1751
Journal of Cell Science 108, 1751-1760 (1995)
Printed in Great Britain © The Company of Biologists Limited 1995
Hormone-induced meiotic maturation in Xenopus oocytes occurs
independently of p70s6k activation and is associated with enhanced initiation
factor (eIF)-4F phosphorylation and complex formation
Simon J. Morley and Virginia M. Pain
School of Biological Sciences, University of Sussex, Brighton BN1 9QG, UK
SUMMARY
Hormone-induced meiotic maturation of the Xenopus
oocyte is regulated by complex changes in protein phosphorylation. It is accompanied by a stimulation in the rate
of translation, manifest at the level of polypeptide chain
initiation. At later times in the maturation process, this
reflects an increased ability for mRNA to interact with the
40 S ribosomal subunit. In mammalian cells there is
growing evidence for the regulation of translation by phosphorylation of ribosomal protein S6 and of initiation
factors responsible for the binding of mRNA to ribosomes.
In this report, we show that although the 70 kDa form of
S6 kinase is activated within 1.5 hours in response to progesterone or insulin, a time critical for protein synthesis, its
activation is not required for hormone-induced stimulation
of translation rates or maturation. In response to progesterone, activation of translation occurs in parallel with
enhanced phosphate labelling of eIF-4α and eIF-4γ and
eIF-4F complex formation, events which are thought to
facilitate the interaction of eIF-4F with the mRNA cap
structure. However, with insulin, activation of translation
occurs prior to detectable de novo phosphorylation of eIF4F, although a small enhancement of turnover of phosphate
on eIF-4α may occur at this early time. With either
hormone, enhanced phosphate labelling of eIF-4α is shown
to reflect activation of eIF-4α kinase(s), which co-incides
temporally with activation of p42 MAP and p90rsk kinases.
The possible role of initiation factor modification on
increased translation rates during meiotic maturation is
discussed.
INTRODUCTION
of translation at or after the binding of mRNA to the 40 S
ribosome (Laskey et al., 1977; Patrick et al., 1989).
In mammalian cells, there is evidence for the regulation of
translation by phosphorylation of ribosomal protein S6
(Morley and Thomas, 1991; Erikson, 1991; Morley, 1994) and
of initiation factors involved in binding mRNA to the 40 S
ribosomal subunit (reviewed by Hershey, 1989; Morley and
Thomas, 1991; Merrick, 1992; Redpath and Proud, 1994;
Morley, 1994). Although S6 phosphorylation may play a role
in the selective binding of mRNA species to ribosomes
(Jefferies et al., 1994a,b), the exact mechanism of this effect
has not been resolved. The activity of certain initiation factors
may influence the selection of mRNA from the cellular pool
for translation, in a process which is believed to be a ratelimiting step in many systems (Morley, 1994). Briefly, the cap
structure facilitates the binding of mRNA to the 40 S ribosomal
subunit, a process mediated by at least three initiation factors
(eIF-4A, eIF-4B, eIF-4F) and ATP hydrolysis. eIF-4F is a cap
binding protein complex composed of three subunits; eIF-4α
(eIF-4E) which specifically recognises the cap structure
(Sonenberg, 1988), eIF-4β (eIF-4A) an ATP-dependent singlestranded-RNA-binding protein with helicase activity, which
appears to recycle through the eIF-4F complex (Pause et al.,
1994) and eIF-4γ (p220), whose integrity is required but whose
Fully grown Xenopus oocytes are physiologically arrested at
the G 2/M border of the first meiotic prophase. Treatment with
progesterone, insulin, or insulin-like growth factor-1 (IGF-1)
can induce progression through meiosis and the production of
the mature unfertilised egg, by activating a number of protein
kinases which initiate germinal vesicle breakdown (GVBD).
The maturation process is associated with quantitative and
qualitative changes in the translation of proteins, including the
mitotic cyclins and the proto-oncogene c-mos (Maller, 1990;
Jessus and Ozon, 1993). This encompasses a change in translation specificity from mRNAs encoding proteins that support
growth to those which support the rapid division of cells, associated with a 2-3-fold increase in overall protein synthesis
(Woodland, 1974; Laskey et al., 1977; Wasserman et al., 1982;
Taylor and Smith, 1985; Patrick et al., 1989; Richter, 1994).
In unstimulated Xenopus oocytes only a small proportion of
their ribosomes and mRNA are being utilised; even following
maturation, only 1-2% of ribosomes are associated with
polysomes, suggesting a severe limitation in protein synthesis
initiation (Woodland, 1974; Laskey et al., 1977; Audet et al.,
1987). Studies have also shown that, relative to the egg, the
immature oocyte system exhibits an impairment of initiation
Key words: protein synthesis, initiation, phosphorylation, S6 kinase,
maturation
1752 S. J. Morley and V. M. Pain
exact function is unclear (Rhoads, 1991; Morley, 1994).
Although little is known about the protein-protein interactions
within the eIF-4F complex (Tuazon et al., 1990), it has been
recently shown that eIF-4F complex formation is increased
following activation of T lymphocytes (Morley et al., 1993). It
is believed that the eIF-4F complex functions to unwind
secondary structure in the mRNA 5′ untranslated region to
facilitate binding of the 40 S ribosome (reviewed by Morley,
1994).
In this report, we have analysed the possible role of the
p70s6k family of S6 kinase and of the phosphorylation and
assembly of the eIF-4F complex in the stimulaton of translation rates during hormone-induced maturation of Xenopus
oocytes. In contrast to the findings with mammalian cells, we
show that activation of the p70s6k family of S6 kinase is not
required for either increased translation rates or hormoneinduced meiotic maturation in Xenopus oocytes. In response to
progesterone, we show that enhanced phosphorylation of eIF4α and eIF-4γ is concomitant with increased eIF-4F complex
formation and increased rates of protein synthesis. However,
in response to insulin, translation rates are substantially
increased prior to time when we can detect the de novo phosphorylation of eIF-4α and eIF-4γ and enhanced eIF-4F
complex formation. These data suggest that the activation of
translational initiation factors can be modulated by multiple
pathways during Xenopus oocyte maturation.
MATERIALS AND METHODS
Isolation of Xenopus oocytes
Xenopus oocytes were isolated from dissected ovaries treated with 1
mg/ml collagenase A (Boehringer Mannheim) in modified Barth’s
medium (88 mM NaCl, 1 mM KCl, 2.4 mM NaHCO3, 0.3 mM
Ca(NO3)2, 0.4 mM CaCl2, 0.82 mM MgCl2, 15 mM Hepes-KOH, pH
7.6) for 1-2 hours ar room temperature, and then extensively washed
with the same buffer. Stage V and VI oocytes were manually sorted
and washed in modified Barth’s medium prior to incubation at 21°C.
Oocyte maturation
Oocytes were incubated in the presence of 5 µg/ml progesterone or
10 µM insulin for the times indicated. For the preparation of extracts
to analyse eIF-4F or protein kinase activation, oocytes were washed
and resuspended in ice-cold homogenisation buffer (50 mM MopsKOH, pH 7.2, 50 mM NaCl, 10 mM sodium pyrophosphate, 50 mM
NaF, 50 mM β-glycerophosphate, 5 mM EDTA, 5 mM EGTA, 2 mM
benzamidine, 1 mM phenylmethylsulphonyl fluoride (PMSF), 14 mM
2-mercaptoethanol, 100 µM GTP), lysed by pipetting and cell debris
removed by centrifugation in a microfuge for 5 minutes at 4°C.
Extracts were frozen in liquid nitrogen prior to use.
Preparation of cell-free extracts for protein synthesis
assays
Following incubation in the presence of hormone as shown in individual figure legends, oocytes were lysed by centrifugation at 10,000
g in the presence of 20 mM Hepes-KOH, pH 7.5, 125 mM KCl, 2
mM magnesium acetate, 2 mM 2-mercaptoethanol, 3 µg/ml leupeptin,
as described previously (Patrick et al., 1989). The cytoplasmic
extracts were made 0.5 mg/ml in soybean trypsin inhibitor (Sigma)
and used immediately. Protein synthesis assays contained 0.6 vol.
extract and 0.4 vol. reaction mix, to give the following concentrations
(in addition to those contributed by the extract): 50 mM KCl, 0.8 mM
magnesium acetate, 40 mM creatine phosphate (+ creatine phosphokinase), 50 µg/ml calf liver tRNA, 1 mM dithiothreitol (DTT), 370
U/ml placental ribonuclease inhibitor, 50 µM unlabelled methionine
and 1 mCi/ml [35S]methionine. Samples were incubated for the
indicated times at 23°C, prior to precipitation of protein with
trichloroacetic acid; total RNA concentration of the extracts was
measured as described (Patrick et al., 1989).
Immunoblotting
After SDS-PAGE, proteins were transferred to polyvinylidene difluoride membrane (PVDF, Immobilon-P, Millipore). p90rsk, MAP
kinase and p70s6k were decorated with specific anti-peptide polyclonal
antisera kindly provided by Dr J. Blenis, Harvard University. eIF-4α
and eIF-4γ were detected with specific rabbit anti-peptide antisera,
raised against the following sequences, coupled to keyhole limpet
haemocyanin using either glutaraldehyde or 1-ethyl-3-(3-dimethylamino-propyl)-carbodiimide (EDC): eIF-4α, WALWFFKNDKSKTWQANL,
synthesised
by
O.
Marin;
eIF-4γ,
KKEAVGDLLDAFKEVN, synthesised in-house. Antisera were
affinity purified using the respective peptide coupled to Affi-gel 10
(Bio-Rad), according to the manufacturers instructions. Immunblots
were visualised with goat anti-rabbit alkaline phosphatase coupled
second antibody (Sigma).
Immunoprecipitation of p70s6k and eIF-4α
Oocyte extracts (20 oocytes in 20 µl homogenisation buffer) were
diluted to 200 µl with Buffer B (50 mM Tris-HCl, pH 8.0, 120 mM
NaCl, 20 mM NaF, 30 mM ρ-nitrophenyl phosphate (NPP), 1% (v/v)
Nonidet P40 (NP40), 1 mM benzamidine, 0.1 mM PMSF), and
incubated with antisera specific to p70s6k or eIF-4α for 1.5 hours on
ice. Immunoglobulins were precipitated and recovered with Protein
A-Sepharose (Pharmacia), as previously described (Lane et al., 1992).
To assay for p70s6k activity in the immunocomplex, the beads were
washed three times with Buffer B, once in assay dilution buffer (50
mM Mops-KOH, pH 7.2, 1 mM DTT, 10 mM MgCl2, 0.2% (v/v)
Triton X-100, 5 mM ρNPP, 2 mM benzamidine) and resuspended in
5 µl assay dilution buffer. Samples were assayed for their ability to
phosphorylate 40 S ribosomal subunits as described (Lane et al.,
1992). To analyse protein co-immunoprecipiated with eIF-4α, beads
were washed as above, but bound proteins eluted directly with SDSPAGE sample buffer.
Vertical slab iso-electric focussing (VSIEF) of eIF-4α
The iso-electric focussing method of O’Farrell (1975), modified for
use in vertical slab gels, was used essentially as described by Jagus
et al. (1993), with the following modifications: the IEF sample buffer
contained 1 µM Microcysin LR in addition to the components listed;
the anode buffer (upper tank) was 0.01 M glutamic acid and the
cathode buffer (lower tank) was 0.05 M histidine. Following transfer
of proteins to PVDF, eIF-4α was visualised by immunoblotting.
m7GTP-Sepharose affinity chromatography
For the isolation of eIF-4α, cell extracts were diluted to 300 µl with
Buffer C (50 mM Mops-KOH, pH 7.2, 0.5 mM EDTA, 0.5 mM
EGTA, 100 mM KCl, 14 mM 2-mercaptoethanol, 50 mM NaF, 80
mM β-glycerophosphate, 0.5 mM PMSF, 100 µM GTP) prior to the
addition of 25 µl of a 50% (v/v) slurry of m7GTP-Sepharose
(Pharmacia), equilibrated in Buffer C. Samples were shaken for 10
minutes at 4°C and the resin isolated by centifugation in a microfuge.
The beads were washed three times with Buffer C and bound protein
eluted with either SDS-PAGE or VSIEF sample buffer, as indicated.
For the isolation of the eIF-4F complex, m7GTP-Sepharose column
chromatography was employed, as described previously (Morley and
Traugh, 1990).
Calmodulin-Sepharose chromatography
Oocyte extracts (50 µl) were adjusted to 2 mM CaCl2 prior to passage
through a 1 ml calmodulin-Sepharose column (Pharmacia) equilibrated in Buffer D (50 mM Tris-HCl, pH 7.4, 200 mM NaCl, 14 mM
eIF-4α phosphorylation during maturation 1753
2-mercaptoethanol, 0.2 mM EDTA, 2 mM benzamidine, 0.5 mM
PMSF, 2 mM CaCl2), Following an extensive wash, bound protein
was eluted with Buffer D, but in the presence of 2 mM EDTA and 4
mM EGTA. Proteins were precipitated and analysed for the presence
of eIF-4α and eIF-4γ by immunoblotting as described above.
RESULTS
Activation of the p70s6k family is not required for
meiotic maturation of Xenopus oocytes
Previous studies with Xenopus oocytes have shown that progesterone-induced meiotic maturation results in a stimulation of
the rate of translation on endogenous mRNA (Woodland, 1974;
Laskey et al., 1977; Patrick et al., 1989). In many cell systems,
activation of protein synthesis is associated with an increase in
the phosphorylation of ribosomal protein S6 and translation
initiation factors (reviewed by Morley, 1994). Whilst more than
one family of S6 kinase is known to exist in Xenopus oocytes,
their activation appears to be differentially regulated (Stefanovic and Maller, 1988; Erikson, 1991). Lane et al. (1992)
have shown that progesterone induces the transient activation
of the p70s6k family of S6 kinase within 1 hour of hormone
treatment of non-primed oocytes, whilst others had shown that
the major activation of the p90rsk family of S6 kinases occurs
late during the maturation time course (Stefanovic and Maller,
1988; Erikson and Maller, 1989; Maller, 1990). We have
utilised the macrolide immunosuppressant rapamycin, which
specifically prevents activation of p70s6k in many cell systems
(Chung et al., 1992; Downward, 1994) to investigate the
potential role of this family of S6 kinases in the activation of
translation. Fig. 1A shows that neither the rate nor the extent of
progesterone- or insulin-induced meiotic maturation is affected
by the presence of extracellular rapamycin (1 µM). To
determine whether rapamycin was entering the oocytes and
functioning as predicted, the activation of p42 MAP kinase,
p90rsk and p70s6k kinases was visualised by polyacrylamide gel
electrophoresis (SDS-PAGE) and immunoblotting. All of these
kinase families are themselves activated by phosphorylation
(Erikson, 1991; Blenis, 1993), which results in a characteristic
reduction in migration on SDS-PAGE, and hence in an apparent
increase in molecular mass (Chung et al., 1992). As shown in
Fig. 1B, this characteristic mobility shift of p90rsk and p42 MAP
kinase induced by either progesterone or insulin occurs late
Fig. 1. Activation of the p70s6k family of S6 kinase is not required for hormone-induced meiotic maturation. (A) Xenopus oocytes were isolated
by collagenase treatment of the isolated ovary and stage VI oocytes selected manually. Oocytes were incubated in modified Barth’s medium in
the presence of vehicle alone (n), 10 µM progesterone (d) or 10 µM insulin (s), in the absence (−RAPA) or presence of 1 µM rapamycin
(+RAPA). At the times indicated, the % of oocytes which had undergone germinal vesicle breakdown (GVBD) was scored. (B) Progesterone
and insulin activate p90rsk and p42 MAP kinase. Extracts were prepared from oocytes, without a prior incubation (a), or following incubation
for 3 hours (b) or 7 hours (c) with progesterone (P) or insulin (I), in the absence or presence of rapamycin as indicated. The activation state of
p90rsk (RSK), p70s6k and p42 MAP kinase (MAPK) was monitored by polyacrylamide gel electrophoresis and immunoblotting. The specific
antisera were kindly provided by Dr J. Blenis, Harvard University, USA. (C) Progesterone and insulin activate p70s6k. Oocytes (20 per time
point) were incubated in the absence of hormone (lanes 1,2) or in the presence of progesterone (lanes 3,4) or insulin (lanes 5,6), in the absence
(lanes 1,2,3,5) or presence (lanes 4,6) of 1 µM rapamycin for 1.5 hours. Extracts were prepared, adjusted to equal protein concentrations with
buffer and p70s6k immunoprecipitated from each using specific antisersa; immunocomplex assays were used to assay p70s6k activity, in the
absence (lane 1) or presence (lanes 2-6) of 40 S ribosomal subunits as substrate, as described. The resulting autoradiogram is presented and an
arrowhead indicates the migration of S6.
1754 S. J. Morley and V. M. Pain
during maturation and is unaffected by rapamycin. Immunoblot
analysis of the activation of p70s6k by this method was inconclusive due to fact that p70s6k is down-regulated within 3 hours
(Lane et al., 1992). Therefore, we directly assayed the activity
of p70s6k in extracts prepared at 1.5 hours post-stimulation
using immunoprecipitation and an immunocomplex assay
(Chung et al., 1992). Fig. 1C shows that both progesterone and
insulin stimulate the activity of p70s6k to a similar extent, as
assayed by the increased ability of the immunocomplex to phosphorylate isolated 40 S ribosomal subunits. The low activity of
p70s6k in the oocyte extract is consistent with the use of nonprimed frogs in this work (Lane et al., 1992). However, in the
presence of rapamycin, the activation of p70s6k is abrogated.
These data have been confirmed utilising a peptide substrate for
p70s6k designed after the known phosphorylation sites on S6
(data not shown). Therefore, we conclude that activation of
p70s6k is not necessary for progesterone- or insulin-induced
meiotic maturation in Xenopus oocytes.
Insulin stimulates protein synthesis at the level of
initiation
We have also measured translation rates in cell extracts
(Patrick et al., 1989), prepared at different times after hormonal
stimulation, in the absence or presence of rapamycin. Table 1
shows that progesterone effects a 3-fold increase in translation
rates on endogenous mRNA evident within 1.5 hours, with a
further increase to maximal levels between 5-9 hours coincident with GVBD. In contrast, insulin causes a maximal
increase in translation rate within 1.5 hours, prior to GVBD.
In both cases, however, the decrease in protein synthesis rate
observed at 5 hours or 3 hours post-hormone treatment, respectively, was not reproducible between experiments. The finding
that increased translation rates were largely unaffected by
rapamycin, suggests that activation of p70s6k is not required
for the stimulation of protein synthesis observed at any stage
during meiotic maturation. Fig. 2A-E shows [35S]methionine
incorporation into protein in extracts prepared at different
times following insulin treatment. Fig. 2A,B show that, in
agreement with the data presented in Table 1, relative to the
unstimulated oocyte system, insulin induces a substantial
increase in the translation rate within 1.5 hours, which was not
prevented by the presence of rapamycin. To determine whether
increased translation rates reflect a stimulation of the initiation
phase of protein synthesis, we have used the initiation inhibitor
edeine; by virtue of the increased sensitivity to edeine,
increased protein synthesis rates reflect, in part, an increased
efficiency of reinitiation of translation on endogenous mRNA.
This effect is further evident at later times of maturation (Fig.
2C,D) and illustrated in Fig. 2E, which shows the increase in
edeine-sensitive (i.e. initiation-dependent) incorporation in
these extracts. To complement these studies, we also examined
the translation products from endogenous mRNA in these
extracts. Fig. 2F shows that in all cases, translated products
encompass a wide range of molecular masses; at early times
(lanes a-c), there was essentially no detectable qualitative
changes in translation. These data suggest that the increased
protein synthesis rates at these times reflects increased utilisation of mRNA already interacting with the translational
apparatus. However, upon prolonged exposure to insulin (lanes
e-g), a number of characteristic changes in the pattern of
protein synthesis occurred, similar to that observed with prog-
Table 1. In vitro protein synthesis by extracts prepared
from Xenopus oocytes at different stages of hormoneinduced maturation
Time
after addition
(hours)
Protein synthesis at 60 min
(cpm/µg total RNA)
− rapamycin
+ rapamycin
Progesterone
0 (no hormone)
1.5
3.0
5.0
7.0
9.0
749
2101
2160
1277
5840
5221
−
n.d.
1597
3788
5198
6453
Insulin
0 (no hormone)
1.5
3.0
7.5
901
4027
2418
5454
−
3658
3796
n.d.
n.d., not determined.
esterone (lane h). These included an increase in the synthesis
of low molecular mass products, co-incident with an increase
in initiation-dependent translation. A more detailed analysis
involving two-dimensional gel electrophoresis would be
required to characterise any early changes in individual translation products.
Progesterone and insulin increase phosphate
labelling of eIF-4α
To examine the potential role for increased phosphorylation of
initiation factors in enhanced translation rates, we have
analysed the phosphate labelling of eIF-4α and -γ in intact
oocytes during meiotic maturation. To this end, Xenopus
oocytes were incubated with [32P]orthophosphate overnight
and removed from the labelling medium prior to hormone
treatment. At various times after hormone treatment, cell
extracts were prepared and eIF-4α isolated by m7GTPSepharose chromatography and equal amounts of eIF-4α
protein visualised by SDS-PAGE. The autoradiogram
presented in Fig. 3A shows that both progesterone and insulin
increase the phosphate labelling of eIF-4α, evident typically at
30-40% GVBD. The increases in labelling are insensitive to
rapamycin, and in the case of progesterone, co-incide roughly
with the major increase in translational activity of the extracts
(Table 1). In the case of insulin, protein synthesis activity had
already increased substantially by 1.5 hours (Table 1, Fig. 2AE), and therefore preceded the major increase in labelling of
eIF-4α. However, a small but reproducible enhancement of
labelling of eIF-4α after 1.5 hours of insulin treatment could
be revealed by prolonged exposure of the gel, an effect not
evident with progesterone (Fig. 3A). In order to assess whether
increased phosphate labelling of eIF-4α reflects increased de
novo phosphorylation or phosphate turnover, we have used one
dimensional iso-electric focussing (VSIEF) and immunoblotting to examine the phosphorylation status of eIF-4α during
maturation. Fig. 3B shows the results from one such experiment following isolation of eIF-4α from oocyte extracts by
m7GTP-Sepharose affinity chromatography. In the unstimulated oocytes, immunoblot analysis of eIF-4α yields three
bands; the lower band is the less-phosphorylated form, the
middle band the more highly phosphorylated form and an
eIF-4α phosphorylation during maturation 1755
Fig. 2. Insulin stimulates protein synthesis at the level of initiation. Oocytes were selected and incubated in the absence of insulin (A), or in the
presence of the hormone for 1.5 hours (B), 5 hours (C) or 7 hours (D), in the absence (sd) or presence (hj) of 1 µM rapamycin. At these
times, oocytes were lysed by centrifugation and cytoplasmic extracts prepared as described above. Protein synthesis assays, in the absence
(closed symbols) or presence (open symbols) of 10 µM edeine were performed at 23°C, with samples removed at the indicated times to
determine the incorporation of radioactive label into protein, which is expressed as cpm/µg total RNA. (E) Edeine-sensitive incorporation of
label into protein as a function of the maturation time course. (F) Protein synthesis assays with extracts prepared from unstimulated oocytes (a),
or oocytes after exposure to insulin for 1.5 hours (b), 3 hours (c), 5 hours (d), 7 hours (e), 9 hours (f), 16 hours (g), or progesterone for 16 hours
(h), were carried out as above. Following a 60 minute incubation, aliquots (1 µl) were analysed by SDS-PAGE; the right hand track shows
molecular mass markers (kDa).
upper band, which is possibly an isoform of eIF-4α retained
during the affinity chromatography step. Following hormonal
stimulation for 5-7 hours, immunoblot analysis shows an
increase in the level of eIF-4α protein migrating as the phosphorylated variant relative to the unstimulated oocyte. The
identity of these species was confirmed by the use of potato
acid phosphatase or following preparation of extracts in the
absence of phosphatase inhibitors; under these conditions, all
of the immunoreactive eIF-4α isolated from hormone-stimulated oocytes migrated largely as the less phosphorylated
variant, as observed with the unstimulated oocyte (data not
shown). These data suggest that progesterone and insulin either
increase the activity of an eIF-4α specific kinase or inactivate
a specific phosphatase, or some combination of both. To
resolve this, extracts were assayed for their ability to phosphorylate bacterially-expressed eIF-4α protein. The autoradiogram presented in Fig. 3C shows that progesterone treatment
enhances the activity of eIF-4α kinase late during maturation,
an event which can be correlated with the mobility shift of eIF4α isolated from intact oocytes on VSIEF (Fig. 3B) and the
activation of p90rsk and p42 MAP kinase (Fig. 1). Insulin
induces a small, reproducible increase in eIF-4α kinase activity
at 1.5 hours, reflected in the low level of phosphate labelling
at this time (Fig. 3A) but prior to any detectable mobility shift
of eIF-4α isolated from intact oocytes (Fig. 3B). A later phase
of eIF-4α kinase activity is seen with similar characteristics to
that observed with progesterone, suggesting that, at least in
part, activation of eIF-4α kinase(s) is responsible for increased
labelling of this protein during maturation.
eIF-4γ phosphorylation and eIF-4F complex
formation are enhanced during meiotic maturation
There is growing evidence that phosphorylation of the eIF-4γ
subunit of the eIF-4F complex plays an important role in translational control (reviewed by Morley, 1994). The formation of
functional complexes between the individual subunits, the
phosphorylation of the eIF-4F complex, and its interaction with
other initiation factors plays a central role in the binding of
mRNA to the 48 S initiation complex (Hershey, 1989; Merrick,
1992; Redpath and Proud, 1994; Morley, 1994). Therefore we
have examined the phosphorylation status during meiotic maturation of eIF-4γ recovered on m7GTP-Sepharose by virtue of
its affinity for eIF-4α (Morley and Traugh, 1990). Fig. 4A
shows that in vivo phosphate-labelling of eIF-4γ is substantially increased in response to either progesterone or insulin, in
a time course which parallels that of eIF-4α. Immunoblot
1756 S. J. Morley and V. M. Pain
Fig. 3. Progesterone and insulin increase phosphate-labelling of eIF-4α and eIF-4γ. (A) Oocytes were incubated in the presence of 500 µCi/ml
[32P]orthophosphate for 16 hours at 21°C, washed and maintained in the same buffer. Oocytes were then incubated in the absence or presence
of progesterone, insulin or rapamycin (1 µM), as indicated. Sets of 20 oocytes were removed at the time indicated, rinsed and lysed in an equal
volume of lysis buffer. eIF-4α was purified by m7GTP-Sepharose affinity chromatography following extensive washing of the resin as
described above, and visualised by SDS-PAGE and autoradiography. The right hand panel shows the phosphate labelling of eIF-4α following
exposure of oocytes to progesterone (upper) or insulin (lower) for 0, 1.5 or 7 hours, as described, but with a 5-fold increase in exposure time to
film over that shown in the left hand panel. (B) Oocyte maturation is associated with enhanced de novo phosphorylation of eIF-4α. Oocytes
were prepared and incubated in the absence of phosphate-labelling, with or without hormone or rapamycin as indicated. At the times shown, 20
oocytes were removed, cell extracts prepared and incubated with 20 µl of a 50% (v/v) slurry of m7GTP-Sepharose for 10 minutes on ice. The
resin was isolated by centrifugation, washed 3 times in buffer containing phosphatase and protease inhibitors, and bound protein eluted with
IEF sample buffer (containing 1 µM Microcystin LR) and resolved by one-dimensional vertical slab isoelectric focussing (VSIEF). Following
transfer to PVDF membrane, eIF-4α was visualised by immunoblotting using affinity-purified rabbit anti-peptide polyclonal antisera. Arrows
indicate the migration of eIF-4α, with P indicating the more highly phosphorylated variant and P′ putative isoforms of eIF-4α.
(C) Progesterone and insulin activate eIF-4α kinase(s). Oocyte extracts prepared as described above were assayed for their ability to
phosphorylate bacterially expressed eIF-4α in vitro. Each assay (10 µl; containing 50 mM Mops-KOH, pH 7.2, 5 mM MgCl2, 200 µM (1320
cpm/pmol) ATP, 500 nM PKI, 2 mM DTT, Xenopus extract (50 µg protein), 2 µg recombinant eIF-4α protein, 5 mM ρNPP) was incubated for
20 mins at 23°C, prior to dilution to 200 µl with lysis buffer. eIF-4α was isolated from the reaction mixture by m7GTP-Sepharose
chromatography as in Fig. 2B, and visualised by gel electrophoresis. The resulting autoradiogram is presented.
analysis of eIF-4α and eIF-4γ in complexes isolated on
m7GTP-Sepharose (Fig. 4A) shows that although the level of
eIF-4α does not change during hormone-induced maturation,
there is a large increase in the association of eIF-4γ with eIF4α, suggesting that the level of eIF-4F complex in the oocyte
increases during maturation. To look more directly at eIF-4F
complex formation during meiotic maturation, eIF-4α was
immunoprecipitated from cell extracts and the amount of associated eIF-4γ visualised by immunoblotting. Fig. 4B shows that
following stimulation with either progesterone or insulin, there
eIF-4α phosphorylation during maturation 1757
Fig. 4. eIF-4γ phosphorylation and
eIF-4F complex formation are
enhanced during meiotic
maturation. (A) Xenopus oocytes
were incubated with [32P]phosphate
as described in Fig. 3, prior to
stimulation with either progesterone
(P) or insulin (I). Without a prior
incubation (a) or following
stimulation with hormone for 1.5
hours (b), 3 hours (c), 5 hours (d), 7
hours (e), 20 oocytes were removed
and eIF-4α was partially purified as
described. The left hand figure
shows the resulting autoradiogram;
arrowheads indicate the migration
of the eIF-4α and eIF-4γ subunits of
the eIF-4F complex. The right hand
figure shows the corresponding
immunoblot; the upper half of the
blot was probed with affinity-purified anti-eIF-4γ antiserum and the lower
half with the corresponding anti-eIF-4α antiserum.
(B) Immunoprecipitation of oocyte extracts with anti-eIF-4α antiserum
indicates enhanced eIF-4F complex formation during maturation. Oocyte
extracts (20 µl) prepared in the absence of phosphate labelling from
unstimulated oocytes (1) or from those stimulated with progesterone (2)
for insulin (3) for 7 hours, were subject to immunoprecipitation with antieIF-4α antisera, as described. Recovered proteins were analysed by
polyacrylamide gel electrophoresis and the amount of recovered eIF-4α
and eIF-4γ visualised by immunoblotting. Arrowheads indicate the
migration of the α and γ subunits of eIF-4F. (C) Fractionation of oocyte
extracts on calmodulin-Sepharose shows enhanced eIF-4F complex
formation. Oocyte extracts were prepared as in Fig. 4B and subjected to
chromatography on calmodulin-Sepharose as described. Following
extensive washing of the resin, specifically eluted proteins were then
visualised by SDS-PAGE and immunoblotting, as described.
was increased recovery of eIF-4γ associated with eIF-4α, coincident with their phosphorylation during maturation. Alternatively, eIF-4γ has been isolated from cell extracts by virtue
of its affinity for calmodulin; in the presence of calcium, eIF4γ is retained on the resin whilst eIF-4α not associated with
eIF-4γ does not bind. Using this resin, the data presented in
Fig. 4C indicates that following progesterone or insulin stimulation of oocytes, there is increased association between eIF4α and eIF-4γ. This was confirmed using immunoprecipitation
with serum specific to eIF-4γ (data not shown). Therefore,
these data show that eIF-4F complex formation is enhanced
during meiotic maturation, co-incident with its increased phosphorylation and the stimulation of protein synthesis initiation.
DISCUSSION
Hormone-induced meiotic maturation of the Xenopus oocyte is
regulated by complex changes in protein phosphorylation
(reviewed by Maller, 1990; Jessus and Ozon, 1993) and is
accompanied by a 2-3-fold increase in the rate of protein
synthesis (Woodland, 1974; Laskey et al., 1977; Wasserman
and Smith, 1978; Wasserman et al., 1982; Taylor and Smith,
1985; Richter, 1994). This enhanced translation rate is associated with increased recruitment of mRNA into polysomes
without a change in polypeptide elongation rates (Richter et
al., 1982). This has been attributed to a number of concomi-
tant events, such as the dissociation of masking proteins
inhibiting the recruitment of mRNAs (Richter, 1986), regulation of the length of mRNA poly(A) tails, (where addition of
poly(A) is correlated with activation of many maternal mRNAs
(reviewed by Jackson and Standart, 1990; Sheets et al., 1994)),
increased phosphorylation of ribosomal protein S6 (Cicirelli et
al., 1988; Erikson and Maller, 1989; Maller, 1990) and the activation of initiation factor eIF-4A (Audet et al., 1987). We have
used maturation of Xenopus oocytes as a model system to study
the potential role of S6 and initiation factor phosphorylation in
the activation of protein synthesis.
In Xenopus oocytes, S6 becomes multiply phosphorylated
on serine residues in response to progesterone, insulin, phorbol
esters and micro-injection of proteins encoded by transforming viruses (Maller, 1990; Morley and Thomas, 1991). We now
show that insulin can activate the p70s6k family of kinase (Fig.
1), and that this effect co-incides with a major increase in
protein synthesis rate, which appears to reflect an increase in
the efficiency of re-initiation of translation on endogenous
mRNA (Fig. 2). To determine whether p70s6k plays an
essential role in the increased rates of translation, we have used
rapamycin, a macrolide immunosuppressant, which specifically prevents the activation of the p70s6k, but does not affect
the activation of either p42 MAPK or p90rsk (Fig. 1). Pretreatment with rapamycin did not prevent either progesterone or
insulin-induced maturation (Fig. 1), and had little effect on the
activation of overall protein synthesis (Fig. 2 and Table 1). Our
1758 S. J. Morley and V. M. Pain
data indicate that activation of p70s6k, which is believed to be
the physiological S6 kinase in mammalian cells (Chung et al.,
1992; Blenis, 1993; Morley, 1994; Downward, 1994) is not
required for the overall activation of translation during progesterone or insulin-induced meiotic maturation. At this time,
based on one-dimensional gel electrophoresis of labelled
proteins (Fig. 2F), we do not know whether rapamycin affects
the translation of specific classes of mRNA (Jefferies et al.,
1994a,b). These data suggest either that S6 phosphorylation
plays no role in the activation of protein synthesis in Xenopus
oocytes or that p90rsk activation is sufficient to satisfy any
requirement for S6 phosphorylation during this period. Indeed,
stimulation of p90rsk activity can be correlated with increased
translation rates, with p90rsk activated within 20-30 minutes in
response to insulin (Cicirelli et al., 1988), an effect not
observed with progesterone (Lane et al., 1992).
The activity of initiation factors can be regulated by phosphorylation in response to growth factors, mitogens, phorbol
esters and transforming viruses. Recently, Xu et al. (1993)
have shown that, co-incident with activation of translation, eIF4α phosphate labelling is increased during maturation of
starfish oocytes, induced by 1-methyladenine. However, the
authors have not determined whether this increased labelling
reflects increased de novo phosphorylation or turnover of
phosphate on eIF-4α. In this study, we show that both progesterone and insulin-induced maturation of Xenopus oocytes are
associated with increased phosphate labelling of eIF-4α (Fig.
3), which is clearly evident at 5 hours after addition of
hormone, co-incident with activation of p42 MAP and p90rsk
kinases. Prolonged exposure of the gel (Fig. 3A) revealed a
small, reproducible increase in eIF-4α labelling within 1.5
hours of insulin treatment. Iso-electric focussing and
immunoblotting indicate that this early phase of labelling
probably reflects turnover of phosphate on eIF-4α, since there
is no detectable change in the distribution of the factor between
phosphorylated and non-phosphorylated forms at this early
time. However, the larger increase in labelling at later times of
maturation primarily reflects de novo phosphorylation (Fig.
3A,B). In some ways, the increased phosphate labelling of eIF4α co-incident with enhanced translation rates and the onset of
GVBD is similar to that reported for starfish oocytes induced
to mature with 1-methyladenine (Xu et al., 1993). However, in
our study the major increase in phosphorylation of eIF-4α in
response to insulin in Xenopus oocytes appears to occur after
the stimulation of translation is already well established (Fig.
3 vs Fig. 2). One possible explanation for these data is that
another factor(s) is limiting for translation at early times
following activation. Although eIF-2/eIF-2B and mRNA have
been shown to be limiting in the starfish (Xu et al., 1993) and
sea urchin oocyte systems (Colin et al., 1987), this does not
appear to be the case in Xenopus oocytes (Laskey et al., 1977;
Patrick et al., 1989). Isolation of germinal vesicles and
immunoblot analysis have indicated that eIF-4α is cytoplasmically localised. Therefore, the possibility that eIF-4α is
sequestered in the germinal vesicle and not readily available as
a substrate until dissolution of the membrane, has been discounted (data not shown).
The data presented in Fig. 3C suggests that increased phosphorylation of eIF-4α reflects, at least in part, the biphasic activation of kinase(s) during maturation, which is largely insensitive to rapamycin. In the experiment presented, increased
labelling of eIF-4α at 5 hours post-progesterone stimulation
was partially inhibited by rapamycin; this was not a reproducible finding between different batches of oocytes. Preliminary data using 32P-labelled eIF-4F as a substrate has
suggested that whilst phosphatase activity towards eIF-4α is
increased in response to either hormone, the rate of dephosphorylation of eIF-4α is similar at 1.5 hours and 7 hours poststimulation (S. J. Morley and V. M. Pain, unpublished data).
Therefore, the enhanced activity of eIF-4α kinase at early
times of insulin treatment are off-set by increased phosphatase
activity, resulting in enhanced phosphate turnover on eIF-4α,
and no de novo phosphorylation (Fig. 2B,C); at later times,
further kinase is activated with no change in phosphatase
activity, resulting in de novo phosphorylation. We have also
assayed extracts for their ability to phosphorylate a bacteriallyexpressed variant of eIF-4α protein, where the major site of
phosphorylation (Ser53) has been mutated to Alanine (Ala53).
Although phosphorylated with a lower stoichiometry, the
Ala53 mutant protein is phosphorylated with a similar time
course to that of the Ser53 protein (data not shown). This
suggests that either the observed increase in eIF-4α kinase
activity reflects, in part, activation of a non-specific kinase or
else that eIF-4α can be phosphorylated at more than one site
(Bu et al., 1993). Kaufman et al. (1993) have shown that the
Ala53 mutant protein is efficiently phosphorylated in COS-1
cells, and that the phosphorylation of Ser53 is not required for
its interaction with eIF-4γ. The physiological significance of
phosphorylation sites additional to Ser53 has not as yet been
determined.
Although the role of eIF-4γ in translation is unclear, it is
known to be required for the efficient translation of capped
mRNA (reviewed by Sonenberg, 1988; Morley, 1994). It is
now believed that eIF-4γ co-ordinates the activity of eIF-4A
and eIF-4α to facilitate the initial interaction with mRNA, in
a sequence independent manner, prior to the specific interaction of eIF-4α with the cap structure (Pause et al., 1994). We
now show that the phosphorylation of eIF-4γ is increased
during the later stages of both progesterone- and insulininduced meiotic maturation, co-incident with enhanced interaction of this protein with eIF-4α (Fig. 4). We have demonstrated this enhanced interaction using m7GTP-Sepharose and
by immunoprecipitation with anti-eIF-4α antiserum and direct
isolation of eIF-4γ on calmodulin-Sepharose (Fig. 4). Mono Q
chromatography, employed to separate free eIF-4α from the
eIF-4F complex (Lamphear and Panniers, 1990), verified
increased levels of eIF-4F complex formation in the mature
oocyte (data not shown). A possible relevance of increased
levels of eIF-4F complex formation on rates of translation is
shown by the finding that the eIF-4F complex has a 20-fold
increased ability over free eIF-4A to unwind the secondary
structure at the 5′ end of mRNA (Pause et al., 1994). Moreover,
phosphorylation in vitro of the eIF-4F complex stimulates its
interaction with the mRNA cap structure (Morley et al., 1991).
Enhanced eIF-4F complex formation has been reported
following the mitogenic activation of protein synthesis in
primary T cells (Morley et al., 1993) and treatment of HepG2
cells with okadaic acid (Bu et al., 1993), both of which result
in increased phosphorylation of eIF-4F.
A major question which still needs to be resolved is the
identity of the physiological eIF-4α kinase(s). In Xenopus
oocytes, both progesterone and insulin can induce meiotic mat-
eIF-4α phosphorylation during maturation 1759
uration through separate, but converging, signalling pathways.
During the process of meiotic maturation, a dramatic increase
in protein phosphorylation occurs due to activation of a
cascade of protein kinases (Maller, 1990; Jessus and
Ozon,1993). These include the de novo synthesis and activation of the c-mos kinase, possibly regulated by increased
polyadenylation of mos maternal mRNA (Sheets et al., 1994).
This is the only protein whose synthesis is essential for the
initiation of meiotic maturation (Matten et al., 1994). Completion of meiosis I and arrest in meiosis II is regulated by a progesterone-induced cascade of events including temporal
changes in the activities of p70s6k (Lane et al., 1992), p34cdc2
kinase, cdc25, p42 MAP kinase, which lags about 2 hours
behind the onset of mos protein synthesis, protein kinase A,
and activation of the p90rsk family of S6 kinases (Erikson and
Maller, 1989; Jessus and Ozon, 1993; Matten et al., 1994). A
ras-dependent pathway of p42 MAP kinase activation also
exists; Xenopus oocytes contain p21ras, which is required for
insulin-stimulated maturation and is activated via the insulin
receptor substrate 1 (IRS-1: Myers et al., 1994). By analogy
with events known to occur in somatic cells (Grigorescu et al.,
1994; Downward, 1994), insulin is believed to lead to the activation of p90rsk via MAP kinase kinase during maturation.
Temporal correlations presented in Figs 1 and 3, suggest a
possible link between the increased de novo phosphorylation
of eIF-4α and eIF-4γ and the activation of the p42 MAP kinase
pathway. In the light of the insensitivity to rapamycin, the
above data also show that p70s6k itself, or any immediately
downstream kinase activated by p70s6k is not responsible for
increased phosphorylation of eIF-4α. Reddy et al. (1993) have
suggested that one possible candidate for an eIF-4α kinase is
a protamine kinase, activated via a MAP kinase pathway
distinct to that of p42 MAP kinase. However, it is unlikely to
be a candidate here as the level of protamine kinase activity
did not increase during meiotic maturation (data not shown).
At this time, the identity of any specific kinase(s) involved in
phosphorylation of eIF-4α awaits determination.
We thank Dr J. Blenis (Harvard University, USA) for providing the
antiserum to p70s6k, p42 MAP kinase and p90rsk; Dr R. Jagus (Center
for Marine Biotechnology, Baltimore USA) for providing the recombinant eIF-4α proteins; Drs O. Marin and L. Pinna (Dipartimento di
Chimica Biologica, Universita di Padova, Italy) for the synthetic
peptide designed after eIF-4α used in the production of rabbit
antiserum; and Dr A. Flynn (University of Bristol, UK) for advice on
analysis of eIF-4α by VSIEF. This work was supported by a project
grant from The Wellcome Trust.
REFERENCES
Audet, R. G., Goodchild, J. and Richter, J. D. (1987). Eukaryotic initiation
factor 4A stimulates translation in micro-injected Xenopus oocytes. Dev.
Biol. 121, 58-68.
Blenis, J. (1993). Signal transduction via the MAP kinases: Proceed at your
own RKS. Proc. Nat. Acad. Sci. USA 90, 5889-5892.
Bu, X., Haas, D. and Hagedorn, C. H. (1993). Novel phosphorylation sites of
eukaryotic initiation factor 4F and evidence that phosphorylation stabilises
interactions with p25 and p220 subunits. J. Biol. Chem. 268, 4975-4978.
Chung, J., Kuo, C. J., Crabtree, G. R. and Blenis, J. (1992). RapamycinFKBP selectively blocks growth-dependent activation of and signalling by
the 70kd S6 kinases. Cell 69, 1227-1236.
Cicirelli, M. F., Pelech, S. L. and Krebs, E. G. (1988). Insulin and
progesterone activate a common synthetic ribosomal protein S6 peptide
kinase in Xenopus oocytes. FEBS Lett. 241, 195-201.
Colin, A. M., Brown, B. D., Dholakia, J. N., Woodley, C. L., Wahba, A. J.
and Hille M. (1987). Evidence for simultaneous derepression of messenger
RNA and the guanine nucleotide exchange factor in fertilised sea urchin
eggs. Dev. Biol. 123, 354-367.
Downward, J. (1994). Regulating S6 kinase. Nature 371, 378-379.
Erikson, R. L. (1991). Structure, expression and regulation of protein kinases
involved in the phosphorylation of ribosomal protein S6. J. Biol. Chem. 266,
6007-6010.
Erikson, R. L. and Maller J. (1989). In vivo phosphorylation and activation of
ribosomal S6 kinases during Xenopus oocyte maturation. J. Biol. Chem. 264,
13711-13717.
Grigorescu, F., Baccara, M.-T., Rouard, M. and Renard, R. (1994). Insulin
and IGF-1 signalling in oocyte maturation. Hormone Res. 42, 55-61.
Hershey, J. W. B. (1989). Protein phosphorylation controls translation rates. J.
Biol. Chem. 264, 20823-20826.
Jackson, R. J. and Standart, N. (1990). Cytoplasmic regulation of mRNA
function; importance of the 3′ untranslated region. Cell 62, 15-24.
Jagus, R., Huang, W. I., Hiremath, L. S., Stern, B. D. and Rhoads, R. E.
(1993). Mechanism of action of developmentally regulated sea urchin
inhibitor of eIF-4. Dev. Gen. 14, 412-423.
Jefferies, H. B. J., Reinhard, C., Kozma, S. C. and Thomas G. (1994a).
Rapamycin selectively represses translation of the polypyrimidine tract
mRNA family. Proc. Nat. Acad. Sci. USA 91, 4441-4445.
Jefferies, H. B. J., Thomas, G. and Thomas, G. (1994b). Elongation factor 1α
mRNA is selectively translated following mitogenic stimulation. J. Biol.
Chem. 269, 4367-4372.
Jessus, C. and Ozon, R. (1993). Regulation of cell division during oogenesis of
vertebrates: the Xenopus oocye paradigm. Comp. Biochem. Physiol. 106A,
431-448.
Kaufman, R. J., Murtha-Riel, P., Pittman, D. O. and Davies, M. W. (1993).
Characterisation of wild-type and ser53 mutant eukaryotic initiation factor
4E over-expression in mammalian cells. J. Biol. Chem. 268, 11902-11909.
Lamphear, B. J. and Panniers, R. (1990). Cap-binding protein complex that
restores protein synthesis in heat-shocked Ehrlich cell lysates contains highly
phosphorylated eIF-4E. J. Biol. Chem. 265, 5333-5336.
Lane, H. A., Morley, S. J., Kozma, S. C. and Thomas, G. (1992).
Identification and early activation of a Xenopus laevis p70S6K following
progesterone induced meiotic maturation. EMBO J. 11, 1743-1749.
Laskey, R. A., Mills, A. D. Gurdon, J. B. and Partington G. A. (1977).
Protein synthesis in oocytes of Xenopus laevis is not regulated by the supply
of mRNA. Cell 11, 345-351.
Maller, J. L. (1990). Xenopus oocytes and the biochemistry of cell division.
Biochemistry 29, 3158-3166.
Matten, W., Daar, I. and Vande Woude, G. F. (1994). Protein kinase A acts
at multiple points to inhibit Xenopus oocyte maturation. Mol. Cell. Biol. 14,
4419-4426.
Merrick, W. C. (1992). Mechanism and regulation of eukaryotic protein
synthesis Microbiol. Rev. 56, 291-315.
Morley, S. J. and Traugh, J. A. (1990). Differential stimulation of
phosphorylation of initiation factors eIF-4E, eIF-4B, eIF-3 and ribosomal
protein S6 by insulin and phorbol esters. J. Biol. Chem. 265, 10611-10616.
Morley, S. J., Dever, T. E., Etchison, D. E. and Traugh, J. A. (1991).
Phosphorylation of eIF-4F by protein kinase C or multipotential S6 kinase
stimulated protein synthesis at initiation. J. Biol. Chem. 266, 4669-4672.
Morley, S. J. and Thomas G. (1991). Intracellular messengers and the control
of protein synthesis. Pharm. Ther. 50, 291-319.
Morley, S. J., Rau, M., Kay, J. E. and Pain, V. M. (1993). Increased
phosphorylation of eukaryotic initiation factor 4E during early activation of
T lymphocytes correlates with increased initiation factor 4F complex
formation. Eur. J. Biochem. 218, 39-48.
Morley, S. J. (1994). Signal transduction mechanisms in the regulation of
protein synthesis. Mol. Biol. Rep. 19, 221-231.
Myers, M. G., Sun, Y. G. and White, M. F. (1994). The IRS-1 signalling
system. Trends Biochem. Sci. 19, 289-293.
O’Farrell, P. H. (1975). High resolution two dimensional electrophoresis of
proteins. J. Biol. Chem. 250, 4007-4021.
Patrick, T. D., Lewer, C. E. and Pain, V. M. (1989). Preparation and
characterisation of cell-free systems from oocytes and eggs of Xenopus
laevis. Development 106, 1-9.
Pause, A., Methot, N., Svitkin, Y., Merrick, W. C. and Sonenberg, N.
(1994). Dominant negative mutants of mammalian translation initiation
factor eIF-4A define a critical role for eIF-4F in cap-dependent and capindependent initiation of translation. EMBO. J. 13, 1205-1215.
Reddy, S. A. G., Gou, H., Tarun, S. Z. and Damuni. Z. (1993).
1760 S. J. Morley and V. M. Pain
Phosphorylation and activation of protamine kinase by two forms of myelin
basic protein kinase from extracts of bovine kidney cortex. J. Biol. Chem.
268, 15298-15304.
Redpath, N. T. and Proud, C. G. (1994). Molecular mechanisms in the control
of translation by hormones and growth factors. Biochim. Biophys. Acta 1220,
147-162.
Rhoads, R. E. (1991). Protein synthesis, cell growth and oncogenesis. Curr.
Opin. Cell Biol. 3, 1019-1024.
Richter, J. D., Wasserman, W. J. and Smith, L. D. (1982). The mechanism
for increased protein synthesis during Xenopus oocyte maturation. Dev. Biol.
89, 159-167.
Richter, J. D. (1986). Molecular mechanisms of translational control during
the early development of Xenopus laevis. In Translational Regulation of
Gene Expression (ed. J. Ilan), pp. 111-139. Plenum Press, New York.
Richter, J. D. (1994). Translational control in development: a perspective. Dev.
Gen. 14, 407-411.
Sheets, M. D., Fox, C. A., Hunt, T., Vande, Woude, G. and Wickens, M.
(1994). The 3′ untranslated region of c-mos and cyclin mRNAs stimulate
translation by regulating cytoplasmic polyadenylation. Genes Dev. 8, 826838.
Sonenberg, N. (1988). CAp-binding proteins of eukaryotic mRNA: functions
in initiation and control of translation. Microbiol. Rev. 44, 173-205.
Stefanovic, D. and Maller J. (1988). Post-transcriptional regulation by insulin
of Xenopus ribosomla protein S6 kinase. Exp. Cell. Res. 179, 104-114.
Taylor, M. A. and Smith, L. D. (1985). Quantitative changes in protein
synthesis during oogenesis in Xenopus laevis. Dev. Biol. 110, 230-237.
Tuazon P. T., Morley, S. J., Dever, T. E., Merrick, W. C., Rhoads, R. E. and
Traugh, J. A. (1990). Association of initiation factor eIF-4E in a cap binding
protein complex is critical for and enhances phosphorylation by protein
kinase C. J. Biol. Chem. 265, 10617-10621.
Wasserman, W. J. and Smith, L. D. (1978). The regulation of Xenopus oocyte
maturation. In The Vertebrate Ovary (ed. R. E. Jones), pp. 443-468. Plenum
Press, New York.
Wasserman, W. J., Richter, J. D. and Smith, L. D. (1982). Protein synthises
during maturation-promoting factor and progesterone-induced maturation of
Xenopus oocytes. Dev. Biol. 89, 152-158.
Woodland, H. (1974). Changes in the polysome content of developing
Xenopus laevis embryos. Dev. Biol. 40, 90-101.
Xu, Z., Dholakia, J. and Hille, M. (1993). Phosphorylation of starfish eIF-4E.
Dev. Gen. 14, 424-439.
(Received 24 October 1994 - Accepted 13 December 1994)