PI-PLCb1b affects Akt activation, cyclin E expression,and caspase cleavage, promoting cell survival inpro-B-lymphoblastic cells exposed to oxidative stress
Manuela Piazzi,*,1 William L. Blalock,†,‡ Alberto Bavelloni,†,§ Irene Faenza,* Mirco Raffini,†,§
Francesca Tagliavini,‡ Lucia Manzoli,* and Lucio Cocco**Cell Signaling Laboratory, Department of Biomedical Sciences, University of Bologna, Bologna, Italy;†Struttura Complessa Laboratory of Musculoskeletal Cell Biology, Rizzoli Orthopedic Institute, Bologna,Italy; ‡Institute of Molecular Genetics, National Research Council of Italy, Bologna, Italy; and§RAMSES Laboratory, Rizzoli Orthopedic Institute, Bologna, Italy
ABSTRACT The phosphoinositide-dependent signaltransduction pathway has been implicated in the control ofa variety of biologic processes, such as the regulation ofcellular metabolism and homeostasis, cell proliferationand differentiation, and apoptosis. One of the key playersin the regulation of inositol lipid signaling is the phospho-lipase Cb1 (PI-PLCb1), that hydrolyzes phosphatidylino-sitol 4,5-bisphosphate [PtIns(4,5)P2], giving rise to thesecondmessengers inositol triphosphateanddiacylglicerol.PI-PLCb1 has been associated with the regulation ofseveral cellular functions, some of which have not yetbeen fully understood. In particular, it has been reportedthat PI-PLCb1 protects murine fibroblasts from oxida-tive stress-induced cell death. Themediators of oxidativestress, reactive oxygen species (ROS), have been shownto regulate major epigenetic processes, causing thesilencing of tumor suppressors and enhancing the pro-liferation of leukemic cells under oxidative stress. In-vestigation of the interplay betweenROS, PI-PLCb1, andtheir signaling mediators in leukemia might thereforereveal innovative targets of pharmacological therapy inthe treatment for leukemia. In this work, we demonstratethat in pro-B-lymphoblastic cells (Ba/F3), treated withH2O2, PI-PLCb1b conferred resistance to cell death,promoting cell cycle progression and cell proliferationand influencing the expression of cyclin A and E. In-terestingly, we found that, expression of PI-PLCb1baffects the activity of caspase-3, caspase-7, and of severalprotein kinases induced by oxidative stress. In particular,PI-PLCb1b expression completely abolished the phos-phorylation of Erk1/2 MAP kinases, down-regulatedphosphatase and tensin homolog (PTEN), and up-regulated the phosphorylation of Akt, thereby sustain-ing cellular proliferation.—Piazzi, M., Blalock, W. L.,Bavelloni, A., Faenza, I., Raffini, M., Tagliavini, F.,
Manzoli, L. Cocco, L. PI-PLCb1b affects Akt activation,cyclin E expression, and caspase cleavage, promoting cellsurvival in pro-B-lymphoblastic cells exposed tooxidativestress. FASEB J. 29, 1383–1394 (2015). www.fasebj.org
OXIDATIVE STRESS IS THE RESULT of an imbalance between theproduction of reactive oxygen species (ROS) and theavailable antioxidant defense. Resistance to oxidative stressis important for cellular survival and cancer prevention. Infact, sublethal levels of ROS (,0.7 mM) can promote pro-liferation, genomic and epigenetic alterations, differentia-tion, and survival in leukemic cells; higher ROS levels(.1 mM), however, can cause severe oxidative stress thatleads to cell death (1). IncreasedROS generation has beenassociated with the onset of cancer, diabetes, and neuro-degenerative diseases (2). An increase in ROS is also ob-served in aging, probably as a consequence of free radicalaccumulation linked to a decreased antioxidant defenseand mithocondrial dysfunction/inefficiency (3). Severalrecent studies have proved conclusively that alteration ofcellular redox omeostasis is a distinctive attribute of varioustypes of leukemia, where ROS levels were observed to besignificantly higher than in normal cells (4–6). Numerousmajor signalingpathwayscanbeactivatedbyoxidative stress(7, 8), eventually resulting in changes in gene expression,which influence the ability of the cell to survive or not.
Among these, changes in the metabolism of phospho-inositides and their elicited signaling pathways have beenassociated with the regulation of cellular responses to oxi-dative stress. Peroxide exposure induced the increase ofphosphoinositide-5-phosphate in various cell lines (9, 10)and stimulated the activity of Akt/PKB, a downstream tar-get of phosphatidylinositol-3,4,5 phosphate (11–13), and
the function of phosphoinositide-specific phospholipase Cisoforms, especially b-1 and g-1 (14, 15). Although the in-crease in phosphoinositide-5-phosphate and the activationof Akt promotes cell viability, as a cellular defense mecha-nism against stress-induced apoptosis, phosphoinositide-dependent phospholipase Cs play a role in promotingapoptosis (16), as well as protecting cells fromdeath (17, 18).
The phosphoinositide-dependent phospholipase Cb1,isoformb (PI-PLCb1b), is one of the 2 existing forms of PI-PLCb1 produced by alternative splicing. The 2 isoforms,a and b, only differ at the C-terminal region, which resultsin a different cellular localization, with PI-PLCb1b beingpreferentially locatedwithin thenucleus (19, 20).Over thelast 2 decades, PI-PLCb1 has been implicated with severalcellular processes, including cell growth and proliferation(21), cell cycle regulation (22, 23), and cellular differen-tiation (24–26). Recently, the protein interaction networkof nuclear PI-PLCb1b was analyzed, highlighting the exis-tence of multiprotein complexes in which PI-PLCb1b wasassociated with proteins involved in cellular metabolicprocesses (e.g., response to oxidative stress), nucleartransport, and regulation of apoptosis (27). Thesefindingsstrengthen the hypothesis that PI-PLCb1 might exert itsfunctionnot only via the generationof secondmessengers,but also directly associatingwith other signalingmediators.Our group has demonstrated that PI-PLCb1 is involved inthe pathophysiology of myelodysplastic syndrome (MDS)and in the induction of both myeloid and erythroid dif-ferentiation(28).Oneof the initial steps in theevolution toleukemia is the loss of sensitivity to either proapoptoticstimuli or the dependence on hematologic growth factors.As an IL-3–dependent pro-B cell line whose reliance onIL-3 for growth and survival can be altered under diverseconditions, the Ba/F3 cell line represents an excellentmodel to study the initial eventsof leukemogenesis. For thisreason, we sought to characterize the function of PI-PLCb1b in Ba/F3 cells. Our results demonstrate that PI-PLCb1b expression promotes entry into S-phase of the cellcycle and reduces the dependence of the cells on IL-3. Inoxidative stressed cells, PI-PLCb1b has an effect on cellcycle regulation, on the activation of effector caspases, andon the activity of several stress-responsive kinases, such asErk1/2, Akt, and protein kinase R (PKR), thus promotingcell survival. As a consequence, cells expressing PI-PLCb1bwhen treated with lethal doses of H2O2, demonstrated aresistance to cell death, confirming a protective role of PI-PLCb1 against apoptosis, as was reported previously (14).
MATERIALS AND METHODS
Cells culture and reagents
Ba/F3 cells were transfected with either an empty pBmz Blast(modified with Blasticidin resistance) (pBB) retroviral vector orpBB containing rat PI-PLCb1 isoform b as previously described(27). The resultant cell lines weremaintained in culture in RPMI
1640 medium supplemented with 10% (v/v) fetal calf serum,2 mM L-glutamine, and 5% (v/v) mouse interleukin-3 (mIL-3)(conditioned media from X63-Ag-653 cells). U73122 and Peri-fosinwere fromSigma-Aldrich(St.Louis,MO,USA),MK-2206wasfrom Selleck Chemicals (Houston, TX, USA). Where not other-wise specified, all the chemicals and reagents were from Sigma-Aldrich.
Cell viability assay
Cells (4 3 103) were washed twice with sterile 13 PBS and werethen cultured in a 96-well plate in 100 ml RPMI 1640 mediumsupplemented with 10% (v/v) fetal calf serum and decreasingconcentrations of recombinant IL-3 (rIL-3) (Life Techonologies,Gaithersburg, MD, USA) ranging from 10 to 0.01 ng/ml. Cellswere incubated for 24, 48, 72, and 96 hours, and cell viabilitywas assessed with a 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tet-razolium bromide (MTT) assay (Roche Diagnostic GmbH,Mannheim,Germany).Briefly,0.5mg/mlofMTTlabelingreagentwas added to eachwell and incubated for 4hours. Purple formazancrystals were solubilized by adding 100 ml of the solubilizationsolution overnight. The plate was subsequently read on an InfiniteM200 photometer (Tecan Group Ltd., Mannedorf, Switzerland)at a wavelength of 570 nm. Colorimetric readings were normalizedagainst plates of nontreated cells under identical culture con-ditions. The experiment was performed 3 times, each time intriplicate.
Apoptosis assay
Apoptosis was evaluated by flow cytometry analysis with the Phy-coerythrin Annexin V detection kit I (BD Biosciences, San Jose,CA,USA), exploiting thebindingofPE-conjugatedAnnexinV forthe detection of apoptotic and necrotic cells. Secondary stainingwith 7-amino-actinomycin D (7-AAD) allowed for the identifica-tion of early apoptotic cells. Cell staining was performed accord-ing to the manufacturer’s instructions. Fluorescence resultingfrom PE and 7-AAD was measured at 530 and 620 nm, re-spectively. Samples were analyzed by BD FACS Canto II (BectonDickinson, San Diego, CA, USA), and data were analyzed usingthe FACS Diva software (Becton Dickinson).
Western blot analysis
Cells were lysed in RIPA lysis buffer containing the CompleteEDTA-free protease and PhosSTOP inhibitor cocktails (RocheDiagnostic GmbH), calpain I and calpain II inhibitors (MerckKGaA, Darmstadt, Germany), and 25 U/ml benzonase. Theprotein concentration was determined by the Bradford ProteinAssay (Bio-Rad, Hercules, CA, USA) according to the manu-facturer’s indications, and80mgof cellular lysatewas separatedbySDS-PAGE on 4–15% gradient gels (Bio-Rad) or 10% standardgels. Gels were transferred onto nitrocellulose membranes andblocked in5%nonfatdrymilk in13PBScontaining0.1%Tween-20 (PBST) and then incubated in primary antibody in 1%nonfatdry milk in PBST. Membranes were washed in 13 PBST andincubated for 1hour in theappropriate secondaryantibody in1%nonfat dry milk in PBST. Primary antibodies were from SantaCruz Biotechnology (Santa Cruz, CA, USA): a-PLCb1 (sc-9050),a-caspase-3 (sc-7148),a-Parp1 (poly (ADP-ribose)polymerase; sc-7150), a-p53 (sc-6243), a-Bcl2 (B cell lymphoma 2; sc-7382),a-Bax (sc-7480), a-p21 (sc-397), a-cyclin A (sc-596), a-Akt (sc-1618); from Cell Signaling Technologies (Danvers, MA, USA):a-caspase-7 (9492),a-cyclinD3(2936),a-cyclinE(4129),a-Erk1/2(527460), a-pErk1/2 (4370), a-pAkt (4058), a-PTEN (9559),
PKR, protein kinase R; PTEN, phosphatase and tensin ho-molog; PtIns(4,5)P2, phosphatidylinositol 4,5-bisphosphate;rIL-3, recombinant IL-3; ROS, reactive oxygen species; STAT,signal transducer and activator of transcription
(continued from previous page)
1384 Vol. 29 April 2015 PIAZZI ET AL.The FASEB Journal x www.fasebj.org
a-pPTEN (9554); from Abcam (Cambridge, United Kingdom):a-PKR (ab58301) anda-pPKR (ab4818); and fromMerck KGaA:a-GAPDH (MAB374). The incubation was performed overnightat 4°C or for the indicated time at room temperature. The blotswere washed 3 times with 13 PBST, detected using SuperSignalWest PicoReagent (Pierce, Rockford, IL,USA), and visualized ina ChemiDoc digital imaging station (Bio-Rad).Membranes usedto detect phosphorylated proteins were blocked in 5% bovineserum albumin and washed in 13 Tris-Buffered Saline contain-ing 0.1% Tween-20. Band intensity was determined by densi-tometry using ImageJ software (National Institutes of Health,Bethesda, MD, USA).
Cell cycle analysis
The number of cells in each phase of the cell cycle was evaluatedwith propidium iodide staining of ethanol-fixed cells, and flowcytometry analysis was performed as previously described (29).Briefly, 13 105 cells were washed in 13 PBS and fixed in 5 ml ofcold 70% ethanol at 4°C for $2 hours. After washing, cells wereincubated with 100 mg/ml RNAse and 20 mg/ml propidium io-dide for 30minutes. Samples were analyzed by BD FACSCanto II(Becton Dickinson), and the proportions of cells in the G1-, S-,and G2-phases of the cell cycle were determined using the FACSDiva software (Becton Dickinson).
Antibody arrays
For antibody arrays, 500mg of cellular extracts was incubatedwiththe Phospho-Kinase Array Kit (Proteome Profiler; R&D Systems,Abingdon, United Kingdom) following the manufacturer’sinstructions. As the array was tested for human phospho-kinases,we considered only those proteins that shared .95% homologywith mouse ortholog (BLAST alignment, www.ncbi.nlm.nih.gov)and in which the phosphorylation site was conserved (Phospho-SitePlus, www.phosphosite.org). Densitometry values were esti-mated by the ImageJ software and were expressed as arbitraryunits. Multiple film exposures were used to verify the linearity ofthe samples analyzed and to avoid saturation of the film. In anti-body arrays, the average signal of the pair of duplicate spots,representing each phosphorylated kinase protein, was calculatedafter subtraction of background values (pixel density) from neg-ative control spots and normalization to average values frompositive control spots.
Statistical analysis
Data are presented as themean6 SD for the indicated number ofindependently performed experiments (at least n = 3) and wereanalyzed by Student t test or 2-way ANOVA. All statistical analyseswere done, and all graphs generated, using the GraphPad Prism,v.5.0 software (GraphPad Software, La Jolla, CA, USA).
RESULTS
Proliferation of IL-3–dependent pro-B lymphoid cellsis influenced by PI-PLCb1b at sublethal doses of IL-3
To elucidate the role of PI-PLCb1b in pro-B lymphoblasticcells, PI-PLCb1b was stably transfected into Ba/F3 cells aspreviously described (27); Ba/F3 cells transfected with PI-PLCb1b or the empty vector are indicated as pBB-PLCb1band pBB-ev, respectively. As Ba/F3 cell proliferation is IL-3dependent, this cell line is often used as a model to study
the effect of protein kinases and downstream signaling oncell proliferation and transformation. Ba/F3 cells werecultured with decreasing concentrations of rIL-3 (Fig. 1),and the rate of mitochondrial metabolism/cell growth (ameasure of cell viability) was evaluatedwith theMTT assay.No differences in pBB-PLCb1b and pBB-ev cell viabilitywere observed between cells cultured in the optimumconcentration of IL-3 conditioned medium (mIL-3, corre-sponding to 20 ng/ml), as well as whendeprived of rIL-3 (1and 0.1 ng/ml). In the latter case, the viability drops to 50%(1 ng/ml of rIL-3) and 20% (0.1 and 0.01 ng/ml of rIL-3)after 24 hours; the cells cultured in 0.1 ng/ml were virtuallyall dead after 48 hours. In contrast, although no significantdifference was observed in the viability of Ba/F3 pBBev orpBB-PLCb1 cultured for 24 h in the presence of 10 ng/mlrIL-3, the viability of pBB-PLCb1b cells was significantlyhigher than pBB-ev by 48 hours, but comparable with thestandard culture condition, indicating that PI- PLCb1b couldexert a proproliferative effect at sublethal doses of mIL-3.
PI-PLCb1b expression in lymphoid cells confersresistance to oxidative stress–induced apoptosis
As different PI-PLC isoforms were previously reported tobe involved in the apoptotic response to oxidative stress,the ability of H2O2 or the DNA damaging agent (mitomy-cin C) to induce apoptosis in Ba/F3 cells expressing pBB-PLCb1b and pBB-ev was examined. Cell viability wasdetermined with cell counting using Trypan blue exclusion
Figure 1. Proliferation assay of pro-B cells expressing PI-PLCb1b. Cells (4 3 103) from pBB-PLCb1b and pBB-ev weregrown for 24, 48, 72, and 96 hours in standard conditions(mIL3, murine IL-3 conditioned medium) and with de-creasing concentrations of recombinant IL-3 (rIL-3): 10, 1,and 0.1 ng/ml. Cell viability was determined with an MTTassay. Graphs are presented with percent cell viability and SD
for the respective independent experiments (n = 3). Statisticalanalysis was performed using a Student’s t test. ***P , 0.001.
(Fig. 2A). In response toH2O2, pBB-PLCb1b cell viabilitywas 30% and 40% higher, after 24 and 48 hours ofstimulation, respectively, compared with pBB-ev (Fig.2A). No differences were observed when inducing DNAdamage with mitomycin C. To confirm that PI-PLCb1bexpression in Ba/F3 cells stimulated with H2O2 con-ferred a resistance to oxidative stress–induced apoptosis,pBB-PLCb1b andpBB-evwere stainedwithannexinVand7-AAD and analyzed by flow cytometry (Fig. 2B). Treatingcells with mitomycin C led to 80% apoptotic cells in bothpBB-PLCb1b and pBB-ev pools within 24 hours. In con-trast, following 24 hours of H2O2 exposure, only 20% ofpBB-PLCb1b cells were positive for apoptosis comparedwith 70% of pBB-ev. Furthermore, these results demon-strate thatPI-PLCb1bexertedaprotective effect specificallyagainst oxidative stress–induced apoptosis.
PI-PLCb1b protects cells from death influencingcaspase-3– and -7–dependent apoptosis onH2O2 exposure
As H2O2-treated pBB-PLCb1b cells were more viablecompared with pBB-ev cells (at 24 hours), we investigatedthe potential cause of the apoptotic delay resulting fromPLCb1 forced expression. Bright field microscopyrevealed that treatment with H2O2 for 6 hours did notaffect either pBB-PLCb1b or pBB-ev cell morphology, as
both exhibited the same phenotype as their respectiveuntreated controls (Fig. 3). In contrast, following 24, 48,and 72 hours of treatment with H2O2, pBB-ev cells wereclustered and presented with the typical signatures ofcells undergoing death (Fig. 3 and Supplemental Data)(30, 31). However, pBB-PLCb1b cells appeared to havethe same phenotype as untreated cells. Peroxide expo-sure eventually induced apoptosis in pBB-PLCb1b cells,but following a delay of $24 hours, as reflected by thechanges in cell morphology (Supplemental Data). Ad-ditionally, the expressionof the typical apoptoticmarkerswas investigated to determine which apoptotic pathwayscould be influenced by PI-PLCb1b. Cells were stimulatedwith 250 mM H2O2 for diverse amounts of time (Fig. 4),and the expression of caspase-3, caspase-7, Parp-1, p53,Bcl2, and Bax was evaluated by Western blot analysis. InH2O2-treated pBB-ev cells, between 24 and 48 hours, theinactive procaspase-3 was completely cleaved to its activeform as was Parp-1, a downstream targets of caspase-3.In cells expressing PLCb1b, a delay of 24 hours for thecomplete cleavage of both caspase-3 and Parp-1 was ob-served, which could in part explain the morphologicdifferences observed between the cell lines followingH2O2 exposure. Notably, pBB-PLCb1b cells showeda complete absence of caspase-7 cleavage, which canexplain themorphologic differences at 24 hours ofH2O2exposure. Unlike pBB-PLCb1b, H2O2-induced apoptosisled to an increase in the levels of p53 in pBB-ev cells, and
Figure 2. PI-PLCb1b expressing Ba/F3 cells are resistant to oxidative stress–induced apoptosis. A) Apoptosis was induced bytreating pBB-PLCb1b and pBB-ev cells with 250 mM of H2O2 or with 5 mg/ml mitomycin C (MMC). Cell viability was determinedby counting cells with Trypan blue exclusion. Cells were analyzed at 0.5, 2, 4, 6, 24, and 48 hours after treatment. T0 representsthe cell viability prior to the induction with apoptotic stimulus. Graphs (black solid and dotted lines for pBB-PLCb1b and pBB-evcells, respectively) are presented with percent cell viability and SD for the respective independent experiments (n = 3). Statisticalanalysis was performed using a Student’s t test. *P , 0.05, **P , 0.01, and ***P , 0.001. B) Cytometric analysis of PE:Annexin Vstaining of PI-PLCb1b and pBB-ev cells treated with 250 mM of H2O2 or with 5 mg/ml MMC for 2, 4, 6, and 24 hours. Cells (13 105)were incubated at room temperature in the dark with 5 ml PE:Annexin V and 5 ml of 7-AAD. Histograms are representative of3 independent experiments, each conducted in triplicate.
1386 Vol. 29 April 2015 PIAZZI ET AL.The FASEB Journal x www.fasebj.org
the expression of bothBax andBcl2 in pBB-PLCb1b cellswas lower compared with pBB-ev, whereas the Bax/Bcl2ratio was higher in pBB-ev cells, characteristic of cellsmore sensitive to apoptosis (32). Taken together, theseresults further confirmed the role of PI-PLCb1 in pro-tecting cells from oxidative stress, giving insight into themolecular pathways that could be regulated.
PI-PLCb1b and the regulation of the cell cycle
PI-PLCb1 expression promotes pro-B lymphoid cells in theS-phase of the cell cycle and augments their proliferation
It has been extensively demonstrated that PI-PLCb1 is in-volved in the cell cycle regulation of erythroleukemia(MEL) cells, both stimulating the progression through theG1-phase andmediating theG2/Mtransition (22, 23).Cellcycle analysis of pBB-PLCb1b and pBB-ev cells, stimulatedwith H2O2 or left untreated, was undertaken to assesswhether PI-PLCb1 acted as a cell cycle regulator inacute lymphoid leukemia cells. Cells were stained withpropidium iodide for the indicated time points and ana-lyzed by flow cytometry (Fig. 5A, B). As it is shown in theplots and statistically represented as histograms, the ex-pression of PI-PLCb1b increased the percentage of cells inthe S-phase of the cell cycle in standard growing con-ditions, only after2hours, comparedwithpBB-ev (Fig. 5A).After 24 hours, 16% of pBB-ev cells were found in S-phase.On the contrary, 46% of pBB-PLCb1b cells were in theS-phase. After 48 hours, although .60% of pBB-ev cellswere arrested in G0/G1, slightly .40% of the cellsexpressingPI-PLCb1bwere in theG1-phaseof thecell cycle.Because H2O2 stress resulted in enhanced S-phase accu-mulation in PI-PLCb1b cells, we evaluated the proliferationof PI-PLCb1b and pBB-ev cells both with Trypan blue ex-clusion and the MTT assay at 0.5, 2, 4, 6, 24, and 48 hoursunder standard growing conditions with newly added IL-3(mIL-3 conditioned medium; Fig. 5C). Whereas no differ-ences between pBB-PLCb1b and pBB-ev cells emerged as
cellular metabolic activity (Fig. 5C, right), there was a dif-ferenceof0.6-foldmorepBB-PLCb1bviablecells comparedwith pBB-ev after 24 hours (Fig. 5C, left). These resultsdemonstrated that PI-PLCb1b induces pro-B cells to cycle,promoting the S-phase and resulting in a proproliferativephenotype, without altering themitochondrial activity. Theexposure to H2O2 induced pBB-ev cells to arrest in G0/G1immediately after 2 hours (+20% compared with untreatedcells; Fig. 5B); however, .50% of PI-PLCb1b cells werefound in the S-phase compared with the 26% of pBB-ev,indicating that the effect of PI-PLCb1b on cell cycle regu-lationovercame the apoptotic stimulus inducedbyH2O2, aswas observed in Fig. 4.
PI-PLCb1 induces the up-regulation of cyclin A and cyclin E
Cells were stimulated with 250mMH2O2, or left untreated,for diverse amounts of time (Fig. 6), and the expression ofp21, cyclin A, cyclin D3, and cyclin E was evaluated byWestern blot analysis. Although p21 was poorly inducedfollowing H2O2 treatment, cyclins appeared to be mainlyaffected inPI-PLCb1bexpressingcells.CyclinAexpressionwas observed to be slightly elevated in pBB-PLCb1b cellscompared with the control, whereas no significant differ-ences were detected in the profile of cyclin D3. Notably,cyclinEwas found tobe constitutively expressed at ahigherlevel in PI-PLCb1b expressing cells compared with pBB-evcells. As the cyclin E:cdk2 complex is required for thetransitions G1/S of the cell cycle, these results couldexplain the elevated percentage of pBB-PLCb1b cells inS-phase.
Inhibition of PI-PLCb1 has an effect on both oxidativestress-induced apoptosis and the cell cycle
Cells were treated with 10mMofU73122 for 1.5 hours, andthen 250 mM of H2O2 was added for 24 hours. Apoptosiswasexaminedby stainingwithAnnnexinV-7AAD(Fig. 7A),
Figure 3. PI-PLCb1b induces morphologic changes in IL-3–dependent lymphoid cells. Cells (pBB-PLCb1b and pBB-ev; 1 3 107)were cultured in a 6-multiwell plate, treated with 250 mMH2O2, or left untreated. After 6 and 24 hours, cells were observed undera bright field light microscope (Axio Observer.A1 microscope, equipped with an AxioCamHR3; Carl Zeiss, Jena, Germany), andimages were acquired using a 332 lens, with the software AxioVision, v.4.7.2.0.
PI-PLCb1B AND OXIDATIVE STRESS 1387
cell proliferation by cell counting with Trypan blue exclu-sion (Fig. 7B), and cell cycle by staining with propidiumiodide (Fig. 7C). As shown in Fig. 7A, B, the PI-PLCsinhibitor U73122 had little effect on apoptosis in pBB-PLCb1b cells not exposed to H2O2, whereas it had a sig-nificant increase in the percentage of cells arrested in G0/G1-phases of the cell cycle (Fig. 7C). In contrast, when cellswere treatedwithH2O2 in thepresenceof the inhibitor, thenumber of cells in S-phase did not significantly differ fromthat observed in both untreated and H2O2-treated cells,whereas the percentage of cells inG2/Mslightly increased.Moreover, H2O2 treatment of pBB-PLCb1b cells in thepresence of U73122 resulted in a significant increase ofAnnexin V-positive cells and in a 20% decrease in viability.
PI-PLCb1b influences the activity of protein kinasesinvolved in the response to cellular stress
ROS produced by H2O2 can be considered as an in-tracellular second messenger that can regulate the activityof several protein kinases. For this reason, changes in the
phosphorylation level of critical residues required for theactivity of common protein kinases were analyzed in pro-Blymphoblastic cells expressing PI-PLCb1b under basalconditions (Fig. 8A) and following exposure to 250 mM ofH2O2 for 1 h (Fig. 8B). We used a phospho-protein arraythat specifically screens for relative levels of phosphoryla-tion of protein kinases involved in cellular proliferationand survival. Cells expressing PI-PLCb1b showed a signifi-cant up-regulation of Akt (Ser473) and cAMP responseelement-binding protein phosphorylation, whereas thesignal transducer and activator of transcription (STAT)5familywas stronglydown-regulated(only STAT3presentedwith an increased phosphorylation level). Interestingly, inBa/F3 cells that expressed PI-PLCb1b and were treatedwith H2O2, 12 of 18 protein kinases exhibited a significantdown-regulation of their phosphorylation levels. Notably,phosphorylation of MAPK (Erk1/2), JNK, p38 MAPK,mitogen- and stress-activatedproteinkinase-1/2, and c-jun,which are involved in the response to cellular stress andH2O2, were down-regulated by PI-PLCb1b. The only ex-ceptionwas representedbyAkt,whosephosphorylationonSer473, required for its catalytic activity, remained higher
Figure 4. Activation of effector caspases, in oxidative stressed cells, is influenced by PI-PLCb1b. Cells (pBB-PLCb1b and pBB-ev; 13 107)were treated with 250 mMH2O2 and cultured in a 6-multiwell plate for 2, 4, 6, 24, 48, and 72 hours. Protein lysate (80 mg) were separatedin a 4–15% gradient SDS-PAGE and immunoblotted with specific antibodies directed against PI-PLCb1 (1:500, overnight),caspase-3 (1:500, overnight), caspase-7 (1:1000, overnight), Parp-1 (1:1000, 2 hours), p53 (1:500, overnight), Bcl2 (1:500,overnight), Bax (1:500, overnight), and GAPDH (1:8000, 1 hour). Graph bars represent the ratio between Bax and Bcl2 in pBB-ev(white histograms) and pBB- PLCb1b cells (black histograms), expressed as densitometric arbitrary unit (a.u.).
1388 Vol. 29 April 2015 PIAZZI ET AL.The FASEB Journal x www.fasebj.org
Figure 5. PI-PLCb1b effect on cell cycle regulation and proliferation of pro-B-lymphoblastic cells. pBB-PLCb1b and pBB-ev cellswere treated with 250 mM H2O2 and cultured in a 6-multiwell plate for 0.5, 2, 4, 6, 24, and 48 hours. Cells (1 3 106) of bothuntreated (A) and H2O2-treated (B) cells, for each time point, were stained with propidium iodide as described in Materials and
(continued on next page)
PI-PLCb1B AND OXIDATIVE STRESS 1389
in pBB-PLCb1b cells compared with pBB-ev, also aftertreatment with H2O2. To further confirm these data, pBB-PLCb1b and pBB-ev cells were exposed to H2O2 for 0.5, 2,4, 6, 24, 48, and 72 hours or left untreated, and the ex-pression and phosphorylation of Akt and Erk1/2 wereevaluated by immunoblotting (Fig. 8C and SupplementalData). The expression of Akt was up-regulated in cellsexpressing PI-PLCb1b and treated with H2O2; p-Akt wasinsteadbarelydetectable inpBB-evuntreatedcells,whereasit was activated shortly after stimulus (up to 2 hours). Inuntreated pBB-PLCb1b cells, however, Akt was highly ac-tivated, and the phosphorylation remained high, com-pared with the control, even following exposure to H2O2.Conversely, the expression and phosphorylation of PTENwas down-regulated in pBB-PLCb1b cells, both untreatedand treated, compared with their respective controls. Al-though the expression of Erk1/2 was not significantlydifferent in pBB-PLCb1b and pBB-ev untreated cells, ex-posure to H2O2 in cells expressing PI-PLCb1b negativelyaffected Erk1 expression. Phosphorylation of Erk1/2 wascompletely inactivated inH2O2-treatedpBB-PLCb1bcells.Another protein kinase activated in response to diversestress signals is the interferon-induced, double-strandedRNA-activated protein kinase PKR. In fact, it was re-ported that PKR activity augmented in leukemic cells,following treatment with H2O2 (33). Because PKR wasnot included in the phospho-array, the expression andphosphorylation of PKR were investigated by immuno-blotting following H2O2 exposure. As expected, PKRactivation was prolonged in H2O2-stressed pBB-ev cellscompared with untreated cells. Surprisingly, both PKRexpression and phosphorylation were influenced by theexpression of PI-PLCb1b. Not only was PKR activitycompletely abolished in untreated pBB-PLCb1b cells,but the kinase, although expressed, was also not re-sponsive to treatment with H2O2.
The allosteric Akt inhibitor, MK-2206, inducespBB-PLCb1b cells to arrest the cell cycle
To assess whether the inhibition of Akt has an effect on cellproliferation and apoptosis of pBB-PLCb1b cells, the orallyactive, allosteric Akt inhibitorMK-2206 was used. Cell weretreated with 1 mM of MK-2206 for 1.5 hours and then ex-posed to 250 mM of H2O2 for 24 hours. Apoptosis wasexamined by staining with Annexin V-7AAD (Fig. 9A), cellproliferation by cell counting with Trypan blue exclusion(Fig. 9B), and cell cycle by staining with propidium iodide(Fig. 9C). As shown in Fig. 9A, B, MK-2206 did not induceany effect on apoptosis and cell viability of both pBB-ev andpBB-PLCb1b cells, whereas in combinationwithH2O2, thepercentage of cells positive to annexin V increased in both
pBB-ev and pBB-PLCb1b cells, with respect to cells onlytreatedwithH2O2.The sameresultswereobtained treatingcells with another Akt inhibitor, the alkylphospholipidperifosine (20 mM; Supplemental Data), whereas no dif-ferences were observed targeting theMAPKpathway usingthe inhibitor PD98059 (20 mM; Supplemental Data). Fi-nally, the effect of MK-2206 on the cell cycle was analyzed(Fig. 9C), and although pBB-PLCb1b cells treated withMK-2206 alone arrested in Go/G1, the combination withH2O2 induced the cell cycle to arrest in G2/M.
DISCUSSION
In this study, we investigated the role of PI-PLCb1b ex-pression inBa/F3 cells exposed to lethal doses (250mM)ofH2O2. Previousfindings reported that different isoformsofphospholipase C can exert an important protective func-tion during the cellular response to oxidative stress. Inparticular, PI-PLCg1 activation enhances cell survival ofmurine fibroblasts exposed to H2O2, through an increaseof PKC-dependent Bcl2 phosphorylation and inhibition ofcaspase-3 (17). Also, PI-PLCb1 has been shown to protectmurine fibroblasts (NIH3T3) from cell death occurring inresponse to oxidative stress, without further investigatingthe signaling pathways that were affected (14). Our resultsshow that PI-PLCb1b exerts a protective effect against ap-optosis in pro-B lymphoblastic cells, and this function wasspecifically observed in response to oxidative stress,whereas cells were not responsive following treatment witha DNA damaging agent, mitomicin C. In Ba/F3 cells, PI-PLCb1b expression causes a delay in caspase-3 activation,abrogates caspase-7 cleavage, and affects the activity ofseveral protein kinases implicated in the response to stress.Exposure of cells to oxidant stress in the form of H2O2induces the activation of diverse signaling pathways,among which are MAPK (Erk1/2), JNK, p38 kinase, andthe PI3K/Akt axis.
JNK and p38 are frequently associated with the in-duction of apoptosis, whereas Akt is implicated in cellsurvival and protection from apoptosis. Erk1/2 has beendemonstrated to contribute to both protecting and/orpromoting apoptosis, depending on the cell type and or-gan (34). In our case, PI-PLCb1b expression in H2O2-treated cells completely abolished the phosphorylation ofErk 1/2, whereas the phosphorylation of Akt on S473, re-quired for its catalytic activity, remained elevated com-pared with control cells. It is known that Erk-mediatedapoptosis can occur through the suppression of PI3K/Aktaxis; thus, the 2 signaling transduction pathways worksinergically to lead cells to apoptosis or survival dependingon the balance of their activity (35). As was previouslyreported, under oxidative stress, inhibition of Erk1/2
Methods and analyzed by flow cytometry. Plots and histograms are representative for the distribution of cells in the differentphases of the cell cycle. Graphs are presented with percent of the total number of cells and SD for the G1, S, and G2 phase ofthe cell cycle and for the number of independent experiments (n = 3). Statistical analysis was performed using a Student’s t test.***P , 0.001. C) pBB-PLCb1b and pBB-ev cell proliferation was assessed in a standard growing condition. Cell viability wasdetermined counting cells with Trypan blue exclusion (left) and performing a MTT assay (right) as detailed in Materials andMethods. Cells were analyzed at 0.5, 2, 4, 6, 24, and 48 hours. T0 represents the cell viability prior to the addition of fresh IL-3conditioned medium. Graphs are presented with fold increase and SD for the respective independent experiments (n = 3). Statisticalanalysis was performed using a Student’s t test. *P , 0.05 and **P , 0.01.
1390 Vol. 29 April 2015 PIAZZI ET AL.The FASEB Journal x www.fasebj.org
potentiates Akt phosphorylation, resulting in an increaseof cell survival; conversely, once Erk1/2 is activated, it canmediate the suppressionofAkt,propagating the signal thatleads to cell death (36, 37). This condition was found to bespecific for cells exposed to H2O2; in fact, treatment withantioxidants or ROS scavengers prevents the activation ofErk1/2. We showed that PI-PLCb1b represents an up-stream mediator that provokes the suppression of Erk1/2phosphorylation causing activation of the Akt signalingpathway. PI-PLCb1b expression induces down-regulationof Erk1/2 and a sustained phosphorylation of Akt undernormal growth conditions.
PKR is a dsRNA-dependent protein kinase activated inresponse to diverse stress signals, including H2O2. It hasbeen reported that PKR activation occurs in acute leuke-mia cell lines, which present as PTEN null (38). In-terestingly, in high-risk MDS patients, the presence ofactive PKR correlates with an increased phosphorylationof Akt and a decrease in PTEN levels (39). AdditionallyPI-PLCb1 exerts an influence on Akt both in high-risk and
low-risk MDS, and an inverse correlation between PI-PLCb1 expression and Akt phosphorylation has beendemonstrated (40). Here we report that the expression ofPI-PLCb1 in pro-B lymphoblastic cells abrogated PKRphosphorylation on T451 while sustaining p-Akt. More-over, PI-PLCb1 prevents PKR phosphorylation followingH2O2 treatment. All together, these findings strengthenour hypothesis that a regulatory axis PI-PLCb1/PKR/Aktexists and is required for leukemic cell fate, balancingcell proliferation/survival, differentiation, and resistanceto stress. However, we found that the prosurvival effects ofPI-PLCb1b expression had little to do with the observedinduction of Akt activity, at least via the classic pathwaywhere Akt is recruited to the phosphatidylinositol (3,4,5)-trisphosphate via the Pleckstrin homology domain, whereit was more related to the regulation of the cell cycle.
Ba/F3 cell cycle analysis revealed that PI-PLCb1b pro-motes cell proliferation, inducing cells in the S-phase andincreasing the proliferation rate. This effect is likely me-diated by Akt, because the inhibition of its activity induced
Figure 6. Cyclin A and cyclin E expression is up-regulated in pBB-PLCb1b cells. Cells (pBB-PLCb1b and pBB-ev; 1 3 107) weretreated with 250 mM of H2O2 or left untreated and cultured in a 6-multiwell plate for 0.5, 2, 4, 6, 24, 48, and 72 hours. Proteinlysates (80 mg) were separated in a 4–15% gradient SDS-PAGE and immunoblotted with specific antibodies directed against: p21(1:500, overnight), cyclin A (1:500, overnight), cyclin D3 (1:1000, overnight), cyclin E (1:1000, overnight), and GAPDH (1:8000,1 hour).
Figure 7. Inhibition of PI-PLCb1 affects both the prosurvival response to oxidative stress-induced apoptosis and the cell cycle ofpBB-PLCb1b cells. Cells (pBB-PLCb1b and pBB-ev; 5 3 106 in a 12-multiwell plate) were treated with 10 mM U73122 for1.5 hours, and then 250 mM H2O2 was added for 24 hours. Apoptosis was evaluated by staining with Annexin V-7AAD (A), cellviability by cell counting with Trypan blue exclusion (B), and cell cycle by propidium iodide staining (C). Graphs arerepresentative of 3 independent experiments, each conducted in triplicate.
PI-PLCb1B AND OXIDATIVE STRESS 1391
Figure 8. Phospho-kinase array analysis of pro-B-lymphoblastic cells exposed to H2O2. A, B) Whole cell lysates were prepared frompBB-PLCb1b and pBB-ev cell lines, untreated or treated with 250 mM H2O2 for 1 h, respectively, and hybridized with a Phospho-Kinase array kit. Spot densities of phospho-proteins were quantified using ImageJ software and normalized to those of positivecontrols on the same membrane. **P , 0.01, ***P , 0.001. C) Cells (pBB-PLCb1b and pBB-ev; 1 3 107) were treated with 250 mMH2O2 for 0.5, 2, 4, 6, 24, and 48 hours or left untreated. Cell lysates (80 mg) were separated by 10% SDS-PAGE and immunoblottedwith specific antibodies directed against: PKR (1 mg/ml, 2 hours), p-PKR (T451) (1:1000, overnight), Akt (1:1000, overnight), p-Akt(S473) (1:1000, overnight), PTEN (1:1000, overnight), p-PTEN (S380/T382/T383) (1:1000, overnight), Erk1/2 (1:1000, overnight),p-Erk 1/2 (T202/T204) (1:1000, overnight), and GAPDH (1:8000, 1 hour).
1392 Vol. 29 April 2015 PIAZZI ET AL.The FASEB Journal x www.fasebj.org
pBB-PLCb1b cells to arrest in G0/G1. Our hypothesis isthat PI-PLCb1b regulation on the cell cycle overwhelmstheeffect ofH2O2, thusprolonging cell survival. In termsofcontrasting the negative side effects associated with oxi-dative stress, for example, in radiotherapy and chemo-therapy, where the production of intracellular ROS likelyaffects healthy cells, the role of PI-PLCb1b could be con-sidered promising: the enzyme activation may correlatewith the use of antioxidants or ROS scavengers to coun-teract the intracellular ROS production. However, a di-minished expression of p53 associated with an increase ofAkt activation, which was observed in PI-PLCb1b ex-pressing cells, might correspond with the Warburg effect,leading cells to transformation and tumor proliferation. Inthe case of leukemic cells, therefore, the expression of PI-PLCb1b in response tooxidative stress couldbe interpretedas favorable to the tumor, spanning oncogenic cell lives.Further investigations are required to determine whetherPI-PLCb1b can be used as a sentry for tumors treatedagainst ROS. The PI-PLCb1b signaling axis can be ade-quately targeted for therapeutic purposes in leukemia.
The authors thank Dr. Aurelio Valmori, Dr. ElisabettaLongo, and Dr. Luca Cattini for technical assistance. Thiswork was supported by the Italian Ministero dell’Istruzione,dell’Universita e della Ricerca-Fondo per gli Investimenti dellaRicerca di Base (MIUR-FIRB) 2010 Human ProteomeNet;Italian MIUR-FIRB 2010 (RBAP10447J); and the “5 3 1000”fund to the SC Laboratory of Musculoskeletal Cell Biology,Rizzoli Orthopedic Institute. The authors declare no conflictsof interest.
REFERENCES
1. Stone, J. R., and Yang, S. (2006) Hydrogen peroxide: a signalingmessenger. Antioxid. Redox Signal. 8, 243–270
2. Waris, G., and Ahsan, H. (2006) Reactive oxygen species: role inthe development of cancer and various chronic conditions.J. Carcinog. 5, 14
3. Finkel, T. (2011) Signal transduction by reactive oxygen species.J. Cell Biol. 194, 7–15
4. Zhou, F., Shen, Q., and Claret, F. X. (2013) Novel roles ofreactive oxygen species in the pathogenesis of acute myeloidleukemia. J. Leukoc. Biol. 94, 423–429
5. Li, L., Li, M., Sun, C., Francisco, L., Chakraborty, S., Sabado,M., McDonald, T., Gyorffy, J., Chang, K., Wang, S., Fan, W., Li,J., Zhao, L. P., Radich, J., Forman, S., Bhatia, S., and Bhatia, R.(2011) Altered hematopoietic cell gene expression precedesdevelopment of therapy-related myelodysplasia/acute mye-loid leukemia and identifies patients at risk. Cancer Cell 20,591–605
6. Toyokuni, S. (2006) Novel aspects of oxidative stress-associatedcarcinogenesis. Antioxid. Redox Signal. 8, 1373–1377
7. Martindale, J. L., and Holbrook, N. J. (2002) Cellular response tooxidative stress: signaling for suicide and survival. J. Cell. Physiol.192, 1–15
8. Shiizaki, S., Naguro, I., and Ichijo, H. (2013) Activationmechanisms of ASK1 in response to various stresses and itssignificance in intracellular signaling. Adv Biol Regul 53, 135–144
9. Jones, D. R., Foulger, R., Keune, W. J., Bultsma, Y., and Divecha,N. (2013) PtdIns5P is an oxidative stress-induced second mes-senger that regulates PKB activation. FASEB J. 27, 1644–1656
10. Keune,W.J., Jones,D.R.,Bultsma,Y.,Sommer,L.,Zhou,X.Z.,Lu,K.P.,and Divecha, N. (2012) Regulation of phosphatidylinositol-5-phosphate signaling by Pin1 determines sensitivity to oxidative stress.Sci. Signal. 5, ra86
11. Chetram, M. A., Bethea, D. A., Odero-Marah, V. A., Don-Salu-Hewage, A. S., Jones, K. J., and Hinton, C. V. (2013) ROS-mediated activation of AKT induces apoptosis via pVHL in prostatecancer cells. Mol. Cell. Biochem. 376, 63–71
12. van Boxtel, R., Gomez-Puerto, C., Mokry, M., Eijkelenboom, A.,van der Vos, K. E., Nieuwenhuis, E. E., Burgering, B. M., Lam,E. W., and Coffer, P. J. (2013) FOXP1 acts through a negativefeedback loop to suppress FOXO-induced apoptosis. Cell DeathDiffer. 20, 1219–1229
13. Van Gorp, A. G., Pomeranz, K. M., Birkenkamp, K. U., Hui, R. C.,Lam, E. W., and Coffer, P. J. (2006) Chronic protein kinase B(PKB/c-akt) activation leads to apoptosis induced by oxidativestress-mediated Foxo3a transcriptional up-regulation. Cancer Res.66, 10760–10769
14. Lee, Y. H., Kim, S. Y., Kim, J. R., Yoh, K. T., Baek, S. H., Kim, M. J.,Ryu, S. H., Suh, P. G., and Kim, J. H. (2000) Overexpression ofphospholipase Cbeta-1 protects NIH3T3 cells from oxidativestress-induced cell death. Life Sci. 67, 827–837
15. Wang, X. T., McCullough, K. D., Wang, X. J., Carpenter, G., andHolbrook, N. J. (2001) Oxidative stress-induced phospholipaseC-gamma 1 activation enhances cell survival. J. Biol. Chem. 276,28364–28371
16. Monturiol-Gross, L., Flores-Dıaz, M., Pineda-Padilla, M. J.,Castro-Castro, A. C., and Alape-Giron, A. (2014) Clostridiumperfringens phospholipase C induced ROS production andcytotoxicity require PKC, MEK1 and NFkB activation. PLoS ONE9, e86475
17. Bai, X. C., Deng, F., Liu, A. L., Zou, Z. P., Wang, Y., Ke, Z. Y., Ji,Q. S., and Luo, S. Q. (2002) Phospholipase C-gamma1 is re-quired for cell survival in oxidative stress by protein kinase C.Biochem. J. 363, 395–401
Figure 9. Inhibition of Akt induces the arrest of the cell cycle in pBB-PLCb1b cells. Cells (pBB-PLCb1b and pBB-ev; 5 3 106 ina 12-multiwell plate) were treated with 1 mM MK-2206 for 1.5 hours, and then 250 mM H2O2 was added for 24 hours. Apoptosiswas evaluated by staining with Annexin V-7AAD (A), cell viability by cell counting with Trypan blue exclusion (B), and cell cycleby propidium iodide staining (C). Graphs are representative of 3 independent experiments, each conducted in triplicate.
PI-PLCb1B AND OXIDATIVE STRESS 1393
18. Yasuda, E., Nagasawa, K., Nishida, K., and Fujimoto, S. (2008)Decreased expression of phospholipase C-beta 1 protein in en-doplasmic reticulum stress-loaded neurons. Biol. Pharm. Bull. 31,719–721
19. Bahk, Y. Y., Song, H., Baek, S. H., Park, B. Y., Kim, H., Ryu, S. H.,and Suh, P. G. (1998) Localization of two forms of phospholipaseC-beta1, a and b, in C6Bu-1 cells. Biochim. Biophys. Acta 1389,76–80
20. Kim, C. G., Park, D., and Rhee, S. G. (1996) The role of carboxyl-terminal basic amino acids in Gqalpha-dependent activation,particulate association, and nuclear localization of phospholipaseC-beta1. J. Biol. Chem. 271, 21187–21192
21. Poli, A., Faenza, I., Chiarini, F., Matteucci, A., McCubrey, J. A.,and Cocco, L. (2013) K562 cell proliferation is modulated byPLCb1 through a PKCa-mediated pathway. Cell Cycle 12,1713–1721
22. Faenza, I., Matteucci, A., Manzoli, L., Billi, A. M., Aluigi, M.,Peruzzi, D., Vitale, M., Castorina, S., Suh, P. G., and Cocco, L.(2000) A role for nuclear phospholipase Cbeta 1 in cell cyclecontrol. J. Biol. Chem. 275, 30520–30524
23. Fiume, R., Ramazzotti, G., Teti, G., Chiarini, F., Faenza, I.,Mazzotti, G., Billi, A. M., and Cocco, L. (2009) Involvement ofnuclear PLCbeta1 in lamin B1 phosphorylation and G2/M cellcycle progression. FASEB J. 23, 957–966
24. Matteucci, A., Faenza, I., Gilmour, R. S., Manzoli, L., Billi, A. M.,Peruzzi, D., Bavelloni, A., Rhee, S. G., and Cocco, L. (1998)Nuclear but not cytoplasmic phospholipase C beta 1 inhibitsdifferentiation of erythroleukemia cells. Cancer Res. 58, 5057–5060
25. Faenza, I., Bavelloni, A., Fiume, R., Lattanzi, G., Maraldi, N. M.,Gilmour, R. S., Martelli, A. M., Suh, P. G., Billi, A. M., and Cocco,L. (2003) Up-regulation of nuclear PLCbeta1 in myogenic dif-ferentiation. J. Cell. Physiol. 195, 446–452
26. O’Carroll, S. J., Mitchell, M. D., Faenza, I., Cocco, L., andGilmour, R. S. (2009) Nuclear PLCbeta1 is required for 3T3-L1adipocyte differentiation and regulates expression of the cyclinD3-cdk4 complex. Cell. Signal. 21, 926–935
27. Piazzi, M., Blalock, W. L., Bavelloni, A., Faenza, I., D’Angelo, A.,Maraldi, N. M., and Cocco, L. (2013) Phosphoinositide-specificphospholipase C b 1b (PI-PLCb1b) interactome: affinity purification-mass spectrometry analysis of PI-PLCb1b with nuclear protein.Mol. Cell. Proteomics 12, 2220–2235
28. Follo, M. Y., Mongiorgi, S., Clissa, C., Paolini, S., Martinelli, G.,Martelli, A. M., Fioravanti, G., Manzoli, L., Finelli, C., and Cocco,L. (2012) Activation of nuclear inositide signalling pathwaysduring erythropoietin therapy in low-risk MDS patients. Leukemia26, 2474–2482
29. Bertagnolo, V., Benedusi, M., Brugnoli, F., Lanuti, P., Marchisio,M., Querzoli, P., and Capitani, S. (2007) Phospholipase C-beta 2
promotes mitosis and migration of human breast cancer-derivedcells. Carcinogenesis 28, 1638–1645
30. Rello, S., Stockert, J. C., Moreno, V., Gamez, A., Pacheco, M.,Juarranz, A., Cañete, M., and Villanueva, A. (2005)Morphological criteria to distinguish cell death inducedby apoptotic and necrotic treatments. Apoptosis 10, 201–208
31. Galluzzi, L., Maiuri, M. C., Vitale, I., Zischka, H., Castedo, M.,Zitvogel, L., and Kroemer, G. (2007) Cell death modalities:classification and pathophysiological implications. Cell DeathDiffer. 14, 1237–1243
32. Raisova, M., Hossini, A. M., Eberle, J., Riebeling, C., Wieder, T.,Sturm, I., Daniel, P. T., Orfanos, C. E., and Geilen, C. C. (2001)The Bax/Bcl-2 ratio determines the susceptibility of humanmelanoma cells to CD95/Fas-mediated apoptosis. J. Invest.Dermatol. 117, 333–340
33. Blalock, W. L., Bavelloni, A., Piazzi, M., Tagliavini, F., Faenza, I.,Martelli, A. M., Follo, M. Y., and Cocco, L. (2011) Multiple formsof PKR present in the nuclei of acute leukemia cells represent anactive kinase that is responsive to stress. Leukemia 25, 236–245
34. Lu, Z., and Xu, S. (2006) ERK1/2 MAP kinases in cell survivaland apoptosis. IUBMB Life 58, 621–631
35. Yang, J. Y., Michod, D., Walicki, J., and Widmann, C. (2004)Surviving the kiss of death. Biochem. Pharmacol. 68, 1027–1031
36. Zhuang, S., Yan, Y., Daubert, R. A., Han, J., and Schnellmann,R. G. (2007) ERK promotes hydrogen peroxide-induced apo-ptosis through caspase-3 activation and inhibition of Akt in renalepithelial cells. Am. J. Physiol. Renal Physiol. 292, F440–F447
37. Sinha, D., Bannergee, S., Schwartz, J. H., Lieberthal, W., andLevine, J. S. (2004) Inhibition of ligand-independent ERK1/2activity in kidney proximal tubular cells deprived of soluble sur-vival factors up-regulates Akt and prevents apoptosis. J. Biol. Chem.279, 10962–10972
38. Blalock, W. L., Grimaldi, C., Fala, F., Follo, M., Horn, S., Basecke,J., Martinelli, G., Cocco, L., and Martelli, A. M. (2009) PKRactivity is required for acute leukemic cell maintenance andgrowth: a role for PKR-mediated phosphatase activity to regulateGSK-3 phosphorylation. J. Cell. Physiol. 221, 232–241
39. Follo, M. Y., Finelli, C., Mongiorgi, S., Clissa, C., Bosi, C.,Martinelli, G., Blalock, W. L., Cocco, L., and Martelli, A. M.(2008) PKR is activated in MDS patients and its subcellularlocalization depends on disease severity. Leukemia 22, 2267–2269
40. Follo, M. Y., Marmiroli, S., Faenza, I., Fiume, R., Ramazzotti, G.,Martelli, A. M., Gobbi, P., McCubrey, J. A., Finelli, C., Manzoli,F. A., and Cocco, L. (2013) Nuclear phospholipase C b1signaling, epigenetics and treatments in MDS. Adv Biol Regul53, 2–7
Received for publication June 23, 2014.Accepted for publication December 1, 2014.
1394 Vol. 29 April 2015 PIAZZI ET AL.The FASEB Journal x www.fasebj.org
