Molecular Human Reproduction Vol.7, No.11 pp. 1039–1046, 2001
Expression of a green fluorescent protein variant in mouse
oocytes by injection of RNA with an added long poly(A)
tail
Takuya Aida1,2, Shoji Oda1, Takeo Awaji1, Koyo Yoshida2 and Shunichi Miyazaki1,3,4
1Department
of Physiology, Tokyo Women’s Medical University School of Medicine, 8-1 Kawada-cho, Shinjuku-ku, Tokyo,
162-8666, 2Department of Obstetrics and Gynecology, Juntendo University School of Medicine, Bunkyo-ku, Tokyo, 113-8241 and
3Laboratory of Intracellular Metabolism, Department of Molecular Physiology, National Institute for Physiological Sciences, Okazaki,
444-8585, Japan
4To
whom correspondence should be addressed. E-mail: [email protected]
In oocytes, cytoplasmic 3⬘ polyadenylation regulates translational activation of dormant mRNA during meiotic
maturation. Thus exogenous proteins are hardly expressed after injection of conventional RNA. To circumvent this,
we synthesized a long polyadenylated (~250 A) tail to encode RNA with an enhanced yellow fluorescent protein
targeted to mitochondria (EYFP-mito), and injected it into mouse oocytes at the germinal vesicle (GV) stage. From
this transcript, EYFP-mito was clearly expressed in ~80% of oocytes, while scarce expression from a transcript
with only 30 A was observed. In strongly expressing oocytes, fluorescence was detected within 1–3 h after RNA
injection, increased linearly up to 12 h, and reached a maximum at 12–15 h. The distribution of EYFP-mito
matched the staining of mitochondria in these oocytes. About 80% of these oocytes underwent GV breakdown and
60% matured in vitro, comparable to non-expressing or non-RNA-injected oocytes. Some of the oocytes which
strongly expressed EYFP-mito remained at the GV stage. Thus, the expression was not always accompanied by
meiotic maturation, nor did it suppress the maturation process. Mature oocytes expressing EYFP-mito possessed
normal fertilizability associated with intracellular Ca2⫹ oscillations, and developed into 2-cell embryos. Thus,
polyadenylated RNA is a useful tool applicable to the expression of EYFP-fused functional proteins or of indicator
protein probes for studies of mammalian fertilization.
Key words: gene expression/GFP variant/mouse oocyte/polyadenylated RNA/mitochondria-targeted protein
Introduction
The artificial expression of a defined protein fused with green
fluorescent protein (GFP) is a powerful technique that enables
us to visualize the localization and trafficking of the protein
in living cells or to monitor concentration changes of second
messengers that interact with the protein probe. The technique
has been applied in a wide variety of cells. However, this
strategy is often difficult in oocytes, particularly in mammalian
oocytes, at meiotic maturation, since the translation of mRNA
is suppressed for the majority of proteins (Davidson, 1986)
other than those necessary during maturation such as cyclin
B1 (Polanski et al., 1998; Tay et al., 2000) and Mos (Gebauer
et al., 1994). It has recently been found that, for the proteins
that are expressed in maturing mouse oocytes, the 3⬘ poly(A)
tail of their mRNA is extended to 100–150 nucleotides
following induction of meiotic maturation (Richter, 1999). It
has also been revealed that the cytoplasmic polyadenylation
is necessary for translational activation of dormant mRNA and
is conferred by the cytoplasmic polyadenylation element and
© European Society of Human Reproduction and Embryology
polyadenylation hexanucleotide AAUAAA, both of which are
located in the distal portion of the 3⬘ untranslated regions of
responding mRNAs (Vassalli et al., 1989; Richter, 1999).
Vassalli et al. (1989) have shown that exogenous plasminogen
activator can be expressed in mouse oocytes, when its RNA
is polyadenylated in vitro by adding 200–400 A and then
injected into oocytes at the germinal vesicle (GV) stage.
In the present study, we applied this in-vitro polyadenylation
method for expression of a GFP variant, enhanced yellow
fluorescent protein (EYFP), in mouse oocytes, as a step towards
further application of EYFP-fused functional proteins or indicator protein probes. RNA of mitochondria-targeted EYFP was
used, because EYFP expression could be clearly determined
by the characteristic distribution of mitochondria. Fluorescence
of EYFP is stronger than that of GFP at a pH near 8 in the
mitochondrial matrix (Llopis et al., 1998). We show here
that the method was successful, without disturbing oocyte
maturation in vitro, fertilizabilty, and Ca2⫹ responses at fertilization.
1039
T.Aida et al
Materials and methods
Plasmid construction
A plasmid encoding EYFP fused with the mitochondria-targeting
sequence from subunit VIII of human cytochrome c oxidase, ‘pEYFPMito’ was used (Clontech, Palo Alto, CA, USA). The cDNA encoding
mitochondria-targeted EYFP (EYFP-mito) in this plasmid was amplified by polymerase chain reaction (PCR) using the following primers;
5⬘-CGC GGT ACC GCC ACC ATG TCC GTC CTG ACG CCG
CTG C-3⬘ and 5⬘-TCG AGA TCG CGG CCG CTA CTT GTA CAG
CTC GTC CAT GC-3⬘. Fragments of EYFP-mito cDNA were digested
with KpnI and NotI and ligated to the KpnI and NotI sites of pBluescript
KS⫹ (Stratagene, La Jolla, CA, USA). For non-mitochondria-targeted
EYFP, pEYFP-Mito was digested with BamHI and NotI, to remove
the mitochondria-targetting sequence, and the released fragments
were introduced into the BamHI and NotI sites of pBluescript KS⫹.
RNA and polyadenylation
The constructed plasmids, containing cDNA encoding EYFP-mito or
EYFP, were digested with SacI and NotI and the resulting fragments
were used as templates for in-vitro transcription. RNA were synthesized by means of T3 polymerase using T3 MessageMachine Kit
(Ambion, Austin, TX, USA). Synthesized RNA were treated with
phenol–chloroform before precipitation with ethanol, and dried RNA
were resolved in TE. RNA was polyadenylated by incubation for
30 min at 37°C in the presence of 100 µmol/l ATP and 30 IU/ml
poly(A) polymerase (Gibco BRL, Rockville, MD, USA) in buffer A
containing (mmol/l) 250 NaCl, 50 Tris–Cl (pH 8.1), 10 MgCl2, 2
dithiothreitol, 1000 IU/ml RNasin (Promega, Madison, WI, USA),
and 100 µg/ml BSA (Vassalli et al., 1989). The RNA was resolved
in 150 mmol/l KCl solution (final concentration, ~1 µg RNA/µl). To
avoid folding of the RNA, the sample was heated at 90°C for 1 min
and then cooled on ice. The polyadenylated RNA was designated as
‘EYFP-mito-polyA’ or ‘EYFP-polyA’. The numbers of nucleotides
in EYFP-mito-polyA and EYFP-mito (no polyA) RNA fragments
were ~1150 and 900 respectively, examined by polyacrylamide gel
electrophoresis. The length of added poly(A) tract was therefore
estimated to be ~250 A. RNA of EYFP-mito added with 30 A
(‘EYFP-mito-short polyA’) was also injected into some oocytes. To
obtain the template for transcription, a fragment containing T3
promoter and EYFP-mito cDNA was amplified by PCR using the
following primers: 5⬘-GGA AAC AGC TAT GAC CAT G-3⬘ and 5⬘TTT TTT TTT TTT TTT TTT TTT TTT TTT TTT CTA TCT AGA
CTT GTA CAG CTC G-3⬘. In-vitro transcription was performed with
the T3 polymerase and the synthesized RNA was designated as
‘EYFP-mito-short poly A’.
Oocytes and injection of RNA
B6D2F1 female mice (6–10 weeks old) were injected intraperitioneally
with 5 IU pregnant mare’s serum gonadotrophin (Teikoku Hormone
Mfg., Tokyo, Japan), and immature oocytes at the GV stage were
collected from mature follicles 48–52 h later. M2 medium (Fulton
and Whittingham, 1978) supplemented with BSA (4 mg/ml) was used
for cell manipulation. Oocytes with cumulus cells only partially
attached were selected and inserted with a micropipette for injection
with 5–10 pl of EYFP-mito-polyA- or EYFP-polyA-containing solution (1 µg/µl), to produce a concentration of 25–50 ng/µl RNA in
the oocyte (assuming an oocyte volume of 200 pl). In each experiment,
5–15 oocytes were injected within 45 min after oocyte collection.
After washing, the RNA-injected oocytes were transferred to a glassbottomed dish and cultured in M16 medium (Whittingham, 1971) at
37°C under 5% CO2 in air for 18–20 h, during which time ~60% of
1040
the oocytes spontaneously matured. Cumulus cells were detached
from the oocytes.
Measurement of EYFP fluorescence
Fluorescence of EYFP expressed in oocytes was investigated under
an inverted fluorescence microscope (TMD; Nikon, Tokyo, Japan) at
a given time after RNA injection, using excitation light passed
through a 470–490 nm bandpass filter. Emission was passed through
a 510 nm dichroic mirror, a 520–560 nm bandpass filter, and a
silicon intensifier target camera (C2400-08; Hamamatsu Photonics,
Hamamatsu, Japan). Images were acquired by an image processor
(Arugas 50; Hamamatsu Photonics). The fluorescence intensity (F)
was averaged in an oocyte and presented as an arbitrary unit of
values between 0 and 255. The relationship between F and the
arbitrary unit was calibrated using fluorescein (Sigma, St Louis, MO,
USA). The optical system and detector were set at the same conditions
throughout the present study.
Mitochondria staining and confocal microscopy
Some RNA-injected oocytes were stained with the mitochondriaspecific fluorescent dye MitoTracker Orange CM-H2TMRos
(Molecular Probes Inc., Eugene, OR, USA) by incubation in the
presence of 0.5 µmol/l MitoTracker for 30 min at 37°C. The
distribution patterns of EYFP-mito and MitoTracker were investigated
in the same cells using a confocal laser scanning microscope (CLSM;
LSM310, Carl Zeiss, Oberkochen, Germany) equipped with an
Achroplan ⫻40 objective water immersion lens (NA 0.75, Carl Zeiss),
as described previously (Shiraishi et al., 1995). EYFP and MitoTracker
were excited by 488 nm and 543 nm laser, respectively, and each
emitted fluorescence was collected through a 510–525 nm bandpass
filter or a 570 nm longpass filter, respectively.
Insemination of zona-free oocytes and Ca2⫹ measurement
RNA-injected oocytes were inseminated after maturation. The oocytes
were freed from the zona pellucida, since the cumulus cells had
detached and cumulus-free zona-intact mouse oocytes are known to
be less fertilizable (Bleil, 1993). To prevent hardening of the zona
and make it removable, 5% fetal bovine serum was added to M16
medium for in-vitro maturation (Downs et al., 1986). The zonae
of mature oocytes were partially dissolved by applying 0.001%
chymotrypsin for several minutes (Boldt and Wolf, 1986) and removed
mechanically by pipetting in acidic Tyrode solution (Nakano et al.,
1997). Spermatozoa obtained from the caudae epididymides of
B6D2F1 male mice (11–15 weeks old) were incubated in M16
medium for more than 2 h at 37°C under 5% CO2 in air. Oocytes
were inseminated by adding a small amount of the sperm suspension
to the oocyte-containing medium.
Prior to insemination, some oocytes were loaded with the Ca2⫹sensitive fluorescent dye fura-2 by incubation in the presence of
5 µmol/l fura-2 AM (acetoxymethyl ester; Molecular Probes) for
8 min at 37°C. Increases in intracellular Ca2⫹ concentration ([Ca2⫹]i
during fertilization were recorded by a conventional Ca2⫹ imaging
method at 20 s intervals using an image processor (Arugas 50;
Hamamatsu Photonics), for details of the method and calibration of
[Ca2⫹]i), (Nakano et al., 1997). Fluorescence of fura-2 was obtained
without interference with that of EYFP, by applying 340 and 380 nm
UV lights alternatively and by passing emission light through a 500–
520 nm bandpass filter.
Fertilization of oocytes was identified after 6–8 h culture in M16
medium by extrusion of the second polar body and formation of the
male and female pronuclei, using differential interference contrast
optics (LSM310, Carl Zeiss). Two-cell embryos were observed
24–28 h after fertilization.
Expression of GFP variant in mouse oocytes
Table I. Expression of EYFP-mito in RNA-injected oocytes
RNA injected
EYFP-mito-polyA
EYFP-mito
No. of
experiments
12
8
No. of oocytes
injected
136
63
No. of oocytes
survived (%)a
95 (70)
44 (70)
No. of expressing oocytes (%)b
Clearc
Weakd
None
79 (83)
0 (0)
12 (13)
0 (0)
4 (4)
44 (100)
aExamined
bExamined
1 h after RNA injection.
20 h after RNA injection.
cClearly expressing oocytes (F ⬎ 10).
dWeakly expressing oocytes (2 ⬍ F 艋 10).
eNon-expressing oocytes (F ⫽ 2).
EYFP-mito ⫽ enhanced yellow fluorescent protein targeted to mitochondria.
Figure 1. The fluorescence intensity (F) of oocytes injected with
EYFP-mito RNA without a poly(A) tail, with a short poly(A) tail
(30 A), or with a long poly(A) tail (~250 A). F averaged in the
entire oocyte was measured at 20 h after RNA injection and
presented as mean ⫾ SD. The numbers in the parentheses indicate
the numbers of oocytes examined. EYFP-mito ⫽ enhanced yellow
fluorescent protein targeted to mitochondria.
Results
Expression of EYFP-mito in oocytes
Table I shows the rate of expression of EYFP-mito examined
20 h after RNA injection. Oocytes that shrank within 1 h after
injection were discarded (30%). The fluorescence intensity (F)
of all oocytes injected with non-polyadenylated RNA was the
same level as autofluorescence seen in non-RNA-injected
oocytes (Figure 1), indicating that EYFP-mito was not
expressed. In contrast, EYFP-mito was clearly expressed in
83% of oocytes (79 of 95) injected with EYFP-mito-polyA
(Table I), as judged by F ⬎ 10 (mean ⫾ SD, 63 ⫾ 35; Figure
1). Another 12 oocytes in which F was between 2 and 10
were considered as ‘weakly’ expressing oocytes, since F did
not increase after the initial 3 h (see later section for details).
With EYFP-mito-short polyA (30 A), fluorescence was detected
in some oocytes, but F was 5 ⫾ 3 (n ⫽ 18; Figure 1) in the
range of weakly expressing oocytes. Thus, the addition of a
long poly(A) tail to RNA is much more efficient for the
expression of EYFP-mito.
Distribution of EYFP-mito in oocytes
Figure 2A shows that the distribution of EYFP-mito fluorescence (green image at 488 nm excitation) in oocytes was
almost consistent with MitoTraker-stained structures (red image
at 543 nm excitation) and was composed of many granular
spots of various sizes at the equatorial section of the oocyte
examined using confocal microscopy ~20 h after RNA injection. No interference by the two excitation lights was involved
in these images, since no fluorescence was detected in the
oocyte injected with EYFP-mito-polyA but not treated with
MitoTracker when examined with 543 nm light (Figure 2B,
right) nor in a non-injected oocyte stained with MitoTracker
and examined under 488 nm excitation (Figure 2C, left).
In the oocytes injected with polyadenylated RNA of nonmitochondria-targeted EYFP, fluorescence was homogeneous
in the entire oocyte without granular structures (Figure 2D) as
the produced EYFP were located in the cytosol. These results
indicate that EYFP-mito was properly targeted to mitochondria
in the oocytes.
The distribution of EYFP-mito appears to be different
between the green images of Figure 2A and B, despite of the
same conditions for image acquisition and presentation. This
may result from a non-homogeneous distribution of mitochondria in mouse oocytes (Van Blerkom and Runner, 1984;
Calarco, 1995), observation at different optical sections and/
or differences in expression. When Figure 2A is precisely
inspected, small faint spots of MitoTracker are visible over
the oocyte in the red image, whereas corresponding small
spots are hardly recognized in the green image. Production of
EYFP-mito is likely to be lower in this oocyte, compared with
the oocyte of Figure 2B which exhibits more abundant granular
structures and brighter background fluorescence. The oocyte
of Figure 2B seems to contain excessively expressed EYFPmito molecules that are not incorporated into mitochondria.
Time course of EYFP-mito expression
Fluorescence intensity was measured successively for 20 h
after RNA injection in individual oocytes under the same
conditions (Figure 3A). F was proportional to fluorescein
concentration, up to the arbitrary unit of 230 (Figure 3C). F
was unaltered for non-polyA EYFP-mito or only slightly
increased for EYFP-mito-short polyA (dotted or broken line
in Figure 3B, respectively). Figure 3A is an example showing
fluorescence changes in individual oocytes injected with EYFP1041
T.Aida et al
Figure 2. The distribution of EYFP fluorescence (green image with
488 nm excitation) and MitoTracker (red image with 543 nm
excitation) in mature oocytes. (A) An oocyte injected with EYFPmito-polyA and stained with MitoTracker. (B) An oocyte injected
with EYFP-mito-polyA but not stained with MitoTracker. (C) An
oocyte stained with MitoTracker but not injected with RNA.
(D) An oocyte injected with EYFP-polyA (not targeted to
mitochondria). EYFP-mito ⫽ enhanced yellow fluorescent protein
targeted to mitochondria.
mito-polyA (n ⫽ 19; three experiments on the same day).
Except in two cases, substantial fluorescence was detected as
early as 1 h after RNA injection, and a significant increase in
F (⬎10) was recognized at 3 h. The succeeding increase in F
was variable among oocytes. In four oocytes, F stayed at
relatively low levels between 20 and 30 after 6 h and up to
20 h. In other oocytes, F increased to ⬎50 in a linear fashion
up to 12 h, and then showed variation of a slight decrease or
increase among oocytes during the next 8 h. Figure 3B is an
average profile of changes in F of these ‘strongly’ expressing
1042
Figure 3. The time course of expression of EYFP-mito.
(A) Changes in fluorescence intensity in individual oocytes injected
with EYFP-mito-polyA at time zero. (B) Solid line: the profile of
changes in F of ‘strongly’ expressing oocytes, presented by mean
⫾ SD of 13 oocytes in which the peak fluorescence intensity
exceeded 50 in (A). Broken and dotted lines: averages of
fluorescence intensity in 18 oocytes injected with EYFP-mito-short
polyA and 18 oocytes injected with non-polyadenylated EYFP-mito
respectively. (C) The relationship between fluorescein concentration
and the arbitrary F unit used. EYFP-mito ⫽ enhanced yellow
fluorescent protein targeted to mitochondria.
oocytes, showing that F linearly increased during 12 h and
reached an almost maximum level at 12 or 15 h.
Figure 4 shows the distribution of EYFP-mito at four stages
Expression of GFP variant in mouse oocytes
Table II. In-vitro maturation of RNA-injected oocytes
RNA injected
EYFP-mito-polyA
EYFP-mito
None
No. of oocytes (%)
Survived
GVa
GVBDb
PB1c
95
44
33
20 (21)
9 (20)
7 (21)
18 (19)
9 (20)
7 (21)
57 (60)
26 (60)
19 (58)
a–cExamined
20 h after RNA injection.
oocytes arrested at the germinal vesicle (GV) stage.
oocytes that underwent germinal vesicle breakdown (GVBD)
but did not form the first polar body.
cMatured oocytes with the extruded first polar body (PB1).
EYFP-mito ⫽ enhanced yellow fluorescent protein targeted to mitochondria.
aImmature
bNon-matured
Figure 4. The distribution of EYFP-mito fluorescence at different
times after RNA injection (four different oocytes). (A) An oocyte
in which the germinal vesicle was still visible (2 h after RNA
injection). (B) An oocyte after germinal vesicle breakdown (6.5 h).
(C) An oocyte immediately before the formation of the first polar
body (12 h; arrow indicates the extruding polar body). (D) A
mature oocyte (18 h). Scale bar ⫽ 20 µm. EYFP-mito ⫽ enhanced
yellow fluorescent protein targeted to mitochondria.
of meiotic maturation. The oocyte in which the GV was still
visible (2 h after RNA injection) showed small patches of
EYFP-mito surrounding the GV and small spots in the peripheral region (Figure 4A), consistent with the distribution of
mitochondria (Van Blerkom and Runner, 1984; Calarco, 1995).
Thus, the trafficking of EYFP-mito into mitochondria occurred
before GV breakdown (GVBD) was completed. After GVBD
(6.5 h), a dense mass of EYFP-mito was localized in the
perinuclear region (Figure 4B). In the oocyte immediately
before formation of the first polar body (PB1) (12 h), EYFPmito was aggregated around a dark region away from the
extruding PB1 (Figure 4C), as with mitochondria which
surround the second metaphase spindle region (Van Blerkom
and Runner, 1984; Calarco, 1995). Mitochondria disperse over
the cytoplasm in mature oocytes, but have a denser distribution
in the central region (Van Blerkom and Runner, 1984; Calarco,
1995). In the mature oocyte, EYFP-mito was densely located
in the central region of the mature oocyte (Figure 4D). Thus,
EYFP-mito molecules were precisely localized in accordance
with the distribution changes of mitochondria during oocyte
maturation.
In-vitro maturation of oocytes
About 80% of fluorescent oocytes (75 of 95) injected with
EYFP-mito-polyA resumed first meiosis indicated by GVBD,
and 60% underwent maturation, judged by extrusion of the
PB1 (Table II). These percentages were consistent with those
in non-EYFP-expressing oocytes injected with non-polyA
EYFP or not injected at all (Table II).
To examine any correlation between the magnitude of EYFP-
Figure 5. Values of F in individual ‘clearly’ expressing oocytes
which were classified into three groups according to maturation
stage. F and the maturation stage were examined at 20 h after
injection of EYFP-mito-polyA. GV ⫽ oocytes arrested at the
germinal vesicle stage; GVBD ⫽ oocytes which underwent
germinal vesicle breakdown but did not form PB1; PB1 ⫽
oocytes having extruded the first polar body.
mito expression and the advance of oocyte maturation, F and
the maturation stage of clearly expressing oocytes examined
at 20 h after RNA injection were plotted in Figure 5. Oocytes
were divided into three groups: oocytes that were arrested at
the GV stage even at 20 h, oocytes that underwent GVBD but
did not form the PB1, and oocytes that were matured. The
values of F were scattered in a wide range in each group of
oocytes, and there was no correlation between F and oocyte
maturation. EYFP-mito was well expressed even in some
oocytes arrested at the GV stage. Oocytes were matured even
when EYFP-mito was strongly expressed.
Fertilization, Ca2⫹ oscillations and first cleavage
The fertilizability of EYFP-mito expressing oocytes was also
examined. Mammalian oocytes exhibit repetitive increases in
[Ca2⫹]i called ‘Ca2⫹ oscillations’ at fertilization (Miyazaki
et al., 1993), and these are responsible for cortical granule
exocytosis and resumption of the second meiosis (Kline and
Kline, 1992). Ca2⫹ oscillations were recorded in strongly
1043
T.Aida et al
expressing oocytes (Figure 6), having the temporal pattern
similar to that seen at normal fertilization of oocytes matured
in vivo (Deguchi et al., 2000).
The resumption of the second meiosis and completion of
fertilization were determined by formation of the second polar
body (PB2) and male and female pronuclei (PN ⫹ PB2)
6–8 h after insemination. About 90% of injected and inseminated oocytes (20 of 22) formed PN ⫹ PB2, and this was
comparable to oocytes matured in vitro without RNA injection
(Table III). The first cleavage was examined in another series
of experiments. The rate of formation of normal 2-cell embryos
was 75% in control oocytes (Table III), probably because
polyspermy tended to occur in zona-free oocytes, although
diluted sperm suspension was used. The rate of cleavage was
67% in EYFP-expressing oocytes, almost comparable to that
in the control oocytes.
The first cleavage in these oocytes occurred 26–28 h after
insemination, whereas it was completed within 24 h in control
experiments. The period tended to be more delayed in the
oocytes with brighter fluorescence. Thus, expression of EYFPmito did not affect the fertilizability of oocytes, although
excessive expression might slightly affect the efficiency of
development into the 2-cell embryo.
Figure 6. Ca2⫹ oscillations at fertilization of a mature oocyte in
which EYFP-mito was expressed. [Ca2⫹]i in the whole oocyte was
measured every 20 s by conventional Ca2⫹ imaging with fura-2.
The measurement was started when a spermatozoon attached to the
surface of the zona-free oocyte (zero time). EYFP-mito ⫽
enhanced yellow fluorescent protein targeted to mitochondria.
Discussion
The expression of GFP variant-fused proteins in mouse oocytes
is a potential tool for analysing mechanisms involved in oocyte
maturation, fertilization, and early embryonic development.
The present study demonstrated that mitochondria-targeted
EYFP is efficiently expressed by injection of RNA with an
added long poly(A) tail. The number of added A (~250)
was in the range (200–400) reported for experiments with
plasminogen activator RNA (Vassalli et al., 1989). The dormant
mRNA in oocytes have short poly(A) tails fewer than 20 A
(Richter, 1999). We confirmed that RNA having 30 A induced
only weak expression. The expression of exogenous proteins
in mouse oocytes by injecting RNA has been performed to
introduce the human m1 muscarinic receptor (Williams et al.,
1992, 1998; Moore et al., 1993) and human platelet-derived
growth factor β receptor (Mehlmann et al., 1998), but polyadenylation of RNA was not described. In these cases, a small
amount of expressed receptor proteins might be sufficient to
examine the involvement of G proteins in agonist-induced
Ca2⫹ release. Polyadenylated RNA have, however, been used
for expression of cyclin B1 (Polanski et al., 1998) or luciferase/
cyclin B1 (Barkoff et al., 2000) in oocytes. For visualization
of GFP variants, a higher expression level is thought to be
required, and this was provided by using polyadenylated RNA.
The majority of isolated immature mouse oocytes undergo
spontaneous maturation. In previous experiments, RNA have
been injected into oocytes arrested at the GV stage in the
presence of dibutyryl cAMP (Vassalli et al., 1989) or
3-isobutyl-1-methylxanthine (IBMX) (Williams et al., 1992).
RNA-injected oocytes have also been kept in IBMX-containing
medium for several hours (Moore et al., 1993) to give a longer
time before maturation for RNA translation. The present
method provided maximal expression of EYFP-mito without
such treatments. EYFP-mito was strongly expressed both in
maturing oocytes and in some oocytes that failed to undergo
GVBD. It is conceivable therefore that RNA translation
advances independently of meiotic maturation, when polyadenylated RNA is injected.
EYFP-mito became detectable as early as 1–3 h after RNA
injection, and its trafficking into mitochondria advanced soon
after production of the protein. Thus, the targeting sequence
from subunit VIII of human cytochrome c oxidase (Rizzuto
et al., 1989) was effective in the mouse oocyte. The expression
of EYFP-mito increased with time, associated with incorporation into mitochondria. In strongly expressing oocytes, the
fluorescence intensity increased in an approximately linear
Table III. Fertilization and first cleavage of EYFP-mito-expressing oocytes
RNA injected
EYFP-mito-polyA
None
No. (%) of oocytes
No. (%) of embryos
No. of
experiments
Inseminated
PN ⫹ PB2
No. of
experiments
Inseminated
2-cell
stage
5
2
52
26
48 (92)
23 (88)
3
2
36
25
24 (67)
18 (75)
EYFP-mito ⫽ enhanced yellow fluorescent protein targeted to mitochondria; PN ⫽ male and female pronuclei; PB2 ⫽ second polar body.
1044
Expression of GFP variant in mouse oocytes
manner and attained a maximal level in 12 or 15 h. The time
course is favourable for studies using expressed GFP variants
in mature oocytes.
EYFP-mito was clearly expressed in ~80% of oocytes.
However, the magnitude of expression indicated by the fluorescence intensity was widely variable among individual
oocytes. Possible factors that affect the magnitude of expression
due to RNA activity include the length of poly(A) tail, the
amount of RNA injected, the cellular activity, which could
depend partly on the grade of cell growth of the oocytes
collected from ovaries, and the injection technique which is
likely to cause cell damage. In our imaging system, the
brightness of EYFP fluorescence corresponding to F values
of 20–50 would probably be sufficient for identifying the
localization of EYFP-fused proteins or for monitoring concentration changes of second messengers using EYFP-fused
probes. A fluorescence intensity above that level may indicate
excessive expression with EYFP-mito remaining in the cytosol,
not incorporated into mitochondria, as seen in Figure 2B.
Regulated expression, by controlling the experimental conditions described above, might therefore be required. It should
be noted that the present method provides such high levels of
EYFP expression that it could be adjustable to various purposes.
The biological inspection of the RNA-injected oocytes
showed that percentages of GVBD and maturation in vitro
were about 80 and 60%, respectively, comparable to that in
non-RNA-injected oocytes. The culture condition could be
improved in future experiments, to obtain higher yields.
Mature oocytes had normal fertilizability associated with Ca2⫹
oscillations leading to oocyte activation. The PB2 and male and
female pronuclei were formed normally in 90% of inseminated
oocytes. Fertilized oocytes developed into 2-cell embryos,
although the first cleavage was a little delayed, compared with
control oocytes. Since development to the 2-cell stage takes 1
day after activation of the oocyte at fertilization, excessive
expression of EYFP in mitochondria could affect the process
via some effects on cellular metabolism.
The present method could be applicable to the expression,
in mouse oocytes, of GFP variant-fused functional proteins,
for example: cell cycle-related proteins such as cyclin (Polanski
et al., 1998; Tay et al., 2000) and Mos (Gebauer et al., 1994);
fertilization-related proteins such as α6β1 integrin (Almeida
et al., 1995), Cd9 (Kaji et al., 2000), and Ca2⫹ oscillationinducing factor derived from spermatozoa (if cloned in the
future) (Parrington et al., 2000); proteins involved in the
meiotic and mitotic apparatus (Wianny et al., 1998), or signal
proteins such as membrane receptors (Mehlmann et al., 1998;
Williams et al., 1998), G proteins, and phospholipase C. It is
also applicable to indicator probes for second messengers such
as Ca2⫹ (Miyawaki et al., 1999), cyclic AMP (Zaccolo et al.,
2000), or inositol 1,4,5-trisphosphate (Hirose et al., 1999), as
have been developed in somatic cells. The present study
provides a step forward for such technology in mammalian
oocytes.
Acknowledgments
We thank Dr H.Shirakawa and Dr Z.Kohchi for discussion and
comments and Mr T.Shikano, and Mr Y.Konuma for their technical
assistance. This work was supported by Grant-in-Aid for General
Scientific Research (B) to S.M. and for Encouragement of Young
Scientists to S.O. from the Japan Ministry of Education, Science,
Sports, and Culture.
References
Almeida, E.A., Huovila, A.P., Sutherland, A.E. et al. (1995) Mouse egg
integrin alpha 6 beta 1 functions as a sperm receptor. Cell, 81, 1095–1104.
Barkoff, A.F., Dickson, K.S., Gray, N.K. and Wickens, M. (2000) Translational
control of cyclin B1 mRNA during meiotic maturation: coordinated
repression and cytoplasmic polyadenylation. Dev. Biol., 220, 97–109.
Bleil, J.D. (1993) In vitro fertilization. Methods Enzymol., 225, 253–263.
Boldt, J. and Wolf, D.P. (1986) An improved method for isolation of fertile
zona-free mouse eggs. Gamete Res., 13, 213–222.
Calarco, P.G. (1995) Polarization of mitochondria in the unfertilized mouse
oocyte. Dev. Genet., 16, 36–43.
Davidson, E.H. (1986) Gene Activity in Early Development. Academic Press,
New York.
Deguchi, R., Shirakawa, H., Oda, S. et al. (2000) Spatiotemporal analysis of
Ca2⫹ waves in relation to the sperm entry site and animal-vegetal axis
during Ca2⫹ oscillations in fertilized mouse eggs. Dev. Biol., 218, 299–313.
Downs, S.M., Schroeder, A.C. and Eppig, J.J. (1986) Serum maintains the
fertilizability of mouse oocytes matured in vitro by preventing hardening
of the zona pellucida. Gamete Res., 15, 115–122.
Fulton, B.P. and Whittingham, D.G. (1978) Activation of mammalian oocytes
by intracellular injection of calcium. Nature, 273, 149–151.
Gebauer, F., Xu, W., Cooper, G.M. and Richter, J.D. (1994) Translational
control by cytoplasmic polyadenylation of c-mos mRNA is necessary for
oocyte maturation in the mouse. EMBO J., 13, 5712–5720.
Hirose, K., Kadowaki, S., Tanabe, M. et al. (1999) Spatiotemporal dynamics
of inositol 1,4,5-trisphosphate that underlies complex Ca2⫹ mobilization
patterns. Science. 284, 1527–30.
Kaji, K., Oda, S., Shikano, T. et al. (2000) The gamete fusion process is
defective in eggs of Cd9-deficient mice. Nature Genet., 24, 279–282.
Kline, D. and Kline, J.T. (1992) Repetitive calcium transients and the role of
calcium in exocytosis and cell cycle activation in the mouse egg. Dev.
Biol., 149, 80–89.
Llopis, J., McCaffery, J.M., Miyawaki, A. et al. (1998) Measurement of
cytosolic, mitochondrial, and Golgi pH in single living cells with green
fluorescent proteins. Proc. Natl. Acad. Sci. USA, 95, 6803–6808.
Mehlmann, L.M., Carpenter, G., Rhee, S.G. and Jaffe, L.A. (1998) SH2
domain-mediated activation of phospholipase Cγ is not required to initiate
Ca2⫹ release at fertilization of mouse eggs. Dev. Biol., 203, 221–232.
Miyawaki, A., Griesbeck, O., Heim, R. and Tsien, R.Y. (1999) Dynamic and
quantitative Ca2⫹ measurements using improved cameleons. Proc. Natl.
Acad. Sci. USA, 96, 2135–2140.
Miyazaki, S., Shirakawa, H., Nakada, K. and Honda, Y. (1993) Essential role
of the inositol 1,4,5-trisphosphate receptor/Ca2⫹ release channel in Ca2⫹
waves and Ca2⫹ oscillations at fertilization of mammalian eggs. Dev. Biol.,
158, 62–78.
Moore, G.D., Kopf, G.S. and Schultz, R.M. (1993) Complete mouse egg
activation in the absence of sperm by stimulation of an exogenous G
protein-coupled receptor. Dev. Biol., 159, 669–678.
Nakano, Y., Shirakawa, H., Mitsuhashi, N. et al. (1997) Spatiotemporal
dynamics of intracellular calcium in the mouse egg injected with a
spermatozoon. Mol. Hum. Reprod., 3, 1087–1093.
Parrington, J., Lai, F.A. and Swann, K. (2000) The soluble mammalian sperm
factor protein that triggers Ca2⫹ oscillations in eggs: evidence for expression
of mRNA(s) coding for sperm factor protein(s) in spermatigenetic cells.
Biol. Cell, 92, 267–275.
Polanski, Z., Ledan, E., Brunet, S. et al. (1998) Cyclin synthesis controls the
progression of meiotic maturation in mouse oocytes. Development, 125,
4989–4997.
Richter, J.D. (1999) Cytoplasmic polyadenylation in development and beyond.
Microbiol. Mol. Biol. Rev., 63, 446–456.
Rizzuto, R., Nakase, H., Darras, B. et al. (1989) A gene specifying subunit
VIII of human cytochrome c oxidase is localized to chromosome 11 and
is expressed in both muscle and non-muscle tissues. J. Biol. Chem., 264,
10595–10600.
1045
T.Aida et al
Shiraishi, K., Okada, A., Shirakawa, H. et al. (1995) Developmental changes in
the distribution of the endoplasmic reticulum and inositol 1,4,5-trisphosphate
receptors and the spatial pattern of Ca2⫹ release during maturation of
hamster oocytes. Dev. Biol., 170, 594–606.
Tay, J., Hodgman, R. and Richter, J.D. (2000) The control of cyclin B1
mRNA translation during mouse oocyte maturation. Dev. Biol., 221, 1–9.
Vassalli, J-D., Huarte, J., Belin, D. et al. (1989) Regulated polyadenylation
controls mRNA translation during meiotic maturation of mouse oocytes.
Genes Dev., 3, 2163–2171.
Van Blerkom, J. and Runner, M.N. (1984) Mitochondrial reorganization during
resumption of arrested meiosis in the mouse oocyte. Am. J. Anat., 171,
335–355.
Whittingham, D.G. (1971) Culture of mouse ova. J. Reprod. Fertil., 14
(Suppl.), 7–21.
1046
Wianny, F., Tavares, Á., Evans M.J. et al. (1998) Mouse polo-like kinase 1
associates with the acentriolar spindle poles, meiotic chromosomes and
spindle midzone during oocyte maturation. Chromosoma, 107, 430–439.
Williams, C.J., Schultz, R.M. and Kopf, G.S. (1992) Role of G proteins in
mouse egg activation: stimulatory effects of acetylcholine on the ZP2 to
ZP2f conversion and pronuclear formation in eggs expressing a functional
m1 muscarinic receptor. Dev. Biol., 151, 288–296.
Williams, C.J., Mehlmann, L.M., Jaffe, L.A. et al. (1998) Evidence that Gq
family G proteins do not function in mouse egg activation at fertilization.
Dev. Biol., 198, 116–127.
Zaccolo, M., De Giorgi, F., Cho, C.Y. et al. (2000) A genetically encoded,
fluorescent indicator for cyclic AMP in living cells. Nature Cell Biol., 2,
25–29.
Received on March 30, 2001; accepted August 23, 2001