14-3-3σ controls mitotic translation to facilitate cytokinesis

Abstract

14-3-3 proteins are crucial in a wide variety of cellular responses including cell cycle progression, DNA damage checkpoints and apoptosis. One particular 14-3-3 isoform, sigma, is a p53-responsive gene, the function of which is frequently lost in human tumours, including breast and prostate cancers as a result of either hypermethylation of the 14-3-3sigma promoter or induction of an oestrogen-responsive ubiquitin ligase that specifically targets 14-3-3sigma for proteasomal degradation. Loss of 14-3-3sigma protein occurs not only within the tumours themselves but also in the surrounding pre-dysplastic tissue (so-called field cancerization), indicating that 14-3-3sigma might have an important tumour suppressor function that becomes lost early in the process of tumour evolution. The molecular basis for the tumour suppressor function of 14-3-3sigma is unknown. Here we report a previously unknown function for 14-3-3sigma as a regulator of mitotic translation through its direct mitosis-specific binding to a variety of translation/initiation factors, including eukaryotic initiation factor 4B in a stoichiometric manner. Cells lacking 14-3-3sigma, in marked contrast to normal cells, cannot suppress cap-dependent translation and do not stimulate cap-independent translation during and immediately after mitosis. This defective switch in the mechanism of translation results in reduced mitotic-specific expression of the endogenous internal ribosomal entry site (IRES)-dependent form of the cyclin-dependent kinase Cdk11 (p58 PITSLRE), leading to impaired cytokinesis, loss of Polo-like kinase-1 at the midbody, and the accumulation of binucleate cells. The aberrant mitotic phenotype of 14-3-3sigma-depleted cells can be rescued by forced expression of p58 PITSLRE or by extinguishing cap-dependent translation and increasing cap-independent translation during mitosis by using rapamycin. Our findings show how aberrant mitotic translation in the absence of 14-3-3sigma impairs mitotic exit to generate binucleate cells and provides a potential explanation of how 14-3-3sigma-deficient cells may progress on the path to aneuploidy and tumorigenesis.

Full text

LETTERS
14-3-3scontrols mitotic translation to facilitate
cytokinesis
Erik W. Wilker
1
, Marcel A. T. M. van Vugt
1
, Stephen C. Artim
1
, Paul H. Huang
2
, Christian P. Petersen
1
,
H. Christian Reinhardt
1
, Yun Feng
1
, Phillip A. Sharp
1
, Nahum Sonenberg
3
, Forest M. White
2
& Michael B. Yaffe
1,2
14-3-3 proteins are crucial in a wide variety of cellular responses
including cell cycle progression, DNA damage checkpoints and
apoptosis. One particular 14-3-3 isoform, s, is a p53-responsive
gene, the function of which is frequently lost in human tumours,
including breast and prostate cancers as a result of either hyper-
methylation of the 14-3-3spromoter or induction of an oestrogen-
responsive ubiquitin ligase that specifically targets 14-3-3sfor
proteasomal degradation
1–9
. Loss of 14-3-3sprotein occurs not
only within the tumours themselves but also in the surrounding
pre-dysplastic tissue (so-called field cancerization), indicating that
14-3-3smight have an important tumour suppressor function
that becomes lost early in the process of tumour evolution
3,9
. The
molecular basis for the tumour suppressor function of 14-3-3s
is unknown. Here we report a previously unknown function for
14-3-3sas a regulator of mitotic translation through its direct
mitosis-specific binding to a variety of translation/initiation fac-
tors, including eukaryotic initiation factor 4B in a stoichiometric
manner. Cells lacking 14-3-3s, in marked contrast to normal
cells, cannot suppress cap-dependent translation and do not
stimulate cap-independent translation during and immediately
after mitosis. This defective switch in the mechanism of trans-
lation results in reduced mitotic-specific expression of the endo-
genous internal ribosomal entry site (IRES)-dependent form of
the cyclin-dependent kinase Cdk11 (p58 PITSLRE), leading to
impaired cytokinesis, loss of Polo-like kinase-1 at the midbody,
and the accumulation of binucleate cells. The aberrant mitotic
phenotype of 14-3-3s-depleted cells can be rescued by forced
expression of p58 PITSLRE or by extinguishing cap-dependent
translation and increasing cap-independent translation during
mitosis by using rapamycin. Our findings show how aberrant
mitotic translation in the absence of 14-3-3simpairs mitotic exit
to generate binucleate cells and provides a potential explanation of
how 14-3-3s-deficient cells may progress on the path to aneu-
ploidy and tumorigenesis.
Because many tumour suppressor proteins function during spe-
cific parts of the cell cycle, we examined whether the binding of
ligands to 14-3-3sshowed a cell cycle dependence. U2OS cells were
synchronized by a double thymidine block, and lysates were prepared
at various times after release (Supplementary Fig. 1a). Immunopre-
cipitation of endogenous 14-3-3sfrom these lysates showed a
marked increase in 14-3-3s-bound proteins in the mitotic and
immediate post-mitotic periods (Fig. 1a, time points at 12 and 16 h
after release). This same mitotic enrichment in ligand binding to 14-
3-3swas also observed in other cell types, including HCT116 cells
and HeLa cells (Supplementary Fig.1b–d), but was not observed when
ligands of another endogenous 14-3-3 isoform, 14-3-3b, were ana-
lysed (Supplementary Fig. 1e) or in 14-3-3simmunoprecipitations
from asynchronous cells after DNA damage (Supplementary Fig. 2).
The mitosis-dependent binding of ligands to 14-3-3sseemed to be
both dependent on phosphorylation, because binding was lost when
lysates were treated with protein phosphatase 1 (Supplementary Fig.
1d), and direct, as shown with a two-dimensional ‘far-western’ assay
(Fig. 1d).
The mitotic targets of 14-3-3swere identified by mass spectro-
metry. This analysis revealed that a large proportion of 14-3-3s
ligands were proteins involved in the process of translation, including
several initiation factors involved in mediating cap-dependent trans-
lation (Fig. 1b and Supplementary Table 1). Two-dimensional far-
western blotting of mitotic 14-3-3simmunoprecipitates showed
that several of these proteins, including eukaryotic initiation factor
4B (eIF4B), eIF-2aand elongation factor 1a(EF1a), bound directly
to 14-3-3s, whereas other 14-3-3s-associated proteins such as eIF4G
seemed to be interacting indirectly (Fig. 1e).
It is well established that translation in mammalian cells is markedly
altered during and immediately after mitosis
10
, with a pronounced
suppression of cap-dependent translation and a corresponding
enhancement of cap-independent translation
11
. The biological mech-
anism and significance of this translational switch, however, are
unclear. Given the abundance of proteins involved in translation in
the 14-3-3simmunoprecipitations, we investigated whether 14-3-3s
has a direct function in this process by using RNA interference
(RNAi). Synchronous 14-3-3sstable knockdown and control cells
were pulse-labelled with [
35
S]methionine, and newly translated pro-
teins were analysed by SDS–polyacrylamide gel electrophoresis
(PAGE) and autoradiography (Fig. 2a, d, and Supplementary Fig.
3a). As expected, in control short hairpin RNA (shRNA)-treated
HeLa and U2OS cells, radiolabel incorporation into nascent polypep-
tides was suppressed at 12 and 16 h after release from the double
thymidine block (Fig. 2a, e, lanes marked ‘con’)
11
. Remarkably, no
suppression of mitotic translation was observed in either of these
cell types when they were depleted of 14-3-3s(lanes marked ‘s’).
Identical results were also obtained with two other distinct 14-3-3s
shRNAs (data not shown). In contrast to 14-3-3sdepletion, RNAi-
mediated knockdown of 14-3-3bhad no effect on the suppression
of translation during and immediately after mitosis (Fig. 2b, d).
Furthermore, the aberrant mitotic translation seen in the 14-3-3s
knockdown cells did not result from a failure of these cells to enter
mitosis (Fig. 2c). Although the global pattern of cell cycle progres-
sion did not change, we observed an increase in the population of 4n
DNA-containing cells and a significant increase in mitotic index
(Fig. 2c, and Supplementary Fig. 3b, c). Coexpression of an shRNA-
resistant construct of 14-3-3stogether with 14-3-3sshRNA restored
the suppression of mitotic translation (Fig. 2e, lanes marked ‘s
r
’,
and Fig. 2f, top). These findings indicate that 14-3-3sis important
1
Center for Cancer Research, Department of Biology and
2
Biological Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA.
3
Department of
Biochemistry, McGill Cancer Centre, McGill University, Montreal, Quebec, H3G 1Y6, Canada.
Vol 446
|
15 March 2007
|
doi:10.1038/nature05584
329
Nature
©2007 Publishing Group

in the physiological downregulation of new protein synthesis during
and immediately after mitosis.
The inhibition of translation observed during mitosis has prev-
iously been shown to result from a global decrease in cap-dependent
translation, whereas cap-independent translation is subsequently
increased
12,13
. We used two different IRES-containing bi-cistronic
vectors, a viral-based IRES from HIV
14
, and a normal cellular IRES
from p27
Kip1
(ref. 15), in a direct investigation of the relative effects of
14-3-3son cap-dependent versus cap-independent translation in
synchronized U2OS and HeLa cells. As shown in Fig. 2g, 14-3-3s-
depleted cells failed to show the robust mitotic-specific increase in
the ratio of cap-independent to cap-dependent translation that
occurred in control cells. Thus, a loss of 14-3-3sresulted in both
aberrant persistence of cap-dependent translation and prevented the
normal enhancement of cap-independent to cap-dependent trans-
lation during mitosis.
The relative amounts of different eukaryotic initiation factors in
cells varies more than 100-fold
16
, and they have distinct roles in the
cap recognition process
17,18
. To investigate which eIF targets of 14-3-
3sare potentially responsible for the 14-3-3s-mediated suppression
of cap-dependent mitotic translation, we immunodepleted mitotic
extracts with the use of anti-14-3-3santibodies and probed the
depleted lysates for residual eIFs. Of the eIFs examined, the most
pronounced result was obtained with eIF4B, which was completely
absent from the 14-3-3s-depleted mitotic lysates (Fig. 1c). eIF4B
facilitates the ATP-dependent helicase activity of eIF4A to promote
ribosome recruitment required for cap-dependent translation
19
.We
further observed that transient overexpression of exogenous Flag-
tagged eIF4B was able to overcome the ability of endogenous 14-3-
3sto suppress mitotic translation in normal cells but had no effect on
the increased translation seen in 14-3-3sknockdown cells (Fig. 2e,
lanes marked ‘4B’, and Fig. 2f, bottom). Taken together, the data in
Figs 1 and 2 indicate that 14-3-3sis important in the physiological
downregulation of new protein synthesis during and immediately
after mitosis by suppressing cap-dependent translation through
binding to eIF4B.
Because loss of 14-3-3soccurs early in the process of tumori-
genesis, we also examined the 14-3-3sknockdown cells for a corres-
ponding cellular phenotype that might correlate with aberrant
regulation of mitotic translation. In the 14-3-3sknockdown cells
we observed significantly increased numbers of cells displaying
persistent cytokinetic bridges, binucleate cells (often containing mid-
body remnants), and cells that seemed to have ‘fused’ after mitosis
with widely separated nuclei, probably reflecting a failure late in
cytokinesis (Fig. 3a–c; P,0.05 for 14-3-3sshRNA versus control
cells, Student’s t-test, two-tailed; and Supplementary Fig. 3d–f).
Intriguingly, these phenotypes seemed to correlate with a failure of
Polo-like kinase-1 (Plk1), a critical mitotic kinase involved in the
completion of cytokinesis
20,21
, to localize to the midbody at the end
of mitosis (Fig. 4a). Both the localization of Plk1 to the midbody and
the mitotic phenotype of 14-3-3s-depleted cells could be signifi-
cantly restored by treatment of the cells with rapamycin, an inhibitor
of cap-dependent, but not cap-independent, translation
22
, immedi-
ately before mitotic entry (Figs 3b, c and 4a).
We searched for proteins that were translated in mitosis in a cap-
independent manner and were postulated to be involved in the mitotic
process. One of the candidates, PITSLRE/Cdk11, is a member of the
Cdc2-like protein kinase family that undergoes cap-independent
translation from an internal ribosome entry site during mitosis to
produce a 58-kDa isoform that facilitates proper mitotic progression
and termination
23–25
. In both U2OS cells and HeLa cells we observed
that 14-3-3sknockdown cells, but not control cells, failed to synthe-
size p58 PITSLRE in mitosis (Fig. 4b, c, 16 h lanes). Depletion of
PITSLRE/Cdk11 from U2OS cells caused an increase in mitotic con-
tent as well as the number of binucleate cells, exactly as observed in
14-3-3s-depleted cells (Supplementary Figs 3 and 4). Addition of
rapamycin to 14-3-3sknockdown cells restored the mitotic trans-
lation of p58 PITSLRE(Fig. 4b, c). Furthermore, transienttransfection
of a complementary DNA encoding the 58-kDa isoform of PITSLRE
into the 14-3-3sknockdown cells was sufficient to relocalize Plk1 at
the midbody and partly rescue the mitosis-defective cell phenotype,
reducing the number of binucleate and fused cells by slightly more
a
b
c
d
e
A048 12162024h
Time after release from T/T block
14-3-3σ
Other
RNA/
transcription
Mitotic/
histones
Structural
Translation
+–+–+–
14-3-3σ depletion
01216h
eIF4B
eIF4G
14-3-3σ
Blot
σ far-
western Immunoblot Merge
EF1α
eIF2α
eIF4B
eIF4G
IgG
pI 311
0 h
12 h
16 h
14-3-3σ
14-3-3σ
14-3-3σ
eIF4B
eIF2α
EF1α
Lysates
Figure 1
|
14-3-3sbinds to its targets during mitosis. a, SDS–PAGE
analysis of 14-3-3simmunoprecipitates from synchronized U2OS cells.
Sypro-Ruby stain. T/T, double thymidine; A, asynchronous cells. b, Mitotic
ligands of 14-3-3sidentified by mass spectrometry categorized into five
major groups. c, 14-3-3squantitatively immunodepletes eIF4B but not
eIF4G from mitotic (12 h), but not from interphase (0 h), U2OS cell extracts.
d, Two-dimensional far-western blotting of 14-3-3simmunoprecipitates,
with purified 14-3-3sas a probe for direct binding. pI, isoelectric point.
e, eIF4B, eIF2aand EF1aas direct mitotic ligands of 14-3-3s. Two-
dimensional far-western blots (green) re-probed with antibodies against the
indicated proteins (red). eIF4G in the 14-3-3simmunoprecipitates does not
directly bind to 14-3-3s.
LETTERS NATURE
|
Vol 446
|
15 March 2007
330
Nature
©2007 Publishing Group

than 50% (Fig. 4d, e, and Supplementary Fig. 4i). Thus, 14-3-3sis
required in normal mitosis to suppress cap-dependent translation
through binding eIF4B, allowing the cap-independent translation of
critical mitotic regulators including p58 PITSLRE kinase.
Two recent reports
26,27
have suggested that the formation of
binucleate cells as a result of cytokinesis failure is an early event in
tumour formation and underlies the subsequent development of
genomic instability. We have observed that 14-3-3s, an important
tumour suppressor protein whose expression is lost in a variety of
epithelial tumours, has a critical function in regulating protein syn-
thesis during and immediately after mitosis and is required for the
IRES-dependent translation of p58 PITSLRE, a protein kinase critical
to the proper completion of cytokinesis. These findings therefore
establish a connection between the aberrant regulation of mitotic
translation and improper cytokinesis resulting in a phenotype that
is associated with early stages of human oncogenesis.
METHODS
Immunoprecipitation studies and mass spectrometry analysis. U2OS, HeLa
and HCT116 cell lysates were prepared and immunoprecipitated overnight at
4uC with an anti-14-3-3smonoclonal antibody (CS112)
28,29
. Immunopre-
cipitates were digested with trypsin and analysed on a QSTAR XL Pro quad-
rupole time-of-flight mass spectrometer (Applied Biosystems).
Two-dimensional gel electrophoresis. 14-3-3simmunoprecipitates were
resolved by isoelectric focusing with IPGphor strips as described previously
29
.
Separation in the second dimension was performed with 12% Anderson gels and
used for far-western blot analyses with bacterially purified 14-3-3s(about
50 mgml
21
) as a probe, followed by immunoblotting with CS112.
Lentiviral shRNA constructs. 14-3-3s, 14-3-3band p58 PITSLRE shRNAs were
subcloned into pLentilox 3.7 for lentiviral production as described
30
. Heter-
geneous populations stably expressing shRNA were established in HeLa and
U2OS cells.
[
35
S]Methionine pulse-labelling. Cells were labelled with 100 mCi of
[
35
S]methionine for 20 min, lysed, and precipitated with acetone. Resuspended
α-tubulin DAPI
Multinucleate
Fused
Merged
Binucleate Fused
Control
14-3-3σ knockdown
+Rap. *
+Rap. *
+Rap. *
+Rap. *
Percentage
10 12 14 16 18 20 24 10 12
Time (h)
14 16 18 20 24
8
4
0
12
16
8
6
4
2
0
a
bc
Figure 3
|
Depletion of 14-3-3sresults in impaired cytokinesis. a, 14-3-3s
knockdown U2OS cells, stained as indicated, show increased numbers of
binucleate cells (short arrows) and cells in terminal cytokinesis
(arrowheads). Close-up views demonstrate persistent midbody structures
(long arrow), and a ‘fused’ post-mitotic cell lacking distinct intercellular
boundaries. Scale bars, 10 mm. DAPI, 4,6-diamidino-2-phenylindole.
b,c, Rapamycin treatment before mitotic entry suppresses binucleate (b) and
fused-cell (c) formation in 14-3-3sknockdown cells. Synchronized U2OS
shRNA cells were scored at the indicated times after release. Green bars,
control shRNA; black bars, 14-3-3sshRNA; 1Rap., 14-3-3sshRNA cells
treated with 20 ng ml
21
rapamycin at 9 h. Asterisk, P,0.02 for rapamycin-
treated versus untreated 14-3-3sknockdown cells. Results are means and
s.d. from three independent experiments.
ab
cd
e
shRNA
150-
kDa kDa
75-
50-
25-
Con
σ
Con
σ
Con
σ
Con
σ
1 h 12 h 16 h 20 h
012 16h h012 16
14-3-3β
Control
shRNA
150-
75-
50-
25-
14-3-3σ
Control
4 h
8 h
10 h
12 h
14 h
16 h
24 h
2n4n2n 4n
Blot
shRNA Cσβ
14-3-3β
14-3-3σ
β-actin
0 h 12 h
Con
σ
σ
Con
σ
σ
Con
σ
shRNA
Plasmid vv vv 4B4B
σrσr
0121610
012
Time (h)
169
Ratio of IRES-dependent to
cap-dependent translation
0
4
1
2
3
Blot
Blot 14-3-3σ
shRNA
shRNA
Con
σCon
σ
++– –
++––
Vector
Flag–eIF4B
Flag
β-actin
++
+––––
–++–
+
pTre vector:
pTre sigma (σr)
Con
σ
Con
σ
σ
σ
0 h 12 h
12 h
fg
0
1.0
2.0
1.5
2.5
0.5
Figure 2
|
14-3-3sknockdown cells fail to suppress cap-dependent
translation during and immediately after mitosis. a,b, Control, 14-3-3s
(a) and 14-3-3b(b) shRNA HeLa cells were pulse-labelled with
[
35
S]methionine. New protein synthesis was assayed by SDS–PAGE and
autoradiography. Numbers at the left indicate molecular masses. c, Cell cycle
distribution of control and 14-3-3sshRNA-treated cells. d, Efficiency of
14-3-3 knockdown. C, control. e, Overexpression of eIF4B overcomes
14-3-3s-dependent suppression of translation in mitosis. V, vector; 4B,
Flag-eIF4B; s
r
, RNAi-resistant 14-3-3s. Synchronized cells pulse-labelled as
in bat indicated times. f, Expression of s
r
and Flag–eIF4B in e.g, Ratio of
cap-independent to cap-dependent translation in synchronized control
(light grey bars) and 14-3-3s(dark grey bars) knockdown cells assayed by
firefly/Renilla luciferase activity at the indicated times. Top, U2OS cells and
HIV IRES; bottom, HeLa cells and p27
Kip1
IRES. Results are means and
s.e.m. from duplicate experiments.
NATURE
|
Vol 446
|
15 March 2007 LETTERS
331
Nature
©2007 Publishing Group

proteins were separated on 12% SDS–PAGE gels and analysed by autoradio-
graphy. Where indicated, cells were treated with rapamycin at a final concentra-
tion of 20 ng ml
21
9 h after release from a double thymidine block.
Additional information and detailed protocols are provided in Supplementary
Methods.
Received 13 July 2006; accepted 9 January 2007; corrected 10 April 2007.
1. Vercoutter-Edouart, A. S. et al. Proteomic analysis reveals that 14-3-3s
is down-regulated in human breast cancer cells. Cancer Res. 61, 76
–
80
(2001).
2. Ferguson, A. T. et al. High frequency of hypermethylation at the 14-3-3slocus
leads to gene silencing in breast cancer. Proc. Natl Acad. Sci. USA 97, 6049
–
6054
(2000).
3. Umbricht, C. B. et al. Hypermethylation of 14-3-3s(stratifin) is an early event in
breast cancer. Oncogene 20, 3348
–
3353 (2001).
4. Urano, T. et al. Efp targets 14-3-3sfor proteolysis and promotes breast tumour
growth. Nature 417, 871
–
875 (2002).
5. Moreira, J. M., Gromov, P. & Celis, J. E. Expression of the tumor suppressor protein
14-3-3sis down-regulated in invasive transitional cell carcinomas of the urinary
bladder undergoing epithelial-to-mesenchymal transition. Mol. Cell. Proteomics 3,
410
–
419 (2004).
6. Iwata, N. et al. Frequent hypermethylation of CpG islands and loss of expression of
the 14-3-3sgene in human hepatocellular carcinoma. Oncogene 19, 5298
–
5302
(2000).
7. Suzuki, H. et al. Inactivation of the 14-3-3sgene is associated with 59
CpG island hypermethylation in human cancers. Cancer Res. 60, 4353
–
4357
(2000).
8. Villaret, D. B. et al. Identification of genes overexpressed in head and neck
squamous cell carcinoma using a combination of complementary DNA
subtraction and microarray analysis. Laryngoscope 110, 374
–
381 (2000).
9. Gasco, M. et al. Coincident inactivation of 14-3-3sand p16INK4a is an early event
in vulval squamous neoplasia. Oncogene 21, 1876
–
1881 (2002).
10. Prescott, A. R. Synthesis of RNA and protein during mitosis in mammalian tissue
culture cells. Exp. Cell Res. 26, 260
–
268 (1962).
11. Pyronnet, S., Dostie, J. & Sonenberg, N. Suppression of cap-dependent translation
in mitosis. Genes Dev. 15, 2083
–
2093 (2001).
12. Pyronnet, S., Pradayrol, L. & Sonenberg, N. A cell cycle-dependent internal
ribosome entry site. Mol. Cell 5, 607
–
616 (2000).
13. Qin, X. & Sarnow, P. Preferential translation of internal ribosome entry site-
containing mRNAs during the mitotic cycle in mammalian cells. J. Biol. Chem. 279,
13721
–
13728 (2004).
14. Brasey, A. et al. The leader of human immunodeficiency virus type 1 genomic RNA
harbors an internal ribosome entry segment that is active during the G2/M phase
of the cell cycle. J. Virol. 77, 3939
–
3949 (2003).
15. Kullmann, M., Gopfert, U., Siewe, B. & Hengst, L. B. ELAV/Hu proteins inhibit p27
translation via an IRES element in the p27 59UTR. Genes Dev. 16, 3087
–
3099
(2002).
16. Hershey, J. W. Expression of initiation factor genes in mammalian cells. Biochimie
76, 847
–
852 (1994).
17. Holcik, M. & Sonenberg, N. Translational control in stress and apoptosis. Nature
Rev. Mol. Cell Biol. 6, 318
–
327 (2005).
18. Merrick, W. C. Cap-dependent and cap-independent translation in eukaryotic
systems. Gene 332, 1
–
11 (2004).
19. Rogers, G. W. Jr, Richter, N. J. & Merrick, W. C. Biochemical and kinetic
characterization of the RNA helicase activity of eukaryotic initiation factor 4A. J.
Biol. Chem. 274, 12236
–
12244 (1999).
20. Neef, R. et al. Phosphorylation of mitotic kinesin-like protein 2 by polo-like kinase 1
is required for cytokinesis. J. Cell Biol. 162, 863
–
875 (2003).
21. Liu, X., Zhou, T., Kuriyama, R. & Erikson, R. L. Molecular interactions of Polo-like-
kinase 1 with the mitotic kinesin-like protein CHO1/MKLP-1. J. Cell Sci. 117,
3233
–
3246 (2004).
22. Beretta, L., Gingras, A. C., Svitkin, Y. V., Hall, M. N. & Sonenberg, N. Rapamycin
blocks the phosphorylation of 4E-BP1 and inhibits cap-dependent initiation of
translation. EMBO J. 15, 658
–
664 (1996).
23. Li, T., Inoue, A., Lahti, J. M. & Kidd, V. J. Failure to proliferate and mitotic arrest of
CDK11
p110/p58
-null mutant mice at the blastocyst stage of embryonic cell
development. Mol. Cell. Biol. 24, 3188
–
3197 (2004).
24. Petretti, C. et al. The PITSLRE/CDK11p58 protein kinase promotes
centrosome maturation and bipolar spindle formation. EMBO Rep. 7, 418
–
424
(2006).
25. Cornelis, S. et al. Identification and charac terization of a novel cell cycle-regulated
internal ribosome entry site. Mol. Cell 5, 597
–
605 (2000).
26. Shi, Q. & King, R. W. Chromosome nondisjunction yields tetraploid
rather than aneuploid cells in human cell lines. Nature 437, 1038
–
1042
(2005).
27. Fujiwara, T. et al. Cytokinesis failure generating tetraploids promotes
tumorigenesis in p53-null cells. Nature 437, 1043
–
1047 (2005).
28. Brunet, A. et al. 14-3-3 transits to the nucleus and participates in dynamic
nucleocytoplasmic transport. J. Cell Biol. 156, 817
–
828 (2002).
29. Wilker, E. W., Grant, R. A., Artim, S. C. &Yaffe, M. B. A structuralbasis for 14-3-3s
functional specificity. J. Biol. Chem. 280, 18891
–
18898 (2005).
30. Rubinson, D. A. et al. A lentivirus-based system to functionally silence genes in
primary mammalian cells, stem cells and transgenic mice by RNA interference.
Nature Genet. 33, 401
–
406 (2003).
Supplementary Information is linked to the online version of the paper at
www.nature.com/nature.
Acknowledgements We thank P. Stern, D. Lowery and W. Merrick for reagents and
technical assistance. This work was supported by postdoctoral fellowships from
the Anna Fuller Fund and the NIH to E.W.W., an EMBO long-term fellowship to
M.A.T.M.v.V., the David H. Koch Cancer Research Fund, NIH grants, and a
Burroughs-Wellcome Career Development Award to M.B.Y.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. The authors declare no competing financial interests.
Correspondence and requests for materials should be addressed to M.B.Y.
(myaffe@mit.edu).
a
b
Anti-Plk1 Phalloidin DAPI Merged
Control
Mock-
treated+ Rapamycin
14-3-3σ knockdown
−
Rapamycin
0 h 10.5 h 16 h
++–– –– ––
CKDKD C KD
16 h 18 h
Blot:
anti-p58
Blot:
anti-p58
+
–– ––
Rapamycin
c
d
e
GFP
GFP
DAPI Merged
Anti-Plk1 DAPI Mer
g
ed
14-3-3σ knockdown
+p58
10
8
6
4
2
0
Percentage
Multinucleate Fused
CC CKD KDKDKDKD Control
σ shRNA
σ shRNA + p58
Figure 4
|
Depletion of 14-3-3sblocks IRES-dependent mitotic translation
of p58 PITSLRE kinase and results in failure of Plk1 to localize at the
midbody. a, Lack of localization of Plk1 to the midbody and cell fusion with
failure to complete cytokinesis in 14-3-3sknockdown U2OS cells is reversed
by rapamycin before mitotic entry. Scale bar, 5 mm. b,c, Synchronized
control (C) or 14-3-3sshRNA-treated (KD) U2OS (b) and HeLa
(c) cells, with and without rapamycintreatment, lysed at the indicated times
and immunoblotted for endogenous p58 PITSLRE expression
25
. IRES-
dependent translation of p58 PITSLRE in 14-3-3sknockdown cells was
restored by rapamycin. d,e, Forced expression of p58 PITSLRE reverts the
mitotic phenotype of 14-3-3sknockdown cells. GFP-expressing 14-3-3s
knockdown U2OS cells were transfected with p58 PITSLRE or vector
control, then imaged (d) and scored (e) 18 h after release from a double-
thymidine block. Results in eare means and s.d. from three independent
experiments.
LETTERS NATURE
|
Vol 446
|
15 March 2007
332
Nature
©2007 Publishing Group