The MLL fusion gene, MLL-AF4, regulates cyclin-dependent kinase inhibitor CDKN1B (p27kip1) expression (original) (raw)

Proc Natl Acad Sci U S A. 2005 Sep 27; 102(39): 14028–14033.

Department of Medicine, Molecular Biology Program, and Oncology Institute, Cardinal Bernardin Cancer Center, Loyola University Medical Center, Maywood, IL 60153

* To whom correspondence should be addressed at: Cardinal Bernardin Cancer Center, Loyola University Medical Center, 2160 South First Avenue, 112-337, Maywood, IL 60153. E-mail: ude.cmul@nzelezn.

Communicated by Janet D. Rowley, University of Chicago Medical Center, Chicago, IL, July 29, 2005

Copyright © 2005, The National Academy of Sciences

Abstract

MLL, involved in many chromosomal translocations associated with acute myeloid and lymphoid leukemia, has >50 known partner genes with which it is able to form in-frame fusions. Characterizing important downstream target genes of MLL and of MLL fusion proteins may provide rational therapeutic strategies for the treatment of MLL-associated leukemia. We explored downstream target genes of the most prevalent MLL fusion protein, MLL-AF4. To this end, we developed inducible MLL-AF4 fusion cell lines in different backgrounds. Overexpression of MLL-AF4 does not lead to increased proliferation in either cell line, but rather, cell growth was slowed compared with similar cell lines inducibly expressing truncated MLL. We found that in the MLL-AF4-induced cell lines, the expression of the cyclin-dependent kinase inhibitor gene CDKN1B was dramatically changed at both the RNA and protein (p27kip1) levels. In contrast, the expression levels of CDKN1A (p21) and CDKN2A (p16) were unchanged. To explore whether CDKN1B might be a direct target of MLL and of MLL-AF4, we used chromatin immunoprecipitation (ChIP) assays and luciferase reporter gene assays. MLL-AF4 binds to the CDKN1B promoter in vivo and regulates CDKN1B promoter activity. Further, we confirmed CDKN1B promoter binding by ChIP in MLL-AF4 as well as in MLL-AF9 leukemia cell lines. Our results suggest that CDKN1B is a downstream target of MLL and of MLL-AF4, and that, depending on the background cell type, MLL-AF4 inhibits or activates CDKN1B expression. This finding may have implications in terms of leukemia stem cell resistance to chemotherapy in MLL-AF4 leukemias.

Keywords: leukemia, acute lymphoblastic leukemias, cell cycle

MLL is involved in chromosomal translocations associated with leukemia. Remarkably, MLL is involved in translocations with >50 different genes (1, 2). MLL is specifically cleaved shortly after translation into two peptides that noncovalently associate with each other (3, 4). The amino-terminal portion of MLL contains a region with AT-hooks that binds DNA, as well as a region with transcriptional repression activity (5) that binds CpG-rich DNA (6) and recruits histone deacetylases, the corepressor CtBP1, and polycomb group proteins (7). The carboxyl-terminal portion contains a transcriptional activation domain (5), which interacts with CBP (2), and a SET domain, with histone methyltransferase activity (3, 8). Different MLL fusion partners are associated with leukemias producing blast cells of various lineages. MLL-AF9 results mainly in acute myeloid leukemia (AML), whereas MLL-AF4 causes almost exclusively B-cell lineage acute lymphoblastic leukemias. These findings suggest that MLL chimeras affect the phenotype of the leukemia by influencing differentiation pathways of uncommitted cells or early progenitors. MLL-AF4, an MLL fusion protein that is associated with infant pro-B acute lymphoblastic leukemias, is the most prevalent of the numerous MLL fusion proteins (9), and it is usually associated with a poor prognosis (10). Numerous data show that MLL fusion genes can transform hematopoietic cells in vitro and cause leukemia in vivo (11, 12). Recent studies suggest some potential mechanisms of MLL fusion protein leukemogenesis. For example, fusion partner dimerization domains and/or activation domains fused to MLL can aberrantly activate downstream targets such as HOX genes and contribute to cell transformation (13, 14). This regulation is mediated at the level of target gene transcription. There is very strict regulation of HOX gene expression during hematopoiesis, therefore misregulated expression of these genes is likely important in MLL leukemogenesis.

During normal hematopoiesis, a tight balance is required between levels of mostly quiescent stem cells that can renew the population and highly proliferating progenitor cells, before final differentiation along a particular lineage. This balance is regulated through cell cycle regulators. Cyclins and cyclin-dependent kinases (CDKs) play important roles in this process (15). CDKs are opposed by CDK inhibitors (CDKIs) (16, 17). There are two related families of CDKIs: (i) the Cip/Kip family (p21, p27, and p57), which inhibits CDK2 and CDK1-containing complexes (cyclin A/E–CDK2, cyclin B–CDK1) and (ii) the INK4 family (p15, p16, p18, and p19), which inhibits cyclin D-containing complexes (cyclin D–CDK4/6). Expression of CDKIs generally causes growth arrest and, when acting as tumor suppressors, may cause cell cycle arrest and apoptosis. Numerous studies have also shown that CDKIs accumulate during cell differentiation (18, 19). However, the CDKIs do not act similarly in all cell lineages. For example, in the myeloid lineage, expression of p27kip1 was required for differentiation of progenitors along this lineage, and p27-deficient marrow accumulated progenitor cells (20). In contrast, in the lymphoid lineage, expression of p27 inhibits differentiation, and p27 expression must be decreased for normal T cell development (21).

The MLL fusion genes can cause leukemia in vivo (11, 12); however, the mechanism is unclear. It has been proposed that the fusions block hematopoietic cell differentiation and reduce cell death, thus contributing to leukemogenesis. Identification of additional gene targets of MLL-AF4 regulation may allow the design of rational therapeutic strategies. To identify potential direct targets of MLL-AF4, we generated cell lines with an inducible MLL-AF4 transgene.

In an epithelial cell background, we observed that MLL-AF4 down-regulated the CDKI p27 but not p21 or p16. Down-regulation of p27 occurred at both the RNA (CDKN1B) and protein levels. Furthermore, chromatin immunoprecipitation (ChIP) assays indicated that MLL-AF4 binds to the CDKN1B promoter in vivo, and reporter gene assays show that it represses transcription of the CDKN1B promoter in an epithelial cell line. Similarly, in a lymphoid cell background and in primary bone marrow progenitor cells, MLL-AF4 also regulates CDKN1B expression, but in this case it is up-regulated. Our results suggest that MLL-AF4 regulates CDKN1B expression directly, but that the outcome of this regulation depends on the cell type.

Materials and Methods

Expression Plasmids. MLL(672) was generated by digestion of pEGFP-MLL2Kb(22) with KpnI and ligating into pcDNA5/FRT/TO (Invitrogen). MLL(1250) was generated by digestion of MSCVneo-MLL-CBP(12) with PseI/BamHI, followed by ligation of the MLL fragment into pcDNA5/FRT/TO-MLL(672). MLL-AF4 was generated by digestion of MSCVneo-MLL-AF4 (N.J.Z.-L., unpublished data) with BamHI, and ligation of AF4 into pcDNA5/FRT/TO-MLL(1250). The constructs were confirmed by sequencing.

Establishment of Cell Lines Expressing MLL and MLL-AF4 Proteins. By using the Flp-In T-REx 293 host cell line (Invitrogen), 9 μg of expression plasmid pOG44 was electroporated along with 1 μg of either pcDNA5/FRT/TO [MLL-AF4, MLL(672), MLL(1250), and MLL] or pCMV5/RPT/TO (vector alone). Hygromycin-resistant clones were obtained under 100 μg/ml hygromycin selection. The cell lines used were β-galactosidase negative and zeocin sensitive and showed inducible expression by RT-PCR and Western blotting. To generate inducible cell lines in the Jurkat background, Flp-In Jurkat cells (Invitrogen) were used to generate a Flp-In T-REx Jurkat cell line, and inducible MLL-AF4 Jurkat lines were developed in a similar manner.

Cell Culture and Maintenance. Stably expressing 293 cell lines were maintained in Dulbecco's modified Eagle's medium (DMEM) plus 10% FCS, 15 μg/ml blasticidin, and 100 μg/ml hygromycin. Stably expressing Jurkat cell lines were maintained in RPMI medium 1640 with the same additives, except with 200 μg/ml hygromycin. Cell lines were induced with 1 μg/ml tetracycline. Leukemia cell lines were maintained in RPMI medium 1640 plus 10% FCS. Murine embryonic fibroblasts (MEFs) were cultured as described in ref. 23. _Mll_-/- MEFs were transfected with MLL or MLL-AF4-expressing plasmids and cultured 4 weeks with 200 μg/ml hygromycin to create stable populations.

Cell Growth, Cell Cycle, and Apoptosis Assays. For growth rate analysis, cells were plated at 1 × 105 cells with or without 1 μg/ml tetracycline. Viable and dead cells were assessed by counting with trypan blue exclusion at 24, 48, 72, 96, 120, and 144 h. For propidium iodide (PI) staining, ≈1 × 106 cells were washed in PBS, fixed in 4:1 (ice-cold methanol/PBS), incubated in PI (5 μg/ml PI in PBS) with 100 μg/ml RNase A at 37°C for 1 h, and analyzed on a Becton Dickinson FACSCalibur with cellquest analysis software. Annexin V-FITC apoptotic detection (catalogue no. 556547; BD Pharmingen) was according to the manufacturer's protocol.

Immunoprecipitation and Western Blot Analysis. FLAG-MLL proteins were expressed by tetracycline induction for 24 h. Cells were lysed in IPH buffer [50 mM Tris·HCl, pH 8.0/150 mM NaCl/5 mM EDTA/0.5% NP-40/10 μl/ml protease inhibitor mix (Sigma)], immunoprecipitated on anti-FLAG beads as described in ref. 7, and washed four times with NETN buffer (20 mM Tris·HCl, pH 8.0/100 mM NaCl/1 mM EDTA/0.5% Nonidet P-40). Proteins were resolved by SDS/PAGE and detected with an enhanced chemiluminescence (ECL) kit (Amersham Pharmacia) according to the manufacturer's protocols. Rabbit anti-MLL antibodies, anti-MLL-AT-hook, anti-MLL-RD, and anti-MLL-AD, are against MLL amino acids 310–400, 1101–1400, and 2771–3114, respectively. Preimmune sera from the same rabbits were used as negative controls. For Western blot analysis of p21 and p16, cells were lysed in hypotonic lysis buffer (HLB; 20 mM Hepes, pH 7.5/10 mM KCl/0.5 mM EDTA/0.1% Triton X-100), followed by HLB/500 mM NaCl. Each buffer contained 10 μlof100× protease inhibitors (catalogue no. P 8340; Sigma), 4 μl of 1 M DTT, 2 μl of phosphatase inhibitor (catalogue no. P 5726; Sigma), and 2 μl of 1 M NaF per ml. Proteins were electrophoresed on SDS/10% or 5% polyacrylamide gels and transferred to polyvinylidene difluoride membrane. Membranes were blocked with 5% milk and were incubated with polyclonal rabbit anti-p21 (sc-397) or polyclonal rabbit anti-p27 (sc-528) (1:1000) (Santa Cruz Biotechnology). Detection was performed with peroxidase-conjugated anti-rabbit Ig (1:3,000) (Amersham Pharmacia) and ECL detection as above. Membranes were stripped and rehybridized as described in ref. 7.

RNA Extraction, RT-PCR, and Real-Time RT-PCR. Total RNA was isolated with TRI Reagent (catalogue no. T 9424; Sigma) and converted to cDNA by using Invitrogen Superscript First Strand Synthesis system. CDKN2A, CDKN1A, and CDKN1B primers were used as described in refs. 24–26. PCR was performed as follows: 94°C for 1 min, 55°C for 1 min, and 72°C for 1 min for 30 cycles, then a final 72°C 10-min extension. The MLL-AF4 primers used were top, 5′-CACCTACTACAGGACCGCCAA-3′; and bottom, 5′-GGGGTTTGTTCACTGTCACTGTCC-3′. PCR was performed as follows: 94°C for 50 sec, 60°C for 50 sec, and 72°C for 1 min for 35 cycles. For quantitative RT-PCR, cDNA was prepared from MLL-AF4 induced and uninduced 293 and Jurkat cell lines and from murine primary bone marrow progenitors infected with MSCVneo-MLL-AF4 or MSCVneo retrovirus and cultured in methylcellulose under G418 selection for 1 week as described in ref. 12. All reactions were performed in triplicate, and CDKN1B expression level was measured by using SYBR green reagents and the Applied Biosystems Prism 5700 sequencer. The relative CDKN1B expression level was determined by normalization to murine (for bone marrow cells) or human (for 293 and Jurkat cells) GAPDH. Primer sequences are available on request.

Luciferase Assays. We seeded 1.0 × 105 cells in six-well plates and transfected empty vector, MLL, and MLL-AF4 cell lines with CDKN1B promoter (-3568 to -12) (0.9 μg) (27), CDKN2A promoter (-2000 to +41 or -970 to -164) (0.9 μg) (M.O.D., unpublished data) linked to the firefly luciferase gene or basic vector (0.9 μg) plus pTKRL(Renilla) (100 ng), with Fugene6 (Roche) according to the manufacturer's protocol. After 48 h, cells were harvested. Dual-Luciferase Reporter assay system (Promega) was used according to manufacturer's protocol.

ChIP Assay. ChIP was performed with Upstate Biotechnology ChIP assay kit (catalogue no. 17-295) following manufacturer's protocol with some modifications. Either uninduced or induced 3 × 106 cells were crosslinked with 1% formaldehyde for 5 min at room temperature, and the reaction was terminated with an excess of glycine. Chromatin was sonicated to an average size of 400 bp and was immunoprecipitated with MLL-specific polyclonal antibodies or preimmune serum (28), or with anti-FLAG antibody (catalogue no. F 3165; Sigma). ChIP DNA was detected by using PCR and ethidium bromide staining after agarose gel electrophoresis. The CDKN1B DNA fragments amplified promoter and coding regions, spanning nucleotides 161–340 and 717–974, respectively (GenBank accession no. AF480891).

Results

MLL and MLL-AF4 Expression in Conditional Cell Lines. To characterize MLL fusion downstream target gene expression, we generated stable cell lines that conditionally expressed FLAG-tagged MLL-AF4, by using the Invitrogen Flp-In system (see Materials and Methods) in 293 human kidney epithelial and Jurkat lymphoblast cell lines. We also generated cell lines based on the Flp-In T-REx 293 cell line that conditionally express FLAG-MLL(672), FLAG-MLL(1250), or full-length FLAG-MLL, or contain the empty expression vector (Fig. 1_A_). MLL expression was confirmed by Western blotting, and MLL-AF4 expression was confirmed by RT-PCR (data not shown) and Western blotting (Fig. 1_B_). MLL-AF4 RNA is expressed from 2 to 48 h after tetracycline induction as determined by RT-PCR (data not shown). MLL-AF4 protein is observed by 12 h after induction and remains detectable 48 h after induction (data not shown). All protein expression was detectable in the nuclear fraction or by immunoprecipitation from whole cell extract. The FLAG full-length MLL and MLL-AF4 proteins are detectable from whole cell lysates only after immunoprecipitation in these cell lines because even the induced levels of expression are quite low. We also confirmed that MLL-AF4 localized in the nucleus by immunostaining (data not shown). We found that cell growth slowed after induction of MLL-AF4, but not after induction of MLL(1250) (Fig. 1_C_ Upper) or MLL(672) (data not shown), relative to the empty vector control in the epithelial cell background and in the Jurkat lymphoid cells (Fig. 1_C_ Lower). However, this slower growth was not due to increased apoptosis in either cell line (Fig. 1 D and E and data not shown).

An external file that holds a picture, illustration, etc. Object name is zpq0370594900001.jpg

Creation and characterization of conditional MLL and MLL-AF4 cell lines. (A) FLAG-tagged MLL and MLL-AF4 fusion proteins expressed in conditional cell lines: full-length MLL (Top), amino portions of MLL (Middle), and MLL-AF4 (Bottom). Numbers refer to amino acid number. MLL-AF4 retains the AT-hook and repression domain of MLL, plus amino acids 347-1221 of AF4. AT, AT hooks; RD1 and RD2, repression domains; PHD, plant homeodomain; AD, activation domain; CS, proteolytic cleavage site; SET, conserved domain with methyltransferase activity. (B) FLAG full-length MLL (a_–_c), FLAG-MLL-AF4 (c, d, and f), or two different amino-terminal FLAG-MLL proteins (e) were expressed in clonally selected stably transfected 293 cell lines or Jurkat cell lines (f) after induction with tetracycline (+), but not without tetracycline (-) for 24 h. After immunoprecipitation with anti-FLAG beads, proteins were detected with anti-FLAG antibodies (a, d, e, and f), anti-MLL activation domain antibodies (b), or anti-AT plus anti-RD antibodies (c). (C) MLL-AF4 inhibits growth of inducible 293 and Jurkat cell lines. Cells were cultured in the absence (-Tet) or presence (+Tet) of 1 μg/ml tetracycline, and cell numbers were counted every 24 h. Results are presented as mean (±SD) of three independent experiments. (D) MLL-AF4 expression does not affect apoptosis and cell cycle. Annexin and PI staining show that the cells were not significantly more apoptotic in 293 cell line with induced MLL-AF4 expression. (E) The relative percent cells in each stage of the cell cycle was not changed after induced MLL-AF4 expression in 293 and Jurkat (data not shown)

MLL-AF4 Represses Endogenous CDKN1B (p27) Expression, but Not CDKN2A (p16) or CDKN1A (p21), in Epithelial Cells. CDKIs are important in regulation of cell cycle transit. Because induced expression of MLL-AF4 caused the cells to grow more slowly, we determined whether MLL-AF4 alters expression of CDKIs in these cell lines. We used inducible MLL(672), MLL(1250), full-length MLL, and MLL-AF4 epithelial cell lines to compare their relative effect on expression of CDKIs. CDKN1B shows down-regulation at both the mRNA level (Fig. 2_A_, lane 6, and B and C) and protein (p27) level (Fig. 2_D_, lane 1) after induced expression of MLL-AF4 compared with cells inducibly expressing MLL(672), MLL(1250), or full-length MLL. CDKN1A (Fig. 2_A_, lane 4) and CDKN2A (Fig. 2_A_, lane 2) levels were not affected by induction of MLL-AF4 in this cell line. Because p27 targets CDK1/cyclin B and CDK2/cyclin A or cyclin E complexes to regulate the cell cycle, over-expression of p27 usually causes cell cycle arrest at the G0/G1 phase or causes cell growth to slow (29). Because we showed that MLL-AF4 down-regulates CDKN1B, we wanted to test whether MLL-AF4 expression could alter the percentage of cells in any phase of the cell cycle. We performed cell-cycle analysis by PI staining and determined that the percentage of cells in each stage of the cell cycle does not seem to be significantly affected after induction of MLL-AF4 at time points from 2 to 16 h (Fig. 1 D and E) and also through 6 days (data not shown). These data suggested that p27 down-regulation by MLL-AF4 does not result in cell-cycle arrest at a specific phase in this cell line.

An external file that holds a picture, illustration, etc. Object name is zpq0370594900002.jpg

MLL-AF4 decreases p27 expression in 293 epithelial and MEF cells. (A) MLL-AF4 expression decreases CDKN1B RNA in 293 cells. Shown is RT-PCR for CDKN2A (lanes 1 and 2), CDKN1A (lanes 3 and 4), and CDKN1B (lanes 5 and 6), in conditionally expressed MLL-AF4 293 cell line with (+) or without (-) tetracycline. (B) Semiquantitative RT-PCR was used to determine changes of relative expression in CDKN2A, CDKN1A, and CDKN1B, normalized to GAPDH. The relative CDKN1B expression (CDKN1B/GAPDH) was significantly reduced after induction of MLL-AF4 compared with noninduced. Reduced expression was not seen with CDKN2A or with CDKN1A. The data were derived from at least three independent experiments (mean ± SD). (C) CDKN1B was decreased 8-fold by MLL-AF4-expressing 293 cells as compared with uninduced 293 cells by quantitative RT-PCR. CDKN1B expression levels were normalized to GAPDH expression and performed in triplicate. (D) p27kip1 protein levels are reduced after MLL-AF4 expression in 293 cells. p27 protein expression was reduced 24 h after induction of MLL-AF4 (lane 1) in 293 cells compared with the noninduced cells (lane 2). p27 protein levels were not changed with induced expression of MLL(672), MLL(1250) clones 1 and 2, or full-length MLL (lanes 3–10). The upper arrow indicates loading control. (E) p27 protein level was reduced in _Mll_-/- MEFs compared with wild-type Mll+/+ MEFs (upper arrow). p21 protein levels were not changed (lower arrow). (F) MLL rescues CDKN1B expression, and MLL-AF4 further reduces CDKN1B expression in murine _Mll_-/- MEFs. RNA was isolated from _Mll_-/- MEFs (lane 1) or from _Mll_-/- MEFs transfected with MLL-AF4-expressing (lane 2) or MLL-expressing (lane 3) plasmids and grown under hygromycin selection. Data are presented as fold-change in CDKN1B expression levels relative to untransfected _Mll_-/- MEFs as determined by quantitative RT-PCR normalized to GAPDH and performed in triplicate

MLL-AF4 Represses CDKN1B Promoter Activity in an Epithelial Cell Background. Because induction of MLL-AF4 results in decreased levels of CDKN1B mRNA, we tested whether this repression occurs through the CDKN1B promoter by using luciferase reporter gene assays. 293 stable cell lines, inducible for full-length MLL or MLL-AF4, or containing empty vector, were transfected with a firefly luciferase reporter plasmid containing the CDKN1B promoter, the CDKN2A promoter, or basic vector, plus Renilla expression construct (for normalization). MLL or MLL-AF4 expression was induced with tetracycline 24 h after transfection. MLL-AF4 expression repressed activity of the CDKN1B promoter (Fig. 3_A_) but not the CDKN2A promoter (Fig. 3_B_). In contrast, empty vector control and full-length MLL had no significant effect on CDKN1B promoter activity in these cells (Fig. 3_A_).

An external file that holds a picture, illustration, etc. Object name is zpq0370594900003.jpg

MLL-AF4 represses CDKN1B, but not CDKN2A, promoter activity in 293 epithelial cells. Reporter constructs basic pGL3, pGL3-CDKN1B (-3568 to -12) (A), pGL3–2kb_CDKN2A_, or pGL3–0.8kb_CDKN2A_ (B), were cotransfected with pTKRL (Renilla) as internal control, in MLL-AF4-, MLL-, or vector-inducible 293 cell lines. Luciferase activity was detected in the absence (dark bars) or presence (light bars) of tetracycline. The data were normalized to the internal control and the basic vector. The data were from at least three independent experiments (mean ± SD)

p27 Expression Is Regulated by Wild-Type MLL. The MLL protein domains responsible for target DNA binding, including the AT hooks and CXXC motifs present in the repression domain, as well as nuclear targeting sequences, are all present in the amino-terminal portion of the protein. These domains are present in wild-type MLL, as well as in all MLL fusions, including MLL-AF4. Because MLL-AF4 could regulate p27 expression acting through the promoter, we hypothesized that wild-type MLL might also regulate p27 expression. MEFs, either wild type (+/+) or mutant (-/-) for Mll (23) were assessed for p27 protein expression levels (Fig. 2_E_). We found an _Mll_-dependent effect on p27 expression, whereas p21 protein levels were unchanged. Surprisingly, however, the absence of wild-type Mll decreased p27 protein levels, whereas in the 293 epithelial cells, induced MLL-AF4 expression decreased p27 expression. This observation was also substantiated by microarray analysis reported by others using these same cells, where RNA levels for CDKN1B were significantly higher in Mll+/+ versus _Mll_-/- MEFs (30). Furthermore, CDKN1B/p27 expression was rescued in _Mll_-/- MEFs stably transfected with MLL, whereas MLL-AF4 further reduced CDKN1B expression in these cells as measured by quantitative RT-PCR (Fig. 2_F_).

MLL-AF4 Induces p27 Expression in Lymphoid Cells. The MLL-AF4 translocation almost always results in acute leukemia with lymphoid phenotype. Therefore, it is relevant to determine whether MLL-AF4 regulates p27 expression in lymphoid cells. After induction of MLL-AF4 expression in Jurkat cells, p27 RNA and protein expression were increased (Fig. 4 A and B). This finding was corroborated by Affymetrix microarray hybridization of MLL-AF4-vs. control-induced Jurkat RNA, where p27 was identified as one of the most significantly increased genes (data not shown). Furthermore, in two cell lines derived from patients with the MLL-AF4 translocation, MV4–11 and RS4;11, p27 was expressed at a high level, but p27 expression was at a lower level in Mono Mac 6 and THP-1, cell lines with the MLL-AF9 translocation (Fig. 4_A_). To determine whether the MLL-AF4 fusion could also regulate p27 expression in primary hematopoietic cells, we infected murine bone marrow progenitor cells with MSCVneo-MLL-AF4 retrovirus and analyzed cells after 1 week under G418 selection. CDKN1B expression was increased 2-fold compared with control by quantitative RT-PCR (Fig. 4_C_).

An external file that holds a picture, illustration, etc. Object name is zpq0370594900004.jpg

p27 expression is increased in MLL-AF4-expressing hematopoietic cells. (A) p27 protein expression is increased in MLL-AF4-expressing hematopoietic cell lines. (Upper) Western blot analysis shows that induced expression of MLL-AF4 results in increased p27 expression in Jurkat cells (compare lanes 1 and 2) and that p27 is increased in human patient-derived leukemia cell lines expressing the MLL-AF4 translocation (MV4–11 and RS4;11) (lanes 5 and 6) but not in lines expressing the MLL-AF9 translocation (Mono Mac 6 and THP-1) (lanes 3 and 4). (Lower) Blots were stripped and rehybridized with antibody recognizing actin. (B) CDKN1B is up-regulated by MLL-AF4 expression in Jurkat cells. Semiquantitative RT-PCR for CDKN1B in MLL-AF4 Jurkat cell line compared with parental Jurkat cell line (control) induced with tetracycline. Two-fold dilutions of cDNA were used as template for RT-PCR. The relative CDKN1B expression (CDKN1B/GAPDH) was significantly increased after induction of MLL-AF4. (C) CDKN1B is up-regulated by MLL-AF4 expression in primary murine bone marrow progenitor cells. Bone marrow progenitors were infected with control (MSCVneo) or MLL-AF4-expressing (MSCVneo-MLL-AF4) retrovirus. After G418 selection for 1 week, RNA was isolated and analyzed by quantitative RT-PCR for CDKN1B expression. Expression levels were normalized to GAPDH expression and performed in triplicate. The relative CDKN1B expression was increased 2-fold after MLL-AF4 expression

MLL-AF4 Binds the Endogenous CDKN1B Promoter in 293 Cells and in MLL-Fusion Leukemia Cell Lines. MLL-AF4 is able to regulate CDKN1B promoter activity (Fig. 3) and to affect the levels of endogenous CDKN1B RNA and protein (Figs. ​2 and ​4). These findings suggest that CDKN1B might be a direct transcriptional target of MLL-AF4 and of MLL. Therefore, we performed ChIP assays by using inducible MLL-AF4 cells and two cell lines developed from patients with MLL-AF4 translocation leukemias (31) to determine whether MLL-AF4 was bound to the CDKN1B promoter in vivo. After crosslinking, chromatin was sheared to an average size of 400 bp and immunoprecipitated with antibodies against the amino-terminal portion of MLL (AT and RD) (Fig. 1_Bc_) (28) or with anti-FLAG antibodies. MLL-AF4 binds to the CDKN1B promoter (Fig. 5_A_) in 293 cells induced to express MLL-AF4; however, MLL-AF4 did not bind to the CDKN1B coding region in the MLL-AF4-induced cell line (Fig. 5_A_). Anti-FLAG antibody did not show binding, but this may be because the FLAG epitope was not accessible in vivo. In similar experiments with the leukemia cell lines, MV4–11 and RS4;11, which express endogenous MLL-AF4, we observed binding to the CDKN1B promoter, whereas no binding was observed with preimmune sera or in the absence of antibody (Fig. 5_B_). MLL binding to the CDKN1B promoter was also observed in cell lines expressing the MLL-AF9 fusion, Mono Mac 6 and THP-1. The MLL fusion proteins and/or MLL bind to the CDKN1B promoter in these cell lines because both endogenous proteins are expressed and recognized by the anti-MLL antibodies used.

An external file that holds a picture, illustration, etc. Object name is zpq0370594900005.jpg

MLL-AF4 binds to the endogenous CDKN1B promoter. ChIP assays were performed by using antibodies to MLL (AT + RD) or FLAG, or preimmune sera. Immunoprecipitated chromatin was analyzed by PCR with primers specific for either the CDKN1B promoter (A Left; B)orthe CDKN1B coding region (A Right). (A) MLL-AF4 binds to the CDKN1B promoter (Left) but not to the CDKN1B coding region (Right) in induced MLL-AF4 293 cells. (B) Endogenous MLL/MLL-AF4 binds to the CDKN1B promoter in MV4–11 and RS4;11 cells. Endogenous MLL/MLL-AF4 binds to the endogenous CDKN1B promoter in MV4–11 and RS4;11 cells (Upper). Endogenous MLL/MLL-AF9 binds to the endogenous CDKN1B promoter in Mono Mac 6 and THP-1 cells (Lower)

Discussion

Our data demonstrate that an oncogenic protein, MLL-AF4, can regulate expression of p27kip1. Interestingly, MLL-AF4 causes either a decrease or an increase in p27 expression, depending on the cell type. Furthermore, p27 expression can also be modulated by wild-type MLL. MLL has previously been shown to bind to the promoters of Hoxc8 and HOXA9, targets of MLL regulation (3, 8), but no other direct targets have previously been identified. We therefore explored whether MLL/MLL-AF4 was bound to the CDKN1B promoter. Our data indicate that MLL-AF4 binds to the CDKN1B promoter in both the inducible MLL-AF4 epithelial cell line and in patient-derived MLL fusion-expressing leukemia cell lines. While this manuscript was under revision, it was reported by another group (32) that MLL and menin cooperatively regulate expression of the CDKIs p27kip1 and p18Ink4C in murine embryonic fibroblasts. Similar to our findings, it was observed that MLL positively regulates p27 expression by directly binding to the p27 locus.

p27 is one of the CDKIs involved in cell-cycle progression and can also act as a tumor-suppressor gene. Functionally, loss of CDKN1B enhances the growth of mice (33) and results in hyperplasia of most organs, including spleen and thymus, causes a selective expansion of progenitor cells populations (CFU-GM and CFU-E or BFU-E) (34), and increases the risk of lymphoma development (35). Furthermore, in the absence of CDKN1B, altered cell kinetics was observed among progenitors (36). In addition, a number of recent studies have demonstrated the prognostic significance of p27 in human cancer. Decreased levels of total p27 protein are associated with high tumor grade and stage in human breast, colorectal, and gastric cancers, among others (37–40).

The expression of p27 during hematopoiesis varies at different stages of hematopoietic cell differentiation. For example, stem cells express high levels of p27, and they are quiescent or cycling at a very low level. Bone marrow progenitor cells express lower levels of p27 and cycle rapidly (29). Terminally differentiated cells often accumulate p27 and are growth arrested; however, the growth arrest itself is not sufficient to cause differentiation (29). Recently, a number of studies show that p27 may be involved in cell differentiation (41), and although the cells differentiate, they accumulate p27, which cooperates with other proteins to modulate cell differentiation (20, 42).

There are several features of MLL-AF4 leukemias that suggest that regulation of p27 by the fusion protein may be important in this disease. Infants with MLL-AF4 leukemia, which is the predominant type of infant leukemia, have a poor prognosis as compared with other types of acute lymphoblastic leukemias (43). If MLL-AF4 potentiates expression of p27 in bone marrow progenitor cells with this translocation, this could have significant ramifications in terms of the cycling of these cells. Increased p27 expression would likely result in lower cycling of these cells, causing them to be resistant to most current types of chemotherapy regimens. Recently, conditional MLL-AF4-expressing U937 myelomonocytic leukemia cell lines were created by another group (44). Similar to our results, it was found that MLL-AF4 expression resulted in increased doubling times of these cells, although expression levels of cell-cycle regulators including p27 were not reported.

Much attention is currently focused on trying to identify and understand leukemia stem cells (45). Some progress has been made in chronic myelogenous leukemia (CML), where it was found that the leukemia stem cells are mostly quiescent (46). Murine models of this disease are useful in trying to determine which strategies might be most effective in destroying the leukemia stem cell to cure disease. Unfortunately, the leukemia stem cell for MLL-AF4 leukemia has not yet been identified, nor have any reliable animal models been created that recapitulate the human disease. This limitation is in contrast to other MLL fusion leukemias, where animal models have been successfully created (11, 12). We have demonstrated that CDKN1B is a direct target of MLL-AF4. Our working hypothesis is that MLL-AF4 aberrantly regulates expression of p27 as compared with wild-type MLL, which might result in an inappropriate either increase or decrease of p27 expression during a critical stage in the hematopoietic differentiation program. An increase in p27 in the leukemia stem cell population would likely result in a less effective response to standard chemotherapy regimens. Alternative strategies that target p27 in leukemia stem cells may prove effective in treatment of this disease. Wild-type MLL also regulates p27 expression. MLL may be involved in regulating the balance between quiescence and proliferation in normal hematopoietic stem cells.

Acknowledgments

We thank Dr. T. Sakai (Kyoto Prefectural University of Medicine) and E. A. Williamson (University of California Los Angeles School of Medicine) for providing pLG3-CDKN1B and Dr. S. Korsmeyer (Dana–Farber Cancer Institute, Boston) for providing MEFs. We thank Drs. Michael Thirman and Stephen Nimer for helpful advice regarding the paper. This work was funded by National Institutes of Health/National Cancer Institute Grants CA78438 and CA40046 (to N.J.Z.-L.) and CA81269 and CA104300 (to M.O.D.), and the Dr. Ralph and Marian Falk Medical Research Trust (to N.J.Z.-L.).

Notes

Author contributions: Z.-B.X. and N.J.Z.-L. designed research; Z.-B.X., R.P., J.C., C.T., T.S., D.A.S., and F.E. performed research; and Z.X.-B., M.O.D., and N.J.Z.-L. analyzed data and wrote the paper.

Abbreviations: ChIP, chromatin immunoprecipitation; CDKs, cyclin-dependent kinases; CDKIs, CDK inhibitors; PI, propidium iodide; MEFs, murine embryonic fibroblasts.

References

2. Ernst, P., Wang, J. & Korsmeyer, S. J. (2002) Curr. Opin. Hematol. 9**,** 282-287. [PubMed] [Google Scholar]

3. Nakamura, T., Mori, T., Tada, S., Krajewski, W., Rozovskaia, T., Wassell, R., Dubois, G., Mazo, A., Croce, C. M. & Canaani, E. (2002) Mol. Cell 10**,** 1119-1128. [PubMed] [Google Scholar]

4. Yokoyama, A., Kitabayashi, I., Ayton, P. M., Cleary, M. L. & Ohki, M. (2002) Blood 100**,** 3710-3718. [PubMed] [Google Scholar]

5. Zeleznik-Le, N. J., Harden, A. M. & Rowley, J. D. (1994) Proc. Natl. Acad. Sci. USA 91**,** 10610-10614. [PMC free article] [PubMed] [Google Scholar]

6. Birke, M., Schreiner, S., Garcia-Cuellar, M. P., Mahr, K., Titgemeyer, F. & Slany, R. K. (2002) Nucleic Acids Res. 30**,** 958-965. [PMC free article] [PubMed] [Google Scholar]

7. Xia, Z. B., Anderson, M., Diaz, M. O. & Zeleznik-Le, N. J. (2003) Proc. Natl. Acad. Sci. USA 100**,** 8342-8347. [PMC free article] [PubMed] [Google Scholar]

8. Milne, T. A., Briggs, S. D., Brock, H. W., Martin, M. E., Gibbs, D., Allis, C. D. & Hess, J. L. (2002) Mol. Cell 10**,** 1107-1117. [PubMed] [Google Scholar]

9. Johansson, B., Moorman, A. V., Haas, O. A., Watmore, A. E., Cheung, K. L., Swanton, S. & Secker-Walker, L. M. (1998) Leukemia 12**,** 779-787. [PubMed] [Google Scholar]

10. Behm, F. G., Raimondi, S. C., Frestedt, J. L., Liu, Q., Crist, W. M., Downing, J. R., Rivera, G. K., Kersey, J. H. & Pui, C. H. (1996) Blood 87**,** 2870-2877. [PubMed] [Google Scholar]

13. So, C. W., Lin, M., Ayton, P. M., Chen, E. H. & Cleary, M. L. (2003) Cancer Cell. 4**,** 99-110. [PubMed] [Google Scholar]

14. Zeisig, B. B., Milne, T., Garcia-Cuellar, M. P., Schreiner, S., Martin, M. E., Fuchs, U., Borkhardt, A., Chanda, S. K., Walker, J., Soden, R., et al. (2004) Mol. Cell. Biol. 24**,** 617-628. [PMC free article] [PubMed] [Google Scholar]

16. Polyak, K., Lee, M. H., Erdjument-Bromage, H., Koff, A., Roberts, J. M., Tempst, P. & Massague, J. (1994) Cell 78**,** 59-66. [PubMed] [Google Scholar]

18. Hsieh, F. F., Barnett, L. A., Green, W. F., Freedman, K., Matushansky, I., Skoultchi, A. I. & Kelley, L. L. (2000) Blood 96**,** 2746-2754. [PubMed] [Google Scholar]

19. Steinman, R. A., Lu, Y., Yaroslavskiy, B. & Stehle, C. (2001) Oncogene 20**,** 6524-6530. [PubMed] [Google Scholar]

20. de Koning, J. P., Soede-Bobok, A. A., Ward, A. C., Schelen, A. M., Antonissen, C., van Leeuwen, D., Lowenberg, B. & Touw, I. P. (2000) Oncogene 19**,** 3290-3298. [PubMed] [Google Scholar]

21. Tsukiyama, T., Ishida, N., Shirane, M., Minamishima, Y. A., Hatakeyama, S., Kitagawa, M., Nakayama, K. & Nakayama, K. (2001) J. Immunol. 166**,** 304-312. [PubMed] [Google Scholar]

22. Erfurth, F., Hemenway, C. S., de Erkenez, A. C. & Domer, P. H. (2004) Leukemia 18**,** 92-102. [PubMed] [Google Scholar]

23. Yu, B. D., Hess, J. L., Horning, S. E., Brown, G. A. & Korsmeyer, S. J. (1995) Nature 378**,** 505-508. [PubMed] [Google Scholar]

24. Graf, E. H., Taube, T., Hartmann, R., Wellmann, S., Seifert, G., Henze, G. & Seeger, K. (2002) Blood 99**,** 4629-4631. [PubMed] [Google Scholar]

25. Li, C. Y., Suardet, L. & Little, J. B. (1995) J. Biol. Chem. 270**,** 4971-4974. [PubMed] [Google Scholar]

26. Williamson, E. A., Dadmanesh, F. & Koeffler, H. P. (2002) Oncogene 21**,** 3199-3206. [PubMed] [Google Scholar]

27. Minami, S., Ohtani-Fujita, N., Igata, E., Tamaki, T. & Sakai, T. (1997) FEBS Lett. 411**,** 1-6. [PubMed] [Google Scholar]

28. Allen, R. J., Smith, S. D., Moldwin, R. L., Lu, M. M., Giordano, L., Vignon, C., Suto, Y., Harden, A., Tomek, R., Veldman, T., et al. (1998) Leukemia 12**,** 1119-1127. [PubMed] [Google Scholar]

30. Schraets, D., Lehmann, T., Dingermann, T. & Marschalek, R. (2003) Oncogene 22**,** 3655-3668. [PubMed] [Google Scholar]

31. Lange, B., Valtieri, M., Santoli, D., Caracciolo, D., Mavilio, F., Gemperlein, I., Griffin, C., Emanuel, B., Finan, J., Nowell, P., et al. (1987) Blood 70**,** 192-199. [PubMed] [Google Scholar]

32. Milne, T. A., Hughes, C. M., Lloyd, R., Yang, Z., Rozenblatt-Rosen, O., Dou, Y., Schnepp, R. W., Krankel, C., Livolsi, V. A., Gibbs, D., et al. (2005) Proc. Natl. Acad. Sci. USA 102**,** 749-754. [PMC free article] [PubMed] [Google Scholar]

33. Kiyokawa, H., Kineman, R. D., Manova-Todorova, K. O., Soares, V. C., Hoffman, E. S., Ono, M., Khanam, D., Hayday, A. C., Frohman, L. A. & Koff, A. (1996) Cell 85**,** 721-732. [PubMed] [Google Scholar]

34. Fero, M. L., Rivkin, M., Tasch, M., Porter, P., Carow, C. E., Firpo, E., Polyak, K., Tsai, L. H., Broudy, V., Perlmutter, R. M., et al. (1996) Cell 85**,** 733-744. [PubMed] [Google Scholar]

35. Hwang, H. C., Martins, C. P., Bronkhorst, Y., Randel, E., Berns, A., Fero, M. & Clurman, B. E. (2002) Proc. Natl. Acad. Sci. USA 99**,** 11293-11298. [PMC free article] [PubMed] [Google Scholar]

36. Cheng, T., Rodrigues, N., Dombkowski, D., Stier, S. & Scadden, D. T. (2000) Nat. Med. 6**,** 1235-1240. [PubMed] [Google Scholar]

37. Catzavelos, C., Bhattacharya, N., Ung, Y. C., Wilson, J. A., Roncari, L., Sandhu, C., Shaw, P., Yeger, H., Morava-Protzner, I., Kapusta, L., et al. (1997) Nat. Med. 3**,** 227-230. [PubMed] [Google Scholar]

38. Ciaparrone, M., Yamamoto, H., Yao, Y., Sgambato, A., Cattoretti, G., Tomita, N., Monden, T., Rotterdam, H. & Weinstein, I. B. (1998) Cancer Res. 58**,** 114-122. [PubMed] [Google Scholar]

39. Sgambato, A., Ratto, C., Faraglia, B., Merico, M., Ardito, R., Schinzari, G., Romano, G. & Cittadini, A. R. (1999) Mol. Carcinog. 26**,** 172-179. [PubMed] [Google Scholar]

40. Fredersdorf, S., Burns, J., Milne, A. M., Packham, G., Fallis, L., Gillett, C. E., Royds, J. A., Peston, D., Hall, P. A., Hanby, A. M., et al. (1997) Proc. Natl. Acad. Sci. USA 94**,** 6380-6385. [PMC free article] [PubMed] [Google Scholar]

41. Quaroni, A., Tian, J. Q., Seth, P. & Ap, R. C. (2000) Am. J. Physiol. 279**,** C1045-C1057. [PubMed] [Google Scholar]

42. McArthur, G. A., Foley, K. P., Fero, M. L., Walkley, C. R., Deans, A. J., Roberts, J. M. & Eisenman, R. N. (2002) Mol. Cell. Biol. 22**,** 3014-3023. [PMC free article] [PubMed] [Google Scholar]

44. Caslini, C., Serna, A., Rossi, V., Introna, M. & Biondi, A. (2004) Leukemia 18**,** 1064-1071. [PubMed] [Google Scholar]

45. Jordan, C. T. & Guzman, M. L. (2004) Oncogene 23**,** 7178-7187. [PubMed] [Google Scholar]

46. Holyoake, T., Jiang, X., Eaves, C. & Eaves, A. (1999) Blood 94**,** 2056-2064. [PubMed] [Google Scholar]

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