Emerin Suppresses Notch Signaling by Restricting the Notch Intracellular Domain to the Nuclear Membrane

Byongsun Lee a, Tae-Hee Lee b and Jaekyung Shim a,b* a Department of Molecular Biology, Sejong University, Seoul 05006, Republic of Korea b Laboratory for Cancer & Stem Cell Biology, Plant Engineering Institute, Sejong University, Seoul 05006, Korea

Contact Information: *All correspondence should be addressed to

Jaekyung Shim , Ph.D.

Department of Molecular Biology, Sejong University, 98, Kunja-Dong, Kwangjin-Gu, Seoul 05006, Korea Phone: 82-2-3408-3944, Fax: 82-2-3408-4336, E-mail: [email protected]

Running Title: Inhibition of Notch signaling by

Keywords: Emerin, Notch signaling, nuclear membrane

Abbreviations: CSL, CBF1/suppressor hairless/Lag-1; NEXT, Notch1 extracellular truncation;

NICD, Notch intracellular domain; ADAM, A disintegrin and metalloproteinase; CoA, Coactivators;

CoR, Corepressors; PM, Plasma membrane; NE, Nuclear envelope; NGPS, Néstor-

Guillermo progeria syndrome; LEM, LAP2-emerin-MAN1; TM, Transmembrane

ABSTRACT Emerin is an inner nuclear membrane that is involved in maintaining the mechanical integrity of the nuclear membrane. Increasing evidence supports the involvement of emerin in the regulation of expression; however, its precise function remains to be elucidated. Here, we show that emerin downregulated downstream of Notch signaling, which are activated exclusively by the Notch intracellular domain (NICD). Deletion mutant experiments revealed that the transmembrane domain of emerin is important for the inhibition of Notch signaling. Emerin interacted directly and colocalized with the NICD at the nuclear membrane. Emerin knockdown induced the phosphorylation of ERK and AKT, increased endogenous Notch signaling, and inhibited hydrogen peroxide-induced apoptosis in HeLa cells. Notably, the downregulation of barrier-to-autointegration factor (BAF) or lamin A/C increased Notch signaling by inducing the release of emerin into the cytosol, implying that nuclear membrane-bound emerin acts as an endogenous inhibitor of Notch signaling. Taken together, our results indicate that emerin negatively regulates Notch signaling by promoting the retention of the

NICD at the nuclear membrane. This mechanism could constitute a new therapeutic target for the treatment of emerin-related diseases. 1. Introduction

The nuclear lamina establishes the mechanical support for the nucleus and provides a platform for protein interactions that contribute to gene regulation, DNA replication, and genome stability [ 1- 4].

Nuclear that bind to the lamina include emerin, MAN1, lamin-associated polypeptide 2

(LAP2), and LEMD, which scaffold potentially hundreds of proteins [5, 6]. Multiple human diseases are caused by loss of individual nuclear lamina proteins, highlighting the importance of this network

[2, 7]. Emerin was originally identified as a 35 kDa protein encoded by the EMD gene, which is located on the human X-. Emery-Dreifuss muscular dystrophy (EDMD) consists of X- linked EDMD (X-EDMD) and autosomal dominant EDMD (AD-EDMD) [4]. X-EDMD is caused by mutations in EMD (encoding Emerin) located on chromosome Xq28 and AD-EDMD is caused by mutations in LMNA (encoding Lamin A) located on chromosome 1q11-q23 [8-10]. EDMD is characterized by skeletal muscle wasting and cardiac defects, and mutations in the emerin gene that cause X-EDMD result in the loss of the emerin protein [11, 12]; however, the role of emerin loss in this disease has not been precisely elucidated. In addition to its known function in supporting the mechanical integrity of the nuclear membrane, emerin plays a role in the regulation of . Emerin interacts with many proteins, including nuclear lamins, germ cell-less (GCL), nesprin-1α, and BAF [13-16]. In particular, GCL proteins localize to the nuclear envelope and bind directly to emerin with high affinity [14, 17, 18]. The emerin bound GCL can also interact with the

DP3 subunit of E2F-DP heterodimers and represses E2F-DP-dependent gene expression [17, 18].

However, emerin downregulation leads to the mislocalization of GCL from the nuclear envelope and the concomitant increase in E2F-DP-dependent gene expression, suggesting that emerin plays an important role in transcriptional repression by GCL. Similarly, the transcriptional repressor Btf (Bcl-

2-associated transcription factor), which is highly expressed in skeletal muscle [19, 20], interacts with emerin to exert its inhibitory role on transcription [21], suggesting that Btf may be relevant to EDMD.

These findings indicate that emerin can negatively modulate gene expression through the recruitment of transcriptional repressors.

Emerin also functions as a negative regulator of gene expression by trapping transcriptional activators at the nuclear membrane. For example, the direct interaction of emerin with β-catenin causes the downregulation of Wnt signaling [22]. In the absence of emerin, the levels of nuclear β- catenin increase, resulting in the upregulation of target genes [22, 23]. Emerin induces the nuclear envelope localization of LIM Domain Only 7 (LMO 7), a transcriptional activator for myogenic differentiation, and suppresses its transcriptional function [24]. The mislocalization of transcriptional activators could represent a new addition to the list of already well-known regulatory mechanisms of gene expression such as transcription, translation, and post-translational modification; however, few studies have explored this regulatory mechanism.

In the present study, we examined the role of emerin as a transcriptional regulator and assessed its effect on gene expression using a transcription factor profiling assay covering 84 genes. We identified

Notch signaling as a potential target pathway regulated by emerin. In Notch signaling, a Notch receptor interacts extracellularly with its canonical ligand on a contacting cell and is cleaved proteolytically by metalloproteinases and secretases. The NICD released from the receptor translocates to the nucleus, where it interacts with a CBF1/suppressor hairless/Lag-1 (CSL) family

DNA-binding protein, resulting in the initiation of transcription of Notch target genes involved in several biological functions, such as development, growth, differentiation, and survival [25].

Here, we show that emerin can modulate Notch signaling by inducing the retention of the NICD at the nuclear membrane; thus, emerin may participate in some genetic laminopathy disorders, at least in part through the modulation of Notch signaling.

2. Materials and methods

2.1. Cell culture and transfection

HeLa cells (American Type Culture Collection, Manassas, VA) were cultured in DMEM (Welgene,

South Korea) supplemented with 10% FBS and 1% penicillin-streptomycin (Welgene, South Korea).

Transfection was performed using the Lipofectamine 2000 reagent (Invitrogen, Grand Island, NY) according to the manufacturer's instruction. The transfected cells were cultured for 24–72 h, washed with DPBS, and harvested with lysis buffer (#FNN0011; Life technology, Grand Island, NY).

2.2. Antibodies

For immunoblotting, primary antibodies specific for emerin (1:5000, #sc-15378,), BAF (1:100, #sc-

166324), lamin A/C (1:5000, #sc-20681), ERK1 (1:3000. #sc-93), pERK (1:2500, #sc-7383), pAKT/Thr 308 (1:3000, #sc-16646), AKT (1:3000, #sc-8312), pSTAT3 (1:2500, #sc-8059), and

STAT3 (1:3000, #sc-482) were purchased from Santa Cruz Biotechnology (Santa Cruz Biotechnology,

Santa Cruz, CA). NICD antibody (1:3000, #4147S) was obtained from Cell Signaling Technology

(Cell Signaling Technology, Beverly, MA). Primary antibodies specific for HA (1:5000, #G036), Flag

(1:1000, #G191), α-tubulin (1:5000, #G094), and β-actin (1:5000, #G043) were purchased from

Applied Biological Materials (Applied Biological Materials, Richmond, BC, Canada). Primary antibodies specific for GAPDH (1:5000, #csb-ma000071m0m) and GST (1:3000, #csb- ma000031m0m) were purchased from Cusabio (Cusabio, Wuhan, China). HES1 (1:1000, #ab 5702) and HES5 (1:1000, #ab 5708) antibodies were purchased from Millipore (Millipore, Darmstadt,

Germany). FITC (1:500, # 209-095-082) and TRITC (1:500, # 209-025-082) antibodies were purchased from Jackson Immuno Research Laboratories (Jackson Immuno Research Laboratories,

West Grove, PA).

2.3. Plasmid constructs

Human emerin and NICD cDNA were provided by 21C Frontier Human Gene Bank (South Korea).

The emerin cDNA was amplified by PCR and inserted into the restriction enzyme sites of HA- pcDNA3 for biochemical studies. The amplified full-length NICD cDNA was inserted into the restriction enzyme sites of pEGFP-C1 (Clontech, Mountain View, CA) for immunocytochemistry.

Alternatively, both amplified genes were inserted into the restriction enzyme sites of pcDNA3 for biological assays. To generate the GST-NICD or GST-YTHDC1 fusion protein, the coding region of each cDNA was amplified by PCR and inserted into the restriction enzyme sites of pGEX-4T-1 (GE

Healthcare, Marlborough, MA). For construction of emerin deletion mutants or NICD deletion mutants, the corresponding regions were amplified by PCR and inserted into the restriction enzyme sites of HA-pcDNA3.

2.4. Transcription factor profiling assay

Total RNA was isolated using Trizol reagent (Life Technologies), and 1 μg of total RNA was used for cDNA synthesis. The human transcription factor profiling PCR array was performed according to the manufacturer’s protocol (#PAHS-075ZC-2; Qiagen, Valencia, CA). Data were obtained using the manufacturer’s software.

2.5. Luciferase reporter assay

The plasmid vectors for 4 × CSL-Luc (a four-time repeating section of the CSL target sequence,

CGTGGGAA, with the luciferase gene), HES1-Luc (-467 to +46 of the Hes1 promoter with the luciferase gene) and HES5-Luc (-800 to +32 of the Hes5 promoter with the luciferase gene) were kindly provided by Dr. HS Park (Chonnam National University, South Korea) [26]. Cells were grown to subconfluency in 6-well plates and transfected with 0.5 µg of luciferase plasmid (4 × CSL-Luc,

HES1-Luc or HES5-Luc). The transfected cells were cultured for 24–48 h, lysed in 5 × Reporter Lysis

Buffer (Promega, Madison, WI), and analyzed for luciferase activity with a luminometer (Promega).

The luciferase reporter activity in each sample was normalized to β-galactosidase (β-gal) activity

(Promega).

2.6. GST pull-down assay BL21 E. coli was transformed with vectors encoding GST, GST-YTHDC1 or GST-NICD, and incubated overnight in LB medium containing ampicillin (100 µg/ml). IPTG was added to the bacterial culture medium to a final concentration of 1 mM and incubated for 8 h at 22°C. After harvest, cells were resuspended in PBS containing 0.05% TritonX-100, 0.5 mg/ml lysozyme, 5 mM

PMSF, and 10 µg/ml aprotinin. Lysates were sonicated for 3 min and centrifuged at 15000 rpm for 30 min at 4°C. The bacterial supernatant was rocked overnight at 4°C with glutathione-agarose beads

(Peptron, South Korea), which were washed five times in PBS containing 0.1% Triton X-100. To verify the purification, the GST-fusion proteins containing beads were subjected to SDS-PAGE and stained with a Coomassie solution (#B8522, Sigma, St. Louis, MO). The GST, GST-NICD or GST-

YTHDC1 containing beads were incubated with HeLa cell lysates and rotated for 4 h at 4°C. The glutathione-agarose beads were then washed three times in PBS buffer containing 0.05% Triton X-

100 and resuspended in 5 × sample buffer. Proteins were released from beads by boiling 5 min, and subjected to western blot analysis with an anti-emerin antibody. Alternatively, methionine S35 labeled emerin protein was produced in vitro using the TNT T7-coupled reticulocyte lysate system according to the manufacturer’s instructions (Promega). The products were incubated with the bead-bound GST or GST-NICD by rotating for 4 h at 4°C. Samples were prepared according to the method described above and subjected to autoradiography.

2.7. Lentivirus production and generation of inducible NICD expressing cells

Lentivirus production was achieved using a transcriptional activator (FUW-M2rtTA; #20342), a doxycycline (Dox)-inducible lentiviral vector encoding NICD cDNA (Tet-O-FUW-NICD; #61540), psPAX2 (#12260), and pMD2.G (#12259), which were obtained from Addgene (Addgene, Cambridge

, MA). Three plasmids (FUW-M2rtTA, psPAX2, and pMD2.G, or Tet-O-FUW-NICD, psPAX2, and pMD2.G) were added to 200 µl of Opti-MEM at a ratio of 2:1.5:0.5 and mixed with 10 µl of polyethylenimine (1 mg/ml). After incubation for 30 min, an aliquot of the mixture containing 4 µg of plasmids was added to 293FT cells (R70007, Invitrogen). After transfection for 12 h, the medium was replaced with fresh DMEM, and samples were incubated for 2 days. The supernatant was harvested and passed through a filter with a pore size of 0.4 µm. The lentiviral supernatant containing FUW-

M2rtTA and Tet-O-FUW-NICD was added to HeLa cells at a ratio of 1:10 (cell:virus) in the presence of polybrene (8 µg/ml) for 24 h. The medium was replaced with fresh DMEM and cells were incubated for 12 h. Thereafter, cells were infected with a second aliquot of lentiviral supernatant at the same ratio.

2.8. Cellular fractionation and immunoblotting

Cellular fractionation was performed as described previously [27, 28]. Briefly, cells were washed twice with cold PBS, harvested by scraping, and incubated in hypotonic buffer (20 mM HEPES, pH

7.0, 10 mM KCl, 2 mM MgCl2, 0.5% NP-40, 1 mM Na3VO4, 10 mM NaF, 1 mM phenylmethanesulfonyl fluoride, 2 μg/ml aprotinin). After 10 min of incubation on ice, cells were homogenized by 15–20 strokes in a Dounce homogenizer. The homogenate was centrifuged for 5 min at 1,500×g to sediment nuclei. The supernatant was then centrifuged at 16,000×g for 20 min to separate cytosolic and total membrane fractions. After washing the nuclei pellets three times with hypotonic buffer, the pellets were incubated with lysis buffer and centrifuged at 16,000×g for 20 min to extract nuclear proteins. To obtain total cell lysates, cells were incubated with the lysis buffer and centrifuged at 16,000 × g for 20 min. For western blot analysis, proteins were resolved by SDS-

PAGE and transferred onto PVDF membranes (Millipore). Membranes were blocked for 1 h at room temperature with a solution of 5% nonfat milk powder or 3% BSA in TBS containing 0.05% Tween-

20 (TBST). The membranes were then incubated with primary antibody in blocking solution overnight at 4°C. The membranes were washed three times with TBST and incubated with secondary antibody for 1 h at room temperature. After washing three times with TBST, the membranes were developed using the ECL detection system (Bio-Rad, Hercules, CA).

2.9. Immunocytochemistry Hela cells were plated on glass cover slips and then transfected with 0.5 μg of vector or 100 nM of siRNA oligonucleotides. After incubation for 48 h, cells were fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100 in PBS and then incubated with blocking solution (Dako,

Glostrup Denmark). After incubation overnight with primary HA antibody (1:200, Applied Biological

Materials) in blocking solution, cells were washed and incubated with FITC or TRITC-conjugated secondary antibodies (1:200, Jackson Immuno Research laboratories) for 1 h at room temperature.

After staining with DAPI (Life technology), cells were observed under a confocal microscope (Leica

TCS SPE, Buffalo Grove, IL).

2.10. Quantitative RT-PCR (qRT-PCR)

Total RNA was isolated using the Trizol reagent, and 1 µg of total RNA was used for cDNA synthesis. cDNA was amplified using primer pairs for human Notch1 (forward 5′-

TACGTGTGCACCTGCCGGG-3′, reverse 5′-CGTTTCTGCAGGGGCTGGGG-3′), human HES1

(forward 5′-ATGACGGCTG CGCTGAGCA C-3′, reverse 5′-TAACGCCCTCGCACGT GAC

-3′), human HES5 (forward 5′-CCGGTGGTGGAGAAGATG-3′, reverse 5′-GACAGCCATCTCCA

GGATGT-3′), human emerin (forward 5'-TACCGAGCTGACCACC-3', reverse 5'-GACCAGGAAAA

GCAGCAG-3'), human BAF (forward 5'-ATGACAACCTCCCAAAAGCAC-3', reverse 5'-TCACAA

GAAGGCGTCGCACC-3'), human Lamin A/C (forward 5’-CGGTTCCCACCAAAGTTCA-3’, reverse 5’-CTCATCCTCGTCGTCCTCAA-3’) and human GAPDH (forward 5′-GTCGGAGTCAA

CGGATTTGG-3′, reverse 5′-AAAAGCAGCCCTGGTGACC-3′). qRT-PCR was performed using the

IQ5 (Bio-Rad) system. Reactions were amplified using the selective primers described above and an

EvaGreen 2× qRT-PCR MasterMix-iCycler (ABM) according to the manufacturer’s instruction.

2.11. Small interfering RNA (siRNA) Emerin (#1047199), BAF #1(#1011344), BAF #2(#1011345), Lamin (#1033333) or Negative control (#SN-1002) siRNA oligonucleotides were purchased from Bioneer (South Korea).

Transfection was performed with Lipofectamine RNAiMAX reagent (Life Technologies) in HeLa cells according to the manufacturer’s protocol. The nucleotide sequences for siRNA targeting were as follows: Emerin (sense sequence, 5′-CUCUGACUUGAAUUCACU-3′ and antisense, 5′-

AGUCGAAUUCAAGUCAG

AG-3′), BAF #1 (sense sequence, 5’-GCUCUAACUGGCUAGAAGU-3’ and antisense, 5’ACUU

CUGCCAGUUAGAGC-3’); BAF #2 (sense sequence, 5’-CAACGUGGAAUGUUUCUUU-3’ and antisense, 5’-AAAGAAACAUUCCACGUUG-3’), and Lamin (sense sequence, 5′- CUCUGACUUG

AAUUCACU-3′ and antisense, 5′-AGUCGAAUUCAAGUCAGAG-3′). Negative siRNA was non- targeting siRNA for human, mouse, and rat.

2.12. Flow cytometry

HeLa cells were transfected with siRNA oligonucleotides against emerin or control for 48 h. Cells were treated with 500 µM H2O2 for 12 h and harvested, then fixed with cold ethanol for 1 h at 4°C, washed twice with PBS, and treated with RNase A. After incubation with propidium iodide (PI) solution (Sigma) in the dark, samples were analyzed by flow cytometry (Becton Dickinson, Franklin

Lakes, NJ). Apoptosis was measured as the percentage of cells with a sub-G0/G1 DNA content in the

PI intensity-area histogram plot.

2.13. Statistical analysis

The results are presented as the mean ± S.D. Statistical significance was determined with the

Student’s t-test with a significance level of P <0.05. The data for transcription PCR array were presented as the mean of two independent experiments. 3. Results

3.1. Emerin regulates the expression of genes downstream of Notch signaling

To evaluate whether emerin can act as a transcriptional regulator, we performed a transcription profiling assay covering 84 genes using qRT-PCR. From this experiment, we observed that emerin overexpression downregulated the transcription of most of the genes examined, including genes regulated by the Wnt and STAT pathways (Sup. Table 1), suggesting that emerin acts as a negative regulator of gene expression. This result was consistent with a previous report showing that Wnt signaling is negatively regulated by emerin [22]. Interestingly, we also found that emerin suppressed the expression of CREBBP and HDAC1 (Sup. Table 1), which are co-regulators of Notch signaling

[29, 30]. With the previous reports showing crosstalk between Wnt and Notch signaling [31-33], these observations suggest the possibility that emerin can act as a negative regulator against Notch signaling. To address this issue, cDNA encoding the full-length NICD, which is a Notch effector [34], was introduced into HeLa cells to specifically stimulate the Notch pathway. To assay Notch activation, we used qRT-PCR analysis or a promoter assay system that detects the transcription of mammalian hairy enhancer of split 1 (Hes1), Hes5 and Notch1, which harbor multiple CSL-binding

DNA sequences in their promoters and are essential targets of Notch signaling in epithelial cells [35].

Our results showed that the NICD significantly upregulated the expression of Hes1 and Hes5 mRNA downstream of the Notch pathway, and emerin suppressed the NICD-induced mRNA expression of

Notch1, Hes1, and Hes5 (Fig. 1A). Emerin also inhibited the NICD-induced promoter activation of

Notch target genes (Fig. 1B). In addition, emerin had a strong inhibitory effect on the endogenous protein expression of Hes1 and Hes5 (Fig. 1C), indicating that Notch signaling can be negatively regulated.

3.2. The transmembrane (TM) domain of emerin is necessary for the inhibition of Notch signaling

To study the molecular mechanism underlying the downregulation of Notch target genes by emerin, several deletion mutants of emerin were generated and linked to HA tags (Fig. 2A). The expression and localization of these mutants were verified by western blotting (Fig. 3F) and immunocytochemistry (Fig. 2B). Promoter assays for Notch target genes were used to examine the effect of these mutants on NICD-induced Notch activation. Our results showed that D2 and D3 mutants, as well as wild-type emerin, significantly inhibited NICD-induced promoter activation, whereas the other mutants had no effect (Fig. 2C). The inhibitory effect of emerin on NICD-induced

Notch activation was abolished by deletion of the TM domain (D1, D4, D5, and D6) or deletion of the lamin binding region (D6 and D7). Assessment of the effect of the mutants on the transcription of

Notch target genes showed that only mutants (D2 and D3) containing the TM domain and the lamin binding region of emerin had an inhibitory effect on the expression of endogenous Notch genes, similar to that of wild-type emerin (Fig. 2D). These data suggest that the localization of emerin proteins to the inner nuclear membrane is important for the suppression of Notch signaling.

3.3. Emerin proteins directly interact with the NICD

The NICD is released from Notch by intracellular cleavage within the plasma membrane and translocated into the nucleus to exert its biological activity as a transcriptional modulator [25]. The fact that emerin can inhibit Wnt signaling by limiting the subcellular localization of β-catenin to the nuclear membrane [22], together with our results showing that the TM domain and the lamin binding region of emerin are critical for the suppression of Notch signaling, led us to hypothesize that emerin prevents NICD binding to the promoter region of target genes by capturing the NICD at the nuclear membrane. To do so, emerin would have to interact directly or indirectly with NICD proteins. To examine this possibility, we constructed a vector encoding GST-NICD or GST-YTHDC1, expressed it in E. coli, and purified the fusion proteins using an affinity column. YTHDC1 was used as a positive control because it interacts with emerin [36].

Incubation of the purified recombinant proteins with crude lysates of HeLa cells showed that both the NICD and YTHDC1 interacted with emerin (Fig. 3A). To exclude the possibility that the NICD interacted indirectly with emerin via other molecules in crude cell lysates, we performed an in vitro binding assay using a reticulocyte translation system. This assay can be used to demonstrate direct interactions between proteins [37]. When emerin proteins were translated from rabbit reticulocytes and subjected to autoradiography, the predicted molecular weight (35 Kda) of emerin was observed as a main band (Fig. 3B, lane 3). A GST pull-down assay with the purified GST-NICD and the in vitro translation products of emerin synthesis showed that the NICD interacted directly with emerin (Fig.

3B, lane 5). Furthermore, transient overexpression of the NICD and emerin in HeLa cells resulted in the colocalization of these two molecules in the nuclear membrane region (Fig. 3C). To examine whether emerin can interact endogenously with the NICD, HeLa cell lysates were incubated with an anti-emerin antibody. As shown in Figure 3D (right panel), this antibody precipitated the NICD proteins, indicating that emerin interacts with the endogenous NICD. Again, when lysates were incubated with an anti-NICD antibody, we observed that antibody against the NICD precipitated emerin protein (Fig. 3D, left panel), clearly showing that both of these molecules interact endogenously in HeLa cells. To determine the NICD region responsible for the emerin-NICD interaction, we performed an immunoprecipitation assay with NICD deletion mutants. The experiments showed that the N-terminal region of NICD is necessary for its interaction with emerin

(Fig 3E). Moreover, to determine the emerin region responsible for the emerin-NICD interaction, we performed a GST pull-down assay with purified GST-NICD and the emerin deletion mutants. The experiments revealed that the lamin binding region (LB, grey color in Fig. 2A) of emerin is essential for its interaction with NICD because D6 and D7 mutants failed to interact with the NICD (Fig. 3F).

Furthermore, together with the results showing that deletion of the lamin binding region blocked the inhibition of Notch by the emerin D3 mutant (Fig. 2C and D), these data strongly support the notion that emerin inhibits Notch signaling by limiting the recruitment of the NICD to target promoters through its direct interaction with the NICD.

3.4. Emerin has no effect on the stability of NICD proteins Notch signaling is modulated post-translationally through the regulation of the stability of its functional mediator, the NICD [38-40]. We therefore examined the possibility that emerin negatively regulates Notch signaling by modulating the stability of NICD proteins. For these experiments, HeLa cells were transfected with a pcDNA vector encoding NICD cDNA and selected with geneticin for stable expression of the NICD. However, HeLa cells stably expressing the NICD showed growth retardation and senescent-like morphology (data not shown), consistent with a previous report in which overexpression of the NICD induced growth-arrest in HeLa cells through oncogene-induced senescence [41]. We therefore generated HeLa cells expressing doxycycline (Dox) inducible NICD using lentivirus encoding rtTA and Tet-NICD. After 2 days of Dox treatment, cells were transiently transfected with vectors encoding emerin for another 2 days, incubated with cycloheximide (CHX) for the indicated times, and the lysed to subject western blotting. CHX is used to study protein stability because it blocks protein synthesis through interaction with the translocase enzyme in eukaryotic cells

[42, 43]. The results showed a time-dependent decrease in the levels of NICD proteins; however, the decrease in protein levels in the presence of CHX was not further affected by emerin (Sup Fig 1A).

We observed a similar result by using HeLa cells transfected with a pcDNA vector encoding emerin cDNA without the NICD (Sup Fig 1B). These data indicated that emerin had no effect on the stability of exogenous or endogenous NICD proteins.

3.5. Endogenously expressing emerin modulates Notch signaling and survival in HeLa cells

To determine whether endogenously expressed emerin modulates Notch signaling, we performed loss-of-function experiments using siRNA against emerin. Treatment of HeLa cells with siRNA targeting emerin effectively downregulated the expression of emerin (Fig. 4A and C). Knockdown of emerin significantly upregulated the expression of Notch target genes at the transcriptional and translational levels (Fig. 4B and C). Furthermore, the siRNA oligonucleotides against emerin increased the phosphorylation level of ERK and AKT compared with that in control transfected cells (Fig. 4C), suggesting that the downregulation of emerin has a favorable effect on cell growth or survival. To test this possibility, HeLa cells were treated with H2O2 to induce apoptotic cell death, and the effect of emerin downregulation was investigated. As shown in Figure 4D, H2O2 significantly induced apoptosis in HeLa cells, and this effect was suppressed by emerin downregulation.

Furthermore, H2O2-induced cell death was suppressed significantly by transient expression of the

NICD. Taken together, these results suggest that emerin plays a role in HeLa cell survival through, at least in part, the modulation of Notch signaling.

3.6. Knockdown of BAF increases Notch signaling via the mislocalization of emerin

Emerin localizes predominantly to the inner nuclear membrane by binding to A-type lamins

(nuclear intermediate filament proteins) and a chromatin protein, BAF [44]. BAF is essential for the localization of emerin to the nuclear envelope because in the absence of BAF emerin is sequestered in the cytosol [45, 46]. Thus, loss of BAF may increase Notch signaling by inducing mislocalization of emerin to the cytosol. To examine this possibility, we performed loss-of-function experiments using siRNA against BAF. When HeLa cells were treated with siRNA targeting BAF, BAF expression decreased (Fig. 5A and C) and emerin was released into the cytosol (Fig. 5D and E). As expected, knockdown of BAF significantly upregulated the expression of Notch target genes at the transcriptional and translational levels (Fig. 5B and C). Lamin A/C is also essential for the localization of emerin to the nuclear envelope via its direct binding to emerin [47]. To confirm the function of

Lamin A/C in Notch signaling, we performed Lamin A/C loss-of-function experiments. The results

(Fig. 6A-D) were similar to those obtained in the BAF loss-of-function experiments. The effect of simultaneous knockdown of both BAF and Lamin A/C was similar to that of knockdown of BAF or

Lamin A/C alone (Fig. 6E and F). These results further demonstrated that retention of emerin at the nuclear membrane is essential for suppression of Notch signaling (Fig. 6G).

4. Discussion In the present study, we explored the role of emerin on Notch signaling and found that emerin can act as a negative modulator against Notch signaling. Specifically, we demonstrated that emerin suppressed Notch signaling through direct interaction with NICD domain of Notch and the TM domain of emeirn was essential for inhibition of Notch signaling. On the other hand, the downregulation of endogenous emerin activated Notch signaling and induced the activation of signaling molecules such as ERK and AKT, which promote cancer cell survival. Our data suggested that emerin modulates Notch signaling through a mechanism similar to that mediating its inhibition of the transcription factor β-catenin or Lmo7 [22, 24]. In Notch signaling, the NICD directly binds the autoregulatory CSL sites in the Notch and can regulate its own transcription [48]. In this way,

Notch signaling can be switched on or off according to the availability of the NICD, suggesting that the NICD is finely regulated at the transcriptional and translational level, as well as by post- translational modification. These regulatory pathways support the importance of our data showing the role of the spatial control of the NICD in the modulation of Notch signaling. Although the mechanism of modulation of Notch signaling by emerin is not a novel concept, our findings suggest a new paradigm in the regulation of gene expression, as shown previously by the role of emerin in the regulation of β-catenin or Lmo7-mediated signaling.

An important question is the pathological or physiological meaning of Notch signaling modulation by emerin. Emerin is anchored at the nuclear membrane via its interactions with A-type lamins and

BAF. A homozygous missense mutation (A12T) in human BAF destabilizes the protein and subsequently induces the mislocalization of emerin, suggesting that this mechanism may underlie the pathogenesis of Néstor-Guillermo progeria syndrome (NGPS) [46, 49]. NGPS patients are characterized by pathological features such as accelerated skin aging, lipoatrophy, osteoporosis, and osteolysis [50]. These symptoms are also found in Notch-related diseases [51-53]. A recent study reported that fibroblasts from NGPS patients have higher levels of NF-kB activity than normal fibroblasts [54]. Notch signaling increases NF-kB activity [55]. In addition, we found that knockdown of BAF significantly upregulated Notch activity in HeLa cells. Therefore, it is possible that the cytosolic sequestration of emerin caused by loss of BAF could be involved in NGPS via activation of Notch signaling (Fig. 6G). Investigating the role of Notch signaling as a clinical target for NGPS patients would be of interest.

In conclusion, we showed that nuclear membrane-bound emerin can act as an endogenous inhibitor of Notch signaling by restricting the NICD to the nuclear membrane. Our findings suggest that the Notch pathway is involved in emerin-related pathologies through a unique system of regulation of spatial localization.

Acknowledgements

This work was supported by Basic Science Research Program through the National Research

Foundation of Korea (NRF) funded by the Korea government (MSIP) (2009-0075847) (to J. K. S.), the Ministry of Education (NRF-2013R1A1A2065820) (to T. H. L.), and the Bio & medical

Technology Development Program of the National Research Foundation (NRF) funded by the

Ministry of Science, ICT & Future Planning (NRF-2015M3A9B6073690) (to T. H. L.)

Competing interests

The authors declare that they have no conflicts of interest with the contents of this article.

Author contributions

B. S. L., acquisition, analysis and interpretation of data, writing of the manuscript; T. H. L., interpretation of data, writing of the manuscript; J. S., conception, design, writing of the manuscript References

[1] T. Shimi, V. Butin-Israeli, S.A. Adam, R.D. Goldman, Nuclear lamins in cell regulation and disease, Cold Spring Harbor symposia on quantitative biology, 75 (2010) 525-531. [2] P.M. Davidson, J. Lammerding, Broken nuclei--lamins, nuclear mechanics, and disease, Trends in cell biology, 24 (2014) 247-256. [3] L. Gerace, M.D. Huber, Nuclear lamina at the crossroads of the cytoplasm and nucleus, Journal of structural biology, 177 (2012) 24-31. [4] A.J. Koch, J.M. Holaska, Emerin in health and disease, Seminars in cell & developmental biology, 29 (2014) 95-106. [5] E.C. Schirmer, L. Gerace, The nuclear membrane proteome: extending the envelope, Trends in biochemical sciences, 30 (2005) 551-558. [6] N. Korfali, G.S. Wilkie, S.K. Swanson, V. Srsen, J. de Las Heras, D.G. Batrakou, P. Malik, N. Zuleger, A.R. Kerr, L. Florens, E.C. Schirmer, The nuclear envelope proteome differs notably between tissues, Nucleus, 3 (2012) 552-564. [7] L.J. Barton, A.A. Soshnev, P.K. Geyer, Networking in the nucleus: a spotlight on LEM- domain proteins, Current opinion in cell biology, 34 (2015) 1-8. [8] M. Raffaele Di Barletta, E. Ricci, G. Galluzzi, P. Tonali, M. Mora, L. Morandi, A. Romorini, T. Voit, K.H. Orstavik, L. Merlini, C. Trevisan, V. Biancalana, I. Housmanowa-Petrusewicz, S. Bione, R. Ricotti, K. Schwartz, G. Bonne, D. Toniolo, Different mutations in the LMNA gene cause autosomal dominant and autosomal recessive Emery-Dreifuss muscular dystrophy, American journal of human genetics, 66 (2000) 1407-1412. [9] A. Jimenez-Escrig, I. Gobernado, M. Garcia-Villanueva, A. Sanchez-Herranz, Autosomal recessive Emery-Dreifuss muscular dystrophy caused by a novel mutation (R225Q) in the lamin A/C gene identified by exome sequencing, Muscle & nerve, 45 (2012) 605-610. [10] A. Helbling-Leclerc, G. Bonne, K. Schwartz, Emery-Dreifuss muscular dystrophy, European journal of human genetics : EJHG, 10 (2002) 157-161. [11] S. Bione, E. Maestrini, S. Rivella, M. Mancini, S. Regis, G. Romeo, D. Toniolo, Identification of a novel X-linked gene responsible for Emery-Dreifuss muscular dystrophy, Nature genetics, 8 (1994) 323-327. [12] A.E. Emery, Emery-Dreifuss muscular dystrophy - a 40 year retrospective, Neuromuscular disorders : NMD, 10 (2000) 228-232. [13] J.M. Mislow, J.M. Holaska, M.S. Kim, K.K. Lee, M. Segura-Totten, K.L. Wilson, E.M. McNally, Nesprin-1alpha self-associates and binds directly to emerin and lamin A in vitro, FEBS letters, 525 (2002) 135-140. [14] J.M. Holaska, K.K. Lee, A.K. Kowalski, K.L. Wilson, Transcriptional repressor germ cell- less (GCL) and barrier to autointegration factor (BAF) compete for binding to emerin in vitro, The Journal of biological chemistry, 278 (2003) 6969-6975. [15] J.M. Holaska, K.L. Wilson, M. Mansharamani, The nuclear envelope, lamins and nuclear assembly, Current opinion in cell biology, 14 (2002) 357-364. [16] K.K. Lee, T. Haraguchi, R.S. Lee, T. Koujin, Y. Hiraoka, K.L. Wilson, Distinct functional domains in emerin bind lamin A and DNA-bridging protein BAF, Journal of cell science, 114 (2001) 4567-4573. [17] E. Nili, G.S. Cojocaru, Y. Kalma, D. Ginsberg, N.G. Copeland, D.J. Gilbert, N.A. Jenkins, R. Berger, S. Shaklai, N. Amariglio, F. Brok-Simoni, A.J. Simon, G. Rechavi, Nuclear membrane protein LAP2beta mediates transcriptional repression alone and together with its binding partner GCL (germ-cell-less), Journal of cell science, 114 (2001) 3297-3307. [18] J.M. Holaska, K.L. Wilson, Multiple roles for emerin: implications for Emery-Dreifuss muscular dystrophy, The anatomical record. Part A, Discoveries in molecular, cellular, and evolutionary biology, 288 (2006) 676-680. [19] G.M. Kasof, L. Goyal, E. White, Btf, a novel death-promoting transcriptional repressor that interacts with Bcl-2-related proteins, Molecular and cellular biology, 19 (1999) 4390- 4404. [20] T. Nagase, N. Seki, K. Ishikawa, A. Tanaka, N. Nomura, Prediction of the coding sequences of unidentified human genes. V. The coding sequences of 40 new genes (KIAA0161-KIAA0200) deduced by analysis of cDNA clones from human cell line KG-1, DNA research : an international journal for rapid publication of reports on genes and genomes, 3 (1996) 17-24. [21] T. Haraguchi, J.M. Holaska, M. Yamane, T. Koujin, N. Hashiguchi, C. Mori, K.L. Wilson, Y. Hiraoka, Emerin binding to Btf, a death-promoting transcriptional repressor, is disrupted by a missense mutation that causes Emery-Dreifuss muscular dystrophy, European journal of biochemistry / FEBS, 271 (2004) 1035-1045. [22] E. Markiewicz, K. Tilgner, N. Barker, M. van de Wetering, H. Clevers, M. Dorobek, I. Hausmanowa-Petrusewicz, F.C. Ramaekers, J.L. Broers, W.M. Blankesteijn, G. Salpingidou, R.G. Wilson, J.A. Ellis, C.J. Hutchison, The inner nuclear membrane protein emerin regulates beta-catenin activity by restricting its accumulation in the nucleus, The EMBO journal, 25 (2006) 3275-3285. [23] A.J. Koch, J.M. Holaska, Loss of emerin alters myogenic signaling and miRNA expression in mouse myogenic progenitors, PloS one, 7 (2012) e37262. [24] Z. Dedeic, M. Cetera, T.V. Cohen, J.M. Holaska, Emerin inhibits Lmo7 binding to the Pax3 and MyoD promoters and expression of myoblast proliferation genes, Journal of cell science, 124 (2011) 1691-1702. [25] E.R. Andersson, R. Sandberg, U. Lendahl, Notch signaling: simplicity in design, versatility in function, Development, 138 (2011) 3593-3612. [26] M.Y. Kim, E.J. Ann, J.Y. Kim, J.S. Mo, J.H. Park, S.Y. Kim, M.S. Seo, H.S. Park, Tip60 histone acetyltransferase acts as a negative regulator of Notch1 signaling by means of acetylation, Molecular and cellular biology, 27 (2007) 6506-6519. [27] S.Y. Lin, K. Makino, W. Xia, A. Matin, Y. Wen, K.Y. Kwong, L. Bourguignon, M.C. Hung, Nuclear localization of EGF receptor and its potential new role as a transcription factor, Nature cell biology, 3 (2001) 802-808. [28] S.C. Hsu, S.A. Miller, Y. Wang, M.C. Hung, Nuclear EGFR is required for cisplatin resistance and DNA repair, American journal of translational research, 1 (2009) 249-258. [29] H.Y. Kao, P. Ordentlich, N. Koyano-Nakagawa, Z. Tang, M. Downes, C.R. Kintner, R.M. Evans, T. Kadesch, A histone deacetylase corepressor complex regulates the Notch signal transduction pathway, Genes & development, 12 (1998) 2269-2277. [30] C.J. Fryer, E. Lamar, I. Turbachova, C. Kintner, K.A. Jones, Mastermind mediates chromatin-specific transcription and turnover of the Notch enhancer complex, Genes & development, 16 (2002) 1397-1411. [31] T. Klein, A.M. Arias, The vestigial gene product provides a molecular context for the interpretation of signals during the development of the wing in Drosophila, Development, 126 (1999) 913-925. [32] Y.H. Jin, H. Kim, H. Ki, I. Yang, N. Yang, K.Y. Lee, N. Kim, H.S. Park, K. Kim, Beta-catenin modulates the level and transcriptional activity of Notch1/NICD through its direct interaction, Biochimica et biophysica acta, 1793 (2009) 290-299. [33] K. Yamamizu, T. Matsunaga, H. Uosaki, H. Fukushima, S. Katayama, M. Hiraoka-Kanie, K. Mitani, J.K. Yamashita, Convergence of Notch and beta-catenin signaling induces arterial fate in vascular progenitors, The Journal of cell biology, 189 (2010) 325-338. [34] E.H. Schroeter, J.A. Kisslinger, R. Kopan, Notch-1 signalling requires ligand-induced proteolytic release of intracellular domain, Nature, 393 (1998) 382-386. [35] T. Ohtsuka, M. Ishibashi, G. Gradwohl, S. Nakanishi, F. Guillemot, R. Kageyama, Hes1 and Hes5 as notch effectors in mammalian neuronal differentiation, The EMBO journal, 18 (1999) 2196-2207. [36] F.L. Wilkinson, J.M. Holaska, Z. Zhang, A. Sharma, S. Manilal, I. Holt, S. Stamm, K.L. Wilson, G.E. Morris, Emerin interacts in vitro with the splicing-associated factor, YT521-B, European journal of biochemistry / FEBS, 270 (2003) 2459-2466. [37] K.P. Mitton, P.K. Swain, S. Chen, S. Xu, D.J. Zack, A. Swaroop, The leucine zipper of NRL interacts with the CRX homeodomain. A possible mechanism of transcriptional synergy in rhodopsin regulation, The Journal of biological chemistry, 275 (2000) 29794- 29799. [38] L. Qiu, C. Joazeiro, N. Fang, H.Y. Wang, C. Elly, Y. Altman, D. Fang, T. Hunter, Y.C. Liu, Recognition and ubiquitination of Notch by Itch, a hect-type E3 ubiquitin ligase, The Journal of biological chemistry, 275 (2000) 35734-35737. [39] M.A. McGill, C.J. McGlade, Mammalian numb proteins promote Notch1 receptor ubiquitination and degradation of the Notch1 intracellular domain, The Journal of biological chemistry, 278 (2003) 23196-23203. [40] P. Chastagner, A. Israel, C. Brou, AIP4/Itch regulates Notch receptor degradation in the absence of ligand, PloS one, 3 (2008) e2735. [41] L. Wang, H. Qin, B. Chen, X. Xin, J. Li, H. Han, Overexpressed active Notch1 induces cell growth arrest of HeLa cervical carcinoma cells, International journal of gynecological cancer : official journal of the International Gynecological Cancer Society, 17 (2007) 1283-1292. [42] P. Zhou, Determining protein half-lives, Methods in molecular biology, 284 (2004) 67- 77. [43] J.S. Mo, M.Y. Kim, S.O. Han, I.S. Kim, E.J. Ann, K.S. Lee, M.S. Seo, J.Y. Kim, S.C. Lee, J.W. Park, E.J. Choi, J.Y. Seong, C.O. Joe, R. Faessler, H.S. Park, Integrin-linked kinase controls Notch1 signaling by down-regulation of protein stability through Fbw7 ubiquitin ligase, Molecular and cellular biology, 27 (2007) 5565-5574. [44] M. Segura-Totten, K.L. Wilson, BAF: roles in chromatin, nuclear structure and retrovirus integration, Trends in cell biology, 14 (2004) 261-266. [45] T. Haraguchi, T. Koujin, M. Segura-Totten, K.K. Lee, Y. Matsuoka, Y. Yoneda, K.L. Wilson, Y. Hiraoka, BAF is required for emerin assembly into the reforming nuclear envelope, Journal of cell science, 114 (2001) 4575-4585. [46] X.S. Puente, V. Quesada, F.G. Osorio, R. Cabanillas, J. Cadinanos, J.M. Fraile, G.R. Ordonez, D.A. Puente, A. Gutierrez-Fernandez, M. Fanjul-Fernandez, N. Levy, J.M. Freije, C. Lopez-Otin, Exome sequencing and functional analysis identifies BANF1 mutation as the cause of a hereditary progeroid syndrome, American journal of human genetics, 88 (2011) 650-656. [47] C.Y. Ho, D.E. Jaalouk, M.K. Vartiainen, J. Lammerding, Lamin A/C and emerin regulate MKL1-SRF activity by modulating actin dynamics, Nature, 497 (2013) 507-511. [48] Y. Yashiro-Ohtani, Y. He, T. Ohtani, M.E. Jones, O. Shestova, L. Xu, T.C. Fang, M.Y. Chiang, A.M. Intlekofer, S.C. Blacklow, Y. Zhuang, W.S. Pear, Pre-TCR signaling inactivates Notch1 transcription by antagonizing E2A, Genes & development, 23 (2009) 1665-1676. [49] R. Cabanillas, J. Cadinanos, J.A. Villameytide, M. Perez, J. Longo, J.M. Richard, R. Alvarez, N.S. Duran, R. Illan, D.J. Gonzalez, C. Lopez-Otin, Nestor-Guillermo progeria syndrome: a novel premature aging condition with early onset and chronic development caused by BANF1 mutations, American journal of medical genetics. Part A, 155A (2011) 2617-2625. [50] N. Paquet, J.K. Box, N.W. Ashton, A. Suraweera, L.V. Croft, A.J. Urquhart, E. Bolderson, S.D. Zhang, K.J. O'Byrne, D.J. Richard, Nestor-Guillermo Progeria Syndrome: a biochemical insight into Barrier-to-Autointegration Factor 1, alanine 12 threonine mutation, BMC molecular biology, 15 (2014) 27. [51] A. Louvi, S. Artavanis-Tsakonas, Notch and disease: a growing field, Seminars in cell & developmental biology, 23 (2012) 473-480. [52] M.J. Hilton, X. Tu, X. Wu, S. Bai, H. Zhao, T. Kobayashi, H.M. Kronenberg, S.L. Teitelbaum, F.P. Ross, R. Kopan, F. Long, Notch signaling maintains bone marrow mesenchymal progenitors by suppressing osteoblast differentiation, Nature medicine, 14 (2008) 306-314. [53] F. Engin, Z. Yao, T. Yang, G. Zhou, T. Bertin, M.M. Jiang, Y. Chen, L. Wang, H. Zheng, R.E. Sutton, B.F. Boyce, B. Lee, Dimorphic effects of Notch signaling in bone homeostasis, Nature medicine, 14 (2008) 299-305. [54] C. Soria-Valles, F.G. Osorio, A. Gutierrez-Fernandez, A. De Los Angeles, C. Bueno, P. Menendez, J.I. Martin-Subero, G.Q. Daley, J.M. Freije, C. Lopez-Otin, NF-kappaB activation impairs somatic cell reprogramming in ageing, Nature cell biology, 17 (2015) 1004-1013. [55] L. Poellinger, U. Lendahl, Modulating Notch signaling by pathway-intrinsic and pathway-extrinsic mechanisms, Current opinion in genetics & development, 18 (2008) 449-454.

Figure Legends

Fig. 1. Emerin regulates Notch signaling. (A) HeLa cells were co-transfected transiently with vectors encoding NICD, emerin-HA, and CSL-Luc (HES1-Luc or HES5-Luc) as indicated. Luciferase activity was measured in cell lysates and normalized to β-gal activity in the same sample. The results represent the mean ± S.D. of three independent experiments performed in triplicate. R.L.U., relative luciferase units. *, P<0.05 (B) HeLa cells were transiently transfected with vectors encoding the

NICD or emerin. Total RNA was isolated and subjected to qRT-PCR analysis. Data were normalized to GAPDH expression. The results represent the mean ± S.D. of three independent experiments performed in triplicate. *, P<0.05 (C) HeLa cells were transiently transfected with vectors encoding

HA-emerin for 48 h and subjected to western blot analysis. The membranes were analyzed using

ImageJ software (NIH, Bethesda).

Fig. 2. The transmembrane (TM) domain of emerin is essential for inhibition of Notch signaling. (A)

Strategy for construction of emerin deletion mutants. LEM, LAP2-emerin-MAN1; TM,

Transmembrane; LB, Lamin binding region (grey color) (B) The localization of emerin deletion mutants was analyzed by immunocytochemistry with anti-HA antibody. Scale bar = m (C) HeLa cells were co-transfected transiently with vectors encoding emerin deletion mutants and the NICD in combination with reporter vectors. Luciferase activity was measured in cell lysates and normalized to

β-gal activity in the same sample. The results represent the mean ± S.D. of three independent experiments performed in triplicate. R.L.U., relative luciferase units. *, P<0.05. (D) HeLa cells were transiently transfected with vectors encoding emerin deletion mutants. Total RNA was isolated and subjected to qRT-PCR analysis, and the results were normalized to GAPDH expression. The results represent the mean ± S.D. of three independent experiments performed in triplicate. R.L.U., relative luciferase units. *, P<0.05.

Fig. 3. Emerin interacts with the NICD. (A) The purified proteins were visualized by Coomassie staining. HeLa cell lysates were subjected to GST pull-down assay with purified GST, GST-NICD, and GST-YTHDC1. Bound proteins were analyzed by western blotting with anti-emerin antibody. (B)

The in vitro translated methionine 35S labeled emerin was subjected to GST pull-down assay with purified GST or GST-NICD. Bound proteins were analyzed by autoradiography. (C) HeLa cells were transiently co-transfected with vectors encoding GFP-NICD (green) and HA-emerin. Cells were fixed, stained with an antibody against HA (red), and observed under confocal microscopy. DAPI (blue) was used to visualize nuclei. Scale bar = 10 µm. The white arrows indicate colocalization of emerin and the NICD (yellow). (D) HeLa cells (0.5 mg) were lysed and immunoprecipitated with an antibody against emerin or the NICD, and the precipitates were subjected to western blotting using antibodies against the NICD or emerin. HC, heavy chain of Ig-G; WCL, whole cell lysates. (E) HeLa cells were co-transfected transiently with vectors encoding NICD deletion mutants and Flag-emerin, Cells were lysed and immunoprecipitated with antibody against HA, and the precipitates were subjected to western blotting using antibody against Flag. F, Full-length; N, N-terminus; C, C-terminus; RAM,

RBP-associated molecule domain; ANK, Ankyrin repeats; TAD, Transactivation domain; PEST,

Proline(P), glutamic acid(E), serine(S), threonine(T)-rich sequence domain. (F) The in vitro translated emerin deletion mutants were subjected to GST pull-down assays with purified GST and GST-NICD.

Bound proteins were analyzed by western blotting with anti-HA antibody. The purified proteins were subjected to western blot analysis with anti-GST antibody.

Fig. 4. Effect of endogenously expressed emerin on Notch signaling and cell survival. (A, B) HeLa cells were treated with siRNA (100 nM) against emerin (si Emerin) or control (si Control) for 48 h.

Total RNA was isolated and subjected to qRT-PCR analysis, and the expression was normalized to that of GAPDH. The results represent the mean ± S.D. of three independent experiments performed in triplicate. *, P<0.05 (C) HeLa cells were treated with si Emerin or si Control for 72 h. Cells were lysed and subjected to western blot analyses. The membranes were analyzed using ImageJ software

(NIH, Bethesda). (D) HeLa cells were transfected with si Emerin or si Control for 48 h and then treated with H2O2 (500 µM) for 12 h. Alternatively, cells were transiently transfected with vector encoding NICD cDNA for 48 h and then treated with H2O2 (500 µM) for 12 h. Cells were stained with propidium iodide (PI) and analyzed by flow cytometry. Apoptosis was measured as the percentage of cells with a sub-G0/G1 DNA content in the PI intensity-area histogram plot. The results represent the means ± S.D. of three independent experiments performed in triplicate. *, P<0.05; **, P<0.01.

Fig. 5. Effect of BAF knockdown on Notch signaling. (A, B) HeLa cells were treated with siRNA

(100 nM) against BAF (si BAF #2) or control (si Control) for 48 h. Total RNA was isolated and subjected to qRT-PCR analysis, and expression was normalized to that of GAPDH. The results represent the mean ± S.D. of three independent experiments performed in triplicate. **, P<0.01. (C)

HeLa cells were treated with two different siRNA oligonucleotides against BAF (si BAF #1 and -#2) or si Control for 72 h. Cells were lysed and subjected to western blot analyses. The membranes were analyzed using ImageJ software (NIH, Bethesda). (D) HeLa cells were treated with si BAF #2 or si

Control for 72 h. Cells were fractionated and subjected to western blot analyses (upper panel).

Fractionation was verified by using Lamin A/C antibody (nuclear fraction) or GAPDH antibody

(cytosolic fraction). The intensity of the cytosolic emerin was analyzed using ImageJ software (lower panel). (E) HeLa cells were treated with si BAF #2 or si Control for 48 h. Cells were fixed, stained, and observed under a confocal microscope. DAPI (blue) was used to visualize nuclei. Scale bar = 10

µm.

Fig. 6. Effect of Lamin A/C knockdown on Notch signaling. (A, B) HeLa cells were treated with siRNA (100 nM) against Lamin A/C (si Lamin) or control (si Control) for 48 h. Total RNA was isolated and subjected to qRT-PCR analysis, and expression was normalized to that of GAPDH. The results represent the mean ± S.D. of three independent experiments performed in triplicate. **,

P<0.01. (C) HeLa cells were treated with two different siRNA oligonucleotides against Lamin A/C or

Control for 48 h. Cells were lysed and subjected to western blot analyses. The membranes were analyzed using ImageJ software (NIH, Bethesda). (D) HeLa cells were treated with si Lamin A/C or si

Control for 48 h. Cells were fractionated and subjected to western blot analyses (upper panel).

Fractionation was verified by using Lamin A/C antibody (nuclear fraction) or tubulin antibody

(cytosolic fraction). The intensity of the cytosolic emerin band was analyzed using ImageJ software

(lower panel). (E) HeLa cells were treated with si BAF, si Lamin A/C, si BAF and si Lamin A/C, or the si control for 48 h. Total RNA was isolated and subjected to qRT-PCR analysis, and expression was normalized to that of GAPDH. The results represent the mean ± S.D. of three independent experiments performed in triplicate. **, P<0.01. (F) HeLa cells were treated with si BAF, si Lamin

A/C, si BAF and Lamin A/C, or si Control for 48 h using two different siRNA oligonucleotides. Cells were lysed and subjected to western blot analyses. The membranes were analyzed using ImageJ software (NIH, Bethesda). (G) Diagram describing the role of emerin in Notch signaling. CSL,

CBF1/suppressor hairless/Lag-1; NEXT, Notch1 extracellular truncation; ADAM, A disintegrin and metalloproteinase; CoA, Coactivators; CoR, Corepressors; PM, Plasma membrane; NE, Nuclear envelope; NGPS, Néstor-Guillermo progeria syndrome.

Supplemental Table. 1.

HeLa cells were transiently transfected with an emerin expression vector and subjected to transcriptional profiling. Data from two independent experiments were normalized to GAPDH expression and are presented as emerin/control ratios.

Supplemental Fig. 1. Effect of emerin on NICD protein stability. Dox-inducible NICD containing

HeLa cells (HeLa/iNICD) were treated with Dox (4 µg/ml) for 2 days. Cells were transiently transfected with vector encoding HA-emerin for another 2 days and then incubated with cycloheximide (CHX, 100 µM) for the indicated time periods. Cell lysates were subjected to western blotting using antibodies against the NICD or HA. The membranes were analyzed using ImageJ software (NIH, Bethesda).

Supplement Table 1. Human transcription factor PCR array Gene name Emerin/Normal Gene name Emerin/Normal Gene name Emerin/Normal

AR (Androgen receptor) 0.85 GTF2F1 0.95 NR3C1 0.24

ARNT 0.76 HAND1 0.84 PAX6 0.04

ATF1 0.86 HAND2 0.54 POU2AF1 0.03

ATF2 0.65 HDAC1 0.48 PPARA 0.18

ATF3 0.58 HIF1A 1.17 PPARG 0.13

ATF4 0.37 HNF1A 0.18 RB1 0.27

CEBP, alpha 0.81 HNF4A 0.59 REL 0.52

CEBP, beta 0.89 HOXA5 0.14 RELA 0.21

CEBP, gamma 0.29 HSF1 0.46 RELB 1.53

CREB1 0.27 ID1 1.83 SMAD1 0.68

CREBBP 0.12 IRF1 0.37 SMAD4 0.13

Catenin, beta 0.22 JUN 0.05 SMAD5 0.22

DR1 0.53 JUNB 0.09 SMAD9 0.17

E2F1 0.10 JUND 0.30 SP1 0.37

E2F6 0.01 MAX 0.36 SP3 0.10

EGR1 0.98 MEF2A 0.82 STAT1 0.02

ELK1 0.02 MEF2C 0.83 STAT2 0.32

ESR1 0.49 MYB 0.97 STAT3 0.06

ETS1 0.47 MYC 0.15 STAT4 0.48

ETS2 0.58 MYF5 0.50 STAT5A 0.16

FOS 0.86 MYOD1 0.06 STAT5B 0.04

FOXA2 0.02 NFAT5 0.12 STAT6 0.12

FOXG1 0.45 NFATC1 0.77 TBP 0.23

FOXO1 0.06 NFATC2 0.38 TCF7L2 0.29

GATA1 0.32 NFATC3 0.37 TFAP2A 0.02

GATA2 0.74 NFATC4 0.56 TGIF1 0.03

GATA3 0.82 NFKB1 0.60 TP53 0.48

GTF2B 0.76 NFYB 0.04 YY1 0.11