Báo cáo khoa học: The use of recombinant protein and RNA interference approaches to study the reproductive functions of a gonad-stimulating hormone from the shrimp Metapenaeus ensis ppt

The use of recombinant protein and RNA interference
approaches to study the reproductive functions of a
gonad-stimulating hormone from the shrimp
Metapenaeus ensis
Shirley Hiu-Kwan Tiu and Siu-Ming Chan
Department of Zoology, The University of Hong Kong, China
Neurosecretory structures in crustacean eyestalks are
known to produce the crustacean hyperglycemic hor-
mone (CHH), molt-inhibiting hormone (MIH) and
gonad-inhibiting hormone (GIH) of the CHH ⁄ MIH ⁄
GIH gene family. These neuropeptides can regulate a
variety of physiologic processes, including molting,
carbohydrate metabolism, and reproduction [1–3].
GIH is one of the most studied neuropeptides of this
group because of its potential importance in shrimp
aquaculture. In penaeid shrimp, GIH is produced in
the X-organs and stored in the sinus glands of eye-
stalks [4–7]. Although the precise mechanism is not
known, GIH is postulated to inhibit reproduction by
suppressing ovary growth or vitellogenesis [1,2]. Eye-
stalk ablation removes the source of GIH and results
in ovary growth. In contrast, when eyestalk-ablated
females were injected with eyestalk extract, the gonad
stimulatory effect of eyestalk ablation was abolished
[1,2]. In addition to GIH, a factor found in the brain
and thoracic ganglion of decapod has been implicated
Keywords
eyestalk neuropeptide hormone; RNA
interference; shrimp; vitellogenin gene
Correspondence
S M. Chan, Department of Zoology, The

University of Hong Kong, Pokfulam Road,
Hong Kong
Fax: +852 2857 4672
Tel: +852 2299 0864
E-mail:
(Received 25 January 2007, revised 15 June
2007, accepted 2 July 2007)
doi:10.1111/j.1742-4658.2007.05968.x
Although the crustacean crustacean hyperglycemic hormone ⁄ molt-inhibi-
ting hormone ⁄ gonad-inhibiting hormone neuropeptides have been studied
extensively in the last two decades and several neuropeptides from the
shrimp Metapenaeus ensis have been cloned, the functions of most of these
neuropeptides remained putative. In this article, we describe the use of
recombinant protein and an RNA interference approach to study the
reproductive function of the previously reported molt-inhibiting hormone
(MeMIH-B) in M. ensis. When hepatopancreas and ovary explants were
cultured in medium containing recombinant MeMIH-B, the vitellogenin
gene (MeVg1) expression level was upregulated in a dose-dependent man-
ner, reaching a maximum in explants treated with 0.3 nm recombinant
MeMIH-B. Shrimp injected with recombinant MeMIH-B showed an
increase in vitellogenin gene expression in the hepatopancreas. Moreover, a
corresponding increase in the vitellogenin-like immunoreactive protein was
detected in the hemolymph and ovary of these females. Injection of
MeMIH-B dsRNA into the female shrimp caused a decrease in MeMIH-B
transcript level in thoracic ganglion and eyestalk. These shrimp also
showed reduction of vitellogenin gene expression in the hepatopancreas
and ovary. Furthermore, the hemolymph vitellogenin level was also
reduced in these animals. In summary, the results from recombinant
protein and RNA interference experiments have demonstrated the gonad-
stimulatory function of MeMIH-B in shrimp.

Abbreviations
CHH, crustacean hyperglycemic hormone; GIH, gonad-inhibiting hormone; GSI, gonadosomatic index; MeVg1, Metapenaeus ensis
vitellogenin gene 1; MIH, molt-inhibiting hormone; RNAi, RNA interference; si, small interfering.
FEBS Journal 274 (2007) 4385–4395 ª 2007 The Authors Journal compilation ª 2007 FEBS 4385
in the stimulation of gonad maturation. Injection of
protein extract from thoracic ganglion or the brain can
stimulate gonad maturation [8].
In sand shrimp Metapenaeus ensis, two forms of
MIH-like cDNA (i.e. MeMIH-A and MeMIH-B) have
been cloned and characterized [4,5]. MeMIH-B shows
only 68% amino acid s imil arity to M eMI H-A, and
amino acid sequence alignment ind icates that M eMIH-B
is more closely related to GIH of the lobster Homarus
americanus [9] than to the mandibular organ-inhibiting
hormones of the crab Cancer pagarus [10]. MeMIH-A
and MeMIH-B are non-sex-specific and are expressed
in the eyestalks of males and females. The expression
of MeMIH-A is molt-stage-related, whereas the expres-
sion of MeMIH-B is correlated with the reproductive
cycle. In addition to the eyestalk, MeMIH-B is also
expressed in the brain [4]. MeMIH-B transcript level is
low in the initial phase of gonad maturation and
increases towards the end of maturation [4]. These
findings suggest that the two neuropeptides should
have different functions. As they share relatively high
sequence similarity, cross-bioactivity also occurs for
these two neuropeptides [4]. For example, injection of
recombinant MeMIH-B also delays the process of
molting [4,11]. At the time when we had characterized
MeMIH-B, only a few CHH type II neuropeptides

were reported [2,3]. Despite its potential involvement
in reproduction, no further research on the reproduc-
tive function of MeMIH-B has been attempted, as
there is a lack of a good bioassay system for the neu-
ropeptide. The recent cloning and characterization of
the gene encoding the major yolk protein, vitellogenin,
may provide a potential biomarker for analysis of
genes that regulate ⁄ control reproduction [12].
Concurrently, the recently developed RNA interfer-
ence (RNAi) technique has been used to define the bio-
logical function of many genes. This technique is based
on the gene-silencing effect of dsRNA [13]. The tech-
nique has revolutionized ‘reverse genetic’ research by
introducing dsRNA to organisms or cells. dsRNA can
knock down a gene and will produce a phenotypic loss
of function of that gene [14–16]. Although the com-
plete mechanism has yet to be revealed, successful
RNAi has been reported for many animal models. For
example, Caenorhabditis elegans can be soaked in
dsRNA or can be fed plasmids that make dsRNA and
consequently exhibit RNAi effects. In many studies,
dsRNA can move across cell boundaries freely. Thus,
it is not necessary to inject dsRNA directly into the
gonad to get progeny that exhibit RNAi effects [13].
As RNAi works in many organisms, it might also
work in shrimp. Gene function analysis by RNAi may
be advantageous as compared to other conventional
approaches. This article describes the production of
recombinant protein and dsRNA for reproduction-
related eyestalk neuropeptide gene, and use of an

in vitro explant culture system and an RNAi technique
to demonstrate the reproductive function of MeMIH-B
in M. ensis.
Results
Expression of MeMIH-B in shrimp
Although we have previously studied the tissue distri-
bution of MeMIH-B in the female shrimp, the expres-
sion pattern of MeMIH-B in the central nervous
system of different reproductive stages has not
been fully investigated. Moreover, to ascertain that
MeMIH-B expression pattern is correlated with repro-
ductive developmental stages in females, we have rein-
vestigated the expression pattern of MeMIH-B in the
eyestalks and other nervous tissues of the adult females
by northern blot analysis. MeMIH-B transcripts could
be detected in the eyestalk, nerve cord, thoracic gan-
glion and brain of shrimp at early to middle stages of
gonad maturation (Fig. 1A). In female eyestalks,
MeMIH-B transcript level was low in immature
shrimp with low gonadosomatic index (i.e. GSI < 2).
As gonad development was in progress, a steady
increase in MeMIH-B transcript level was observed.
Similarly, the expression pattern of MeMIH-B in the
thoracic ganglia also followed that of the eyestalk
(Fig. 1B). For example, in both eyestalk and thoracic
ganglion, the highest MeMIH-B transcript level was
recorded at the late maturation stage in shrimp with
GSI ¼ 10. Similar to the previous results, expression
of MeMIH-B is sex-nonspecific, as the males also
expressed MeMIH-B (Fig. 1B).

Functional study of recombinant MeMIH-B
in vitro and in vivo
The rMeMIH-B produced by pRSET expression was
purified on an Ni
2+
-charged column. To study the
function of rMeMIH-B in reproduction, hepatopan-
creas explants from females at early stage of gonad
maturation (GSI < 2) were used. A dose-dependent
increase of MeVg1 expression was recorded when the
concentration of rMeMIH-B was increased (i.e.
0.3 pm,3pm and 30 pm). The maximum increase of
MeVg1 transcript level was observed in the hepatopan-
creas explants treated with 0.3 nm rMeMIH-B; further
increase of rMeMIH-B (i.e. 3 nm,30nm and 300 nm)
resulted in a decrease in the overall MeVg1 expression
level (Fig. 2A). When the ovary explants were treated
Functional study of crustacean neuropeptide S. H K. Tiu and S M. Chan
4386 FEBS Journal 274 (2007) 4385–4395 ª 2007 The Authors Journal compilation ª 2007 FEBS
with 0.3 nm rMeMIH-B, an increase of about 25% of
MeVg1 expression was recorded (Fig. 2B).
Next, we performed an in vivo injection of
rMeMIH-B into females to further confirm its gonad-
stimulatory effect. To demonstrate the specificity of
rMeMIH-B in gonad maturation, a control group
injected with rMeMIH-A was included. As compared
to the NaCl ⁄ P
i
control, injection of an equal amount
(6.6 nmol) of rMeMIH-A did not cause any change in

the overall expression of vitellogenin in the hepatopan-
creas and ovary (Fig. 3A,B). In contrast, injection
of 6.6 nmol of rMeMIH-B stimulated an increase
(2–3-fold) in MeVg1 expression by the hepatopancreas
and ovary (Figs 3A,B) at 72 h, but only weakly for the
24 h time point (data not shown).
It is well accepted that the vitellogenin produced in
the hepatopancreas serves as an extraovarian source
for the final synthesis of vitellin. The newly made vitel-
logenin is expected to be secreted rapidly into the
hemolymph and transported to the ovary for oocyte
uptake. To demonstrate that the increase in expression
of the MeVg1 gene could also result in the appearance
of vitellogenin in the hemolymph for transport, we
also collected hemolymph samples of these injected
shrimp and analyzed the increase in vitellogenin-spe-
cific protein. As shown in Fig. 3C,D, when females
were injected with rMeMIH-B (i.e. 6.6 nmol), the
hemolymph and ovaries of most animals contained a
much higher level of vitellogenin (i.e. 148 kDa)
(Fig. 3C, left panel). These vitellogenin-specific pro-
teins are presumably derived from the translation of
the MeVg1 gene from the hepatopancreas after
rMeMIH-B stimulation. The results from SDS ⁄ PAGE
and western blot analysis of the hemolymph and
ovarian proteins from shrimp injected with 6.6 nmol of
rMeMIH-B demonstrated an increase in the overall
Es Br Tg Vn Hp Mu Ov
noisserpxe evitaleR
1.0

A
B
0.75
0.5
0.25
%2
%4
%
6
%7
%9
M
1.0
0.75
0.5
0.25
noisserpxe evitaleR
Es MeMIH-B
Tg MeMIH-B
Fig. 1. Expression of MeMIH-B in different tissues of early
(GSI < 2) mature females (N ¼ 5). (A) The relative expression levels
of MeMIH-B in nervous tissues (ES, eyestalk; Br, brain; Tg, thoracic
ganglia; Vn, ventral nerve) and non-nervous tissues (Hp, hepatopan-
creas; Mu, muscle; Ov, ovary); the bar indicates the SE. (B) The
expression pattern of MeMIH-B (N > 20) at different gonad matura-
tion stages of the eyestalk (open bar) and thoracic ganglia (diago-
nally shaded bars) of females. The percentage indicates the GSI of
the females. M (B) indicates the expression pattern of MIH-B in
the same tissues in males (N ¼ 5). The lower panel is the northern
blot analysis of MeMIH-B expression in the eyestalk (Es) and tho-

racic ganglia during the gonad maturation cycle. Each lane repre-
sents an RNA sample from the eyestalk or the thoracic ganglion of
one shrimp. The last lane shows the RNA samples from a male.
The bar indicates the SE.
Relative expressionRelative expression
1.4
B
A
1.2
1.0
0.8
0.6
0.4
0.2
0
Concentrations of rMeMIH-B
Concentrations of rMeMIH-B
*
*
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0
Fig. 2. Histogram showing the relative expression levels of MeVg1
in (A) hepatopancreas and (B) ovary explants after exposure to dif-
ferent concentrations (i.e. from 0.3 p

M to 0.3 lM) of rMeMIH-B.
The sample size (or numbers of shrimp) is 10 for the in vitro assay.
Relative MeVg1 mRNA levels (A) are shown as means + SEM of
10 prawns. The shrimp that show significant differences (P<0.05)
in the relative MeVg1 mRNA levels are indicated by an asterisk.
S. H K. Tiu and S M. Chan Functional study of crustacean neuropeptide
FEBS Journal 274 (2007) 4385–4395 ª 2007 The Authors Journal compilation ª 2007 FEBS 4387
0
0.5
1
1.5
2
2.5
3
3.5
4
A
B
C
D
treatment
0
0.5
1
1.5
2
2.5
3
rMeMIH-BCtrl rMeMIH-A
Ctrl rMeMIH-A rMeMIH-B

0
2
4
6
8
12
10
0
0.5
1
1.5
2
2.5
3
Ctrl rMeMIH-A rMeMIH-B
Ctrl rMeMIH-A rMeMIH-B
148 kDa
Hycn
148 kDa
NSP
MeVg1
rRNA
MeVg1
rRNA
Me MIH-Actrl Me MIH-B
treatment
Me MIH-Actrl
Me MIH-B
treatment
Me MIH-Actrl

Me MIH-B
treatment
Me MIH-Actrl
Me MIH-B
Fold changes of Vg transcript
level in hepatopancreas
Fold changes of Vg transcript
level in ovary
Fold changes of Vg
level in hemolymph
Fold changes of Vg
content in ovary
Fig. 3. Effect of recombinant MeMIH-B and MeMIH-A on vitellogenin expression in shrimp. (A) Left: relative expression levels of MeVg1 in
hepatopancreas of females (N ¼ 10) at 48 h after injecting NaCl ⁄ P
i
, rMIH-A and rMIH-B. Right: a typical northern blot analysis of the shrimp
MeVg1 transcript level after injection of NaCl ⁄ P
i
, rMIH-A, and rMIH-B. (B) Left: relative expression levels of MeVg1 in ovary of females
(N ¼ 10) at 48 h after injection of NaCl ⁄ P
i
, rMIH-A, and rMIH-B. Right: a typical northern blot analysis of the shrimp MeVg1 transcript level
after injection of NaCl ⁄ P
i
, rMIH-A, and rMIH-B. (C) Left: relative levels of vitellogenin in hemolymph of females (N ¼ 10) at 48 h after injec-
tion of NaCl ⁄ P
i
, rMIH-A, and rMIH-B. Right: western blot analysis (upper) of the hemolymph level of vitellogenin for shrimp injected with
rMIH-B. The 148 kDa protein is one of the vitellogenin subunits recognized by the shrimp antibody to vitellogenin [27]. The lower panel
shows the Coomassie blue staining of the hemocyanin (Hcy) corresponding to the same protein samples. (D) Left: relative levels of vitelloge-

nin in ovary of shrimp at 48 h after injection of NaCl ⁄ P
i
, MIH-A, and rMIH-B. Right: western blot detection (upper) of vitellogenin (148 kDa)
in ovary of shrimp injected with rMIH-B. NSP is the nonspecific protein unrelated to vitellogenin of the ovary samples. In the northern blot
(or western blot) analysis, each lane represents RNA (or protein) samples collected from individual shrimps.
Functional study of crustacean neuropeptide S. H K. Tiu and S M. Chan
4388 FEBS Journal 274 (2007) 4385–4395 ª 2007 The Authors Journal compilation ª 2007 FEBS
vitellogenin-specific protein (Fig. 3C,D). Unlike the
rMeMIH-B-injected group, shrimp injected with
rMeMIH-A (6.6 nmol) did not show any changes in
the overall MeVg1 transcript level in the hepatopan-
creas or a significant increase in MeVg1 protein level
in the hemolymph and ovary (Fig. 3A–D).
Inhibition of vitellogenin expression after RNAi
We have performed preliminary experiments using a
nonspecific dsRNA (from Tiger frog virus), and the
results show no effect on MeMIH-B gene silencing
(data not shown). In the following study, individual
shrimp (N ¼ 40; average GSI < 3) were injected with
3 lg of dsRNA for MeMIH-B, and RNA samples were
collected after 24, 48, 72, 96 and 120 h. Northern blot
results from the eyestalk (Fig. 4A) indicated no signifi-
cant reduction in MeMIH-B transcript level at all time
points. However, when we used RT-PCR to analyze
the same samples, a significant reduction of MeMIH-B
transcript was observed (Fig. 4B). In fact, by RT-PCR,
the MeMIH-B dsRNA appeared to knock down most
of the transcripts after 72 h of treatment (Fig. 4B). In
addition, hybridization signals representing small-size
RNAs were strong and persisted from 24 to 120 h after

injection (Fig. 4A). This suggests that dsRNAs are very
stable, as residual MeMIH-B dsRNA remained. Unlike
in the eyestalk, there was a significant decrease in the
MeMIH-B transcript level in the nerve cord as early as
24 h after MeMIH-B dsRNA injection. The knock-
down also persisted 120 h after dsMIH-B injection
(Fig. 5A). MeMIH-B transcript level was lowest in
nerve cord at 72 h after injection, but started to
increase afterwards (Fig. 5B). With regard to the effect
of MeMIH-B dsRNA on hepatopancreas MeVg1
expression, it was observed that there was a significant
drop in MeVg1 transcript level in the hepatopancreas.
For example, at 24, 48 and 72 h after dsRNA treat-
ment, drops of 20%, 71% and 23% of the overall
MeVg1 transcript level were recorded. (Fig. 6A).
Unlike in the hepatopancreas, the reduction of MeVg1
expression in ovaries of these female was small after
the injection of MeMIH-B dsRNA. For example, the
reduction in MeVg1 transcript level in the ovary repre-
sented only 6%, 7% and 22% decreases at 24, 48 and
72 h post-dsRNA treatment (Fig. 6B).
Similar SDS ⁄ PAGE analysis and western blot analy-
sis were performed for these females. In the hemolymph
sample of the NaCl ⁄ P
i
-injected control, vitellogenin-spe-
cific protein could be detected using antibody to vitel-
logenin. In contrast, no vitellogenin-specific protein was
detected in the hemolymph of the dsRNA-injected
females (Fig. 7A). In the ovary, the amount of vitelloge-

nin remained relatively constant. However, only minute
quantities of vitellogenin subunits (i.e. 148, 97 and
78 kDa) were detected in the ovaries of the dsRNA-
injected females (Fig. 7B). These proteins were immuno-
reactive to the antibody to vitellogenin of M. ensis [27].
B-HIMeM
β nitca-
0 24 48 72 96 120
0 24487296120
+-+-+-+-+-Ctr
120967248240
Relative change in
transcript level
100
80
60
40
20
Time after injection (h)
Time after injection (h)
AB
Fig. 4. Effects of MeMIH-B RNAi in eyestalk of female shrimp. (A) Northern blot detection of eyestalk MeMIH-B transcript level in control (–)
and dsRNA-injected (+) females from animals at different time intervals (i.e. 0, 24, 48, 72, 96 and 120 h); the arrow indicates the MIH-B
transcript, and the smear indicates the residual dsRNA. (B) Top panel: RT-PCR detection of MeMIH-B gene knockdown using MIH-B-specific
primers. Lower panel: relative change in MIH-B transcript level at different time intervals. The bar diagram indicates the relative transcript level
of MeMIH-B after normalization with b-actin gene.
S. H K. Tiu and S M. Chan Functional study of crustacean neuropeptide
FEBS Journal 274 (2007) 4385–4395 ª 2007 The Authors Journal compilation ª 2007 FEBS 4389
Discussion
Structure–function research on crustacean eyestalk

CHH family neuropeptides remains a challenging
endeavor. This is mainly due to the existence of highly
similar neuropeptides in the same species [1,2]. For
example, it is now known that there are at least four
or five CHH-like genes in M. ensis. These genes may
share high sequence similarity and ⁄ or analogous
function. Some of these genes may be expressed in
0 h 24 h 48 h 72 h 96 h 120 h
MIH-B
β
-actin
+-+-+-+-+-Ctr
120967248240
Relative change in
transcript level
100
80
60
40
20
120967248240
Time after injection
Time after injection (h)
AB
Fig. 5. Effects of MeMIH-B RNAi in the ventral nerve cord of female shrimp. (A) Northern blot detection of eyestalk MeMIH-B transcript
level in control (–) and dsRNA-injected (+) females from animals at different time intervals; the arrow indicates the MIH-B transcript, and the
smear indicates the residual dsRNA. (B) RT-PCR detection of MeMIH-B transcript using specific primers. The bar diagram indicates the rela-
tive transcript level of MeMIH-B after normalization with b-actin gene.
1gVeM
ANRr

72h48h24h
72h48h24h
1g
V
eM
ANR
r
yravoninoisserpxe1gVeM
% decrease in MeVg I
transcript level
% decrease in MeVg
I
transcript level
30
20
10
724824 724824
100
80
60
40
20
+-+-+-
+-+-+-
MeVg1 expression in hepatopancreas
Time (h) after injection Time (h) after injection
AB
Fig. 6. Expression of vitellogenin in hepatopancreas and ovary of dsRNA-injected (+) and control (–) females. (A) Upper: northern blot detec-
tion of hepatopancreas MeVg1 transcript level in shrimp. Lower: bar indicates the relative decrease in expression level of MeVg1. (B) Upper:
RT-PCR detection of ovary MeVg1 transcript level in shrimp. Lower: bar indicates the relative decrease in expression level of MeVg1.

Functional study of crustacean neuropeptide S. H K. Tiu and S M. Chan
4390 FEBS Journal 274 (2007) 4385–4395 ª 2007 The Authors Journal compilation ª 2007 FEBS
non-neuronal (or non-eyestalk) tissues [3]. Likewise,
additional MIH subtypes have also been found in one
species. For example, in addition to the known MIH
and GIH subtypes, there is also a report suggesting the
existence of novel MIH subtypes. In the GenBank data-
base ( there are at least
four different entries for MIH-like neuropeptides in the
tiger prawn Penaeus monodon. As there is a high degree
of sequence similarity and structural conservation of
these genes and gene products, confirmation of the true
identity and function of these neuropeptides remains
difficult. Unfortunately, most of these MIH-like
peptides are produced in very low quantities, and they
function as inhibitory regulators in a physiologic
process, making it difficult to develop a good bioassay.
The production of large quantities of recombinant
protein for bioassay may circumvent the lack of active
material for structure–function studies. However, the
inhibitory nature of these hormones remains a challenge
for the successful development of a bioassay system.
Attempts at developing biological assays for inhibi-
tory factors such as MIH have been reported, and the
inhibitory function on molting has been demonstrated
convincingly. Most biological assays of the gonad inhi-
bition of this neuropeptide rely on its ability to inhibit
ovary development and reduce oocyte size, cause a
decrease in total ovary protein incorporation, and sup-
press ovary total protein synthesis [11,17,18]. Biologi-

cal assays using these criteria are nonspecific and
provide little information on the mechanism of GIH
regulation of reproduction. Previously, we have pro-
duced rMeMIH-B (formerly MeeMIH-B), but little
progress was made in developing a biological assay for
the recombinant protein. This is mainly attributed to
the lack of a biomarker for the reproductive process.
With the recent cloning of the vitellogenin gene in dif-
ferent crustaceans [19–23], a more precise role for GIH
can be defined with the vitellogenin as a biomarker.
Recently, there was a report on the effect of sinus
gland extract and neuropeptide on vitellogenin gene
expression in M. japonicus. In that study, the effect of
a CHH peptide and two MIH-like peptides on ovary
vitellogenin gene expression was investigated; the
results indicated that CHH causes inhibition, whereas
the MIH-like neuropeptides have no effect on vitello-
genin gene epression [23]. Unlike the CHH of M. japo-
nicus, MeMIH-B (a type II neuropeptide) has a
stimulatory effect on vitellogenin synthesis. Taken
together, the results suggest that a CHH-like (or type I)
neuropeptide may be inhibitory for gonad maturation,
321M65
4
detcejni-ANRsddetcejni-SBP
321M654
detcejni-ANRsdde
tcejni-SBP
yravOh
p

mylom
AB
eH
gni
n
ia
ts
BCgniniatsBC
tolbnretseWtolbnretseW
321M654
321M
6
54
aD
k
841
aDk79
aD
k6
7
a
D
k
84
1
aD
k
79
a
D

k67
Fig. 7. Western blot analysis of the hemolymph and ovary total protein of dsRNA-injected females. (A) Hemolymph sample of NaCl ⁄ P
i
-
injected and dsRNA-injected females. Lanes 1–3: NaCl ⁄ P
i
-injected and 4–6 dsRNA-injected animals. Individual lanes represent protein sam-
ples collected from injected shrimps, and the arrows indicate the vitellogenin-specific protein (148, 97 and 76 kDa) using antibody to vitellog-
enin [19,27]. (B) Ovary sample of NaCl ⁄ P
i
-injected and dsRNA-injected females. Lanes 1–3: Ovary from the corresponding NaCl ⁄ P
i
-injected
individual and 4–6 dsRNA-injected animals. Each lane represents protein samples collected from individuals, and the arrow indicates the vitel-
logenin-specific protein determined using antibody to vitellogenin [19].
S. H K. Tiu and S M. Chan Functional study of crustacean neuropeptide
FEBS Journal 274 (2007) 4385–4395 ª 2007 The Authors Journal compilation ª 2007 FEBS 4391
whereas the MIH-like neuropeptide (i.e MeMIH-B,
type II) may be a vitellogenin-stimulatory factor of
shrimp. Because of the existence of multiple forms of
the CHH family neuropeptides, there may be a dis-
crepancy in defining the function of many cDNAs
cloned using a molecular biology approach. For
example, although the MIH-like gene of Litopenaeus
vannamei was reported [24], detailed analysis revealed
that the MIH-like deduced protein was more closely
related to the CHH as described in other crustaceans.
Since our report of a second form of MIH subtype
cDNA in M. ensis, the naming of this MeMIH-B as a
GIH was based on its similarity to the lobster GIH, as

other MIH subtypes in shrimp had not been reported
[4]. In re-evaluating the expression pattern of
MeMIH-B in the eyestalk and CNS, we further recon-
firmed that its expression level increases during active
vitellogenesis. Our result suggests that a high transcript
level of the MeMIH-B gene (or the protein) may
be needed for vitellogenin production during active
vitellogenesis.
The use of recombinant protein to study the func-
tion of crustacean neuropeptides has been reported
for a number of species [4,11,19,25]. Recombinant
MeMIH-B has been produced and used in a biologi-
cal assay for molt inhibition [4]. Except for the role
of CHH in the increase of glycemia, the use of
recombinant protein to study molt-inhibiting function
and gonad maturation remains a challenge for crusta-
cean endocrinologists [2,4]. The difficulty is due to the
lack of a good biological marker for consistent
results. As the hepatopancreas and ovary express
MeVg1, explants from these tissues were used in the
explant culture. Given the fact that the hepatopan-
creas culture lasts for 3–4 h and active vitellogenin
expression can be detected in shrimp, the explant
culture system was successfully developed. This is
the first demonstration of the stimulatory effects of a
neuropeptide in vitellogenin gene stimulation. As
rMeMIH-B can stimulate MeVg1 expression in hepa-
topancreas and ovary in vitro, the result may provide
information on the initial mechanism of hormone
action. In other words, the in vitro results indicate

that MeMIH-B may act directly on the hepatopan-
creas and ovary to increase the rate of vitellogenin
gene expression, indicating that both the hepatopan-
creas and ovary are the targets of MeMIH-B. This
result will provide the basis for identifying and
characterizing the receptor for the neuropeptide.
Furthermore, rMeMIH-B acted on the hepatopan-
creas and ovary in a dose-dependent manner. As the
optimal concentration (i.e. 30 nm in vitro and
6.6 nmol in vivo) for the stimulatation of MeVg1
expression is low, the result also suggests that
rMeMIH-B is highly potent in stimulating vitellogenin
gene expression. As subadult (i.e. < 15 g) and adult
females also responded to rMeMIH-B in a similar
dose-dependent manner, the results would be useful
for us to develop a strategy to induce gonad matura-
tion in shrimp aquaculture.
RNAi is defined as the gene-silencing effect medi-
ated by dsRNA. RNAi technology was developed in
the mid-1990s, based on the antisense RNA techno-
logy developed in the 1980s. RNAi can silence or
knock down the expression of a gene, and the phe-
nomenon appears to be universal, as it has been
reported in both plants, animals, and even cultured
cells. There are two major types of RNAi, with slight
differences in the mechanism. They are mediated by
either: (a) dsRNA; or (b) small interference (si)RNA.
The longer dsRNA may generate a large population
of siRNA (with 21–23 nucleotides), and the use of
longer dsRNA may be advantageous over siRNA. In

this study, the longer dsRNA was produced and used
in the RNAi experiments. It has been reported that
the longer dsRNA of approximately 600–800 nucleo-
tides works best for most genes. The GIH-specific
dsRNA, however, was synthesized from the 238 bp
coding region of the mature peptide. As the coding
sequences of all the neuropeptides are short (< 350
nucleotides) and there are scattered repetitive
sequences in the noncoding region of MeMIH-B,
selection of the effective gene region to produce
dsRNA is limited and restricted only to the coding
region. As the siRNA produced by the endogeneous
Dicer (assuming a mechanism similar to the verte-
brates) is small, the siRNA has to be specific to cause
an effect. For example, the RNAi will not work even
with a 1–2 bp mismatch. Thus, another highly similar
gene (MeMIH-A) will not be affected. This may
explain why there are hybridization signals in the eye-
stalk, as the eyestalk is also known to produce
MeMIH-A. The apparent lack of MeMIH-B knock-
down (northern blot result) may simply indicate that
another very similar but abundant neuropeptide (i.e.
MeMIH-A) may hybridize to the MeMIH-B probe.
When the more specific RT-PCR was used, the
decrease in MeMIH-B transcript level was evident.
The amount of dsRNA injected in the animal may
vary, depending on the expression level of the gene; a
much higher dose of dsRNA was injected into the
shrimp L. schmitti to silence the CHH gene [26]. In
our study, the amount of dsRNA injected into each

animal was about 3–5 lg for each shrimp (23–28 g).
At present, the mechanism of RNAi in shrimp is not
known, but we expected that these dsRNA molecules
Functional study of crustacean neuropeptide S. H K. Tiu and S M. Chan
4392 FEBS Journal 274 (2007) 4385–4395 ª 2007 The Authors Journal compilation ª 2007 FEBS
could circulate by way of hemolymph and would be
taken up by a variety of tissues. In the target tissues
that express MeMIH-B (i.e. neuronal cells of the eye-
stalk and ⁄ or central nervous system), the dsRNA
enters the cell to initiate gene knockdown. The
dsRNA appears to be stable in the target tissues. For
example, a strong hybridization signal representing the
residual MeMIH-B dsRNA still remains in the eye-
stalk at 120 h after injection (Figs 4A and 5A).
In conclusion, the use of recombinant protein or
RNAi alone may not be sufficient to confirm the func-
tion of a neuropeptide. The combined use of recombi-
nant protein and RNAi described in this study has
provided unequivocal evidence for a stimulatory func-
tion of MeMIH-B in vitellogenesis. We can also apply
similar approaches to study the structure–function
relationships of other CHH ⁄ MIH ⁄ GIH neuropeptide
members.
Experimental procedures
Animals
Shrimp were purchased from a local seafood market. They
were acclimated in the laboratory at 25–28 °C in an indoor
aquarium for 2 days before rMeMIH-B or dsRNA injec-
tion. The GSI was calculated as the percentage of ovary
weight per total body weight.

Production of the recombinant MeMIH-B
The cDNA encoding the mature peptide of the MeMIH-B
was amplified by PCR using restriction enzyme-linked
gene-specific primers (forward, 5¢-GACGAATTCTTCGG
CCTTCGC-3¢; reverse, 5¢-AGGAGATCTAAGCTTACCA
CGCTCCACCAGGG-3¢) that contained restriction sites
EcoRI and BamHI, respectively. The PCR product was first
digested with EcoRI and HindIII, and was then ligated into
the cloning vector pRSET-B containing a T7 lac promoter
site (Invitrogen, Carlsbad, CA, USA). The constructs were
transformed into Escherichia coli XL1 Blue cells, and the
bacteria were grown at 37 °C. The plasmids were purified
by alkaline lysis DNA minipreparation. The insertion of
the desired gene in the plasmid was verified by automated
DNA sequencing. The clones were transformed into E. coli
BL21 ⁄ DE3. For large-scale expression, bacterial cells were
grown in 1 L of LB broth at 37 °C. Protein expression was
induced by the addition of isopropyl 1-thio-b-d-galactopyr-
anoside (Sigma, St Louis, MO, USA) to a final concentra-
tion of 1 mm. The culture was allowed to continue for 4 h.
The bacteria were pelleted by centrifugation (5000 g for 15
min) and resuspended in 50 mL of binding buffer (20 mm
Tris ⁄ HCl, 0.5 m NaCl, and 5 mm imidazole, pH 7.9). The
cells were homogenized with a polytron, and centrifuged at
5000 g for 15 min. The pellet was then resupended in
50 mL of denaturing binding buffer (8 m urea in binding
buffer), sonicated, and agitated overnight at 4 °C. Follow-
ing centrifugation as above, the supernatant was collected
and loaded into an Ni
2+

–nitrilotriacetic acid–agarose
(Qiagen, Hilden, Germany) affinity column that was pre-
equilibrated with denaturing binding buffer. The column
was then washed three times with washing buffer (8 m urea,
20 mm Tris ⁄ HCl, 0.5 m NaCl, and 25 mm imidazole,
pH 7.9). The fusion protein was eluted with elution buffer
(6 mm Tris ⁄ HCl, 0.5 m NaCl, 8 m urea, and 300 mm imid-
azole, pH 7.9). The denatured recombinant protein was
refolded by both dilution and dialysis. The concentration of
urea present in the solubilized protein was decreased step-
wise by addition of an equal volume of renaturing buffer
(6 mm Tris ⁄ HCl, 0.5 m NaCl, and 300 mm imidazole,
pH 7.9) for every 3 h until the concentration of urea was
decreased to 1 m. The diluted recombinant protein was then
dialyzed in a dialysis bag (Sigma; cut-off 6–7 kDa) in a
large volume of renaturing buffer for 16 h at 4 °C, with
three changes of buffer. The recombinant protein was
further dialyzed in a large volume of 0.1 · NaCl ⁄ Tris over-
night, and the final dialysis was against 0.1 · NaCl ⁄ P
i
over-
night. A Bradford protein assay (Bio-Rad, Hercules, CA,
USA) was performed to determine the concentration of the
refolded protein. The recombinant MeMIH-A was
expressed using the same strategy and was used as negative
control in the following in vitro and in vivo bioassay.
Functional study of rMeMIH-B by explant assay
and in vivo injection
The functional study of rMIH-B involving a shrimp in vitro
explant culture system was based on a previously developed

method [20–22]. Briefly, hepatopancreas and ovary were
dissected, cut into small fragments, and placed in the wells
of 24-well plate containing 1.5 mL of Medium 199 (Sigma)
prepared in crab saline [28]. Different concentrations of
rMeMIH-B were added to the explants, and the culture
was incubated at 23–25 °C for 4 h. At the end of the
culture period, the tissues were collected for total RNA
extraction followed by northern blot hybridization [25] or
RT-PCR. For northern blot analysis, the nylon membrane
was hybridized in hybridization buffer (50% formamide)
containing a nonradioactive (digoxigenin) probe at 50 °C
overnight. The probe (derived from partial MeVg1 cDNA)
was synthesized as per kit instructions (Roche, Mannheim,
Germany). After hybridization, the membrane was washed
twice in 2 · NaCl ⁄ Cit (20 · NaCl ⁄ Cit: 3 m NaCl, 0.3 m
Na
3
-citrate) and 0.1% SDS for 15 min, and then washed
twice in 0.5 · NaCl ⁄ Cit and 0.1% SDS at 58 °C for
15 min. The signals were detected by adding the antidigoxi-
genin–AP conjugate. Chemiluminescent substrate CDP-Star
(Roche, Mannheim, Germany) was added, and the mem-
brane was exposed to X-ray film. For PCR, the PCR mix
S. H K. Tiu and S M. Chan Functional study of crustacean neuropeptide
FEBS Journal 274 (2007) 4385–4395 ª 2007 The Authors Journal compilation ª 2007 FEBS 4393
(20 lL) consisted of 10 mm Tris ⁄ HCl (pH 8.0), 1.5 mm
MgCl
2
,50mm KCl, 2 mm dNTP, and 10 pmol of forward
and reverse primers. The PCR conditions were one cycle of

3 min at 95 °C, followed by 35 cycles of denaturation at
95 °C for 1 min, annealing at 58 °C for 1 min, and exten-
sion at 72 ° C for 1 min. In the last cycle, the PCR product
was incubated at 72 °C for 10 min to allow the completion
of DNA synthesis. PCR products were analyzed on a 1.5%
agarose gel, and Southern Blot was performed to determine
the specific amplification of the cDNA.
For in vivo injection of rMIH-B, adult females in the
nonreproductive stage were injected with 20 lL of either
6.6 nmol or 0.66 nmol of rMIH-B at the arthropodial
membrane of the periopod and returned to the culture
tanks. At 24, 48 and 72 h after injection, the hepatopan-
creas and ovary of the shrimp were dissected for total
RNA preparation, and the hemolymph samples were col-
lected for SDS ⁄ PAGE and western blot analysis.
Functional study of MeMIH-B by RNAi
To prepare a DNA template for the synthesis of dsRNA,
DNA corresponding to the mature peptide of MeMIH-B
was amplified by PCR using T7 promoter-linked primers
(forward, 5¢-TAATACGACTCACTATAGGTACTATG
TATCGCATGCCAAT-3¢; reverse, 5¢-TAATACGACTC
ACTATAGGTACTTTAAAGTCCCGGGTTGA-3¢). For
PCR, the reaction mix consisted of 1 · Taq buffer contain-
ing 1.5 mm MgCl
2
, 0.2 mm dNTP mix, 0.5 lm MIH
T
7
-linked primers, and 0.25 lLofTaq DNA polymerase
(Life Technologies, Carlsbad, CA, USA). PCR conditions

were denaturation at 95 °C for 1 min, annealing at 55 °C
for 1 min and extension at 72 ° C for 1.5 min for the first
five cycles, annealing at 60 °C for another 30 cycles, and
extension at 72 °C for 10 min. The resulting PCR products
were analyzed by 2% agarose gel electrophoresis. The tar-
get PCR product band was purified with the GENE-
CLEAN II Kit (Biogene, Vista, CA, USA). For
transcription of MeMIH-B dsRNA, 1–2 lg of purified T7
promoter-linked MeMIH-B DNA was used as template in
an in vitro transcription reaction with the T7 Megascript
RNAi Kit (Ambion, Austin, TX, USA), according to the
manufacturer’s recommendations. In 20 lL of reaction mix-
ture, MeMIH-B DNA template was mixed with appropriate
amounts of nuclease-free water, 2 lLof10· T7 reaction
buffer, 2 lL of ATP solution, 2 lL of CTP solution, 2 lL
of GTP solution, 2 lL of UTP solution, and 2 lLof
T7 Enzyme Mix. The mixture was incubated at 37 °C for
18 h. During the transcription, the two RNA strands were
hybridized to form dsRNA.
For RNAi experiments, shrimp (25–35 g) with similar
GSI values were purchased from a local seafood market.
They were acclimated in a culture tank overnight prior to
injection. Then, 50 lL of MeMIH-B dsRNA (3 lgin
1 · NaCl ⁄ P
i
) was injected into shrimp through the arthro-
dial membrane of a periopod. The controls received an equal
volume of NaCl ⁄ P
i
injection. Shrimps were returned to the

tanks for culture before being killed for total RNA prepara-
tion from different tissues. The relative level of MeMIH-B
expression in the nerve cord was used as an indication of the
RNAi effect between the treatment and control groups.
Statistical analysis of northern blots and western
blots
Northern blot or western blot signals from the films (or
blots) were scanned with the free software imagej (Image
processing and analysis in Java: />to obtain quantitative numbers representing expression lev-
els of the gene or protein. The expression level was normal-
ized with either the rRNA (for the northern blot) or
hemocyanin (for the western blot). Either simple t-test or
anova was used to perform statistical analysis.
Acknowledgements
This research was supported by the Research Grant
Council of the Hong Kong Special Administrative
Region, China (HKU 7214 ⁄ 05M) awarded to S M.
Chan.
References
1 Keller R (1992) Crustacean neuropeptides: structures,
functions and comparative aspects. Experientia 48, 439–
448.
2 De Kleijn DP & Van Herp F (1995) Molecular biology
of neurohormone precursors in the eyestalk of Crusta-
cea. Comp Biochem Physiol B Biochem Mol Biol 112,
573–579.
3 Chan SM, Gu PL, Chu KH & Tobe SS (2003) Crusta-
cean neuropeptide genes of the CHH ⁄ MIH ⁄ GIH family:
implications from molecular studies. Gen Comp Endo-
crinol 134, 214–219.

4 Gu PL, Tobe SS, Chow BK, Chu KH, He JG & Chan
SM (2002) Characterization of an additional molt inhibi-
ting hormone-like neuropeptide from the shrimp
Metapenaeus ensis. Peptides 23, 1875–1883.
5 Gu PL & Chan SM (1998) Cloning of a cDNA encod-
ing a putative molt-inhibiting hormone from the eye-
stalk of the sand shrimp Metapenaeus ensis. Mol Mar
Biol Biotechnol 7, 214–220.
6 Gu PL & Chan S-M (1998) The shrimp hyperglycemic
hormone-like neuropeptide is encoded by multiple
copies of genes arranged in a cluster. FEBS Lett 441,
397–403.
7 Gu PL, Yu KL & Chan SM (2000) Molecular charac-
terization of an additional shrimp hyperglycemic hor-
mone: cDNA cloning, gene organization, expression
Functional study of crustacean neuropeptide S. H K. Tiu and S M. Chan
4394 FEBS Journal 274 (2007) 4385–4395 ª 2007 The Authors Journal compilation ª 2007 FEBS
and biological assay of recombinant proteins. FEBS
Lett 472, 122–128.
8 Fingerman M (1997) Roles of neurotransmitters in regu-
lating reproductive hormone release and gonadal matu-
ration in decapod crustaceans. Invertebr Reprod Dev 31,
47–54.
9 de Kleijn DP, Janssen KP, Waddy SL, Hegeman R, Lai
WY, Martens GJ & Van Herp F (1998) Expression of
the crustacean hyperglycaemic hormones and the
gonad-inhibiting hormone during the reproductive cycle
of the female American lobster Homarus americanus.
J Endocrinol 156, 291–298.
10 Lu W, Wainwright G, Webster SG, Rees HH & Turner

PC (2000) Clustering of mandibular organ-inhibiting
hormone and moult-inhibiting hormone genes in the
crab, Cancer pagurus, and implications for regulation of
expression. Gene 253, 197–207.
11 Gu PL, Chu KH & Chan SM (2001) Bacterial
expression of the shrimp molt-inhibiting hormone
(MIH): antibody production, immunocytochemical
study and biological assay. Cell Tissue Res 303,
129–136.
12 Tsang WS, Quackenbush LS, Chow BK, Tiu SH, He
JG & Chan SM (2003) Organization of the shrimp vitel-
logenin gene: evidence of multiple genes and tissue spe-
cific expression by the ovary and hepatopancreas. Gene
303, 99–109.
13 Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE
& Mello CC (1998) Potent and specific genetic interfer-
ence by double-stranded RNA in Caenorhabditis
elegans. Nature 391, 806–811.
14 Amdam GV, Simoes ZL, Guidugli KR, Norberg K &
Omholt SW (2003) Disruption of vitellogenin gene func-
tion in adult honeybees by intra-abdominal injection of
double-stranded RNA. BMC Biotechnol 3, 1–8.
15 Cogoni C & Macino G (2000) Post-transcriptional gene
silencing across kingdoms. Genes Dev 10, 638–643.
16 Blandin S, Moita LF, Kocher T, Wilm M, Kafatos FC
& Levashina EA (2002) Reverse genetics in mosquito
Anopheles gambiae: targeted disruption of the defensin
gene. EMBO Rep 9, 852–856.
17 Ohira T, Tsutsui N, Nagasawa H & Wilder MN (2006)
Preparation of two recombinant crustacean hyperglyce-

mic hormones from the giant freshwater prawn,
Macrobrachium rosenbergii, and their hyperglycemic
activities. Zool Sci 23, 383–391.
18 Ohira T, Okumura T, Suzuki M, Yajima Y, Tsutsui N,
Wilder MN & Nagasawa H (2006) Production and
characterization of recombinant vitellogenesis-inhibiting
hormone from the American lobster Homarus americ-
anus. Peptides 27, 1251–1258.
19 Tiu SH, Hui JH, He JG, Tobe SS & Chan S-M (2006)
Characterization of vitellogenin in the shrimp Metape-
naeus ensis: expression studies and hormonal regulation
of MeVg1 transcription in vitro. Mol Reprod Dev 73,
424–436.
20 Mak AS, Choi CL, Tiu SH, Hui JH, He JG, Tobe SS,
Chan SM (2005) Vitellogenesis in the red crab
Charybdis feriatus: hepatopancreas-specific expression
and farnesoic acid stimulation of vitellogenin gene
expression. Mol Reprod Dev 70, 288–300.
21 Tsutsui N, Saido-Sakanaka H, Yang WJ, Jayasankar V,
Jasmani S, Okuno A, Ohira T, Okumura T, Aida K &
Wilder MN (2004) Molecular characterization of a
cDNA encoding vitellogenin in the coonstriped shrimp,
Pandalus hypsinotus and site of vitellogenin mRNA
expression. J Exp Zool A Comp Exp Biol 301, 802–814.
22 Tiu SHK, Hui JHL, Mak ASC, He J & Chan S-M
(2006) Equal contribution of hepatopancreas and ovary
to the production of vitellogenin (PmVg1) transcripts in
the tiger shrimp, Penaeus monodon. Aquaculture 254,
666–674.
23 Tsutsui N, Katayama H, Ohira T, Nagawasawa H,

Wilder M & Katsumi A (2005) The effects of crustacean
hyperglycemic hormone-family peptides on vitellogenin
gene expression in the kuruma prawn, Marsupenaeus
japonicus. Gen Comp Endocrinol 144, 232–239.
24 Sun PS (1995) Expression of the molt-inhibiting hor-
mone-like gene in the eyestalk and brain of the white
shrimp Penaeus vannamei. Mol Mar Biol Biotechnol 4,
262–268.
25 Maniatis T, Sambrook J & Fritsch EF (1989) Molecular
Cloning: a Laboratory Manual. Cold Spring Harbor
Laboratory Press, Cold Spring Harbor, NY.
26 Lugo JM, Morera Y, Rodriguez T, Huberman A,
Ramos L & Estrada MP (2006) Molecular cloning and
characterization of the crustacean hyperglycemic hor-
mone cDNA from Litopenaeus schmitti. Functional
analysis by double-stranded RNA interference tech-
nique. FEBS J 273, 5669–5677.
27 Kung SY, Chan SM, Hui JH, Tsang WS, Mak A, He
JG (2004) Vitellogenesis in the sand shrimp, Metapena-
eus ensis: the contribution from the hepatopancreas-
specific vitellogenin gene (MeVg2). Biol Reprod 71,
863–870.
28 Duan S & Cooke IM (1999) Selective inhibition of tran-
sient K
+
current by La
3+
in crab peptide-secretory
neurons. J Neurophysiol 81, 1848–1855.
S. H K. Tiu and S M. Chan Functional study of crustacean neuropeptide

FEBS Journal 274 (2007) 4385–4395 ª 2007 The Authors Journal compilation ª 2007 FEBS 4395