1 in various skeletal
muscles and other tissues of normal and dystrophic chickens was analyzed
by Northern blotting and reverse transcription (RT)-peR analysis to know
the differences in the general expression pattern between these chickens.
- 38-
Materials & Methods
Chickens. A two-month-old dystrophic chicken (New Hampshire: NH-413)
and an age-matched normal chicken (White Leghorn: WL- F) were used in
this study. The New Hampshire (NH-413) strain is a homozygous dystrophic
line introduced from University of California, Davis to Japan in 1976 (Kondo
et ai., 1982). The disease in this strain is transmitted co-dominantly by a
single gene, but the phenotype is modified by other background genes
(Kikuchi et ai., 1981, 1987; Wilson et al., 1979). The White Leghorn (WL-F)
strain was established in 1970s, and maintained as closed colony in the
Nippon Institute of Biological Science in Yamanashi, Japan. This study was
carried out according to the guidelines of Animal Experimentation of Kobe
University.
Expression analysis_ For Northern blotting, mRNAs were isolated from M
pectoralis superficialis (PS), M tensor fascia lata (TFL), M biceps femoris
(BF), M triceps surae (TS), M peroneus longus (PL), heart (H), brain (B),
liver (L), kidney (K) and whole embryo (E) with PolyATtract mRNA Isolation
kit (Promega, Madison, WI, USA). The 2 llg ofmRNAs, which were measured
with NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies,
Wilmington, DE, USA), were resolved by 1.2 % agarose gel electrophoresis in
the presence of formaldehyde and blotted on to Hybon:d-N+ membrane (GE
Healthcare Bio-Sciences AB, Uppsala, Sweden). The mRNAs were visualized
using digoxigenin (DIG) reagents, and kits for non-radioactive nucleic acid
- 39-
labeling and detection system (Roche Diagnostics, Basel, Switzerland)
according to the procedure specified by the manufacturer excepting that the
washing was done with 4 x SCC 0.1 % SDS at room temperature for 10 min,
4 x SCC 0.1 % SDS at 40°C for 8 min and then 2 x SCC 0.1 % SDS at 40°C
for 8 min twice. The DIG-labeled DNA probes were prepared by PCR using
DIG-dUTP using pectorals cDNA sample of a WL-F strain female as a
template. The prImers applied In this procedure were
Figure 111-1 Expression of chicken WWP1 in various tissues. A WWP1 cDNA probe was used to detect v\-wP1 mRNA transcripts by Northern blotting using blots containing 2 f.J. g of mRNAs from chicken muscles or various other tissues. M. pectoralis superficialis (PS), M. tensor fascia lata (TFL) , M. biceps femoris (BF) , M. triceps surae (TS), M. peroneus longus (PL), heart (H), brain (B), liver (L), kidney (K) and embryo (E) were analyzed. A doublet band is detected at variable levels in all tissues. "A" indicates mRNAs from affected, dystrophic chickens. "N" indicates mRNAs from normal chickens. The numbers below the GAPDH bands represent the relative ratios of IrVvVP1IGAPDH.
dystrophic chicken was less than that of normal chicken (Fig. III-I). The PS
of chicken is a fast twitch muscle composed of two types of fast twitch fibers
(aW and BW). TFL, BF, TS and PL muscles from wing and leg are mixed
muscles co-existing fast twitch (aW and BW) with slow twitch fibers (BR) in a
mosaic pattern (Ashmore & Doerr, 1971a), except that the ALD and M
adductor magnus are composed of slow tonic fibers (ST) innervated multiply
(Ashmore et al. , 1978; Kikuchi et al., 1986). In chicken muscular dystrophy,
fast twitch fibers are initially and most severely affected, while slow twitch
and slow tonic muscles persist relatively harmless throughout the life span
(Ashmore & Doerr, 1971b; Barnard et al. , 1982) .
- 43-
a dystrophic PS ALD H
normal PS ALD H
35 cycles
40 cycles
a; E o c:
PS ALO
Figure 111-2 RT -peR detection of WWP1 gene and histological analysis for three representative muscle types. M. pectoralis superficia/is (PS), M. anterior latissimus dorsi (ALD) and heart (H) expressed WWP11ess in muscular dystrophic chicken, but only dystrophic PS was severely harmed. (a) Expression of WWP1 in PS, ALD and H was analyzed by RT-PCR method. PCR was performed for 35 or 40 cycles. (b) The PS, ALD and H of dystrophic (NH-413) and normal (WL-F) chickens were analyzed with HE staining. Vacuoles (arroVv'S) and fatty infiltration (asterisk) are observed in PS of dystrophic chickens. It is also remarkable that, in dystrophic PS, many muscle fibers have many nuclei in cytoplasm and vary widely in size. These pathological features are not observed in ALD and H of dystrophic chicken. Scale bar = 120 t1 m.
The WWP 1 expression in dystrophic BF, TS and PL showed a similar
downward trend as observed in dystrophic PS (Fig. III-I). These data
indicate that there might not be a causal relationship between the alteration
of WWPl expression level and the severity of muscular dystrophy, since not
only affected muscles but unaffected ones exhibited the same pattern.
Moreover, the alteration of WWPl expression level was observed in other
unaffected tissues, such as Band K (Fig. III-I), which reinforces our
hypothesis that the alteration of WWPl expression levels does not link
directly to the dystrophic phenotype.
To assess the genetic influence of mutant WWPl upon chicken
- 44-
H
muscular dystrophy, we examined WWPl gene expression and histological
changes in three distinct muscle types, PS as a fast twitch type, ALD as a
slow tonic type, and H as a different type of muscle. RT·PCR was applied to
this study since ALD was not enough quantity of mRNA for Northern
blotting. RT·PCR analysis indicated that WWPl gene was expressed in slow
tonicALD, not only in PS and H of both phenotypes (Fig. III-2a).
Figure III-2b shows HE stained sections of PS, ALD and H from
normal and dystrophic chicken. The dystrophic PS was severely affected,
while ALD and heart of dystrophic chicken remained relatively intact as
described in a previous study (Kikuchi et a1., 1981). The pathological findings
in dystrophic PS were characterized by the degenerating fibers with many
vacuoles in cytoplasm, the fatty infiltration into connective tissue, and the
proliferation of nuclei within muscle fibers with large variation in sizes.
However, no such lesions were observed in ALD and H from age-matched
dystrophic chickens (Fig. lII-2b).
The WWPl was expressed even in unaffected muscles and the
downward alteration of WWPl expression was observed commonly in almost
all dystrophic muscles examined (Figs. III-1, 2). The observation suggests
that the alteration of WWPl might not be the cause of the pathological
change in chicken muscular dystrophy. Hence, the mutation identified in
chapter II (Matsumoto et a1., 2008) might playa crucial role in leading the
onset of chicken muscular dystrophy. The detected mutation lay between
- 45-
WW domains, highly conserved region among tetrapods (Matsumoto et a1.,
2008), which has been predicted as substrate binding region (Pirozzi et a1.,
1997; Flasza et a1., 2002). This suggests that mutated WWP1 could not
recognize its substrates.
Many known substrates of the HECT-type E3s including WWP1 are
membrane proteins which aberrations contribute to oncogenesis (Chen &
Matesic, 2007). Though most of known WWP1's substrates have also been
reported as oncogenic proteins, not as muscle-related ones, one of known
proteins interacting with WWP1 is Notch (Flasza et a1., 2006). Kitamura et
a1. (2007) have reported that the ablation of Notch results in abnormality of
muscle differentiation, suggesting that WWP1 play a role in muscle
differentiation process. Moreover, B-dystroglycan, an important muscle
protein consisting of membrane, could undergo proteasomal regulation via
ubiquitinaion by WWP1 (Pirozzi et a1., 1997). The function of WWP1 in
muscle mechanism needs elucidating.
In this chapter, we analyzed the mRNA expression of WWPl in
various skeletal muscles and other tissues of normal and dystrophic chickens.
The results suggest that WWPl expression level lowered in dystrophic
phenotype is not directly related to the cause of disease in chicken muscular
dystrophy, whereas mutated WWPl does not function properly to cause the
onset of chicken muscular dystrophy.
- 46-
CHAPTER IV
Mutated WWP1 induces an aberrant expression of
myosin heavy chain gene in C2C12 skeletal muscle cells
Introduction
The WW domain containing E3 ubiquitin protein ligase 1 (WWP1) IS
classified into a ubiquitin ligase (E3) which plays an important role m
ubiquitin ·proteasome pathway (UPp) to degrade unneeded or damaged
proteins (Scheffner & Staub, 2007). The WWP1 and similar E3 ligases play
important roles in cancer development, bone remodeling and central nervous
system regeneration (Chen & Matesic, 2007; Glimcher et a1., 2007;
Bernassola et a1., 2008; Qin et a1., 2008).
In the previous chapters, we identified WWPl gene as the candidate
gene responsible for the chicken muscular dystrophy, and the R441Q
missense mutation in the WWPl gene was found to be the cause of muscular
dystrophic phenotype (Matsumoto et a1., 2007, 2008). It is generally known
that myosin is the principal protein of the contractile apparatus in muscle,
and myosin diversity is primarily produced by the different expression of
multiple isoforms of myosin heavy chain (MyHC) subunits which undergo
transition during development in a variety of muscle systems (Schiaffino &
Reggiani, 1994). The switching of adult phenotype in fast muscle is inhibited
in chicken muscular dystrophy, resulting in the continued expression of slow
·47·
twitch MyHC isoform in adult fast muscles (Bandman, 1985; Bandman &
Bennett, 1988; Kaprielian et a1., 1991; Tidyman et a1., 1997).
The WWPl gene is expressed strongly in skeletal muscles (Flasza et
a1., 2002; Komuro et a1., 2004; Matsumoto et a1., 2009), but the relationship
between WWP1 and MyHC proteins has not been examined. To investigate
the effects of the overexpression and the expression of mutated WWP 1 gene
on the MyHC genes expressions, we transfected the wild and mutated types
of WWPl gene into C2C12 cells which are myoblasts derived from mice
skeletal muscle. The expression of four muscle-differentiation markers,
Myogenin (Myog) , myogenic differentiation 1 (MyoD) and MyHCs, was
analyzed by real-time PCR. In addition, expression of caveolin-3 (cav-3J and
muscle specific ring finger protein 1 (MuRF1) was analyzed as muscular
dystrophic-phenotype-related genes.
- 48-
Materials & Methods
Cell culture. The C3H murine skeletal muscle cell line C2C12 (CRL-1722) was
commercially obtained from the American Type Culture Collection (ATCC) ,
VA, USA. Cells were cultured in the growth medium: Dulbecco's modified
Eagle's medium (DMEM) (Nissui, Tokyo, Japan) containing 0.2 % sodium
at 60°C, and 30 sec elongation at 72 °C. The last cycle was performed at 95
°C for 10 sec, 60°C for 30 sec and 72 °C for 60 sec. Reactions were run on a
TaKaRa PCR Thermal Cycler Dice® Real Time System (Takara) using cycling
parameters defined by the manufacturer.
Each assay included a standard curve for each gene with five serial
dilution points of a standard cDNA and a no-template control. Expression
levels for each gene were calculated as relative expression levels toward
GAPDH mRNA expression. For each gene, three individual samples were
prepared and reactions were repeated two times for each sample. Statistical
analysis was carried out using Statcel2 (oms-publishing, Tokyo, Japan).
Statistical analysis. Values in this paper represent means +/- S.D.
Differences between two groups were examined for statistical significance
- 52-
using Student's t test. A P value less than 0.05 denoted the presence of a
statistically significant difference.
- 53 .
Results & Discussion
We transfected the WWPl genes into C2C12 cells to analyze the expressions
of muscle-differentiation markers. To evaluate the vector capacity, the
WWPl expression of each group (pCAGGS-empty-: control, pCAGGS-WT-:
WT and pCAGGS-R436Q-transfected cells: R436Q) was analyzed by RT-PCR
using cells at day 0 (Fig. IV-Ia). Although WT and R436Q groups expressed
3.83 and 5.47 fold higher than control level, clear difference was not observed
in the myotube morphology and the proliferation rate of myoblasts (data not
shown).
FigUrE; IV-Ib shows the diachronic analysis of WWPl expression in
control cells during myogenic differentiation. Since in vitro myogenesis is
completed within a week (Sultan et a1., 2006), WWPl expressions analyzed
by real-time PCR were made at day 0,2,4,6 and 8. The WWPl gene seems
to be expressed stably in each time point, though upward trend was observed
in day 2 and 4.
Subsequently, we analyzed the expreSSIons of the
muscle-differentiation markers by real-time PCR to access the influence of
WWPl overexpression and WWPl with R436Q mutation on the gene
expressions of Myog, MyoD, MyHC Ia and MyHC lIb using C2C12 cells at day
6 (Fig. IV-2). The former two were analyzed as the markers for early stages
(Langlands et a1., 1997) and the latter two for later stages of muscle
- 54-
a
b G) I: 2.5 G) til ....
~ 2.0
'5 1.5 I:
'1 1.0 l5. )( G) 0.5 ~
I 0 2 4 6 8
Figure IV-1 Exogenous and endogenous expression of WWP1 gene in C2C12 cells at day O. (a) The vector capacity was assessed by RT-PCR method using GAPDH as internal standard. Each expression of WWP1 gene in WWP1-transfected cells 0NT and R436Q) was greater than empty vectortransfected cells (control). (b) Endogenous expression of the WWP1 gene in differentiating control cells. The WWP1 expression levels analyzed by realtime PCR were not significantly changed through day 0 to 8. Y-axis indicates relative expression level of WWP1 gene to the GAPDH gene expression. Bars indicate standard deviations.
differentiation (Silberstein et ai., 1986).
The expressions of Myogin WT and R436Q groups were 1.04 fold (±
0.02) higher and 0.64 fold (± 0.26) lower compared to the control group. The
R436Q group was significantly lower than other groups. These results
indicate that R436Q mutation in WWPl gene affects early stages of muscle
differentiation through the reduction of the Myog expression. There was no
clear difference in MyoD gene expression among three groups.
It was, however, interesting to indicate that the WWP 1 gene
overexpression influenced upon MyHC genes expression. The MyRes are
(day)
Q) c Q) Q) c 1.4 Ol 1.4 Q)
.i; Ol
i 1 a a 0 1.2 0
~1 :t 1.0 .... 0 ....
0 c 0.8 0 0 c .~ 0 0 0.6 .~ ~
~ 0 Q. ~ 0.4
~ Q) Q) 0 .~ 0.2
Q)
.~ iii Ri ! Gi control wr R436Q control WT R436Q ...
Q) Q)
c c ~ 4.0
Q)
b Ol
..!!! ~ 5.0
l> 3.5 l>
~ 3.0 ~ 4 .0
'0 2.5 -0 3.0 c 2.0 c 0 a 0
.~ 1.5
.~ 2.0
Q) Q) ... ... ~ 1.0 Q. a x Q) Q) 1.0 Q) 0.5 Q) b
.~ .~ ..... iii III Gi control wr R4360 Gi control wr R4360 ... ...
Figure IV-2 Muscle-differentiation markers in WWP1-transfected (WT and R436Q) and empty vector-transfected (control) C2C12 cells. Expression levels of muscle-differentiation markers (myogen in , MyoD, MyHC fa and MyHC lib) in differentiating day 6 C2C12 cells were analyzed by real-time PCR. The expression of Myogenin in R436Q group was significantly lO'Ner than other groups, while no significant difference was seen in the expression level of MyoD gene. HO'Never, the expression of fast MyHC fa in WWP1-transfected cells was significantly higher than in control cells, while that of slO'N MyHC lib was significantly lO'Nered in WWP1-transfected cells compared to control cells. The R436Q-transfected cells retained the high expression of both fast MyHC fa and slO'N MyHC lib isoforms compared to.control cells. Y-axis indicates relative expression level of each gene to the GAPDH gene expression. Bars indicate standard deviations. Different letters indicate significantly differences (p < 0.05) among column graphs.
among muscle proteins increasing during the course of myogenesls
(Silberstein et a1., 1986), and are divided into two classes, type I composed of
. 56'
fast twitch fibers and type II of slow twitch fibers (Larsson & Salviati, 1989).
The MyHC la expression levels in WT and R436Q groups increased 2.81 (±
0.95) and 2.18 (± 0.24) fold compared to control group, respectively.
The expression level of MyHC la gene in WT group was significantly
higher (2.81 fold ± 0.95) compared to control group, but MyHC lIb gene
expression was significantly lower (0.26 fold ± 0.04) in WT group, indicating
that WWP1 promotes to transform C2C12 cells into fast twitch characteristics.
However, R436Q-transfected cells persisted in the high expression of both
fast MyHC la and slow MyHC lIb isoforms compared to control cells,
suggesting retainment in slow and fast twitch isoforms characteristic.
Additionally, we analyzed the expressions of cav-3 and MuRFl genes
by real-time PCR as muscular dystrophic phenotype-associated genes (Fig.
IV-:3). The expression of cav-3 in both WWP1-transfected cells was
significantly higher (2.56 fold ± 0.16 in WT and 1.87 ± 0.43 in R436Q cells)
than in control cells. The caveolin-3 expression is known to increase by
pathological change of dystrophin-deficient type muscular dystrophy (Vaghy
et a1., 1998; Repetto et a1., 1999), suggesting WWP1-transfected cells might
be affected. However, since the amount of caveolin-3 protein is regulated by
posttranscriptional manner (Galbiati et a1., 2000a), further analysis is
required. Caveolin-3 is also known as the marker for later stages of muscle
differentiation (Schubert et a1., 2007), which suggests that WWP1 promotes
C2C12 cells to differentiate, while the differentiation is suppressed in R436Q
- 57-
Q) c Q) Q) c CI 8, "? b 3.0 .S ..-. b
j & 2.5
~ B '0 2.0 '0 c c 1 0 1.5 .2 a .~ a m ~ 1 ~ ~ ~ Q) Q) Q) 0.5 Q) .~ .~ ni ni li li control OE R436Q .... control OE R436Q ....
Figure IV-3 Dystrophic phenotype-associated genes in WWP1-transfected (WT and R436Q) and empty vector-transfected (control) C2C12 cells. Expression levels of muscular dystrophic phenotype-associated genes (caveolin-3 and MuRF1) in differentiating day 6 C2C12 cells v.tere analyzed by real-time PCR. The expression of caveolin-3 in both WWP1-transfected cells 0NT and R436Q) was significantly higher than in control cells. The expression of MuRF1 in R436Q group was significantly higher than other groups. Y-axis indicates relative expression level of each gene to the GAPDH gene expression. Bars indicate standard deviations. Different letters indicate significantly differences (p < 0.05) among column graphs.
due to the mutation in WWPl gene.
The expressions of the MuRFl in WT and R436Q groups were 1.53
fold (± 0.50) higher and 2.32 fold (± 0.28) higher compared to the control
group. The R436Q group was significantly higher than the other groups.
MuRFl belongs to E3s which degrade MyRCs and known as one of muscle
atrophy markers (Clarke et al., 2007). The data indicate that R436Q
mutation in WWPl gene causes the transfected cells muscle atrophy, though
clear difference was not observed in the myotube morphology and the
proliferation rate (data not shown) .
. 58'
One of known proteins interacting with WWPI is Notch (Flasza et
a1., 2006), whose ablation in skeletal muscle results in increased formation of
fast twitch fibers and altered fiber type distribution at the expense of slow
twitch fibers (Kitamura et a1., 2007). WWPI might control skeletal muscle
fiber types via the regulation of MyHC genes expression by Notch signaling.
The R436Q-transfected cell group showed the highest MyHC lIb expression
(2.80 ± 1.92 fold than control level) among three groups, suggesting that
C2C12 cells persisted in the slow twitch character. Taken together, the R436Q
mutation in WWPl gene seems to inhibit the normal fiber type
differentiation and to induce atrophy. The results suggest that WWPI plays
an important role in myoblasts' differentiating process.
- 59-
CHAPTER V
Overexpression of caveolin-3 protein is limited
in damaged muscle in chicken muscular dystrophy
Introduction
Caveolae are vesicular invaginations of plasma membrane which range
50-100 nm in size (Anderson, 1993). The inner surface of caveolae is coated
with scaffolding protein formed by members of the caveolin family
(caveolin-1, -2 and -3) (Thomas & Smart, 2008). Caveolin-1 and -2 are
coexpressed and form heterooligomers in nonmuscle cells (Scherer et ai.,
1997), whereas caveolin-3 is muscle specific and forms homooligomers in
muscle cells (Song et ai., 1996). Though caveolae are known to regulate
endocytosis, exocytosis, cholesterol homeostasis and signal transduction
(Thomas & Smart, 2008), the role of caveolae and cave olin proteins of the
etiology are not fully understood.
Caveolin-3 co-localizes with elements of the dystrophin complex
(Rahkila et ai., 2001) and is proposed to function with the dystrophin
complex at the cell surface (Song et ai., 1996; Crosbie et ai., 1998). A WW-like
domain in caveolin-3 can recognize and bind a PPXY motif of B-dystroglycan
(Sotgia et ai., 2000). While caveolin-3 is predominantly associated with the
sarcolemma of mature muscle, caveolin-3 in differentiating muscle cells
associates with the developing transverse tubule (T-tubule) system (Parton
- 60-
et a1., 1997; Ralston & Ploug, 1999). Caveolin-3-deficiency leads T-tubule
abnormality, suggesting that caveolin-3 may be involved in the organization
of the T-tubules but is not essential for their formation (Galbiati et a1., 2001).
Accumulating data suggest that caveolin-3 reqUIres an accurate
regulation. Caveolin-3 is known as the responsible gene for limb-girdle
muscular dystrophy type lC (LGMD-IC) and other muscle diseases (Dowling
et a1., 2008). LGMD-IC mutations in caveolin-3 gene result in a loss of
~90-95 % of caveolin-3 protein expression (Galbiati et a1., 2000a). Caveolin-3
deficient transgenic mice also exhibit muscle degeneration (Hagiwara et a1.,
2000). On the other hand, Duchenne muscular dystrophy (DMD) patients
show increase of caveolae and caveolin-3 overexpression (Repetto et a1.,
1999). The concordant phenomenon is observed in dystrophin-deficient mdx
mice (Vaghy et a1., 1998). Conversely, transgenic overexpression of caveolin-3
induces a DMD-like phenotype (Galbiati et a1., 2000b).
In chapter I-III, we identified WWPl (WW domain containing E3
ubiquitin protein ligase 1) gene as the responsible gene for chicken muscular
dystrophy (Matsumoto et a1., 2007, 2008, 2009). The WWPI is classified into
a ubiquitin ligase (E3) to degrade unneeded or damaged proteins (Chen &
Matesic, 2007), suggesting that muscular dystrophic phenotype might be
triggered by aberrant regulation of some WWPl's substrates. An electron
microscopic analysis revealed increase of caveolae in dystrophic chickens
(Costello & Shafiq, 1979). Although increased expression of caveolin-3
- 61-
induces a muscular dystrophic phenotype (Galbiati et ai., 2000b), its
underlying mechanism to cause the onset remains elucidated. Therefore,
molecular biological analysis of caveolae in chickens with muscular
dystrophy has been required.
In this chapter, we analyzed the expression pattern of caveolin-3 and
other caveolae-related proteins in dystrophic chickens. Our result revealed
the correlation between caveolin-3 expression and the severity of dystrophic
phenotype, and gave new insights into the effect of caveolin-3 overexpression
in muscle and the potential role of caveolae in causative role of muscular
dystrophies.
·62·
Materials & Methods
Chickens & Muscles. '!\vo month -old (2M) dystrophic chickens (New
Hampshire, line 413: NH-413) and age-matched normal chickens (White
Leghorn-F: WL- F) were analyzed. For diachronic study, embryonic 7 day-old
(E7), 14 day-old (E14), 1 day-old (lD), 1 week-old (lW) and 2-week old (2W)
chickens were added to them. In each group, more than three birds were
utilized for this study according to the guideline of Animal Experimentation
of Kobe University. The NH-413 strain was a homozygous dystrophic line
introduced from University of California Davis in 1976 (Kondo et a1., 1982).
The disease in this strain is transmitted co-dominantly by a single gene, but
the phenotype is modified by other backcross genes (Wilson et a1., 1979;
Kikuchi et a1., 1981, 1987). The WL-F strain was established around 1971,
and maintained in closed colonies in the Nippon Institute of Biological
Science in Yamanashi, Japan.
M pectoralis superficialis (PS), M anterior latissimus dorsi (ALD)
M anterior tibialis (AT), M soleus (S) and heart (H) were obtained from
chickens and immediately stored in liquid nitrogen. The PS is a fast twitch
muscle composed of two types of fast twitch fibers (aW and 8W) (Ashmore &
Doerr, 1971a). The ALD is composed of slow tonic fibers (ST) innervated
multiply (Ashmore et a1., 1978; Kikuchi et a1., 1986). The AT and S are mixed
muscles of both fibers (Maier, 1993; Matsuoka & Inoue, 2008). In chicken
muscular dystrophy, fast twitch fibers are initially and most severely
affected, while slow twitch and slow tonic muscle fibers persist relatively
- 63-
harmless throughout the life span (Ashmore & Doerr, 1971b; Barnard et a1.,
1982).
Histology. The PS, ALD and H were snap-frozen in cooled isopentane and
sectioned in a cryostat (Leica Microsystems Japan, Tokyo, Japan). For
pathological observation and evaluation, the sections were stained with
hematoxylin and eosin (HE) (Kikuchi et al., 1981).
Antibodies. Antibodies used for the primary incubation were: anti-B-Actin
(B-actinr-F) and 5'-TACCACAGGACTCCATACCCAAG-3' (B-actinr-R)
(NM_205518). The PCR was done for 31, 33 and 35 cycles at 94°C for 30 sec,
60 °c for 30 sec, 72°C for 30 sec using TaKaRa Ex Taq® Hot Start Version
(Takara Bio Inc., Tokyo, Japan). Quantitative analysis was performed as
shown above. For each gene, three individual samples were examined.
- 65-
Statistical analysis. Values in this paper represent means +/- S.D.
Differences between two groups were examined for statistical significance
using Student's t test. A P value less than 0.05 denoted the presence of a
statistically significant difference.
·66·
Results
The expreSSIOn pattern of caveolin-3 protein was assessed by Western
blotting with a specific caveolin-3 antibody in three distinct types (PS, ALD
and H) and two mixed types (AT and S) of muscles from normal and
muscular dystrophic chickens (Fig. V-la). The PS of dystrophic chickens
shows higher caveolin-3 expression than those of age-matched control
chickens. Same tendency as PS was observed in AT. Caveolin-3 expression of
ALD, Sand H was undetectable level in both groups, even though higher
amount of protein extracts were blotted onto the same membrane.
Figure V-lb shows histopathological changes in PS, ALD and H of
normal and dystrophic chickens. The dystrophic PS showed various
pathological lesions such as degenerating fibers with many vacuoles in
cytoplasm, the fatty infiltration into connective tissue, and the nucleic
proliferation within muscle fibers with large variation in sizes. However, the
ALD and H from dystrophic chickens did not exhibit such pathological
alterations. The overexpression of caveolin-3 was obvious only in affected
muscle in dystrophic chickens, which suggested that there is a close
correlation between caveolin-3 overexpression and dystrophic phenotype.
The overexpression of caveolin-3 observed in dystrophic PS was
noted only at the protein level. Although the expression of caveolin-3 was
much higher (7.12 ± 3.31 fold) in dystrophic PS than in normal one at the
- 67-
a PS
PS A
ALD H b N A N A N PS ALD
.... ~
~ - I ~ I caveoIln-3 A
* ~ - -I /i-a~n N
- l~~ -AT S
N A N A N
I caveolin-3
T .. Figure V-1 Expression of caveolin-3 and histopathological observation in three representative and two mixed types of muscles. M. pectoralis superficia/is (PS) expressed higher amount of caveolin-3 protein in muscular dystrophy chicken (A), while the expression in M. anterior latissimus dorsi (ALD) and heart (H) was undetectable as in normal chicken (N). The expression in M. anterior tibialis (AT), one of mixed muscles, was similar as in PS, while that in M. soleus (S), the other one, as in ALD. (a) Expression of caveolin-3 in PS, ALD, H (upper panel), AT and S (lower panel) was analyzed by Western blotting. (b) The PS, ALD and H of dystrophic and normal chickens were analyzed with HE staining. Vacuoles (arrows) and fatty infiltration (asterisk) were noted in PS of dystrophic chickens. It was also remarkable that, in dystrophic PS, many muscle fibers have many nuclei in cytoplasm and vary widely in size. These pathological features are not observed in ALD and H of dystrophic chicken. Scale bar = 120 .Lt m.
protein level (Fig. V-2a, C), our semi-quantitive RT-PCR analysis indicated
that dystrophic PS did not show such higher caveolin-3 expression at the
mRNA level (1.09 ± 0.11 fold) (Fig. V2b, c). The difference of caveolin-3
expression between at protein level and at mRNA expression suggested that
the overexpression of caveolin-3 in dystrophic PS is induced in a
Figure V-2 Enhanced expression of caveolin-3 at protein level, but not at mRNA level. The caveolin-3 expression in M. pectoralis superficialis (PS) was assessed at protein level (a) and mRNA level (b) . Significantly higher expression of caveolin-3 protein was detected in dystrophic PS (7.12 + 3.31 fold) , On the other hand, its mRNA expression was similar level between affected (A) and normal (N) PS (1 .09 + 0.11 fold) . (c) Quantification of caveolin-3 protein and mRNA expression was performed with applying the 13 -actin expression as an internal control. Values are represented as fold changes of the A expressions relative to N levels. Y-axis indicates relative expression level of caveolin-3 to the 13 -actin expression. Bars indicate standard deviations, Asterisk means without a common superscript letter differ (p < 0.05).
The possibility remained that the increase of caveolin-3 protein
might be the secondary event, rather than the primary one, of the
pathological change of chicken muscular dystrophy. To examine this
- 69-
C? c
2 0 > c ~~ _.2 o.s:::. c 0 0.2 .- .s:::. 0a. o 0 ~ a-a.1i) ~~ CD c > .-
Figure V-3 Diachronic study of caveolin-3 expression in dystrophic muscle. The expression of caveolin-3 protein in M. pectoralis superficialis (PS) was analyzed with embryonic 7 day-old (E7), 14 day-old (E14), 1 dayold (1 D), 1 week-old (1W), 2-week old (m) and 2-month old (2M) individuals. As early as 10 muscular dystrophic chickens expressed higher caveolin-3. (a) Expression of caveolin-3 in PS from dystrophic (A) and normal (N) chickens was analyzed by Western blotting. (b) Quantification of caveolin-3 expression was performed with applying the {3 -actin expression as an internal control. Values are represented as fold changes of the A expressions relative to N levels. Y-axis indicates relative expression level of caveolin-3 to the {3 -actin expression. Sequential line graph indicates average value of three individuals. Each X means actual measurement.
possibility, diachronic study with E7, E14, ID, lW, 2W and 2M chickens was
carried out (Fig. V-8a). Immunoblot analysis revealed that as early as 1
. 70·
day-old muscular dystrophic chickens expressed higher caveolin-3 protein
(Fig. V-iSh). The caveolin-3 expression levels in muscular dystrophic chickens
increased 1.09 (± 0.15) fold at E7, 1.04 (± 0.07) at E14, 4.33 (± 5.85) at ID,
1.57 (± 0.77) at lW, 7.51 (± 6.57) at 2W and 6.84 (± 3.18) at 2M compared to
control group, respectively.
Subsequently, we analyzed the expreSSIOn of other proteins
consisting caveolae such as BinI (bridging integrator 1) and PTRF
(polymerase I and transcript release factor) (Parton et ai., 2006; Thomas &
Smart, 2008). These two caveolae-related proteins showed a similar
expression pattern as in caveolin-3 (Fig. V~4a). Western blotting study
revealed increase of both proteins in fast twitch dystrophic PS muscles, but
not in slow twitch and tonic muscles. BinI of ALD and H from dystrophic and
normal chickens was undetectable level. Nor drastic difference of PTRF was
detected between dystrophic and control groups.
It was speculated that Akt (v-akt murine thymoma viral oncogene
homologs) signal is also influenced by the elevation of caveolin-3 expression
in dystrophic PS. The Akt signal was examined with the antibody detecting
phosphorylated Aktl (S473), Akt2 and Akt3 (the corresponding residues)
(Fig. V-4b). With like caveolin-3 and other caveolae-related proteins,
phosphorylated forms of Akt increased only in dystrophic PS, but not in ALD
and H and in muscles from normal chickens.
-71'
a PS A N
ALD A N
H A N
181n1
~======================~ IpTRF
~--~~==~============~
IJ -actin
b PS A N
ALD A N
H A N
~====================---======~I A~ 1p-AA1
~~==~~==============~ ..... S1_ .. +t_,,..~.., ... -~_-""_ I p-actin
Figure V-4 Enhanced expression of other caveolae-related proteins in dystrophic muscle. (a) The expression of BIN1 and PTRF in M. pectoralis superficialis (PS), M. anterior latissimus dorsi (ALD) and heart (H) was analyzed by Western blotting. The increase of both proteins was observed in dystrophic PS, not in the other types of muscles. BIN1 expression of ALD and H from dystrophic (Al and normal (N) chickens was not confirmed. In ALD and H, no drastic difference of PTRF was observed between dystrophic and control groups either. (b) The expression of Akt and its phospholilation in PS, ALD and H was analyzed by Western blotting . The phosphorylated forms of Akt were observed only in PS from muscular dystrophy chickens, though total Akt protein was detectable in all muscles examined.
·72·
Discussion
Marked elevation of caveolin-3 expression in dystrophic PS was confirmed by
Western blotting analysis with three distinct types of muscles (PS, ALD and
H) (Fig. V-1a). This change was undetectable in the other muscles from
dystrophic chickens. In addition, Western blotting analysis with two mixed
muscles (AT and S) revealed that dystrophic AT expressed higher caveolin-3
than normal counterparts' while caveolin-3 expression in S of both groups
was undetectable as in ALD. AT is a mixed muscle mainly composed of fast
twitch fibers (Maier, 1993) and S is one mainly composed of slow tonic fibers
(Matsuoka & Inoue, 2008), suggesting elevation of caveolin-3 expression
depends on muscle fiber types.
The difference was limited only in the affected muscle of dystrophic
chickens (Fig. V-1b), suggesting that there is a close relationship between the
expression of caveolin-3 and etiology of chicken muscular dystrophy. Since
increased number of caveolae in dystrophic chickens is notable before the
onset (Costello & Shafiq, 1979), the caveolin-3 protein should be expressed
primarily and intensively in this disease. Supporting this hypothesis, the
diachronic study of caveolin-3 expression revealed thatas early as 1 day-old
muscular dystrophic chickens expressed higher caveolin-3 protein (Fig. V-3b).
Since pathological changes in chicken muscular dystrophy can be observed
after 1 week from hatching (Nonaka, 1987b), elevation of caveolin-3 \
expression might play a leading role for the onset of chicken muscular
- 73-
dystrophy.
The comparison between caveolin-3 protein and mRNA expression
suggested that the overexpression of caveolin-3 in dystrophic PS might be a
posttranslational event (Fig. V-2). The LGMD-1C phenotype occurred
through the possible dysfunction of caveolin-3 due to the ubiquitination and
proteasomal degradation (Galbiati et ai., 2000a). Moreover, we previously
identified WWPl gene, a ubiquitin ligase, as the responsible gene for chicken
muscular dystrophy (Matsumoto et ai., 2007, 2008, 2009), and suggested the
mutant WWP1 protein could not ubiquitinate and degrade caveolin-3
properly.
Mutations of caveolin-3 are responsible for muscle diseases
including LGMD-1C, in any of which the caveolin-3 expression is mostly
vanished (Woodman et ai., 2004). Reduction of caveolin-3 has also been
reported in another type of muscular dystrophy, LGMD-2B (Walter et ai.,
2003) and may influence the muscle function negatively and lead to the
muscle degeneration (Hagiwara et ai., 2000). Likewise, forced expression of
caveolin-3 leads to severe muscle damage in mice by decreasing the
expression of dystrophin and B-dystroglycan (Galbiati et ai., 2000b). It was
reported that dystrophin-deficiency leads to the excess expression of
caveolin-3 in human and mouse with muscular dystrophy (Repetto et ai.,
1999; Vaghy et ai., 1998). Smythe & Rando (2006) reported that the effect
upon downstream by aberrant caveolin-3 expression differs depending on
either increased or decreased expression of caveolin-3.
The expression pattern of two other proteins consisting of caveolae,
BinI and PTRF, was similar to that of caveolin-3 and elevated only m
dystrophic PS (Fig. V-4a). It has been proposed that the vacuoles m
cytoplasm observed in degenerating fibers (Fig. V-lb) are derived from
T-tubules (Nonaka, 1987b), and that endocytosis from T-tubules is an early
and essential pathological phenomenon in chicken muscular dystrophy
(Libelius et aI., 1979). Enhanced endocytosis in dystrophic muscle may be
related to the overexpression of BinI which can induce apoptosis via c-Myc
(Galderisi et aI., 1999; DuHadaway et aI., 2001). BinI is known to be
involved in T-tubule biogenesis (Parton et aI., 1997; Lee et aI., 2002) and
mutations in Bini gene are responsible for autosomal recessive
centronuclear myopathy (Nicot et aI., 2007). Although various studies proved
that PTRF is required for caveolae organization and function (Vinten et aI.,
2005; Hill et aI., 2008; Liu & Pilch, 2008), little is known about this molecule.
Total Akt protein was detectable in all muscles examined, but only
dystrophic PS exhibited the activation of Akt (Fig. V-4b). The fact that the
Akt signal is one of the downstreams of caveolin-3 (Fanzani et aI., 2007)
suggests that the Akt activation observed in dystrophic PS was induced by
the elevated caveolin-3 expression, as shown in previous studies (Fanzani et
aI., 2007; Kim et aI., 2008) and in the various muscular dystrophies (Lai et
aI., 2004; Peter & Crosbie, 2006).
- 75-
The study in this chapter indicated 1) caveolin-3 was expressed
higher in damaged muscle of dystrophic chickens, 2) the amount of
caveolin-3 protein was controlled in a posttranslational fashion and 3) the
expression of other caveolae-related proteins showed· similar tendency as
that of caveolin-3_ The overexpression of caveolin-3 protein seems to be
involved in the causative process of chicken muscular dystrophy. The
relationship between the overexpression of caveolin-3 protein and dystrophic
phenotype, and the interaction between WWP1 and caveolin-3 should be
clarified in the future.
- 76-
GENERAL DISCUSSION
Muscular dystrophies, the group of inherited diseases with progressIve
weakness and degeneration of skeletal muscle, are mainly caused by
abnormalities and/or defects of muscle proteins to compose of the linkage
between sarcolemma and basal lamina (Imamura et a1., 2000). Many animal
models corresponding to this type of muscular dystrophies have been
established and utilized intensively all over the world, so that the
understanding of this field has became wider and deeper for these years.
However, many of the diseases need elucidating the responsible genes and
mechanisms to trigger the onsets (Terri & Kunkel, 2000; Lisi & Cohn, 2007).
Muscular dystrophic chicken with abnormal muscle (AAd) is one of
the oldest animal models for the diseases (Asmundson & Julian, 1956). Any
known causative genes are unlikely to cause muscular dystrophic phenotype
in this model animal, according to our previous studies applying linkage
analysis (Lee et a1., 2002; Yoshizawa et a1., 2004), and the expressions of
known causative proteins are not disturbed in this model (Saito et a1., 2005).
Hence, it has been expected that chicken muscular dystrophy is the one with
a different mechanism from known muscular dystrophies. From this
perspective, more precise linkage analysis was carries out to narrow down
the AM candidate region and each candidate gene was examined in chapter
I -III. Subsequent chapters focused on the function of the AM responsible
gene.
-77 -
In chapter I, a resource family for AM gene was established with 487
F2 individuals and 22 gene markers, including micro satellite and
insertion-deletion markers, were developed. The haplotypes were analyzed
with these markers for the candidate region of GGA2q described m a
previous study (Yoshizawa et ai., 2004). The candidate region was
successfully narrowed down to approximately 1 Mbp. The region included
seven functional genes predicted as the most likely AM candidates.
Sequence comparison and gene expression analysis to elucidate the
responsible gene was performed in chapter II. One missense mutation was
detected in AM candidate genes, while no remarkable alteration of
expression patterns was observed. The mutation was identified in WWP1,
detected only in dystrophic chickens within several tetrapods. These results
suggested WWPl (WW domain containing E3 ubiquitin protein ligase 1) is
responsible for chicken muscular dystrophy.
In chapter III, we analyzed the WWPl expression in various muscles
and tissues of normal chickens, and compared with those from muscular
dystrophic chickens. Two mRNA isoforms were detected in all tissues
examined and revealed almost equal expression level. The WWPl expression
of dystrophic chickens was decreased in almost all skeletal muscles including
unaffected muscles. These data indicate that there might not be a causal
relationship between the alteration of WWPl expression level and the
. 78-
severity of muscular dystrophy, and that mutated WWPl does not function
normally to cause the onset of chicken muscular dystrophy.
Despite of intensive studies on its oncogenic characters, the role of
WWPl, an enzyme to degrade unneeded or damaged proteins, to muscular
diseases has not yet been fully understood. Since it is generally known that
the switching of myosin heavy chain (MyHC) isoforms from neonatal isoform
to adult one is inhibited in chicken muscular dystrophy, we transfected
either of wild and mutated types of WWPl gene into C2C12 cells to monitor
the expression pattern of muscle-differentiation markers including MyHCs
by real-time PCR in chapter IV. Excessive WWP 1 expression enhanced the
expression of MyHC Ia gene but lowered the expression of MyHC lIb gene.
On the other hand, mutated WWPl gene transfected into myoblasts was
distinct from these cases in that MyHC gene or genes expression inhibited
the normal myoblast differentiation. These data suggest that WWPl
promotes myoblast differentiation from embryonic into fast twitch phase
while mutation in WWPl results to retain slow and fast twitch isoforms
characteristic of dystrophic fast twitch muscles.
In chapter V, caveolin-3 expression was analyzed as the most likely
substrate of WWPl. The caveolin-3 protein is the· main component of
caveolae in skeletal muscle. Caveolin-3 deficiency induces a muscular
dystrophic phenotype, while its overexpression also does harm to muscle
tissue. In chicken muscular dystrophy, increase of caveolae was observed by
- 79-
an electron microscopic analysis, but its underlying mechanism to cause the
onset remains elucidated. Therefore, we analyzed the expression of
caveolin-3 and other caveolae-related proteins in dystrophic chickens.
Western blotting and semi-quantitive RT-PCR analysis revealed 1)
caveolin-3 is expressed higher in damaged muscle of dystrophic chickens, 2)
the amount of caveolin-3 protein is controlled in a posttranslational fashion
and 3) the expression of other caveolae-related proteins shows similar
tendency as that of caveolin-3. The results indicated that the overexpression
of caveolin-3 protein seems to be involved in the causative process of chicken
muscular dystrophy.
This study identified WWPl as the causative gene of chicken
muscular dystrophy. All E3 ubiquitin ligases involved in muscular
dystrophies reported so far are classified into RING-type E3
(Pallares-Trujillo et ai., 1997; Acharyya et ai., 2005; Kudryashova et ai.,
2005); this is the first report all over the world that WWP1, a HECT-type E3
can cause a muscular dystrophy. This information will provide a new aspect
to understand the function of WWPl gene, famous for its oncogenic
character.
Furthermore, this study showed that the aberration of WWPl gene
influences myoblast differentiation and indicated that abnormal regulation
of caveolin-3 protein leads the onset of chicken muscular dystrophy. It is
proposed that vacuoles in degenerating fibers of muscular dystrophic
- 80-
chickens (Fig. I-I, III-2, V-lh) be derived from abnormal proliferation, fusion,
enlargement of T-tubules (Nonaka, 1987b) and that the muscle degeneration
in this disease start with this vacuole formation (Libelius et al., 1979). BinI
consisting of caveolae with caveolin-3 plays a leading role in T-tubule
biogenesis (Parton et al., 1997; Lee et al., 2002). Since caveolin-3 undergoes
ubiquitination and proteasomal degradation (Galbiati et a1., 2000a), we are
currently hypothesizing 1) excessive amount of caveolin-3 proteins retain on
sarcolemma due to dysfunction of WWPl, 2) caveolin-3 overexpression
results in increase of caveolae, which causes an abnormality in T-tubule
biogenesis, 3) endocytosis from T-tubules is hyper-activated and leads the
onset of chicken muscular dystrophy.
In the future, the relationship between the overexpreSSlOn of
caveolin-3 protein and dystrophic phenotype, and the interaction between
WWPI and caveolin-3 should be clarified. The mechanism leading the onset
by excess amount of caveolin-3 protein needs also elucidating. These
information will contribute to the study of chicken muscular dystrophy and
the corresponding human dystrophies.
- 81-
ACKNOWLEDGMENTS
I would like to express my sincere gratitude to Prof. H. Mannen and
Prof. F. Mukai for giving me an opportunity to undertake this study and lots
of helpful suggestions. I also gratefully acknowledge helpful discussions with
Dr. S. Sasazaki on many points in this study.
This study owes much to the thoughtful and helpful comments of
Prof. T. Kikuchi, Prof. A. Nakamura and Prof. M. Imamura of National
Institute of Neuroscience.
Appreciation is expressed to Prof. N. Ichihara of Azabu University
and Dr. A. Fujiwara of Nippon Institute for Biological Science for providing
me with great technical helps.
Thanks are extended to members of laboratory for their invaluable
encouragements.
Finally, I thank my family, Ryoma, Hiroka and Minatsu, who have
always supported and understood me and my dream.
·82·
REFERENCE
Acharyya S, Butchbach ME, Sahenk Z, Wang H, Saji M, Carathers M, Ringel