Cold inducible RNA binding protein Source
Cold-inducible RNA-binding protein(Cirp)interacts with Dyrk1b/Mirk and promotes proliferation of immature male germ cells in mice
Tomoko Masuda a,Katsuhiko Itoh a,1,Hiroaki Higashitsuji a,Hisako Higashitsuji a,Noa Nakazawa a,Toshiharu Sakurai a,
Yu Liu a,Hiromu Tokuchi a,Takanori Fujita a,Yan Zhao a,Hiroyuki Nishiyama b,Takashi Tanaka c,Manabu Fukumoto d, Masahito Ikawa e,Masaru Okabe e,and Jun Fujita a,1
a Department of Clinical Molecular Biology,Graduate School of Medicine,Kyoto University,Kyoto606-8507,Japan;
b Department of Urology,Graduate School of Comprehensive Human Sciences,University of Tsukuba,Tsukuba3058575,Japan;
c Department of Molecular Genetics,Graduate School of Medicine,Kyoto University,Kyoto606-8507,Japan;
d Department of Pathology,Institut
e o
f Development,Aging,and Cancer,Tohoku University,Sendai980-8575,Japan;and e Department of Experimental Genome Research,Genome Information Research Center,Osaka University,Osaka565-0871,Japan
Edited by Ralph L.Brinster,University of Pennsylvania,Philadelphia,PA,and approved May24,2012(received for review December30,2011)
Cold-inducible RNA-binding protein(Cirp)was the?rst cold-shock protein identi?ed in mammals.It is structurally quite different from bacterial cold-shock proteins and is induced in response to mild,but not severe,hypothermia.To clarify the physiological function of Cirp in vivo,we produced cirp-knockout mice.They showed neither gross abnormality nor defect in fertility,but the number of undifferentiated spermatogonia was signi?cantly re-duced and the recovery of spermatogenesis was delayed after treatment with a cytotoxic agent,busulfan.Cirp accelerated cell-cycle progression from G0to G1as well as from G1to S phase in cultured mouse embryonic?broblasts.Cirp directly bound to dual-speci?city tyrosine-phosphorylation–regulated kinase1B(Dyrk1b, also called Mirk)and inhibited its binding to p27,resulting in de-creased phosphorylation and destabilization of p27.Cirp did not affect binding of Dyrk1b to cyclin D1but inhibited phosphoryla-tion of cyclin D1by Dyrk1b,resulting in cyclin D1stabilization.In the spermatogonial cell line GC-1spg,suppression of Cirp expres-sion increased the protein level of p27,decreased that of cyclin D1, and decreased the growth rate,which depended on Dyrk1b.Con-sistent changes in the protein levels of p27and cyclin D1as well as the percentage of cells in G0phase were observed in undifferen-tiated spermatogonia of cirp-knockout mice.In undifferentiated spermatogonia of wild-type mice,Cirp and Dyrk1b colocalized in the nucleus.Thus,our study demonstrates that Cirp functions to ?ne-tune the proliferation of undifferentiated spermatogonia by interacting with Dyrk1b.
I n most mammals,the testes are located in the scrotum,wherein the temperature is below that of the body cavity by2°C to7°C (1).Scrotal testes become aspermatogenic if their temperature is raised to body temperature or above,either by arti?cial heat or by retaining them in the abdomen(2,3).We previously identi?ed the?rst mammalian cold-shock protein in the testis and named it cold-inducible RNA-binding protein(Cirp,also called Cirbp or hnRNP A18)(4).It is constitutively expressed in the male germ cells and inducibly expressed in almost all cell types by lowering the culture temperature to32°C but not below20°C.Expression of Cirp is also induced by cellular stresses such as UV irradiation and hypoxia(5–7).In response to the stress,Cirp migrates from the nucleus to cytoplasm and affects expression of its target mRNAs(8,9).Cirp affects cell growth and protects cells from damages induced by tumor necrosis factorα,genotoxic stress,or cryptorchidism(4,10–13).
In an adult mouse testis,spermatogenesis is sustained by germ-line stem cells(A single)that undergo mitosis and self-renew or differentiate into committed A paired and A aligned spermato-gonia,which are collectively called undifferentiated spermato-gonia(14).Undifferentiated spermatogonia constitute the smallest population,less than1%,of total testicular cells,in-clude cells with stem-cell function,and differentiate into differ-entiating spermatogonia(A1–4,intermediate,and B),which will undergo meiosis after the?nal mitosis(15,16).Recently, methods to enrich undifferentiated spermatogonia by?ow cytometry using combinations of markers such as c-Kit andα6-integrin and the side population phenotype have been developed (17,18).
Dual-speci?city tyrosine-phosphorylation–regulated kinase1B (Dyrk1b,also called Mirk)is a member of the conserved Dyrk/ minibrain family of tyrosine-regulated,arginine-directed serine/ threonine protein kinases(19).The highest level of Dyrk1b mRNA is observed in the testis and skeletal muscle,and Dyrk1b protein mediates muscle differentiation,cell survival,and cell migration(19).Biochemical analysis has revealed that Dyrk1b phosphorylates and blocks degradation of the cyclin-dependent kinase(CDK)inhibitor p27in nontransformed cells that helps to maintain a quiescent state(20).Dyrk1b blocks quiescent cells from traversing G1phase of the cell cycle by phosphorylating and destabilizing cyclin D1and cyclin D3(21).
In the present study,we have generated Cirp-de?cient mice to clarify the biological function of mammalian cold-shock protein in vivo.We found that Cirp interacts with Dyrk1b and that it is required to fully maintain the undifferentiated spermatogonia. Results
Number of Undifferentiated Spermatogonia Is Decreased in Cirp-De?cient Mice.We generated Cirp-de?cient mice by homologous recombination in ES cells(Fig.1A).Intercrossing of heterozygous mutant mice generated progeny at nearly Mendelian ratios (cirp+/+:cirp+/?:cirp?/?=58:110:47,n=215),and cirp?/?mice were indistinguishable from wild-type littermates in appearance. When we examined fertility of cirp?/?male mice by mating them with wild-type females,the litter size(average=5.1mice,n=18) was not different from that of wild-type mating(average=4.9mice, n=15).We observed no reproductive defect in cirp?/?females, either.Although the testis of cirp?/?mice looked histologically normal and the numbers of total testicular cells and differentiating spermatogonia as well as the size and weight of the testis were comparable to wild type,the percentage and number of c-Kit?and highlyα6-integrin+undifferentiated spermatogonia in the
Author contributions:K.I.and J.F.designed research;T.M.,K.I.,Hiroaki Higashitsuji, Hisako Higashitsuji,N.N.,T.S.,Y.L.,H.T.,T.F.,Y.Z.,T.T.,M.F.,M.I.,and M.O.performed research;T.M.,K.I.,H.N.,and J.F.analyzed data;and T.M.,K.I.,M.O.,and J.F.wrote the paper.
The authors declare no con?ict of interest.
This article is a PNAS Direct Submission.
1To whom correspondence may be addressed.E-mail:jfujita@virus.kyoto-u.ac.jp or katsu@virus.kyoto-u.ac.jp.
This article contains supporting information online at https://www.wendangku.net/doc/508417337.html,/lookup/suppl/doi:10. 1073/pnas.1121524109/-/DCSupplemental.
https://www.wendangku.net/doc/508417337.html,/cgi/doi/10.1073/pnas.1121524109PNAS|July3,2012|vol.109|no.27|10885–10890C E L L B I O L O G Y
testicular epithelial cell adhesion molecule (EpCAM)+cells (18,22)were decreased in cirp ?/?mice compared with those in wild-type mice (P <0.05)(Fig.1B and C ).The number of un-differentiated spermatogonia in cirp +/?mice,however,was not different from that in wild-type mice (P =0.925).
To determine whether the observed decrease in the number of undifferentiated spermatogonia has biological meaning,we ana-lyzed the recovery process of spermatogenesis after i.p.injection of the alkylating agent busulfan (30mg/kg body weight).The per-centage of seminiferous tubules showing spermatogenesis was less than 3%and not different between wild-type and cirp ?/?mice at 5wk after the injection (Fig.1D ).At 7and 10wk,however,sig-ni ?cantly fewer tubules were reconstituted in cirp ?/?mice com-pared with wild-type mice (Fig.1E ).Consistently,the number of undifferentiated spermatogonia was signi ?cantly less in cirp ?/?mice at 4and 5wk after the busulfan treatment (Fig.1F ).
To assess whether stem cells of other cell lineages are affected by de ?ciency of Cirp,we analyzed puri ?ed bone marrow cells enriched with hematopoietic stem cells (23,24).Their numbers were comparable between cirp ?/?and wild-type mice (Fig.S1A and B ).The competitive reconstitution assay did not reveal any signi ?cant difference between them even with an additional ir-radiation of recipient mice (Fig.S1C and D ).In contrast to spermatogonia,cirp mRNA expression was barely detectable in the analyzed hematopoietic cells (Fig.S1E ).
Cirp Accelerates Cell-Cycle Progression in Cultured Mouse Embryonic Fibroblasts (MEFs).To analyze the effect of Cirp in vitro,we made
several MEF cell lines that inducibly change the protein level of Cirp (Fig.S2A and B ).As shown in Fig.2A ,cirp ?/?MEFs proliferated faster when Cirp was expressed both at 32°C and 37°C.When expression of Cirp was suppressed by shRNAs at 32°C,proliferation of cirp +/+MEFs was suppressed (Fig.2B ).The suppressive effect was reversed by expression of shRNA-insensitive Cirp construct (Fig.S2C and D ),suggesting that the observed effect was not attributable to off-target effects of shRNAs.
We next analyzed the effect of Cirp on the cell cycle by BrdU incorporation and propidium iodide staining.When exogenous Cirp was expressed in cirp ?/?MEFs at 32°C,the percentage of cells in G0/G1phase was reduced and that in S phase increased signi ?cantly (Fig.2C ).Conversely,when expression of Cirp was suppressed in cirp +/+MEFs at 32°C,the percentage of cells in G0/G1phase was increased and that in S phase reduced signif-icantly (Fig.2C ).The changes in the Cirp protein levels did not affect the percentage of the sub-G1population (Fig.S3A ).These results indicate that expression of Cirp accelerates the G0/G1-to-S phase transition without affecting apoptosis.
The cells with 2N DNA content and that are not positively stained with anti-Ki67antibody are considered to be in the G0phase of the cell cycle (25).The G0state is characterized by signi ?cantly decreased ribosomal RNA synthesis and lowered total cellular protein content.When Cirp was expressed in cirp ?/?MEFs,the percentage of cells in the fraction of Ki67low or Ki67?was signi ?cantly decreased,and the percentage of cells highly positive for pyronin Y staining,which stains polysomal RNA,was increased (Fig.2D ).Conversely,the percentage of G0cells was signi ?cantly increased when Cirp expression was suppressed in cirp +/+MEFs at 32°C (Fig.2E ).These results demonstrate that Cirp accelerates cell-cycle progression from G0to G1as well as from G1to S phase in cultured MEFs.
Cirp Binds to Dyrk1b and Inhibits Its Kinase Activity.Yeast two-hy-
brid screening of a mouse testis cDNA library using Cirp as bait suggested that the C-terminal glycine-rich domain of Cirp binds to Dyrk1b (Fig.S4A ).
When exogenously expressed as GFP-fusion proteins in HEK293T cells,full-length Cirp and its glycine-rich domain,but not its N-terminal RNA-binding domain,were coimmu-noprecipitated with Dyrk1b as expected (Fig.S4B and C ).The region of Dyrk1b that interacted with Cirp was the kinase do-main (Fig.S4D ).Binding of endogenous Cirp and Dyrk1b was detected in MEFs of the cirp +/+genotype cultured at 32°C (Fig.S4E ).The recombinant Cirp protein bound to recombi-nant Dyrk1b protein in vitro,indicating a direct
interaction
Fig.1.Reduced number of undifferentiated spermatogonia in cirp -knockout mice.(A )Targeted disruption of the cirp gene.Structures of the wild-type allele,targeting vector,and mutant allele are shown together with the relevant restriction sites.The ?lled rectangles indicate the exons of the cirp gene.Neo,TK,and arrows (a and b)indicate a pgk-neo selection cassette,a thymidine kinase cassette,and the locations of primers for genomic PCR,respectively.(B )Flow cytometric analysis of undifferentiated spermatogonia.The fractions of undifferentiated spermatogonia (boxed)were compared between wild-type (+/+)and cirp -knockout (?/?)mice.*Signi ?cantly different from +/+mice (P <0.05).Data are from four +/+and three ?/?12-wk-old mice.(C )Numbers of undifferentiated spermatogonia in +/+,?/?,and heterozygous knockout (+/?)mice.Signi ?cant decrease (*P <0.05)was seen in ?/?mice but not in +/?mice (P =0.93).Values represent mean ±SD.Data are from eight +/+,three ?/?,and four +/?mice.(D )Percentage of reconstituted seminiferous tubules after busulfan treatment.*Signi ?cantly different (P <0.05).**Signi ?cantly different (P <0.01).Values are mean ±SD (n =3each).(E )Histological sections of the testes at 7wk after busulfan treatment.Sections are stained with hematoxylin and eosin.(Scale bar:50μm.)(F )Numbers of undifferentiated spermatogonia after busulfan treatment.●and ○indicate the numbers in +/+and ?/?mice,respectively.Values represent mean ±SD (n =3–7).**Signi ?cantly different between +/+and ?/?mice (P <0.01).
10886|https://www.wendangku.net/doc/508417337.html,/cgi/doi/10.1073/pnas.1121524109
Masuda et al.
(Fig.S4F ).Immunocytochemistry demonstrated that Cirp was induced and colocalized with Dyrk1b in the nucleus at 32°C in cirp +/+MEFs (Fig.S4G ).In undifferentiated and differen-tiating spermatogonia,Cirp and Dyrk1b mRNAs were ex-pressed,and their proteins were colocalized in the nucleus (Fig.S4H and I ).
Dyrk1b phosphorylates p27and cyclin D1to affect cell-cycle progression (19).When increasing amounts of Cirp were expressed in HEK293T cells,decreasing amounts of p27were coimmunoprecipitated with Dyrk1b (Fig.3A ).In contrast,the amount of cyclin D1coimmunoprecipitated with Dyrk1b was not decreased in the presence of Cirp and its binding to Dyrk1b (Fig.S5A ).The presence of cyclin D1did not affect the binding of Cirp to Dyrk1b either,and Cirp was coimmunoprecipitated with cyclin D1as well (Fig.S5B and C ).These results indicate that Cirp forms a complex together with Dyrk1b and cyclin D1.
Dyrk1b immunoprecipitated from HEK293T cells could phosphorylate recombinant p27in vitro (Fig.3B ).The amount of phosphorylated p27was dose-dependently decreased in the presence of Cirp.When we incubated recombinant cyclin D1with recombinant Dyrk1b,phosphorylation of cyclin D1was observed (Fig.3C ).The phosphorylation was dose-dependently
suppressed by Cirp.Interestingly,Cirp was also phosphorylated by Dyrk1b.
These data demonstrate that Cirp binds to and inhibits the kinase activity of Dyrk1b toward both p27and cyclin D1,al-though Cirp competitively blocks the binding of Dyrk1b to p27but not to cyclin D1.
Effects of Cirp on Cell-Cycle Progression Are Mediated by p27and Cyclin D1via Dyrk1b.Rapid removal of p27at the G0/G1transi-
tion and expression of cyclin D1at the G1/S transition are re-quired for effective progression of the cell cycle to S phase (26,27).When Cirp was inducibly expressed with doxycycline (Dox)in cirp ?/?MEFs,expression of p27was decreased and that of cyclin D1was increased (Fig.4A )because of the increased and decreased degradation rates of p27and cyclin D1,respectively (Fig.S4J and K ).Conversely,when expression of Cirp was suppressed in cirp +/+MEFs,expression of p27was increased and that of cyclin D1was decreased (Fig.4A ).
Because p27and cyclin D1are known to be phosphorylated by kinases besides Dyrk1b (26,27),we examined whether the effects of Cirp on p27and cyclin D1depended on Dyrk1b.We stably expressed shRNAs targeted to Dyrk1b mRNAs or scrambled shRNAs in cirp ?/?MEFs that inducibly express Cirp (Fig.S5D ).Induction of Cirp decreased the protein level of
p27
Fig.2.Accelerated proliferation of MEFs expressing Cirp.(A and B )Effects of Cirp on cell growth.Cells from Cirp(?/?)/teton-Cirp clones 2,5,and 6(Cl.2,Cl.5,and Cl.6,respectively)were derived from cirp ?/?MEFs that express Cirp in the presence of Dox.Cirp(+/+)/teton-shCirp cells were derived from cirp +/+MEFs that Dox-inducibly express shRNAs that knockdown Cirp expression (shCirp-1and shCirp-2)or scrambled RNAs (shSCR).Cells were cultured in the presence (+)or absence (?)of Dox at the indicated temperatures.Cell numbers were assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)assays.Values are mean ±SD (n =3or 4).*P <0.05.**P <0.01.(C )Analysis of the cell cycle by ?ow cytometry.BrdU incorporation and DNA content were analyzed in indicated cells cultured at 32°C in the presence (+)or absence (?)of Dox.Percentages of cells in indicated phases of cell cycle are shown.Values are mean ±SD (n =3).*P <0.05.**P <0.01.(D and E )Percentages of cells in G0phase in the presence (+)or absence (?)of Dox.Ki67expression,stainability with pyronin Y,and DNA content were analyzed in indicated cells by ?ow cytometry.Values are mean ±SD (n =3).*P <
0.05.
Fig.3.Inhibition of Dyrk1b kinase activity by Cirp.(A )Effects of Cirp on binding of p27to Dyrk1b.Plasmids were cotransfected into HEK293T cells as in-dicated.Proteins immunoprecipitated (IP)with anti-FLAG antibody and inputs were analyzed by Western blotting (IB)using the indicated antibodies.+,Present;?,absent.(B )Effects of Cirp on in vitro kinase activity of Dyrk1b toward p27.Immunoprecipitants (IP)were prepared from HEK293T cells expressing FLAG-Dyrk1b by using anti-FLAG antibody.Their kinase activity was assayed in the presence of [γ-32P]ATP,GST-p27,and increasing amounts of GST-Cirp or GST.Reaction mixtures were separated by SDS/PAGE and analyzed by autoradiography.Proteins coprecipitated with protein G Sepharose beads (*)from the reaction mixtures without 32P and inputs (2%)were analyzed by IB using the indicated antibodies.(C )Effects of Cirp on in vitro kinase activity of Dyrk1b toward cyclin D1.Recombinant GST-Dyrk1b was incubated with [γ-32P]ATP,GST-cyclin D1,and increasing amounts of His-tagged Cirp and analyzed as in B .The lower images show Coomassie blue staining (CBB)of the indicated proteins used in the assay and IB of proteins coprecipitated with glutathione Sepharose beads from the reaction mixtures without 32P (af ?nity-puri ?ed).*1/4of input.Experiments were repeated three (A )and four (B and C )times,with similar results.
Masuda et al.PNAS |July 3,2012|vol.109|no.27|10887
C E L L B I O L O G Y
and increased that of cyclin D1in cirp ?/?MEFs expressing scrambled shRNAs (Fig.4B and Fig.S5E ).When expression of Dyrk1b was suppressed,however,Cirp did not show these effects in cirp ?/?MEFs.The growth rate of these cells was not affected by Cirp either (Fig.4C ).These data demonstrate that expression of Cirp accelerates cell-cycle progression by inhibiting the kinase activity of Dyrk1b toward p27and cyclin D1in MEFs.
Down-Regulation of Cirp Suppresses Cell-Cycle Progression in a Spermatogonial Cell Line in Vitro and Undifferentiated Spermatogonia in Vivo.To analyze the effect of Cirp on male germ cells,we ?rst
used the mouse spermatogonial cell line GC-1spg (28).When GC-1spg cells were transduced with control retrovirus expressing scrambled shRNAs,expression of Cirp was induced at 32°C like parental cells (Fig.S6A ).In GC-1spg cells transduced with ret-rovirus expressing shRNAs to cirp mRNAs,the induction of Cirp was partly suppressed,the level of p27protein was increased,and the level of cyclin D1protein was decreased (Fig.S6B ).These cells proliferated more slowly than the control cells (Fig.5A ).To con ?rm that growth-suppressive effect of Cirp down-regulation depends on Dyrk1b,we further down-regulated Dyrk1b expression
(Fig.S6C ).As shown in Fig.5B ,the impaired growth was rescued by suppressing the expression of Dyrk1b.
Next,we examined the effects of Cirp de ?ciency on the cell cycle in undifferentiated spermatogonia in vivo.At 4wk after injection of busulfan,the percentage of Ki67+cells was signi ?-cantly smaller and the percentage of pyronin Y low cells was sig-ni ?cantly larger in undifferentiated spermatogonia isolated from cirp ?/?mice than those from cirp +/+wild-type mice (Fig.5C ).RT-PCR analysis demonstrated that undifferentiated sper-matogonia expressed mRNAs for p27and cyclin D1(Fig.5D )as well as for Cirp and Dyrk1b.Furthermore,the protein level of p27was higher and that of cyclin D1was lower in un-differentiated spermatogonia of cirp ?/?mice compared with those of wild-type mice,although the mRNA level for p27was not different and that for cyclin D1was higher in cirp ?/?mice (Fig.5E and Fig.S7A and B ).Altogether,these data indicate that Cirp destabilizes p27and stabilizes cyclin D1at the protein level in undifferentiated spermatogonia,resulting in an acceler-ated cell-cycle progression in wild-type mice and decreased proliferation in Cirp-de ?cient
mice.
Fig.4.Effects of Cirp on p27and cyclin D1via Dyrk1b in cultured MEFs.(A )Effects of Cirp on p27and cyclin D1protein levels.Cirp(?/?)/teton-Cirp Cl.5and Cirp(+/+)/teton-shCirp-1cells were cultured at 32°C in the presence (+)or absence (?)of Dox.Cell lysates were analyzed by IB using the indicated antibodies.(B )Dyrk1b dependency of protein levels.Indicated cells,both derived from Cirp(?/?)/teton-Cirp Cl.5cells,were cultured at 37°C and analyzed as in A .(C )Dyrk1b and cell growth.Indicated cells were cultured at 32°C and 37°C in the presence (+)or absence (?)of Dox.Cell numbers were assessed by MTT assays.Values are mean ±SD (n =3).*P <0.05.**P <0.01.Experiments were repeated four (B )and three (C )times,with similar
results.
Fig.5.Impaired cell-cycle progression in Cirp-de ?cient spermatogonia.(A )Effects of Cirp on growth of cultured GC-1spg cells.Cells were transduced with retrovirus [murine stem cell virus-LTRmiR30-PIG (MSCV-LMP)]expressing RNAi sequences targeted to Cirp (Cirp-i-1and Cirp-i-2)or scrambled sequence (Cirp-SCR).Cells were cultured at 32°C,and cell numbers were assessed by MTT assays.wt,untransduced cells.Values are mean ±SD (n =3).**P <0.01.(B )Dyrk1b-dependent effects of Cirp on growth of GC-1spg cells.Cirp-i-1cells were further transduced with retrovirus [Friend spleen focus-forming virus-based hybrid vector 2(SF2)]expressing RNAi sequences targeted to Dyrk1b (Dyrk-i-1and Dyrk-i-2)or scrambled sequence (Dyrk-SCR)and analyzed as in A .(C –E )Un-differentiated spermatogonia of cirp ?/?and cirp +/+genotypes were sorted and analyzed at 4wk after busulfan treatment.(C )The percentages of Ki67+(Left )and pyronin Y ?or pyronin Y low (Right )cells were assessed by ?ow cytometry.Values are mean ±SD [n =3(Left )or 4(Right )].*P <0.05and **P <0.01,respectively,different between wild-type (+/+)and cirp -knockout (?/?)mice.(D )Detection of p27and cyclin D1transcripts in spermatogonia isolated from cirp ?/?mice.cDNAs were analyzed by RT-PCR.(E )Protein levels of p27and cyclin D1.Cell lysates were analyzed by IB using the indicated antibodies.Experiments were repeated three times (D and E ),with similar results.
10888|https://www.wendangku.net/doc/508417337.html,/cgi/doi/10.1073/pnas.1121524109
Masuda et al.
Discussion
Numbers and Proliferative Activity of Undifferentiated Spermatogonia Are Decreased in cirp-Knockout Mice.In mammals,there have been only two established cold-shock proteins reported to date,Cirp and RNA-binding motif protein3(Rbm3),which are structurally quite similar(5).In the mouse testis,Cirp is constitutively ex-pressed in the germ cells and the level varies depending on the stage of differentiation,whereas Rbm3is suggested to be expressed mainly in the Sertoli cells(3,29).When mouse testis is exposed to heat stress,expression of Cirp and Rbm3is decreased, suggesting that the decreased expression might be related to the heat-induced testicular damage.Indeed,overexpression of Cirp has recently been reported to reduce the damage by cryptorchi-dism(13).In the present study,we found that the Cirp-de?cient mice have reduced numbers of undifferentiated spermatogonia, although the total numbers of testicular germ cells and fertility were not signi?cantly decreased.When they were exposed to the cytotoxic agent busulfan,the recovery of spermatogenesis was affected slightly but signi?cantly.In addition to Cirp,Rbm3was expressed in spermatogonia(Fig.S7C and D).Because its ex-pression level was not increased in spermatogonia of the cirp?/?genotype compared with those of wild type,it seems unlikely that the relatively mild defects observed in the knockout mice are a consequence of redundancy/compensation by Rbm3.Analysis of the cirp and rbm3double-knockout mice would clarify this possibility.
A vast majority of stem cell functions reside in the un-differentiated spermatogonia population(30).The?ndings that the undifferentiated spermatogonia expressed Cirp and that proliferation of the spermatogonial cell line was suppressed when Cirp expression was decreased suggest that the observed defect in cirp?/?mouse testis is mainly intrinsic to the sper-matogonia rather than their environment.Consistent with this notion,no effect of Cirp de?ciency was observed in hemato-poietic stem cell populations that normally do not express Cirp. Cells overexpressing Cirp are more resistant to various stresses,and down-regulation of Cirp inhibits cell growth in some cancer cells(31).We found that cell-cycle progression is inhibited at both G1-to-S and G0-to-G1phase transitions with no change in the sub-G1population in Cirp-de?cient MEFs. Consistently,the proportion of undifferentiated spermatogonia in G0phase was increased in cirp?/?mice compared with wild-type mice after treatment with busulfan.The numbers of apo-ptotic testicular cells were not apparently different between cirp?/?and wild-type mice after busulfan treatment,and the sensitivity of cultured MEFs of the cirp?/?genotype to busulfan was not affected by expression of Cirp(Fig.S3
B and C).In addition,expression of Cirp stimulated proliferation of the spermatogonial cell line.Thus,the decrease in the number of undifferentiated spermatogonia in cirp?/?mice could be con-sidered mainly attributable to decreased cell proliferation rather than increased cell death.
The growth-stimulatory effects of Cirp on spermatogonia and MEFs seem contradictory to our original?nding in BALB/3T3 cells(4)and consistent with the?nding in CHO cells(32).We have also observed growth-suppressive effects of Cirp in NIH/ 3T3and U-2OS cells(Fig.S8A).Growth-stimulatory effects have been observed by others in PC3and LNCaP human pros-tate cancer cells and MEFs,but not in NIH/3T3,HeLa,IMR90, or human mammary epithelial cells(10,12).These two classes of cell lines that differently respond to Cirp expression were not distinguished by the protein levels of cyclin D1,CDK inhibitors such as p16and p27,Dyrk1b,or Cirp(Fig.S8B and C).Further studies are thus necessary to clarify this issue.Cirp probably?ne-tunes the proliferation of cells by promoting and suppressing growth signals through distinct protein–protein and protein–RNA interactions,which depend on multiple factors,including the cell types,stresses,and conditions of cells.
Cirp Binds to and Inhibits the Kinase Activity of Dyrk1b Toward p27 and Cyclin D1.Dyrk1b autophosphorylates the second tyrosine in the Tyr-X-Tyr motif,an event that is essential for full kinase activity(33).Once phosphorylated,Dyrk1b functions as a serine/ threonine kinase.Ran-binding protein M(RanBPM)directly binds to Dyrk1b and inhibits its kinase activity toward hepatocyte nuclear factor1α(HNF1α)(34).In contrast,direct binding to dimerization cofactor of HNF1from muscle(DCoHm)poten-tiates Dyrk1b’s activity as an activator of HNF1α(35).Although phosphorylation of RanBPM or DCoHm by Dyrk1b has not been detected(34),we found that Cirp directly binds to and is phosphorylated by Dyrk1b,resulting in inhibition of its kinase activity toward p27and cyclin D1.
p27opposes cell-cycle progression by inhibiting cyclin E/Cdk2 (27).The expression level of p27is high during G0phase but decreases rapidly on reentry of the cells into G1phase.Ser10is the major site of p27phosphorylation by Dyrk1b in resting cells that leads to protein stability(20).Previous studies have sug-gested a role of p27in male germ-cell development.Mice de-?cient in S-phase kinase-associated protein2(Skp2),which mediates ubiquitin-dependent degradation of p27,exhibit a pro-gressive loss in spermatogenic cells(36).Conversely,testes from p27-knockout mice are signi?cantly larger(37)and contain in-creased numbers of type A spermatogonia(38),and EpCAM+ spermatogonia from the knockout mice proliferate more actively than those from wild-type mice do(39).We found that Cirp decreases the protein level of p27and increases cell proliferation via Dyrk1b in MEFs and the spermatogonial cell line and that the p27protein level and cells in G0phase were increased in Cirp-de?cient undifferentiated spermatogonia.These results strongly suggest that the absence of Cirp increases phosphory-lation of p27by Dyrk1b,resulting in its stabilization and reduced cell-cycle progression in undifferentiated spermatogonia of the cirp?/?genotype.
D-type cyclins are G1-speci?c cyclins that associate with Cdk4 or Cdk6and promote restriction point progression during G1 phase(40).In accordance with the immunohistochemical study by Beumer et al.(41),we detected expression of cyclin D1in undifferentiated spermatogonia.The protein level of cyclin D1 was decreased in undifferentiated spermatogonia of cirp?/?mice compared with wild-type mice.Although binding to Cirp decreases the activity of Dyrk1b to phosphorylate cyclin D1as well as to p27,the binding inhibits the binding of Dyrk1b to p27 but not to cyclin D1.After phosphorylation by Dyrk1b,p27 remains a functional CDK inhibitor,does not translocate from the nucleus,and colocalizes with Dyrk1b in G0(20).Mitogen stimulation then causes cells to enter G1,reduces Dyrk1b levels (42),and initiates the translocation of p27to the cytoplasm to be degraded.Thus,induction at moderately low temperatures, colocalization in the nucleus with Dyrk1b,and inhibition of Dyrk1b’s binding to p27would contribute to the ability of Cirp to enhance degradation of p27.Cyclin D1is translocated into the cytoplasm during S phase,where it is destroyed by the protea-some after phosphorylation at Thr286by glycogen synthase ki-nase3.Dyrk1b phosphorylates cyclin D1at Thr288,which also enhances its rapid turnover(26).The nuclear complex formation of Dyrk1b,Cirp,and cyclin D1would thus inhibit the nuclear export and degradation of cyclin D1.Although Cirp itself is a substrate and seems to compete with cyclin D1for phosphory-lation(Fig.3C),it is also possible that Cirp has an independent mechanism for inhibiting Dyrk1b activity.The precise molecular mechanism by which Cirp inhibits the kinase activity of Dyrk1b remains to be elucidated.
In summary,we have demonstrated unique physiological function of the mammalian cold-shock protein Cirp that might
Masuda et al.PNAS|July3,2012|vol.109|no.27|10889C E L L B I O L O G Y
partly explain why testis should be kept cool.Cirp?ne-tunes the cell-cycle progression of undifferentiated spermatogonia as well as?broblasts and cancer cells by interacting with and suppressing kinase activity of Dyrk1b to modulate the protein levels of p27 and cyclin D1(Fig.S9).Because Cirp affects proliferation of immature male germ cells and could act as an oncoprotein in some tumors,these?ndings may pave the way for development of new diagnoses and treatments for male infertility and cancers. Materials and Methods
Generation of the Targeting Vector.The6.4-kb XbaI/EcoRI fragment of the genomic DNA clones of the cirp locus(Fig.1A)was inserted between the neomycin-resistance cassette and the HSV thymidine kinase gene,both of which are driven by the phosphoglycerate kinase1(PGK1)promoter,in the pPNT vector.The1.4-kb EcoRI/SacI fragment was inserted upstream of the neomycin-resistance gene(Fig.1A).
Isolation of Undifferentiated Spermatogonia.Isolation of undifferentiated spermatogonia was performed as described by Takubo et al.(18)with minor modi?cations.
Additional materials and experimental procedures are provided in SI Materials and Methods.
ACKNOWLEDGMENTS.We thank Dr.C.Toniatti and Dr.H.Bujard for the Bs/ IRES-M2plasmid;Dr.T.Shinohara,Dr.M.Sugai,and Dr.R.J.Mayer for helpful comments;and Dr.K.Nonoguchi for technical assistance.This work was supported in part by the Smoking Research Foundation of Japan,the Cooperative Research Project Program at the Institute of Development, Aging,and Cancer at Tohoku University;and Grants-in-Aid for Scienti?c Research(KAKENHI)from the Japan Society for the Promotion of Science.
1.Harrison RG,Weiner JS(1948)Abdomino-testicular temperature gradients.J Physiol
107:48–49.
2.Carrick F,Setchell B(1977)The evolution of the scrotum.Reproduction and Evolution,
eds Calaby J,Tyndale-Biscoe T(Australian Academy of Science,Canberra),pp165–170.
3.Nishiyama H,et al.(1998)Decreased expression of cold-inducible RNA-binding pro-
tein(CIRP)in male germ cells at elevated temperature.Am J Pathol152:289–296. 4.Nishiyama H,et al.(1997)A glycine-rich RNA-binding protein mediating cold-in-
ducible suppression of mammalian cell growth.J Cell Biol137:899–908.
5.Fujita J(1999)Cold shock response in mammalian cells.J Mol Microbiol Biotechnol
1:243–255.
6.Wellmann S,et al.(2004)Oxygen-regulated expression of the RNA-binding proteins
RBM3and CIRP by a HIF-1-independent mechanism.J Cell Sci117:1785–1794.
7.Yang C,Carrier F(2001)The UV-inducible RNA-binding protein A18(A18hnRNP)plays
a protective role in the genotoxic stress response.J Biol Chem276:47277–47284.
8.De Leeuw F,et al.(2007)The cold-inducible RNA-binding protein migrates from the
nucleus to cytoplasmic stress granules by a methylation-dependent mechanism and acts as a translational repressor.Exp Cell Res313:4130–4144.
9.Yang R,Weber DJ,Carrier F(2006)Post-transcriptional regulation of thioredoxin by
the stress inducible heterogenous ribonucleoprotein A18.Nucleic Acids Res34: 1224–1236.
10.Sakurai T,et al.(2006)Cirp protects against tumor necrosis factor-alpha-induced
apoptosis via activation of extracellular signal-regulated kinase.Biochim Biophys Acta 1763:290–295.
11.Artero-Castro A,et al.(2009)Cold-inducible RNA-binding protein bypasses replicative
senescence in primary cells through extracellular signal-regulated kinase1and2 activation.Mol Cell Biol29:1855–1868.
12.Zeng Y,Kulkarni P,Inoue T,Getzenberg RH(2009)Down-regulating cold shock
protein genes impairs cancer cell survival and enhances chemosensitivity.J Cell Bio-chem107:179–188.
13.Zhou KW,Zheng XM,Yang ZW,Zhang L,Chen HD(2009)Overexpression of CIRP may
reduce testicular damage induced by cryptorchidism.Clin Invest Med32:E103–E111.
14.Meistrich ML,van Beek MEAB(1993)Spermatogonial stem cells.Cell and Molecular
Biology of the Testis,eds Desjardins C,Ewing LL(Oxford Univ Press,Oxford,UK),pp 266–294.
15.Nakagawa T,Nabeshima Y,Yoshida S(2007)Functional identi?cation of the ac-
tual and potential stem cell compartments in mouse spermatogenesis.Dev Cell 12:195–206.
16.Oatley JM,Brinster RL(2006)Spermatogonial stem cells.Methods Enzymol
419:259–282.
17.Barroca V,et al.(2009)Mouse differentiating spermatogonia can generate germinal
stem cells in vivo.Nat Cell Biol11:190–196.
18.Takubo K,et al.(2008)Stem cell defects in ATM-de?cient undifferentiated sper-
matogonia through DNA damage-induced cell-cycle arrest.Cell Stem Cell2:170–182.
19.Friedman E(2007)Mirk/Dyrk1B in cancer.J Cell Biochem102:274–279.
20.Deng X,Mercer SE,Shah S,Ewton DZ,Friedman E(2004)The cyclin-dependent kinase
inhibitor p27(Kip1)is stabilized in G(0)by Mirk/dyrk1B kinase.J Biol Chem279: 22498–22504.
21.Deng X,Ewton DZ,Friedman E(2009)Mirk/Dyrk1B maintains the viability of quies-
cent pancreatic cancer cells by reducing levels of reactive oxygen species.Cancer Res 69:3317–3324.22.Anderson R,Schaible K,Heasman J,Wylie C(1999)Expression of the homophilic
adhesion molecule,Ep-CAM,in the mammalian germ line.J Reprod Fertil116: 379–384.
23.Osawa M,Hanada K,Hamada H,Nakauchi H(1996)Long-term lymphohematopoietic
reconstitution by a single CD34-low/negative hematopoietic stem cell.Science 273:242–245.
24.Harrison DE(1980)Competitive repopulation:A new assay for long-term stem cell
functional capacity.Blood55:77–81.
25.Kubbutat MH,et al.(1994)Epitope analysis of antibodies recognising the cell pro-
liferation associated nuclear antigen previously de?ned by the antibody Ki-67(Ki-67 protein).J Clin Pathol47:524–528.
26.Alao JP(2007)The regulation of cyclin D1degradation:Roles in cancer development
and the potential for therapeutic invention.Mol Cancer6:24–39.
https://www.wendangku.net/doc/508417337.html,rrea MD,Wander SA,Slingerland JM(2009)p27as Jekyll and Hyde:Regulation of
cell cycle and cell motility.Cell Cycle8:3455–3461.
28.Hofmann MC,Narisawa S,Hess RA,Millán JL(1992)Immortalization of germ cells and
somatic testicular cells using the SV40large T antigen.Exp Cell Res201:417–435.
29.Danno S,Itoh K,Matsuda T,Fujita J(2000)Decreased expression of mouse Rbm3,
a cold-shock protein,in Sertoli cells of cryptorchid testis.Am J Pathol156:1685–1692.
30.Nakagawa T,Sharma M,Nabeshima Y,Braun RE,Yoshida S(2010)Functional hier-
archy and reversibility within the murine spermatogenic stem cell compartment.
Science328:62–67.
31.Lleonart ME(2010)A new generation of proto-oncogenes:Cold-inducible RNA
binding proteins.Biochim Biophys Acta1805:43–52.
32.Hong JK,Kim YG,Yoon SK,Lee GM(2007)Down-regulation of cold-inducible RNA-
binding protein does not improve hypothermic growth of Chinese hamster ovary cells producing erythropoietin.Metab Eng9:208–216.
33.Lochhead PA,Sibbet G,Morrice N,Cleghon V(2005)Activation-loop autophosphor-
ylation is mediated by a novel transitional intermediate form of DYRKs.Cell121: 925–936.
34.Zou Y,Lim S,Lee K,Deng X,Friedman E(2003)Serine/threonine kinase Mirk/Dyrk1B is
an inhibitor of epithelial cell migration and is negatively regulated by the Met adaptor Ran-binding protein M.J Biol Chem278:49573–49581.
35.Lim S,Jin K,Friedman E(2002)Mirk protein kinase is activated by MKK3and functions
as a transcriptional activator of HNF1alpha.J Biol Chem277:25040–25046.
36.Fotovati A,Nakayama K,Nakayama KI(2006)Impaired germ cell development due to
compromised cell cycle progression in Skp2-de?cient mice.Cell Div1:4–13.
37.Nakayama K,et al.(1996)Mice lacking p27(Kip1)display increased body size,multiple
organ hyperplasia,retinal dysplasia,and pituitary tumors.Cell85:707–720.
38.Beumer TL,et al.(1999)Regulatory role of p27kip1in the mouse and human testis.
Endocrinology140:1834–1840.
39.Kanatsu-Shinohara M,Takashima S,Shinohara T(2010)Transmission distortion by loss
of p21or p27cyclin-dependent kinase inhibitors following competitive spermato-gonial transplantation.Proc Natl Acad Sci USA107:6210–6215.
40.Kim JK,Diehl JA(2009)Nuclear cyclin D1:An oncogenic driver in human cancer.J Cell
Physiol220:292–296.
41.Beumer TL,Roepers-Gajadien HL,Gademan IS,Kal HB,de Rooij DG(2000)In-
volvement of the D-type cyclins in germ cell proliferation and differentiation in the mouse.Biol Reprod63:1893–1898.
42.Deng X,Ewton DZ,Pawlikowski B,Maimone M,Friedman E(2003)Mirk/dyrk1B is
a Rho-induced kinase active in skeletal muscle differentiation.J Biol Chem278:
41347–41354.
10890|https://www.wendangku.net/doc/508417337.html,/cgi/doi/10.1073/pnas.1121524109Masuda et al.