Abstract
The ALOX5 gene encodes 5-lipoxygenase (5-LO), a key enzyme of inflammatory reactions, which is transcriptionally activated by trichostatin A (TSA). Physiologically, 5-LO expression is induced by calcitriol and/or transforming growth factor-β. Regulation of 5-LO mRNA involves promoter activation and elongation control within the 3′-portion of the ALOX5 gene. Here we focused on the ALOX5 promoter region. Transcriptional initiation was associated with an increase in histone H3 lysine 4 trimethylation in a TSA-inducible manner. Therefore, we investigated the effects of the MLL (mixed lineage leukemia) protein and its derivatives, MLL-AF4 and AF4-MLL, respectively. MLL-AF4 was able to enhance ALOX5 promoter activity by 47-fold, which was further stimulated when either vitamin D receptor and retinoid X receptor or SMAD3/SMAD4 were co-transfected. In addition, we investigated several histone deacetylase inhibitors (HDACi) in combination with gene knockdown experiments (HDAC1-3, MLL). We were able to demonstrate that a combined inhibition of HDAC1-3 induces ALOX5 promoter activity in an MLL-dependent manner. Surprisingly, a constitutive activation of ALOX5 by MLL-AF4 was inhibited by class I HDAC inhibitors, by relieving inhibitory functions deriving from MLL.Conversely, a knockdown of MLL increased the effects mediated by MLL-AF4. Thus, HDACi treatment seems to switch ‘inactive MLL’ into ‘active MLL’ and overwrites the dominant functions deriving from MLL-AF4.
Similar content being viewed by others
Introduction
The human 5-lipoxygenase (5-LO), which is encoded by the ALOX5 gene, is an enzyme that catalyzes the first two steps in the biosynthesis of leukotrienes from arachidonic acid. In the pathophysiological context, leukotrienes are associated with inflammatory, allergic and cardiovascular diseases, as well as certain types of cancer.1
The human ALOX5 gene is organized by 14 exons.2 The ALOX5 promoter contains eight GC boxes, but lacks TATA and CAAT boxes.3 As such, the ALOX5 promoter resembles a promoter structure, which is typically found for housekeeping genes. Expression of the ALOX5 gene is regulated by transcriptional initiation as well as elongation. 5-LO transcript elongation and mRNA maturation is strongly stimulated by calcitriol (1,25(OH)2D3) and transforming growth factor-β (TGFβ), respectively, and is controlled by regulatory elements outside the promoter in a ligand-dependent manner, whereas regulatory elements in the promoter region seem to act ligand independent.4, 5, 6, 7, 8 In addition to this promoter-independent mechanism, induction of 5-LO mRNA transcripts in undifferentiated myeloid cells can be strongly enhanced by the pan-histone deacetylase inhibitor (HDACi) trichostatin A (TSA).9, 10 Furthermore, we observed that upregulation of ALOX5 promoter activity by TSA correlates with the recruitment of the transcription factors Sp1 and Sp3, to a promoter proximal Sp1-binding site next to the transcript initiation site.11
The status of histone acetylation, as a marker for active gene transcription, is regulated by histone acetyl transferases and counteracted by HDACs.12 HDACs deacetylate histones as well as other proteins such as transcription factors and can be divided into different classes, namely class I (comprising HDACs 1, 2, 3, 8), class II (4, 5, 6, 7, 9, 10) and class IV (11).13 In addition to acetylation, phosphorylation and ubiquitination, the methylation of histones at certain residues contributes to a sophisticated system called the ‘histone code’ that is directly linked to the regulation and transcriptional memory of cellular gene expression.14, 15, 16 Histone H3 lysine 4 trimethylation (H3K4me3) represents the general signature for active promoters. The enzymatic reaction of H3 lysine 4 trimethylation is catalyzed, for example, by the SET domain of the MLL (mixed lineage leukemia) protein.17, 18 For the MLL gene, a large number of chromosomal rearrangements are described. In particular, the chromosomal translocation t(4;11)(q21;q23) with the AF4 gene is the most frequently diagnosed reciprocal chromosomal translocation of the human MLL gene.19 The resulting fusion proteins MLL-AF4 (der11) and AF4-MLL (der4) are able to induce and maintain the onset of high-risk acute lymphoblastic leukemia. Possible mechanisms that explain the strong oncogenic behavior has been recently summarized in several publications.20, 21, 22, 23
In this study, we demonstrate that HDAC inhibition induces 5-LO mRNA expression, which is concomitantly associated with H3K4 trimethylation of the ALOX5 promoter by the MLL protein. We also show evidence that the MLL-AF4 fusion protein acts as a strong, HDAC-independent transcriptional activator, which acts in a dominant-positive manner over endogenous or transfected MLL. Interestingly, when endogenous MLL becomes activated by HDAC inhibition, the high constitutive activity of MLL-AF4 is diminished to the level of wild-type MLL. We conclude from our study that the ALOX5 promoter/gene system provides a unique tool to dissect regulatory properties of MLL and its derivative proteins.
Results
Time-dependent induction of ALOX5 mRNA expression and histone H3K4me3 after HDAC inhibition
We recently demonstrated that ALOX5 promoter activity is upregulated by HDAC inhibition.9, 10, 11 Here we analyzed the TSA-mediated induction of 5-LO mRNA expression in a time-dependent manner in MM6 and HL-60 cells. Cells were grown in the presence or absence of TSA (330 nM) for the indicated time points (Figure 1a). Cells were harvested and the amount of ALOX5 mRNA was determined by quantitative reverse transcription–PCR. In MM6 as well as HL-60 cells, a distinct increase in 5-LO mRNA was detected already after 4 h. Maximal 5-LO mRNA expression was observed after 16 and 24 h of TSA treatment in HL-60 and MM6 cells, respectively. Only HL-60 cells displayed a decline at 24 h (Figure 1a). We additionally investigated the H3K4me3 signature as a surrogate marker for an activated promoter by chromatin immunoprecipitation (ChIP) assays after various time points of TSA treatment (Figure 1b). An increase in H3K4me3 levels was detectable in both tested cell lines, with a maximum after 16–24 h of TSA treatment. Moreover, H3K4me3 levels correlated with the amount of ALOX5 mRNA. Therefore, we concluded that TSA treatment seems to induce the recruitment of the H3K4 histone methyltransferase MLL to the ALOX5 promoter.
MLL-AF4 (der11) shows a strong, constitutive 5-LO promoter activation, whereas 5-LO promoter activation by MLL is HDAC dependent
In order to recapitulate the previous published data with TSA and its regulatory function on the ALOX5 promoter, we used the pN10 (ALOX5 promoter-luciferase) reporter construct. This construct covers the ALOX5 promoter from −778 to +53 relative to the transcriptional start site. In addition, we used the MLL fusion proteins, MLL-AF4 and AF4-MLL, which were known to strongly enhance transcription (MLL-AF4/der11) or transcriptional elongation (AF4-MLL/der4). All three expression constructs were co-transfected with the corresponding reporter construct into HeLa cells. Vehicle treatment (dimethyl sulfoxide) or TSA (330 nM) were used to analyze the effects of all three proteins on the ALOX5 promoter. As shown in Figure 2, MLL-AF4 strongly induced transcription from the ALOX5 promoter by about 47-fold. However, the addition of TSA caused an upregulation of MLL-mediated transcription, while the strong activation of the 5-LO promoter by MLL-AF4 disappeared, reaching a final activity level that was quite similar to that one observed for transfected MLL.
MLL and t(4;11) fusion proteins stimulate 5-LO promoter activity in the presence of VDR/RXR and of SMAD3/4 in a ligand-independent manner
As ALOX5 is a vitamin D/TGFβ target gene, we investigated possible interactions of the vitamin D3 and TGFβ signaling pathways with MLL on the ALOX5 promoter. HeLa cells were transfected with the pN10 reporter construct in conjunction with either pTarget (empty vector control) or pTarget-based expression vectors for MLL, AF4-MLL or MLL-AF4. As the ALOX5 promoter contains vitamin D response elements and SMAD-binding sites, cells were additionally co-transfected with the nuclear receptors vitamin D receptor/retinoid X receptor (VDR/RXR) or SMAD3/SMAD4 and incubated with or without 1,25(OH)2D3 or TGFβ.
In the absence of VDR/RXR or SMAD3/SMAD4, neither empty vector, MLL nor der4 stimulated the ALOX5 promoter (Figure 3a, white bars), while MLL-AF4 expression activated the ALOX5 promoter by about 10-fold. Interestingly, when VDR/RXR (Figure 3a) or SMAD3/SMAD4 (Figure 3b) were co-expressed, MLL, der4 and der11 stimulated 5-LO promoter activity, whereas no effect of VDR or SMADs was observed with the empty vector. The up-to-six-fold induction of ALOX5 promoter activity by VDR/RXR or SMAD3/SMAD4 is displayed in the right panels of Figures 3a and b, respectively.
Interestingly, neither the addition of 1,25(OH)2D3 nor TGFβ enhanced the VDR/RXR or SMAD3/SMAD4-mediated effects. This indicated that the observed effects are ligand independent. In contrast, when we used the VDR/RXR or SMAD3/SMAD4 luciferase reporter control plasmids, a ligand-dependent stimulation of about 4- to 8-fold was observed (Figures 3c and d). Thus, the MLL-mediated regulation of the ALOX5 promoter is stimulated by VDR/RXR and SMAD3/SMAD4, but does not depend on the presence of the respective ligand.
Inhibition of HDAC1-3 indirectly upregulates the ALOX5 promoter via the activation of the MLL protein
Next we wanted to identify the HDAC isoform, which is responsible for the MLL-dependent upregulation of 5-LO promoter activity. To address this question, our reporter construct was co-transfected with pTarget or the pTarget expression vectors for MLL, AF4-MLL or MLL-AF4. Transfected cells were then incubated with different HDACi that display distinct inhibitory profiles for the various HDAC isoforms: MC-1568 is an inhibitor of class IIa HDACs; Apicidin acts on HDAC1, HDAC2, HDAC3 and HDAC8; MS-275 (Entinostat) preferentially inhibits HDAC1, but also inhibits HDAC2 and HDAC3 at micromolar concentrations; PCI-34051 is a known HDAC8 inhibitor; Mocetinostat (MGCD0103) is an HDAC1, HDAC2 and HDAC3 inhibitor; and Droxinostat is an HDAC3, HDAC6 and HDAC8 inhibitor.24, 25, 26, 27, 28 Similar to TSA, all tested inhibitors that interfere with HDACs 1–3 strongly increased ALOX5 promoter activity, whereas the inhibition of class IIa HDACs or HDAC8 had no effect (Figures 4a and b). Co-transfection of AF4-MLL leads to similar promoter activity as pTarget (negative control), suggesting that in contrast to MLL and MLL-AF4, the AF4-MLL fusion protein does not activate the basal ALOX5 promoter. Interestingly, some HDACi caused an activation of the reporter construct (see Figure 4a) in the control cells (transfected with pTarget) and in cells where MLL was co-transfected. For example, Apicidin at 100 nM, which inhibits HDAC1, HDAC2 and HDAC3, leads to full promoter activation when MLL is co-transfected. Similar observations were made with MS275 at 1 μM, which inhibits HDAC1 and HDAC2. When the inhibitors were added at concentrations that inhibit HDAC1–3, for example, MS275 (10 μM) and Mocetinostat (1 and 10 μM), the ALOX5 promoter activation was comparable under all four conditions (control, AF4-MLL, MLL and MLL-AF4). Interestingly, co-transfection of MLL-AF4 led to strong activation of the ALOX5 promoter and the addition of HDACi did not enhance this effect. This suggests that the observed activation of the ALOX5 promoter was independent of any HDACi. Most important, all HDACi that lead to an MLL-dependent ALOX5 promoter activation (Apicidin, MS-275, Mocetinostat and Droxinostat) reduced the effect of the MLL-AF4 fusion protein to ~50%, similar to what was observed before with TSA (Figure 2). These results suggested again that a functional activation of (endogenous) MLL by distinct HDACi reduces the non-physiological effects of the MLL-AF4 fusion protein on gene transcription, possibly by a competition of MLL that disables binding of MLL-AF4 at the ALOX5 promoter.
Knockdown of HDACs1-3 induces 5-LO promoter activity
As the available HDACi only displayed a limited selectivity for HDAC1–3, transient knockdown experiments for these HDACs were performed to directly test their influence on ALOX5 promoter activity. Knockdown of each HDAC was performed by co-transfection of the respective small interfering RNA (siRNA) and the ALOX5 promoter-luciferase reporter construct. As displayed in Figure 5a, siRNA co-transfection leads to a reduction of HDAC1–3 mRNA in the range of about 40–75%. Unfortunately, we could not verify the knockdown of the HDACs at the protein level, as HDAC expression was too low to be detected by conventional western blotting experiments (data not shown). This HDAC knockdown experiment revealed that 5-LO promoter activity was per se not induced by the knockdown of single HDAC enzymes (Figure 5b). However, the combined knockdown of HDAC3 in conjunction with either HDAC1 or HDAC2 leads to slightly higher promoter activity (Figure 5b), supporting the assumption that HDAC3 is important for the function of the VDR/RXR that binds to ALOX5 promoter fragment, while HDAC1 or HDAC2 probably modulate genuine MLL functions.
Next, we wanted to confirm the synergistic effects between the various HDACs. To this end, we combined the HDAC knockdown with a TSA treatment in order to inhibit remaining HDAC activity and to assess this effect on ALOX5 promoter activation (Figure 5c). Interestingly, the combination of HDAC3 knockdown and TSA treatment led to the highest 5-LO promoter induction, indicating that HDAC3 has a key role in the regulation of 5-LO promoter activity but seems to require inhibition of additional HDACs (here, inhibited by TSA) for the maximal activation of the ALOX5 promoter. The lowered induction of the ALOX5 promoter in the presence of TSA (Figure 5c) was due to the higher basal activity of the ALOX5 promoter under the knockdown conditions (compare with Figure 5b). Taken together, our data suggest that inhibition of HDAC1–3 upregulates the ALOX5 promoter activity, presumably in an MLL-dependent manner.
Concentration-dependent effects of HDACi on 5-LO promoter activity and SEM cells
Finally, we used three different HDACi and recorded dose–response curves to compare their MLL-AF4-blocking activities with the concurrent activation of the endogenous MLL protein. For this experiment we used three representative inhibitors: Apicidin (3–300 nM) inhibits HDAC1, 2 and 3), MS-275 that preferentially inhibits HDAC1 (30 nM–3 μM), but at micromolar concentrations also HDAC2 and 3. Finally, we used Mocetinostat (3–300 nM) that inhibits all three HDACs. As shown in Figure 6a, all three inhibitors were able to decrease the promoter-stimulating activity of MLL-AF4 (pT-der11), while the inhibitors increased 5-LO promoter activity with endogenous MLL (pTarget) or the combination of endogenous and transfected MLL (pT-MLL), but all ending at similar promoter activities (compare Figure 4a). Interestingly, reduction of MLL-AF4 activity correlates with HDAC1 inhibition by the three inhibitors, respectively, whereas 5-LO promoter activation correlates with inhibition of all three HDACs. The data suggest an HDAC-dependent mechanism that blocks the promoter-stimulating activity of endogenous or transfected MLL, whereas MLL-AF4 is constitutively active and overrides endogenous MLL. However, if the intrinsic blocking mechanism of MLL is relieved by the addition of HDAC1/2 inhibitors, endogenous or transfected MLL is able to outcompete and thus inhibit the MLL-AF4 functions, thereby blocking its ectopic properties on gene transcription.
This observation was also tested on cells that bear the t(4;11) translocation. In this case, we used the SEM cell line that expresses both MLL fusion proteins, MLL-AF4 and AF4-MLL, together with wild-type MLL and AF4. As summarized in Figure 6b, cell counting kit-8 (CCK-8) assays and growth analysis demonstrated that the HDAC1–3 inhibitors Droxinostat and Apicidin display the most potent effects on the viability of these leukemic cells. The HDAC1–3 inhibitors Droxinostat (at 100 μM), Apicidin (300 nM) and MS-275 (3 μM) were able to inhibit SEM cell growth but also compromised cell viability, whereas the HDAC1, 2 inhibitor Mocetinostat (300 nM) as well as MC1568 (1 μM) and PCI-34051 (5 μM) were less active. The data suggest that inhibition of HDAC1–3 activities and subsequent blockage of the MLL-AF4 oncofusion protein results in reduced cell viability and inhibition of cell proliferation.
Knockdown of endogenous MLL enhances functions of MLL-AF4
In order to validate our hypothesis that activation of endogenous MLL by HDAC class I inhibitors neutralizes the oncogenic properties of MLL-AF4, we performed two additional experiments. First, we recorded dose–response curves for the class I HDACi Apicidin, MS-275 and Mocetinostat in the presence of transfected MLL and MLL-AF4. As shown in Figure 7a, Apicidin, MS-275 and Mocetinostat were all able to reduce the strong transcriptional activation function deriving from MLL-AF4 to the level that is also reached either by endogenous or transfected MLL.
Second, we used an short hairpin RNA-mediated knockdown of endogenous MLL to demonstrate that endogenous MLL is responsible for the observed HDACi effects on MLL-AF4. As shown in Figure 7b, an MLL knockdown of about 30% could be reached. This was sufficient to increase the MLL-AF4-mediated activation of the 5-LO promoter, indicating that endogenous MLL antagonizes the transcriptional activity of MLL-AF4. Furthermore, reduction of MLL-AF4 transcriptional activity by the class I HDACi was impaired by the MLL knockdown.
HDACi inhibit recruitment of MLL-AF4 and stimulate recruitment of MLL to the promoter
In order to verify that activation of endogenous MLL by HDACi can displace MLL-AF4 from the ALOX5 promoter, ChIP analyses of MLL and MLL-AF4 were performed. HeLa cells were co-transfected with the ALOX5 promoter reporter plasmid pN10, and either the expression construct for MLL or MLL/MLL-AF4. In cells transfected with MLL, HDAC inhibition by Apicidin slightly increases the recruitment of MLL to the 5-LO promoter, but strongly triggers binding of endogenous AF4 (Figure 8b), which is reflected by ALOX5 promoter activation (Figure 4a). When cells are transfected with MLL and MLL-AF4, there is a prominent binding of endogenous MLL to the ALOX5 promoter as detected by the MLL antibody directed against the MLL C-terminus, which recognizes MLL but not MLL-AF4. Interestingly, Apicidin does not seem to alter the abundance of MLL at the 5-LO promoter but the compound strongly downregulates binding of MLL-AF4 to the ALOX5 promoter (Figure 8c vs d), which is in accordance to our observation that Apicidin downregulates the ability of MLL-AF4 to bind to the 5-LO promoter, while MLL does.
Discussion
Expression of the ALOX5 gene is upregulated during myeloid cell maturation. The gene is regulated at the level of transcript initiation and transcriptional elongation. In the present study, we focused only on the regulation of transcriptional initiation. For the purpose of our studies, we used a promoter fragment (−778 to +53) of ALOX5 that was fused to a luciferase reporter gene.9, 10, 11 In order to understand transcript initiation in more detail, we investigated the chromatin signatures at the endogenous promoter and observed that TSA treatment leads to a profound upregulation of the canonical H3K4me3 signature. Therefore, we started to investigate the function of the H3K4 histone methyltransferase MLL and its role in the control of ALOX5 transcription. As the ALOX5 gene expression has previously been linked to the onset of leukemia development, we also investigated two fusion genes of MLL, MLL-AF4 and AF4-MLL, which produce the most common fusion proteins found in MLL-rearranged leukemia patients.19 From earlier studies, it was known that MLL-AF4 acts on promoters and enhances gene transcription, while AF4-MLL has an important role for transcriptional elongation.23
We were able to demonstrate that either selective inhibition of class I HDACs or the siRNA-mediated knockdown of HDAC1–3 has significant effects on the activation of the ALOX5 promoter in an MLL-dependent manner. As the MLL protein complex is known to bear an intrinsic regulatory mechanism, enabling MLL to act either as transcriptional activator or repressor,29, 30, 31our findings about HDACi and the MLL functions are quite intriguing. The activation and inhibitory functions are mediated by distinct domains of the MLL protein. The N-terminal portion of MLL exhibits the inhibitory MT domain that binds BMI-1 and HDAC1 and 2,32 whereas the C-terminal activation domain can recruit CBP and potentially other members of the coactivator complex.33 The MLL PHD3 domain, which is located between both regulatory domains on the MLL protein, has been shown to act as a molecular switch for the transition between activation and repression of target genes.30, 31 The MLL PHD3 domain binds either to H3K4me3 signatures or to the prolyl isomerase CYP33 (mutually exclusive). CYP33 regulates the conformation of MLL by proline isomerization, which facilitates an interaction to the complex of BMI-1 and HDACs1/2.30, 31 Therefore, CYP33 modulates the function of MLL and allows it to act as a transcriptional repressor.32
Interestingly, the MLL-AF4 fusion protein still retains the MT domain, but lacks the PHD domain. According to our data, MLL-AF4 acts similar to a constitutively active version of the MLL protein on the tested ALOX5 promoter. Addition of different HDACi that specifically block HDAC1–3 activities allows efficient activation of the reporter construct by endogenous MLL, while binding to and activation of the ALOX5 promoter by MLL-AF4 is concomitantly reduced. This mode of action is displayed in Figure 9, where class I HDACs (presumably HDAC1 and 2) seem to act on the activator/repressor switch of MLL, which significantly reduce binding of MLL-AF4 to the ALOX5 promoter. Interestingly, the presence of MLL-AF4 seems to dramatically enhance the binding of MLL. This may indicate an additional and novel function of MLL-AF4, which could be ‘chromatin opening’, for example, by recruiting the SWI/SNF complex. This needs to be investigated in future experiments.
However, our current data suggest that the application of class I HDACi should have a therapeutic benefit in MLL-rearranged leukemias, as they abolish activities deriving from MLL-AF4 product by activation of endogenous MLL. This assumption was validated by the experiments displayed in Figure 6b, where proliferation and viability of t(4;11) cells could be strongly inhibited by the class I HDACi.
Our data also confirm earlier findings that the AF4-MLL fusion protein has effects mainly on transcriptional elongation and acts less on transcriptional initiation. AF4-MLL lacks the DNA-binding domain of MLL. Instead, the AF4-MLL fusion protein links the N-terminal part of AF4, which mediates binding to the transcription elongation factor p-TEFb, to the C-terminus of MLL, which can interact with the transcription factors such as CBP or E2F2, and can stimulate transcript elongation, if a given transcript is elongation controlled. When we tested the p-TEFb inhibitor Flavopiridol (30 nM), we could not inhibit reporter gene activity, indicating that transcription from the ALOX5 promoter reporter plasmid is not elongation controlled (data not shown).
By contrast, VDR/RXR as well as SMAD3/SMAD4 support the induction of ALOX5 promoter activity by MLL, AF4-MLL and MLL-AF4. Interestingly, their respective ligands 1,25(OH)2D3 and TGFβ are strong inducers of 5-LO mRNA expression in myeloid cells5, 6 and it was shown that the 5-LO promoter contains vitamin D-responsive elements and binds the VDR,34 but that these ligand-dependent effects are not mediated by the ALOX5 promoter and are due to stimulation of transcriptional elongation, which seems to be controlled by regulatory elements outside of the ALOX5 promoter.35, 36 These data are in line with our observation here that VDR/RXR as well as SMAD3/SMAD4 support MLL induction of ALOX5 promoter in a ligand-independent manner. Our results suggest that there is an interaction between VDR/RXR as well as SMADs and MLL, which supports the model of an MLL associated transcription-regulating multiprotein complex.37
In conclusion, our data show that both MLL and MLL-AF4 are able to activate the ALOX5 promoter. Of note, we found that wild-type MLL has low activating activity in the absence of HDAC inhibition and that HDAC class I inhibitors stimulate 5-LO promoter activity in an MLL-dependent manner. Our findings are line with recent findings where gene expression profiling data of t(4;11) leukemia cells were related to connectivity maps. This type of analyses revealed that MLL leukemia cells could profit from HDACi.38 Here we demonstrate for the first time by using a bona fide target gene of the MLL complex that the high constitutive oncogenic activity of MLL-AF4 can be diminished by class I HDACi. This observation might be of clinical importance and it will be of interest to elucidate the benefit of class I HDACi as an add-on treatment for t(4;11) leukemia patients.
Materials and methods
Cells and cell culture
Mono Mac 6 (MM6) cells (DSMZ number: ACC-124) were grown at 37 °C in a humidified atmosphere with 5% CO2 in RPMI 1640 medium supplemented with 10% (v/v) fetal calf serum, L-glutamine (2 mM), 1 × non-essential amino acids, sodium pyruvate (1 mM), oxalacetate (1 mM), penicillin (100 U/ml), streptomycin (100 μg/ml) and insulin (10 μg/ml). HL-60 cells (DSMZ number: ACC-3) and SEM K2 cells (DSMZ number: ACC-546) were grown under same conditions in RPMI 1640 medium, 10% (v/v) fetal calf serum, penicillin (100 U/ml) and streptomycin (100 μg/ml). HeLa cells (DSMZ number: ACC-57) were grown in Dulbecco’s modified Eagle medium supplemented with 10% (v/v) fetal calf serum, L-glutamine (2 mM), sodium pyruvate (1 mM), penicillin (100 U/ml) and streptomycin (100 μg/ml). Cultivation of the HeLa cells was carried out in a humidified atmosphere of 5% CO2 at 37 °C.
Reagents
TSA (Sigma-Aldrich, Schnelldorf, Germany: T1952), MC-1568 (Sigma-Aldrich: M1824), Apicidin (Sigma-Aldrich: A8851), MS-275 (Enzo Life Sciences, Lörrach, Germany: ALX-270-378-M005), PCI-34051 (Cayman Chemical, Ann Arbor, MI, USA: 10444), Mocetinostat (Selleckchem, München, Germany: S1122) and Droxinostat (Selleckchem: S1422) were dissolved in dimethyl sulfoxide (AppliChem, Darmstadt, Germany: A3006). Calcitriol was purchased from Sigma-Aldrich. Human TGFβ1 was purified from platelets according to Werz et al.39
RNA extraction and cDNA synthesis
Two micrograms of total RNA were extracted from HL-60, MM6 or HeLa cells by the Total RNA Kit from Omega Bio-Tek (Norcross, GA, USA: R6834-02). RNA was reverse transcribed into cDNA using the iScript cDNA synthesis kit (Bio-Rad, München, Germany: 170–8897) or the High-Capacity cDNA Reverse Transcription Kit (Life Technologies, Darmstadt, Germany: 4368814) according to the manufacturer’s protocol.
Real-time PCR analysis for detection of time-dependent induction of 5-LO mRNA expression
Quantitative PCR analysis was performed with a MyiQ cycler (Bio-Rad). The sequences of the 5-LO primers were 5′-GTTCCTGAATGGCTGCAAC-3′ (forward) and 5′-GGCAATGGGGACAATCTTG-3′ (reverse). Results were normalized to riboprotein P0 Ct values. Sequences of the riboprotein P0 primers were as follows: 5′-AGATGCAGCAGATCCGCAT-3′ (forward) and 5′-GTGGTGATACCTAAAGCCTG-3′ (reverse). Each sample was set up in triplicates. The expression was quantified by comparative ΔΔCT method.
ChIP assay for H3K4me3
ChIP assays were performed as described previously, with some modifications.11 Cells were incubated with 330 nM TSA for the indicated times before harvest. The collected immune complexes were washed twice with RIPA buffer (10 mM Tris–HCl pH 8.1, 1% Triton-X, 0.1% SDS, 0.25% sodium deoxycholate) and equilibrated with TE buffer (10 mM Tris-HCl pH 8, 1 mM EDTA pH 8) before harvest. The antibodies used for immunoprecipitation were anti-histone H3 lysine 4 trimethyl (Upstate, Merck Millipore, Schwalbach, Germany: 07–473) and normal rabbit IgG (Santa Cruz Biotechnology, Heidelberg, Germany: sc-2027). The DNA was analysed by PCR (32 cycles) using the following primers: 5′-AGGAACAGACACCTCGCTGAGGAG-3′ (forward) and 5′-GAGGCTGAGGTAGATGTAGTCGTCAGTG-3′ (reverse), which cover the 5-LO promoter from bp −286 to +78. PCR was visualized after separation on 1% agarose gels containing 0.5% ethidium bromide with the Gel Doc 1000 System (Bio-Rad).
ChIP assay on the recruitment of MLL and MLL-AF4 to the 5-LO promoter region
Twenty-four hours before transfection, HeLa cells were seeded at a density of 2 × 106 cells in a 145-mm cell culture plate with the cell line-specific medium. Transient transfection of the cells with 13.75 μg 5-LO promoter reporter gene construct pN10 and 6.875 μg expression vector of pTarget/pT-MLL or pT-MLL/pT-der11 was performed by calcium phosphate precipitation method. The medium was changed 16 h after transfection and the cells were incubated without or with 100 nM Apicidin. Forty hours after transfection, 1 × 107 cells were collected and protein–DNA cross-linking was conducted by 2 mM disuccinimidyl glutarate, followed by 1% formaldehyde. The reaction was stopped by adding 125 mM glycine directly to the media. Chromatin was extracted by cell lysis and subsequent sonification steps. Sheared chromatin samples were incubated overnight at 4 °C with protein A/G beads and with the following primary antibodies: goat IgG (Santa Cruz Biotechnology: sc-2028); MLL-C (Active Motif, Carlsbad, CA, USA: 61295) and AF4-C (Abcam, Cambridge, UK: ab31812). After elution, the cross-link was reversed by treating the eluates with RNase A and proteinase K. Quantification of all samples was performed by quantitative PCR analysis in a StepOnePlusTM System (Life Technologies) with following primers 5′-CTTCTCCACACCCTTCCAGGCA-3′ (forward) and 5′-AGAATGGCGCCGGGCCT-3′ (reverse).
Plasmids
The 5-LO promoter firefly luciferase reporter gene vector pN10, containing the ALOX5 promoter sequence from −778 to +53 (relative to the transcriptional start site), was described previously.9, 10 The pSG5-VDR and pSG5-RXR expression plasmids for the human VDR and RXRα, and the control plasmid p(DR3)4tk-Luc were obtained from C Carlberg (Kuopio, Finland). The pCGN-Smad3 and pCGN-Smad4 expression plasmids were obtained from XF Wang (Durham, NC, USA). The control plasmids p3TP-Lux and pSBE4-Luc were obtained from J Massagué (New York, NY, USA) and B Vogelstein (Baltimore, MD, USA). The mammalian expression vector pTarget and Renilla luciferase control vector pRL-SV40 were purchased from Promega (Mannheim, Germany). Expression vectors for full-length MLL, MLL-AF4 and AF4-MLL were established in the group of Professor Marschalek. All expression cassettes are flanked by the rare-cutting Sfi1 sites and were cloned in a Sfi1 restriction site-modified pTarget vector (Promega).40
Knockdown of HDAC1, HDAC2 and HDAC3 with siRNA
Twenty-four hours before transfection, HeLa cells were seeded into 24-well plates at a density of 4 × 104 cells per well. siRNA for HDAC1, 2 and 3, GAPDH and AllStars Negative Control was from Qiagen, Hilden, Germany (FlexiTube siRNA: catalog numbers si02663472, si00434952, si00057316, si02653266 and 1027281). Co-transfection with 100 ng 5-LO promoter construct pN10 and 20 ng control plasmid pRL-SV40 was carried out with HiPerfect Transfection Reagent (Qiagen: 301705) according to the manufacturer’s protocol for co-transfection. The optimized conditions for knockdown were as follows: siRNA for GAPDH 5 nM, 3 μl HiPerfect; siRNA for HDAC1 10 nM, 4.5 μl; siRNA for HDAC2 7.5 nM, 4.5 μl HiPerfect; siRNA for HDAC3 7.5 nM, 4.5 μl. Medium was changed 16 h after transfection. Cells were collected 48 h after transfection and reporter gene assay was performed.
Calcium phosphate transfection and reporter gene assay
Twenty-four hours before transfection, HeLa cells were seeded at a density of 4 × 104 cells per well in 24-well plates by using the cell line-specific medium without phenol red. Transient transfection of the cells with 225 or 275 ng 5-LO promoter firefly luciferase reporter gene construct pN10, 225 or 275 ng expression vector (pTarget, pT-MLL, pT-der4 or pT-der11), 50 ng nuclear receptor expression vector (pSG5-VDR and pSG5-RXR or pCGN-SMAD3 and pCGN-SMAD4) and 20 ng Renilla luciferase control vector (pRL-SV40) per well was performed by calcium phosphate precipitation method.41 The medium was changed 16 h after transfection and the cells were incubated without or with 50 nM calcitriol and 1 ng/ml TGFβ or inhibitors, as indicated in the text. Forty hours after transfection, firefly and Renilla luciferase activity was determined with the Dual-Glo Luciferase Assay System (Promega) and a TECAN infinite M200 luminometer. Data were calculated as relative light units by normalization of transfection efficiency with the values of Renilla luciferase activity.
Stable knockdown of MLL in HeLa cells by lentiviral transduction
MISSION short hairpin RNA plasmid for MLL knockdown (NM_005933.1-2990s1c1: TRCN0000005954) was obtained from Sigma-Aldrich. HEK293T cells were transfected with 10 μg short hairpin RNA plasmid, 6.5 μg lentiviral packaging vector psPAX2 (Addgene, Cambridge, MA, USA: 12260) and 3.5 μg envelope plasmid pMD2.G (Addgene: 12259) by calcium phosphate precipitation method. The medium was changed 4 h after transfection and supernatant was collected after 72 h. For transduction, HeLa cells were prepared in 24-well plates and were treated after 24 h with 4 μg/ml protamine sulfate and 10, 50, 100 or 500 μl supernatant including lentiviral particles. The cells were centrifuged (90 min, 2500 r.p.m., 32 °C) and cultured for 72 h. Selection of transduced cells was performed by puromycin (0.75 μg/ml) and knockdown efficiency was checked by quantitative PCR analysis with a StepOnePlus System (Life Technologies). The sequences of the MLL primers were 5′-AGGAGAATGCAGGCACTTTGAA-3′ (forward) and 5′-TTTGCTTAGAACTATTGCCATTGG-3′ (reverse). Results were normalized to β-actin Ct values. Sequences of the β-actin primers were as follows: 5′-CGGGACCTGACTGACTACCTC-3′ (forward) and 5′-CTTCTCCTTAATGTCACGCACG-3′ (reverse). Each sample was set up in triplicates. The expression was quantified by comparative ΔΔCT method.
Statistics
Results were calculated as mean±s.e.m. of at least three independent experiments. Two-tailed Student’s t-test was performed to determine the significance between two groups. One sample t-test was performed to analyze significance of MLL knockdown. Thereby, statistical significance was illustrated as follows: *P<0.05, **P<0.01 and ***P<0.001.
References
Radmark O, Werz O, Steinhilber D, Samuelsson B . 5-Lipoxygenase: regulation of expression and enzyme activity. Trends Biochem Sci 2007; 32: 332–341.
Funk CD, Hoshiko S, Matsumoto T, Radmark O, Samuelsson B . Characterization of the human 5-lipoxygenase gene. Proc Natl Acad Sci USA 1989; 86: 2587–2591.
Hoshiko S, Radmark O, Samuelsson B . Characterization of the human 5-lipoxygenase gene promoter. Proc Natl Acad Sci USA 1990; 87: 9073–9077.
Steinhilber D, Radmark O, Samuelsson B . Transforming growth factor beta upregulates 5-lipoxygenase activity during myeloid cell maturation. Proc Natl Acad Sci USA 1993; 90: 5984–5988.
Brungs M, Radmark O, Samuelsson B, Steinhilber D . On the induction of 5-lipoxygenase expression and activity in HL-60 cells: effects of vitamin D3, retinoic acid, DMSO and TGF beta. Biochem Biophys Res Commun 1994; 205: 1572–1580.
Brungs M, Radmark O, Samuelsson B, Steinhilber D . Sequential induction of 5-lipoxygenase gene expression and activity in Mono Mac 6 cells by transforming growth factor beta and 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 1995; 92: 107–111.
Harle D, Radmark O, Samuelsson B, Steinhilber D . Calcitriol and transforming growth factor-beta upregulate 5-lipoxygenase mRNA expression by increasing gene transcription and mRNA maturation. Eur J Biochem 1998; 254: 275–281.
Stoffers KL, Sorg BL, Seuter S, Rau O, Radmark O, Steinhilber D . Calcitriol upregulates open chromatin and elongation markers at functional vitamin D response elements in the distal part of the 5-lipoxygenase gene. J Mol Biol 2010; 395: 884–896.
Uhl J, Klan N, Rose M, Entian KD, Werz O, Steinhilber D . The 5-lipoxygenase promoter is regulated by DNA methylation. J Biol Chem 2002; 277: 4374–4379.
Klan N, Seuter S, Schnur N, Jung M, Steinhilber D . Trichostatin A and structurally related histone deacetylase inhibitors induce 5-lipoxygenase promoter activity. Biol Chem 2003; 384: 777–785.
Schnur N, Seuter S, Katryniok C, Radmark O, Steinhilber D . The histone deacetylase inhibitor trichostatin A mediates upregulation of 5-lipoxygenase promoter activity by recruitment of Sp1 to distinct GC-boxes. Biochim Biophys Acta 2007; 1771: 1271–1282.
Pokholok DK, Harbison CT, Levine S, Cole M, Hannett NM, Lee TI et al. Genome-wide map of nucleosome acetylation and methylation in yeast. Cell 2005; 122: 517–527.
Yang XJ, Seto E . The Rpd3/Hda1 family of lysine deacetylases: from bacteria and yeast to mice and men. Nat Rev Mol Cell Biol 2008; 9: 206–218.
Margueron R, Trojer P, Reinberg D . The key to development: interpreting the histone code? Curr Opin Genet Dev 2005; 15: 163–176.
Kouzarides T . Chromatin modifications and their function. Cell 2007; 128: 693–705.
Scholz B, Marschalek R . Epigenetics and blood disorders. Br J Haematol 2012; 158: 307–322.
Milne TA, Briggs SD, Brock HW, Martin ME, Gibbs D, Allis CD et al. MLL targets SET domain methyltransferase activity to Hox gene promoters. Mol Cell 2002; 10: 1107–1117.
Wang P, Lin C, Smith ER, Guo H, Sanderson BW, Wu M et al. Global analysis of H3K4 methylation defines MLL family member targets and points to a role for MLL1-mediated H3K4 methylation in the regulation of transcriptional initiation by RNA polymerase II. Mol Cell Biol 2009; 29: 6074–6085.
Meyer C, Hofmann J, Burmeister T, Groger D, Park TS, Emerenciano M et al. The MLL recombinome of acute leukemias in 2013. Leukemia 2013; 27: 2165–2176.
Bursen A, Schwabe K, Ruster B, Henschler R, Ruthardt M, Dingermann T et al. The AF4.MLL fusion protein is capable of inducing ALL in mice without requirement of MLL.AF4. Blood 2010; 115: 3570–3579.
Wilkinson AC, Ballabio E, Geng H, North P, Tapia M, Kerry J et al. RUNX1 is a key target in t(4;11) leukemias that contributes to gene activation through an AF4-MLL complex interaction. Cell Rep 2013; 3: 116–127.
Benedikt A, Baltruschat S, Scholz B, Bursen A, Arrey TN, Meyer B et al. The leukemogenic AF4-MLL fusion protein causes P-TEFb kinase activation and altered epigenetic signatures. Leukemia 2011; 25: 135–144.
Marschalek R . Mixed lineage leukemia: roles in human malignancies and potential therapy. FEBS J 2010; 277: 1822–1831.
Khan N, Jeffers M, Kumar S, Hackett C, Boldog F, Khramtsov N et al. Determination of the class and isoform selectivity of small-molecule histone deacetylase inhibitors. Biochem J 2008; 409: 581–589.
Huber K, Doyon G, Plaks J, Fyne E, Mellors JW, Sluis-Cremer N . Inhibitors of histone deacetylases: correlation between isoform specificity and reactivation of HIV type 1 (HIV-1) from latently infected cells. J Biol Chem 2011; 286: 22211–22218.
Balasubramanian S, Ramos J, Luo W, Sirisawad M, Verner E, Buggy JJ . A novel histone deacetylase 8 (HDAC8)-specific inhibitor PCI-34051 induces apoptosis in T-cell lymphomas. Leukemia 2008; 22: 1026–1034.
Fournel M, Bonfils C, Hou Y, Yan PT, Trachy-Bourget MC, Kalita A et al. MGCD0103, a novel isotype-selective histone deacetylase inhibitor, has broad spectrum antitumor activity in vitro and in vivo. Mol Cancer Ther 2008; 7: 759–768.
Wood TE, Dalili S, Simpson CD, Sukhai MA, Hurren R, Anyiwe K et al. Selective inhibition of histone deacetylases sensitizes malignant cells to death receptor ligands. Mol Cancer Ther 2010; 9: 246–256.
Hom RA, Chang PY, Roy S, Musselman CA, Glass KC, Selezneva AI et al. Molecular mechanism of MLL PHD3 and RNA recognition by the Cyp33 RRM domain. J Mol Biol 2010; 400: 145–154.
Park S, Osmers U, Raman G, Schwantes RH, Diaz MO, Bushweller JH . The PHD3 domain of MLL acts as a CYP33-regulated switch between MLL-mediated activation and repression. Biochemistry 2010; 49: 6576–6586.
Rossler T, Marschalek R . An alternative splice process renders the MLL protein either into a transcriptional activator or repressor. Pharmazie 2013; 68: 601–607.
Xia ZB, Anderson M, Diaz MO, Zeleznik-Le NJ . MLL repression domain interacts with histone deacetylases, the polycomb group proteins HPC2 and BMI-1, and the corepressor C-terminal-binding protein. Proc Natl Acad Sci USA 2003; 100: 8342–8347.
Ernst P, Wang J, Huang M, Goodman RH, Korsmeyer SJ . MLL and CREB bind cooperatively to the nuclear coactivator CREB-binding protein. Mol Cell Biol 2001; 21: 2249–2258.
Sorg BL, Klan N, Seuter S, Dishart D, Radmark O, Habenicht A et al. Analysis of the 5-lipoxygenase promoter and characterization of a vitamin D receptor binding site. Biochim Biophys Acta 2006; 1761: 686–697.
Seuter S, Sorg BL, Steinhilber D . The coding sequence mediates induction of 5-lipoxygenase expression by Smads3/4. Biochem Biophys Res Commun 2006; 348: 1403–1410.
Seuter S, Vaisanen S, Radmark O, Carlberg C, Steinhilber D . Functional characterization of vitamin D responding regions in the human 5-Lipoxygenase gene. Biochim Biophys Acta 2007; 1771: 864–872.
Nakamura T, Mori T, Tada S, Krajewski W, Rozovskaia T, Wassell R et al. ALL-1 is a histone methyltransferase that assembles a supercomplex of proteins involved in transcriptional regulation. Mol Cell 2002; 10: 1119–1128.
Stumpel DJ, Schneider P, Seslija L, Osaki H, Williams O, Pieters R et al. Connectivity mapping identifies HDAC inhibitors for the treatment of t(4;11)-positive infant acute lymphoblastic leukemia. Leukemia 2012; 26: 682–692.
Werz O, Brungs M, Steinhilber D . Purification of transforming growth factor beta 1 from human platelets. Pharmazie 1996; 51: 893–896.
Gaussmann A, Wenger T, Eberle I, Bursen A, Bracharz S, Herr I et al. Combined effects of the two reciprocal t(4;11) fusion proteins MLL.AF4 and AF4.MLL confer resistance to apoptosis, cell cycling capacity and growth transformation. Oncogene 2007; 26: 3352–3363.
Sambrook J, Russell DW . Molecular Cloning: A Laboratory Manual, 3rd edn. Cold Spring Harbor Laboratory Press: Cold Spring Harbor, NY, 2001.
Acknowledgements
This work was supported by grants from Else Kröner-Fresenius-Stiftung (Dr Hans Kröner-Graduiertenkolleg) and from CEF (EXC115).
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
Competing interests
The authors declare no conflict of interest.
Rights and permissions
Oncogenesis is an open-access journal published by Nature Publishing Group. This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-sa/4.0/
About this article
Cite this article
Ahmad, K., Katryniok, C., Scholz, B. et al. Inhibition of class I HDACs abrogates the dominant effect of MLL-AF4 by activation of wild-type MLL. Oncogenesis 3, e127 (2014). https://doi.org/10.1038/oncsis.2014.39
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/oncsis.2014.39
- Springer Nature Limited