Inhibition of BCL11B expression leads to apoptosis of malignant but ...

Oncogene (2007) 26, 3797–3810

& 2007 Nature Publishing Group All rights reserved 0950-9232/07 $30.00 www.nature.com/onc

ORIGINAL ARTICLE

Inhibition of BCL11B expression leads to apoptosis of malignant but not normal mature T cells P Grabarczyk1, GK Przybylski1,2, M Depke3, U Vo¨lker3, J Bahr1, K Assmus1, BM Bro¨ker4, R Walther5 and CA Schmidt1 1 Clinic for Internal Medicine C, University of Greifswald, Greifswald, Germany; 2Institute of Human Genetics, Polish Academy of Sciences, Poznan, Poland; 3Laboratory for Functional Genomics, University of Greifswald, Greifswald, Germany; 4Institute for Immunology and Transfusion Medicine, University of Greifswald, Greifswald, Germany and 5Department of Medical Biochemistry and Molecular Biology, University of Greifswald, Greifswald, Germany

The B-cell chronic lymphocytic leukemia (CLL)/lymphoma 11B gene (BCL11B) encodes a Kru¨ppel-like zincfinger protein, which plays a crucial role in thymopoiesis and has been associated with hematopoietic malignancies. It was hypothesized that BCL11B may act as a tumorsuppressor gene, but its precise function has not yet been elucidated. Here, we demonstrate that the survival of human T-cell leukemia and lymphoma cell lines is critically dependent on Bcl11b. Suppression of Bcl11b by RNA interference selectively induced apoptosis in transformed T cells whereas normal mature T cells remained unaffected. The apoptosis was effected by simultaneous activation of death receptor-mediated and intrinsic apoptotic pathways, most likely as a result of tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) upregulation and suppression of the Bcl-xL antiapoptotic protein. Our data indicate an antiapoptotic function of Bcl11b. The resistance of normal mature T lymphocytes to Bcl11b suppression-induced apoptosis and restricted expression pattern make it an attractive therapeutic target in T-cell malignancies. Oncogene (2007) 26, 3797–3810. doi:10.1038/sj.onc.1210152; published online 18 December 2006 Keywords: BCL11B; T-ALL; apoptosis; TRAIL; Bcl-xL

Introduction The B-cell chronic lymphocytic leukemia (CLL)/lymphoma 11B (BCL11B/CTIP2/RIT1/hRita) gene, encodes a Kru¨ppel-like C2H2 zinc-fingers protein that has been initially identified as a transcriptional repressor. It has been shown to mediate its repressor function either directly via binding to a GC-rich consensus Correspondence: Professor CA Schmidt, KIM C, Molecular Haematology, Universita¨t Greifswald, Sauerbruchstr., Greifswald 17475, Germany. E-mail: [email protected] Received 1 February 2006; revised 29 September 2006; accepted 23 October 2006; published online 18 December 2006

sequence (Avram et al., 2002), or via interacting with chicken ovalbumin upstream promoter-transcription factor nuclear receptors (Avram et al., 2000), binding with SIRT1 histone deacetylase (HDAC) (Avram et al., 2002; Senawong et al., 2003, 2005) and nucleosome remodeling and HDAC complex (NuRD), one of the major transcriptional co-repressor complexes in mammalian cells (Cismasiu et al., 2005; Topark-Ngarm et al., 2006). Recently, the first evidence has been provided, that Bcl11b activates transcription of the interleukin 2 gene in stimulated CD4 þ T lymphocytes (Cismasiu et al., 2006). However, the precise function and endogenous transcriptional targets of this transcription factor still remain poorly characterized. In vivo, BCL11B was shown to be expressed exclusively in the central nervous system and immune system (Leid et al., 2004). Lack of BCL11B, as shown in germline knockout mice, leads to abnormal thymopoiesis and the complete absence of mature ab T cells (Wakabayashi et al., 2003b). Specifically, the transition from the double-negative stage 3 (DN3) to DN4 stage was blocked, the rearrangement of T-cell receptor b locus (TRB) was incomplete and extensive apoptosis was observed. Decreased levels of Bcl-xL and Bcl-2 antiapoptotic proteins in sorted DN4 thymocytes obtained from BCL11B/ mice were suggested to contribute to the observed block of thymocyte development (Inoue et al., 2006). BCL11B has been also associated with lymphoproliferative disorders. The translocation t(14:14)(q11.2; q32.31) or the inversion inv(14)(q11.2q32.31) disrupting the BCL11B locus were found in one adult T-cell leukemia/lymphoma (ATLL) case and in two cases of Tcell acute lymphoblastic leukemias (T-ALL), respectively (Gesk et al., 2003; Przybylski et al., 2005). Further translocations involving the BCL11B locus were reported in three T-ALL-derived cell lines and in one case of acute myelocytic leukemia (MacLeod et al., 2003; Nagel et al., 2003; Bezrookove et al., 2004). In mice, deletions within BCL11B were found in normal thymocytes and in irradiation-induced lymphomas (Wakabayashi et al., 2003a; Sakata et al., 2004). We and others reported high levels of BCL11B mRNA expression in the majority of T-ALL and T-cell

Inhibition of BCL11B leads to apoptosis P Grabarczyk et al

3798 BCL11B mRNA/1x105 β-2-MG

a

BCL11B mRNA levels

1500

674 1000 *

500

**

0 Jurkat

b nt

mock

huT-78 48h 674 sc

72h sc 674

Bcl11b β-actin

huT78

Bcl11b β-actin

Results BCL11B-specific siRNAs suppress BCL11B expression in Jurkat and huT78 human T-cell leukemia/lymphoma cells To determine the efficiency of BCL11B inhibition after small interfering RNA (siRNA) (BCL11B-674) treatment, BCL11B expression was analysed on the mRNA and protein level. An approximately three to five-fold reduction in mRNA levels was observed 24 h after nucleofection in both cell lines, compared to scrambled non-silencing siRNA (BCL11B-sc) (Figure 1a). On the protein level, the Bcl11b knock-down effect reached approximately 80% after 48 h and the suppression persisted 72 h post-transfection (Figure 1b). In order to rule out possible off-target effects of the BCL11B-674 duplex, a different Jurkat-tetracycline-inducible BCL11B-RNA interference system was developed. In this system, short hairpin RNA (shRNA-1415), targeting a different BCL11B mRNA region, was expressed from pSuperior-pgk-Neor-enhanced green fluorescence protein (EGFP) vector after doxycycline induction. Again, Bcl11b protein levels were reduced by 90% upon induction (Figure 1c). BCL11B suppression induces apoptosis in Jurkat and huT78T cell lines The biological consequences of BCL11B silencing were further studied in T-ALL (Jurkat), T-cell-lymphoma (huT78)-derived cell lines and the BCL11B-negative Raji Burkitt lymphoma cell line as control. The percentage of apoptotic cells was determined by an Annexin-V-fluorescein isothiocyanate (FITC)-binding assay in all three cell lines 48 and 72 h after siRNA treatment. Jurkat and huT78 cells showed a large increase in Annexin-V-positive cells after BCL11B suppression, whereas Raji cells remained unaffected. The effect was detectable 48 h after transfection (not shown), reaching 50 and 80% after 72 h in Jurkat and huT78 cell lines, respectively (Figure 2a). Control siRNA-treated, untreated or mock-transfected cells showed 5–10% apoptotic cells (not shown). The proapoptotic effect of BCL11B knock down was further confirmed by the determination of the active form of Oncogene

sc

Jurkat

leukemia/lymphoma cell lines (Nagel et al., 2003; Przybylski et al., 2005). Furthermore, BCL11B overexpression was found in the acute type of ATLL in contrast to no or low expression of the gene in the lymphoma type (Oshiro et al., 2006). Thus, although the mechanisms are not clear, Bcl11b seems to be implicated in the process of oncogenesis. Here, we report the induction of apoptosis in Jurkat and huT78 human T-cell lymphoma cell lines upon Bcl11b suppression by RNA interference, whereas peripheral blood T cells are not affected. Our findings do not support the hypothesis of BCL11B being a tumor suppressor, but indicate an antiapoptotic activity. Therefore, downregulation of Bcl11b could represent a new therapeutic option in T-cell malignancies.

c

-dox 1415 sc

+dox 1415 sc

Bcl11b β-actin Figure 1 Inhibition of Bcl11b in Jurkat and huT78 cells by RNA interference. (a) Suppression of BCL11B mRNA measured by QRT–PCR 24 h upon transient nucleofection of BCL11B-specific 674 duplex compared to non-silencing scrambled control siRNAtreated cells, b-2-MG: b-2-microglobulin (reference gene). The results represent mean values of three independent experiments7s.d. *Po0.01; **Po0.001. (b) Bcl11b protein levels in Jurkat and huT78 cells 48 and 72 h after BCL11B-674 siRNA nucleofection (5 mg). Non-treated (nt), mock-transfected (mock) and non-silencing scrambled siRNA (sc)-treated cells were used as controls. (c) Conditional Bcl11b knock down in Jurkat cells. BCL11B-specific shRNA (1415) and non-silencing scrambled control shRNA (sc) expression was induced with 0.5 mg/ml doxycycline for 96 h. b-actin served as loading control. The results shown are representative of three experiments.

caspase 3 (Figure 2a) and DNA degradation visualized by staining with propidium iodide (Figure 2a). Again, BCL11B-siRNA did not influence Raji cells, whereas in Jurkat and huT78, fractions of sub-diploid cells and those of cells positive for active caspase 3 correlated well with Annexin-V-binding measurements. In the conditional shRNA knock-down system, Jurkat cells showed 60% Annexin-V-positive cells 96 h after induction of BCL11B-specific short hairpin RNA, whereas in the control expressing a non-silencing shRNA species, more than 95% of cells remained viable (Figure 2b). No significant loss of viability was observed in untreated cells. The caspase 3 activity measured at the same time point was over six times higher in Bcl11bdeficient cells than in controls (Figure 2b). Even in the absence of the inducer, the caspase 3 activity was slightly increased in cells harboring the functional shRNA vector, most likely as a result of leaky shRNA transcription. We next examined the dose dependence of BCL11BsiRNA-induced apoptosis in huT78 cells. The percentages of Annexin-positive cells correlated well with the


3799

amount of BCL11B-674 siRNA used and the BCL11B knock-down efficacy. As expected, no influence on viability was observed in BCL11B-sc treated cells (Figure 2c). In order to analyse the upstream cascades leading to activation of the effector caspase 3, huT78 cells were treated with specific caspase inhibitors after BCL11B suppression, and apoptosis was measured after 48 and

a

-dox

60 30

Count

60%

90

Count

90

Count

+dox

120 60

RLU

120

Count

Count

Count

250 200 150 100 50 0

54%

Count

Raji Count

Jurkat Count huT78 Count

250 200 150 100 50 0

b

Count

Active Caspase 3 Propidium Iodide (DNA) 300 400 53% 50% 250 300 200 150 200 100 100 50 0 0 1 2 3 1 2 3 1 200 400 600 10 10 10 10 10 10 Annexin V Active Caspase 3 Propidium Iodide (DNA) 300 200 80% 85% 80% 250 150 200 150 100 100 50 50 0 0 1 2 3 1 2 3 1 200 400 600 800 1000 10 10 10 10 10 10 Propidium Active Caspase 3 Annexin V Iodide (DNA) 300 400 250 300 200 200 150 100 100 50 0 1 2 3 1 2 3 1 200 400 600 800 10 10 10 10 10 10 Active Caspase 3 Propidium Annexin V Iodide (DNA) siRNA 674 Control siRNA

Annexin V 250 200 150 100 50 0

30

0

0

Annexin V-APC

c Count

300

Casapase 3/7 Activity * 1415 sc

25000 20000 15000 10000 5000 0

+dox

Annexin V-APC shRNA 1415 Control shRNA

15%

300

28%

300

-dox

300

36%

200

200

200

200

100

100

100

100

0

0

0.5 µg

0

1 µg

50%

0

2 µg

5 µg

Annexin V-FITC Control siRNA siRNA 674 0.5µg 1µg 2µg 5µg

sc

Bcl11b β-actin

d

48h

72h

80 *#

60 40

*#

20

sc-negCi

sc-DMSO

sc-C9i

sc-NT

sc-C8i

674-DMSO

674-negCi

674-C9i

674-NT

0 674-C8i

% Annexin V-positive cells

100

72 h. The caspase 8-specific inhibitor benzyloxycarbonyl-Ile-Glu(OMe)-Thr-Asp(OMe)-fluoromethylketone (IETD) targeting the major regulatory caspase downstream of the death receptor pathway, reduced the percentage of apoptotic cells by approximately 40% at 48 h and 65% at 72 h (Figure 2d). The specific caspase 9 inhibitor benzyloxycarbonyl-Leu-Glu(OMe)-His-Asp(OMe)-fluoromethylketone (LEHD), blocking the intrinsic pathway, reduced Annexin-V-positive cells by approximately 30% at both time points. Vehicle as well as a non-blocking peptide Z-FA controls had no effect on viability. This confirmed the involvement of caspases in BCL11B siRNA-induced apoptosis. BCL11B-siRNA induces TRAIL expression and suppresses BCLxL To elucidate the molecular mechanisms accounting for BCL11B-siRNA-mediated cell death, global gene expression profiling was performed using the Affymetrix HG U133 Plus 2.0 GeneChip. Jurkat cells were transiently nucleofected with BCL11B-674 and BCL11B-sc siRNAs. Total RNA was isolated 48 h post-transfection, when approximately 80% suppression of Bcl11b protein was detected (Figure 1b) but ratio of apoptotic to viable cells was low (Figure 2a). One hundred and ninety-one probe sets were found to be at least twofold regulated upon Bcl11b suppression. A set of 15 genes was measured by quantitative reverse transcription–polymerase chain reaction (QRT–PCR) to assess the accuracy of the array data from which 14 were confirmed (Supplementary data). Two directly apoptosis-related genes were studied in detail: tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) and BCLxL. Array data exhibited fourfold upregulation of the TRAIL mRNA after BCL11B siRNA treatment providing further evidence for the involvement of the Figure 2 Induction of apoptosis by Bcl11b suppression. (a) Jurkat, huT78 and Raji cells were nucleofected with BCL11B674 (filled histograms) and non-silencing scrambled control siRNA (black lines) duplexes. Annexin-V-binding, activation of caspase 3 and DNA content (propidium iodide) were measured 72 h after siRNA treatment. (b) Apoptosis after shRNA-mediated Bcl11b suppression in Jurkat cells. Annexin-V-APC measured before and 96 h after induction of BCL11B-specific (1415) or of non-silencing (sc) shRNAs with 0.5 mg/ml doxycycline. Caspase 3/7 activity (RLU – relative luminescence units). The results represent mean values of three independent experiments7s.d. *Po0.001. (c) Dosedependent effect of Bcl11b knock down in huT78 cells. Cells were nucleofected with increasing doses of BCL11B-674 (filled histograms) and scrambled control (sc; black lines) duplexes. Apoptosis was measured by Annexin-V-binding assay 48 h after treatment. Bcl11b levels were analysed by Western blotting. Equal protein load was confirmed by b-actin detection. (d) Suppression by caspase inhibitors of apoptosis induced by Bcl11b depletion. HuT78 cells were nucleofected with BCL11B-674 or with sc duplexes and either left untreated or cultured in the presence 50 mM of caspase 8 (C8i), caspase 9 (C9i) and negative control (negCi) inhibitors. DMSO in dilution corresponding to the amount of inhibitors was used as a control. Apoptosis was analysed by Annexin-V-binding 48 and 72 h after treatment. The results represent averaged values of three independent experiments7s.d. *Po0.05; #Po0.001. Oncogene


150 100

BCLxL mRNA/1x105 β-2-MG

a


3800 BCLxL mRNA levels *

**

**

50 0 24h

48h

72h

huT78-674

BCLxL mRNA levels

150

*

100

0 24h

48h

Jurkat-674

b mock

674

48h sc

huT78

Bcl-xL

-dox 1415

sc

+dox 1415

sc Jurkat

Bcl-xL

TRAIL mRNA levels *

Count

*

200 100 0 24h

48h

huT78-674

d

TRAIL mRNA/1x103 β-2-MG


β-actin

300

250 200 150 100 50 0 0 10

0.5 µg

250 200 150 100 50 0 3 10 100

27%

101 102 TRAIL

72h Jurkat-sc

72h 674 sc

β-actin

c

*

50

hut78-sc

nt

**

72h

TRAIL mRNA levels

300 *

**

100 0 24h

hut78-sc

48h

Jurkat-674

1.0 µg

250 200 150 100 50 0 3 10 100

41%

101 102 TRAIL

siRNA674

**

200

2.0 µg

250 200 150 100 50 0 3 10 100

51%

101 102 TRAIL

72h Jurkat-sc 5.0 µg 68%

101 102 TRAIL

103

Control siRNA

Figure 3 Expression of apoptosis-related genes after BCL11B inhibition. (a) Decreased BCL-xL mRNA expression measured by QRT–PCR upon transient suppression of Bcl11b in Jurkat and huT78 cells compared to sc-RNA duplex-treated cells. The results represent mean values of three experiments7s.d. *Po0.05; **Po0.001. (b) Suppression of Bcl-xL protein resulting from transient (huT78) and induced (Jurkat) Bcl11b knock down. 674: BCL11B-specific siRNA, sc: non-silencing scrambled control siRNA/shRNA, nt: non-treated cells, mock: mock nucleofected, 1415: BCL11B-specific shRNA, dox: cells before induction of shRNA expression, þ dox: cells induced with 0.5 mg/ml doxycycline for 96 h. Equal protein load was confirmed by b-actin staining. (c) Induction of TRAIL expression measured by QRT–PCR upon transient suppression of Bcl11b in Jurkat and huT78 cells compared to sc-RNA-duplex treated cells. The results represent mean values of three experiments7s.d. *Po0.05; **Po0.01. (d) Dose-dependent effect of Bcl11b knock down on the expression of Trail in huT78 cell line. Cells were nucleofected with increasing doses of BCL11B-674 (filled histograms) and sc (black lines) duplexes. Membrane-bound Trail was measured by FACS 48 h post-transfection.

death-receptor-mediated apoptosis pathway. Further, the antiapoptotic BCLxL gene’s levels decreased approximately threefold. For both genes, the microarray data were subsequently verified by TRAIL- and BCLxLspecific QRT–PCR assays in three independent experiments performed with both T-cell lines (Figure 3a and c). The results were confirmed in the tetracyclineinducible knock-down system (not shown). Next, proteins levels of both genes were investigated. Western blot analysis of Bcl-xL in the huT78 cell line revealed significant downregulation of the protein after transient suppression of Bcl11b. Similar results were Oncogene

obtained in Jurkat cells with the conditional shRNAmediated Bcl11b knock-down system (Figure 3b). Cell-surface-bound Trail was measured by fluorescence-activated cell sorting (FACS) analysis in huT78 cells. The percentages of Trail-positive cells correlated very well with the amount of BCL11B-674 duplexes employed (Figure 3d), the number of apoptotic cells, and with Bcl11b knock-down efficiency (Figure 2c). To assess the individual contribution of Bcl-xL to the pro-apoptotic effects of BCL11B downregulation, a plasmid expression vector encoding Bcl-xL followed by internal ribosome entry site (IRES) and an EGFP


3801

reporter gene was constructed and co-transfected with BCL11B-specific or control siRNA into Jurkat cells. The ‘empty’ plasmid containing the IRES-EGFP elements was used as vector control. As shown in Figure 4a, co-transfection of the reporter constructs did not influence the efficiency of Bcl11b suppression, but restored Bcl-xL levels in Bcl11b-deficient cells to baseline. In Bcl11b-deficient cells, apoptosis – quantified in the EGFP-positive sub-population 48 h after treatment was reduced to 10% when the BCLxL vector was applied, compared to over 20% observed in the vector

sc+BCLxL

674+CV

a

sc+CV

674+BCLxL

control (Figure 4b). However, the protective Bcl-xL effect was transient. After 72 h transfection, apoptosis was similar to that in Jurkat cells treated with siRNA and the vector control probably due to the transient nature of BCL-xL expression. The decreased number and mean fluorescence of EGFP-positive cells support this hypothesis. To clarify this issue, Jurkat and huT78 cells overexpressing Bcl-xL in a stable manner were developed. As shown in Figure 4c, stable overexpression of Bcl-xL almost completely rescued the cells from apoptosis upon Bcl11b suppression.

Bcl11b Bcl-xL β - actin

b

48h

COUNT

100

674 22% 10%

75

100 75

72h sc 5% 5%

674

100

50% 47%

75

100 75

50

50

50

50

25

25

25

25

sc 5% 5%

COUNT

0 0 0 0 100 101 102 103 100 101 102 103 100 101 102 103 100 101 102 103 Annexin V-APC Annexin V-APC Annexin V-APC Annexin V-APC 100

100

100

100

75

75

75

75

50

50

50

50

25

25

25

25

0 0 0 0 100 101 102 103 100 101 102 103 100 101 102 103 100 101 102 103 EGFP EGFP EGFP EGFP BCL-xL-IRES-EGFP

c

Jurkat BCL-xL mock

75

Hut-78 BCL-xL mock

β-actin

β-actin

Bcl-xL

Bcl-xL 50% 7%

100 75

50

50

25

25

5% 5%

0 0 100 101 102 103 100 101 102 103 Annexin V-APC Annexin V-APC BCL-xL

100 COUNT

100 COUNT

IRES-EGFP

75

80% 16%

100 75

50

50

25

25

5% 5%

0 0 100 101 102 103 100 101 102 103 Annexin V-APC Annexin V-APC Empty vector

Figure 4 Contribution of BCLxL to apoptosis mediated by BCL11B suppression (a) Bcl11b, Bcl-xL and b-actin Western blot analysis 48 h after co-transfection of BCL11B-specific siRNA with pBCL-xL-IRES-EGFP expression vector into Jurkat cells. 5 mg of siRNA and 5 mg of plasmid DNA were used for each experiment. 674, BCL11B-specific siRNA; sc, scrambled control siRNA; cv, control vector (pIRES2-EGFP); BCLxL, pBCLxL-IRES-EGFP vector. (b) Apoptosis measured in EGFP-positive cells 48 and 72 h after cotransfection of BCL11B-specific siRNA with pBCL-xL-IRES-EGFP expression vector into Jurkat cells (c) Apoptosis (lower panel) in Jurkat and huT78 cells stably overexpressing Bcl-xL (upper panel) measured 72 h after BCL11B knock down. Oncogene


3802

In order to test the contribution of increased TRAIL expression to apoptosis following Bcl11b suppression, we attempted to inactivate the cytokine with Trailneutralizing antibody. First, the neutralizing properties of the antibody were confirmed by simultaneous exposure of huT78 cells to increasing concentrations of human recombinant Trail and 50 ng/ml of the antibody (Figure 5a). When Bcl11b-deficient huT78 cells were cultured in the presence of the antibody, the fraction of Annexin-V-positive cells dropped dramatically compared to the untreated cells. As expected, the antibody neither influenced TRAIL induction following BCL11B knock down nor reduced the basal rate of apoptosis in cells with normal BCL11B expression (Figure 5b). The contribution of TRAIL was further analysed by simultaneous double knock down of BCL11B and TRAIL in huT78 cells, abrogating the increase of TRAIL expression, which usually followed BCL11B reduction. As shown in Figure 5c, the reduction of BCL11B mRNA was unaffected by the simultaneous nucleofection of a second siRNA. However, the increase of TRAIL transcription following Bcl11b suppression was abolished and TRAIL mRNA amounts reverted to basal levels upon co-transfection of TRAIL-specific siRNAs. In the latter case, the percentage of apoptotic cells was reduced twofold 48 h after targeting TRAIL in addition to BCL11B (Figure 5d). TRAIL suppression provided only temporary protection and the effects of TRAIL co-inhibition vanished after 72 h (data not shown). BCL11B suppression does not induce apoptosis of peripheral blood T lymphocytes We and others reported, previously, increased BCL11B transcription in acute T-cell leukemia cell lines and primary leukemic T cells in comparison to normal mature T cells. However, Bcl11b protein levels in Jurkat, huT78 and in peripheral blood T cells isolated from three different healthy individuals were similar (Figure 6a). To test effects of BCL11B suppression in nontransformed peripheral blood T cells, knock-down experiments were performed. Purified T cells obtained from three healthy individuals were stimulated for proliferation in vitro before transient BCL11B knock down. QRT–PCR and Western blot analysis confirmed the knock-down efficiency (Figure 6b). Apoptosis measured 72 h after transfection by Annexin-V-binding assay revealed low levels of spontaneous cell death and slight differences between T cells obtained from different donors, but no differences between Bcl11b-deficient and control cells were detected (Figure 6c). A similar experiment was performed with freshly isolated, purified, non-stimulated and non-proliferating T cells. Once again, no increase of apoptotic cells was seen in Bcl11bdeficient cells (not shown). These findings prompted us to examine the status of the apoptosis-related genes whose expression was changed after BCL11B knock down in Jurkat and huT78 cells. In the in vitro expanded T lymphocytes, Oncogene

the levels of BCL-xL mRNA and protein were significantly reduced in BCL11B siRNA-treated cells (Figure 7a and b). TRAIL mRNA was induced significantly in mature T cells but to a lesser extent than in the transformed T-cell lines, not exceeding a twofold upregulation in Bcl11b-deficient cells (Figure 7a). Although our FACS-based assay did not show any induction of surface-bound Trail following inhibition of BCL11B (Figure 7c), intracellular Trail measured by enzyme-linked immunosorbent assay (ELISA) was approximately doubled after Bcl11b inhibition compared to control siRNA-treated cells (Figure 7d). In summary, whereas it reduced Bcl-xL expression and slightly increased Trail, BCL11B knock down did not influence the survival of normal, mature T cells, regardless of their proliferation or activation status. Mature normal T cells are known to express Bcl-2 antiapoptotic protein. This may substitute for decreased levels of Bcl-xL after BCL11b knock down and make lymphocytes insensitive to Bcl11b suppression. To clarify this issue, we inhibited both genes either alone or simultaneously and showed significant downregulation of Bcl-xL and Bcl-2 expression (Figure 8a). Apoptosis measured 72 h after treatment confirmed, that downregulation of Bcl-xL via BCL11B knock down does not influence cells’ viability. Inhibition of Bcl-2 alone slightly, but significantly decreased cell survival, whereas simultaneous suppression of both genes increased the number of apoptotic cells twofold (Figure 8b). Treatment of the cells with the apoptosisinducing agent etoposide showed in turn, that only T cells deficient for both antiapoptotic genes were slightly more prone to undergo apoptosis (Figure 8c). These results support our hypothesis concerning compensatory role of Bcl-2 in BCL11B-deficient normal T lymphocytes, but also indicate, that lack of Bcl11b does not substantially alter their sensitivity to DNA damageinduced cell death.

Discussion Although much progress has been made in elucidating the mechanisms of BCL11B-mediated transcriptional regulation in vitro and several different potential modes of action have been proposed, the course of events in vivo remains controversial. The lack of BCL11B expression in knockout mice completely blocks the differentiation of ab T lymphocytes (Wakabayashi et al., 2003b). The authors hypothesized that this could be due to the pronounced sensitivity of thymocytes to apoptotic signals, a lack of pre-T-cell receptor (TCR) signaling, or a functional impairment of parallel pro-survival pathways. These findings emphasize the essential role of BCL11B in thymopoiesis. Most researchers currently favor the idea that BCL11B mainly functions as a tumor-suppressor gene, protecting the cells from uncontrolled cell proliferation and the accompanying massive apoptosis. Support for this hypothesis was


3803 200

8% 9%

200

13% 40% 150

100

100

100

100

50

50

50

50

0

Annexin V - FITC

b

0

Annexin V - FITC

27% 57%

200

150

150 Count

200

0

43% 60%

150

0

Annexin V - FITC

Annexin V - FITC

-hrTRAIL, +Ab

100ng/ml hrTRAIL, +Ab



-hrTRAIL, -Ab

100ng/ml hrTRAIL, -Ab



Count

100 80 60 40 20 0 100

674 +TRAIL Ab - 30% -TRAIL Ab - 78% 674, +Ab 674, -Ab

101 102 103 Annexin V

Count

a

104

TRAIL mRNAl evels

sc +TRAIL Ab - 9,7% -TRAIL Ab - 8,6% SC, +Ab SC, -Ab

101 102 103 Annexin V

104

200


160 140 120 100 80 60 40 20 0

100 80 60 40 20 0 100

SC, +Ab

Count

150

SC, -Ab

100 50

674, +Ab 674, -Ab

0 100

1

10 10 TRAIL -PE

674 +Ab 674 -Ab sc +Ab sc -Ab

300 250 200 150 100 50 0

BCL11b mRNA levels *

d

*

200 150

Count

350 300 250 200 150 100 50 0

3

10

20% 40%

100 50 0 100

674 + sc + 674 + sc + siTRAIL siTRAIL scTRAIL scTRAIL TRAIL mRNA levels *

200

*

150 Count


BCL11b mRNA/1x105 β-2-MG

c

2

Annexin V - FITC 674, +siTRAIL 674, +scTRAIL 5% 8%

100 50 0 100

674 + 674 + sc + sc + siTRAIL scTRAIL siTRAIL scTRAIL

Annexin V - FITC SC, +siTRAIL SC, +scTRAIL

Figure 5 Role of TRAIL in apoptosis mediated by BCL11B suppression. (a) Neutralization of human recombinant Trail (hrTRAIL) by Trail-neutralizing antibody (TRAIL Ab). Annexin-V-FITC measured 24 h after administration of increasing amounts of soluble recombinant cytokine and 50 ng/ml of Trail Ab. (b) Protective effect of Trail inhibition. HuT78 cells were nucleofetced with BCL11B674 and sc-siRNA and either left untreated (black lines, TRAIL Ab) or cultured in the presence of 25 ng/ml of Trail-neutralizing antibody ( þ TRAIL Ab). Apoptosis was analysed by Annexin-V-binding assay 72 h after nucleofection. TRAIL mRNA levels measures 48 h after Bcl11b suppression. Membrane bound Trail analysed by FACS 72 h after siRNA transfection. 674, BCL11Bspecific siRNA; sc, scrambled control siRNA; Ab, Trail-neutralizing antibody. The illustrated data are representative of three independent experiments. (c) The relative BCL11B and TRAIL mRNA levels analysed 24 h after simultaneous double knock down of both genes in huT78 cells. 674, BCL11B-specific siRNA; sc, scrambled control siRNA; siTRAIL, TRAIL-specific siRNAs; scTRAIL, corresponding scrambled control siRNAs. *Po0.001. (d) Protective effect of TRAIL suppression. Annexin-V measured 48 h after simultaneous double knock down of BCL11B and TRAIL. 674, BCL11B-specific siRNA; sc, scrambled control siRNA; siTRAIL, TRAIL specific siRNAs; scTRAIL, corresponding scrambled control siRNAs. The illustrated data are representative of three independent experiments.

Oncogene


Donor 3

*

Donor 2

*

800

Donor 1

BCL11B mRNA levels huT78

1200

Jurkat


a

MW marker

3804

Bcl11b

400

β-actin

0 Jurkat


b

huT78

T cells BCL11B mRNA levels

80

*

60 40 20 0

674

sc 48h

nt

mock

674

72h sc

674

sc

Bcl11b β-actin

48h Count

c

250 200 150 100 50 0 100

Donor 1 250 13% 200

101

102

150 100 50 0 103 100

72h Count

Annexin V 250 200 150 100 50 0 100

102

101

102

150 100 50 0 103 100

Annexin V

Donor 1 250 13% 200

101

Donor 2 250 15% 200

150 100 50 0 3 10 100

Annexin V

101

102

siRNA 674

101

102

103

Annexin V

Donor 2 250 15% 200

Annexin V

Donor 3 9%

150 100 50 0 3 10 100

Donor 3 10%

101

102

103

Annexin V Control siRNA

Figure 6 Suppression of Bcl11b in normal peripheral T cells. (a) Comparison of BCL11B mRNA and corresponding protein in Jurkat, huT78 and normal in vitro cultured T cells obtained from healthy individuals. BCL11B mRNA measured by QRT–PCR, protein assayed by Western Blot. *Po0.001. (b) Suppression of BCL11B mRNA measured by QRT–PCR 24 h upon transient nucleofection of BCL11B-674 duplex compared to non-silencing scrambled control siRNA treated cells. *Po0.05 Bcl11b protein levels in T cells 48 and 72 h after BCL11B-674 siRNA nucleofection (5 mg). Non-treated (nt), mock transfected (mock) and non-silencing scrambled control siRNA (sc) treated cells were used as controls, b-actin served as protein loading control. (c) Lack of apoptosis induction after Bcl11b suppression in T cells. Normal T cells from three healthy individuals were nucleofected with BCL11B-674 (filled histograms) and non-silencing scrambled control siRNA (black lines) duplexes. Annexin-V-binding was determined 48 and 72 h after siRNA treatment. These results are representative of at least three independent experiments.

provided by the same group, who demonstrated deletions in the BCL11B locus in g-irradiation-induced lymphomas in mice (Wakabayashi et al., 2003a). The deletions were later found in normal thymocytes and occurred with same frequency in g-irradiated and nonirradiated mice, indicating rather an indirect role of BCL11B mutations in lymphomagenesis (Sakata et al., 2004) Interestingly, the deletions occurred with much higher frequencies in p53-proficient than in p53-deficient lymphomas, which suggested that loss of BCL11B might contribute to oncogenesis only in p53-proficient lymphocytes or was not required any more in p53-null lymphomas. Three possibilities were considered for the Oncogene

relationship between these two proteins: (i) BCL11B functions on the p53-signaling pathway by regulating its expression or being regulated by p53; (ii) BCL11B interacts directly with p53; (iii) both proteins have a similar function but act independently. Whereas the first option is improbable (no changes of BCL11B levels in p53-null mice and p53 expression in BCL11B knockout animals were observed), the latter ones have not been verified yet. Further, contribution of p53 to the effects described here can be excluded, as T-cell lines used are known to carry inactivating p53 mutations. In keeping with the idea that Bcl11b might act as a tumor suppressor, we recently reported a T-ALL case


3805 250 200

BCLxL mRNA levels 674

sc



a

*

150 100 50 0 24h

48h

72h

b

TRAIL mRNA levels

500

674

400

mock

*

48h

72h

300 200 100 0 24h

48h nt

*

sc

72h

674

sc

674

sc

Bcl-xL β-actin

48h Count

c

250 200 150 100 50 0 100

Donor 1

101

102

250 200 150 100 50 0 3 10 100

72h Count

TRAIL 250 200 150 100 50 0 100

101

102

102

250 200 150 100 50 0 103 100

TRAIL

Donor 2

101

102

Donor 3

101

102

103

TRAIL 250 200 150 100 50 0 103 100

TRAIL

siRNA 674

Donor 3

101

102

103

TRAIL

Control siRNA

Intracellular Trail

300 Trail [pg/ml]

250 200 150 100 50 0 3 10 100

TRAIL

Donor 1

101

Donor 2

**

**

*

200 100 0 Jurkat

huT78 674

T cells sc

Figure 7 Expression of apoptosis related genes after BCL11B inhibition in normal T lymphocytes. (a) BCL-xL and TRAIL mRNA changes measured by QRT–PCR upon transient suppression of Bcl11b in normal T cells compared to scrambled control RNA duplex treated cells. The results represent mean values of three experiments7s.d. *Po0.05. (b) Suppression of Bcl-xL protein resulting from transient Bcl11b knock down in normal T cells. 674, BCL11B-specific siRNA; sc, non-silencing scrambled control siRNA; nt, nontreated cells; mock, mock nucleofected cells. Equal protein load was confirmed by the staining of b-actin. (c) Lack of membrane-bound Trail after Bcl11b inhibition in normal T cells. Cells were nucleofected with 5 mg of BCL11B-674 (filled histograms) and sc (black lines) duplexes. Membrane-bound Trail was measured by FACS 48 and 72 h post-transfection. The illustrated results are representative of three experiments. (d) Levels of intracellular Trail protein measured by ELISA assay in the cell lysates from Jurkat, huT78 and T cells from three different healthy individuals after Bcl11b inhibition. *Po0.05, **Po0.01.

carrying an inversion involving the BCL11B locus, which abolished the expression of the wild-type gene (Przybylski et al., 2005). However, our data indicate that this is a rare event in T-ALL. In the vast majority of TALL cases and T-cell lines, BCL11B is expressed. Taken together, these observations cast some doubt on the tumor-suppressor hypothesis. To clarify the role of BCL11B in T-cell malignancies, we used RNA interference to inhibit its function in two different human T-cell leukemia and lymphoma-derived cell lines. Our results provide evidence for a major role of BCL11B in promoting the survival of these malignant

cells. Moreover, we have at least partially delineated the molecular mechanisms of cell death caused by Bcl11b deficiency. Two different RNA interference strategies targeting different regions of the BCL11B sequence gave congruent results. Reduction of BCL11B expression resulted in dramatically increased apoptosis in both T-cell leukemia and lymphoma cell lines. This is consistent with the findings of Wakabayashi et al. (2003b), who detected elevated apoptosis in the thymuses of neonatal BCL11B knockout mice. The pro-apoptotic effect of BCL11B deficiency was pronounced and led to the Oncogene


3806

a sc

bcl2-si+ 674 bcl2-si 674

Bcl11b

400 300 200

* *

100

*

0 BCL-xL

BCL-2

bcl2-si+674-si

bcl2-si

*

+ ETOPOSIDE *

bcl2-si+674-si

*

60 50 40 30 20 10 0

bcl2-si+674-si

bcl2-si

bcl2-si

-ETOPOSIDE

674-si

60 50 40 30 20 10 0

SC

% Annexin Vpositive cells

b

674-si

674-si

sc

SC

BCLxL and BCL-2 mRNA/1x105 β-2-MG

β-actin

Figure 8 The role of Bcl-xL and Bcl-2 genes in prevention of apoptosis upon BCL11B knock down in normal T-cell. (a) Indirect downregulation of Bcl-xL by BCL11B knock down and direct inhibition of Bcl-2 expression in peripheral blood T cells obtained from three different individuals. Upper panel: Bcl11b protein levels 48 h after siRNA treatment. Lower panel: relative Bcl-xL and Bcl-2 mRNA levels 24 h after nucleofection. *Po0.05. (b) Apoptosis measured by Annexin-V-binding assay: left panel – 72 h after knock down, without apoptosis induction (etoposide); right panel: 6 h treatment with 10 mM etoposide added 48 h after knock down ( þ etoposide). sc, non-silencing control siRNA; 674, BCL11B-specific siRNA; Bcl2-si, Bcl2-specific siRNA; *Po0.05.

complete block of thymopoiesis in the early stages in these animals. The analysis of the expression and phosphorylation status of many different apoptosis mediators revealed that only a few of them were involved, among them the antiapoptotic protein BclxL. This observation was confirmed in a recent report by the same group (Inoue et al., 2006). The authors demonstrated reduced Bcl-xL and Bcl-2 protein levels in sorted DN4 BCL11B (/) thymocytes. Interestingly, the transgenic TCRab in those cells reduced apoptosis and raised Bcl-xL and Bcl-2 protein expression without increasing thymus cellularity. These results suggest that Bcl11b influences multiple signaling pathways during thymopoiesis (Inoue et al., 2006). Our results confirm in part the latter findings: whereas we did not observe changes in Bcl-2 levels, we showed a marked decrease of Bcl-xL in T-cell lines and in normal mature T cells upon Bcl11b suppression. The restoration of normal Bcl-xL levels in Bcl11b-deficient leukemic cells significantly protected these cells from apoptosis. These observations indicate that Bcl-xL is positively Oncogene

regulated by Bcl11b and that this might contribute to the enhanced survival of transformed T cells. However, the underlying molecular mechanisms by which Bcl11b regulates the expression of Bcl-xL remain to be elucidated. Our findings are in line with the known role of Bcl-xL in thymopoiesis and malignant transformation. The knockout mouse model showed that, as with Bcl11b, deficiency of Bcl-xL led to massive apoptosis of hematopoietic cells, and impaired maturation and shortened lifespan of lymphocytes (Motoyama et al., 1995). In turn, overexpression of Bcl-xL within the Tcell lineage resulted in accumulation of thymocytes in lymphoid organs and increased resistance to apoptosis (Grillot et al., 1995). Moreover, overexpression of BclxL was detected in T-ALL and ATLL cell lines and samples, and its suppression by anti-sense approaches induced apoptosis and sensitized T-ALL cells to chemotherapeutic drugs (Mori et al., 2001; Broome et al., 2002). Taken together, published data and our findings implicate that Bcl11b and Bcl-xL may act on a common antiapoptotic axis and therefore contribute to malignant transformation in the T-cell lineage. In addition to Bcl-xL, we found another important apoptosis mediator involved in Bcl11b deficiencyinduced cell death. Transcriptional activation of TRAIL and markedly elevated levels of membrane-bound and intracellular protein have been found in tumor T-cell lines as a result of Bcl11b inhibition. Both the cytokine Trail, which acts via the death-receptor pathway, and agonistic antibodies that bind to its receptors are known to selectively induce apoptosis in various malignant cells, including leukemias. Recently, the upregulation of TRAIL was identified as a molecular basis for the tumor-selective action of HDAC inhibitors in acute myeloid leukemia (Insinga et al., 2005; Nebbioso et al., 2005). Bcl11b in turn has been found to interact with HDAC1 and HDAC2 within the NuRD nucleosome remodeling complex, and its transcriptional repression activity was abrogated by HDAC inhibitors (Cismasiu et al., 2005). The authors envisioned that Bcl11b would specifically recruit the NuRD complex to unknown targeted promoters and as a consequence repress gene transcription. This was confirmed recently by detecting the Bcl11b/NuRD complexes on the promoter of p57KIP putative tumor-suppressor gene in neuroblastoma cells (Topark-Ngarm et al., 2006). Therefore, it is tempting to speculate that TRAIL might be one of the Bcl11b/NuRD target genes, at least in Tcell lineage. Caspase and neutralization assays indicated that TRAIL might be as important for cell death as Bcl-xL in our system. Although blockade of caspase 9, which is located downstream of Bcl-xL in the intrinsic apoptotic pathway, resulted in only minor increase of viability of Bcl11b-deficient cells, inhibition of caspase 8, acting downstream of TRAIL, markedly reduced the effect of Bcl11b knock down. Similarly, inactivation of TRAIL by neutralizing antibodies or simultaneous knock down also significantly diminished the percentage of apoptotic cells upon Bcl11b suppression in huT78 cells. However,


3807

it has been shown that Bcl-xL may be responsible for TRAIL resistance in cancer cells (Sun et al., 2002; Kim et al., 2003; Zhang and Fang, 2005a) and that Bcl-xL suppression is required for TRAIL-mediated apoptosis in different experimental systems (Bradham et al., 1998; Medema et al., 1998). Therefore, Bcl-xL and TRAIL probably act in concert through a crosstalk between the intrinsic and death receptor-mediated pathways of apoptosis. It is known that Bcl11b is required for differentiation and survival of ab thymocytes. Our results show that survival of transformed T-cell lines in vitro is dramatically decreased upon Bcl11b suppression. Therefore, we next investigated the consequences of Bcl11b inhibition in normal, mature T cells. First, we showed that elevated BCL11B mRNA levels in T-cell lines and leukemia samples did not correspond with increased protein levels. Obviously, transcription and translation are regulated differently in normal and malignant T cells. Interestingly, we did not observe any decrease in viability or increased sensitivity to DNA damageinduced apoptosis of in vitro cultured Bcl11b-deficient peripheral blood T cells. Although a marked downregulation of Bcl-xL was observed, TRAIL mRNA was elevated marginally compared to the transformed T-cell lines, and Trail protein could not be detected on the cell surface upon Bcl11b inhibition. However, intracellular Trail increased slightly but significantly, which is in line with findings, that normal T cells upon stimulation secrete Trail rather than retain it on their surface (Ehrlich et al., 2003). It has been reported that, in contrast to thymocytes and transformed T cells, normal mature T cells may be independent of Bcl-xL as a result of high Bcl-2 expression (Grillot et al., 1995). This was elegantly confirmed in a Bcl-xL conditional knockout mouse. In this model, no alterations in cell survival and immune function were found in either effector or memory T cell (Zhang and He, 2005b). Our results confirm the compensatory role of Bcl-2 in Bcl-xL deficient normal T cells. In addition to Bcl-xL independence, normal peripheral T cells are insensitive to TRAIL treatment, because they do not express TRAIL receptors (Jeremias et al., 1998; Simon et al., 2001; Klose et al., 2005). Assuming the profound role of Bcl-xL and TRAIL in BCL11BRNAi mediated cell death, one should expect a lack of apoptosis induction in non-transformed mature T cells upon Bcl11b suppression, as demonstrated here. Notably, as Jurkat is derived from immature (Schneider et al., 1977) and huT-78 (Gootenberg et al., 1981) from fully differentiated T cells, the selective apoptogenic effect of the BCL11B suppression seems to be not only a property of the differentiation status (DN thymocytes vs mature T cells) but also of the transformed phenotype (malignant vs normal T cells). Our study provides evidence for an antiapoptotic function of Bcl11b in T-cell malignancies. The selective Bcl11b dependence of transformed T cells makes it an attractive target for novel therapeutic strategies directed against T-ALL and T-cell lymphomas.

Materials and methods Reagents All biochemicals were from Sigma Chemicals (Munich, Germany) unless otherwise specified. Caspase 8 inhibitor, IETD-fluoromethyl ketone (FMK) and negative control peptide Z-FA-FMK were purchased from R&D Systems (Abingdon, UK). Caspase 9 inhibitor LEHD-FMK was provided by BD Pharmingen (Heidelberg, Germany). Human recombinant Trail was purchased from PeproTech (Peprotech Ltd, London, UK). Antibodies Primary antibodies for the following proteins were used: affinity-purified, anti-human Bcl11b rabbit polyclonal antibody (Biogenes, Berlin, Germany); anti-human Trail-PE mouse monoclonal antibody (R&D Systems, Abingdon, UK); anti-human Bcl-xl mouse monoclonal antibody (BD Transduction Laboratories, Erembodegem, Belgium); antiactin, affinity-purified antibody (Sigma, St Louis, MO, USA); anti-human caspase 3-FITC, active form, (BD Pharmingen). Secondary alkaline-phosphatase-conjugated donkey antimouse IgG and goat anti-rabbit antibodies were purchased from Jackson ImmunoResearch Laboratories Inc. (Soham, UK). Goat anti-mouse horseradish peroxidase (HRP) conjugated antibody was provided by Dako Cytomation (Hamburg, Germany). Cell culture Jurkat human T-cell leukemia (DSMZ, Braunschweig, Germany), huT78 human T-cell lymphoma cell lines (ATCC, Manassas, VA, USA) and Raji human Burkitt lymphoma cell line (ATCC) were grown in Rosewell Park Memorial Institute medium (RPMI) 1640 medium (Invitrogen, Paisley, UK), supplemented with 1 mM sodium pyruvate, 1 non-essential amino acids (Invitrogen), 10% fetal calf serum (FCS) (Invitrogen) and 5 mg/ml Plasmocin (Amaxa, Cologne, Germany) and maintained in a humidified incubator at 371C and 5% CO2. All experiments were performed using cells in exponential phase growth. Jurkat TetR-IRES-DsRed cell line expressing the tetracycline repressor (a kind gift of DW Kowalczyk) was cultured in RPMI 1640 medium supplemented with 1 mM sodium pyruvate, 1 non-essential amino acids (Invitrogen), 10% TetOn System approved FCS (BD Clontech, Heidelberg, Germany). Purified human T cells were prepared using the Pan T-Cell Isolation kit II (Miltenyi, Bergisch Gladbach, Germany) and stimulated by T-cell Activation/Expansion Kit (Miltenyi) according to the manufacturer’s instructions. Cells were cultured in RPMI 1640 medium supplemented with 1 mM sodium pyruvate, 1 non-essential amino acids (Invitrogen), 10% FCS (PanBiotech, Aidenbach, Germany) and 20 U/ml recombinant human IL-2 (Sigma). siRNA, plasmids and cell transfections BCL11B-specific (BCL11B-674), TRAIL-specific Stealth siRNAs and the corresponding non-silencing control (sc) RNA duplexes were designed and synthesized by Invitrogen. Prevalidated Bcl-2-specific synthetic siRNA was purchased from MWG Biotech AG (Ebersberg, Germany). Five micrograms of each siRNA were used for each transfection unless otherwise specified. Cells were transfected with Amaxa Nucleofection Device II (Amaxa), using the Amaxa nucleofection Kit V or T-cell Oncogene


3808 nucleofection kit according to the manufacturer’s instructions. Other experimental conditions are given in the figure legends. To develop the tetracycline-inducible BCL11B-specific RNA interference system, the DNA oligonucleotides were designed and purchased from Oligoengine (Seattle, WA, USA). The oligonucleotides were annealed and ligated into BglII/XhoI sites of pSuperior-Neor-EGFP plasmid (Oligoengine) using standard cloning procedures. Plasmids were transfected into Jurkat TetR-IRES-DsRed cells using Lipofectamine2000 (Invitrogen), and stably transfected cells were selected by culturing in the presence of 1 mg/ml geneticin (Invitrogen). The BCL-xL coding sequence was amplified and ligated into XhoI and HindIII sites of pIRES2-EGFP vector (BD Clontech, Heidelberg, Germany) using T4 DNA Ligase (Promega, Madison, WI, USA). For stable transfection, Jurkat and huT78 cells were transfected and selected in the presence of 1 mg/ml geneticin (Invitrogen) The sequences of all siRNAs and shRNAs used in this study are provided in the Supplementary Information. RNA isolation, expression profiling, reverse transcription and realtime QRT–PCR Total RNA was isolated from the cells with Trizol (Invitrogen). Affymetrix microarray analysis was performed using the one-cycle target Labeling and Control Reagents and Human Genome U133 Plus 2.0 DNA Microrrays (Affymetrix, Santa Clara, CA, USA) according to the manufacturer’s instructions. The subsequent scanning was performed with the GeneChip Scanner 3000 (Affymetrix). The probe sets displaying signal log ratios of X1 or p1 and signal differences of X200 were regarded as regulated in the knock-down sample compared to the control. For QRT–PCR, RNA was reverse transcribed with PowerScript reverse transcriptase (BD Clontech) using random hexamers (Promega). All primers and probes were synthesized by TibMolBiol (Berlin, Germany). Quantification of BCL11B mRNA was performed as described elsewhere (Przybylski et al., 2005). BCLxL TRAIL and b-2-microglobulin (b-2-MG) reference gene mRNA expression was measured with the 7500 RealTime PCR System (Applied Biosystems, Foster City, CA, USA) using the Platinum SYBR Green qPCR SuperMix-uracil-DNA glycosylase (Invitrogen) according to the manufacturer’s instructions. The sequences of all primers used in this study are provided in the Supplementary Information. Apoptosis assays Cells were treated with siRNA as described in the figure legends; 24–72 h after siRNA treatment, DNA content was analysed by propidium iodide incorporation assay. Cells were washed twice with phosphate-buffered saline (PBS), and fixed in 70% ethanol at 201C for at least 30 min. After centrifugation, cells were re-suspended in PBS containing 1% glucose, 50 mg/ml RNase A and 50 mg/ml propidium iodide (Sigma) and incubated in the dark at room temperature for 30 min. Flow cytometry analysis was performed on a Becton Dickinson (Heidelberg, Germany) FACSCalibur using CellQuest software. Apoptotic cells were detected as sub-G0/G1 peak. Annexin-V-binding assays were performed at 24, 48 and 72 h after siRNA nucleofection using the BD Clontech ApoAlert Annexin-V-FITC or Annexin-V-APC (BD Pharmingen) kits according to the manufacturer’s instructions. The percent apoptotic cells was quantified by FACS analysis. Activation of caspase 3 was measured by FACS with FITClabeled active caspase 3 rabbit IgG, after fixation and permeabilization of the cells with CytoFix/CytoPerm (BD Pharmingen). Caspase 8 and 9 inhibition was achieved by Oncogene

IETD-FMK (R&D) and LEHD-FMK (BD Pharmingen), respectively; Z-FA-FMK peptide (R&D) was used as a negative control. Dimethylsulfoxide (DMSO) vehicle delivered at the dilutions corresponding to the concentrations of inhibitors was used as an additional control. To inhibit Trail, increasing doses (2–50 ng/ml) of Trail-neutralizing antibody (Abcam, Cambridge, UK) were added to cell cultures upon Bcl11b suppression. Caspase 3 activity was detected with Caspase-Glo 3/7 assay (Promega) and measured with a GENios luminometer (Tecan, Crailsheim, Germany). Immunoblotting Cells were lysed on ice for 15 min (1 107 cells/50 ml radioimmunoprecipitation assay buffer; PBS containing 1% Igepal, 0.5% deoxycholic acid and 0.1% sodium dodecyl sulfate (SDS)), supplemented with a freshly added complete protease inhibitor cocktail (Roche, Palo Alto, CA, USA) according to the supplier’s instructions. After the cells had been lysed, samples were centrifuged at 10 000 g. Protein concentrations of supernatants were determined by Micro BCA assay (Pierce, Rockford, IL, USA) and samples were separated by SDS–10% polyacrylamide gel electrophoresis (PAGE) or SDS–15% PAGE (BioRad, Munich, Germany). The proteins (50 or 25 mg) were blotted onto polyvinylidine membranes (ROTH, GmbH, Karlsruhe, Germany) using a tank blotting system (BioRad). Equal protein loads were confirmed by Ponceau staining and by immunodetection of b-actin. The membranes were subsequently blocked in 50 mM Tris-buffered saline (pH ¼ 7.6) containing 3% NaCl and 3% non-fat dry milk (BioRad). Afterwards, the membranes were incubated with polyclonal rabbit anti-human Bcl11b antibodies (0.2 mg/ml, BioGenes, Berlin, Germany) and rabbit anti-actin antibodies (0.5 mg/ml, Sigma, Munich, Germany), followed by alkaline phosphatase-conjugated goat anti-rabbit antibodies (Jackson ImmunoResearch Europe Ltd). The binding was visualized by the Sigma Fast BCIP/NBT buffered substrate. Bcl-xL protein was detected using anti-human Bcl-xL mouse antibodies (0.5 mg/ml, BD Transduction Laboratories, Erembodegem, Belgium) followed by goat anti-mouse IgG-HRP (Dako Cytomation GmbH, Hamburg, Germany) and visualized by enhanced chemiluminescence Plus detection reagents (Amersham Biosciences, Uppsala, Sweden) using Image Station 440CF (Kodak, Rochester, NY, USA). ELISA assay Intracellular human Trail levels were measured with Human TRAIL/TNFSF10 Quantikine ELISA Kit (R&D Systems) in Jurkat, huT78 and normal T cells subjected to BCL11B siRNA-mediated suppression. Cells were counted and 1 106 of cells were lysed and measured 48 h after transfection according to the manufacturer’s instructions. Cytofluorometric analysis Cells were prepared for FACS measurements according to the protocols provided with the assays or antibodies and measured with FACSCalibur (Becton Dickinson). The results were analysed using CellQuest and WinMDI 2.8 software. Statistical analysis All experiments were performed at least in triplicate. Statistical significance of the observed differences was analysed by Statistica Software (StatSoft Inc., Tulsa, OK, USA), using the Student’s t-test for unpaired (for Jurkat and huT78 cells) or Student’s t-test for paired data (normal T cells from healthy individuals).


3809 Acknowledgements We thank R Jack, J Sonnemann, J Beck, J Mayerle, FU Weiss and M Lerch from the University of Greifswald for helpful discussion and critical reading of the manuscript. This work

was supported by the Alfried Krupp von Bohlen and Halbach Stiftung (PG, GKP, JB and CAS), the Committee for Scientific Research, Poland (Grant KBN 2 P05A 057 26; GKP), the German Jose˘ Carreras Leukemia Foundation (PG, KA and CAS) and BMBF-NBL3 (PG, CAS and UV).

References Avram D, Fields A, Pretty On Top K, Nevrivy DJ, Ishmael JE, Leid M. (2000). Isolation of a novel family of C(2)H(2) zinc finger proteins implicated in transcriptional repression mediated by chicken ovalbumin upstream promoter transcription factor (COUP-TF) orphan nuclear receptors. J Biol Chem 275: 10315–10322. Avram D, Fields A, Senawong T, Topark-Ngarm A, Leid M. (2002). COUP-TF (chicken ovalbumin upstream promoter transcription factor)-interacting protein 1 (CTIP1) is a sequence-specific DNA binding protein. Biochem J 368: 555–563. Bezrookove V, van Zelderen-Bhola SL, Brink A, Szuhai K, Raap AK, Barge R et al. (2004). A novel t(6;14)(q25q27;q32) in acute myelocytic leukemia involves the BCL11B gene. Cancer Genet Cytogenet 149: 72–76. Bradham CA, Qian T, Streetz K, Trautwein C, Brenner DA, Lemasters JJ. (1998). The mitochondrial permeability transition is required for tumor necrosis factor alphamediated apoptosis and cytochrome c release. Mol Cell Biol 18: 6353–6364. Broome HE, Yu AL, Diccianni M, Camitta BM, Monia BP, Dean NM. (2002). Inhibition of Bcl-xL expression sensitizes T-cell acute lymphoblastic leukemia cells to chemotherapeutic drugs. Leuk Res 26: 311–316. Cismasiu VB, Adamo K, Gecewicz J, Duque J, Lin Q, Avram D. (2005). BCL11B functionally associates with the NuRD complex in T lymphocytes to repress targeted promoter. Oncogene 24: 6753–6764. Cismasiu VB, Ghanta S, Duque J, Albu D, Chen HM, Kasturi R et al. (2006). BCL11B participates in the activation of interleukin-2 gene expression in CD4+ T lymphocytes. Blood 108: 2695–2702. Ehrlich S, Infante-Duarte C, Seeger B, Zipp F. (2003). Regulation of soluble and surface-bound TRAIL in human T cells, B cells, and monocytes. Cytokine 24: 244–253. Gesk S, Martin-Subero JI, Harder L, Luhmann B, Schlegelberger B, Calasanz MJ et al. (2003). Molecular cytogenetic detection of chromosomal breakpoints in T-cell receptor gene loci. Leukemia 17: 738–745. Gootenberg JE, Ruscetti FW, Mier JW, Gazdar A, Gallo RC. (1981). Human cutaneous T cell lymphoma and leukemia cell lines produce and respond to T cell growth factor. J Exp Med 154: 1403–1418. Grillot DA, Merino R, Nunez G. (1995). Bcl-XL displays restricted distribution during T cell development and inhibits multiple forms of apoptosis but not clonal deletion in transgenic mice. J Exp Med 182: 1973–1983. Inoue J, Kanefuji T, Okazuka K, Watanabe H, Mishima Y, Kominami R. (2006). Expression of TCRalphabeta partly rescues developmental arrest and apoptosis of alphabeta T cells in Bcl11b/ mice. J Immunol 176: 5871–5879. Insinga A, Minucci S, Pelicci PG. (2005). Mechanisms of selective anticancer action of histone deacetylase inhibitors. Cell Cycle 4: 741–743. Jeremias I, Herr I, Boehler T, Debatin KM. (1998). TRAIL/ Apo-2-ligand-induced apoptosis in human T cells. Eur J Immunol 28: 143–152.

Kim IK, Jung YK, Noh DY, Song YS, Choi CH, Oh BH et al. (2003). Functional screening of genes suppressing TRAILinduced apoptosis: distinct inhibitory activities of Bcl-XL and Bcl-2. Br J Cancer 88: 910–917. Klose R, Kemter E, Bedke T, Bittmann I, Kelsser B, Endres R et al. (2005). Expression of biologically active human TRAIL in transgenic pigs. Transplantation 80: 222–230. Leid M, Ishmael JE, Avram D, Shepherd D, Fraulob V, Dolle P. (2004). CTIP1 and CTIP2 are differentially expressed during mouse embryogenesis. Gene Expr Patterns 4: 733–739. MacLeod RA, Nagel S, Kaufmann M, Janssen JW, Drexler HG. (2003). Activation of HOX11L2 by juxtaposition with 30 -BCL11B in an acute lymphoblastic leukemia cell line (HPB-ALL) with t(5;14)(q35;q32.2). Genes Chromosomes Cancer 37: 84–91. Medema JP, Scaffidi C, Krammer PH, Peter ME. (1998). BclxL acts downstream of caspase-8 activation by the CD95 death-inducing signaling complex. J Biol Chem 273: 3388– 3393. Mori N, Fujii M, Cheng G, Ikeda S, Yamasaki Y, Yamada Y et al. (2001). Human T-cell leukemia virus type I tax protein induces the expression of anti-apoptotic gene Bcl-xL in human T-cells through nuclear factor-kappaB and c-AMP responsive element binding protein pathways. Virus Genes 22: 279–287. Motoyama N, Wang F, Roth KA, Sawa H, Nakayama K, Negishi I et al. (1995). Massive cell death of immature hematopoietic cells and neurons in Bcl-x-deficient mice. Science 267: 1506–1510. Nagel S, Kaufmann M, Drexler HG, MacLeod RA. (2003). The cardiac homeobox gene NKX2–5 is deregulated by juxtaposition with BCL11B in pediatric T-ALL cell lines via a novel t(5;14)(q35.1;q32.2). Cancer Res 63: 5329–5334. Nebbioso A, Clarke N, Voltz E, Germain E, Ambrosino C, Bontempo P et al. (2005). Tumor-selective action of HDAC inhibitors involves TRAIL induction in acute myeloid leukemia cells. Nat Med 11: 77–84. Oshiro A, Tagawa H, Ohshima K, Karube K, Uike N, Tashiro Y et al. (2006). Identification of subtype-specific genomic alterations in aggressive adult T-cell leukemia/lymphoma. Blood 107: 4500–4507. Przybylski GK, Dik WA, Wanzeck J, Grabarczyk P, Majunke S, Martin-Subero JI et al. (2005). Disruption of the BCL11B gene through inv(14)(q11.2q32.31) results in the expression of BCL11B-TRDC fusion transcripts and is associated with the absence of wild-type BCL11B transcripts in T-ALL. Leukemia 19: 201–208. Sakata J, Inoue J, Ohi H, Kosugi-Okano H, Mishima Y, Hatakeyama K et al. (2004). Involvement of V(D)J recombinase in the generation of intragenic deletions in the Rit1/Bcl11b tumor suppressor gene in gamma-ray-induced thymic lymphomas and in normal thymus of the mouse. Carcinogenesis 25: 1069–1075. Schneider U, Schwenk HU, Bornkamm G. (1977). Characterization of EBV-genome negative ‘null’ and ‘T’ cell lines derived from children with acute lymphoblastic leukemia Oncogene


3810 and leukemic transformed non-Hodgkin lymphoma. Int J Cancer 19: 621–626. Senawong T, Peterson VJ, Avram D, Shepherd DM, Frye RA, Minucci S et al. (2003). Involvement of the histone deacetylase SIRT1 in chicken ovalbumin upstream promoter transcription factor (COUP-TF)-interacting protein 2mediated transcriptional repression. J Biol Chem 278: 43041–43050. Senawong T, Peterson VJ, Leid M. (2005). BCL11A-dependent recruitment of SIRT1 to a promoter template in mammalian cells results in histone deacetylation and transcriptional repression. Arch Biochem Biophys 434: 316–325. Simon AK, Williams O, Mongkolsapaya J, Jin B, Xu XN, Walczak H et al. (2001). Tumor necrosis factor-related apoptosis-inducing ligand in T cell development: sensitivity of human thymocytes. Proc Natl Acad Sci USA 98: 5158–5163. Sun XM, Bratton SB, Butterworth M, MacFarlane M, Cohen GM. (2002). Bcl-2 and Bcl-xL inhibit CD95-mediated apoptosis by preventing mitochondrial release of Smac/

DIABLO and subsequent inactivation of X-linked inhibitorof-apoptosis protein. J Biol Chem 277: 11345–11351. Topark-Ngarm A, Golonzhka O, Peterson VJ, Barrett Jr B, Martinez B, Crofoot K et al. (2006). CTIP2 associates with the NuRD complex on the promoter of p57KIP2, a newly identified CTIP2 target gene. J Biol Chem 281: 32272–32283. Wakabayashi Y, Inoue J, Takahashi Y, Matsuki A, Kosugi-Okano H, Shinbo T et al. (2003a). Homozygous deletions and point mutations of the Rit1/Bcl11b gene in gamma-ray induced mouse thymic lymphomas. Biochem Biophys Res Commun 301: 598–603. Wakabayashi Y, Watanabe H, Inoue J, Takeda N, Sakata J, Mishima Y et al. (2003b). Bcl11b is required for differentiation and survival of alphabeta T lymphocytes. Nat Immunol 4: 533–539. Zhang L, Fang B. (2005a). Mechanisms of resistance to TRAIL-induced apoptosis in cancer. Cancer Gene Ther 12: 228–237. Zhang N, He YW. (2005b). The antiapoptotic protein Bcl-xL is dispensable for the development of effector and memory T lymphocytes. J Immunol 174: 6967–6973.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc).

Oncogene