Activating signal cointegrator-2 is an essential adaptor to recruit histone H3 lysine 4 methyltransferases MLL3 and MLL4 to the liver X receptors

Abstract

Activating signal cointegrator-2 (ASC-2), a coactivator of multiple nuclear receptors and transcription factors, including the liver X receptors (LXRs), is associated with histone H3 lysine 4 (H3K4) methyltransferase (H3K4MT) MLL3 or its paralogue MLL4 in a steady-state complex named ASCOM (ASC-2 complex). ASCOM belongs to Set1-like complexes, a conserved family of related H3K4MT complexes. ASC-2 binds to many nuclear receptors in a ligand-dependent manner through its two LXXLL motifs. In particular, the second motif has been shown to specifically recognize LXRs. However, the exact role for neither ASC-2 nor MLL3/4 in LXR transactivation is clearly defined. Here, we show that the key function of ASC-2 in transactivation by LXRs is to present MLL3 and MLL4 to LXRs. Thus, ASC-2 is required for ligand-induced recruitment of MLL3 and MLL4 to LXRs, and LXR ligand T1317 induces not only expression of LXR-target genes but also their H3K4-trimethylation. Strikingly, both of these ligand effects are ablated in ASC-2-null cells but only partially suppressed in cells expressing an enzymatically inactivated mutant MLL3. Our results also reveal that transactivation by LXRs does not appear to require other Set1-like complexes. Taken together, these results suggest that ASCOM-MLL3 and ASCOM-MLL4 play redundant but essential roles in ligand-dependent H3K4 trimethylation and expression of LXR-target genes, and that ASC-2 is likely a key determinant for LXRs to function through ASCOM but not other Set1-like complexes.

Figure 1

ASC-2 Is Essential for LXR Transactivation A, Set1-like complexes. Unique subunits in each complex are denoted in bold. B, RT-PCR analysis of E9.5 MEFs from wild-type (wt) and ASC-2^−/− mice reveal that LXR-target genes SREBP-1c, ABCA1, and ABCG1 are enhanced in expression by T1317 in wild-type MEFs but not in ASC-2^−/− MEFs. ASC-2^−/− cells indeed lack ASC-2, as shown. Glyceraldehyde-3-phosphate dehydrogenase represents negative control. Similar results were obtained from three independent experiments. C, Q-PCR using mRNAs in panel A confirms the essentiality of ASC-2 in expression of LXR-target gene, ABCA1. D, HeLa cells stably expressing control siRNA or specific siRNA against ASC-2 (33) were immunoblotted with antibodies against ASC-2 and β-actin (control), and monitored for ABCA1 transcripts using Q-PCR. GAPDH, Glyceraldehyde-3-phosphate dehydrogenase.

Figure 2

ASC-2 Is Essential for H3K4-Trimethylation of LXR-Target Genes A, Transfection assays for LXRE:LUC reporter in wild-type (wt) and ASC-2^−/− cells established from E9.5 MEFs (15). B, ChIP assays using anti-ASC-2 antibody reveal that ASC-2 is recruited to the LXREs of ABCA1 in wild-type MEFs. Similar results were obtained in HEK293, HepG2, and HeLa cells (data not shown). C, ChIP assays using antitrimethylated H3K4 demonstrate that T1317-enhanced H3K4-trimethylation of ABCA1-LXREs is observed in wild-type but not in ASC-2^−/− MEFs. D, ChIP assays using anti-MLL3 and MLL4 antibodies show that ASC-2^−/− cells no longer support the T1317-enhanced recruitment of MLL3/4 to ABCA1-LXREs, which is observed readily in wild-type cells. MLL3 and MLL4 were equally expressed in both cell types, as measured by Q-PCR. Inputs indicate 2% of the reactions (B–D).

Figure 3

LXR Specifically Requires ASCOM But Not Other Set1-Like Complexes A, Sequences of five and six potential NR boxes found in MLL3 and MLL4, respectively. In the yeast two-hybrid assays, ERα interacts with MLL3-NR1, MLL4-NR1, and MLL4-NR5. In contrast, RARα and LXRα interact with none of these NR boxes. B, Expression of ABCG1 was readily enhanced by T1317 in both wild-type and Menin^−/− MEFs, indicating that Menin-containing MLL1/2 H3K3MT complexes are not essential for LXR transactivation. C, HepG2 cells were transfected with control siRNA, siRNA against MLL3 or MLL4, or siRNAs for both MLL3 and MLL4. Cells were treated 2 d after transfection with vehicle or 10 μm T1317 for 12 h, and they were then tested for FAS transcript levels by Q-PCR. Similar results were obtained in three independent experiments. con, Control; E2, estradiol.

Figure 4

MLL3 Is Involved with Expression of LXR-Target Genes A, The liver sections from 2-month-old wild-type (wt) and MLL3^Δ/Δ male mice (n = 3∼5) fed normal chow diet or high-fat diet for 2 months were analyzed with hematoxylin and eosin and oil-red-O staining. All MLL3^Δ/Δ mice accumulated less lipid droplets than wild-type mice. A set of representative images is shown. B, Analyses of total mRNA isolated from these mice by Q-PCR show that FAS and SREBP-1c are significantly down-regulated in MLL3^Δ/Δ mice under both conditions. Values are the mean ± se (*, P < 0.05; **, P < 0.01). C, T1317-enhanced expression of ABCA1, SREBP-1c, and FAS was only partially impaired in E13.5 MLL3^Δ/Δ MEFs. Similar results were obtained in two independent experiments. D, ChIP assays using antitrimethylated H3K4 demonstrate that both basal and T1317-enhanced level of H3K4-trimethylation of ABCA1-LXREs is significantly reduced, but still present, in E13.5 MLL3^Δ/Δ MEFs in comparison with E13.5 wild-type MEFs. In contrast, H3K4-trimethylation of Hoxa9 was readily detected in both cell types. Inputs indicate 2% of the reactions. H&E, Hematoxylin and eosin.

Figure 5

The Working Model This study, along with our previous results (24,25,29), suggest that ASC-2 tethers ASCOM-MLL3 and ASCOM-MLL4 to LXR-RXR heterodimer through ligand-dependent direct interactions of ASC-2-NR2 and LXR. This, in turn, leads to ligand-dependent H3K4 trimethylation (me) of LXR-target genes. Importantly, expression of LXR-target genes is correlated to their H3K4 trimethylation, supporting the important role for this histone H3 modification in LXR transactivation.