Introduction

About 40% of acute myeloid leukemias (AML) are caused by the activity of chimeric proteins encoded by fusion genes generated by chromosomal translocations (Grignani et al, 1993; Look, 1997; Tenen, 2003). Fusion proteins contribute to a combination of genetic lesions that produce differentiation block and altered growth of leukemia cells. A number of fusion proteins function as transcriptional repressors (Melnick and Licht, 1999; Tenen, 2003), due to their ability to firmly bind transcriptional corepressor molecules. Retinoic acid receptor (RAR) fusion proteins in acute promyelocytic leukemia (APL) and AML1/ETO in AML M2–M4 aberrantly bind the corepressors N-CoR or SMRT. These molecules engage to the promoter of RAR or AML1 target genes a multiprotein complex including Sin3A and histone deacetylases (HDACs) (Gelmetti et al, 1998; He et al, 1998; Lin et al, 1998; Lutterbach et al, 1998; Grignani et al, 1998a; Wang et al, 1998; Melnick and Licht, 1999). HDACs deacetylate histones and can secondarily enroll DNA methyltransferases (Baylin, 2002; Di Croce et al, 2002). This pathologic sequence is caused by the ability of AML1/ETO, PML/RAR or other RAR fusion proteins to generate oligomeric structures that produce an abnormally stable corepressor binding (Kastner et al, 1992; Nervi et al, 1992; Grignani et al, 1996; Lin and Evans, 2000; Minucci et al, 2000). The overall effect is hypoacetylation of histones and DNA methylation that rearrange chromatin structure hampering transcription of crucial myeloid differentiation genes (Ferrara et al, 2001; Di Croce et al, 2002; Alcalay et al, 2003; Tenen, 2003). Murine bone marrow transduction/transplantation models and cell line studies have shown that the activity of these leukemia fusion proteins can block cell differentiation (Grignani et al, 1993, 1998a, 2000; Gelmetti et al, 1998; He et al, 1998; Lin et al, 1998; Lutterbach et al, 1998; Wang et al, 1998; Melnick and Licht, 1999; Minucci et al, 2002; Schwieger et al, 2002; Tenen, 2003). However, in transgenic mouse models, PML/RAR and AML1/ETO are not sufficient to induce differentiation block, which appears to require further molecular damages (reviewed in Melnick and Licht, 1999; Bernardi et al, 2002). In this context, a crucial question is whether fusion protein functions are required to maintain a fully malignant leukemia phenotype, including the block of differentiation. In such a case, targeting the primary fusion protein/corepressor interaction could reverse gene silencing and restore the differentiation potential of leukemia cells. This would represent a therapeutic goal, specifically directed to leukemia cells, where abnormal corepressor recruitment occurs. Actually, differentiation treatment of APL with retinoic acid (RA) is based on the ability of this drug to detach corepressors from the PML/RAR protein, inducing coactivators recruitment on both the fusion protein and the normal RAR (Grignani et al, 1998a; He et al, 1998; Lin et al, 1998). However, most acute leukemia blast cells are not induced to differentiate by RA. Moreover, APL themselves include RA-resistant cases, due to the expression of the PLZF/RAR fusion protein or of mutant PML/RAR proteins (Warrell, 1993; Shao et al, 1997; Melnick and Licht, 1999). Thus, a general strategy to modify fusion protein/corepressor interactions and induce leukemia cell differentiation is currently unavailable.

Targeting fusion protein/corepressor contact can be achieved by interfering with the protein interaction interfaces. The corepressor interaction surfaces of the PML/RAR and AML1/ETO proteins have been mapped and contact well-defined domains of the corepressors (Gelmetti et al, 1998; Lutterbach et al, 1998; Wang et al, 1998; Glass and Rosenfeld, 2000). Peptides representing the corepressor core interaction domains inhibit the RAR/N-CoR association, and are released by RA (Hu and Lazar, 1999; Nagy et al, 1999; Perissi et al, 1999). Therefore, expression in leukemia cells of protein sequences that are representative of the interaction surfaces could block fusion protein/corepressor interactions.

In this work, we investigated whether interrupting the contact between fusion proteins and corepressors could modify the leukemia phenotype. We took advantage of two well-studied models: the leukemias derived from the expression of the PML/RAR or the AML1/ETO fusion proteins. We show that expression of short sequences representing the fusion protein interaction domains of N-CoR efficiently blocks these interactions and restores the differentiation response of leukemia cells. These effects were also obtained by direct protein transduction. Our data indicate that fusion protein activity is a permanent requirement for the leukemia phenotype and that it is necessary for the leukemia differentiation block. The block of pathogenetic protein–protein interaction could represent a novel strategy for cancer treatment.

Results

Expression of interaction domain peptides restores differentiation response in leukemia cells

To block fusion protein/N-CoR interactions, we expressed in PML/RAR or the AML1/ETO positive leukemia cell lines protein fragments representing the interaction sequences: the IDN and IDC domains of N-CoR that include the nuclear receptors corepressor binding sites (Hu and Lazar, 1999; Nagy et al, 1999; Perissi et al, 1999), and the N-CoR RD3 domain that interacts with ETO (Gelmetti et al, 1998; Lutterbach et al, 1998; Wang et al, 1998) (Figure 1A).

Using bicistronic retroviral vectors, we coexpressed green fluorescence protein (GFP) and the IDC or IDN domains of N-CoR in the APL cell lines NB4, containing PML/RAR, and NB4R4, a derivative of NB4 made RA-resistant by a mutation in the RA binding region of PML/RAR (Lanotte et al, 1991; Shao et al, 1997). In these cell lines, we also expressed a mutated IDC (M10) whose ability to bind nuclear receptors is greatly reduced (Nagy et al, 1999) and a RAR fragment spanning the region of the corepressor binding surface (D403). We then expressed these protein fragments in U937 and HL60 cells, which do not contain fusion proteins. As a model of AML1/ETO leukemia, we utilized the SKNO1 cell line that has high and stable expression of the fusion protein (Matozaki et al, 1995). In SKNO1 cells, we expressed the RD3 and IDC domains of N-CoR, with the latter as a negative control. We added an HA tag to IDN, D403 and RD3 domains to make them recognizable by antibodies. GFP-expressing cells were purified by fluorescence-activated cell sorting (FACS). These cell lines expressed the different fragments (Figure 1B).

The expression of HDAC1, HDAC3 SMRT and N-CoR protein (Figure 1C) and the basal differentiation stage of all these lines were not significantly affected by the expression of any of these fragments (Figure 2). We next measured by real-time PCR the mRNA expression of selected fusion protein target genes (Zhu et al, 2001; Linggi et al, 2002; Alcalay et al, 2003) (Figure 1D). IDC and IDN expression in both NB4 and NB4R4 cells increased the PML/RAR target genes RAR and granulocyte colony-stimulating factor receptor (G-CSF-R) mRNA, whereas RD3 expression in SKNO1 cells increased the AML1/ETO target genes G-CSF-R and p14^ARF mRNA. Since these genes participate in the differentiation process, we studied the cell response to differentiation inducers, vitamin D3 (D3) or RA. As expected, D3-induced differentiation is impaired in NB4, NB4R4 and SKNO1 cell lines. In contrast, in NB4-IDC/IDN, NB4R4-IDC/IDN and SKNO1-RD3 cells expressing the interaction peptides, we observed a strong D3-mediated differentiation induction as measured by the expression of the myeloid differentiation surface markers CD11b and CD14 (Figure 2A and B and Supplementary Figure 1). Confirming that the N-CoR fragments exerted their activity on the PML/RAR-RAR pathway, the expression of the RAR D403 fragment restored D3-induced differentiation in NB4 D403 and NB4R4D403 cells (Figure 2A and B). Differentiation-unblocking effects were specific since they were not detectable in NB4 and R4 cells expressing the IDC M10 mutant, or in SKNO1 cells expressing IDC.

When treated with RA, both control NB4 and NB4-IDC/IDN cells differentiated efficiently, while NB4R4 cells were unresponsive. However, NB4R4-IDC/IDN cells displayed a strong induction of myeloid differentiation, as measured by morphology, NBT reduction assay and expression of surface differentiation markers (Figure 3A, C and E). In NB4R4D403, RA-induced differentiation was also clear. U937 and HL60 cells expressing IDC did not show any impairment of RAR expression or RA-induced differentiation, indicating that IDC expression does not decrease the activity of wild-type RAR (not shown). SKNO1 cells are unresponsive to RA. However, when these cells expressed the RD3 N-CoR fragment, RA treatment induced their differentiation, although it was less complete than in NB4, as shown by growth arrest, CD11b expression and NBT assay (Figure 3B, D and F). The cell growth of NB4 and NB4R4 cells was not significantly affected by the expression of the N-CoR fragments (not shown). SKNO1-RD3 cells grew significantly less rapidly than control SKNO1 cells (Figure 3F).

Taken together, these data suggest that the expression of specific fusion protein/corepressor interaction domains restores differentiation response in leukemia cells by derepression of fusion protein targets.

Expression of interaction domain peptides triggers specific PML/RAR degradation

We next asked whether the altered interactions with N-CoR could affect stability and expression of the fusion proteins. Indeed, in both NB4IDC or IDN and NB4R4IDC or IDN cells, the PML/RAR protein expression was markedly reduced with respect to control cells (Figure 4A). Notably, the RAR protein was not affected. In RT–PCR assays, however, the expression of the PML/RAR fusion mRNA was still abundant (Figure 4B), suggesting that protein degradation was occurring. The phenomenon is ligand independent since it occurred in the absence of RA and was maintained in the absence of serum (Supplementary Figure 2). To test whether proteasomal degradation could be responsible for the reduced PML/RAR protein expression, we treated IDC-expressing cells with the proteasome inhibitors MG132 and lactacystein (Zhu et al, 1999). Fusion protein expression was restored (Figure 4C). Since elastases and caspases can proteolyse PML/RAR in vivo (Nervi et al, 1998; Lane and Ley, 2003), we treated these cells with the serine protease inhibitor PMSF, and the caspase inhibitors Z-VAD and DEVD. Caspase inhibitors had no effect (not shown), in agreement with the PML/RAR-dependent caspase activation by RA (Nervi et al, 1998). PMSF treatment increased the expression of PML/RAR in NB4 and NB4R4 cells, but failed to restore PML/RAR expression in IDC or IDN cells (Figure 4C and not shown). We concluded that IDC/IDN expression triggers degradation of the PML/RAR fusion protein by proteasomal enzymes. Proteolysis was PML/RAR specific, since SKNO1 cells expressing the N-CoR RD3 fragment (Figure 4A) displayed only a marginally reduced amount of AML1/ETO protein.

We then asked whether the restored differentiation responsiveness obtained in NB4 and NB4R4 cells was due to fusion protein degradation. We restored PML/RAR protein expression in NB4 and NB4R4IDC or IDN cells by proteasomal inhibitors and induced cell differentiation with D3. Despite the re-established expression of PML/RAR, both these cell lines were efficiently differentiated by treatment with D3 (Figure 4C and D), indicating that the disturbance of fusion protein association with N-CoR rather than fusion protein degradation was responsible for the restored differentiation potential of the leukemia cells.

Block of fusion protein/corepressor interactions is responsible for the restored differentiation potential of leukemia cells

To verify whether expression of the N-CoR fragments can block N-CoR /fusion protein association in live cells, we used the IDC, IDN or the RD3 vector to infect U937 cells with Zn-inducible expression of PML/RAR (U937 PR9 cells) (Grignani et al, 1993) or AML1/ETO (U937 A/E), respectively (Gelmetti et al, 1998). Infected cells expressed the different N-CoR domains (Figure 5A). Notably, fusion protein expression was maintained at 6 h after Zn induction (Figure 5B and C), allowing the analysis of the fusion protein/corepressor complex formation. Thus, we performed co-immunoprecipitation experiments using anti-N-CoR antibodies, followed by Western blotting with anti-RAR or -AML1 antibodies. Reciprocally, we immunoprecipitated the fusion proteins by anti-PML or -ETO antibodies and Western blotted the immunoprecipitated proteins with anti-Sin3A antibodies. The expression of the N-CoR fragments abolished co-precipitation of the PML/RAR or AML1/ETO proteins with the N-CoR/Sin3A complex (Figure 5B and C). Since the same domains of the fusion proteins that bind N-CoR also recruit the SMRT corepressor, we tested whether the N-CoR fragments could impair SMRT interaction with the fusion proteins. Co-immunoprecipitation of SMRT with both PML/RAR and AML1/ETO was markedly reduced by the expression of IDC or RD3 N-CoR fragments in PML/RAR- or AML1/ETO-expressing cells, respectively (Figure 5D). To further prove that the disruption of the corepressor complex was occurring in vivo, we performed chromatin immunoprecipitation (ChIP) with anti-N-CoR and anti-SMRT antibodies and searched for specific sites in the promoter of fusion protein target genes. In PML/RAR cells, we searched for the RA-responsive element on RAR2, and in AML1/ETO cells for an AML1 site on p14^ARF. Again, expression of the N-CoR interaction fragments markedly reduced the amount of N-CoR and SMRT on the promoter region of these genes (Figure 5E). Consistently, mRNA expression of RAR, which is reduced by Zn-induced PML/RAR expression, is increased in cells expressing IDC. Similarly, p14^ARF expression is repressed by Zn-induced AML1/ETO, but is increased by the presence of the RD3 fragment (Figure 5F). We concluded that fusion protein interaction with corepressors can be effectively competed in vivo by peptides representing specific interaction domains of N-CoR.

Block of fusion protein/N-CoR interactions can be obtained by protein transduction strategies

We next explored protein transduction as a tool for delivering therapeutic molecules into the cells (Schwarze et al, 1999, 2000). We produced in bacteria the N-CoR IDC, IDN, M10 and RD3 proteins fused to an HIV TAT protein transduction domain (PTD) and to an SV40-derived nuclear localization signal (NLS). We repeatedly exposed NB4, NB4R4 and SKNO1 cells to the TAT fusion peptides. Western blotting analysis after trypsin treatment of the cells (see Materials and methods) showed the presence of protein fragments in the cells (Figure 6A). The transduced TAT-IDC induced PML/RAR degradation after 3 days of treatment, although partial degradation products, reacting with an anti-RAR antibody, were still visible (Figure 6B). To verify whether the protein fragments were biologically active, we performed D3 and RA differentiation experiments as described above (Figure 6C and D). NB4 and NB4R4 cells transduced with TAT-IDC and TAT-IDN cells displayed a high level of D3-induced differentiation (Figure 6C) as measured by CD11b and CD14 surface expression. Again, NB4R4 cells became highly sensitive to the differentiation effect of RA (Figure 6D). The effects of transducing SKNO1 cells by the TAT-RD3 protein were qualitatively similar to those obtained by expressing the N-CoR RD3 fragment by retroviral vectors. However, in SKNO1 cells, differentiation induction by D3 and RA was less effective, suggesting that the amount of TAT-RD3 protein entering the cells was not sufficient to warrant a complete release of the leukemia differentiation block (Figure 6E). Overall, these results indicate that protein transduction of interaction domain peptides can be envisaged as an effective strategy to unblock differentiation in leukemia cells.

Discussion

In this study, we restored leukemia cell response to RA and D3, two physiological inducers of myeloid differentiation, by expressing short protein sequences representative of the surfaces used by N-CoR to bind the PML/RAR and AML1/ETO proteins. Probably, these protein fragments saturate the N-CoR binding sites on the RAR and ETO moiety of the fusion proteins, impairing N-CoR interaction. Importantly, saturation of fusion protein–corepressor binding sites also displaces SMRT, increasing the effectiveness of this approach. Our co-immunoprecipitation and CHIP data indicate that N-CoR, SMRT and Sin3A may all contribute to the repressor activity of AML1/ETO and PML/RAR. N-CoR fragments dislocate all these members of the repressor complex from the fusion proteins. This occurs in vivo at specific sites on the promoter of fusion protein target genes involved in the regulation of cell differentiation, such as RAR, G-CSF-R and p14^ARF, resulting in their derepression. Although the AML1 moiety of AML1/ETO binds Sin3A (Lutterbach et al, 2000), the ETO-directed N-CoR fragment reduces the amount of fusion protein-bound Sin3A below the sensitivity of our assay. Overall, the disruption of the repressor complex on direct fusion protein targets is likely to be a major contribution in the increased differentiation potential of the cells. The differentiation-unblocking effects of a segment of RAR spanning the region that binds N-CoR further confirm that PML/RAR activity is the major target of IDC/IDN.

Modification of other molecular pathways might contribute to the overall phenotype. Recruitment of N-CoR by vitamin D receptor may be altered, contributing to unblock differentiation. We observed an increased expression of the osteocalcin gene, a target of vitamin D receptor, in U937 cells overexpressing IDC (not shown). Activation of cAMP pathway can restore RA-induced differentiation in RA-resistant APL-derived cells (Kamashev et al, 2004). We cannot exclude that N-CoR fragments may activate the cAMP pathway, but our data suggest that their major effects are actually exerted through modulation of fusion protein target genes.

Expression of the N-CoR fragments specifically induces degradation of the PML/RAR protein. HDAC1, HDAC3, N-CoR and SMRT, important components of the repressor complex, are unmodified. Degradation is ligand independent, since it occurs in serum-free medium, in the absence of retinoids. In fact, inhibitors of caspases, which participate in RA-induced degradation, do not restore PML/RAR expression. Conversely, proteasomal inhibitors abolish this phenomenon, suggesting that the fusion protein is degraded in the proteasome. Serine protease inhibitors should prevent PML/RAR degradation by elastases, recently implicated in the pathogenesis of APL (Lane & Ley, 2003). Actually, these agents appear to increase the expression of PML/RAR and its SUMO-modified forms in NB4 cells, but do not restore PML/RAR expression in IDC/IDN cells, indicating that the pathogenetic degradation is different from the proteolysis seen in this study. Here, PML/RAR degradation is triggered by the loss of corepressor interaction. Likely, interaction with corepressors maintains the fusion protein in a steric conformation that makes it inaccessible to proteasomal enzymes. Likewise, release of corepressors from the fusion protein may also contribute to RA- and arsenic trioxide-induced PML/RAR degradation (Yoshida et al, 1996; Zhu et al, 2001; Hong et al, 2003).

The AML1/ETO protein is only modestly degraded when separated from N-CoR, indicating that the degradation is PML/RAR specific. The RD3 fragment may induce the same conformational changes as the entire corepressor molecule. Alternatively, a fraction of Sin3A protein below the sensitivity of our assays may remain bound to the AML1 moiety of the fusion protein and may be sufficient to stabilize it (Lutterbach et al, 2000; Imai et al, 2004).

Receptor proteolysis plays a role in ligand-dependent transcriptional activation by nuclear receptors (Zhu et al, 2001; Seeler and Dejean, 2003; Perissi et al, 2004). In NB4-IDC/IDN cells, PML/RAR is proteolysed and the cells respond to RA. Although we cannot exclude a contribution of partially degraded forms of the fusion protein, RA response probably derives from the activity of the RAR protein. Its expression is not modified in IDC/IDN cells although its mRNA is increased, possibly implying limited proteolysis that may contribute to ligand-dependent activation of target genes. Overall, in the absence of PML/RAR, NB4-IDC/IDN cells may simply behave like other RA-responsive myeloid cells, since RA binding releases from RAR peptides representing the N-CoR interaction regions, allowing coactivator recruitment (Hu and Lazar, 1999; Nagy et al, 1999; Perissi et al, 1999). Thus, the N-CoR fragments, which specifically target the abnormal protein interactions underlying leukemia transformation, may not affect normal cells. In agreement, IDC expression does not impair RA-induced differentiation of cells that do not express the PML/RAR fusion protein.

Expression of the N-CoR fragments can convert leukemia cells from RA-resistant to RA-responsive. In NB4R4IDC or IDN cells, the dominant-negative effect on RAR of a mutant PML/RAR is abolished. The kinetics of RA-induced differentiation in NB4-IDC/IDN and NB4R4-IDC/IDN cells was somewhat slower than in NB4 cells (Figure 3), possibly due to the loss of PML/RAR contribution to differentiation (Grignani et al, 1993; Kogan et al, 2000) and to stable molecular alterations due to the long-term block of the RAR pathway in NB4R4 cells. Also SKNO1/RD3 cells, which express AML1/ETO, became RA-responsive. These data are in agreement with our previous findings, suggesting that the AML1/ETO fusion protein is able to block the RAR pathway (Ferrara et al, 2001), and show that this is the direct consequence of the AML1/ETO protein interaction with corepressors.

Overall, loss of N-CoR/SMRT interactions, rather than fusion protein degradation, is primarily responsible for restored differentiation response in cells expressing N-CoR fragments. Treatment of NB4IDC and NB4R4IDC cells with proteasome inhibitors re-establishes PML/RAR expression, but the fusion protein cannot recruit N-CoR, due to the overexpression of interaction peptides, and, as a result, cannot block differentiation. Moreover, the SKNO1-RD3 cells differentiate efficiently despite the fact that AML1/ETO protein is only slightly degraded.

Our data have implications regarding the role of fusion proteins in the construction of the leukemia phenotype, a critical issue in the selection of targets for molecular therapy. Transgenic animal models indicate that fusion protein activity is not sufficient to cause differentiation block (reviewed in Melnick and Licht, 1999; Bernardi et al, 2002). However, fusion proteins block differentiation more effectively in murine bone marrow transduction–transplantation models and in cell lines (Melnick and Licht, 1999; Tenen, 2003). We show that fusion protein function is necessary to block leukemia cell response to physiologic myeloid differentiation inducers. Ligand-induced receptor stimulation is still required to trigger maturation of the cells. Overall, full malignant features in leukemia require fusion protein activity. This phenomenon has been referred to as 'addiction' to oncogenes and has been shown for myc-dependent cancers (Weinstein, 2002; Jain et al, 2003). In our model system, it implies that the removal of fusion protein function may restore leukemia cell differentiation response. Thus, fusion proteins are important targets for molecular therapy of leukemia.

Searching for a method to interfere with leukemia transformation in vivo, we obtained direct transduction of ID fragments by fusing them with HIV TAT PTD. The transduced fragments had effects that were qualitatively similar to those obtained by retroviral transduction including PML/RAR protein degradation, although partial degradation products were still visible after 3 days of treatment. SKNO1 cells appear to be less accessible to TAT PTD-mediated transduction than APL-derived cell lines. Nevertheless, we show that this strategy is feasible and can be applied to diverse protein–protein interactions. Protein transfer has already proven to be therapeutically effective in live animals (Asoh et al, 2002; Kilic et al, 2003). Our data establish the foundation for a targeted treatment approach to leukemia, based on its molecular pathogenesis (Rabbitts and Stocks, 2003). As a further support to the relevance of this approach, while this paper was in preparation, it has been published that TAT-mediated transfer of SMRT fragments inactivates the repressor activity of Bcl-6 in vitro and in vivo, leading to growth arrest and apoptosis of lymphoma cells (Polo et al, 2004). Future improvements of protein transfer efficiency or the development of small interfering molecules that act on protein interactions may render this strategy applicable in human therapy.

Materials and methods

Cell culture

The amphotropic packaging cell line Phoenix, the APL cell lines NB4 and NB4R4, the myeloid cell lines HL60 and U937 and its derivatives were cultured in RPMI medium with 10% FBS. For SKNO1 cells, 10 ng/ml GM-CSF was added to the medium.

PCR and RT–PCR, interaction domain fragments, retroviral vector construction, cell infection and cell sorting

cDNA fragments encoding N-CoR interaction domain peptides IDC, IDN and RD3 were cloned by PCR on the N-CoR cDNA (NM_011308) with the following oligonucleotides: RD3 sequence (amino acids 1071–1309) 5'-GCCACCATGGTTCGGCTTCCGACAACTCGACC
AAC-3' and 5'-TCACATCCCTTGCTTTATATTTCCTTCCAC-3
'; IDN sequence (amino acids 2059–2085) 5'-CGCCACCATGGCCAGGACCCATCGACTG-3' and 5'-TCAATTTCTAGCAAAATCTTGTGA-3', IDC sequence (amino acids 2217–2323) 5'-ACCGCGGCCACCATGGTTAAATCAAAG-3' and 5'-ATCTCACCGTGCCTCGCTGCTCGTCAC-3'. In the IDC mutant M10, the amino acids 2275, 2278 and 2279 were mutagenized to alanine by a Quick Change mutagenesis kit (Stratagene, La Jolla, CA). The RAR D403 fragment was obtained by PCR on a RAR cDNA (NM_000964) with the following oligonucleotides: 5'-GCCGCCACCATGGTGACCCGGAAC-3' and 5' CATGGATCACGGGATCTCCATCTT-3' (amino acids 133–403). An HA tag was subcloned in-frame in the IDN, RD3 and D403 vectors. The cDNAs were cloned in a PINCO vector (Grignani et al, 1998b) where the CMV promoter was substituted for by an encephalomyocarditis virus internal ribosomal entry site (IRES). The resulting bicistronic vector encoded both the interaction peptides and GFP. Retroviral vector production and usage have been described previously (Grignani et al, 1998b, 2000; Minucci et al, 2002). Cells infected with empty control vectors and vectors encoding the described protein fragments were purified by FACS as reported (Grignani et al, 1998b, 2000). RT–PCR for detection of PML/RAR mRNA was performed as described (Biondi et al, 1992) with the following oligonucleotides: 5'-CAGTGTACGCCTTCTCCATCA-3' and 5'-AGAACTGCTGCTCTGGGTCTCAAT-3'.

Antibodies, Western blotting, immunoprecipitation, caspase and protease inhibitors

Western blotting and co-immunoprecipitation experiments were performed as described (Grignani et al, 1996, 1998a) using the following antibodies: anti-RAR-F (a gift of P Chambon), anti-Sin3A AK-11, anti-N-CoR for immunoprecipitation, N-19 and anti-PML PG-M3 (Santa Cruz Biotechnology, Santa Cruz, CA), anti-HDAC-1 and anti-N-CoR for Western blotting rabbit polyclonal IgG 06-720 and 06-892 respectively (Upstate Biotechnology, Lake Placid, NY), anti-AML-1/RHD Ab-2 (Oncogene Science, Boston, MA), anti-ETO Ab-1 (Oncogene Science), anti-HA.11 (Babco, Richmond, CA), anti-HDAC-3 CHIP grade and anti-SMRT-1542 (Abcam Ltd, Cambridge, UK).

The proteasome inhibitors MG132 and lactacystein (Biomol Research Laboratories Butler Pike Plymouth Meeting, PA) were used at a concentration of 10^-6 M. PMSF (Sigma-Aldrich, Milano, Italy) was used at 0.5 mM. DEVD and Z-VAD (Sigma-Aldrich, Milano, Italy) were used at 10^-6 M.

Real-time PCR analysis

Quantitative real-time PCR was performed as published (Linggi et al, 2002) in ABI PRISM 7000 Sequence Detection System (Applied Biosystems) using Taqman oligonucleotides for GAPDH, G-CSF-R and RAR (Applied Biosystem) according to the manufacturer's instructions. p14^ARF primers have been previously described (Linggi et al, 2002). Gene expression, normalized for GAPDH expression serving as endogenous control, was calculated using the CT method.

Growth and differentiation and immunophenotyping experiments

Differentiation of U937, HL60, NB4, NB4R4 and SKNO1 cells and derivatives was induced with 250 ng/ml D3 (a gift from Hoffmann-La Roche, Basel, Switzerland) for 3 days. RA (Sigma-Aldrich, Milano, Italy) was used at 10^-6 M. Cells were seeded at 10⁵/ml. Immunophenotyping was performed as published (Grignani et al, 1993) using PE-conjugated Serotech antibodies (Serotech, Oxford, UK). Nitro blue tetrazolium (NBT) assay was performed as described (Grignani et al, 1993).

Chromatin immunoprecipitation

ChIP was performed using previously described oligonucleotides and methods (Linggi et al, 2002) (see the antibodies above) using the same number of cells for each sample. PCR amplification of the RAR2 promoter (sequence AF283809) containing the RA-responsive element was obtained with the following oligonucleotides: Fwd 5'-ACAATGACACAAGCCGGTGTCTCA-3'; Rev 5'-CTTACAGATCAGACGTCAAGCCC-3'. PCR on -actin (NM_001101) was used to detect nonspecific DNA: Fwd 5'-CTTCTACAATGAGCTGCGTGTGG-3'; Rev 5'-CATGGATCACGGGATCTCCATCTT-3'. PCR products were run on an agarose gel, Southern blotted and probed with a cloned and sequenced DNA fragment amplified from genomic DNA with the same oligonucleotides.

TAT fusion protein production and usage

The N-CoR cDNA fragments were subcloned in a 6xHis-based bacterial expression vector, in-frame with the HIV TAT PTD (Schwarze et al, 1999), an HA tag and an NLS from SV40 (Hodel et al, 2001), added by PCR. TAT fusion proteins were purified as described (Vocero-Albani et al, 2001) and added to cell culture in serum-free medium at a final concentration of 200 nM. FCS (10%) was added 30 min after protein addition. During differentiation experiments, addition of TAT fusion protein was repeated four times a day. To detect protein uptake by Western blotting, cells were washed three times in PBS, treated with 2.5 mg/ml trypsin at 37°C for 20 min, washed again as before and lysed in sample buffer. In all, 10^-6 M RA or 250 ng/ml D3 was added to cell culture medium 5 h after the first TAT protein treatment.

Supplementary data

Supplementary data are available at The EMBO Journal Online.

Acknowledgements

We thank Dr PG Pelicci for helpful advice and reagents, Dr M Cioccoloni and Dr S De Matteis for preliminary experiments, Professor P Chambon for the anti-RAR antibody, Dr W Miller Jr for the NB4R4 cell line, Dr Y Honma and Dr J Licht for the SKNO1 cell line and Roche Pharmaceuticals for the supply of D3. This work was supported by grants from AIRC, Italian Ministry for Instruction University and Research (MIUR and FIRB) and Ministry of Health to FG and CN and Fondazione Cenci Bolognetti to CN. SR and MP are recipients of an FIRC fellowship.

References

Alcalay M, Meani N, Gelmetti V, Fantozzi A, Fagioli M, Orleth A, Riganelli D, Sebastiani C, Cappelli E, Casciari C, Sciurpi MT, Mariano AR, Minardi SP, Luzi L, Muller H, Di Fiore PP, Frosina G, Pelicci PG (2003) Acute myeloid leukemia fusion proteins deregulate genes involved in stem cell maintenance and DNA repair. J Cin Invest 112: 1751−1761 | Article |

Asoh S, Ohsawa I, Mori T, Katsura K, Hiraide T, Katayama Y, Kimura Y, Ozaki D, Yamagata K, Ohta S (2002) Protection against ischemic brain injury by protein therapeutics. Proc Natl Acad Sci USA 99: 17107−17112 | Article |

Baylin SB (2002) Mechanism underlying epigenetically mediated gene silencing. Semin Cancer Biol 12: 331−337 | Article |

Bernardi R, Grisendi S, Pandolfi PP (2002) Modelling haematopoietic malignancies in the mouse and therapeutical implications. Oncogene 21: 3445−3458 | Article |

Biondi A, Rambaldi A, Pandolfi PP, Rossi V, Giudici G, Alcalay M, Lo Coco F, Diverio D, Pogliani EM, Lanzi EM (1992) Molecular monitoring of the myl/retinoic acid receptor-alpha fusion gene in acute promyelocytic leukemia by polymerase chain reaction. Blood 80: 492−497

Di Croce L, Raker VA, Corsaro M, Fazi F, Fanelli M, Faretta M, Fuks F, Lo Coco F, Kouzarides T, Nervi C, Minucci S, Pelicci PG (2002) Methyltransferase recruitment and DNA hypermethylation of target promoters by an oncogenic transcription factor. Science 295: 1079−1082 | Article |

Ferrara FF, Fazi F, Bianchini A, Padula F, Gelmetti V, Minucci S, Mancini M, Pelicci PG, Lo Coco F, Nervi C (2001) Histone deacetylase targeted treatment restores retinoic acid signaling and differentiation in acute myeloid leukemia. Cancer Res 61: 2−7

Gelmetti V, Zhang J, Fanelli M, Minucci S, Pelicci PG, Lazar MA (1998) Aberrant recruitment of the nuclear receptor corepressor−histone deacetylase complex by the acute myeloid leukemia fusion partner ETO. Mol Cell Biol 18: 7185−7192

Glass CK, Rosenfeld MG (2000) The coregulator exchange in transcriptional functions of nuclear receptors. Genes Dev 14: 121−141

Grignani F, De Matteis S, Nervi C, Tomassoni L, Gelmetti V, Cioce M, Fanelli M, Ruthardt M, Ferrara FF, Zamir I, Seiser C, Grignani Fa, Lazar MA, Minucci S, Pelicci PG (1998a) Fusion proteins of the retinoic acid receptor- recruit histone deacetylase in promyelocytic leukaemia. Nature 391: 815−818 | Article |

Grignani F, Ferrucci PF, Testa U, Talamo G, Fagioli M, Alcalay M, Mencarelli A, Peschle C, Nicoletti I, Pelicci PG (1993) The acute promyelocytic leukaemia specific PML/RAR fusion protein inhibits differentiation and promotes survival of myeloid precursor cells. Cell 74: 423−429 | Article |

Grignani F, Kinsella T, Mencarelli A, Valtieri M, Riganelli D, Grignani F, Lanfrancone L, Peschle C, Nolan GP, Pelicci PG (1998b) High-efficiency gene transfer and selection of human hematopoietic progenitor cells with a hybrid EBV/retroviral vector expressing the green fluorescence protein. Cancer Res 58: 14−19

Grignani F, Testa U, Rogaia D, Ferrucci PF, Samoggia P, Pinto A, Aldinucci D, Gelmetti V, Fagioli M, Alcalay M, Seeler J, Grignani Fa, Nicoletti I, Peschle C, Pelicci PG (1996) Effects on differentiation by the promyelocytic leukemia PML/RAR protein depend on the fusion of the PML protein dimerization and RAR DNA binding domains. EMBO J 15: 4949−4958

Grignani F, Valtieri M, Gabbianelli M, Gelmetti V, Botta R, Luchetti L, Masella B, Morsilli O, Pelosi E, Samoggia P, Pelicci PG, Peschle C (2000) PML/RAR alpha fusion protein expression in normal human hematopoietic progenitors dictates myeloid commitment and the promyelocytic phenotype. Blood 96: 1531−1537

He LZ, Guidez F, Tribioli C, Peruzzi D, Ruthardt M, Zelent A, Pandolfi PP (1998) Distinct interactions of PML-RARalpha and PLZF-RARalpha with co-repressors determine differential responses to RA in APL. Nat Genet 18: 126−135 | Article |

Hodel MR, Corbett AH, Hodel AE (2001) Dissection of a nuclear localization signal. J Biol Chem 276: 1317−1325 | Article |

Hong SH, Yang Z, Privalsky ML (2003) Arsenic trioxide is a potent inhibitor of the interaction of SMRT corepressor with its transcription factor partners, including the PML-retinoic acid receptor alpha oncoprotein found in acute promyelocytic leukemia. Mol Cell Biol 21: 7172−7182 | Article |

Hu X, Lazar MA (1999) The CoRNR motif controls the recruitment of corepressors by nuclear hormone receptors. Nature 402: 93−96 | Article |

Imai Y, Kurokawa M, Yamaguchi Y, Izutsu K, Nitta E, Mitani K, Satake M, Noda T, Ito Y, Hirai H (2004) The corepressor mSin3A regulates phosphorylation-induced activation, intranuclear location, and stability of AML1. Mol Cell Biol 24: 1033−1043 | Article |

Jain M, Arvantitis C, Chu K, Dewey W, Leonhardt E, Trinh M, Sunderberg CD, Bishop JM, Felsher DW (2003) Sustained loss of a neoplastic phenotype by brief inactivation of MYC. Science 297: 104

Kamashev D, Vitoux D, De The H (2004) PML-RARA-RXR oligomers mediate retinoid and rexinoid/cAMP cross-talk in acute promyelocytic leukemia cell differentiation. J Exp Med 199: 1163−1174 | Article |

Kastner P, Perez A, Lutz Y, Rochette-Egly C, Gaub MP, Durand B, Lanotte M, Berger R, Chambon P (1992) Structure, localization and transcriptional properties of two classes of retinoic acid receptor fusion proteins in acute promyelocytic leukemia (APL): structural similarities with a new family of oncoproteins. EMBO J 11: 629−642

Kilic U, Kilic E, Dietz GP, Bahr M (2003) Intravenous TAT-GDNF is protective after focal cerebral ischemia in mice. Stroke 34: 1304−1310 | Article |

Kogan SC, Hong SH, Shultz DB, Privalsky ML, Bishop JM (2000) Leukemia initiated by PMLRARalpha: the PML domain plays a critical role while retinoic acid-mediated transactivation is dispensable. Blood 95: 1541−1550

Lane AA, Ley TJ (2003) Neutrophil elastase cleaves PML-RARalpha and is important for the development of acute promyelocytic leukemia in mice. Cell 115: 305−318 | Article |

Lanotte M, Martin-Thouvenin V, Najman S, Ballerini P, Valensi F, Berger R (1991) NB4, a maturation inducible cell line with t(15;17) marker isolated from a human acute promyelocytic leukemia (M3). Blood 77: 1080−1086

Lin RJ, Evans RM (2000) Acquisition of oncogenic potential by RAR chimeras in acute promyelocytic leukemia through formation of homodimers. Mol Cell 5: 821−830 | Article |

Lin RJ, Nagy L, Inoue S, Shao W, Miller WHJ, Evans RM (1998) Role of the histone deacetylase complex in acute promyelocytic leukaemia. Nature 391: 811−814 | Article |

Linggi B, Muller-Tidow C, van de Locht L, Hu M, Nip J, Serve H, Berdel WE, van der Reijden B, Quelle DE, Rowley JD, Cleveland J, Jansen JH, Pandolfi PP, Hiebert SW (2002) The t(8;21) fusion protein, AML1 ETO, specifically represses the transcription of the p14(ARF) tumor suppressor in acute myeloid leukemia. Nat Med 8: 743−750 | Article |

Look AT (1997) Oncogenic transcription factors in the human acute leukemias. Science 278: 1059−1064 | Article |

Lutterbach B, Westendorf JJ, Linggi B, Isaac S, Seto E, Hiebert SW (2000) A mechanism of repression by acute myeloid leukemia-1, the target of multiple chromosomal translocations in acute leukemia. J Biol Chem 275: 651−656 | Article |

Lutterbach B, Westendorf JJ, Linggi B, Patten A, Moniwa M, Davie JR, Huynh KD, Bardwell VJ, Lavinsky RM, Rosenfeld MG, Glass C, Seto E, Hiebert SW (1998) ETO, a target of t(8;21) in acute leukemia, interacts with the N-CoR and mSin3 corepressors. Mol Cell Biol 18: 7176−7184

Matozaki S, Nakagawa T, Kawaguchi R, Aozaki R, Tsutsumi M, Murayama T, Koizumi T, Nishimura R, Isobe T, Chihara K (1995) Establishment of a myeloid leukemia cell line (SKNO-1) from a patient with t(8;21) who acquired monosomy 17 during disease progression. Br J Haematol 89: 805−811

Melnick A, Licht JD (1999) Deconstructing a disease: RAR, its fusion partners, and their roles in the pathogenesis of acute promyelocytic leukemia. Blood 93: 3167−3215

Minucci S, Maccarana M, Cioce M, De Luca P, Gelmetti V, Segalla S, Di Croce L, Giavara S, Matteucci C, Gobbi A, Bianchini A, Colombo E, Schiavoni I, Badaracco G, Hu X, Lazar MA, Landsberger N, Nervi C, Pelicci PG (2000) Oligomerization of RAR and AML1 transcription factors as a novel mechanism of oncogenic activation. Mol Cell 5: 811−820 | Article |

Minucci S, Monestiroli S, Giavara S, Ronzoni S, Marchesi F, Insinga A, Diverio D, Gasparini P, Capillo M, Colombo E, Matteucci C, Contegno F, Lo Coco F, Scanziani E, Gobbi A, Pelicci PG (2002) PML-RAR induces promyelocytic leukemias with high efficiency following retroviral gene transfer into purified murine hematopoietic progenitors. Blood 100: 2989−2995 | Article |

Nagy L, Kao HY, Love JD, Li C, Banayo E, Gooch JT, Krishna V, Chatterjee K, Evans RM, Schwabe JW (1999) Mechanism of corepressor binding and release from nuclear hormone receptors. Genes Dev 13: 3209−3216 | Article |

Nervi C, Ferrara FF, Fanelli M, Rippo MR, Tomassini B, Ferrucci PF, Ruthardt M, Gelmetti V, Gambacorti-Passerini C, Diverio D, Grignani F, Pelicci PG, Testi R (1998) Caspases mediate retinoic acid induced degradation of the acute promyelocytic leukemia PML-RAR fusion protein. Blood 92: 2244−2251

Nervi C, Poindexter EC, Grignani F, Pandolfi PP, Lo Coco F, Avvisati G, Pelicci PG, Jetten AM (1992) Characterization of the PML-RAR chimeric product of the acute promyelocytic leukemia specific t(15;17) translocation. Cancer Res 52: 3687−3692

Perissi V, Aggarwal A, Glass CK, Rose DW, Rosenfeld MG (2004) A corepressor/coactivator exchange complex required for transcriptional activation by nuclear receptors and other regulated transcription factors. Cell 116: 511−526 | Article |

Perissi V, Staszewski LM, McInerney EM, Kurokawa R, Krones A, Rose DW, Lambert MH, Milburn MV, Glass CK, Rosenfeld MG (1999) Molecular determinants of nuclear receptor−corepressor interaction. Genes Dev 13: 3198−3208 | Article |

Polo JM, Dell'oso T, Ranuncolo SM, Cerchietti L, Beck D, Da Silva GF, Prive GG, Licht JD, Melnick A (2004) Specific peptide interference reveals BCL6 transcriptional and oncogenic mechanisms in B-cell lymphoma cells. Nat Med 10: 1329−1335 | Article |

Rabbitts TH, Stocks MR (2003) Chromosomal translocation products engender new intracellular therapeutics technologies. Nat Med 9: 383−386 | Article |

Raelson JV, Nervi C, Rosenauer A, Benedetti L, Monczak Y, Pearson M, Pelicci PG, Miller WHJ (1996) The PML/RAR oncoprotein is a direct molecular target of retinoic acid in acute promyelocytic leukemia cells. Blood 88: 2826−2832

Schwarze SR, Ho A, Vocero-Albani A, Dowdy SF (1999) In vivo protein transduction: delivery of a biologically active protein into the mouse. Science 285: 1569−1572 | Article |

Schwarze SR, Hruska KA, Dowdy SF (2000) Protein transduction: unrestricted delivery into all cells? Trends Cell Biol 10: 290−295 | Article |

Schwieger M, Lohler J, Friel J, Scheller M, Horak I (2002) AML1-ETO inhibits maturation of multiple lymphohematopoietic lineages and induces myeloblast transformation in synergy with ICSBP deficiency. J Exp Med 196: 1227−1240 | Article |

Seeler JS, Dejean A (2003) Nuclear and unclear functions of SUMO. Nat Rev Mol Cell Biol 4: 690−699 | Article |

Shao W, Benedetti L, Lamph WW, Nervi C, Miller WHJ (1997) A retinoid-resistant acute promyelocytic leukemia subclone expresses a dominant negative PML-RAR mutation. Blood 89: 4282−4289

Tenen DG (2003) Disruption of differentiation in human cancer: AML shows the way. Nat Rev Cancer 3: 89−101 | Article |

Vocero-Albani A, Chellaiah MA, Hruska KA, Dowdy SF (2001) Protein transduction: delivery of Tat-GTPase fusion proteins into mammalian cells. Methods Enzymol 332: 36−49

Wang J, Hoshino T, Redner RL, Kajigaya S, Liu JM (1998) ETO, fusion partner in t(8;21) acute myeloid leukemia, represses transcription by interaction with the human N-CoR/mSin3/HDAC1 complex. Proc Natl Acad Sci USA 95: 10860−10865 | Article |

Warrell RPJ (1993) Retinoid resistance in acute promyelocytic leukemia: new mechanisms, strategies, and implications. Blood 82: 1949−1953

Weinstein IB (2002) Addiction to oncogenes—the Achilles heal of cancer. Science 297: 63−64 | Article |

Yoshida H, Kitamura K, Tanaka K, Omura S, Miyazaki T, Hachiya T, Ohno R, Naoe T (1996) Accelerated degradation of PML-retinoic acid receptor (PML-RARA) oncoprotein by all-trans-retinoic acid in acute promyelocytic leukemia: possible role of the proteasome pathway. Cancer Res 38: 2945−2948

Zhu J, Gianni M, Kopf E, Honore N, Chelbi-Alix MK, Koken M, Quignon F, Rochette-Egly C, de Thè H (1999) Retinoic acid induces proteasome-dependent degradation of retinoic acid receptor alpha (RARalpha) and oncogenic RARalpha fusion proteins. Proc Natl Acad Sci USA 96: 14807−14812 | Article |

Zhu J, Lallemand-Breitenbach V, de Thè H (2001) Pathways of retinoic acid- or arsenic trioxide-induced PML/RARalpha catabolism, role of oncogene degradation in disease remission. Oncogene 20: 7257−7265 | Article |