Báo cáo khoa học: Polypyrimidine tract-binding protein is essential for early mouse development and embryonic stem cell proliferation potx

mouse development and embryonic stem cell proliferation Masaki Shibayama*, Satona Ohno*, Takashi Osaka, Reiko Sakamoto, Akinori Tokunaga, Yuhki Nakatake, Mitsuharu Sato and Nobuaki Yoshi

mouse development and embryonic stem cell proliferation Masaki Shibayama*, Satona Ohno*, Takashi Osaka, Reiko Sakamoto, Akinori Tokunaga,

Yuhki Nakatake, Mitsuharu Sato and Nobuaki Yoshida

Laboratory of Developmental Genetics, Center for Experimental Medicine and Systems Biology, Institute of Medical Science, University of Tokyo, Japan

Introduction

Mouse embryonic stem (ES) cells are established from

the inner cell mass (ICM) of blastocysts ES cells are

defined by their ability to give rise to a variety of

mature progeny while maintaining their capacity to

self-renew Self-renewal is the process by which a stem

cell divides to generate one or two daughter stem cells

with developmental potentials that are

indistinguish-able from that of the mother cell This process is

cen-tral to development, as well as to the maintenance of

adult tissues in complex and long-lived organisms

Self-renewal of ES cells is coordinated by multiple pathways, some of which are conserved among diverse types of stem cells, but others of which are restricted

to certain cell types or tissues [1] In some of these pathways, alternatively spliced gene products have a variety of functions across multiple developmental stages [2] In addition, computational and experimental analyses have suggested that alternative splicing is important for ES cell self-renewal and differentiation [3] However, the mechanisms by which molecules that

Keywords

cell cycle; embryonic stem cells; knockout

mouse; polypyrimidine tract-binding protein;

proliferation

Correspondence

N Yoshida, Laboratory of Developmental

Genetics, Center for Experimental Medicine

and Systems Biology, Institute of Medical

Science, University of Tokyo, 4-6-1

Shirokanedai, Minato-ku, Tokyo 108-8639,

Japan

Fax: +81 3 5449 5455

Tel: +81 3 5449 5753

E-mail: nobuaki@ims.u-tokyo.ac.jp

*These authors contributed equally to this

work

(Received 15 July 2009, revised 11

September 2009, accepted 15 September

2009)

doi:10.1111/j.1742-4658.2009.07380.x

Polypyrimidine tract-binding protein (PTB) is a widely expressed RNA-binding protein with multiple roles in RNA processing, including the splic-ing of alternative exons, mRNA stability, mRNA localization, and internal ribosome entry site-dependent translation Although it has been reported that increased expression of PTB is correlated with cancer cell growth, the role of PTB in mammalian development is still unclear Here, we report that a homozygous mutation in the mouse Ptb gene causes embryonic lethality shortly after implantation We also established Ptb) ⁄ ) embryonic stem (ES) cell lines and found that these mutant cells exhibited severe defects in cell proliferation without aberrant differentiation in vitro or

in vivo Furthermore, cell cycle analysis and a cell synchronization assay revealed that Ptb) ⁄ ) ES cells have a prolonged G2⁄ M phase Thus, our data indicate that PTB is essential for early mouse development and ES cell proliferation

Abbreviations

AP, alkaline phosphatase; E, embryonic day; EB, embryoid body; ES, embryonic stem; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; ICM, inner cell mass; IRES, internal ribosome entry site; LIF, leukemia inhibitory factor; PI, propidium iodide; PTB, polypyrimidine

tract-binding protein; SCID, severe combined immunodeficiency; SD, standard deviation; SSEA-1, stage-specific embryonic antigen-1.

regulate alternative splicing contribute to ES cell

func-tion are still elusive

Polypyrimidine tract-binding protein (PTB; also

known as PTBP1⁄ hnRNP I) is an alternative splicing

regulator that is also widely expressed also in the early

embryo [4,5] PTB regulates alternative exon inclusion

in many genes, including Ptb itself [6,7] PTB has also

been implicated in many aspects of mRNA regulation,

including polyadenylation [8], stabilization [9,10],

tran-scription [11], and localization [12,13] In addition,

PTB is involved in internal ribosomal entry site

(IRES)-dependent translation of cellular and viral

genes [14,15] PTB has two paralogs, nPTB (also

known as brPTB or PTBP2) and ROD1, which are

expressed in a tissue-restricted manner nPTB is mostly

expressed in neurons [16,17], and ROD1 is expressed

in hematopoietic cells [18]

Recently, it has been reported that increased

expres-sion of PTB is associated with ovarian tumor cell

growth [19], and that PTB differentially affects cancer

cell malignancy, depending on the cell line [20] In the

context of development, PTB has been shown to be

involved in germ cell differentiation in Drosophila

mel-anogaster [4], and is essential for the development of

Xenopus laevis [5] Although the importance of PTB

for multiple biological processes has been reported, it

is still unclear how PTB contributes to mammalian

development and organogenesis To address these

questions, we disrupted Ptb in mouse ES cells and

gen-erated Ptb knockout mice Homozygous mutation of

Ptb resulted in embryonic lethality and revealed the

importance of PTB in mouse development To

eluci-date the function of PTB in ES cells, we generated

Ptb) ⁄ ) ES cells Although Ptb) ⁄ ) ES cells are viable,

they form compact colonies and exhibit severe defects

in cell proliferation without precocious differentiation

Our data clearly demonstrate that PTB is essential for

mouse development and ES cell proliferation

Results

Homozygous mutation of Ptb leads to embryonic

lethality

Previous reports have shown that Ptb is expressed in a

wide variety of mouse tissues [16,21] and has multiple

functions in somatic cells [6,8,10,12,15] However, the

expression pattern and function of PTB in early

devel-opment have not yet been elucidated To determine the

role of PTB in mouse development, we generated

Ptb-deficient mice through targeted gene disruption

To introduce the null mutation for Ptb, we designed a

targeting vector to replace a 1.9 kb region of Ptb

on chromosome 10C1, including the promoter and transcriptional start site, with a neomycin resistance gene (Fig 1A; see detail in Doc S1) We introduced the targeting vector into E14.1 ES cells by electropora-tion, and screened G418-resistant clones for homolo-gous recombination Southern blot analysis showed that seven of 240 clones were positive for homologous recombination To generate chimeric mice, we inde-pendently injected two heterozygous ES cell clones into C57BL⁄ 6 mouse blastocysts The chimeric mice derived from both clones successfully transmitted the mutated allele, and heterozygous mutant mice were produced by breeding Both male and female Ptb+⁄) mice were fertile, and showed no apparent defects To generate Ptb) ⁄ ) mice, we intercrossed heterozygous mutant mice, and analyzed the genotypes of the offspring by Southern blot and PCR Among the 16 neonatal mice examined, no homozygous mutants were observed (Table 1), indicating that Ptb) ⁄ )embryos do not survive to birth To determine the developmental stage of lethality, we genotyped embryos from embry-onic day (E) 3.5 (blastocyst stage) to E10.5 As sum-marized in Table 1, no homozygous mutants were observed after E6.5, whereas the genotype ratio of embryos from E3.5 fitted the expected Mendelian ratio Thus, we deduced that homozygous mutation for Ptb leads to embryonic lethality shortly after implantation

Characterization of Ptb–/–blastocysts

To assess the protein expression of PTB in mouse early development, we performed immunohistochemical analysis on wild-type blastocysts We detected the immunoreactivity of PTB both in the ICM and in the trophectoderm (Fig 2A) In contrast, the expression of Oct3⁄ 4 and Cdx-2 was restricted exclusively to the ICM or the trophectoderm (Fig 2A) In order to investigate the events surrounding implantation, we performed the blastocyst outgrowth assay We cultured the blastocysts for 5 days and analyzed the genotypes

by PCR Wild-type blastocysts exhibited normal out-growth formations and were positive for alkaline phos-phatase (AP) activity (Fig 2B) In contrast, although Ptb) ⁄ ) blastocysts were positive for AP activity, the growth rate of the ICM was reduced (Fig 2B) These results suggest that PTB is essential for embryonic development during the peri-implantation period

Generation of the Ptb–/–ES cells The above data led us to analyze PTB function in ES cells, as these cells are derived from the ICM To

gain further insight into the function of PTB in ES cells, we first tried to establish Ptb) ⁄ ) ES cells from Ptb) ⁄ ) blastocysts; however, we could not obtain the Ptb) ⁄ ) ES cells (Table 2), probably owing to the cell proliferation defect Then, we attempted to disrupt both alleles of Ptb, using a conditional gene-targeting approach We constructed the second conditional tar-geting vector with a hygromycin resistance gene to mutate the wild-type allele and make heterozygous (Ptb) ⁄ flox-hyg) ES cells (Fig 1A; see detail in Doc S1)

In this vector, we designed three loxP sequences to

MC1 DT-A PGK Neo

1 kb

Targeting vector 1 Targeting vector 2

Wild-type locus (Chr 10C1)

exon 1

PGK Neo

PGK Neo

PGK Neo PGK Hyg

exon 1

Cre recombinase

Cre recombinase

+/+ +/–

–/–

26.0 kb

11.4 kb

9.1 kb

9.4 kb 7.1 kb

2.7 kb

–/flox-hyg

Sm

wild-type: 9.1 kb

1st targeting: 26.0 kb

2nd targeting: 11.4 kb

Probe A

Probe B

–/flox-hyg

–/flox

–/–

7.1 kb 5.2 kb

2.7 kb

Wild-type: 7.1 kb 1st targeting: 2.7 kb 2nd targeting: 9.4 kb

PGK Neo

exon 1

–/flox Probe B

A

B

D

C

Fig 1 Gene targeting of mouse Ptb (A) Targeting strategy of Ptb The mutated Ptb allele was generated by homologous recombination (A)

A, AflII, Bg, BglII; BH, BamHI; E, EcoRI; Sm, SmaI; Ss, Sse8387I (B) Southern blot analysis using the probes described in (A) Left panel: digested with AflII and detected by probe A Right panel: digested with BamHI and detected by probe B (C) Conditional disruption of Ptb in

ES cells Ptb) ⁄ floxES cells were generated by expression of Cre in Ptb) ⁄ flox-hygES cells Ptb) ⁄ )ES cells were generated by infection of Ptb) ⁄ floxES cells with a retroviral vector expressing Cre recombinase (D) Southern blot analysis of Ptb) ⁄ )ES cells; digested with BamHI and detected by probe B.

Table 1 Genotypes of offspring from Ptb+⁄) intercross The

het-erozygous mutant mice were intercrossed The genotypes of

off-spring were analyzed by Southern blot and PCR analysis Among

16 neonatal mice examined, no homozygous mutant was observed.

flank the hygromycin resistance gene cassette and the

1.9 kb genomic fragment containing the promoter

and transcriptional start site of Ptb (Fig 1A) We

introduced the vector into Ptb+⁄) ES cells, which we

generated using the first targeting vector We screened the hygromycin-resistant colonies for homologous recombination, and obtained positive clones, which were mutated at the wild-type allele (Ptb) ⁄ flox-hyg; Fig 1B) To generate Ptb) ⁄ )ES cells, we introduced a Cre expression vector into Ptb) ⁄ flox-hygcells by electro-poration (Fig 1C) Although we obtained Ptb) ⁄ flox cells, we failed to establish Ptb) ⁄ ) ES cells Then, we expressed Cre by retrovirus infection into Ptb) ⁄ floxcells (Fig 1C) To identify Ptb) ⁄ ) ES cells, we screened those cells by PCR and confirmed their genotypes by Southern blot analysis, and we successfully identified two independent Ptb) ⁄ ) ES cell clones () ⁄ )1 and ) ⁄ )2) (Fig 1D) The expression of Ptb mRNA and protein was completely abolished in both Ptb) ⁄ ) ES cells (Fig 5A and Fig S2)

–/– +/+

ICM

TG

Gata4

+/+ +/– –/– –/–

Fgf5

Gata6 GAPDH PTB

SSEA-1

+/+

–/–1

–/–2

Oct3/4

Cdx2 PTB

D

0

0.5

1

1.5

2

C

–/–2

–/–1

+/+ +/+–/–1–/–2 +/+–/–1–/–2 +/+–/–1–/–2

Fig 2 Characterization of blastocysts and Ptb) ⁄ )ES cells (A) Immunostaining of PTB, Oct3 ⁄ 4 and Cdx2 in wild-type blastocysts PTB is expressed in the ICM and trophectoderm (left column) Oct3 ⁄ 4 (red) and Cdx-2 (green) indicate the ICM and the trophectoderm, respectively (right column) (B) In vitro outgrowth assay of blastocysts Intercrossed embryos at E3.5 were collected and cultured for 5 days The mor-phology and AP activity of Ptb) ⁄ )blastocysts were compared with those of wild-type cells Reduced proliferation of the ICM from Ptb) ⁄ ) blastocysts was observed TG, trophoblastic giant cells (C) Quantitative real-time PCR analysis comparing the expression of undifferentiated markers in wild-type cells and two Ptb) ⁄ )ES cell clones Oct3 ⁄ 4, Sox2, Nanog and Rex-1 transcripts were normalized to Gapdh transcripts Mean values ± standard deviation (SD) were plotted from data obtained in at least three independent experiments *P > 0.05, **P > 0.005 (D) Northern blot analysis of differentiated marker expression Total RNA isolated from wild-type, heterozygous and Ptb) ⁄ )cells was hybrid-ized with radiolabeled cDNA probes Five micrograms of total RNA was loaded onto each lane (E) SSEA-1 expression in wild-type and Ptb) ⁄ )ES cells ES cells were cultured on a feeder layer The expression of SSEA-1 was maintained in both wild-type cells and the two Ptb) ⁄ )ES cell clones GAPDH, glyceraldehyde-3-phosphate dehydrogenase Scale bar: 100 lm.

Table 2 Genotypes of established ES cell lines from blastocysts.

Fifty-one blastosysts from heterozygous intercrossing were used

for ES cell derivation After 2–3 weeks of culture, ES cell lines

were established from 14 blastosysts All of the ES cell lines were

positive for AP activity Whereas heterozygous or wild-type cell

lines were obtained with the expected Mendelian ratio, no Ptb) ⁄ )

ES cell line was found.

Genotype

Total

Ptb–/–ES cells maintain the undifferentiated state

To address the expression profile of the

undifferenti-ated markers between wild-type and Ptb) ⁄ ) ES cells,

the relative abundance of selected mRNAs was

deter-mined by quantitative real-time PCR analysis The

expression of Nanog was slightly decreased in both

Ptb) ⁄ ) ES cell clones, and Rex-1 expression was

reduced in one of the Ptb) ⁄ )ES cell clones () ⁄ )2) as

compared with that of wild-type ES cells (Fig 2C)

Although it has been reported that the expression of

undifferentiated marker genes, such as Oct3⁄ 4 and

Sox2, is decreased in Nanog-deficient ES cells [22], the

expression of Oct3⁄ 4 and Sox2 mRNA was maintained and not different between the two Ptb) ⁄ )ES cell clones (Fig 2C) Furthermore, both Ptb) ⁄ )ES cell clones also expressed another ES cell marker, stage-specific embry-onic antigen-1 (SSEA-1) (Fig 2E), and were positive for AP activity (Fig 3A) In contrast, the expression of SSEA-1 was not detected in the differentiated ES cells (Fig S1) On the other hand, the northern blotting and quantitative real-time PCR analysis also showed that the expression of differentiation marker genes such as fgf5, gata4 and gata6 was not increased in wild-type cells or either of the Ptb) ⁄ )ES cell clones (Figs 2D and 4A) Taken together, these results indicate that

Ptb Tg IB:anti-PTB

vector

100

50

+/– –/–

+/+

1

0

2

3

4

6 )

1 2 3 4 5 Day

–/– –/– –/– –/– –/–

1.4

1.2

1

0.8

0.6

0.4

0.2

0

0 3 6 9

Days

+/+

–/–1

+/+

+/–

–/–1

(Serum free)

–/–2

–/–2

+/+

PI TUNEL

A

E

B

C

D

+/+

–/–

+/–

+/+

–/–1

Fig 3 Reduced proliferation of Ptb) ⁄ )ES cells (A) AP staining of wild-type and Ptb) ⁄ )ES cells The AP activities were positive in both wild-type cells and the two Ptb) ⁄ )ES cell clones Scale bar: 100 lm (B) Cell proliferation assay Cells (5 · 10 4 ) were seeded (d0) and counted every day for 5 days of culture The proliferation of Ptb) ⁄ )ES cells was reduced as compared with wild-type cells (C) Impairment

of cell proliferation seen in the Ptb) ⁄ )ES cells was rescued by ectopic expression of PTB Ptb) ⁄ )ES cell clones were stably transfected with a PTB expression vector or control plasmid, and subjected to a cell proliferation assay as in (B) Bars indicate fold increase in cell num-ber after 5 days of cell culture The amount of ectopically expressed PTB was comparable to that expressed by heterozygous ES cells (lower panel) The concentration of lysates was quantified, and the same volume was loaded into each lane Tg, transgene (D) Apoptosis assay Left, bright field; middle, PI staining; right, fluorescence-labeled DNA fragmented by terminal deoxynucleotidyl transferase TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labeling Wild-type ES cells cultured under serum-free conditions was used as a positive control for the apoptosis assay Scale bar: 200 lm (E) Proportion of viable cells Cell viability was calculated as the ratio of the number of Trypan blue-staining-negative cells to that of total cells Cells (3 · 10 5 ) were seeded on growth medium and counted every day for 3 days of culture The circles indicate the values for wild-type cells and the triangles indicate those for Ptb) ⁄ )ES cells Mean values ± SD were plotted from data obtained in experiments conducted in triplicate *P > 0.05.

although the expression of a part of ES cell-specific

markers was reduced in Ptb) ⁄ ) ES cells, both Ptb) ⁄ )

ES cell clones still remained in undifferentiated state

and did not lead to precocious differentiation

Ptb) ⁄ )ES cells exhibit reduced cell proliferation

Although Ptb) ⁄ )ES cell clones were viable and formed

typical oval-shaped compact colonies on feeder layers,

Ptb) ⁄ ) ES cell colonies were smaller than control cell

colonies (Fig 3A) A cell proliferation assay showed

that wild-type and parental Ptb+⁄)ES cells were able

to expand more than 60-fold after 5 days of culture,

whereas both Ptb) ⁄ )ES cell clones showed only a

five-fold to seven-five-fold increase in the same period (Fig 3B)

To confirm whether the reduced proliferation rate of

Ptb) ⁄ )ES cells was due to loss of PTB expression, we

introduced the PTB expression vector into Ptb) ⁄ ) ES

cells (Fig 3C) The proliferation defect seen in Ptb) ⁄ )

ES cells was recovered by PTB re-expression (Fig 3C),

suggesting that the defect in cell proliferation of Ptb) ⁄ )

ES cells was due to the loss of PTB expression As we

observed no signs of apoptosis (Fig 3D) or massive cell

death (Fig 3E) in Ptb) ⁄ )ES cell cultures, the small size

of Ptb) ⁄ )ES cell colonies and the results of our

proli-feration assay indicate a reduced proliproli-feration rate in

Ptb) ⁄ )ES cells

To further investigate the proliferative ability of

Ptb) ⁄ ) ES cells, we assessed their teratoma formation

ability in vivo We transplanted wild-type or Ptb) ⁄ )ES

cells under the kidney capsules of five severe combined

immunodeficiency (SCID) mice, and examined the

kid-neys 3 weeks after transplantation (Fig 4B) The wet

weight of teratomas resulting from transplantation

with Ptb) ⁄ ) ES cells after 3 weeks was more than

20-fold reduced as compared with wild-type teratomas

(Fig 4C) To determine whether the teratoma

forma-tion defect of Ptb) ⁄ )ES cells is due to loss of

pluripo-tency, we performed embryoid body (EB) formation

assay and quantified the expression of differentiation

marker genes by quantitative real-time PCR (Fig 4A)

The EBs were formed by suspension culture of ES cells

for 7 days without leukemia inhibitory factor (LIF)

The quantitative real-time PCR analysis revealed that

differentiation markers such as Fgf5, Gata4 and Gata6

were expressed in EBs from Ptb) ⁄ )ES cells, as well as

wild-type ES cells (Fig 4A) These results indicate that

the defect of teratoma formation from Ptb) ⁄ )ES cells

is not due to the loss of pluripotency Interestingly, the

expression levels of differentiation markers in EBs

from Ptb) ⁄ ) ES cells were higher than those in

wild-type cells, indicating that Ptb) ⁄ )ES cells may have a

greater tendency to differentiate than wild-type ES

cells Collectively, our data demonstrate that PTB is one of the critical factors for proliferation but not pluripotency of ES cells both in vitro and in vivo

5 mm +/+

–/–

0 1 2

3 3.5

2.5

1.5

0.5

0 20 40 60 80 100

+/+–/–1–/–2 –/–1+/+ +/+–/–1–/–2 –/–1 +/++/+ –/–1–/–2 –/–1+/+

Fgf5

Gata4 Gata6

*

*

A

Fig 4 Ptb) ⁄ ) ES cells have a severe defect in cell proliferation

in vivo and in vivo (A) Quantitative real-time PCR analysis compar-ing the expression of differentiated markers in wild-type and Ptb) ⁄ )

ES cells and EBs from wild-type and Ptb) ⁄ )ES cells ( ) ⁄ )1) Fgf5, Gata4 and Gata6 transcripts were normalized to Gapdh transcripts Mean values ± SD were plotted from data obtained in at least three independent experiments *P > 0.05 (B) Teratoma formation

by ES cell transplantation Wild-type or Ptb) ⁄ )ES cells were trans-planted under the kidney capsules of SCID mice Three weeks after transplantation, teratoma formation was examined In four of five mice transplanted with wild-type ES cells, a teratoma formed around the kidney (right) However, no teratoma formation was observed in the five mice transplanted with Ptb) ⁄ )ES cells (left) Scale bar: 5 mm (C) Wet weight of teratomas The average wet weight of teratomas resulting from transplantation with wild-type

ES cells was approximately 2.3 g Mean values ± SD were plotted from data obtained in experiments conducted in triplicate.

*P > 0.01.

Ptb–/–ES cells have prolonged G2/M progression

To further characterize the reduced proliferation

phe-notype seen in Ptb) ⁄ ) ES cells, we measured the

expression of several well-known cell cycle regulators

by western blot analysis (Fig 5A) Although it has

been reported that PTB modulates the G1 to S

transi-tion through enhancement of IRES-dependent

transla-tion of p27kip1in differentiated cells such as 293T cells

[23], the protein level of p27kip1in Ptb) ⁄ )ES cells was

not different from that in wild-type ES cells (Fig 5A)

Moreover, no alterations in cyclin A, B or E protein

expression were found in Ptb) ⁄ ) ES cells (Fig 5A)

These results indicate that the cause of the

prolifera-tion defect in Ptb) ⁄ ) ES cells is not aberrant

expres-sion of these cell cycle regulators To further

investigate the mechanism of the cell proliferation

defect seen in Ptb) ⁄ )ES cells, we performed cell cycle

analysis We fixed and stained cells with propidium

iodide (PI), after which we analyzed the DNA content

by flow cytometry (Fig 5B) The peak of the cell

pop-ulation mapped in the G2⁄ M phase was higher in

Ptb) ⁄ ) ES cells than in wild-type ES cells (Fig 5B)

This result suggests that the cause of the proliferation defect in Ptb) ⁄ ) ES cells may be G2⁄ M phase delay

We next analyzed cell cycle progression in Ptb) ⁄ ) ES cells by arresting the cells in the early S phase with a double thymidine block We released cells from the block and fixed them at the time points indicated in Fig 5C, and then analyzed the DNA contents of the cells by flow cytometry Up until 4 h after the release, the DNA content patterns were essentially the same in Ptb) ⁄ ) and wild-type ES cells, and cells were at the end of the S phase in this period (Fig 5C, shaded in gray) These results indicate that progression through the S phase is not affected by PTB deficiency How-ever, the number of cells returning to the G1 phase through the G2⁄ M phase was smaller in Ptb) ⁄ ) ES cells than in control cells, as seen at 8 h after release (Fig 5C, indicated by arrows) As the pattern of DNA contents at 8 h after release in Ptb) ⁄ )ES cells was the same as that after 6 h in control cells, we estimated the delay in G2⁄ M progression in Ptb) ⁄ ) ES cells to be approximately 2 h Taking these data together, we conclude that the proliferation defect in Ptb) ⁄ ) ES cells is a result of delayed G2⁄ M progression

Fig 5 G2⁄ M progression is delayed in Ptb) ⁄ )ES cells (A) Expression of cell cycle-related proteins in Ptb) ⁄ )ES cells Expres-sion of p27, cyclin A, cyclin B and cyclin E was examined by western blotting Whole cell extracts from wild-type, heterozygous and Ptb) ⁄ )ES cells were subjected to SDS ⁄ PAGE No significant difference was observed in Ptb) ⁄ )ES cells (B) Cell cycle analysis of asynchronous Ptb) ⁄ )ES cell populations by flow cytometry Cells were fixed in ethanol and stained by PI The percentage of cells in the G2⁄ M stage is described in the histograms %G1: + ⁄ +, 10.0%; ) ⁄ )1, 8.99%; ) ⁄ )2, 9.76% %S: + ⁄ +, 67.9%; ) ⁄ ), 64.2%; ) ⁄ )2, 65.9% The experiment was independently repeated

at least three times (C) Cell synchronization assay for cell cycle progression analysis Wild-type and Ptb) ⁄ )ES cells were synchro-nized by a double thymidine block Cells were fixed at the indicated time points after release, and the DNA content of the cells was analyzed by flow cytometry Arrows indicate differences in G 1 peak appearance between wild-type and Ptb) ⁄ )ES cells.

We have shown that PTB, which has multiple functions

in RNA metabolism, is an essential factor in mouse

early development and ES cell proliferation To assess

the function of PTB in vivo and in vitro, we used a

strat-egy in which Ptb was mutated by homologous

recombi-nation, and determined that the Ptb knockout mice

exhibited embryonic lethality shortly after implantation

(Table 1) We then established two Ptb) ⁄ )ES cell lines,

and found that Ptb) ⁄ )ES cells showed severe defects in

cell proliferation in vivo and in vitro (Figs 3B and 4B)

As Ptb) ⁄ ) ES cells exhibit a low proliferation rate,

Ptb) ⁄ ) ES cells may not be established from Ptb) ⁄ )

blastocysts or Cre-transfected Ptb) ⁄ flox-hygro ES cells

Lower proliferation rates are also found in

sall4-dis-rupted, klf5-dissall4-dis-rupted, HDAC1-dissall4-dis-rupted,

ronin-dis-rupted and dicer-disronin-dis-rupted ES cells relative to wild-type

ES cells, and mice with knockout of these genes also

show embryonic lethality at the peri-implantation stage

[24–28], a phenotype similar to that of the Ptb knockout

mice Although the phenotypes of ES cells with

disrup-tion of these genes differ, these reports suggest that a

lower ES cell proliferation rate can cause critical defects

in embryonic development As PTB is expressed in both

the ICM and the trophectoderm, we could not exclude

the possibility of a failure of implantation due to

defec-tive trophectoderm development, as in the case of the

klf5knockout mice [25] In klf5) ⁄ )ES cells, expression

of differentiation-related genes and spontaneous

differ-entiation are increased [25] However, these phenotypes

are not observed in Ptb) ⁄ )ES cells Furthermore, the

expression of Oct3⁄ 4 was not disturbed in Ptb) ⁄ ) ES

cells, and this is different from what is seen in sall4) ⁄ )

ES cells [24] These reports suggest that regulation of

proliferation occurs through more than one mechanism

in ES cells One likely reason for the embryonic lethality

of the Ptb knockout mice is the prolonged G2⁄ M

progression seen in Ptb) ⁄ )ES cells (Fig 5B,C) As

pro-posed in a recent review [29], mitosis is a key process in

which transcriptional programs are altered From our

results showing that Ptb) ⁄ )ES cells have a prolonged

G2⁄ M phase and Ptb knockout mice exhibit embryonic

lethality, it appears that irregular control of the mitotic

phase may affect nuclear reorganization processes,

resulting in loss of control of transcriptional programs

This difference in developmental regulation may also

apply to the mechanisms of promiscuous gene

expres-sion and other phenotypes seen in cancer cells In

ovar-ian cancer, a high level of expression of PTB is

correlated with tumor cell growth and malignancy

[19,20] This may be due to disruption of the gene

expression program in tumor cells resulting from

augmented PTB expression Taken together, these data suggest that PTB is a key factor in switching of cell identity through mitotic phase modulation PTB is a multifunctional protein that is involved in transcription, polyadenylation, alternative splicing, and IRES-depen-dent translation, and these steps are all known to be targets for mitotic inhibition [30,31] The regulatory mechanism of the ES cell cycle is still unclear We are currently investigating whether PTB is one of the important regulators for the G2⁄ M phase in ES cells Cell proliferation and differentiation are highly coor-dinated processes during development, and it is well known that, in many systems, terminal differentiation

is coupled with growth arrest The low proliferation rate may be responsible for the rapid differentiation potential of Ptb) ⁄ ) ES cells, and result in higher expression levels of differentiated marker genes in EBs from Ptb) ⁄ )ES cells than in EBs from wild-type cells (Fig 4A) The expression of the undifferentiated stem cell marker Nanog is downregulated in both Ptb) ⁄ )ES cell clones (Fig 2C) We observed that recombinant PTB protein can bind to a pyrimidine-rich sequence in the Nanog promoter region (Y Nakatake, unpublished data) These data suggest that PTB may partially regu-late the expression of Nanog The difference in Nanog expression between the two Ptb) ⁄ )ES cell clones may

be due to the effect of factor(s) other than PTB As the expression of Rex-1 is regulated by Nanog [32,33], the reduction of Rex-1 expression in Ptb) ⁄ ) ES cells () ⁄ )2) may be caused by downregulation of Nanog In the other clone () ⁄ )1), the expression level of Nanog may be enough to activate Rex-1 expression Although the expression levels of Nanog and Rex-1 are different between the two Ptb) ⁄ ) ES cell clones, we did not observe any differences in phenotypes such as prolifer-ation (Fig 3B), apoptosis (Fig 3D), or undifferenti-ated state (Figs 2E and 3A) Furthermore, in Ptb) ⁄ )

ES cells, we did not observe any spontaneous differen-tiation (Figs 2D and 4A) or downregulation of Oct3⁄ 4 (Fig 2C), as is seen in Nanog-deficient ES cells [22] Collectively, these data suggest that the phenotypes resulting from the absence of PTB are due to a distinct mechanism that is independent of Nanog and Oct3⁄ 4 PTB regulates nonsense-mediated decay of transcripts

of nPTB, which is one paralog of PTB [34] We investi-gated whether the expression of nPTB was increased in Ptb) ⁄ )ES cells (Fig S2) The level of nPTB in Ptb) ⁄ )

ES cells was higher than in wild-type ES cells Although

it has been reported that PTB and nPTB have functional overlap [35] in HeLa cells, the increase of nPTB expres-sion did not rescue the proliferation defect in ES cells Our study has revealed the importance of PTB in cell proliferation Questions that still need to be answered

are what the identity is of the target protein regulated

by PTB in the mitotic phase and how this target protein

modulates mitosis and cell proliferation The answers to

these questions will provide novel insights into gene

reg-ulation through mitosis Another interesting approach

would be to clarify the significance of PTB in cells

with-out a mitotic cycle Heart and brain tissues may be

interesting in this respect, as they express PTB [16,21]

but do not engage in massive cell growth These

experi-ments are now possible, owing to our establishment of

conditional targeting of Ptb in mice The molecular

mechanisms of PTB regulation of early mouse

develop-ment and ES cell proliferation are important questions

that are worthy of further investigation

Experimental procedures

Cell culture

ES cells were cultured in DMEM (Nissui, Tokyo, Japan)

supplemented with LIF, 15% fetal bovine serum, 100 nm

2-mercaptoethanol, 0.06% l-glutamine, and glucose (to a final

concentration of 4500 mgÆL)1) Mouse embryonic fibroblasts

were maintained in DMEM supplemented with 10% fetal

bovine serum and 0.06% l-glutamine Hygromycin-resistant

MEFs were prepared from mice generously provided by

Y Iwakura (IMSUT, Japan)

Proliferation assay, apoptosis assay, and AP

staining

For the proliferation assay, 3· 105cells were seeded in

growth medium and counted every day over 3 days of

cul-ture Viable and total cells were counted with and without

Trypan blue solution The value of relative viable cells was

calculated as the ratio of the number of Trypan

blue-nega-tive cells to that of total cells The apoptosis assay was

per-formed using an ApopTag Fluorescein Direct In Situ

Apoptosis Detection Kit (Chemicon), following the

manu-facturer’s instructions AP staining was performed using an

AP leukocyte kit (Sigma-Aldrich, St Louis, MO, USA),

following the manufacturer’s instructions

PTB expression vector and plasmid transfection

The coding sequence for PTB was obtained by PCR

ampli-fication using relevant primers (Table S1) The resulting

cDNA fragment was digested with HindIII and SlaI, and

then subcloned into pBluescript II (Stratagene, La Jolla,

CA, USA) and sequenced For the PTB expression vector,

the cDNA was ligated into pBPCAGGS, in which the

pHPCAGGS hygromycin resistance gene cassette (kindly

provided by H Niwa, RIKEN, Japan) was replaced with

a blasticidin resistance gene cassette from pcDNA6⁄ TR

(Invitrogen, Carlsbad, CA, USA) Linearized PTB expres-sion vector or pBPCAGGS was then transfected into cells with Lipofectamine2000 (Invitrogen), and the cells were selected in the presence of 3 lgÆmL)1 blasticidin (Invivo-Gen, San Diego, CA, USA) for 5 days

Northern blotting

Total RNA was isolated by ultracentrifugation [36] or extracted using sepasol RNA I (Nacalai Tesque, Kyoto, Japan) Agarose gel electrophoresis and blotting were per-formed as previously reported [37] Hybridization and washing of the blotted filter were performed according to previously described methods [38] Probes for Fgf5, Gata4 and Gata6 were obtained by PCR amplification Primer sequences are described in Table S1 cDNA templates for probes were synthesized by SuperScriptII⁄ III (Invitrogen) according to the manufacturer’s instructions

Quantitative real-Time PCR analysis

For the RT-PCR analysis, first-strand cDNA was synthe-sized from 1 lg of total RNA that had been treated with DNase I in 10 lL of reaction mixture using the High Capac-ity RNA-to-cDNA Kit (ABI, Foster CCapac-ity, CA, USA) The quantitative real-time PCR reaction was performed with a Fast SYBR Green Master Mix (ABI) and analyzed on a Ste-pOnePlus (ABI) Relative gene expression was calculated using the standard curve method The sequences of primers for quantitative real-time PCR are listed in Table S1

Antibodies and immunodetection

Rabbit anti-Oct3⁄ 4 (Santa Cruz Biotechnology, Santa Cruz,

CA, USA), rabbit anti-Oct3⁄ 4 [39], mouse anti-PTB (Zymed, Invitrogen), rabbit anti-SSEA-1 (Chemicon, Millipore, Billerica, MA, USA), mouse anti-p27kip1 (BD Pharmingen, Franklin Lakes, NJ, USA), rabbit anti-cyclin A (Santa Cruz Biotechnology) and rabbit anti-cyclin E (Santa Cruz Biotech-nology) sera were used for immunodetection For immuno-fluorescent staining, Alexa Fluor 488 anti-rabbit IgG (Molecular Probes, Invitrogen) and Alexa Fluor 562 mouse IgG (Molecular Probes) were used as secondary anti-bodies For western blotting, horseradish peroxidase-linked anti-mouse IgG and anti-rabbit IgG (GE Healthcare, Chal-font St Giles, UK) were used Immunoreactivity was detected using an enhanced chemiluminescence kit (GE Healthcare) and X-ray film (Fuji Film, Kanagawa, Japan)

Cell cycle analysis

A double thymidine block was performed as follows Thy-midine (MP Biomedicals, Illkirch, France) was added to each ES cell culture to a final concentration of 2 mm After

16 h, the cells were washed twice with NaCl⁄ Piand released

for 8 h in growth medium A second block was initiated by

adding thymidine to a concentration of 2 mm and was

maintained for 16 h Cells were washed twice with NaCl⁄ Pi,

released in fresh growth medium for the indicated periods

of time, and then fixed in cold 70% ethanol Fixed cells

from the double thymidine block were treated with

5 mgÆmL)1 RNaseA (Sigma, St Louis, MO, USA) and

50 lgÆmL)1 PI (Nacalai Tesque) for 30 min at room

tem-perature Cell cycle analysis was carried out using a

FAC-SCalibur (Becton Dickinson, Franklin Lakes, NJ, USA)

and flowjo software (TreeStar, Ashland, OR, USA)

Mice and teratoma formation

C57BL⁄ 6J mice and MCH:ICR mice were purchased from

CLEA Japan (Tokyo, Japan) All of the mice were

main-tained under specific pathogen-free conditions in the animal

facility of the IMSUT, the University of Tokyo For

tera-toma formation, wild-type or Ptb) ⁄ )ES cells were suspended

in NaCl⁄ Piand transplanted (3· 105

cells per kidney) under the kidney capsules of adult male C.B-17⁄ Icr scid Jcl mice

(CLEA Japan) Three weeks after transplantation, the

kid-neys were collected and examined All of the work with mice

conformed to guidelines approved by the Institutional

Ani-mal Care and Use Committee of the University of Tokyo

Acknowledgements

We thank R Ku¨hn for providing us with E14.1 ES

cells, H Niwa for the pHPCAGGS plasmid and rabbit

anti-Oct3⁄ 4 serum, and Y Iwakura for

hygromycin-resistant mouse embryonic fibroblasts This research

was supported by a Research Grant (2000–2004, to N

Yoshida) for the Future Program (‘Mirai Kaitaku’)

from the Japanese Society for the Promotion of

Science (JSPS) and by grants from the Ministry of

Education, Culture, Sports, Science and Technology

of Japan (to N Yoshida and M Sato)

References

1 Molofsky AV, Pardal R & Morrison SJ (2004) Diverse

mechanisms regulate stem cell self-renewal Curr Opin

Cell Biol 16, 700–707

2 Mattaj I & Hamm J (1989) Regulated splicing in early

development and stage-specific U snRNPs Development

105, 183–189

3 Pritsker M, Doniger TT, Kramer LC, Westcot SE &

Lemischka IR (2005) Diversification of stem cell

molec-ular repertoire by alternative splicing Proc Natl Acad

Sci USA 102, 14290–14295

4 Robida MD & Singh R (2003) Drosophila

poly-pyrimidine-tract binding protein (PTB) functions

specifically in the male germline EMBO J 22, 2924– 2933

5 Hamon S, Le Sommer C, Mereau A, Allo MR & Hardy S (2004) Polypyrimidine tract-binding protein is involved in vivo in repression of a composite inter-nal⁄ 3¢-terminal exon of the Xenopus alpha-tropomyosin Pre-mRNA J Biol Chem 279, 22166–22175

6 Black DL (2003) Mechanisms of alternative pre-messen-ger RNA splicing Annu Rev Biochem 72, 291–336

7 Wollerton MC, Gooding C, Wagner EJ, Garcia-Blanco

MA & Smith CW (2004) Autoregulation of polypyrimi-dine tract binding protein by alternative splicing leading

to nonsense-mediated decay Mol Cell 13, 91–100

8 Castelo-Branco P, Furger A, Wollerton M, Smith C, Moreira A & Proudfoot N (2004) Polypyrimidine tract binding protein modulates efficiency of polyadenylation Mol Cell Biol 24, 4174–4183

9 Knoch KP, Bergert H, Borgonovo B, Saeger HD, Altkruger A, Verkade P & Solimena M (2004) Poly-pyrimidine tract-binding protein promotes insulin secretory granule biogenesis Nat Cell Biol 6, 207–214

10 Kosinski PA, Laughlin J, Singh K & Covey LR (2003)

A complex containing polypyrimidine tract-binding pro-tein is involved in regulating the stability of CD40 ligand (CD154) mRNA J Immunol 170, 979–988

11 Rustighi A, Tessari MA, Vascotto F, Sgarra R, Giancotti V & Manfioletti G (2002) A polypyrimi-dine⁄ polypurine tract within the Hmga2 minimal promoter: a common feature of many growth-related genes Biochemistry 41, 1229–1240

12 Cote CA, Gautreau D, Denegre JM, Kress TL, Terry

NA & Mowry KL (1999) A Xenopus protein related to hnRNP I has a role in cytoplasmic RNA localization Mol Cell 4, 431–437

13 Zang WQ, Li B, Huang PY, Lai MM & Yen TS (2001) Role of polypyrimidine tract binding protein in the function of the hepatitis B virus posttranscriptional regulatory element J Virol 75, 10779–10786

14 Cornelis S, Tinton SA, Schepens B, Bruynooghe Y & Beya-ert R (2005) UNR translation can be driven by an IRES element that is negatively regulated by polypyrimidine tract binding protein Nucleic Acids Res 33, 3095–3108

15 Mitchell SA, Brown EC, Coldwell MJ, Jackson RJ & Willis AE (2001) Protein factor requirements of the Apaf-1 internal ribosome entry segment: roles of polypyrimidine tract binding protein and upstream of N-ras Mol Cell Biol 21, 3364–3374

16 Polydorides AD, Okano HJ, Yang YY, Stefani G & Darnell RB (2000) A brain-enriched polypyrimidine tract-binding protein antagonizes the ability of Nova to regulate neuron-specific alternative splicing Proc Natl Acad Sci USA 97, 6350–6355

17 Kikuchi T, Ichikawa M, Arai J, Tateiwa H, Fu L, Higuchi K & Yoshimura N (2000) Molecular cloning and characterization of a new neuron-specific