Tài liệu Báo cáo khoa học: Molecular defect of isovaleryl-CoA dehydrogenase in the skunk mutant of silkworm, Bombyx mori ppt

Molecular defect of isovaleryl-CoA dehydrogenase in the
skunk mutant of silkworm, Bombyx mori
Kei Urano
1
, Takaaki Daimon
1
, Yutaka Banno
2
, Kazuei Mita
3
, Tohru Terada
4
, Kentaro Shimizu
4,5
,
Susumu Katsuma
1
and Toru Shimada
1,4
1 Department of Agricultural and Environmental Biology, Graduate School of Agricultural and Life Sciences, University of Tokyo, Japan
2 Institute of Genetic Resources, Graduate School of Bioresource and Bioenvironmental Science, Kyushu University, Fukuoka, Japan
3 Division of Insect Sciences, National Institute of Agrobiological Sciences, Tsukuba, Ibaraki, Japan
4 Agricultural Bioinformatics Research Unit, Graduate School of Agricultural and Life Sciences, University of Tokyo, Japan
5 Department of Biotechnology, Graduate School of Agricultural and Life Sciences, University of Tokyo, Japan
Introduction
Isovaleryl-CoA dehydrogenase (IVD; EC 1.3.99.10) is
a tetrameric, mitochondrial flavoenzyme that catalyses
the third step of leucine degradation in which isovale-
ryl-CoA is converted to 3-methylcrotonyl-CoA. IVD is
a member of the acyl-CoA dehydrogenase (ACAD)
family of enzymes, all of which share significant

sequences and employ a similar enzyme mechanism for
the a,b-dehydrogenation of acyl-CoA substrates [1].
Keywords
Bombyx mori; branched-chain amino acid;
isovaleric acidemia; isovaleryl-CoA
dehydrogenase; responsible gene
Correspondence
T. Shimada, Laboratory of Insect Genetics
and Bioscience, Department of Agricultural
and Environmental Biology, Graduate School
of Agricultural and Life Sciences, University
of Tokyo, Yayoi 1-1-1, Bunkyo-ku, Tokyo
113-8657, Japan
Fax: +81 3 5841 8011
Tel: +81 3 5841 8124
E-mail:
(Received 2 March 2010, revised 1 August
2010, accepted 25 August 2010)
doi:10.1111/j.1742-4658.2010.07832.x
The isovaleric acid-emanating silkworm mutant skunk ( sku) was first stud-
ied over 30 years ago because of its unusual odour and prepupal lethality.
Here, we report the identification and characterization of the gene responsi-
ble for the sku mutant. Because of its specific features and symptoms simi-
lar to human isovaleryl-CoA dehydrogenase (IVD) deficiency, also known
as isovaleric acidaemia, IVD dysfunction in silkworms was predicted to be
responsible for the phenotype of the sku mutant. Linkage analysis revealed
that the silkworm IVD gene (BmIVD) was closely linked to the odorous
phenotype as expected, and a single amino acid substitution (G376V) was
found in BmIVD of the sku mutant. To investigate the effect of the G376V
substitution on BmIVD function, wild-type and sku-type recombinants

were constructed with a baculovirus expression system and the subsequent
enzyme activity of sku-type BmIVD was shown to be significantly reduced
compared with that of wild-type BmIVD. Molecular modelling suggested
that this reduction in the enzyme activity may be due to negative effects of
G376V mutation on FAD-binding or on monomer–monomer interactions.
These observations strongly suggest that BmIVD is responsible for the sku
locus and that the molecular defect in BmIVD causes the characteristic
smell and prepupal lethality of the sku mutant. To our knowledge, this is,
aside from humans, the first characterization of IVD deficiency in metazoa.
Considering that IVD acts in the third step of leucine degradation and the
sku mutant accumulates branched-chain amino acids in haemolymph, this
mutant may be useful in the investigation of unique branched-chain amino
acid catabolism in insects.
Abbreviations
ACAD, acyl-CoA dehydrogenase BmIVD, Bombyx mori isovaleryl-CoA dehydrogenase; EST, expressed sequence tag; IVD, isovaleryl-CoA
dehydrogenase; PMS, phenazinemethosulfate; SNP, single nucleotide polymorphism.
4452 FEBS Journal 277 (2010) 4452–4463 ª 2010 The Authors Journal compilation ª 2010 FEBS
IVD dysfunction is well characterized in humans as
the first recognized organic acidaemia (isovaleric acida-
emia), which causes recurrent episodes of vomiting,
lethargy, developmental delay and sometimes acute
neonatal death [2,3]. Isovaleric acid, one of the deriva-
tives of isovaleryl-CoA, is abnormally excreted in
blood and causes a characteristic sweaty feet odour in
patients.
The isovaleric acid-emanating silkworm mutant
skunk (sku) was first described over 30 years ago as an
‘odorous silkworm’ [4,5]. The sku gene is an autosomal
recessive lethal gene and the sku mutant exhibits
prepupal lethality (Fig. 1). However, the gene responsi-

ble for the sku mutant has not yet been identified. The
sku mutant has two physiological characteristics. First,
the distinctive odour of the mutant is caused by the
accumulation of isovaleric acid in the larval excrement.
Second, the mutant exhibits abnormally high accumu-
lations of branched-chain amino acids, including
leucine, in haemolymph [6]. From the second feature,
it can be assumed that the branched-chain amino acid
degradation mechanism might be dysfunctional in
the sku mutant, whereas it is apparent that the first
symptom resembles isovaleric acidaemia in humans.
Because the isovaleric acid-accumulating mechanism in
animals is restricted to isovaleryl-CoA decomposition
disorder (Kyoto Encyclopedia of Genes and Genomes;
it can be expected that IVD
deficiency in silkworm may account for the odorous
phenotype, similar to the IVD deficiency observed in
human isovaleric acidaemia.
In this study, the IVD gene of the silkworm Bomb-
yx mori was identified and a single nucleotide substitu-
tion in the highly conserved site (G376V) in BmIVD of
the sku allele was determined. Genetic and biochemical
analyses indicated that this substitution caused the sku
phenotype. Because G376V was a novel mutation in
IVD, the effect of the mutation was further investi-
gated by molecular modelling. Together with the previ-
ously reported traits of the sku mutant, the molecular
and physiological effects of IVD dysfunction in silk-
worms are compared with those observed in humans.
Results

Identification of the B. mori isovaleryl-CoA
dehydrogenase (BmIVD) gene as a candidate for
the sku mutant
A search of the silkworm expressed sequence tag
(EST) database revealed the existence of several puta-
tive acyl-CoA dehydrogenase genes in silkworm.
Among them, one EST clone, fdpeP14_F_F20, exhib-
ited the highest homology (69%) to human IVD. After
analysis of the full-length sequence of this EST clone,
it was apparent that two key residues distinguishing
IVD from other ACAD family members are both con-
served. One is the catalytic base E254, which abstracts
sku / +
sku
sku / sku sku / +
sku
sku / sku
B A
Fig. 1. Phenotypes of control silkworm
(sku ⁄ +
sku
) and the skunk mutant (sku ⁄ sku)
at (A) day 2 of fifth instar and (B) 10 days
after spinning. Until spinning, the skunk
mutant larva develops normally (A). After
spinning, the mutant dies without pupation,
whereas control larva successfully moults to
pupa (B). Scale bar, 10 mm.
K. Urano et al. Odorous silkworm mutant
FEBS Journal 277 (2010) 4452–4463 ª 2010 The Authors Journal compilation ª 2010 FEBS 4453

the a-hydrogen of the substrate, and the other is
G374, which forms the binding pocket for the
branched-chain acyl moiety of substrate isovaleryl-
CoA in human IVD [7,8]. As represented in Fig. 2C,
both residues were conserved in silkworm proteins
(E276 and G396, respectively). Thus, the EST clone
was named BmIVD in this study. A homology search
of BmIVD on the silkworm revealed that this gene was
mapped on the 22nd linkage group [9]. Because sku is
also mapped on the same chromosome [10], the genetic
loci of sku and BmIVD were compared. In the single
nucleotide polymorphism (SNP) linkage map, the
genetic distance between BmIVD and ptth genes was
estimated to be 10.5 cM [11]. This value strongly
agreed with the distance between the sku and ptth loci
in the genetic linkage map (13.5 cM), encouraging
further investigation of BmIVD as a candidate gene
for sku.
Comparison of BmIVDs from wild-type (wt) and
sku strains
RT-PCR analysis revealed that BmIVD mRNA was
expressed in both wild-type and sku mutant silkworms
with the same molecular size and expression levels
(Fig. 2A). Determination of full-length cDNA
sequences using the RACE method in both strains
revealed the presence of a single point mutation in
1337 nucleotides of BmIVD. In the sku mutant, the
1127th guanine from the start codon was substituted
by thymine and none of the other sites were altered
(Fig. 2B). A 1127G>T mutation is missense, changing

2.3
2.0
1.1
(kb)
B
C
A
wt
sku
Skunk-RT2 gPCRsku-R2
T
poly(A)
Start codon Stop codon
Probe
skunk-RT1
+1 +1127 +1251 +1300–37
G
poly(A)
(nt)
rp49
BmIVD
*
G376V
Fig. 2. Cloning of the BmIVD gene. (A) RT-
PCR analysis revealed the BmIVD band
amplified from whole-body RNA of standard
strains p50T and c108T as well as the sku
mutant. Migrations of the molecular mass
marker and control gene rp49 are indicated.
(B) Full-length mRNA of wild-type and sku

mutant BmIVD are represented. Grey boxes
depict the open reading frame (ORF) with
blank arrowheads indicating the start and
stop codons at the edge. The nucleotide
length of each part is also shown. The sin-
gle nucleotide substitution 1127G>T in the
sku mutant is represented as a dashed line.
The position of the primers used for RT-PCR
and that of the probe used for northern blot-
ting are indicated with black arrowheads
and a black arrow, respectively. (C) Amino
acid sequence alignment of IVDs from Pseu-
domonas aeruginosa PAO1 (bacteria,
NP_250705), Arabidopsis thaliana (arabidop-
sis, NP_190116), Homo sapiens (human,
NP_002216), Caenorhabditis elegans (nema-
tode, NP_500720) and B. mori (silkworm,
AB458683). Alignment was generated using
CLUSTAL W algorithm v1.83 and shaded using
the
BOXSHADE program. The arrow indicates
the position of the 376th glycine residue (G)
of BmIVD, which is replaced by a valine (V)
in the sku mutant. The catalytic base (D),
IVD-specific residue in the acyl binding
pocket (*) and major functional domains
(thick underlines) of IVD are also indicated.
Odorous silkworm mutant K. Urano et al.
4454 FEBS Journal 277 (2010) 4452–4463 ª 2010 The Authors Journal compilation ª 2010 FEBS
the codon for glycine residue at position 376 to that

for valine (G376V) in BmIVD. An investigation of the
nucleotide sequence at the 1127th position in eight
+
sku
⁄ +
sku
silkworm strains revealed that all strains
conserved the canonical guanine at this site (data not
shown). To examine other possible variations, genome
sequences of BmIVD were determined and 11 SNPs
between BmIVDs of wild-type and sku were found, in
addition to 1127G>T. However, all were located on
the introns of BmIVD (data not shown). Thus, these
11 SNPs appeared to have no functional influence on
this gene.
Alignment of the amino acid sequences of IVDs
(Fig. 2C) exhibited that BmIVD is highly homologous
to other IVDs throughout the entire region. Notably,
the glycine residue corresponding to G376 of BmIVD,
which was substituted to valine in the sku mutant, is
highly conserved from bacterial to mammalian IVDs,
indicating the importance of this residue.
Linkage analysis between sku and BmIVD genes
To determine the consistency between the sku pheno-
type and the BmIVD genotype, linkage analysis
between the wild-type and sku strains was performed.
For this, crossing of the strain a85, in which sku locus
is marked with or, a recessive ‘oily’ gene that causes
translucent epidermis, was performed. As represented
in Fig. 3, four genotypes from F

1
progenies were
obtained and SNPs from a total of 133 individuals
were sequenced for BmIVD at nucleotide 1127
(Table 1). As expected, all the odorous individuals
were homozygous for T ⁄ T at nucleotide 1127. How-
ever, none of the nonodorous individuals had the T ⁄ T
genotype at this site (G ⁄ GorG⁄ T) (Fig. 3), suggesting
no recombination between the sku locus and BmIVD
gene.
Expression pattern of the BmIVD gene
Northern blot analysis was performed to investigate
the expression profile of the BmIVD gene. For both
wild-type and sku strain, a single band of 1.35 kb, cor-
responding to the predicted molecular size of BmIVD
mRNA, was detected in all the tissues tested (Fig. 4A).
The spatial expression pattern of BmIVD mRNA was
similar between wild-type and sku and densitometric
analysis of three independent experiments showed that
relative expression levels of BmIVD (normalized to
Actin3) in each tissue were not statistically significant
between wild-type and sku strain (P > 0.05, t-test)
(data not shown). Therefore, it is likely that the differ-
ence in the regulation of BmIVD expression between
+
sku
and sku alleles is not responsible for the odorous
phenotype. To further characterize the spatial expres-
sion of BmIVD in the wild-type strain, 15 tissues were
investigated using RT-PCR analysis. The result showed

that BmIVD is expressed in various tissues, ranging
from digestive organs such as midgut to reproductive
organs such as ovary and testis or the respiratory
organ trachea (Fig. 4B). Among these tissues, fat body
and midgut showed higher expression levels than other
tissues. It is noteworthy that both tissues play essential
roles in nutrient turnover in insects. Namely, nutrients
are digested and absorbed in the midgut and stored
and metabolized in the fat body which is equivalent to
liver in mammals. Thus, it is likely that BmIVD may
G/G G/T T/T
1127
|
P
F
1
1127
|
1127
|
or +
++
or sku
++
or +
or sku
or sku
or sku
or sku
or +

or sku
++
Non-odorous Odorous
SNP of BmIVD
Oily Non-oily Oily
G/T
1127
|
Phenotypes
Fig. 3. Linkage analysis between sku and BmIVD. Recessive gene
or linked to sku on the 22nd linkage group was utilized to distin-
guish sku heterozygous mutants. The upper part indicates that sku
heterozygous mutants were crossed and three kinds (nonodorous-
non-oily, nonodorous-oily and odorous-oily) of F
1
generation were
distinguished by combination of or and sku phenotypes. The lower
part indicates the representative results of genomic DNA direct
sequencing of PCR products harbouring the SNP 1127G>T of
BmIVD from each of the three phenotypes.
Table 1. Results of linkage analysis between sku and BmIVD.A
single nucleotide polymorphism at the 1127th base pair of BmIVD
ORF was analysed from 133 individuals obtained from an F
1
inter-
cross (for details, see Fig. 3).
Phenotype (genotype)
Number of
larvae
screened

1127th base pair
of BmIVD ORF
G ⁄ GG⁄ TT⁄ T
Normal (+ ⁄ +, sku ⁄ +)30 12180
Oily and nonodorous (sku ⁄ +) 60 0 60 0
Oily and odorous (sku ⁄ sku)43 0 043
K. Urano et al. Odorous silkworm mutant
FEBS Journal 277 (2010) 4452–4463 ª 2010 The Authors Journal compilation ª 2010 FEBS 4455
participate in amino acid catabolism for energy pro-
duction in the silkworm.
Expression of recombinant BmIVD by the
baculovirus expression system
To evaluate the effect of the G376V amino acid substi-
tution in the sku mutant on the IVD activity and sub-
strate specificity, we first carried out overexpression
and purification of wild-type and sku-type (G376V)
BmIVDs. Because the expression level in IVD-recom-
binant Escherichia coli was previously reported to be
extremely low and 5¢-end alteration to mimic codon
usage of E. coli is necessary for improved expression
levels [12], a baculovirus expression system was
employed to overexpress the BmIVD. Sf9 cells were
infected with a recombinant baculovirus that expresses
the full-length BmIVD with His-tagged sequences at
the C-terminus. As indicated in Fig. 5A, the expression
level of recombinant BmIVD was sufficiently high that
a putative BmIVD band could be observed in Coomas-
sie Brilliant Blue staining. Western blot analysis
revealed that the molecular mass of the expressed
BmIVD is apparently lower than that of the predicted

full-length recombinant protein (46.5 kDa). However,
a size difference between wild-type and sku-type
BmIVDs was not observed (Fig. 5B). This suggests
that the point mutation at Gly376 does not have an
effect on the processing of mitochondrial leader pep-
tide common in IVDs [13,14]. The recombinant protein
was successfully purified to homogeneity by a single-
step, nickel-chelating chromatography procedure
(Fig. 5C, D) and used for enzymological studies.
Enzymatic activity and substrate specificity of
wild-type and sku-type recombinant BmIVDs
The enzymatic activity of purified recombinant
BmIVD was measured with a variety of acyl-CoA sub-
strates using a dye-reduction assay (Fig. 6). When iso-
valeryl-CoA was used as a substrate, significant
activity (614 nmol of 2,6-dichloroindophenol reduced
mg protein
)1
Æmin
)1
) was observed in wild-type
BmIVD. Meanwhile, when substrates for other
ACADs such as isobutyryl-CoA for isobutyryl-CoA
dehydrogenase and hexanoyl-CoA for medium-chain
acyl-CoA dehydrogenase were used, wild-type BmIVD
exhibited residual but much lower activities against
these substrates compared with isovaleryl-CoA. This
confirms that BmIVD specifically functions in isovale-
ryl-CoA dehydrogenation, similar to IVDs observed in
other species [15].

Next, sku-type BmIVD (G376V) was examined to
determine if it retained enzymatic activities. As indicated
in Fig. 6, sku-type BmIVD exhibited only faint ACAD
activities against all the substrates investigated. This
1.3 kb
FB MG MT EP
wt
FB MG MT EP
sku
Actin3
BmIVD
rp49
BmIVD
B
A
Fig. 4. Expression profiles of BmIVD. (A) Northern blotting compares the expression levels of BmIVD from several tissues obtained from
fifth instar larvae at day 2 of wild-type (p50T) and mutant (sku) strains. Total RNA (5 lg) prepared from fat body (FB), midgut (MG), Malpi-
ghian tubule (MT) and epidermis (EP) were blotted and hybridized with the digoxigenin (DIG)-labelled probe. The arrowhead indicates the
positive signal. Silkworm Actin3 is represented as a control. (B) RT-PCR analysis using cDNAs from 15 tissues of wild-type p50T strain are
indicated. Lane 1, brain (BR); lane 2, prothoracic gland (PG); lane 3, salivary gland (SaG); lane 4, central nervous system (CNS); lane 5, tra-
chea (TR); lane 6, fat body (FB); lane 7, ovary (OV); lane 8, testis (TES); lane 9, anterior silk gland (ASG); lane 10, middle silk gland (MSG);
lane 11, posterior silk gland (PSG); lane 12, midgut (MG); lane 13, hindgut (HG); lane 14, Malpighian tubule (MT); lane 15, epidermis (EP).
Silkworm rp49 was the control.
Odorous silkworm mutant K. Urano et al.
4456 FEBS Journal 277 (2010) 4452–4463 ª 2010 The Authors Journal compilation ª 2010 FEBS
result demonstrates that the point mutation at Gly376
in BmIVD almost totally disrupts the function of
BmIVD as an enzyme, indicating BmIVD dysfunction
in sku mutants.
Sequence alignment and position of the

mutation in the 3D structure
The results of the sequence alignment within the
ACAD family revealed that the mutated residue in the
sku mutant was strictly conserved and positioned
within the loop structure connecting helices I and J
(Fig. 7A). The loop region consists of 10 amino acids,
seven of which are conserved within the species
(Fig. 2C) and two of which are conserved within all
ACAD family of enzymes, forming a highly conserved
motif of Gly–Gly–X–Gly. The second glycine of this
motif (Gly374) is known to make a hydrogen bond
with the pyrophosphate moiety of the FAD of the next
monomer in the tetramer [16]. However, the third gly-
cine (Gly376), which is conserved within all ACAD
family members and is substituted to valine in the sku
mutant, has not been functionally clarified. To investi-
gate how the protein loses its enzymatic activity with
the G376V mutation (Fig. 6), comparative models of
wild-type and mutant BmIVDs were constructed by
using the crystal structure of human IVD (PDB
ID: 1IVH) as a template [7]. The model of wild-type
BmIVD indicated that all the side chains in the loop
connecting helices I and J are exposed on the surface
of the monomer, pointing toward FAD or its neigh-
bouring monomer in the tetramer. By contrast, the
side chain of the mutated residue (Val376) points
toward the inside of the helix–loop–helix structure.
Consequently, the side-chain atoms of Val376 overlap
with those of Ile369, Leu372 and Thr383 with inter-
atomic distances of < 3 A

˚
in the mutant structure
(Fig. 7B). To avoid these overlaps, the mutant proba-
bly has a different structure in this region.
Discussion
In this study, a candidate gene approach was utilized
to discover the gene responsible for the odorous silk-
worm mutant sku. The candidate gene BmIVD was
identified and a single nucleotide substitution was
found in the codon of a highly conserved residue, not
only in the species, but also in all enzyme family mem-
bers (Figs 2C and 7A). It was demonstrated that this
substitution perfectly cosegregated with the sku loci
(Fig. 3 and Table 1) and dramatically decreased the
enzymatic activity (Fig. 6). These genetic and biochem-
ical data, along with previous observations that the
sku mutant accumulates isovaleric acid and branched-
chain amino acids, strongly indicate that a single
amino acid substitution (G376V) in BmIVD is respon-
sible for the sku mutant. In the sku mutant, dysfunc-
tion of BmIVD would cause hydrolytic degradation of
75
50
37
25
(kDa)
1 2 3 4 5 6 7 8 9 10
AB
C
D

75
50
37
25
(kDa)
50
37
Control
Control
Fig. 5. Expression and purification of His-tagged BmIVD protein. Protein samples were electrophoresed by SDS ⁄ PAGE and analysed by Coo-
massie Brilliant Blue staining (A,C) or western blotting with the anti-His IgG (B,D). The molecular mass markers are indicated on the left.
(A,B) Confirmation of baculovirus-expressed recombinant BmIVD. Three kinds of whole cells, Sf9 cells infected with parental AcMNPV
(control), wild-type BmIVD (wt) and sku-type BmIVD (sku) under the polyhedrin promoter, were electrophoresed. Arrows indicate the BmIVD
band around 40 kDa. (C,D) Recombinant His-tagged BmIVD was purified from virus-infected cells by nickel chromatography. Arrows indicate
the position of the recombinant BmIVD. Lane 1, cell lysate; lane 2, proteins not binding to the column; lanes 3, 4 and 5, wash fraction (5, 20
and 40 m
M imidazole, respectively) and lanes 6–10, eluate fraction (500 mM imidazole).
K. Urano et al. Odorous silkworm mutant
FEBS Journal 277 (2010) 4452–4463 ª 2010 The Authors Journal compilation ª 2010 FEBS 4457
isovaleryl-CoA to isovaleric acid, instead of the dehy-
drogenation of the substrate, resulting in the character-
istic odour of the mutant. To our knowledge, this is
the first report of IVD deficiency in animals aside from
humans.
IVD is a member of the ACAD family of enzymes,
all of which employ a similar enzyme mechanism for
a,b-dehydrogenation of acyl-CoA substrates [1]. Con-
servation of the substituted glycine residue in all
ACAD members (Fig. 7A) and significant reduction of
ACAD activities in G376V protein (Fig. 6) suggest

that Gly376 (corresponding to position 354 in human
IVD) is essential for common mechanisms in ACADs.
ACAD is a homotetrameric or homodimeric flavopro-
tein with each monomer containing one molecule of
FAD [17]. FAD not only serves as a catalyst, but also
bridges the monomers; it is located at the interface
between the monomers and forms many hydrogen
bonds with both [16,18]. Molecular modelling revealed
that the G376V mutation causes the side chain of
Val376 to overlap with other residues, possibly result-
ing in changes in the loop structure. Although the
loop has only been characterized to make a hydrogen
bond with FAD [16], the model structures indicated
that it is also involved in interactions with the neigh-
bouring monomer. As shown in Fig. 7B, Tyr377 and
Asn379 in the loop region interact with Leu236 and
Asp234, respectively, of the neighbouring monomer
via a hydrogen bond. These results suggest that the
G376V mutation alters the structure of the loop
region and affects the interactions between monomers
and with FAD. Imperfect FAD-binding or tetramer
formation would lead to disappearance of the enzy-
matic activity (Fig. 6). Recent clinical mutation stud-
ies about ACAD deficiency in humans support this
prediction [17]. An identical substitution at the
homologous position (G371V) in human short-chain
acyl-CoA dehydrogenase has also been reported and,
though this protein’s enzymatic activity was not men-
tioned, in vitro import studies revealed that this muta-
tion led to a temperature-dependent inability to form

tetramers [19].
The sku larvae begin emanating isovaleric acid
odour from the first day after hatching, but do not
show any signs of developmental abnormality until the
onset of spinning (Fig. 1A). The mutants start
spinning after the normal duration of the final instar
(6–8 days) but stop after a short time and develop a
very thin cocoon. They eventually die without becom-
ing pupae in about a week after spinning (Fig. 1B).
Isovaleric acid seems to be the cause of prepupal
lethality because injection of isovaleric acid into
normal spinning larva induces a phenocopy of the
pupation defect observed in the sku mutant [5].
Because the silkworm larvae cannot excrete after the
onset of spinning, highly accumulated isovaleric acid
in sku prepupae may have toxic effects and cause
prepupal lethality. In humans, patients with isovaleric
acidaemia suffer from recurrent episodes of vomiting,
lethargy, developmental delay and sometimes acute
neonatal death [2,3]. These symptoms are also thought
to be caused by isovaleric acid, but the underlying
mechanisms are largely unknown. Because a narcotic
effect of short chain fatty acids has been known
[20,21] and isovaleric acid is also toxic to silkworm [5],
there may be common mechanisms between silkworms
and humans in how isovaleric acid causes severe symp-
toms in addition to the characteristic odour.
One of the most intriguing features of the sku mutant
is that the mature larva accumulates branched-chain
amino a cids, leucine, isoleucine and valine, in haemolymph

at levels  4 times higher in females and  7–12 times
wt
120
100
80
60
40
20
0
sku
Relative activity (%)
*
*
**
IV-CoA IB-CoA HX-CoA
Fig. 6. Relative enzymatic activities of wild-type (wt) and sku-type
BmIVD (sku). Enzymatic activities of isovaleryl-CoA (IV-CoA), isobu-
tyryl-CoA (IB-CoA) and hexanoyl-CoA (HX-CoA) were assayed by
the 2,6-dichloroindophenol ⁄ PMS dye-reduction method. The data
show means ± SD of pooled data from two independent experi-
ments each performed in triplicate. (**P < 0.0001, *P < 0.05, one-
tailed, Student’s t-test).
Odorous silkworm mutant K. Urano et al.
4458 FEBS Journal 277 (2010) 4452–4463 ª 2010 The Authors Journal compilation ª 2010 FEBS
higher in males than in normal silkworms [6]. In human
cases, however, patients with isovaleric acidaemia do
not exhibit such accumulation of amino acids [2]
because an enzyme reaction one step before IVD dehy-
drogenation, in which a-ketoisocaproate is catalysed by
branched-chain a-keto acid dehydrogenase, is irrevers-

ible in humans [22]. This indicates that, in the silk-
worm, there might be an additional mechanism
bypassing branched-chain a-keto acid dehydrogenase
irreversibility, which enables indirect accumulation of
upstream leucine in the sku mutant. It is interesting
that isoleucine and valine also accumulate to the same
degree as leucine [6]. This phenomenon suggests
that leucine catabolism plays an important role in regu-
lating all three branched-chain amino acid levels in
haemolymph.
Thus far, little is known about amino acid catabolism
in insects. It is hoped that the sku mutant will give
insights into the unique catabolism of branched-chain
amino acids in insects. Using recently released B. mori
genome data [9] (also see KAIKObase, http://
sgp.dna.affrc.go.jp/KAIKObase/), several genes respon-
sible for amino acid turnover in silkworm mutants have
been identified [23,24]. The genetic resources of the silk-
worm, together with its advantageous large body size
for physiological study, will facilitate the further study
of amino acid catabolism in insects.
Materials and methods
Materials
B. mori strains p50T and c108T were used as wild-type silk-
worms, which are maintained at the University of Tokyo.
Odorous silkworm-segregating strain a85, which is main-
tained at Kyushu University, was also used. To identify
homozygous sku mutants, larvae were individually reared
in Petri dishes and the odour was determined by sniffing.
wt sku

A
G376V
B
G376
V376 I369
I369
BmIVD
IVD
IBD
SBCAD
GCD
SCAD
MCAD
LCAD
VLCAD
ACAD9
ACAD10
398
376
399
415
415
369
377
413
423
427
938
755
J

T383
T383
L372 L372
Y377
L236 D234
N379
G374 G374
L236
N379
Y377
D234
I
Loop region
Fig. 7. Sequence alignment and predicted BmIVD structure. (A) Alignment of BmIVD with all 11 acyl-CoA dehydrogenase family members
found in humans. The arrow indicates the substituted residue in the sku mutant and the thick line represents conserved helices I and J. The
loop region is also shown. IVD (isovaleryl-CoA dehydrogenase, NP_002216); IBD (isobutyryl-CoA dehydrogenase, NP_055199); SBCAD
(short ⁄ branched-chain acyl-CoA dehydrogenase, NP_001600); GCD (glutaryl-CoA dehydrogenase, NP_000150); SCAD (short-chain acyl-CoA
dehydrogenase, NP_000008); MCAD (medium-chain acyl-CoA dehydrogenase, NP_000007); LCAD (long-chain acyl-CoA dehydrogenase,
NP_001599); VLCAD (very long-chain acyl-CoA dehydrogenase, NP_000009); ACAD9 (acyl-CoA dehydrogenase 9, NP_054768); ACAD10
(acyl-CoA dehydrogenase 10, NP_001130010) and ACAD11 (acyl-CoA dehydrogenase 11, NP_115545). (B) Close-up view of residue 376 to
comparative models of wild-type (wt) and sku-type (sku) BmIVD based on an X-ray structure of human IVD. The main chains are represented
by ribbons and the atoms of key residues and FAD are shown with a stick model. Helices I and J are coloured gray and the loop between
the two helices is coloured yellow, except for mutation site 376 which is coloured pink. FAD of the neighbouring monomer is coloured
orange and main chain of neighbouring monomer is coloured blue. Two-headed arrows indicate distances from the side chain of residue 376
to side chains of other residues that are < 3 A
˚
apart. Hydrogen bonds between Gly374 and FAD, Tyr377 and Leu236, and Asn379 and Asp
234 are represented as a dashed line.
K. Urano et al. Odorous silkworm mutant
FEBS Journal 277 (2010) 4452–4463 ª 2010 The Authors Journal compilation ª 2010 FEBS 4459

The silkworms were reared on fresh mulberry leaves in an
insect rearing chamber under standard conditions (25 °C,
12L : 12D photoperiod). Sf9 cells were cultured at 27 °Cin
TC-100 insect medium (SAFC Biosciences, Lenexa, KS,
USA) supplemented with 10% fetal bovine serum. Autogra-
pha californica multiple nucleopolyhedrovirus (AcMNPV)
was propagated in the Sf9 cells as described previously [25].
All tissues and cells to be examined were washed twice in
NaCl ⁄ P
i
(137 mm NaCl, 2.7 mm KCl, 8.1 mm Na
2
HPO
4
,
1.5 mm KH
2
PO
4
), immediately frozen in liquid nitrogen
and stored at )80 °C. PCRs were performed using the
ExTaq Kit (Takara Bio, Shiga, Japan), unless otherwise
mentioned.
Isolation of B. mori cDNA encoding the IVD-like
gene
To identify the Bombyx gene which is homologous to the
IVD gene, the EST database was screened [26]. The cDNA
clone fdpeP14_F_F20 exhibited the highest homology to
human IVD and was subjected to further analysis. Assess-
ment of the genetic loci of the EST clone was performed

using KAIKObase ( />The nucleotide sequence was determined using the ABI
PRISM BigDye Terminator v3.1 Cycle Sequencing Kit
(Applied Biosystems, Foster City, CA, USA) and the ABI
Prism 3130 DNA Sequencer (Applied Biosystems).
Sequence data were analysed using the program package
genetyx-mac version 12.0 (Genetyx Corporation, Tokyo,
Japan). PCR primers used in this study are listed in
Table S1.
RT-PCR of BmIVD
Total RNA was extracted using Trizol reagent (Invitrogen,
Carlsbad, CA, USA). One lg of total RNA was reverse
transcribed using the RNA PCR Kit (Takara Bio). PCR
was performed using skunkRT1 and skunkRT2 primers
(Table S1). Temperature cycling consisted of 40 cycles of
denaturing at 94 °C for 30 s, annealing at 54 °C for 30 s
and extension at 72 °C for 90 s.
5¢- and 3¢-rapid amplification of cDNA ends
(RACE)
In order to examine differences in the full-length cDNA
nucleotide sequences between odorous and normal silk-
worm BmIVD,5¢- and 3¢-RACE was performed using the
GeneRacer Kit (Invitrogen). Five micrograms of total
RNA was used to dephosphorylate, remove the 5¢ cap,
ligate the RNA Oligo and reverse-transcribe the nucleotide
sequences. The PCR primers used in this experiment are
listed in Table S1. PCRs were carried out according to the
manufacturer’s instructions. PCR products were subcloned
into the pGEM-T Easy vector (Promega, Madison, WI,
USA). The nucleotide sequences were determined as
described above.

Preparation and sequencing of the BmIVD
genomic clone
Genomic DNA was extracted from the silk glands of fifth
instar larvae according to standard methods [27]. Because
the genomic structure of BmIVD is long ( 11 kb), the gen-
ome sequence was divided into two parts and sequenced
separately. The genomic sequence of BmIVD was PCR-
amplified using the TaKaRa LA Taq Kit (Takara Bio).
Amplified PCR fragments were subcloned and sequenced as
described above. Full-length cDNA and genomic sequences
of wild-type BmIVD were deposited into the GenBank ⁄
EMBL ⁄ DDBJ data bank with accession numbers
AB458683 for cDNA and AB462483 for genomic DNA.
Linkage analysis between sku and BmIVD
The heterozygous mutant of sku can be identified using the
sku-linked recessive oily gene or. Crossing was performed
as indicated in Fig. 3. Thirty normal larvae, 60 oily but
nonodorous larvae and 43 oily and odorous larvae were
screened at fifth instar. To extract genomic DNA, caudal
portions of the larvae were cut and homogenized with a
pestle and DNeasy Blood and Tissue Kit (Qiagen, Venlo,
The Netherlands) was used. The genomic DNA was ampli-
fied by PCR with primers PCRseqF and gPCRsku-R1,
which were designed to amplify the fragment that contains
the substitution site in BmIVD. The PCR product was then
cleaned using the QIAquick PCR Purification Kit (Qiagen)
and directly sequenced as described above.
Northern blot analysis
Total RNA from the fat body, midgut, Malpighian tubule
and epidermis of day 2 fifth instar larvae was prepared

using Trizol reagent (Invitrogen). Probes for BmIVD
mRNA were amplified by PCR using the DIG probe syn-
thesis Kit (Roche, Basel, Switzerland) with primers
skunkRT1 and gPCRsku-R2. The vector synthesized in the
protein expression experiment was used as the template.
Northern blot analysis was performed according to proce-
dures described previously [28,29].
Production of recombinant baculoviruses
Recombinant AcMNPVs were constructed using the
Bac-to-Bac Baculovirus Expression System (Invitrogen).
Two recombinant viruses were constructed, one expressing
wild-type BmIVD and the other expressing sku-type
BmIVD. For this, the coding region of BmIVD was
Odorous silkworm mutant K. Urano et al.
4460 FEBS Journal 277 (2010) 4452–4463 ª 2010 The Authors Journal compilation ª 2010 FEBS
PCR-amplified using either the cDNA from p50T or sku
mutant as a template with primers recombi-IVD-F2 and
recombi-IVD-R2. In this procedure, a high-fidelity DNA
polymerase, KOD-Plus (TOYOBO Life Science, Osaka,
Japan), was used for PCR. The PCR products were
digested with EcoRI and XbaI and ligated into the corre-
sponding site of the pFastBac1 vector (Invitrogen). The
construction and propagation of recombinant AcMNPVs
were performed according to the manufacturer’s instruc-
tions (Invitrogen).
Expression and purification of recombinant
BmIVD protein
Monolayers of Sf9 cells in a 150-mm dish were infected with
BmIVD-recombinant AcMNPV. After 72 h, the culture med-
ium was discarded and cells were suspended in 10 mL

NaCl ⁄ P
i
. The suspension was centrifuged at 3000 g for
10 min and the cell pellet was washed and stored at )80 °C
until use. To purify recombinant proteins, harvested cells
were resuspended in 5 mL of 50 mm potassium phosphate
buffer (pH 8.0) per dish, together with protease inhibitor
cocktail tablets (Roche). The cells were lysed by sonication
for 1 min in the Branson Sonifier 250 (Branson, Danbury,
CT, USA) and added with the same volume of binding buffer
(5 mm imidazole, 20 mm sodium phosphate, 500 mm NaCl,
pH 7.4). After centrifugation at 14 000 g for 10 min, the
resulting supernatants were loaded onto a HisGraviTrap col-
umn (GE Healthcare Bioscience, Little Chalfont, UK). The
eluate was dialysed twice in 100 mm potassium phosphate
(pH 8.0) and 100 mm NaCl using the Slide-A-Lyser Dialysis
Cassette (Pierce, Rockford, IL, USA). The protein concentra-
tion was determined using the Coomassie Plus Protein Assay
Reagent (Pierce) with bovine serum albumin as the standard.
Expression and purification of recombinant protein was
confirmed by SDS ⁄ PAGE [30] and western blot as
described previously [31].
Enzyme assays
The isovaleryl-CoA dehydrogenase activity was assayed
spectrophotometrically by the dye-reduction method using
2,6-dichloroindophenol as an electron acceptor and phenaz-
inemethosulfate (PMS) as an intermediate electron carrier
as described previously [32,33], with slight modifications.
The incubation medium was composed of 50 mm potassium
phosphate buffer (pH 8.0), 1.5 mm PMS, 0.05 mm 2,6-di-

chloroindophenol, 0.1 mm FAD and 0.1 mm acyl-CoA sub-
strate. The final volume was 100 lL. The enzyme reaction
was carried out at 25 °C and the reaction was started with
the addition of the acyl-CoA substrate. A reduction rate of
600 nm absorbancy, resulting from bleaching of 2,6-dichlo-
roindophenol, was measured for 2 min using a Beckman
DU 640 spectrophotometer (Beckman Coulter, Brea, CA,
USA). Enzymatic activity was calculated by subtracting the
reduction rate of the enzyme-excluded solution from that of
the enzyme-containing solution and was expressed as nmols
of 2,6-dichloroindophenol reduced per mg of protein per
min. The extinction coefficient of 2,6-dichloroindophenol
(21 000 MÆcm
)1
) at 600 nm was used to compute the
amount of 2,6-dichloroindophenol reduced. FAD, PMS
and 2,6-dichloroindophenol were obtained from Wako Pure
Chemical Industries (Osaka, Japan) and isovaleryl-CoA,
isobutyryl-CoA and hexanoyl-CoA substrates were
obtained from Sigma-Aldrich (St. Louis, MO, USA).
Comparative modelling of BmIVD structure
Comparative models of wild-type and mutant BmIVDs
were generated based on the crystal structure of human
IVD (PDB ID: 1VH) [7]. The primary sequence of BmIVD
was aligned with that of human IVD using blast [34]. The
model structures were generated to have the same tetra-
meric structure as the human IVD protein in the crystal
structure. FAD and a substrate in the crystal structure were
also included in the model. modeller 9v3 was used to gen-
erate the models [35]. Conformations of the side chains

were refined with SCWRL 3.0 [36] and the quality of the
models was evaluated with Verify3D [37].
Acknowledgements
This work was supported by grants from MEXT (Nos.
17018007 to T.S.), JSPS (21248006 to TD and TS),
MAFF-NIAS (Agrigenome Research Program) and
JST (Professional Program for Agricultural Bioinfor-
matics), Japan. The silkworm strains and DNA clones
were provided by the National Bioresource Project
(NBRP), Japan.
References
1 Thorpe C & Kim JJ (1995) Structure and mechanism of
action of the acyl-CoA dehydrogenases. FASEB J 9,
718–725.
2 Tanaka K, Budd MA, Efron ML & Isselbacher KJ
(1966) Isovaleric acidemia: a new genetic defect of leu-
cine metabolism. Proc Natl Acad Sci USA 56, 236–242.
3 Vockley J & Ensenauer R (2006) Isovaleric acidemia:
new aspects of genetic and phenotypic heterogeneity.
Am J Med Genet C Semin Med Genet 142, 95–103.
4 Yoshitake N, Kobayashi M & Miyashita T (1978) On
the ‘skunk’ mutant in the silkworm. J Sericult Sci Japan
47, 32–34.
5 Yoshitake N, Kobayashi M & Ogawa Y (1978) On a
smell factor existing in faeces from the skunk silkworm,
Bombyx mori. J Sericult Sci Japan 47, 161–165.
6 Inokuchi T & Yoshitake N (1978) Abnormality of
amino acid metabolism in the mutant-skunk of the
K. Urano et al. Odorous silkworm mutant
FEBS Journal 277 (2010) 4452–4463 ª 2010 The Authors Journal compilation ª 2010 FEBS 4461

silkworm, Bombyx mori. J Sericult Sci Japan 47,
154–160.
7 Tiffany KA, Roberts DL, Wang M, Paschke R,
Mohsen AW, Vockley J & Kim JJ (1997) Structure of
human isovaleryl-CoA dehydrogenase at 2.6 A
˚
resolution: structural basis for substrate specificity.
Biochemistry 36, 8455–8464.
8 Mohsen AW & Vockley J (1995) Identification of the
active site catalytic residue in human isovaleryl-CoA
dehydrogenase. Biochemistry 34, 10146–10152.
9 The International Silkworm Genome Consortium
(2008) The genome of a lepidopteran model insect, the
silkworm Bombyx mori. Insect Biochem Mol Biol 38,
1036–1045.
10 Ninagi O, Doira H & Yoshitake N (1996) Genetical
studies of the ‘skunk’ mutant in the silkworm,
Bombyx mori. J Sericult Sci Japan 65, 436–440.
11 Shimomura M, Minami H, Suetsugu Y, Ohyanagi H,
Satoh C, Antonio B, Nagamura Y, Kadono-Okuda K,
Kajiwara H, Sezutsu H et al. (2009) KAIKObase: an
integrated silkworm genome database and data mining
tool. BMC Genomics 10, 486.
12 Mohsen AW & Vockley J (1995) High-level expression
of an altered cDNA encoding human isovaleryl-CoA
dehydrogenase in Escherichia coli. Gene 160, 263–267.
13 Ikeda Y, Keese SM, Fenton WA & Tanaka K (1987)
Biosynthesis of four rat liver mitochondrial acyl-CoA
dehydrogenases: in vitro synthesis, import into mito-
chondria, and processing of their precursors in a cell-

free system and in cultured cells. Arch Biochem Biophys
252, 662–674.
14 Volchenboum SL & Vockley J (2000) Mitochondrial
import and processing of wild type and type III mutant
isovaleryl-CoA dehydrogenase. J Biol Chem 275, 7958–
7963.
15 Mohsen AW, Navarette B & Vockley J (2001) Identifi-
cation of Caenorhabditis elegans isovaleryl-CoA
dehydrogenase and structural comparison with other
acyl-CoA dehydrogenases. Mol Genet Metab 73, 126–
137.
16 Kim JJ, Wang M & Paschke R (1993) Crystal structures
of medium-chain acyl-CoA dehydrogenase from pig
liver mitochondria with and without substrate. Proc
Natl Acad Sci USA 90, 7523–7527.
17 Ikeda Y & Tanaka K (1983) Purification and character-
ization of isovaleryl coenzyme A dehydrogenase from
rat liver mitochondria. J Biol Chem 258, 1077–1085.
18 Saijo T & Tanaka K (1995) Isoalloxazine ring of FAD
is required for the formation of the core in the Hsp60-
assisted folding of medium chain acyl-CoA dehydroge-
nase subunit into the assembly competent conformation
in mitochondria. J Biol Chem 270, 1899–1907.
19 Pedersen CB, Kølvraa S, Kølvraa A, Stenbroen V,
Kjeldsen M, Ensenauer R, Tein I, Matern D, Rinaldo
P, Vianey-Saban C et al. (2008) The ACADS gene vari-
ation spectrum in 114 patients with short-chain acyl-
CoA dehydrogenase (SCAD) deficiency is dominated by
missense variations leading to protein misfolding at the
cellular level. Hum Genet 124, 43–56.

20 Dahl DR (1968) Short chain fatty acid inhibition of rat
brain Na
+
–K
+
adenosine triphosphatasei. J Neurochem
15, 815–820.
21 Ribeiro CA, Balestro F, Grando V & Wajner M (2007)
Isovaleric acid reduces Na
+
,K
+
-ATPase activity in syn-
aptic membranes from cerebral cortex of young rats.
Cell Mol Neurobiol 27, 529–540.
22 Shimomura Y, Honda T, Shiraki M, Murakami T,
Sato J, Kobayashi H, Mawatari K, Obayashi M &
Harris RA (2006) Branched-chain amino acid
catabolism in exercise and liver disease. J Nutr 136,
250–253.
23 Meng Y, Katsuma S, Mita K & Shimada T (2009)
Abnormal red body coloration of the silkworm, Bomb-
yx mori, is caused by a mutation in a novel kynurenin-
ase. Genes Cells 14, 129–140.
24 Meng Y, Katsuma S, Daimon T, Banno Y, Uchino K,
Sezutsu H, Tamura T, Mita K & Shimada T (2009)
The silkworm mutant lemon (lemon lethal) is a potential
insect model for human sepiapterin reductase deficiency.
J Biol Chem 284, 11698–11705.
25 Maeda S (1989) Gene transfer vectors of a baculovirus,

Bombyx mori nucleopolyhedrovirus, and their use for
expression of foreign genes in insect cells. In Inverte-
brate Cell System Application (Mitsuhasi J ed), pp. 167–
181. CRC Press, Boca Raton, FL.
26 Mita K, Morimyo M, Okano K, Koike Y, Nohata J,
Kawasaki H, Kadono-Okuda K, Yamamoto K,
Suzuki MG, Shimada T et al. (2003) The construction
of an EST database for Bombyx mori and
its application. Proc Natl Acad Sci USA 100, 14121–
14126.
27 Sambrook J, Fritsch EF & Maniatis T (1989) Molecular
Cloning: A Laboratory Manual. Cold Spring Harbor
Laboratory Press, Cold Spring Harbor, NY.
28 Suzuki MG, Shimada T & Kobayashi M (1998)
Absence of dosage compensation at the transcription
level of a sex-linked gene in a female heterogametic
insect, Bombyx mori. Heredity 81, 275–283.
29 Daimon T, Hamada K, Mita K, Okano K, Suzuki MG,
Kobayashi M & Shimada T (2003) A Bombyx mori
gene, BmChi-h , encodes a protein homologous to bacte-
rial and baculovirus chitinases. Insect Biochem Mol Biol
33, 749–759.
30 Laemmli UK (1970) Cleavage of structural proteins
during assembly of head of bacteriophage-T4. Nature
227, 680–685.
31 Daimon T, Katsuma S, Iwanaga M, Kang W &
Shimada T (2005) The BmChi-h gene, a bacterial-type
chitinase gene of Bombyx mori, encodes a functional
exochitinase that plays a role in the chitin degradation
Odorous silkworm mutant K. Urano et al.

4462 FEBS Journal 277 (2010) 4452–4463 ª 2010 The Authors Journal compilation ª 2010 FEBS
during the molting process. Insect Biochem Mol Biol 35,
1112–1123.
32 Hall CL (1978) Acyl-CoA dehydrogenases and elec-
tron-transferring flavoprotein. Methods Enzymol 53,
502–518.
33 Ikeda Y, Dabrowski C & Tanaka K (1983) Separation
and properties of five distinct acyl-CoA dehydrogenases
from rat liver mitochondria. Identification of a new
2-methyl branched chain acyl-CoA dehydrogenase.
J Biol Chem 258, 1066–1076.
34 Altschul SF, Gish W, Miller W, Myers EW & Lipman
DJ (1990) Basic local alignment search tool. J Mol Biol
215, 403–410.
35 S
ˇ
ali A & Blundell TL (1993) Comparative protein
modeling by satisfaction of spatial restraints. J Mol Biol
234, 779–815.
36 Canutescu AA, Shelenkov AA & Dunbrack RL Jr
(2003) A graph-theory algorithm for rapid protein
side-chain prediction. Protein Sci 12, 2001–2014.
37 Lu
¨
thy R, Bowie JU & Eisenberg D (1992) Assessment
of protein models with three-dimensional profiles.
Nature 356, 83–85.
Supporting information
The following supplementary material is available:
Table S1. The sequences and its purposes of PCR

primers used in this study.
This supplementary material can be found in the
online version of this article.
Please note: As a service to our authors and readers,
this journal provides supporting information supplied
by the authors. Such materials are peer-reviewed and
may be re-organized for online delivery, but are not
copy-edited or typeset. Technical support issues arising
from supporting information (other than missing files)
should be addressed to the authors.
K. Urano et al. Odorous silkworm mutant
FEBS Journal 277 (2010) 4452–4463 ª 2010 The Authors Journal compilation ª 2010 FEBS 4463