Molecular Cell
Article
Pyruvate Kina M2Regulates Gene Transcription by Acting as a Protein Kina
Xueliang Gao,1,3Haizhen Wang,1,3Jenny J.Yang,2Xiaowei Liu,1and Zhi-Ren Liu 1,*
1Department
of Biology 2Department of Chemistry
Georgia State University,Atlanta,GA 30303,USA 3The authors contributed equally to this work *Correspondence:zliu8@gsu.edu DOI 10.lcel.2012.01.001
SUMMARY
Pyruvate kina isoform M2(PKM2)is a glycolysis enzyme catalyzing conversion of phosphoenolpyr-uvate (PEP)to pyruvate by transferring a phosphate from PEP to ADP.We report here that PKM2localizes to the cell nucleus.
校园生活二三事The
levels of nuclear PKM2corre-late with cell proliferation.PKM2activates transcrip-tion of MEK5by phosphorylating stat3at Y705.In vitro phosphorylation assays show that PKM2is a protein kina using PEP as a phosphate donor.ADP competes with the protein substrate binding,indicating that the substrate may bind to the ADP site of PKM2.Our experiments suggest that PKM2promotes cell prolifer-ation,indicating that protein kina activity of PKM2plays a role in promoting cell proliferation.Our study reveals an important link between metab-olism alteration and gene expression during tumor transformation and progression.
INTRODUCTION
An important molecular feature of tumor development is the expression of glycolysis enzyme pyruvate kina isoform M2(PKM2)(Elbers et al.,1991;Hacker et al.,1998).Catalytically active PKM2exists as a tetramer and associates with a glycolytic enzyme complex (Altenberg and Greulich,
周春梅法官
2004;Dombrauckas et al.,2005;Zwerschke et al.,1999).In tumor cells,PKM2forms a dimer and appears to be catalytically inactive for conversion of PEP to pyruvate (Ashizawa et al.,1991;Mazurek et al.,2005).It is believed that the inactive PKM2actually provides growth advan-tage for tumor progression as it helps to channel the carbon source from glycolytic intermediates to biosynthesis,especially synthes of nucleic acids,lipids,and proteins,to meet the demands for tumor cell proliferation.Recent studies have suggested that PKM2changes its pyruvate kina activity,conquently facilitating cell proliferation by binding to tyrosine phosphopeptide,or becoming phosphorylated under growth stimulation (Christofk et al.,2008a,2008b;Hitosugi et al.,
2009),indicating that growth signals regulate the activity of PKM2in gluco metabolism.
It has long been noted that veral enzymes that act in different metabolic pathways posss so-called moonlight activity (Jeffery,1999).The moonlight activities may play important roles in a number of physiological and pathological process (Sriram et al.,2005).Using the pyruvate kina M2purified from chromatin extracts of hepatoma tumor cells,Jan Ignacak and coworkers noticed that histone H1is capable of activation of the enzymatic reaction of PKM2of conversion of PEP to pyruvate in the abnce of ADP (Ignacak and Stachurska,2003).Although the study did not demonstrate the protein kina activity of PKM2,it suggested that PKM2may have an enzymatic acti
vity that is not directly involved in transferring phosphate from PEP to ADP.Recently,veral independent studies demonstrated that PKM2localized in cell nucleus in respon to different signals (Hoshino et al.,2007;Stetak et al.,2007).The nuclear PKM2is shown to participate in regulation of gene transcriptions (Lee et al.,2008;Luo et al.,2011;Yang et al.,2011).Interestingly,PKM2can function as a coactivator of HIF-1a (Luo et al.,2011).Thus,it is suggested that PKM2may act in a feedback loop in respon to hypoxia condition to repro-gram cancer cell metabolism.Furthermore,growth stimulation by EGFR induces PKM2nuclear translocation and activates the activity of PKM2in the gene transcription regulation (Yang et al.,2011),suggesting that the role of PKM2in gene transcrip-tion is regulated by growth stimulations.
We report here that PKM2can act as a protein kina.PKM2dimer is an active protein kina,while the tetramer is an active pyruvate kina.The nuclear PKM2is a dimeric form.The nuclear PKM2directly phosphorylates stat3independent of JaK2and c-Src pathways.Phosphorylation of stat3by nuclear PKM2leads to activation of transcription of a number of stat3targeted genes.Expression of a PKM2mutant R399E that exists as dimer dramatically promotes cell proliferation and tumor growth,indicating the important role of the dimeric PKM2in cancer progression.RESULTS
虾仁烧豆腐
Nuclear PKM2Activates Transcription of mek5by Phosphorylating stat3
We and others have obrved that the PKM2localizes to the cell nucleus (Hoshino et al.,2007;Schneider et al.,2002;Stetak
598Molecular Cell 45,598–609,March 9,2012ª2012Elvier Inc.
et al.,2007).To understand the functional significance of nuclear localization of PKM2,we examined nuclear PKM2levels in ten cancer cell lines by immunoblot using an antibody raid against a peptide spanning aa 399–412of PKM2(PabPKM2).The ten cell lines reprent different cancer progression stages and are derived from cancers of different organs/tissue types.Cell prolif-eration analys demonstrated that M4C1,SW620,WM266,and H146are more proliferated than their corresponding lines in the matched pairs (e Figure S1A available online).Immunoblot analys indicated much higher nuclear levels of PKM2in the more proliferated cancer cells than tho in the corresponding less proliferated cells in the matched pair (Figure 1A and Fig-ure S1B).The higher nuclear PKM2levels were not becau of differences in PKM2expression (Figure 1A).The detection of PKM2in the nuclear extracts was not due to contamination of cytoplasmic proteins,as demonstrated by lack of GAPDH in the immunoblot analys of the nuclear extracts (Figure S1B).To
explore a possibility that nuclear PKM2is involved in regu-lation of gene expressions,we carried out gene expression array analys with SW620cells expressing HA-PKM2.Expressions of 350genes were upregulated,and expressions of 359genes were downregulated,by at least 2-fold (Table S1).Among tho affected genes,expression of mek5was affected by over 6-fold.Expression of mek5plays an important role in SW620cell prolif-eration.Thus,we verified the role of PKM2in expression of mek5by RT-PCR.Consistent with the gene array analys,RT-PCR demonstrated a dramatic change in the mek5expression in SW620cells but only a marginal increa in SW480cells,a colon cancer cell line that is derived from the same patient with sub-stantially lower proliferation rate,following the expression
of
Figure 1.PKM2Regulates mek5Transcrip-tion
(A)Nuclear localization of PKM2in different cancer cells.(Left)Immunoblot analys of PKM2using the antibody PabPKM2(IB:PKM2)in nuclear ex-tracts (NE),cytoplasmic extracts (Cyto),and whole-cell lysates (WCL)of ten cell lines (indi-cated).Blot of GAPDH (IB:GAPDH)in nuclear extracts is a control indicating no cytoplasmic protein contaminations.(Right)Quantization of the nuclear and cytoplasmic PKM2in different cell lines bad on the immunoblots.The results are prented as the ratio of nuclear versus cyto-plasmic (NE/Cyto)levels.
(B)(Left)RT-PCR analys of cellular mek5mRNA in two pairs of cancer cells (indicated)in which PKM2was expresd (HAPK)using the adenoviral expression kit.The results are prented as fold increa in PCR products before and 48hr after PKM2expression in the same cells.The level of PCR product from each cell line before PKM2expression was defined as 1.(Right)The levels of expresd HA-PKM2(HAPK)in the nuclear (NE)and cytoplasm (Cyto)in four cell lines (indicated)were analyzed by immunoblot of nuclear or cyto-plasmic extracts using anti-HA antibody (IB:HA).Vec means that the cells express vector alone.(C)(Left)Expressions of mek5(IB:MEK5)in SW480and SW620cells in
which PKM2was knocked down (PK)were analyzed by immunoblot using antibody against mek5.The cellular levels of PKM2were analyzed by immunoblot of PKM2(IB:PKM2).NT means the cells were transfected with nontarget siRNA.(Middle)RT-PCR analys of cellular mek5mRNA levels and (right)immuno-blot analys of the cellular PKM2levels (IB:PKM2)in cancer cells (indicated)in which PKM2was knocked down (PK).The results of mRNA levels are prented as fold increa in PCR products before and 48hr after PKM2expression in the same cells.The level of PCR product from each cell line before PKM2expression was defined as 1.(D)ChIP of the mek5promoter (MEK5promoter)using antibody against PKM2in SW620cells
(a PK).ChIP using rabbit IgG was a negative control.ChIP using PCR primer pair targeting a region of exon 1(MEK5exon 1)of mek5gene (nt 342–542)and targeting GAPDH promoter (GAPDH promoter)using antibody against PKM2were negative controls.The primer pair positions are indicated.Inputs were PCR products from DNA extracts without ChIP.Error bars in (B)and (C)are standard deviations of three measurements.
In (A)–(C),immunoblot of b -actin (IB:b -actin),immunoblots of H2B (IB:H2B),and immunoblots of Lamin A/C (IB:Lamin A/C)are loading controls.
Molecular Cell
PKM2Phosphorylates Protein Substrate
alookMolecular Cell 45,598–609,March 9,2012ª2012Elvier Inc.599
PKM2.Consistently,higher nuclear levels of exogenously ex-presd HA-PKM2were obrved in more proliferated cells (Fig-ure 1B).Similar patterns were also obrved with another pair of melanoma cell lines,WM115and WM266.To further verify the role of PKM2in transcriptional regulation of mek5,PKM2was knocked down in SW620and SW480cells (Figure 1C).Quantita-tive RT-PCR analys demonstrated a strong reduction in cellular mek5mRNA (Figure 1C).We further ud chromatin immunoprecipitation (ChIP)to probe whether PKM2interacted with the mek5promoter.The ChIP was carried out with SW620cells using the PCR primer pair that spanned the region of nucle-otides (nt)–1,621to À1,366of mek5promoter.Our ChIP assays clearly demonstrated that PKM2indeed interacted with the mek5promoter (Figure 1D).
We next tested whether upregulation of mek5by PKM2con-tributed to k5was first knocked down in SW620and WM266cells.As a result,the cell proliferation rate was dramatically reduced (Figure S1C).The increas in prolifer-ation by expression of HA-PKM2in the mek5knockdown SW620and WM266cells were also largely reduced (Figure S1C),sug-gesting that u
pregulation of MEK5,at least partially,mediated the effects of HA-PKM2overexpression on cell proliferation.Moreover,we analyzed the MEK5expression levels in the ten cancer cell lines.MEK5was expresd in all ten cell lines,but there were substantially higher levels of MEK5in the more proliferated cancer cells than in the less proliferated cancer cells within the matched pair (Figure S1D).Clearly,there is a clo correlation between cellular MEK5levels and nuclear PKM2levels,and both were correlated cloly with the cell proliferation status of the cells (comparing Figure S1D with Figure 1A).To rule
out a possibility that high cellular levels of mek5promote PKM2nuclear localization,mek5was knocked down,and the nuclear/cytoplasmic PKM2was examined.It was clear that knockdown of mek5did not affect the nuclear levels of PKM2(Figure S1E).We concluded from the studies that upregulation of mek5by nuclear PKM2contributed to cell proliferation.
How would PKM2function in regulation of gene transcription?Sequence analys did not reveal any known DNA binding domain/motifs in PKM2.One possibility is that PKM2may activate particular transcription factors.Thus,we attempted to probe the interaction of PKM2with veral known transcription factors in nuclear extracts of SW620and SW480cells by immu-noprecipitation,such as Oct-1,Oct-4,Gadd45,SOX4-1,and stat3.Among the lected targets,stat3and SOX4were involved i
n transcription of mek5(Aaboe et al.,2006;Song et al.,2004),while Oct-4was reported to interact with PKM2in extracts made from embryonic stem cells (ESCs)(Lee et al.,2008).Our experiments showed that stat3was the only tran-scriptional factor among the lected targets that interacted with PKM2(data not shown,Figure 2A,and Figures S2A and S2B).Possibly due to the differences between ESCs and cancer cells,we did not detect the PKM2and Oct-4interaction in the extracts prepared from cancer cells (data not shown).We noted that stat3interacts with mek5promoter at the À1,776to À1,520nt region by ChIP (Song et al.,2004).Interestingly,PKM2also interacts with mek5promoter at the same region (Fig-ure 1D).To address whether the activation of stat3mediates the effects of PKM2on upregulation of mek5transcription,we first knocked down stat3in SW620/SW480and WM266/WM115cells,and HA-PKM2was subquently expresd.It was
clear
Figure 2.PKM2Regulates mek5Transcription via Activation of stat3
海螺共和国(A)Coimmunoprecipitation of PKM2with stat3.PKM2was immunoprecipitated from nuclear extracts of SW620and SW480cells using the antibody PabPKM2(IP:a PK).The immunoprecipitates were analyzed by immunoblot using antibodies against stat3(IB:Stat3)or PKM2(IB:PKM2).Ponceau S staining of IgG heavy chain (IgG HC)in the immunoprecipitates is the loading control.Immunopre-cipitation using rabbit IgG (IP:IgG)was the control immu-noprecipitation.
(B and D)Expressions of mek5(IB:MEK5)in SW620/SW480(B)and WM115/WM266cells (D)were analyzed by immunoblot of the cell lysate (WCL)of the cells in which stat3was knocked down by RNAi (RNAi-Stat3,IB:Stat3).HA-PKM2(HAPK)was expresd in the stat3knockdown cells (IB:HA).
(C and E)RT-PCR analys of cellular mek5mRNA levels in SW620/SW480(C)and WM266/WM115cells (E)in which stat3was knocked down (RNAi-Stat3,IB:Stat3).HA-PKM2(HAPK)was expresd in the stat3knockdown cells (IB:HA).The results are prented as fold changes in PCR products before and 48hr after stat3knockdown and HA-PKM2expression in the same cells.The level of PCR products from each cell line that was treated with nontarget RNAi was d
efined as 1.
In (B)–(E),NT reprents the cells treated with nontarget RNAi.Vec reprents the cells infected with virus that carry the empty vector.Error bars in (C)and (E)are stan-dard deviations of three measurements.In (B)and (D),immunoblots of b -actin (IB:b -actin)are loading controls.
Molecular Cell
PKM2Phosphorylates Protein Substrate
600Molecular Cell 45,598–609,March 9,2012ª2012Elvier Inc.
that expression of HA-PKM2could no longer upregulate MEK5 as measured by cellular levels of both mRNA and protein of MEK5(Figures2B–2E).To exclude a possibility that downregula-tion of mek5is solely due to stat3knockdown,HA-PKM2was expresd in the cells.Immunoblots indicated that MEK5was upregulated upon the HA-PKM2expression(Figure S2D).To further test the role of stat3in mediating the effects of PKM2on upregulation of mek5transcription,we ud a dominant-nega-tive mutant stat3(Y705F,ref to as stat3-DN)(Xie et al.,2006). The mutant was coexpresd with HA-PKM2in SW620/SW480 and WM266/WM115cells.It was evident that expression of the stat3-DN larg
ely diminished the effects of PKM2on upregulating mek5transcription(Figures S2E–S2H).As a control,expression of the stat3-DN in PKM2knockdown SW480/SW620cells did not result in upregulation of mek5expression(Figure S2E).The activity of stat3can be inhibited by specific inhibitor(Xu et al., 2008).We therefore tested the effects of a stat3inhibitor on the upregulation of mek5by expression of HA-PKM2in SW620and WM266cells.Treatment of the HA-PKM2-expressing cells with the inhibitor resulted in a decrea in mek5transcription(Figures S2I and S2J).Thus,we concluded that activation of stat3medi-ated the regulatory effects of PKM2on mek5transcription.Con-sistently,reanalys of our expression array data revealed changes of a number of stat3regulatory genes in the PKM2-over-expressing SW620cells(Table S2).
We analyzed whether expression of PKM2affected the stat3 binding to its target DNA at the mek5promoter.To probe whether knockdown or expression of PKM2affected the interac-tion of stat3with the mek5promoter,we performed ChIP in SW620cells in which the endogenous PKM2was knocked down or HA-PKM2was expresd.Clearly,the stat3and mek5 promoter interaction was strengthened by HA-PKM2expression and was weakened by PKM2knockdown(Figure3A).We then analyzed the effects of PKM2on the interaction between stat3 and its target DNA by gel mobility shift assays using a32P-labled oligonucleotide duplex containing a stat3binding quence (Xie et al.,2006)
.The experiments were carried out with nuclear extracts of SW620cells with or without PKM2knockdown or with or without HA-PKM2expression.It was clear that a slow migra-tion complex was asmbled with the labeled probe,and addi-tion of the antibody against stat3resulted in a supershift complex (Figure3B).Interestingly,knockdown of PKM2substantially decread the asmbly of the oligo-protein complex and asmbly of stat3in the complex(the weak supershift)(Fig-ure3D),while expression of HA-PKM2incread asmbled complex and asmbly of stat3in the complex(Figure3C). Stat3is activated by phosphorylation at Y705.The phosphor-ylation incread its DNA binding affinity(Sehgal,2008).We obrved that PKM2incread the stat3binding to its target quence both in vitro and in vivo.We suspected that PKM2 may play a role in stat3phosphorylation at Y705.To test this conjecture,we analyzed the stat3phosphorylation in the nuclear extracts prepared from SW620cells in which the HA-PKM2was expresd using an antibody against the Y705-phosphorylated stat3(P-y705/stat3).Clearly,a significant increa in Y705phos-phorylation of stat3was evident.Examination of the cellular levels of stat3indicated that the expression levels of stat3 were not affected(Figure3E).We also obrved a significant reduction in the stat3phosphorylation in SW620cells upon PKM2knockdown(Figure3F).JAK2and c-Src are the most common protein tyrosine kinas that phosphorylate stat3.We therefore probed whether JAK2and c-Src became more acti-vated by the HA-PKM2expression.Using antibodies against JAK2,the phosph
orylated JAK2,c-Src,and the phosphorylated c-Src,our experiments showed that there was no increa in JAK2and c-Src activation(Figure S3A).Furthermore,treatment of cells with JAK2and c-Src inhibitors did not lead to any signif-icant changes of the PKM2-dependent stat3phosphorylations (Figures S3B–S3D),suggesting that stat3phosphorylation was not due to activation of JAK2and Src.
Since PKM2acts in the glycolysis pathway,we analyzed cellular gluco,lactate,and pyruvate in two pairs of cells, SW620/SW480and WM266/WM115,in which the HA-PKM2 was expresd.Upon expression of HA-PKM2,there were no significant changes in cellular gluco or production of lactate and pyruvate for either WM266or SW620cells(the more prolif-erated cell lines).Cellular pyruvate and lactate incread slightly in SW480and WM115cells(the less proliferated cell lines),indi-cating a slight increa in glycolytic pyruvate kina activity in the less aggressive cancer cells,while expression of PKM2did not lead to any changes in glycolytic pyruvate kina activity in the more proliferated cancer cells.Overall,no substantial changes in cellular pyruvate and lactate were obrved(Figures S4A–S4F),suggesting that the differences in stimulating cell proliferation by expression of HA-PKM2were not simply due to the effects on the changes in carbohydrate metabolism. Dimeric PKM2Is the Active Protein Kina
We next sought to test whether PKM2could directly phosphor-ylate stat3.An in vitro phosphorylation
assay using both li-expresd recombinant PKM2(rPKM2)and the HA-PKM2immunopurified from nuclear extracts of SW620cells in the prence of ATP did not yield phosphorylation of a commer-cially available GST-stat3.Since PKM2us PEP as phosphate donor to phosphorylate ADP in the glycolysis,we reasoned that the protein may u the same phosphate donor to phosphorylate a protein substrate.Thus,we replaced ATP by PEP in our in vitro reaction.Immunoblot using the antibody P-y705/stat3demon-strated that the GST-stat3was phosphorylated by the HA-PKM2in the prence of PEP.Consistently,stat3was not phos-phorylated in the prence of ATP(Figures4A and4B).The results indicated that PKM2is a protein kina using PEP as the phosphate donor.
The kina activity of the nuclear HA-PKM2in phosphorylation of stat3was substantially higher than that of the rPKM2ex-presd li.The results led us to compare the protein kina activity of the nuclear PKM2and the cytoplasmic PKM2.The same in vitro phosphorylation reactions were per-formed with the HA-PKM2immunopurified from nuclear or cyto-plasmic extracts of SW620cells in the prence of PEP or ATP. The HA-PKM2from the nuclear extracts had much higher activity than that of protein from the cytoplasmic extracts(Figures4C and4D).To test whether the Y705of stat3is the only phosphor-ylation site by PKM2in cells,we expresd a stat3mutant (Y705A)and GFP-PKM2in SW620cells.Phosphorylations of endogenous and exogenously expresd stat3were examined
Molecular Cell
PKM2Phosphorylates Protein Substrate
Molecular Cell45,598–609,March9,2012ª2012Elvier Inc.601
by immunoprecipitation of HA-tagged stat3mutant or endoge-nous stat3followed by immunoblot using an antibody against phorphotyrosine.It was clear that the endogenous stat3was phosphorylated,while the exogenously expresd mutant was not phosphorylated (Figure S4G),indicating that Y705is the only site.
It was reported that the tetramer and dimer of PKM2coexist in proliferation cells (Mazurek et al.,2005).We therefore ques-tioned whether the differences in the protein kina activity of nuclear/cytoplasmic HA-PKM2and the rPKM2were due to dimer or tetramer of the protein.To investigate whether PKM2is a dimer or a tetramer in the nucleus and in the cytoplasm,we first fractioned the nuclear and cytoplasmic extracts of SW620cells by size-exclusion chromatography.The levels of PKM2in each fraction were examined by immunoblot using
the antibody PabPKM2.Nuclear PKM2was only detected in fractions 14–16,while cytoplasmic PKM2
高考励志对联was mainly detected in fractions 11–16,with the highest concentrations in fractions 11–13.According to the MW calibration standard (Figures S5A and S5B),fraction 11coelutes with a MW near 240kDa,while fraction 14coelutes with a MW near 120kDa (Figure 5A).The gel filtration chromatography suggested that nuclear PKM2was completely dimer,while the cytoplasmic PKM2existed in both dimer and tetramer.The same procedure was also employed to analyze whether the rPKM2is a dimer or a tetramer.It was evident that the rPKM2was mostly tetramer with a very small amount of dimer (Figure 5B).It is well documented that FBP functions as an allosteric regulatory factor that stabilizes the tetramer PKM2.We therefore asked whether FBP could convert the dimer nuclear PKM2to a tetramer form.To
this
Figure 3.PKM2Upregulates mek5Transcription by Promoting stat3DNA Interaction and Phosphor-ylation of stat3
(A)(Upper panels)ChIP of the mek5promoter (MEK5promoter)using antibody against stat3in SW620cells (a Stat3).The cells were treated with nontarget RNAi (left,NT)or with RNAi target PKM2(right,RNAi-PK),or HA-PKM2was expresd in the cells (middle,HAPK).Inputs were PCR products from DNA extracts without ChIP.The primer pair positions are indicated.ChIP using rabbit IgG (IgG)was a negative control.ChIP targeting GAPDH promoter (GAPDH promoter)using antibody against stat3was another negative control.(Lower panels)The cellular PKM2(right)and HA-PKM2(left)levels in SW620cells that were treated with RNAi target PKM2(PK)or with nontarget RNAi (NT),or infected with virus that carry HA-PKM2expression vector (HAPK)or vector alone (Vec),were analyzed by immunoblots using anti-HA antibody (IB:HA)or anti-PKM2antibody (IB:PKM2).
(B)DNA-protein complex (Complex)asmbled on a 32
P-labeled oligo containing the stat3targeting quence in nuclear extracts of SW620cells was detected by gel shift.Free probe indicates the 32P-labeled oligo probe without addition of nuclear ext
racts.The antibodies against PKM2(a PK),stat3(a Stat3),or no antibody (NoAb)were added to the complex to create supershift (Supershift).(C and D)Supershift complex asmbled with the 32
P-labeled oligo and anti-stat3antibody (a Stat3)in the nuclear extracts of SW620cells in which (C)HA-PKM2was expresd (HAPK)or (D)PKM2was knocked down (PK)was detected by gel shift.Probe only (Probe)is the free probe without addition of nuclear extracts.The free probe (low exposure)is the loading control with 1/10of exposure time in autoradiography.The quantification of the asmbled complex and supershift complex was prented as percentage (shifted/total 3102)of probe in the shift complex calculated by intensities of complex (complex [gray bars]or supershift [open bars])divided by intensities of total (free probe +complex +supershift).Immunoblots at the bottom of each panel indicate levels of HA-PKM2(IB:HA)and endogenous PKM2(IB:PKM2)in the cells from which the extracts were prepared for the above gel shift experiments.
(E and F)The levels of Y705-phosphorylated stat3(IB:pY-Stat3)in the cell nucleus were analyzed by immunoblot of nuclear extracts (NE)of SW620cells in which HA-PKM2was expresd (E,HAPK)or the PKM2was knocked down (F,PK).The total cellular stat3levels were analyzed by immunoblot analys of stat3(IB:Stat3)in whole-cell lysate (WCL).In (E),immunoblot of HA-tag (IB:HA)indicates the HA-PKM2expression levels in the cells.In (F),immunoblot of PKM2(IB:PKM2)reprents cellular
骑车去拉萨
PKM2levels in the cells.Immunoblot of Lamin A/C in (E)and (F),and b -actin in (A)and (C)–(F)are the loading controls.NTs in (A)and (C)–(F)mean the cells were treated with nontarget RNAi.Vec in (A)and (C)–(F)means the cells were infected with virus that carry the empty vector.
Molecular Cell
PKM2Phosphorylates Protein Substrate
602Molecular Cell 45,598–609,March 9,2012ª2012Elvier Inc.
end,nuclear extracts of SW620cells were incubated with 5mM FBP at room temperature for 2hr.The dimeric/tetrameric status of PKM2in the nuclear extracts was analyzed by the same procedure.It was evident that FBP did not convert PKM2from the dimeric to the tetrameric form (Figure 5C).
Clo examination of the crystal structure of the tetramer human PKM2(Dombrauckas et al.,2005)reveals that a positively charged residue R399may play a critical role in forming the tetramer of PKM2.It is notable that the R399forms stable charge-charge interactions with residues E418and E396of PKM2located on the other dimer of the tetramer PKM2(Fig-ure 6A).We therefore created a mutant R399E to disrupt the interactions.Size-exclusion chromatography analys demon-
海报设计大赛strated that the R399E mutant was mostly dimer (Figure 5B).We reasoned that the dimeric R399E would be more active in phosphorylating stat3.Thus,the in vitro phosphorylation reactions were carried out with the rPKM2and the R399E.It was evident that the protein kina activity of the R399E was substantially higher than that of the wild-type rPKM2(Fig-ure 6B),while the pyruvate kina activity of the mutant was dramatically lower than that of the rPKM2(Figure 6C).The results supported our speculation that dimeric PKM2is an active protein kina.
To further verify the phosphorylation of stat3by PKM2,we prepared [32P]-labeled PEP (Roossien et al.,1983).The same in vitro phosphorylation reaction was carried out with rPKM2,
rPKM1,and R399E using the [32P]-PEP.Autoradiography indi-cated that the GST-stat3was phosphorylated by the rR399E.The phosphorylation of the GST-stat3by the rPKM2was very weak (was only visualized by a substantial overexposure).The GST-stat3was not phosphorylated by the rPKM1in the pres-ence of the [32P]-PEP (Figure 6D).No phosphorylation can be detected even under very high overexposure.To ensure that the in vitro phosphorylation of stat3by PKM2and the R399E was at comparable physiological conditions,we compared the phosphorylation of the GST-stat3by JAK2and R399E.Clearly,very similar levels of stat3phosphorylations were obrved by both kinas (Figure S6A).The R399E could not phosphorylate other proteins,such as stat5and BSA,und
er the same condi-tions (Figures S6B and S6C),indicating substrate specificity.Extensive analys of enzyme kinetics of pyruvate kina in various tissues (K m =0.07–1.2mM range)indicate that the phys-iological concentration of PEP is likely to be from micromolar to millimolar (Mazurek et al.,2007;van Veelen et al.,1978).To test whether PKM2and the R399E would phosphorylate stat3at the physiological PEP concentrations,we carried out the in vitro phosphorylation at various PEP concentrations.The levels of phosphorylated stat3remained almost constant down to 100m M of PEP.However,there was a clear decrea in stat3phosphorylation when PEP concentration fell below 10m M (Fig-ure S6D).The experiments indicated that phosphorylation of stat3by PKM2was at physiologically comparable
conditions.
Figure 4.Phosphorylation of GST-stat3by the rPKM2
Phosphorylation of GST-stat3by the rPKM2(A)and HA-PKM2(HAPK[NE])immunopurified from nuclear extracts of SW620(B)in the prence of 5mM ATP (ATP)or 5mM PEP (PEP)was revealed by immunoblot assays using antibody against Y705-phosphorylated stat3(IB:pY-stat3).Immu-noblot analys using antibody against stat3(IB:Stat3)indicate the amounts of GST-stat3ud in each reaction.The bottom panels in (A)and (B)are the quantitative analys of immunoblot sig-nals.The error bars reprent the standard devi-ations of four measurements.Phosphorylation of GST-stat3by the HA-PKM2immunopurified from the nuclear (HAPK[NE]in C)and the cytoplasmic (HAPK[Cyto]in D)extracts of SW620cells in the prence of 5mM ATP (ATP)or 5mM PEP (PEP)was revealed by immunoblots using antibody against Y705-phosphorylated stat3(IB:pY-stat3).The reactions were also carried out in the pres-ence or abnce of 5mM FBP (FBP)or 5mM ADP (ADP).The immunoblot of HA (IB:HA)indicates the amounts of HA-PKM2ud in each reaction.IgG HC is the Ponceau S stain of antibody heavy chain,reprenting the amounts of antibody ud in immunopurification of HA-PKM2.Coomassie blue staining (CBS)indicates the amounts of GST-stat3and BSA ud in each phosphorylation reaction.Vec was the cells infected with virus that carry an empty vector.
Molecular Cell
PKM2Phosphorylates Protein Substrate
Molecular Cell 45,598–609,March 9,2012ª2012Elvier Inc.603
