Skip to main content
NCBI home page
Transcription logoLink to Transcription
. 2019 Jun 5;10(3):147–156. doi: 10.1080/21541264.2019.1624467

PML–RARα induces all-trans retinoic acid-dependent transcriptional activation through interaction with MED1

Tomoya Fukuoka a, Asami Kawai a, Taku Takahara a, Mahiro Mori a, Robert G Roeder b, Natsumi Hasegawa a, Mitsuhiro Ito a,b,✉
PMCID: PMC6619921  PMID: 31135261

ABSTRACT

Transcriptional activation by PML–RARα, an acute promyelocytic leukemia-related oncofusion protein, requires pharmacological concentrations of all-trans retinoic acid (ATRA). However, the mechanism by which the liganded PML–RARα complex leads to the formation of the preinitiation complex has been unidentified. Here we demonstrate that the Mediator subunit MED1 plays an important role in the ATRA-dependent activation of the PML–RARα-bound promoter. Luciferase reporter assays showed that PML–RARα induced significant transcription at pharmacological doses (1 μM) of ATRA; however, this was submaximal and equivalent to the level of transcription driven by intact RARα at physiological doses (1 nM) of ATRA. Transcription depended upon the interaction of PML–RARα with the two LxxLL nuclear receptor recognition motifs of MED1, and LxxLL→LxxAA mutations led to minimal transcription. Mechanistically, MED1 interacted ATRA-dependently with the RARα portion of PML–RARα through the two LxxLL motifs of MED1. These results suggest that PML–RARα initiates ATRA-induced transcription through its interaction with MED1.

KEYWORDS: PML–RARα, transcriptional activation, Mediator, MED1, LxxLL nuclear receptor recognition motifs

Acute promyelocytic leukemia (APL), a highly aggressive subtype of acute myeloid leukemia (AML), is characterized by an arrest in myeloid maturation due to chimeric oncofusion proteins containing retinoic acid receptor α (RARα). Almost 98% of cases with APL carry promyelocytic leukemia (PML)-fused RARα due to chromosomal translocation t(15;17)(q22;q12) [[1,2]; reviewed in [3]]. In these cases, the breakpoint clustering within the PML gene (intron 3, exon 6 and intron 6) causes the formation of three different isoforms of PML–RARA, termed the short (S or bcr2), long (L or bcr1), and variant (V or bcr3) forms, with frequencies of 40%, 55% and 5%, respectively [[4]; reviewed in [5]]. Although the affinity of all-trans retinoic acid (ATRA) for these forms of PML–RARα is similar to that for RARα [6], at physiological levels of ATRA, PML–RARα does not transactivate target genes and behaves as a constitutive repressor [reviewed in [7–9]]. Under these conditions, PML–RARα interacts with epigenome-modifying corepressors, such as NCoR- and SMRT-containing histone deacetylase complexes, as well as enzymes modifying DNA and histone methylations, causing the hypoacetylation and silencing of chromatin structures [[10,11]; reviewed in [8,9]]. Based on ChIP-Seq analyses, the recruitment of PML–RARα to promoter sequences appears to be less stringently regulated than that of intact RARα [12]. Hence, PML–RARα may suppress a broader range of genes than previously anticipated on the basis of promoters bearing canonical retinoic acid response elements (RARE), and thus may contribute to both the inhibition of ATRA-induced differentiation and leukemogenesis.

Although PML–RARα functions as a constitutive repressor under physiological concentrations of ATRA, in the presence of pharmacological concentrations of ATRA, PML–RARα liberates corepressor complexes, and functions as an activator to induce the differentiation of APL cells [[2,13]; reviewed in [7,8]]. Both RARα and PML–RARα are subject to degradation upon exposure to ATRA, likely through a ligand-induced desensitizing mechanism [reviewed in [8]]. Upon degradation of PML–RARα in the presence of pharmacological concentration of ATRA, previously disrupted PML nuclear bodies are restored, leading to the eradication of leukemic stem cells in vivo and curing the disease [reviewed in [7,8]]. Thus, the therapeutic effects of pharmacological ATRA are derived from targeting the oncoprotein PML–RARα in two ways: PML–RARα-mediated transactivation of target genes in an ATRA-dependent manner followed by the rapid differentiation of APL cells, and ATRA-dependent degradation of the oncoprotein resulting in the restoration of PML nuclear bodies. Regarding the former mechanism, it is still unclear how PML–RARα-mediated transcription is initiated after the release of chromatin from silencing.

The multisubunit complex Mediator is a master transcriptional coregulatory complex, and an end-point hub of intracellular signals. As a constituent of the RNA polymerase II holoenzyme, Mediator facilitates the formation of a functional preinitiation complex (PIC) [reviewed in [14–16]]. The Mediator subunit MED1 serves as an interface for ligand-dependent interaction with nuclear receptors through its two nuclear receptor recognition (NR) motifs and functions as a nuclear receptor specific coactivator [reviewed in [15,17]]. The function of MED1/Mediator, which is related to PIC formation and distinct from histone modifying activities, suggests that the MED1/Mediator acts subsequent to the action of chromatin-modifying coactivators [reviewed in [17,18]]. MED1 is known to be required for optimal RARα function and RARα-mediated neutrophilic differentiation [19], but it is unknown how PML–RARα leads to the formation of functional PICs and subsequent activation of target genes and whether MED1 is involved in this process.

Herein we present evidence that the initiation of transcription by PML–RARα depends upon both of the MED1 NR motifs as well as pharmacological concentrations of ATRA. We propose that MED1 is a specific coactivator for PML–RARα-mediated transcription and is dependent upon pharmacological doses of ATRA, indicating a role for MED1 in differentiation therapy using ATRA in APL.

Results

MED1 is necessary for optimal PML–RARα-initiated transcription

To understand whether MED1 is responsible for PML–RARα-initiated transcription in an ATRA dose-dependent manner, we first performed luciferase reporter assays using the Gal4–luciferase reporter and constructs expressing Gal4 DNA-binding domain (DBD) fusion proteins in Med1+/+ and Med1–/– mouse embryonic fibroblasts (MEFs) [20]. When Gal4 DBD-fused truncated human (h) PML (Gal4-hPML; encoded by PML exons 1–6) (Figure 1a) was introduced to MEFs, transcription levels indicated by luciferase activity remained at a background level regardless of the presence or absence of ATRA or MED1 (Figure 1b). As expected, Gal4–hRARα induced high transcriptional activation in a ligand-dependent manner in Med1+/+ MEFs, reaching a 12-fold increase at the physiological concentration of ATRA (10−9 M) and a 35-fold increase at the pharmacological concentration of ATRA (10−6 M) (Figure 1b). Both the S and L forms of hPML–RARα fused to Gal4 DBD induced significant and similar levels of transcription, reaching a maximum of 5- to 7-fold at the pharmacological ATRA concentration (Figure 1b); however, these levels were much weaker than those induced by Gal4–hRARα. The transcriptional activation in Med1+/+ MEFs also depended on the amounts of these fused oncoproteins (Figure 1c). In contrast, in Med1–/– MEFs, the transcription levels induced by Gal4–hRARα and Gal4–hPML–RARα(S and L) were greatly attenuated to about one-half of the levels observed in Med1+/+ MEFs (Figure 1b).

Figure 1.

Figure 1.

Open in a new tab

MED1 is required for optimal transcription by PML–RARα. (a) Schematic representation of the structures of genes encoding hPML and hRARα, and of their derived oncoprotein products hPML–RARα(S, V and L). Clustered translocation sites (S, V and L) within PML locus are shown. utr, untranslated region. (b-c) Gal4-based luciferase reporter assays in MEFs. Gal4 DBD-fused proteins were tested with 5× Gal4 luciferase reporter in the absence or presence of indicated concentrations of ATRA (b). The same assays were performed in the presence of 10−6 M ATRA and different amounts of Gal4–hPML–RARα(S or L) as indicated (c). MED1 was found to be required for transactivation depending upon the pharmacological ATRA concentrations and Gal4–hPML–RARα(S or L) dose. (d) RARE-based luciferase reporter assays in MEFs. hPML and intact or mutant (oncofusion) hRARα together with their heterooligomerization partner hRXRα were tested with the 3× RARE luciferase reporter in the absence or presence of the indicated concentrations of ATRA. MED1 was found to be required for transactivation initiated by PML–RARα(S or L)/RXRα and dependent on the pharmacological ATRA concentration-dependent transactivation.

Next, similar experiments were performed using a luciferase reporter bearing natural RARE sequences derived from the RARβ promoter [21]. Truncated hPML (derived from exons 1–6) did not activate transcription regardless of the presence of MED1 or ATRA. When hRARα and hRXRα were introduced in Med1+/+ MEFs, transcription was activated over 60-fold in an ATRA dose-dependent manner. However, the levels of activation by the S and L forms of hPML–RARα, while similar, were less than half of that by hRARα. In contrast, the transcriptional activities of these activators were severely attenuated (reduced over three-fold) when analyzed in Med1–/– MEFs (Figure 1d). These results suggest (i) that the S and L forms of PML–RARα similarly activate ATRA-dependent transcription and (ii) that transcriptional activation by PML–RARα is much less efficient compared to that induced by intact RARα and, further, is initiated only at pharmacological doses of ATRA. We also found that MED1 is necessary for optimal transcription induced by PML–RARα and by RARα.

PML–RARα interacts with MED1 through LxxLL NR motifs of MED1

MED1 has two LxxLL NR motifs that serve as ligand-dependent interfaces with nuclear receptors (Figure 2a) [reviewed in [17,18]]. PML–RARα was previously reported to interact with GST-fused MED1 using a GST-pulldown assay [22]. To identify whether PML–RARα(S and L) binds intracellularly with MED1 through the NR motifs, we performed mammalian two-hybrid assays using mutants of the first NR motif (mutant I), the second NR motif (mutant II), and both NR motifs (mutant I+II) of hMED1 (Figure 2a).

Figure 2.

Figure 2.

Open in a new tab

MED1 interacts with PML–RARα through MED1 NR motifs. (a) Schematic representation of hMED1 mutants used. Either one or both of the LxxLL NR motifs were mutated to LxxAA. (b-c) Mammalian two-hybrid assays. Gal4 DBD-fused wild-type or mutant hMED1 and VP16–hPML–RARα(S or L) were tested in HEK 293T cells with the Gal4-luciferase reporter in the absence or presence of the indicated concentrations of ATRA (b). The same assays were performed in the presence of 10−6 M ATRA and different amounts of VP16–hPML–RARα(S or L).

Gal4–hMED1(wild-type) interacted with VP16–hPML–RARα(S and L forms) in an ATRA dose-dependent manner, with a seemingly stronger affinity for VP16–hPML–RARα(S) than for VP16–hPML–RARα(L) (Figure 2b). Gal4–hMED1(mutant I) and Gal4–hMED1(mutant II) interacted likewise with VP16–hPML–RARα(S and L), but less efficiently (especially at lower levels of expression) than with Gal4–hMED1(wild-type). Notably, however, an interaction was barely observed between Gal4–hMED1(mutant I+II) and VP16–hPML–RARα(L and S) even at pharmacological ATRA concentrations (10−6 M) (Figure 2b).

Next, assays were performed with different doses of VP16–hPML–RARα. At a pharmacological ATRA concentration (10−6 M), Gal4–hMED1(wild-type) interacted with VP16–hPML–RARα(S and L) in a VP16–hPML–RARα dose-dependent manner, with a seemingly stronger affinity for VP16–hPML–RARα(S) than for VP16–hPML–RARα(L) (Figure 2c). Both Gal4–hMED1(mutant I) and Gal4-hMED1(mutant II) likewise interacted with VP16–hPML–RARα(S and L) but less strongly than Gal4–hMED1(wild-type). In contrast, Gal4–hMED1(mutant I+II) barely interacted with VP16–hPML–RARα(S and L) (Figure 2c). These results suggest that PML–RARα(S and L) binds to MED1 in an ATRA-dependent manner through both NR motifs of MED1, and that the affinity for PML–RARα(S) may be higher than that for PML–RARα(L).

MED1 LxxLL NR motifs are necessary for optimal PML–RARα-initiated transcription

Since we established that MED1 NR motifs are required for PML–RARα(S and L)-MED1 interactions, we next analyzed the function of MED1 NR motifs using luciferase reporter assays in MEFs bearing LxxLL to LxxAA mutations in MED1 NR motifs (Med1 LxxAA or Med1(lx)KI/KI MEFs) [23]. First, Gal4 DBD-fused proteins were tested using a 5× Gal4–luciferase reporter. We found that Gal4 DBD-fused truncated hPML was completely inactive in Med1 LxxAA MEFs (Figure 3a). When compared to its transcription activity in wild-type (Med1(lx)+/+) MEFs (Med1 WT MEFs), the activity induced by Gal4–hRARα in Med1 LxxAA MEFs was reduced by more than two-fold (Figure 3a) and to levels similar to those observed in Med1–/– MEFs (Figure 1b). Transcriptional activation by Gal4–hPML–RARα(S and L) was also strongly attenuated (reduced over three-fold) in Med1 LxxAA MEFs compared to wild-type MEFs (Figure 3a), similar to above-described observations in Med1–/– MEFs (Figure 1b).

Figure 3.

Figure 3.

Open in a new tab

MED1 NR motifs are required for optimal transcription by PML–RARα. (a) Gal4-based luciferase reporter assays in MEFs. Gal4 DBD-fused proteins were tested with the 5× Gal4 luciferase reporter in the absence or presence of the indicated concentrations of ATRA. (b) RARE-based luciferase reporter assays in MEFs. hPML and intact or mutant (oncofusion) hRARα together with their heterooligomerization partner hRXRα were tested with 3× RARE luciferase reporter in the absence or presence of the indicated concentrations of ATRA. MED1 NR motifs were found to be required for transactivation initiated by PML–RARα(S or L)/RXRα and dependent upon pharmacological ATRA concentrations.

When the luciferase reporter bearing RARβ promoter-derived natural RARE sequences was used, truncated hPML did not activate transcription in Med1 LxxAA MEFs (Figure 3b), consistent with findings described above. RARα-mediated transcription levels were severely attenuated (less than one-fifth) in Med1 LxxAA MEFs compared to those in Med1 WT MEFs (Figure 3b), as was observed in Med1–/– MEFs (Figure 1c). However, there was a residual (6-fold) activation by ATRA at the pharmacological concentrations (10−6 and 10−5 M) (Figure 3b). Notably, however, hPML–RARα(S and L)–mediated transcription was barely detected (1.5- to 3-fold) in Med1 LxxAA MEFs at the pharmacological ATRA concentration (Figure 3b), as seen earlier in Med1–/– MEFs (Figure 1c). These results provide compelling evidence that both RARα and PML–RARα(S and L) require MED1 NR motifs for optimal transcription.

Discussion

PML–RARα-targeted differentiation therapy by ATRA in APL has become the standard of care in clinics and has been so successful that APL is now a curable disease. This study demonstrates, for the first time, that transactivation by PML–RARα is executed through its ATRA-dependent interaction with the Mediator subunit MED1. Our model provides a mechanism of direct transcriptional activation by PML–RARα, and signifies the importance of MED1 NR motifs in ATRA therapy in APL.

Interaction between PML–RARα and MED1

PML–RARα, through its RARα portion, appears to bind equally well to both NR motifs of MED1 (this study). However, this contrasts with the model proposed for the interaction of the receptors TRα, VDR, and PPARα with MED1. These receptors interact strongly with the second NR motif of MED1, while their heterodimerization partner RXR interacts with the first NR motif. The RXR receptor heterodimer has been proposed to simultaneously interact with MED1 through both NR motifs in a unidirectional manner (Figure 4a) [reviewed in [18]]. PML–RARα dimerizes (or oligomerizes) through PML coiled-coil domains, and RXRs associate with the homo-oligomerized PML–RARα, forming a heterotetramer (or heterooligomer) (Figure 4b,c) [reviewed in [8,9]]. Unifying the above ideas, the PML–RARα/RXR heterooligomer may bind MED1 in vivo in two ways: either through the PML–RARα homodimer (Figure 4b) or through the PML–RARα/RXR heterodimer (Figure 4c).

Figure 4.

Figure 4.

Open in a new tab

Model for transcription initiation by PML–RARα. (a) Liganded RARα/RXRα heterodimer interacts with MED1 through two NR motifs, with NR1 bound to RXRα and NR2 bound to RARα. Then Mediator recruits general transcription factors to the promoter leading to the formation of the functional PIC. (b, c) In the case of PML–RARα, the PML–RARα homodimer (or homooligomer) associates with RXRα to form a heterotetramer (or heterooligomer). In this model, pharmacological ATRA doses are assumed to be required to occupy all PML–RARα molecules within the heterooligomer, following which MED1 may interact with the oligoheteromer either through the PML–RARα homodimer (b) or through the PML–RARα/RXRα heterodimer (c). The recruited Mediator then leads to the formation of the functional PIC.

The affinity of ATRA has been reported to be slightly higher for PML–RARα(L) than for PML–RARα(S) [6], and liganded PML–RARα(S) appears to interact with MED1 more efficiently than liganded PML–RARα(L) (this study). The net binding efficiencies of these interfaces may explain the similar levels of transactivation achieved by liganded PML–RARα(S) and PML–RARα(L) (this study), which may lead to similar and extremely high rates of complete remission achieved by ATRA therapy for APL cases with both PML–RARα(S) and PML–RARα(L) [reviewed in [3,8]].

Interestingly, PML–RARα(S and L) showed a residual ATRA-dependent interaction with mutant MED1 having no intact NR motifs in our sensitive mammalian two-hybrid assays (Figure 2). One explanation is the existence of a hitherto unknown ATRA-dependent interface within MED1. A putative intermediating factor that binds to MED1 at a domain other than NR motifs and, simultaneously, to the RARα portion of PML–RARα in an ATRA-dependent fashion may be proposed. One candidate for this factor might be CCAR1, which was originally discovered as a bridging coactivator bypassing estrogen receptors to MED1 [24] and was subsequently implicated in bridging other nuclear receptors and other activators to MED1 [reviewed in [25]].

MED1 as coactivator for PML–RARα: implications in APL cell differentiation therapy

This study establishes MED1 as a coactivator for PML–RARα. However, the coactivation function of MED1 is weak and requires a thousand-fold molar excess of ATRA to induce transcriptional levels comparable to those observed using intact RARα at physiological ATRA doses. Therefore, PML–RARα-initiated transactivation requires pharmacological ATRA concentrations. The fact that the affinity of ATRA for PML–RARα is close to that of intact RARα [6] suggests that all PML–RARα molecules within an oligoheteromer complex must be liganded for transcription initiation.

According to the classical model of ATRA therapy in APL, treatment with pharmacological doses of ATRA converts the constitutive repressor oncoprotein PML–RARα to an activator, dissociates corepressors, recruits histone acetyltransferases, and restores differentiation of APL cells [reviewed in [7]]. APL cells indeed differentiate under this condition, as shown in in vitro cell cultures [13], and MED1 is involved in this process, i.e., rapid APL cell clearance in remission induction therapy. In the view that the lethal bleeding tendency that characterizes APL should be alleviated as soon as possible, the ATRA-mediated rapid differentiation is indeed welcomed at the remission induction stage. However, according to the revised model, eradication of APL stem cells (leading to the subsequent cure of the disease) requires degradation of PML–RARα and restoration of PML nuclear bodies, which are achieved by ATRA and arsenic trioxide [reviewed in [7–9]]. MED1, nevertheless, still plays a fundamental role in the revised model, as the restored intact RARα requires MED1 to resume transcription activation.

In conclusion, MED1 is required for optimal transactivation by PML–RARα. The coactivation function of MED1 requires pharmacological ATRA concentrations and depends upon the interaction of MED1 NR motifs with PML–RARα.

Materials and methods

Plasmids

The mammalian expression vector containing Gal4–hRARα in pCDM8 (Invitrogen) (pGal–hRARα) was described [20]. The cDNA for truncated hPML encoded by PML exons 1–6 was prepared by reverse transcriptase-PCR (RT-PCR) using the ReverTra Ace qPCR RT kit and KOD FX (Toyobo, Japan). PML cDNA was either cloned in pcDNA3.1(+) (Thermo Fisher) (pcDNA3.1–hPML), or fused to Gal4 DBD and cloned in pCDM8 (pGal–hPML). hRARα and hRXRα cDNAs were cloned in pcDNA3.1(+) (pcDNA3.1–hRARα and pcDNA3.1–hRXRα). The cDNA encoding fusion oncoprotein hPML–RARα(S or L) was prepared by 2-step PCR: the first step involved the use of the 5ʹ-forward primer of hPML and the 3ʹ-reverse primer encoding the chimeric sequences of each of the translocation sites of PML and RARα cDNAs, with PML cDNA as a template, and the use of the 5ʹ-forward primer encoding each corresponding reverse chimeric sequence and the 3ʹ-reverse primer of hRARα with RARα cDNA as a template; the second step involved mixing these PCR products and performing PCR without adding primers. The amplicons that encoded chimeric cDNAs were either cloned in pcDNA3.1(+) (pcDNA3.1–hPML–RARα(S and L)) or fused to Gal4 DBD and cloned in pCDM8 (pGal–hPML–RARα(S and L)). The Gal4-luciferase reporter consisted of an SV40 promoter-luciferase reporter pGL3 (Promega) with five Gal4-binding sites, as described [20]. For the 3× RARE luciferase reporter, an adaptor pair encoding the hRARβ2 promoter (−54/-28), 5ʹ-tcgagaagggttcaccgaaagttcactcgcataagggttcaccgaaagttcactcgcataagggttcaccgaaagttcactcgcata-3ʹ and 5ʹ-agcttatgcgagtgaactttcggtgaacccttatgcgagtgaactttcggtgaacccttatgcgagtgaactttcggtgaacccttc-3ʹ, was inserted in the luciferase reporter pGL4.10 (Promega).

For mammalian two-hybrid assays, Gal4–hMED1(wild-type) in pCDM8 (pGal4–hMED1) was described previously [26]. Mutant MED1 cDNAs with the first NR LxxLL motif mutated to encode LxxAA (hMED1(mutant I)), the second NR motif mutated to LxxAA (hMED1(mutant II)), and both NR motifs mutated to LxxAA (hMED1(mutant I+II)), were prepared by site-directed mutagenesis according to the manufacturer’s protocol (Agilent). Mutant MED1 cDNAs were fused to Gal4 DBD and cloned in pCDM8 to generate pGal4–hMED1(mutant I), pGal4–hMED1(mutant II) and pGal4–hMED1(mutant I+II). cDNAs encoding hPML–RARα(S and L) were fused to VP16 and subcloned into pcDNA3.1(+) to generate pVP16–hPML–RARα(S and L).

Generation of MEFs

Stable lines of Med1+/+ p53−/- and Med1−/− p53−/− MEFs, and primary Med1 LxxAA MEFs, were described previously [23,26]. These MEFs were derived from embryonic day (E) 10.5 or 11.5. Animal experiments were performed according to the institutional guidelines of the Animal Research Center, Kobe University, Japan.

Cell culture

MEFs and human embryonic kidney epithelial cell line HEK 293T cells were cultured in Dulbecco’s modified Eagle’s Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) under 5% CO2 at 37°C in a humidified atmosphere.

Transient transfection and luciferase reporter assays

Cells were plated in 24-well plates at 2.0 × 104 cells/well in DMEM supplemented with 10% charcoal-stripped FBS (Biological Industries, Israel). MEFs were used for luciferase reporter assays. For Gal4-reporter-based luciferase reporter assays, mammalian expression vectors for Gal4 DBD-fused proteins (100 ng or the indicated amounts) and Gal4-luciferase reporter (100 ng) were cotransfected with the pRL-CMV control vector (5 ng, Promega) using Lipofectamine 2000 Reagent (Thermo Fisher). For RARE-reporter-based assays, pcDNA3.1–hPML, pcDNA3.1–hRARα, pcDNA3.1–hPML–hRARα(S) or pcDNA3.1–hPML–hRARα(L) (50 ng), with or without pcDNA3.1–hRXRα (50 ng), and 3× RARE luciferase reporter were transfected together with the pRL-CMV control vector (5 ng). ATRA (0, 10−9, 10−8, 10−7, 10−6, or 10−5 M) was then added, and 24 h after transfection, firefly and Renilla luciferase activities were measured using the Dual-Luciferase Reporter Assay System (Promega). Values were normalized by Renilla luciferase activities.

For mammalian two-hybrid assays, Gal4 DBD-fused proteins (50 ng) and VP-16-fused proteins (50 ng) were transfected into HEK 293T cells together with the Gal4-reporter (100 ng) and pRL-CMV control vector (5 ng), and cultured without or with various concentrations of ATRA. Luciferase activities were likewise measured 24 h after transfection.

All numerical data (N = 3) are presented as average ± S.D.

Funding Statement

This work was supported by the Japan Society for the Promotion of Science [26460677 and 17K09012]; NIH [R01-DK071900].

Acknowledgements

We thank members of our laboratories for helpful discussions and suggestions.

Disclosure statement

No potential conflict of interest was reported by the authors.

References

  • [1].Kakizuka A, Miller WH Jr, Umesono K, et al. Chromosomal translocation t(15;17) in human acute promyelocytic leukemia fuses RARα with a novel putative transcription factor, PML. Cell. 1991;66(4):663–674. [DOI] [PubMed] [Google Scholar]
  • [2].de Thé H, Lavau C, Marchio A, et al. The PML-RARα fusion mRNA generated by the t(15;17) translocation in acute promyelocytic leukemia encodes a functionally altered RAR. Cell. 1991;66(4):675–684. [DOI] [PubMed] [Google Scholar]
  • [3].Licht JD. Acute promyelocytic leukemia–weapons of mass differentiation. N Engl J Med. 2009;360(9):928–930. [DOI] [PubMed] [Google Scholar]
  • [4].Pandolfi PP, Alcalay M, Fagioli M, et al. Genomic variability and alternative splicing generate multiple PML/RARα transcripts that encode aberrant PML proteins and PML/RARα isoforms in acute promyelocytic leukaemia. Embo J. 1992;11(4):1397–1407. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [5].van Dongen JJ, Macintyre EA, Gabert JA, et al. Standardized RT-PCR analysis of fusion gene transcripts from chromosome aberrations in acute leukemia for detection of minimal residual disease. Report of the BIOMED-1 Concerted Action: investigation of minimal residual disease in acute leukemia. Leukemia. 1999;13(12):1901–1928. [DOI] [PubMed] [Google Scholar]
  • [6].Benedetti L, Levin AA, Scicchitano BM, et al. Characterization of the retinoid binding properties of the major fusion products present in acute promyelocytic leukemia cells. Blood. 1997;90(3):1175–1185. [PubMed] [Google Scholar]
  • [7].de Thé H, Chen Z. Acute promyelocytic leukaemia: novel insights into the mechanisms of cure. Nat Rev Cancer. 2010;10(11):775–783. [DOI] [PubMed] [Google Scholar]
  • [8].de Thé H, Pandolfi PP, Chen Z. Acute promyelocytic leukemia: a paradigm for oncoprotein-targeted cure. Cancer Cell. 2017;32(5):552–560. [DOI] [PubMed] [Google Scholar]
  • [9].Saeed S, Logie C, Stunnenberg HG, et al. Genome-wide functions of PML-RARα in acute promyelocytic leukaemia. Br J Cancer. 2011;104(4):554–558. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [10].Lin RJ, Nagy L, Inoue S, et al. Role of the histone deacetylase complex in acute promyelocytic leukaemia. Nature. 1998;391(6669):811–814. [DOI] [PubMed] [Google Scholar]
  • [11].Grignani F, De Matteis S, Nervi C, et al. Fusion proteins of the retinoic acid receptor-α recruit histone deacetylase in promyelocytic leukaemia. Nature. 1998;391(6669):815–818. [DOI] [PubMed] [Google Scholar]
  • [12].Martens JH, Brinkman AB, Simmer F, et al. PML-RARα/RXR alters the epigenetic landscape in acute promyelocytic leukemia. Cancer Cell. 2010;17(2):173–185. [DOI] [PubMed] [Google Scholar]
  • [13].Ablain J, Leiva M, Peres L, et al. Uncoupling RARA transcriptional activation and degradation clarifies the bases for APL response to therapies. J Exp Med. 2013;210(4):647–853. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Kornberg RD. Mediator and the mechanism of transcriptional activation. Trends Biochem Sci. 2005;30(5):235–239. [DOI] [PubMed] [Google Scholar]
  • [15].Malik S, Roeder RG. The metazoan Mediator co-activator complex as an integrative hub for transcriptional regulation. Nat Rev Genet. 2010;11(11):761–772. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [16].Allen BL, Taatjes DJ. The Mediator complex: a central integrator of transcription. Nat Rev Mol Cell Biol. 2015;16(3):155–166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Chen W, Roeder RG. Mediator-dependent nuclear receptor function. Semin Cell Dev Biol. 2011;22(7):749–758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [18].Ito M, Roeder RG. The TRAP/SMCC/Mediator complex and thyroid hormone receptor function.Trends. Endocrinol Metab. 2001;12(3):127–134. [DOI] [PubMed] [Google Scholar]
  • [19].Urahama N, Ito M, Sada A, et al. The role of transcriptional coactivator TRAP220 in myelomonocytic differentiation. Genes Cells. 2005;10(12):1127–1137. [DOI] [PubMed] [Google Scholar]
  • [20].Ito M, Yuan CX, Okano HJ, et al. Involvement of the TRAP220 component of the TRAP/SMCC coactivator complex in embryonic development and thyroid hormone action. Mol Cell. 2000;5(4):683–693. [DOI] [PubMed] [Google Scholar]
  • [21].Umesono K, Murakami KK, Thompson CC, et al. Direct repeats as selective response elements for the thyroid hormone, retinoic acid, and vitamin D3 receptors. Cell. 1991;65(7):1255–1266. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [22].Shao W, Rosenauer A, Mann K, et al. Ligand-inducible interaction of the DRIP/TRAP coactivator complex with retinoid receptors in retinoic acid-sensitive and -resistant acute promyelocytic leukemia cells. Blood. 2000;96(6):2233–2239. [PubMed] [Google Scholar]
  • [23].Jiang P, Hu Q, Ito M, et al. Key roles for MED1 LxxLL motifs in pubertal mammary gland development and luminal-cell differentiation. Proc Natl Acad Sci USA. 2010;107(15):6765–6770. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Kim JH, Yang CK, Heo K, et al. CCAR1, a key regulator of mediator complex recruitment to nuclear receptor transcription complexes. Mol Cell. 2008;31(4):510–519. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [25].Mizuta S, Minami T, Fujita H, et al. CCAR1/CoCoA pair-mediated recruitment of the Mediator defines a novel pathway for GATA1 function. Genes Cells. 2014;19(1):28–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [26].Sumitomo A, Ishino R, Urahama N, et al. The transcriptional mediator subunit MED1/TRAP220 in stromal cells is involved in hematopoietic stem/progenitor cell support through osteopontin expression. Mol Cell Biol. 2010;30(20):4818–4827. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Transcription are provided here courtesy of Taylor & Francis

RESOURCES