Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Acute Leukemias

PP2A impaired activity is a common event in acute myeloid leukemia and its activation by forskolin has a potent anti-leukemic effect

Abstract

Protein phosphatase 2A (PP2A) is a human tumor suppressor that inhibits cellular transformation by regulating the activity of several signaling proteins critical for malignant cell behavior. PP2A has been described as a potential therapeutic target in chronic myeloid leukemia, Philadelphia chromosome-positive acute lymphoblastic leukemia and B-cell chronic lymphocytic leukemia. Here, we show that PP2A inactivation is a recurrent event in acute myeloid leukemia (AML), and that restoration of PP2A phosphatase activity by treatment with forskolin in AML cells blocks proliferation, induces caspase-dependent apoptosis and affects AKT and ERK1/2 activity. Moreover, treatment with forskolin had an additive effect with Idarubicin and Ara-c, drugs used in standard induction therapy in AML patients. Analysis at protein level of the PP2A activation status in a series of patients with AML at diagnosis showed PP2A hyperphosphorylation in 78% of cases (29/37). In addition, we found that either deregulated expression of the endogenous PP2A inhibitors SET or CIP2A, overexpression of SETBP1, or downregulation of some PP2A subunits, might be contributing to PP2A inhibition in AML. In conclusion, our results show that PP2A inhibition is a common event in AML cells and that PP2A activators, such as forskolin or FTY720, could represent potential novel therapeutic targets in AML.

Introduction

Acute myeloid leukemia (AML) is a heterogeneous clonal disease that disrupts normal hematopoiesis. Leukemic cells are characterized by a block in differentiation and apoptosis, together with an enhanced proliferation. Despite progressive advances in our understanding of the molecular biology of AML, patient outcomes are still very poor. Complete remission occurs in up to half of these patients; however, relapse is generally expected and prognosis is dismal.1 Therefore, it is necessary to develop more effective treatment strategies to improve the survival of these patients.2

The unrestricted growth of transformed cells is caused by the cumulative deregulation of multiple cellular pathways involved in normal growth control.3 The ubiquitously expressed protein phosphatase 2A (PP2A) is a major serine/threonine phosphatase that accounts for most of the serine/threonine phosphatase activity in eukaryotic cells, and participates in many mammalian signaling pathways.4 PP2A represents a family of heterotrimeric holoenzyme complexes consisting of an active core composed of the scaffold PP2A-A subunit, the catalytic PP2A-C subunit and a regulatory PP2A-B subunit. There are two closely related isoforms of the PP2A-A (Aα/PPP2R1A and Aβ/PPP2R1B),5, 6 and of the PP2A-C (Cα/PPP2CA and Cβ/PPP2CB) subunits.7, 8 The scaffold subunit mediates interaction of the core dimer with a wide variety of regulatory B subunits that regulate both the specific substrate and the localization of the holoenzyme. Four unrelated families of regulatory B subunits have been identified, including at least 26 different alternative transcripts and splice forms.9 Therefore, PP2A has the ability to form complexes with many different substrates.3, 10 A variety of mechanisms that inhibit PP2A are present in transformed cells, including alterations in structural or regulatory PP2A subunits, and also the overexpression of specific endogenous inhibitors.3, 10 Somatic mutations of the PP2A structural subunits Aα and Aβ have been described in several types of cancer, causing a defective binding of the B and C subunits and thus inhibiting PP2A activity.11, 12, 13, 14, 15, 16, 17, 18, 19, 20 Moreover, suppression of PP2A Aβ expression permits immortalized human cells to achieve a tumorigenic state through the deregulation of RaIA GTPase activity. Cancer-associated Aβ mutants fail to reverse this tumorigenic phenotype, indicating that these mutants function as null alleles.21 In addition, both Aα mutants and Aα downregulation lead to a functional haploinsufficiency that seems to induce human cell transformation by activating AKT/PI3 K signaling pathway.22, 23 However, it is likely that different sets of genetic aberrations during tumor formation require the loss of different PP2A holoenzyme complexes for the tumor progression, and this would involve the regulatory subunits that are having a key role directing PP2A to dephosphorylate and regulate key tumor suppressors or oncogenes.9 In this regard, several members of the B56 family of regulatory PP2A subunits seems to have a main role in directing PP2A potential tumor-suppressive activity.24, 25, 26, 27, 28, 29, 30

With regard to the endogenous PP2A inhibitors, upregulation of SET by the BCR/ABL oncogene leads to the suppression of PP2A, and contributes to leukemogenesis in chronic myeloid leukemia (CML) and Philadelphia chromosome-positive acute lymphoblastic leukemia.31, 32 In addition, Junttila et al.33 provide strong evidence that cancerous inhibitor of PP2A (CIP2A) selectively targets PP2A associated with c-Myc to inhibit its phosphatase activity and protect Ser62 from dephosphorylation. Interestingly, CIP2A expression is upregulated in transformed cell lines and cancer tissue samples. Finally, it has been reported that JAK2 directly phosphorylate PP2A at tyrosine 307 of its catalytic subunit, making PP2A inactive.34

Few studies have investigated the role of PP2A in AML. Gallay et al.35 reported that the intensity of phospho-Akt on Thr308 in AML was significantly correlated with high-risk cytogenetics, particularly with a complex karyotype, and they found correlation between decreased PP2A activity and Thr308 phosphorylation in this subgroup (seven cases). Moreover, a recent study shows that activating c-KIT mutations inhibit PP2A, and that reactivation of PP2A effectively suppresses the in vitro and in vivo growth of imatinib-sensitive and imatinib-resistant c-KIT-positive cells, indicating that functional inactivation of PP2A tumor suppressor activity could represent a key step in the induction and maintenance of KIT-positive leukemias.36 Our group has previously reported that SETBP1 overexpression is a recurrent event in AML, which impairs PP2A activity via SET and promotes proliferation of AML cells.37 In addition, it has been reported that the activity of PP1 and PP2A is enhanced in the arsenic sulphide-induced differentiation of the AML cell line HL-60,38 and that the alkylphosphocholine erucylphosphohomocholine is cytotoxic to AML cells through JNK- and PP2A-dependent mechanisms.39 Interestingly, PP2A activators such as FTY720 in CML, acute lymphoblastic leukemia and chronic lymphocytic leukemia, and forskolin in CML show promising anti-leukemic effects in both in vitro and in vivo models.

In this study, we show that PP2A activity is reduced in both myeloid Philadelphia-negative cell lines and AML patient samples. Treatment with the PP2A-activator forskolin restores PP2A activity, affecting proliferation and inducing changes in the phosphorylation status of AKT and ERK1/2. Moreover, we found an additive effect between PP2A activation by forskolin and the chemotherapy reagents Idarubicin and cytosine arabinoside (Ara-c), suggesting that treatment with PP2A activators could be a therapeutic target in AML in combination with standard induction therapy. Finally, deregulated expression of endogenous PP2A inhibitors, together with aberrations affecting the expression of PP2A subunits, were identified as possible mechanisms of PP2A inhibition in AML.

Materials and methods

Cell cultures

EOL-1, HL-60, Kasumi-1, OCI-AML2, MOLM13, MV4-11, HEL, KG-1, KYO-1, K562 and MEG-01 cells were maintained in RPMI-1640 (Invitrogen, Breda, The Netherlands) with 10% fetal bovine serum (FBS); NOMO-1 and KU-812 in RPMI-1640 with 20% FBS; F-36P in RPMI-1640 with 20% FBS, and 10 ng/ml GM-CSF; UT-7 in alpha-MEM (Invitrogen) with 20% FBS and 5 ng/ml GM-CSF; MUTZ-3 in 80% alpha-MEM with 20% FBS and 10 ng/ml GM-CSF; and TF-1 in RPMI-1640 with 20% FBS and 10 ng/ml GM-CSF. Cell lines were grown at 37 °C in a 5% CO2 atmosphere. Media were supplemented with penicillin G (100 U/ml) and streptomycin (0.1 mg/ml). Cells were treated with the following reagents: Idarubicin (15 nM) (Sigma-Aldrich, St Louis, MO, USA), Ara-c (2.5 μM) (Sigma-Aldrich), forskolin (40 μM) (Calbiochem, San Diego, CA, USA), FTY720 (10 μM) (Calbiochem), Z-VAD-fmk (150 mM) (Promega, Madison, WI, USA) and okadaic acid (2.5 nM) (Calbiochem).

Patient samples

The study comprised BM samples of 37 patients with AML at diagnosis. All patients were treated with standard induction chemotherapy. High-dose cytarabine, and autologous or allogenic stem cell transplantation, when possible, were used as consolidation therapy. The BM samples of normal healthy donors were used as controls. This study is part of a project approved by the Comisión de Ética de Investigación Clínica (School of Medicine, University of Navarra) (037/2008). The samples used in this study were anonymous.

Nucleic acid isolation and real-time RT-PCR

Total RNA was isolated using the RNeasy minikit (Qiagen, Hilden, Germany). cDNA was synthesized with SuperScriptIII Reverse Transcriptase (Invitrogen). Quantification of the expression of SETBP1, SET, CIP2A, PP2A catalytic subunits α (PPP2CA) and β (PPP2CB), PP2A scaffold subunit PPP2R1B, and PP2A regulatory subunits PPP2R5B and PPP2R5C were performed using TaqMan Gene Expression Assays (Applied Biosystems, Foster City, CA, USA), specific for each gene. Glyceraldehyde 3-phosphate dehydrogenase was used as internal control. Analysis of relative gene expression data was performed using the 2 - ΔΔ C T method,40 where ΔΔCT=(CT,target geneCT,GAPDH)Cell line−(CT,target geneCT,GAPDH)normal control. A gene was considered deregulated if its expression value was higher or lower than the cutoff value established for each gene (mean+3 s.d.), defined by the analysis of 10 normal BM samples.

Western blot analysis

Cells were lysed in 100 μl of Lysis buffer containing 1% Triton X-100 and the protease inhibitor cocktail Complete Mini (Roche Diagnostics, Mannheim, Germany). After incubation on ice (30 min), protein extracts were clarified (12 000 × g, 15 min, 4 °C), denatured and subjected to SDS-PAGE and western blot. Antibodies used were mouse monoclonal anti-PP2A (clone 1D6, Upstate, Temecula, CA, USA), rabbit monoclonal anti-PP2AY307 (Epitomics, Burlingame, CA, USA), rabbit polyclonal anti-Akt, rabbit polyclonal anti-ERK1/2 (Cell Signaling Technology Inc., Beverly, MA, USA), rabbit polyclonal anti-pAktThr308 rabbit polyclonal anti-pERK1/2Thr202/Tyr204 (Santa Cruz Biotechnology, Santa Cruz, CA, USA), mouse monoclonal anti-β-actin (Sigma), rabbit polyclonal anti-CIP2A (Novus Biologicals, Littleton, CO, USA), goat polyclonal anti-SET (Santa Cruz Biotechnology), goat monoclonal anti-PPP2R5B (Novus Biologicals) and rabbit polyclonal anti-PPP2R1B (Novus Biologicals). Proteins were detected with the appropriate secondary antibodies by chemiluminescence (ECL kit, GE Healthcare, Piscataway, NJ, USA).

Proliferation assay and cell viability

Cell proliferation was measured in triplicate wells by MTS assay in 96-well plates using the CellTiter 96 AQueous One Solution Cell Proliferation Assay (Promega) and following the manufacturer's indications. Results of cell viability were confirmed by the Trypan Blue dye exclusion test.

PP2A phosphatase activity assays

PP2A assays were performed with cell lysates (50 μg) using a PP2A immunoprecipitation phosphatase assay kit (Millipore, Temecula, CA, USA) as previously described.37

Analysis of apoptosis and caspase activation

Caspase 3/7 activities were measured on untreated and forskolin-treated cells using the caspase Glo-3/7 assay kit (Promega). Briefly, 5 × 103 cells were plated in a white-walled 96-well plate, and the Z-DEVD reagent, the luminogenic caspase 3/7 substrate, containing a tetrapeptide Asp-Glu-Val-Asp, was added in a 1:1 ratio of reagent to sample. After 90 min at room temperature, the substrate cleavage by activated caspase-3 and -7, and the intensity of a luminescent signal was measured by a FLUOstar OPTIMA luminometer (BMG Labtech, Offenburg, Germany). Differences in caspase-3/7 activity in forskolin-treated cells compared with untreated cells are expressed as fold-change in luminescence. Apoptosis was measured using the Annexin-V-FLUOS Staining Kit (Roche) and following manufacturer's instructions.

Statistical analysis

Data represented are mean of three independent experiments± s.d. Statistical comparisons were carried out by t-test analysis and significance was considered when P<0.05. Chou–Talalay analysis was performed using the CalcuSyn Software (Biosoft, Cambridge, UK) to determine additivity between forskolin and Idarrubicin/Ara-c treatments.

Microarray analysis

RNA samples were processed following manufacturer protocols (Affymetrix, Santa Clara, CA, USA) and hybridized to the Affymetrix Human Genome-U133 Plus-2.0, which contains 54 676 probe sets (47 000 transcripts). Microarray data analysis consisted in background correction and normalization using RMA algorithm,41 and a filtering process to eliminate low-expression probe sets. Linear Models for Microarray Data42 was used to identify the probe sets with significant differential expression. Samples were distributed in three different groups: ‘Control’ including three normal control samples, ‘AML’ including the AML cell lines EOL-1, HL-60, Kasumi-1, OCI-AML2, MOLM13, MV4-11, HEL, KG-1, NOMO-1, F-36P and TF-1, and CML including the CML in blast crisis (BC-CML) cell lines KU-812, KYO-1, K562 and MEG-01. Data represented are the gene expression mean of each group±s.d. Genes were selected as significant between the different groups using a B-statistic cutoff (B>0) or a less stringent adjusted P-value cutoff (P<0.05).

Results

PP2A is inactivated in AML cell lines

Phosphorylation of tyrosine-307 is responsible for more than 90% of the phosphatase activity of PP2A. Moreover, it has been shown that this phosphatase is inactive when tyrosine-307 is phosphorylated.43 Thus, we assessed the phosphorylation levels on tyrosine-307 of PP2A by western blot in a panel of AML BCR/ABL-negative cell lines. A total of 11 out of 13 cell lines presented increased phosphorylation in this tyrosine. Notably, the two cell lines (Kasumi-1 and MUTZ-3) that had reduced PP2A phosphorylation were the only ones with reduced expression of the catalytic subunit (PP2A-C) (Figure 1). Four BCR/ABL-positive CML cell lines, including K562, were used as positive controls for phosphorylation on PP2A tyrosine-307.31 To confirm whether the differences observed in PP2A protein expression in Kasumi-1 and MUTZ-3 were because of reduced transcriptional levels of PP2A, we analyzed by real-time PCR (qRT-PCR) the expression of the PP2A catalytic subunits PPP2CA and PPP2CB separately (Supplementary Table 1). We found no significant differences in PP2A-C transcriptional levels in these two cell lines, indicating that differences observed in western blot must be because of post-transcriptional regulation.

Figure 1
figure1

Analysis of PP2A activation and expression in 17 myeloid cell lines. Comparison of the PP2A-C expression and phosphorylation levels on tyrosine 307 by western blot in normal donors and in myeloid cell lines. N1-3, normal donors.

Treatment with forskolin leads to reduced proliferation that is dependent on PP2A activation

To assess whether increased PP2A activity affects cell proliferation of AML cells, KG-1 and HEL cell lines were treated with the PP2A activator forskolin or vehicle (dimethyl sulfoxide). Phosphatase assays to quantify PP2A activity levels confirmed that forskolin treatment activates PP2A: forskolin induced a 1.5- to 2-fold increase in PP2A activity (Figure 2a). To study whether a higher PP2A activity was associated with activation of PP2A protein, we pretreated KG-1 and HEL cells with the PP2A inhibitor okadaic acid for 2 h, followed by incubation with forskolin or vehicle for 48 h. Forskolin-induced PP2A activity in KG-1 and HEL cells was inhibited by okadaic acid (Figure 2a). Western blot analysis showed that similar levels of PP2Ac protein were immunoprecipitated in the PP2A phosphatase assays (Figure 2b), suggesting that forskolin-induced PP2A activity is not because of changes in PP2Ac expression levels.

Figure 2
figure2

Forskolin treatment induces an inhibition of the proliferation that is dependent on PP2A activation. Forskolin was used at 40 μM and okadaic acid at 2.5 nM concentrations. Data represented are mean of three independent experiments±s.d. (a) Forskolin-induced PP2A activity in KG-1 and HEL cell lines is inhibited by okadaic acid treatment. (b) Western blot analysis showing the levels of immunoprecipitated PP2A from the KG-1 and HEL lysates used in the phosphatase assays. (c) Inhibited proliferation induced by forskolin treatment is partially rescued by okadaic acid; *P<0.05; **P<0.01.

We next analyzed the effect of PP2A activation on cell growth using MTS assay. We observed a decrease in the proliferation of forskolin-treated KG-1 cells compared with vehicle treated (Figure 2c). In addition, total cell counts and cell viability were confirmed with the Trypan Blue method (data not shown). These results show that PP2A activation by forskolin treatment induces toxicity in KG-1 cells. Similar results were obtained with the HEL cell line (Figure 2c). In addition, we observed that the impaired proliferation induced by forskolin was partially rescued by the treatment with the phosphatase inhibitor okadaic acid used at a concentration that inhibits PP2A but no other phosphatases.44

To confirm that the mechanism was mainly dependent on PP2A activation, we performed the same experiments in HEL and KG-1 cell lines using the PP2A-activator FTY720 observing a similar effect with this drug than with forskolin (Supplementary Figure 1). Moreover, treatment with the PP2A-activators, forskolin and FTY720, had less effect in MUTZ-3 and Kasumi-1, the cell lines with PP2A low expression, than in HEL and KG-1, suggesting that the effect of those drugs in AML cells is mainly by PP2A activation (Supplementary Figure 2).

Increased PP2A activity by forskolin induces apoptosis in AML cells

To further investigate the biological effect of the forskolin-induced PP2A activation in AML, KG-1 cells were treated with forskolin for 48 h, and we assessed apoptosis with the caspase Glo-3/7 assay kit. Vehicle-treated cells were used as controls. Consistent with its ability to enhance PP2A activity and suppress cell proliferation, forskolin had a caspase-dependent proapoptotic effect, increasing caspase activity by 6.5-fold in forskolin-treated KG-1 cells compared with vehicle-treated cells (Figure 3a). In addition, caspase activity in forskolin-treated cells was markedly reduced when cells were pretreated with okadaic acid or the caspase inhibitor Z-VAD-fmk. Effect in apoptosis was confirmed by an Annexin-V-based assay (Figure 3b).

Figure 3
figure3

Forskolin induces caspase-dependent apoptosis together with changes in the phosphorylation status of PP2A targets. Forskolin was used at 40 μM and okadaic acid at 2.5 nM concentrations. Z-VAD-fmk was used at 150 mM. Data represented are mean of three independent experiments±s.d. (a) Caspase 3/7 assays in untreated, forskolin-treated, forskolin/okadaic acid-treated, Z-VAD-fmk-treated and forskolin/Z-VAD-fmk-treated KG-1 cells. (b) Annexin-V/propidium iodide assays in untreated, forskolin-treated, forskolin/okadaic acid-treated and forskolin/Z-VAD-fmk-treated KG-1 cells. (c) Western blot showing the effect of forskolin and forskolin/okadaic acid treatments in KG-1 cells on PP2A, AKT and ERK1/2 activity and expression.

Forskolin induces changes in the phosphorylation status of PP2A targets

We next analyzed by western blot the effects of the forskolin-induced PP2A activation at protein level. As expected, we observed that phosphorylation on tyrosine 307 of PP2Ac was negatively affected in cells treated with forskolin compared with cells treated with vehicle (dimethyl sulfoxide) (Figure 3c). These data confirmed the results obtained with the PP2A phosphatase assays (Figure 2a). In addition, PP2Ac phosphorylation was restored when cells were pretreated with the PP2A inhibitor okadaic acid. Consistent with previous reports about the effects of PP2A activation in myeloid BCR/ABL-positive cells,31 forskolin treatment in AML BCR/ABL-negative cells, decreased phosphorylation (activity) of the PP2A targets AKT and ERK1/2, without affecting their expression levels. Moreover, treatment with okadaic acid rescued AKT and ERK1/2 phosphorylation in forskolin-treated KG-1 cells (Figure 3c).

Additive effect of PP2A activation with Idarubicin and Ara-c treatments in AML cells

To assess the effect of a combination between standard induction chemotherapy drugs in AML and a PP2A activator, we treated KG-1 cells with either Idarubicin or Ara-c, alone or in combination with forskolin. Of importance, we observed that PP2A activation enhanced the anti-leukemic effects mediated by both Idarubicin (Figure 4a) and Ara-c (Figure 4b) treatments in KG-1 and Hel cell lines. Moreover, Chou–Talalay analyses showed that PP2A activation has an additive anti-tumoral effect when combined with either Idarubicin or Ara-c.

Figure 4
figure4

PP2A activation boosts anti-leukemic effects of Idarubicin and Ara-c treatments. Data represented are mean of three independent experiments±s.d.; *P<0.05; **P<0.01; (a) MTS assays showing the effect of PP2A activation by forskolin (40 μM) on the cell growth ratio of KG-1 and HEL cells treated with Idarubicin (15 nM). (b) MTS assays showing the effect of PP2A activation by forskolin (40 μM) on the cell growth ratio of KG-1 and HEL cells treated with Ara-c (2.5 μM). Cells treated with vehicle (DMSO) were used as controls.

PP2A inhibition is a recurrent event in AML

To further evaluate the importance of PP2A in AML, we analyzed at protein level the prevalence of PP2A inhibition in a series of 37 patients with AML at diagnosis. Patient characteristics are presented in Table 1. Increased phosphorylation of tyrosine 307 was observed in 29 out of 37 cases (78.4%) (Figure 5a and Supplementary Figure 3). In addition, PP2A activity was compared between samples of eight AML patients and three normal controls, and we observed a significant reduction of PP2A activity in all the eight patient samples analyzed (Supplementary Figure 4A). These results would indicate that PP2A inhibition is a recurrent event in AML. Moreover, we analyzed samples of three patients at diagnosis, complete remission and relapse, observing that PP2A phosphorylation decreases at complete remission and increases at relapse (Figure 5b and Supplementary Figure 4B).

Table 1 Clinical and molecular characteristics of a series of 37 patients with AML
Figure 5
figure5

Comparison by western blot of the PP2Ac expression and activity levels between patient samples and normal donors. (a) Analysis of PP2A in 16 samples of AML patients at diagnosis. (b) Comparison of PP2A activation in samples of a patient with AML at diagnosis, complete remission and relapse, including a densitometric analysis of the p-PP2A/PP2A ratio.

To investigate the possible causes of PP2A phosphorylation, we analyzed the expression of SET, SETBP1 and CIP2A, and the presence of the constitutive-activating mutation JAK2-V617F in this series of patients. We found overexpression of SET and/or SETBP1 in 55% cases with increased PP2A phosphorylation (16/29), and in none of the cases with low phosphorylated levels (Table 2). CIP2A overexpression was detected only in two cases with high phosphorylation. Analysis by western blot confirmed CIP2A and SET overexpression also at the protein level (Supplementary Figure 5). In addition, JAK2-V617F was detected in one of the patients with high PP2A phosphorylation, although also in three cases with low PP2A phosphorylation (Supplementary Table 2). As detected in the cell lines, six out of eight patient samples with low PP2A phosphorylation had a reduced expression of PP2A-C (Figure 5a). Interestingly, two patients had both PP2A inhibition and FLT3, constitutively activated by the FLT3-ITD mutation (Supplementary Table 2), two aberrations that led to the activation of the transduction pathways JAK/STAT, ERK and AKT.45

Table 2 Overexpression of SET and SETBP1, and presence of JAK2-V617F in a series of 37 patients with AML

Genome-wide gene expression analysis of PP2A subunits in myeloid cell lines

The fact that 12 of the patients included in our series had no known mechanism that could explain PP2A inactivation prompted us to perform expression arrays of 16 myeloid cell lines to obtain an overview of the expression of the PP2A subunits. One of these cases had a deletion del(11)(q13q23), a region in which the locus of the PP2A scaffold subunit PP2A-Aβ (PPP2R1B) (11q23) and the regulatory subunit PPP2R5B (11q13) are located. We hypothesized that as previously described,9 downregulation of some scaffold and regulatory subunits could affect the activity of PP2A.

Expression arrays of the cell lines showed a significant downregulation of the regulatory subunits PPP2R5B and PPP2R5C in both AML and BC-CML cell lines when compared with normal controls; interestingly, PPP2R5B was significantly more downregulated in AML than in CML (Supplementary Figure 6). Data were validated by qRT-PCR (Supplementary Table 1). Moreover, analysis of the patient samples revealed that PP2A-Aβ (PPP2R1B), PPP2R5B or/and PPP2R5C were downregulated in several AML patient samples (Supplementary Table 3). We could analyze at protein level 16 cases that had PPP2R5B downregulation by real time RT-PCR (qRT-PCR), and 15 out of 16 had decreased PPP2R5B protein expression. Moreover, three cases with no PPP2R5B downregulation (P17-19) were included in the study and had normal PPP2R5B protein levels (Supplementary Figure 7A). We could not perform the study of PPP2R5C at protein level. With regard to PPP2R1B, there was no good correlation between mRNA and protein: only 8 out of the 17 cases analyzed had decreased PPP2R1B protein levels, the other 9 were normal. Moreover, we included in the study five cases with no PPP2R1B downregulation by qRT-PCR: four had normal PPP2R1B protein levels (P6, P13, P17, P19), and one case had PPP2R1B low (Supplementary Figure 7B). Taken together, these results show that downregulation of these subunits is a common event in AML that could contribute to PP2A inactivation.

Discussion

PP2A is a human tumor suppressor that inhibits cellular transformation by regulating the activity of several signaling proteins critical for malignant cell behavior. We report here that PP2A inhibition is a recurrent event that could have an important role in AML. We demonstrate that PP2A activation by forskolin induce growth inhibition, caspase-dependent apoptosis, and the modification of downstream targets such as AKT and ERK1/2. Of importance, our data provides evidence that PP2A activation could be a promising therapeutic target in combination with drugs used in standard induction therapy, such as Idarubicin and Ara-c.

It has been reported that impaired PP2A activity has a key role in BCR/ABL-positive leukemias, such as CML and acute lymphoblastic leukemia,31, 32 in myeloid precursors expressing imatinib-sensitive (V560G) and imatinib-resistant (D816 V) mutant c-KIT,36 and also in chronic lymphocytic leukemia;46 moreover, activation of PP2A by either FTY720 or forskolin seems to have promising therapeutic effects in these diseases. Many kinases have been reported to be deregulated in AML; however, the role of phosphatases in the cellular transformation of this disease remains underexplored.10 As indicated above, only few studies have reported reduced PP2A activity in AML.35, 36 Our group has recently reported an impaired PP2A activity by SET as a consequence of SETBP1 overexpression, a recurrent event in AML (27%).37 This leads us to hypothesize that PP2A inhibition could be a recurrent event in AML. Our results in both cell lines and patient samples confirm this hypothesis and show that PP2A inhibition has an important role in AML transformation, as the pharmacological activation of PP2A in vitro reverse some of the leukemogenic features (Figures 2 and 3; Supplementary Figures 1 and 2). When we investigated the PP2A status at protein level in 37 patients with AML at diagnosis, we observed PP2A inhibition in 78% of cases. Interestingly, we detected reduced PP2Ac expression in six out of eight patient samples with no PP2A phosphorylation; this observation is consistent with the reduced PP2Ac levels observed in Kasumi-1 and MUTZ-3, and could suggest that a reduced protein expression could be a mechanism to decrease PP2A activity in AML. Furthermore, we also observed an additive effect of PP2A activation by forskolin with the anti-leukemic effects, induced by Idarubicin or Ara-c, in both KG-1 and HEL cells, suggesting that PP2A activation could be a new alternative for treating AML in combination with standard induction chemotherapy (Figure 4). It has been reported that forskolin-induced effects on cell growth and apoptosis of AML cells at the concentrations used in this study do not impair viability of normal BM cells;31 this observation would support the use of PP2A activators in future therapies for patients with AML. Moreover, we show that forskolin suppresses Akt and ERK1/2 function in an okadaic acid sensitive manner, indicating that its action in AML is dependent on PP2A activation (Figure 3). These results are also interesting, as several PP2A targets have been reported to be deregulated in AML and some of them, as AKT, have been associated to poor outcome.35 Our results support the premise that PP2A inactivation might be one of the events that contribute to these alterations, as PP2A has an integral role in the regulation of a number of major signaling pathways whose deregulation can contribute to cancer.9

Although evidence suggested that PP2A might be a tumor suppressor protein, recent findings provide convincing evidence that suppression of PP2A activity cooperates with other oncogenic changes to cause transformation of multiple cell types.3, 23, 47 In our series, we found that 34% of patients with PP2A inactivated, (10/29) had either FLT3 and/or NPM1 mutated (Supplementary Table 2). Interestingly, the constitutive activation of FLT3 in cases with FLT3-ITD also activates both Akt and ERK1/2,48 suggesting that in some patients PP2A inactivation and FLT3-ITD could cooperate in the transformation of AML.

PP2A can be inhibited by the small-tumor antigen of DNA tumor viruses, by upregulation of endogenous PP2A inhibitors, through mutational inactivation of the structural subunits, or by decreased expression of either the scaffold or the regulatory subunits.9, 10 On the other hand, JAK2 constitutive activity49, 50 and SETBP1 overexpression in BCR/ABL-negative AML37 might independently contribute to PP2A inactivation. Our results show that the mechanisms of PP2A inactivation in AML might be the overexpression of the physiological PP2A inhibitors CIP2A33 and SET,51 the overexpression of SETBP1, or the downregulation of PP2A subunits, suggesting that dysfunction of several distinct PP2A complexes may contribute to cell transformation. Overexpression of CIP2A, SET or SETBP1 could explain the mechanism of PP2A inactivation in 58% of our cases (17/29). SET is upregulated in multiple solid tumors,52 and has been reported to be fused to NUP214/CAN in a patient with AML.53 Importantly, Neviani et al.31 demonstrated that PP2A inactivation in CML-BC results from increased expression of SET, which is induced by BCR/ABL in a dose- and kinase-dependent manner and, that like BCR/ABL, SET progressively increases during transition to blast crisis. In fact, imatinib treatment and SET downregulation restored PP2A activity back to normal levels. Our results show that high expression of SET also leads to PP2A inactivation in AML, independently of BCR/ABL induction. In addition, the activating mutation JAK2-V617F was detected in one of the samples analyzed; however, three cases with JAK2-V617F had no PP2A hyperphosphorylation, suggesting that either this mutation would need other additional changes to inactivate PP2A or that JAK2-V617F is not causal in activating PP2A in AML.

As in 12 cases, the mechanism remained undetermined, and one case had a deletion del(11)(q13q23), a region in which the loci of the PP2A scaffold subunit PP2A-Aβ (PPP2R1B) (11q23) and the regulatory subunit PPP2R5B (11q13) are located, we hypothesized that downregulation of some subunits could affect the activity of PP2A, as previously reported.9 Analysis of 16 myeloid cell lines and AML patient samples showed genetic aberrations affecting PP2A subunits that could be having an important role in the PP2A inhibition observed in AML. We found downregulation of PPP2R5B and PPP2R5C in both AML and CB-CML cell lines (Supplementary Figure 6). It has been reported that PPP2R5B is a tumor suppressor that negatively regulates Pim-1 protein kinase, which is known to enhance the ability of c-Myc to induce lymphomas.26 Furthermore, it has been described that suppression of PPP2R5C expression contributes to the experimental transformation of human cells.28 Our data suggest that loss of PPP2R5B and PPP2R5C could be having a role in AML development, contributing to deregulate the correct PP2A function. We also found that downregulation of the Aβ subunit is a common event in AML. Most cellular PP2A holoenzymes contain the Aα isoform of the scaffold subunit, but a small fraction (10%) contain a second isoform-termed Aβ. Although mutations that disrupt the ability of Aβ to form holoenzymes in vitro were identified in several types of cancer,12, 54, 55 the report by Sablina et al.21 provides the first hard evidence that loss of functional Aβ caused by these cancer-associated mutations contribute to transformation.3 Further studies are necessary to clarify the importance of the downregulation of these PP2A subunits in AML.

Taking together, our results suggest that functional inactivation of the PP2A tumor suppressor is a recurrent event that seems to represent an important mechanism in the leukemogenic transformation of AML. We show that functional loss of PP2A activity could occur through different contributing mechanisms such as enhancement of endogenous PP2A inhibitors, decreased PP2A expression of the structural or regulatory subunits of PP2A. Moreover, although PP2A activators are not still clinically available, the results obtained in this study suggest that PP2A activation could be considered as a future therapeutic alternative for AML. The knowledge that pharmacological restoration of PP2A activity is able to antagonize leukemogenesis and has an additive effect with other drugs used in the treatment of AML highlights PP2A as a potential target for future therapies combined with PP2A activators.

References

  1. 1

    Rowe JM . Optimal induction and post-remission therapy for AML in first remission. Hematology Am Soc Hematol Educ Program 2009; 1: 396–405.

    Article  Google Scholar 

  2. 2

    Haferlach T . Molecular genetic pathways as therapeutic targets in acute myeloid leukemia. Hematology Am Soc Hematol Educ Program 2008; 1: 400–411.

    Article  Google Scholar 

  3. 3

    Mumby M . PP2A: unveiling a reluctant tumor suppressor. Cell 2007; 130: 21–24.

    CAS  Article  Google Scholar 

  4. 4

    Millward TA, Zolnierowicz S, Hemmings BA . Regulation of protein kinase cascades by protein phosphatase 2A. Trends Biochem Sci 1999; 24: 186–191.

    CAS  Article  Google Scholar 

  5. 5

    Hemmings BA, Adams-Pearson C, Maurer F, Muller P, Goris J, Merlevede W et al. α- and β-Forms of the 65-kDa subunit of protein phosphatase 2A have a similar 39 amino acid repeating structure. Biochemistry 1990; 29: 3166–3173.

    CAS  Article  Google Scholar 

  6. 6

    Zhou J, Pham HT, Ruediger R, Walter G . Characterization of the Aα and Aβ subunit isoforms of protein phosphatase 2A: differences in expression, subunit interaction, and evolution. Biochem J 2003; 369: 387–398.

    CAS  Article  Google Scholar 

  7. 7

    Arino J, Woon CW, Brautigan DL, Miller Jr TB, Johnson GL . Human liver phosphatase 2A: cDNA and amino acid sequence of two catalytic subunit isotypes. Proc Natl Acad Sci USA 1988; 85: 4252–4256.

    CAS  Article  Google Scholar 

  8. 8

    Cohen P . The structure and regulation of protein phosphatases. Annu Rev Biochem 1989; 58: 453–508.

    CAS  Article  Google Scholar 

  9. 9

    Eichhorn PJ, Creyghton MP, Bernards R . Protein phosphatase 2A regulatory subunits and cancer. Biochim Biophys Acta 2009; 1795: 1–15.

    CAS  PubMed  Google Scholar 

  10. 10

    Westermarck J, Hahn WC . Multiple pathways regulated by the tumor suppressor PP2A in transformation. Trends Mol Med 2008; 14: 152–160.

    CAS  Article  Google Scholar 

  11. 11

    Ruediger R, Pham HT, Walter G . Disruption of protein phosphatase 2A subunit interaction in human cancers with mutations in the A alpha subunit gene. Oncogene 2001; 20: 10–15.

    CAS  Article  Google Scholar 

  12. 12

    Ruediger R, Pham HT, Walter G . Alterations in protein phosphatase 2A subunit interaction in human carcinomas of the lung and colon with mutations in the A beta subunit gene. Oncogene 2001; 20: 1892–1899.

    CAS  Article  Google Scholar 

  13. 13

    Wang SS, Esplin ED, Li JL, Huang L, Gazdar A, Minna J et al. Alterations of the PPP2R1B gene in human lung and colon cancer. Science 1998; 282: 284–287.

    CAS  Article  Google Scholar 

  14. 14

    Calin GA, di Iasio MG, Caprini E, Vorechovsky I, Natali PG, Sozzi G et al. Low frequency of alterations of the alpha (PPP2R1A) and beta (PPP2R1B) isoforms of the subunit A of the serine-threonine phosphatase 2A in human neoplasms. Oncogene 2000; 19: 1191–1195.

    CAS  Article  Google Scholar 

  15. 15

    Esplin ED, Ramos P, Martinez B, Tomlinson GE, Mumby MC, Evans GA . The glycine 90 to aspartate alteration in the Abeta subunit of PP2A (PPP2R1B) associates with breast cancer and causes a deficit in protein function. Genes Chromosomes Cancer 2006; 45: 182–190.

    CAS  Article  Google Scholar 

  16. 16

    Takagi Y, Futamura M, Yamaguchi K, Aoki S, Takahashi T, Saji S . Alterations of the PPP2R1B gene located at 11q23 in human colorectal cancers. Gut 2000; 47: 268–271.

    CAS  Article  Google Scholar 

  17. 17

    Tamaki M, Goi T, Hirono Y, Katayama K, Yamaguchi A . PPP2R1B gene alterations inhibit interaction of PP2A-Abeta and PP2A-C proteins in colorectal cancers. Oncol Rep 2004; 11: 655–659.

    CAS  PubMed  Google Scholar 

  18. 18

    Kalla C, Scheuermann MO, Kube I, Schlotter M, Mertens D, Dohner H et al. Analysis of 11q22-q23 deletion target genes in B-cell chronic lymphocytic leukaemia: evidence for a pathogenic role of NPAT, CUL5, and PPP2R1B. Eur J Cancer 2007; 43: 1328–1335.

    CAS  Article  Google Scholar 

  19. 19

    Hemmer S, Wasenius VM, Haglund C, Zhu Y, Knuutila S, Franssila K et al. Alterations in the suppressor gene PPP2R1B in parathyroid hyperplasias and adenomas. Cancer Genet Cytogenet 2002; 134: 13–17.

    CAS  Article  Google Scholar 

  20. 20

    Chou HC, Chen CH, Lee HS, Lee CZ, Huang GT, Yang PM et al. Alterations of tumour suppressor gene PPP2R1B in hepatocellular carcinoma. Cancer Lett 2007; 253: 138–143.

    CAS  Article  Google Scholar 

  21. 21

    Sablina AA, Chen W, Arroyo JD, Corral L, Hector M, Bulmer SE et al. The tumor suppressor PP2A Abeta regulates the RalA GTPase. Cell 2007; 129: 969–982.

    CAS  Article  Google Scholar 

  22. 22

    Chen W, Arroyo JD, Timmons JC, Possemato R, Hahn WC . Cancer-associated PP2A Aalpha subunits induce functional haploinsufficiency and tumorigenicity. Cancer Res 2005; 65: 8183–8192.

    CAS  Article  Google Scholar 

  23. 23

    Sablina AA, Hahn WC . The role of PP2A A subunits in tumor suppression. Cell Adh Migr 2007; 1: 140–141.

    Article  Google Scholar 

  24. 24

    Arnold HK, Sears RC . Protein phosphatase 2A regulatory subunit B56alpha associates with c-myc and negatively regulates c-myc accumulation. Mol Cell Biol 2006; 26: 2832–2844.

    CAS  Article  Google Scholar 

  25. 25

    Margolis SS, Perry JA, Forester CM, Nutt LK, Guo Y, Jardim MJ et al. Role for the PP2A/B56delta phosphatase in regulating 14-3-3 release from Cdc25 to control mitosis. Cell 2006; 127: 759–773.

    CAS  Article  Google Scholar 

  26. 26

    Ma J, Arnold HK, Lilly MB, Sears RC, Kraft AS . Negative regulation of Pim-1 protein kinase levels by the B56beta subunit of PP2A. Oncogene 2007; 26: 5145–5153.

    CAS  Article  Google Scholar 

  27. 27

    Letourneux C, Rocher G, Porteu F . B56-containing PP2A dephosphorylate ERK and their activity is controlled by the early gene IEX-1 and ERK. EMBO J 2006; 25: 727–738.

    CAS  Article  Google Scholar 

  28. 28

    Chen W, Possemato R, Campbell KT, Plattner CA, Pallas DC, Hahn WC . Identification of specific PP2A complexes involved in human cell transformation. Cancer Cell 2004; 5: 127–136.

    CAS  Article  Google Scholar 

  29. 29

    Shouse GP, Nobumori Y, Liu X . A B56gamma mutation in lung cancer disrupts the p53-dependent tumor-suppressor function of protein phosphatase 2A. Oncogene 2010; 29: 1669–1681.

    Article  Google Scholar 

  30. 30

    Li HH, Cai X, Shouse GP, Piluso LG, Liu X . A specific PP2A regulatory subunit, B56gamma, mediates DNA damage-induced dephosphorylation of p53 at Thr55. EMBO J 2007; 26: 402–411.

    CAS  Article  Google Scholar 

  31. 31

    Neviani P, Santhanam R, Trotta R, Notari M, Blaser BW, Liu S et al. The tumor suppressor PP2A is functionally inactivated in blast crisis CML through the inhibitory activity of the BCR/ABL-regulated SET protein. Cancer Cell 2005; 8: 355–368.

    CAS  Article  Google Scholar 

  32. 32

    Neviani P, Santhanam R, Oaks JJ, Eiring AM, Notari M, Blaser BW et al. FTY720, a new alternative for treating blast crisis chronic myelogenous leukemia and Philadelphia chromosome-positive acute lymphocytic leukemia. J Clin Invest 2007; 117: 2408–2421.

    CAS  Article  Google Scholar 

  33. 33

    Junttila MR, Puustinen P, Niemela M, Ahola R, Arnold H, Bottzauw T et al. CIP2A inhibits PP2A in human malignancies. Cell 2007; 130: 51–62.

    CAS  Article  Google Scholar 

  34. 34

    Yokoyama N, Reich NC, Miller WT . Determinants for the interaction between Janus kinase 2 and protein phosphatase 2A. Arch Biochem Biophys 2003; 417: 87–95.

    CAS  Article  Google Scholar 

  35. 35

    Gallay N, Dos Santos C, Cuzin L, Bousquet M, Simmonet Gouy V, Chaussade C et al. The level of AKT phosphorylation on threonine 308 but not on serine 473 is associated with high-risk cytogenetics and predicts poor overall survival in acute myeloid leukaemia. Leukemia 2009; 23: 1029–1038.

    CAS  Article  Google Scholar 

  36. 36

    Roberts KG, Smith AM, McDougall F, Carpenter H, Horan M, Neviani P et al. Essential requirement for PP2A inhibition by the oncogenic receptor c-KIT suggests PP2A reactivation as a strategy to treat c-KIT+ cancers. Cancer Res 2010; 70: 5438–5447.

    CAS  Article  Google Scholar 

  37. 37

    Cristobal I, Blanco FJ, Garcia-Orti L, Marcotegui N, Vicente C, Rifon J et al. SETBP1 overexpression is a novel leukemogenic mechanism that predicts adverse outcome in elderly patients with acute myeloid leukemia. Blood 2010; 115: 615–625.

    CAS  Article  Google Scholar 

  38. 38

    Luo LY, Huang J, Gou BD, Zhang TL, Wang K . Induction of human promyelocytic leukemia HL-60 cell differentiation into monocytes by arsenic sulphide: involvement of serine/threonine protein phosphatases. Leuk Res 2006; 30: 1399–1405.

    CAS  Article  Google Scholar 

  39. 39

    Martelli AM, Papa V, Tazzari PL, Ricci F, Evangelisti C, Chiarini F et al. Erucylphosphohomocholine, the first intravenously applicable alkylphosphocholine, is cytotoxic to acute myelogenous leukemia cells through JNK- and PP2A-dependent mechanisms. Leukemia 2010; 24: 687–698.

    CAS  Article  Google Scholar 

  40. 40

    Livak KJ, Schmittgen TD . Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods 2001; 25: 402–408.

    CAS  Article  Google Scholar 

  41. 41

    Irizarry RA, Bolstad BM, Collin F, Cope LM, Hobbs B, Speed TP . Summaries of affymetrix genechip probe level data. Nucleic Acids Res 2003; 31: e15.

    Article  Google Scholar 

  42. 42

    Smyth GK . Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat Appl Genet Mol Biol 2004; 3: Article 3.

    Article  Google Scholar 

  43. 43

    Chen J, Martin BL, Brautigan DL . Regulation of protein serine-threonine phosphatase type-2A by tyrosine phosphorylation. Science 1992; 257: 1261–1264.

    CAS  Article  Google Scholar 

  44. 44

    Saydam G, Aydin HH, Sahin F, Selvi N, Oktem G, Terzioglu E et al. Involvement of protein phosphatase 2A in interferon-alpha-2b-induced apoptosis in K562 human chronic myelogenous leukaemia cells. Leuk Res 2003; 27: 709–717.

    CAS  Article  Google Scholar 

  45. 45

    Gilliland DG, Griffin JD . Role of FLT3 in leukemia. Curr Opin Hematol 2002; 9: 274–281.

    Article  Google Scholar 

  46. 46

    Liu Q, Zhao X, Frissora F, Ma Y, Santhanam R, Jarjoura D et al. FTY720 demonstrates promising preclinical activity for chronic lymphocytic leukemia and lymphoblastic leukemia/lymphoma. Blood 2008; 111: 275–284.

    CAS  Article  Google Scholar 

  47. 47

    Junttila MR, Westermarck J . Mechanisms of MYC stabilization in human malignancies. Cell Cycle 2008; 7: 592–596.

    CAS  Article  Google Scholar 

  48. 48

    Kornblau SM, Singh N, Qiu Y, Chen W, Zhang N, Coombes KR . Highly phosphorylated FOXO3A is an adverse prognostic factor in acute myeloid leukemia. Clin Cancer Res 2010; 16: 1865–1874.

    CAS  Article  Google Scholar 

  49. 49

    Samanta AK, Chakraborty SN, Wang Y, Kantarjian H, Sun X, Hood J et al. Jak2 inhibition deactivates Lyn kinase through the SET-PP2A-SHP1 pathway, causing apoptosis in drug-resistant cells from chronic myelogenous leukemia patients. Oncogene 2009; 28: 1669–1681.

    CAS  Article  Google Scholar 

  50. 50

    Yokoyama N, Reich NC, Miller WT . Involvement of protein phosphatase 2A in the interleukin-3-stimulated Jak2-Stat5 signaling pathway. J Interferon Cytokine Res 2001; 21: 369–378.

    CAS  Article  Google Scholar 

  51. 51

    Li M, Makkinje A, Damuni Z . The myeloid leukemia-associated protein SET is a potent inhibitor of protein phosphatase 2A. J Biol Chem 1996; 271: 11059–11062.

    CAS  Article  Google Scholar 

  52. 52

    Cervoni N, Detich N, Seo SB, Chakravarti D, Szyf M . The oncoprotein Set/TAF-1beta, an inhibitor of histone acetyltransferase, inhibits active demethylation of DNA, integrating DNA methylation and transcriptional silencing. J Biol Chem 2002; 277: 25026–25031.

    CAS  Article  Google Scholar 

  53. 53

    Rosati R, La Starza R, Barba G, Gorello P, Pierini V, Matteucci C et al. Cryptic chromosome 9q34 deletion generates TAF-Ialpha/CAN and TAF-Ibeta/CAN fusion transcripts in acute myeloid leukemia. Haematologica 2007; 92: 232–235.

    CAS  Article  Google Scholar 

  54. 54

    Xu Y, Xing Y, Chen Y, Chao Y, Lin Z, Fan E et al. Structure of the protein phosphatase 2A holoenzyme. Cell 2006; 127: 1239–1251.

    CAS  Article  Google Scholar 

  55. 55

    Cho US, Xu W . Crystal structure of a protein phosphatase 2A heterotrimeric holoenzyme. Nature 2007; 445: 53–57.

    CAS  Article  Google Scholar 

Download references

Acknowledgements

This work was supported by the Ministerio Educación y Ciencia (SAF2005/06425 and AP2006-03038), Ministerio Ciencia e Innovación (PI081687), Departamento Salud del Gobierno de Navarra (14/2008), ISCIII-RTICC (RD06/0020/0078) and Fundación para la Investigación Médica Aplicada y UTE (Spain). We thank Dr Anna Sablina for useful discussion, Elizabeth Guruceaga of the Unit of Proteomics, Genomics and Bioinformatics of CIMA for bioinformatics analysis, and Leyre Urquiza and Marisol Gonzalez-Huarriz for technical assistance.

Author information

Affiliations

Authors

Corresponding author

Correspondence to M D Odero.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies the paper on the Leukemia website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Cristóbal, I., Garcia-Orti, L., Cirauqui, C. et al. PP2A impaired activity is a common event in acute myeloid leukemia and its activation by forskolin has a potent anti-leukemic effect. Leukemia 25, 606–614 (2011). https://doi.org/10.1038/leu.2010.294

Download citation

Keywords

  • AML
  • PP2A
  • forskolin
  • therapy
  • SET

Further reading

Search

Quick links