Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2013 Jul 9;110(30):12420–12425. doi: 10.1073/pnas.1305656110

PTEN loss defines a PI3K/AKT pathway-dependent germinal center subtype of diffuse large B-cell lymphoma

Matthias Pfeifer a, Michael Grau b, Dido Lenze c, Sören-Sebastian Wenzel a, Annette Wolf a, Brigitte Wollert-Wulf a, Kerstin Dietze a, Hendrik Nogai a, Benjamin Storek a, Hannelore Madle a, Bernd Dörken a, Martin Janz a, Stephan Dirnhofer d, Peter Lenz b,✉, Michael Hummel c, Alexandar Tzankov d, Georg Lenz a,1
PMCID: PMC3725065  PMID: 23840064

Abstract

Diffuse large B-cell lymphoma (DLBCL) represents a heterogeneous diagnostic category with distinct molecular subtypes that can be defined by gene expression profiling. However, even within these defined subtypes, heterogeneity prevails. To further elucidate the pathogenesis of these entities, we determined the expression of the tumor suppressor phosphatase and tensin homolog (PTEN) in 248 primary DLBCL patient samples. These analyses revealed that loss of PTEN was detectable in 55% of germinal center B-cell-like (GCB) DLBCLs, whereas this abnormality was found in only 14% of non-GCB DLBCL patient samples. In GCB DLBCL, the PTEN status was inversely correlated with activation of the oncogenic PI3K/protein kinase B (AKT) pathway in both DLBCL cell lines and primary patient samples. Reexpression of PTEN induced cytotoxicity in PTEN-deficient GCB DLBCL cell line models by inhibiting PI3K/AKT signaling, indicating an addiction to this pathway in this subset of GCB DLBCLs. PI3K/AKT inhibition induced down-regulation of the transcription factor MYC. Reexpression of MYC rescued GCB DLBCL cells from PTEN-induced toxicity, identifying a regulatory mechanism of MYC expression in DLBCL. Finally, pharmacologic PI3K inhibition resulted in toxicity selectively in PTEN-deficient GCB DLBCL lines. Collectively, our results indicate that PTEN loss defines a PI3K/AKT-dependent GCB DLBCL subtype that is addicted to PI3K and MYC signaling and suggest that pharmacologic inhibition of PI3K might represent a promising therapeutic approach in these lymphomas.

Diffuse large B-cell lymphoma (DLBCL) represents the most frequent lymphoma subtype and is considered a heterogeneous diagnostic category (1). Using gene expression profiling, two major molecular subtypes can be distinguished termed germinal center B-cell-like (GCB) DLBCL and activated B-cell-like (ABC) DLBCL (2). GCB DLBCLs are derived from germinal center B cells, whereas ABC DLBCLs originate from postgerminal center B cells that are in the transition of being differentiated into plasma cells. However, full plasma cell maturation is blocked in ABC DLBCL by different genetic abnormalities inhibiting the function of BLIMP1 that regulates plasmacytic differentiation (3–5).

Recent work suggested constitutive activation of the PI3K/protein kinase B (AKT) pathway that plays a crucial role in mediating growth, proliferation, and cell survival in a substantial number of DLBCL patient samples determined by immunohistochemical staining for phospho-AKT (p-AKT) (6, 7). However, these studies did not investigate the molecular mechanisms leading to constitutive PI3K/AKT signaling. The tumor suppressor PTEN is the major negative regulator of PI3K/AKT. PTEN functions as a lipid phosphatase dephosphorylating the 3′ position of phosphatidyl-inositol-3,-4,-5-trisphosphate, which serves as a trigger for AKT activation (8, 9). However, recent studies showed that PTEN has additional PI3K/AKT-independent tumor suppressor functions. Nuclear PTEN, for example, acts as guardian of genome integrity by up-regulation of RAD51 that is involved in DNA repair (10). Furthermore, nuclear PTEN can enhance the E3 ligase activity of APC/C by promoting the association of APC/C with CDH1. These complexes degrade oncoproteins, such as polo-like kinase 1 and aurora kinases (11, 12).

The molecular mechanisms leading to constitutive activation of PI3K/AKT signaling in DLBCL remain largely unknown. In ABC DLBCL, ∼20% of patient samples harbor CD79A or CD79B mutations, leading to chronic active B-cell receptor signaling promoting PI3K/AKT activation (13). In contrast, ∼10% of GCB DLBCLs are characterized by heterozygous deletions of the PTEN locus. Intriguingly, these aberrations were not detectable in ABC DLBCL, suggesting a role of PTEN in the pathogenesis of GCB DLBCL (14).

In the present study, we investigated the functional role of PTEN in the biology of DLBCL. We detected that PTEN expression is lost in the majority of GCB DLBCLs. PTEN loss was inversely correlated with activation of PI3K/AKT in these lymphomas, thus identifying a PTEN-deficient PI3K-dependent GCB DLBCL subset. Reexpression of PTEN in PTEN-deficient cells induced toxicity that is caused by inhibition of PI3K/AKT signaling and down-regulation of the transcription factor MYC. Finally, we show that pharmacologic inhibition of PI3K is toxic to PTEN-deficient GCB DLBCLs, identifying PI3K as a potential target for these lymphomas.

Results

PTEN Is Differentially Expressed in DLBCL Subtypes.

To investigate PTEN expression in molecular DLBCL subtypes, we performed immunohistochemical staining for PTEN in 34 DLBCL patient samples that were classified into ABC, GCB, and unclassified DLBCLs by gene expression profiling (cohort 1). Detection of PTEN by immunohistochemistry was established by staining tonsils (n = 8) and normal lymph nodes (n = 2) (Fig. 1A) as well as PTEN-positive and -negative DLBCL cell lines. As expected, the vast majority of cells stained positive in the tonsils and the normal lymph nodes. PTEN expression was detectable in up to 80% of centrocytes and up to 50% of centroblasts in both the light and the dark zone. Furthermore, 60–80% of the cells in the inter- and perifollicular area stained for PTEN. To detect samples that completely lost PTEN expression in DLBCL, we applied a cutoff of <5% of positive lymphoma cells to differentiate PTEN-positive and -negative samples. The DLBCL cell lines TMD8, OCI-Ly10, and OCI-Ly19 expressed PTEN by immunohistochemistry, whereas all other cell lines (OCI-Ly1, BJAB, and HT) had no detectable PTEN levels (Fig. 1 B and C). As shown below, these immunohistochemistry results confirmed our Western blot data (Fig. 1H). In cohort 1, a total of 22 primary DLBCL samples (65%) expressed PTEN. However, the frequency of PTEN expression was different in molecular DLBCL subtypes. Seven of eight (88%) ABC DLBCLs and six of eight (75%) unclassified DLBCLs expressed PTEN. In contrast, only 9 of 18 (50%) GCB DLBCLs had detectable PTEN levels (P = 0.08; PTEN expression in GCB DLBCL vs. other subtypes; Fisher exact test).

Fig. 1.

Fig. 1.

Open in a new tab

PTEN is differentially expressed in DLBCL subtypes. (A–F) Immunohistochemical PTEN staining of (A) a tonsil, (B) the ABC DLBCL cell line TMD8 that expresses PTEN, (C) the PTEN-negative cell line HT, (D) a PTEN-positive DLBCL case, (E) a PTEN-negative DLBCL case with positive internal control (tumor-infiltrating lymphocytes), and (F) a PTEN-negative DLBCL case with positive internal control (blood vessels). (Magnification: A, 120×; B and C, 400×; D–F, 320×.) (G) PTEN loss determined by immunohistochemistry is more frequent in GCB compared with non-GCB DLBCLs (P = 2.1 × 10−11). (H) Analysis of PTEN expression and AKT phosphorylation in DLBCL cell lines. PTEN expression and p-AKT status are inversely correlated in GCB DLBCL cell lines. Bars indicate different blots. OCI-Ly1 cells were used as internal positive controls for p-AKT on both blots. OCI-Ly1 cells treated with the pan-PI3K inhibitor Ly294002 were used as negative control for p-AKT. (I) PTEN expression and p-AKT status are inversely correlated in GCB DLBCL primary samples. Bars indicate different blots. Patient sample 12 was used as an internal positive control for p-AKT on both blots.

To investigate a second and significantly larger cohort of patients, we analyzed PTEN expression in an independent cohort of 249 primary DLBCL patient samples (cohort 2); 214 samples (86%) were evaluable for PTEN staining (Fig. 1 D–F). In this cohort, only 31 of 71 (44%) GCB DLBCLs were PTEN-positive. In contrast, 124 of 143 (87%) of non-GCB DLBCL cases expressed PTEN (P = 1.0 × 10−10; Fisher exact test). Combining the results of both cohorts, we detected PTEN loss in 49 of 89 (55%) GCB DLBCLs compared with only 22 of 159 (14%) in non-GCB DLBCL patient samples (P = 2.1 × 10−11; results of PTEN expression using a cutoff of <10% of positive lymphoma cells are additionally summarized in Table S1) (Fig. 1G). These results indicate that (i) loss of PTEN is detectable almost exclusively in GCB DLBCL and (ii) PTEN expression divides GCB DLBCLs into two subsets.

Next, we analyzed whether established DLBCL cell lines represent adequate models for functional analyses. To this end, we screened 14 DLBCL cell lines for PTEN expression by Western blotting (Fig. 1H). All three ABC DLBCL cell lines had detectable PTEN expression, whereas only four GCB DLBCL lines expressed PTEN at comparable levels (OCI-Ly2, OCI-Ly7, OCI-Ly19, and DB). In contrast, 7 of 11 (64%) GCB DLBCL cell lines had complete PTEN loss, suggesting that these cell lines represent adequate functional models.

Different Molecular Mechanisms Contribute to PTEN Loss.

To determine if genomic deletions of the PTEN locus on 10q23 contribute to loss of PTEN, we performed array comparative genomic hybridization (aCGH) in 12 DLBCL cell lines (TMD8, OCI-Ly10, HBL-1, OCI-Ly1, OCI-Ly2, OCI-Ly7, BJAB, HT, SUDHL-10, K422, DB, and RL). We detected heterozygous PTEN deletions only in GCB DLBCL cell lines (SUDHL-10, OCI-Ly1, K422, and DB) (Fig. 2A, Fig. S1A, and Table S2). In OCI-Ly1, K422, and SUDHL-10, PTEN protein expression was not detectable, indicating silencing of the second allele (Fig. 1H). In contrast, in DB cells, PTEN protein levels were not reduced, suggesting expression of the second PTEN allele (Fig. 1H). We performed a quantitative genomic PCR to confirm these results as well as determine the PTEN copy number in two additional cell lines (OCI-Ly4 and OCI-Ly19). This assay reproduced the aCGH data in all cell lines except OCI-Ly1, in which a homozygous deletion was detected (Fig. S1B and Table S2). PCR analysis of OCI-Ly1 DNA confirmed an internal PTEN deletion of the second allele affecting exons 6–9 (Fig. S1C) that was not detected by aCGH. To determine the PTEN copy number status in primary patient samples, we applied this quantitative PCR assay to all 34 DLBCL samples of cohort 1. We detected heterozygous PTEN deletions in 3 of 18 GCB DLBCLs and 1 of 8 ABC DLBCLs (Fig. 2B and Table S3).

Fig. 2.

Fig. 2.

Open in a new tab

PTEN is deregulated by different molecular mechanisms in GCB DLBCL. (A) SUDHL-10 cells are characterized by a heterozygous deletion of the PTEN locus. (B) Analysis of PTEN DNA copy number using quantitative PCR. Each bar represents a patient sample. (C) Schematic overview of PTEN domains. Mutations detected in GCB (orange) and ABC (blue) DLBCL samples are shown.

To investigate if somatic mutations contribute to PTEN loss in DLBCL, we sequenced the PTEN coding exons in our panel of 14 DLBCL cell lines and 34 primary DLBCL patient samples (cohort 1, Table S4). Previous work had indicated that the frequency of PTEN mutations is rather low (15–17). However, these analyses did not investigate whether mutations occur preferentially in certain molecular DLBCL subtypes. We detected mutations in four cell lines, all derived from GCB DLBCL patients (OCI-Ly4, SUDHL-10, K422, and RL) (Fig. 2C and Fig. S2 A–D). These mutations lead to amino acid changes, stop codons, or frame shifts. Two cell lines (SUDHL-10 and RL) harbored two mutations on the same allele, which was shown by cloning and sequencing of PCR products (Fig. 2C and Fig. S2 A and B). Interestingly, all affected cell lines did not have detectable PTEN expression (Fig. 1H). Furthermore, we detected PTEN mutations in one primary GCB and one ABC DLBCL patient sample. Both aberrations were somatically acquired, because they were not detectable in corresponding nontumor DNA (Fig. S2 E and F). The affected ABC DLBCL patient sample had detectable PTEN levels by immunohistochemistry. In contrast, the GCB DLBCL sample was completely negative for PTEN (Fig. S2G). All PTEN mutations were confined to the phosphatase and the C2 domain. The phosphatase domain is critical for the phosphatase activity of PTEN, whereas the C2 domain has been shown to affect membrane binding and productive orientation of the phosphatase domain at the membrane surface (12, 18). Collectively, our results indicate that PTEN mutations are rare but seem to be associated with PTEN loss in GCB DLBCL.

Combining our mutational and copy number analyses, we detected monoallelic PTEN inactivation in DB, OCI-Ly4, and RL cells as well as two primary GCB and two ABC DLBCL samples, whereas K422, SUDHL-10, OCI-Ly1, and the GCB DLBCL sample 028 are characterized by biallelic PTEN loss.

Next, we investigated if the transcription factor MYC is involved in PTEN regulation in DLBCL. MYC has been shown to down-regulate PTEN expression by up-regulation of microRNA-17-92 (miR-17-92) (19, 20). To this end, we evaluated 128 of 283 (45%) patient samples from cohorts 1 and 2 for the presence of MYC translocations using FISH. Within the GCB DLBCLs, we detected seven cases with an MYC translocation and one case harboring an MYC amplification (8/46; 17%). In contrast, only 2 of 82 (2%) non-GCB DLBCLs harbored MYC translocations (P = 0.004). However, MYC aberrations were not associated with PTEN loss, because only 2 of 8 GCB DLBCLs with MYC aberrations were PTEN-negative compared with 18 of 38 (47%) cases without MYC aberrations (P = 0.44).

PTEN Loss Defines a GCB DLBCL Subset that Is Addicted to PI3K/AKT Signaling.

To study whether PTEN loss is associated with constitutive activation of the PI3K/AKT signaling pathway, we performed Western blotting for p-AKT that we used as a surrogate for PI3K/AKT activation. In GCB DLBCL cell lines, we detected an inverse correlation between PTEN expression and p-AKT levels. Cell lines without PTEN expression had high basal phosphorylation of AKT (OCI-Ly1, OCI-Ly4, BJAB, HT, SUDHL-10, and K422) with the exception of RL, which showed only low p-AKT levels (Fig. 1H). In contrast, all GCB DLBCL cell lines with PTEN expression had either low or undetectable p-AKT levels (OCI-Ly2, OCI-Ly7, OCI-LY19, and DB) (Fig. 1H). To determine if this inverse pattern of PTEN loss and phosphorylation of AKT is also detectable in primary DLBCL samples, we obtained protein samples from 8 PTEN-positive and 12 PTEN-deficient GCB DLBCL patient samples (Table S5). Western blotting for p-AKT confirmed the inverse correlation, because all samples with PTEN loss had detectable p-AKT levels (12/12), whereas only two of eight samples with PTEN expression had phosphorylation of AKT. The remaining cases had either undetectable or extremely low p-AKT levels (P = 7.2 × 10−4; two-sided Fisher exact test) (Fig. 1I). In contrast, constitutive AKT phosphorylation was not associated with PTEN loss in ABC DLBCL cell lines. Despite PTEN expression, all lines had detectable phosphorylation of AKT, suggesting that different molecular mechanisms activate the PI3K/AKT pathway in ABC and GCB DLBCLs (Fig. 1H).

To functionally investigate the significance of PTEN in DLBCL, we transduced ABC and GCB DLBCL cell lines with a PTEN cDNA. Our expression vector coexpressed GFP, allowing us to monitor the proportion of GFP+ vs. GFP− cells over time as a measure of toxicity of the coexpressed PTEN cDNA. Expression of PTEN was lethal to all PTEN-deficient GCB DLBCL cell lines (Fig. 3A). In contrast, PTEN-positive GCB and ABC DLBCL cell lines did not show any toxicity after PTEN introduction, despite similar expression levels (Fig. 3A and Fig. S3A). To obtain additional insights into the nature of the growth inhibitory effect of PTEN expression, we measured cell proliferation and apoptosis. To this end, we transduced BJAB, HT, and K422 cells with PTEN cDNA, and SNARF-1 staining was used to analyze cell proliferation. Cellular divisions were investigated by measuring SNARF-1 dilution in viable cells by flow cytometry. PTEN expression significantly inhibited proliferation of HT (P = 1.6 × 10−17) and K422 (P = 5.1 × 10−33) cells compared with an empty vector control (Fig. 3B). In contrast, for BJAB cells, an inhibition of proliferation was not detectable (P = 0.52). To investigate whether PTEN induces apoptosis, we determined apoptotic Annexin-V+/7AAD− cells after PTEN cDNA transduction. Interestingly, PTEN expression induced apoptosis selectively in BJAB cells, whereas an increase in apoptosis was not detectable in HT and K422 (Fig. 3C). Induction of apoptosis exclusively in BJAB cells was confirmed by an increase of cleaved caspases 3 and 9, as markers of apoptosis after PTEN reexpression (Fig. S3B).

Fig. 3.

Fig. 3.

Open in a new tab

PTEN-deficient GCB DLBCLs are addicted to PI3K/AKT signaling. (A) PTEN reexpression induces toxicity in PTEN-deficient GCB DLBCL cell lines. Representative results from at least eight independent replicates are shown. (B) SNARF-1 dilutions after reexpression of PTEN or an empty vector control are measured after 2 and 6 d. In BJAB cells, no difference in cell proliferation was detectable (P = 0.52). In contrast, HT (P = 1.6 × 10−17) and K422 (P = 5.1 × 10−33) cells show reduced proliferation after reexpression of PTEN. (C) Reexpression of PTEN induces apoptosis in BJAB but not in HT or K422 measured by an increase in Annexin V staining. Error bars indicate SD. (D) Different PTEN mutants detected in DLBCL samples show reduced expression levels compared with PTEN WT in HT cells. (E) PTEN mutants induce reduced or no toxicity in PTEN-deficient GCB DLBCL cell lines BJAB and HT compared with PTEN WT. (F) WT PTEN but not phosphatase-inactive PTEN mutants G129E and C124S inhibit PI3K/AKT signaling measured by Western blotting of p-AKT in HT cells. (G) Phosphatase-inactive PTEN mutant G129E does not induce toxicity in DLBCL cell lines. (H) Constitutive active AKT allele rescues PTEN-deficient GCB DLBCL cell lines BJAB, HT, and RL from PTEN-induced toxicity.

To investigate the functional consequences of the PTEN mutations that we identified in GCB DLBCL samples, we introduced these mutants in GCB DLBCL cell lines. Western blotting after retroviral transduction showed that PTEN mutants, with the exception of the D162H mutant detected in SUDHL-10, were expressed at significantly lower levels compared with PTEN WT, suggesting reduced stability of these mutants (Fig. 3D). This expression pattern was mirrored by reduced or lacking toxicity of the mutants in PTEN-deficient GCB DLBCL cell lines, again with the exception of the D162H aberration (Fig. 3E and Fig. S3C). These analyses indicate that the vast majority of detected mutations are loss-of-function mutations. In contrast, the D162H mutation might represent a PTEN variant.

To elucidate if addiction to PTEN loss in GCB DLBCL results from constitutive activation of the PI3K/AKT signaling pathway, we introduced previously described phosphatase-inactive PTEN mutants (G129E and C124S) (11) that cannot inhibit PI3K/AKT signaling in PTEN-deficient GCB DLBCL cell lines (Fig. 3F). These analyses revealed that PTEN-deficient GCB DLBCL cell lines were not affected by expression of these mutants, suggesting that PTEN-induced toxicity is caused by inhibition of PI3K/AKT signaling (Fig. 3G and Fig. S3D). This hypothesis was further supported by our observation that ectopic expression of a previously described constitutive active AKT allele (21) completely rescued PTEN-deficient GCB DLBCL cells from PTEN-induced toxicity (Fig. 3H).

PI3K/AKT Signaling Regulates MYC Expression in GCB DLBCL.

Next, we investigated the molecular effects of PTEN expression in the PTEN-deficient GCB DLBCL cell lines BJAB, HT, and K422. As expected, PTEN expression significantly inhibited PI3K/AKT signaling measured by p-AKT as well as downstream mechanistic target of rapamycin (mTOR) signaling measured by phosphorylation of p70S6K (Fig. 4A). To gain additional insights into which biologic processes are affected by PTEN expression, we profiled gene expression changes after PTEN introduction in HT cells. Generally, more genes were up-regulated (546 genes; P < 0.01) than down-regulated (279 genes; P < 0.01) by PTEN. For biological interpretation, we used tighter cutoffs and identified 72 genes that were significantly down-regulated (P < 0.0025) and 82 genes that were up-regulated (P < 0.00025) across all time points after PTEN induction (Fig. 4B and Table S6). PTEN induced the expression of various genes involved in general cellular processes, such as cell cycle control (CCNG2, CDKN1B, and CDKN2C), cell proliferation (ACTN4, RASAL1, and FBXO32), and B- and T-cell receptor signaling (FYN, PAG1, and ADA). To obtain a better understanding of the global gene expression changes, we performed an unbiased gene set enrichment analysis using a previously described gene expression signature database (22). This analysis revealed that the most significantly down-regulated gene set was a previously described MYC target gene signature (P ≤ 0.001; false discovery rate = 0.001) (Fig. 4C and Table S7). In addition, another independent MYC target gene set was significantly enriched with down-regulated genes, indicating that MYC activity is inhibited by PTEN induction (P ≤ 0.001; false discovery rate = 0.008) (Table S7). Because MYC mRNA levels were only marginally down-regulated after PTEN induction (P = 0.054), we investigated if MYC was modulated posttranscriptionally by PTEN expression. Indeed, MYC protein was down-regulated by PTEN reexpression in GCB DLBCL cell lines (Fig. 4D) or PI3K inhibition using the pan-PI3K inhibitor Ly294002 (Fig. S4A). To investigate if this effect was mediated by modulation of GSK3β activity and subsequent phosphorylation of MYC at Threonine 58 (T58), we performed Western blotting for phospho-GSK3β and phospo-T58-MYC in PTEN-transduced HT cells. Indeed, we detected a decrease of phospho-GSK3β and subsequent increase of phospho-T58-MYC after PTEN expression (Fig. 4E). These data indicate that PTEN expression leads to increased proteasomal degradation of MYC.

Fig. 4.

Fig. 4.

Open in a new tab

PTEN loss up-regulates MYC in GCB DLBCLs. (A) PTEN reexpression decreases phosphorylation of AKT and p70S6K. (B) Gene expression profiling after PTEN reexpression in HT cells. Changes of gene expression were profiled at the indicated time points after induction of PTEN cDNA. Gene expression changes are depicted according to the color scale shown. Genes that are involved in critical biological processes are highlighted. (C) Gene set enrichment analysis of a previously described MYC gene expression signature. The MYC signature is significantly down-regulated after PTEN reexpression. (D) PTEN reexpression down-regulates MYC protein in PTEN-deficient GCB DLBCL cell lines BJAB, HT, and K422. (E) PTEN reexpression activates GSK3β, leading to MYC phosphorylation at T58. (F) MYC expression rescues BJAB and HT cells from PTEN-induced toxicity. Representative experiments are shown. (G) shRNA-mediated MYC knockdown is toxic to PTEN-positive and -deficient GCB DLBCL cell lines. Representative results are shown. (H) PTEN-deficient GCB DLBCL cell lines are sensitive to PI3K inhibition using the pan-PI3K inhibitor Ly294002. Representative results are shown.

To evaluate the degree to which MYC knockdown contributes to PTEN-induced toxicity, we performed a rescue experiment, in which we introduced an MYC cDNA or an empty vector in PTEN-transduced BJAB and HT cells. We detected a substantial MYC-induced rescue, indicating that MYC knockdown significantly contributes to the lethal effect of PTEN expression (Fig. 4F and Fig. S4B). To confirm the critical role of MYC for survival of PTEN-deficient GCB DLBCL cell lines, we knocked down its expression using two different specific shRNAs (Fig. S4C). As expected, MYC knockdown was toxic to all PTEN-deficient cell lines (Fig. 4G and Fig. S4 D and E). However, PTEN-positive GCB DLBCL cell lines were equally affected by MYC knockdown, because they express similar MYC protein levels (Fig. 4G and Fig. S4 D and F).

Finally, we asked whether the addiction to the PI3K/AKT pathway can be used therapeutically specifically in PTEN-deficient GCB DLBCLs. To this end, we treated PTEN-positive and -deficient GCB DLBCL cell lines with the pan-PI3K inhibitor Ly294002 and determined cell viability after 4 d of incubation. Concentration of 5 μM Ly294002 significantly reduced cell viability of all three PTEN-deficient GCB DLBCL cell lines (K422, OCI-Ly1, and HT) (Fig. 4H), whereas cell viability of PTEN-positive cell lines was unaffected (OCI-Ly2, OCI-Ly7, and OCI-Ly19). These data reveal an important role of PI3K/AKT signaling for cell survival that can be blocked pharmacologically and therefore, might represent a therapeutic approach for GCB DLBCL patients with loss of PTEN.

Discussion

In the present study, we detected (in two independent primary DLBCL patient cohorts) that the tumor suppressor PTEN is differentially expressed in molecular DLBCL subtypes. We found that a majority of GCB DLBCL patient samples has lost PTEN expression, whereas this abnormality rarely occurs in other DLBCL subtypes. In GCB DLBCL, PTEN loss was inversely correlated with constitutive activation of the PI3K/AKT signaling pathway. In contrast, GCB DLBCL patient samples and cell lines with PTEN expression rarely showed activation of this pathway, indicating that loss of PTEN is the predominant mechanism of PI3K/AKT activation in GCB DLBCL. Reexpression of PTEN was lethal to PTEN-deficient GCB DLBCL cell lines and significantly inhibited PI3K/AKT signaling. These results indicate that these GCB DLBCLs are dependent on PI3K/AKT signaling and therefore, might be considered as a previously unappreciated subset of GCB DLBCL that is addicted to PI3K/AKT signaling.

Several molecular mechanisms contribute to PTEN loss in DLBCL. We detected heterozygous deletions of PTEN. Additionally, we found somatically acquired mutations of PTEN in a similar range as earlier studies (15–17). Functional analyses indicated that most of these mutations constitute loss-of-function alterations caused by reduced stability. Interestingly, all affected cell lines were derived from GCB DLBCL patient samples, supporting a role of PTEN loss in the pathogenesis of this entity. Furthermore, PTEN protein was lost in all affected cell lines, indicating additional molecular mechanisms of PTEN silencing. Likewise, PTEN protein was not detectable by immunohistochemistry in the GCB DLBCL patient that harbored a 3-bp deletion. The finding of complete PTEN loss in the majority of GCB DLBCL is remarkable. Previous work in prostate cancer indicated that, in the context of WT p53, PTEN haploinsufficiency can be more tumorigenic than complete loss of PTEN (23, 24). In contrast, in advanced malignancies frequently with loss of p53, complete PTEN inactivation is more oncogenic than PTEN haploinsufficiency. Therefore, additional work is required to differentiate oncogenic effects caused by PTEN haploinsufficiency compared with complete PTEN loss in the setting of DLBCL.

In the majority of GCB DLBCL patient samples, we could not uncover the molecular mode of PTEN silencing. In only three cell lines and one primary GCB DLBCL patient sample, we were able to uncover biallelic PTEN inactivation, whereas various cell lines and patient samples showed monoallelic silencing. It is unclear if loss of one PTEN allele is sufficient to induce PI3K/AKT activation in DLBCL, because the DB cell line that harbors a heterozygous PTEN deletion did not have detectable p-AKT levels. As GCB and ABC DLBCLs express similar levels of PTEN mRNA (P = 0.69; n = 350 patient samples) (25), a posttranscriptional or posttranslational regulation of PTEN in DLBCL seems highly likely. Various miRs, such as miR-17-92 or miR-21, have been implicated in the regulation of PTEN expression (20, 26). Similarly, the PTEN pseudogene PTENP1 has been shown to regulate PTEN levels by serving as a decoy for PTEN-targeting miRs (27). Interestingly, the PTENP1 locus is selectively lost in various malignancies. Therefore, it seems conceivable that these molecular mechanisms could contribute to PTEN silencing in GCB DLBCL.

Our work highlights a previously unappreciated molecular mechanism of MYC regulation in DLBCL. In GCB DLBCL, PTEN loss leads to MYC up-regulation by constitutive activation of the PI3K/AKT pathway. Both introduction of PTEN as well as pharmacologic inhibition of the PI3K pathway significantly down-regulated MYC expression in various PTEN-deficient GCB DLBCL cell lines. Recent work in Burkitt lymphoma showed that MYC and the PI3K pathway cooperate to counterbalance the proapoptotic properties of MYC (28, 29). In GCB DLBCL, PI3K/AKT signaling increases MYC expression by inhibition of GSK3β, a mechanism that has previously been detected in other hematologic malignancies, such as B- and T-cell acute lymphoblastic leukemia, but not DLBCL (30, 31).

We evaluated if PTEN loss and constitutive PI3K/AKT pathway activation can be exploited therapeutically in GCB DLBCL. PI3K inhibition was selectively toxic to PTEN-deficient GCB DLBCL models. In contrast, none of the PTEN-positive cell lines responded to PI3K inhibition. Our results suggest that GCB DLBCL patients could be stratified therapeutically according to their PTEN status. PTEN-negative patients are predicted to respond to inhibition of PI3K/AKT, whereas PTEN expression most likely is associated with therapy resistance. These results underscore the necessity of patient sample characterization within clinical trials using specific inhibitors to detect subgroups of patients that preferentially respond to these compounds.

Materials and Methods

Patient Samples.

Tumor biopsy specimens were obtained from patients with de novo DLBCL before treatment. All samples were studied according to protocols approved by the local institutional ethics review board of the University Hospital Basel and the Charité Universitätsmedizin Berlin, respectively.

Cell Culture, Retroviral Constructs, and Transductions.

Cell culture conditions and retroviral transductions are described in SI Materials and Methods.

Gene Expression Profiling.

The experimental setup and the analysis algorithm of gene expression profiling is described in SI Materials and Methods. The gene expression data have been deposited in the Gene Expression Omnibus database (http://www.ncbi.nlm.nih.gov/geo/; accession no. GSE45495).

Immunohistochemistry.

Previously described tissue microarrays with 249 primary DLBCLs were used in the present study (32). Cases were classified as GCB or non-GCB DLBCL applying the Tally algorithm (33). In addition, 34 primary DLBCL samples classified as ABC, GCB, or unclassified DLBCL were investigated. PTEN staining was established on PTEN-positive and -negative DLBCL cell lines, eight tonsils, and two normal lymph nodes using the primary monoclonal antibody 138G6 from Cell Signaling at a dilution of 1:100 as previously described (34). PTEN staining was established by two expert hematopathologists. Cases were evaluable when an internal positive control was present. To identify completely PTEN-negative DLBCL cases, we applied a low cutoff value of <5% positive cells. Using this cutoff, we were able to reproduce expression levels detected by Western blotting in various DLBCL cell lines.

Western Blotting.

Western blotting was performed as previously described (35), and detailed protocols are described in SI Materials and Methods.

Supplementary Material

Supporting Information
supp_110_30_12420__index.html (7KB, html)

Acknowledgments

This work was supported by a Doctoral Scholarship from the Philipps-University Marburg (to M.G.) and research grants from the German Research Foundation, the Deutsche Krebshilfe, and the Else Kröner-Fresenius-Stiftung (to G.L.).

Footnotes

The authors declare no conflict of interest.

*This Direct Submission article had a prearranged editor.

Data deposition: The data reported in the paper have been deposited in the Gene Expression Omnibus (GEO) database, www.ncbi.nlm.nih.gov/geo (accession no. GSE45495).

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1305656110/-/DCSupplemental.

References

  • 1.Nogai H, Dörken B, Lenz G. Pathogenesis of non-Hodgkin’s lymphoma. J Clin Oncol. 2011;29(14):1803–1811. doi: 10.1200/JCO.2010.33.3252. [DOI] [PubMed] [Google Scholar]
  • 2.Alizadeh AA, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature. 2000;403(6769):503–511. doi: 10.1038/35000501. [DOI] [PubMed] [Google Scholar]
  • 3.Pasqualucci L, et al. Inactivation of the PRDM1/BLIMP1 gene in diffuse large B cell lymphoma. J Exp Med. 2006;203(2):311–317. doi: 10.1084/jem.20052204. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Tam W, et al. Mutational analysis of PRDM1 indicates a tumor-suppressor role in diffuse large B-cell lymphomas. Blood. 2006;107(10):4090–4100. doi: 10.1182/blood-2005-09-3778. [DOI] [PubMed] [Google Scholar]
  • 5.Schmidlin H, et al. Spi-B inhibits human plasma cell differentiation by repressing BLIMP1 and XBP-1 expression. Blood. 2008;112(5):1804–1812. doi: 10.1182/blood-2008-01-136440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Uddin S, et al. Role of phosphatidylinositol 3′-kinase/AKT pathway in diffuse large B-cell lymphoma survival. Blood. 2006;108(13):4178–4186. doi: 10.1182/blood-2006-04-016907. [DOI] [PubMed] [Google Scholar]
  • 7.Hasselblom S, et al. High immunohistochemical expression of p-AKT predicts inferior survival in patients with diffuse large B-cell lymphoma treated with immunochemotherapy. Br J Haematol. 2010;149(4):560–568. doi: 10.1111/j.1365-2141.2010.08123.x. [DOI] [PubMed] [Google Scholar]
  • 8.Cantley LC, Neel BG. New insights into tumor suppression: PTEN suppresses tumor formation by restraining the phosphoinositide 3-kinase/AKT pathway. Proc Natl Acad Sci USA. 1999;96(8):4240–4245. doi: 10.1073/pnas.96.8.4240. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Carracedo A, Alimonti A, Pandolfi PP. PTEN level in tumor suppression: How much is too little? Cancer Res. 2011;71(3):629–633. doi: 10.1158/0008-5472.CAN-10-2488. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Shen WH, et al. Essential role for nuclear PTEN in maintaining chromosomal integrity. Cell. 2007;128(1):157–170. doi: 10.1016/j.cell.2006.11.042. [DOI] [PubMed] [Google Scholar]
  • 11.Song MS, et al. Nuclear PTEN regulates the APC-CDH1 tumor-suppressive complex in a phosphatase-independent manner. Cell. 2011;144(2):187–199. doi: 10.1016/j.cell.2010.12.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Song MS, Salmena L, Pandolfi PP. The functions and regulation of the PTEN tumour suppressor. Nat Rev Mol Cell Biol. 2012;13(5):283–296. doi: 10.1038/nrm3330. [DOI] [PubMed] [Google Scholar]
  • 13.Davis RE, et al. Chronic active B-cell-receptor signalling in diffuse large B-cell lymphoma. Nature. 2010;463(7277):88–92. doi: 10.1038/nature08638. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Lenz G, et al. Molecular subtypes of diffuse large B-cell lymphoma arise by distinct genetic pathways. Proc Natl Acad Sci USA. 2008;105(36):13520–13525. doi: 10.1073/pnas.0804295105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Sakai A, Thieblemont C, Wellmann A, Jaffe ES, Raffeld M. PTEN gene alterations in lymphoid neoplasms. Blood. 1998;92(9):3410–3415. [PubMed] [Google Scholar]
  • 16.Butler MP, Wang SI, Chaganti RS, Parsons R, Dalla-Favera R. Analysis of PTEN mutations and deletions in B-cell non-Hodgkin’s lymphomas. Genes Chromosomes Cancer. 1999;24(4):322–327. [PubMed] [Google Scholar]
  • 17.Grønbaek K, Zeuthen J, Guldberg P, Ralfkiaer E, Hou-Jensen K. Alterations of the MMAC1/PTEN gene in lymphoid malignancies. Blood. 1998;91(11):4388–4390. [PubMed] [Google Scholar]
  • 18.Georgescu MM, et al. Stabilization and productive positioning roles of the C2 domain of PTEN tumor suppressor. Cancer Res. 2000;60(24):7033–7038. [PubMed] [Google Scholar]
  • 19.O’Donnell KA, Wentzel EA, Zeller KI, Dang CV, Mendell JT. c-Myc-regulated microRNAs modulate E2F1 expression. Nature. 2005;435(7043):839–843. doi: 10.1038/nature03677. [DOI] [PubMed] [Google Scholar]
  • 20.Xiao C, et al. Lymphoproliferative disease and autoimmunity in mice with increased miR-17-92 expression in lymphocytes. Nat Immunol. 2008;9(4):405–414. doi: 10.1038/ni1575. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Kharas MG, et al. Constitutively active AKT depletes hematopoietic stem cells and induces leukemia in mice. Blood. 2010;115(7):1406–1415. doi: 10.1182/blood-2009-06-229443. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Shaffer AL, et al. A library of gene expression signatures to illuminate normal and pathological lymphoid biology. Immunol Rev. 2006;210:67–85. doi: 10.1111/j.0105-2896.2006.00373.x. [DOI] [PubMed] [Google Scholar]
  • 23.Chen Z, et al. Crucial role of p53-dependent cellular senescence in suppression of Pten-deficient tumorigenesis. Nature. 2005;436(7051):725–730. doi: 10.1038/nature03918. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Berger AH, Knudson AG, Pandolfi PP. A continuum model for tumour suppression. Nature. 2011;476(7359):163–169. doi: 10.1038/nature10275. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Lenz G, et al. Stromal gene signatures in large-B-cell lymphomas. N Engl J Med. 2008;359(22):2313–2323. doi: 10.1056/NEJMoa0802885. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Meng F, et al. MicroRNA-21 regulates expression of the PTEN tumor suppressor gene in human hepatocellular cancer. Gastroenterology. 2007;133(2):647–658. doi: 10.1053/j.gastro.2007.05.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Poliseno L, et al. A coding-independent function of gene and pseudogene mRNAs regulates tumour biology. Nature. 2010;465(7301):1033–1038. doi: 10.1038/nature09144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Sander S, et al. Synergy between PI3K signaling and MYC in Burkitt lymphomagenesis. Cancer Cell. 2012;22(2):167–179. doi: 10.1016/j.ccr.2012.06.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Schmitz R, et al. Burkitt lymphoma pathogenesis and therapeutic targets from structural and functional genomics. Nature. 2012;490(7418):116–120. doi: 10.1038/nature11378. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Malempati S, et al. Aberrant stabilization of c-Myc protein in some lymphoblastic leukemias. Leukemia. 2006;20(9):1572–1581. doi: 10.1038/sj.leu.2404317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Bonnet M, et al. Posttranscriptional deregulation of MYC via PTEN constitutes a major alternative pathway of MYC activation in T-cell acute lymphoblastic leukemia. Blood. 2011;117(24):6650–6659. doi: 10.1182/blood-2011-02-336842. [DOI] [PubMed] [Google Scholar]
  • 32.Nagel S, Hirschmann P, Dirnhofer S, Günthert U, Tzankov A. Coexpression of CD44 variant isoforms and receptor for hyaluronic acid-mediated motility (RHAMM, CD168) is an International Prognostic Index and C-MYC gene status-independent predictor of poor outcome in diffuse large B-cell lymphomas. Exp Hematol. 2010;38(1):38–45. doi: 10.1016/j.exphem.2009.10.010. [DOI] [PubMed] [Google Scholar]
  • 33.Meyer PN, et al. Immunohistochemical methods for predicting cell of origin and survival in patients with diffuse large B-cell lymphoma treated with rituximab. J Clin Oncol. 2011;29(2):200–207. doi: 10.1200/JCO.2010.30.0368. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Sangale Z, et al. A robust immunohistochemical assay for detecting PTEN expression in human tumors. Appl Immunohistochem Mol Morphol. 2011;19(2):173–183. doi: 10.1097/PAI.0b013e3181f1da13. [DOI] [PubMed] [Google Scholar]
  • 35.Wenzel SS, et al. MCL1 is deregulated in subgroups of diffuse large B-cell lymphoma. Leukemia. 2013;27(6):1381–1390. doi: 10.1038/leu.2012.367. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information
supp_110_30_12420__index.html (7KB, html)
1305656110_pnas.201305656SI.pdf (929.9KB, pdf)

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES