Macroautophagy regulates nuclear NOTCH1 activity through multiple p62 binding sites

Abstract

NOTCH1 is the prototype of the NOTCH family of single-pass transmembrane receptors and regulates many basic processes during embryonic development and human pathogenesis. In core to NOTCH1 activation are proteolytic cleavages that release its intracellular domain (NICD1), which in turn translocates to the nucleus to regulate gene transcription. Macroautophagy (hereafter autophagy) has been shown to promote the degradation of NOTCH1, but the underlying mechanisms remain elusive. Here, we show that autophagy promotes the degradation of NOTCH1 by p62-dependent binding between NICD1 and LC3, a component of the autophagosomes that execute autophagy. Strikingly, deleting any of the structural NICD1 domains fails to block the degradation of NICD1 by autophagy, and p62 binds to almost all these domains independently, indicating that p62 binds to multiple sites on NICD1 to promote its degradation. Intriguingly, inhibition of autophagy induces the accumulation of NICD1 in not only the cytoplasm but also the nucleus and increases the transcriptional activity of NICD1, and such regulation of nuclear NICD1 by autophagy is unique to NICD1 and not observed for all other NICDs (NICD2-4). Collectively, our results suggest that autophagy tightly controls nuclear NOTCH1 activity through multiple p62 binding sites, and that modulating autophagy activity may be useful for treating NOTCH1 related human diseases.

INTRODUCTION

Autophagy is a self-eating process through which the cellular contents (e.g. proteins or organelles) are delivered to autophagosomes, which then fuse with lysosomes to form autolysosomes, where the degradation of proteins or organelles occurs. Under certain conditions, autophagy functions as an essential process for recycling cellular contents and maintaining nutrient and energy levels (1–4). Mechanistically, autophagy is executed by a set of autophagy related proteins (ATGs), which are involved in different stages during the formation and maturation of autophagosomes (1). Autophagy can be highly selective for certain types of cellular contents, for instance the ubiquitinated proteins, which can be recognized and bound by the ubiquitin binding protein p62 (SQSTM1). Since p62 can simultaneously bind to LC3 (also called ATG8), a key component of the autophagosome membrane, the binding of p62 to the ubiquitinated proteins will lead them to autophagosomes for subsequent delivery to lysosomes and degradation (1–4).

Both NOTCH receptors (NOTCH1-4) and their ligands (JAGs and DLLs) are transmembrane proteins (5–7). In multicellular tissues, the cell-cell contacts enable the interactions between NOTCH1 and the NOTCH ligands, followed by a conformational change of NOTCH1 that triggers proteolytic cleavage to generate a membrane-bound shorter form of NOTCH1, called NTM1 (NOTCH1 transmembrane fragment) (5–7). NTM1 is short-lived and subsequently cleaved by γ-secretase to release NICD1, which then translocates into the nucleus to regulate gene expression by forming transcription complexes with RBPJ and co-factors (5–7). Besides such ligand-dependent mechanism, NOTCH1 can also be activated in a ligand-independent manner under certain circumstances. For instance, the endocytosed NOTCH1 can be directly cleaved on the endosomal membrane to release NICD1 (7, 8).

Once being generated, the nuclear NICD1 needs to be quickly degraded to prevent the excessive transcriptional activities, which are known to promote developmental abnormalities and pathogenesis (6, 9). Several mechanisms have been show to promote ubiquitination-dependent degradation of NICD1 (10–12). For instance, the Cyclin dependent kinase CDK8 can be recruited to the NICD1 transcription complex and phosphorylates the NICD1’s C-terminal PEST domain (rich in Proline, Glutamic acid (E), Serine, and Threonine), leading to subsequent SCF^FBXW7-mediated ubiquitination and proteasomal degradation of NICD1 (10). Besides the PEST domain, a recent report has shown that a short N-terminal motif, called N1-box, also mediates the degradation of NICD1 in Xenopus egg extracts and human cancer cells (13). Moreover, autophagy has been shown to promote the degradation of NICD1 through MEKK1-mediated phosphorylation of the PEST domain, which leads to FBW7-mediated ubiquitination (14), followed by p62 binding, thereby linking autophagy to previously identified ubiquitination-dependent mechanism driving NICD1 degradation.

In the present study, we show that the binding of NICD1 to LC3 is completely dependent on p62. Unexpectedly, deleting any of the structural domains, including the PEST domain, does not abolish the degradation of NICD1 through autophagy. Accordingly, p62 binds to almost all NICD1 domains independently, suggesting that multiple p62 binding sites on NICD1 independently but collectively promote its degradation. Furthermore, we show evidence that the autophagic degradation of NICD1 may be initiated in the nucleus and completed in the cytoplasm, and that inhibition of autophagy increases not only the abundance of NICD1 in the nucleus but also its transcriptional activity, suggesting a novel nuclear role for autophagy, which may be useful for developing NOTCH1-targeting therapeutics.

RESULTS

Two recent studies (14, 15) have shown distinct mechanisms by which NOTCH1 is degraded by autophagy: while one reports that the full length NOTCH1 but not NICD1 is degraded by autophagy (15), the other shows that NICD1 can be directly degraded by autophagy (14).

To address such discrepancy, we expressed both full length NOTCH1 and NICD1 in H1299 cells and treated the transfectants with chloroquine (CQ), an autophagy inhibitor (1). Our results show that CQ not only effectively inhibited autophagy, as evidenced by the accumulation of LC3-II (Fig.1A), an autophagosomal component that has been used as a marker for autophagy (1), but also increased the levels of both full length NOTCH1 and NICD1 (Fig.1A, B), suggesting that targeting NICD1 should serve as a common mechanism for the autophagic degradation of full length NOTCH1 and NICD1, which is the intracellular part of NOTCH1. In further support of this idea, CQ also inhibited the degradation of NTM1 (Fig.1C), the intermediate during the processing of full length NOTCH1 to generate NICD1.

Fig. 1. Autophagy promotes the degradation of NOTCH1, NTM1, and NICD1.

(A–C) Western blotting for H1229 cells transfected with full length NOTCH1 (A), FLAG tagged NICD1 (B), and FLAG tagged NTM1 (C), and treated with CQ (50 μM overnight).

(D) Western blotting for the nuclear and cytoplasmic lysates from H1299 cells transfected with FLAG tagged NICD1 and treated with CQ (50 μM overnight).

Strikingly, although autophagy is considered as a cytoplasmic process, CQ induced the accumulation of NICD1 in both the cytoplasm and the nucleus (Fig.1D). Similarly, knockdown of ATGs also increased NICD1 in both the cytoplasm and the nucleus (Fig.2A–E), promoted the binding of NICD1 to the RBPJ sites on the promoter of its previously reported target genes (16, 17), such as ERBB3 and HES1 (Fig.3A, B), and suppressed the ERBB3 promoter activity as validated by reporter assays (Fig.3C–E). Notably, knockdown of ATGs (ATG5 and ATG7) increased the expression level of p62, suggesting that autophagy is impaired in the knockdown cells (Fig.2B). The knockdown of ATGs also moderately increased LC3-II expression but did not consistently increase LC3-l expression, suggesting that the LC3-conversion may also be defective in the knockdown cells (Fig.2C). Collectively, these findings suggest that autophagy has a novel role promoting protein degradation in the nucleus; and the inhibition of autophagy not only increases the abundance of nuclear NICD1, but also promotes its transcription activity.

Fig. 2. ATGs regulate the degradation of NICD1.

(A) Western blotting for HCC827 cells transfected with non-targeting siRNA (NT) or siRNAs against ATGs.

(B, C) Western blotting for HCC827 cells transfected with NT and ATG5/7 siRNAs.

(D, E) Western blotting for nuclear and cytoplasmic lysates from HCC827 cells transfected with NT and ATG5 siRNAs (D) or ATG7 siRNAs (E).

Fig. 3. ATGs regulate the transcriptional activity of NICD1.

(A, B) ChIP assays showing that ATG16L1 siRNAs promote the binding of NICD1 to ERBB3 promoter (A) or HES1 promoter (B).

(C–E) Reporter assays showing that ATG16L1 and ATG5 siRNAs inhibit the transcription activity of ERBB3 promoter.

In contrast to its effect on NICD1, CQ did not affect the nuclear abundance of all others NICDs (NICD2-4) (Fig.4A, B); it also did not affect the cytoplasmic levels of NICD3 and NICD4, but did increase the cytoplasmic NICD2 (Fig.4B). Given that both NICD1 and NICD2 have a TAD domain (transactivation domain), which is absent in NICD3 and NICD4 (18), we examined whether the TAD domain is required for NICD1 degradation. To that end, we generated a TAD-deleted NICD1 mutant construct. As controls, we also generated additional NICD1 deletion mutants, in which each of other well-characterized NICD1 structural domains (Fig.5A, B) are deleted individually (for RAM, ANK, NCR, and PEST domains) or in combination (for ANK and NCR domains). Unexpectedly, deleting TAD or any of other NICD1 domains failed to inhibit the CQ-induced NICD1 accumulation (Fig.5C), both in the cytoplasm and in the nucleus, suggesting that the degradation of NICD1 is likely mediated by multiple sites on distinct domains, rather than depends on any single one of them. Note that the nuclear localization sequences of NICD1 are also deleted in the ANK and NCR deletion mutants, resulting in the largely absence of them in the nucleus (Fig.5C).

Fig. 4. Autophagy distinctly regulates the degradation of NICDs.

(A, B) H1299 cells were transfected with FALG tagged NICD1 (A) or NICD2-4 (B), and then treated with CQ (50 μM overnight). Western blotting was performed for the nuclear and cytoplasmic lysates from each of the transfectants.

Fig. 5. The degradation of NICD1 by autophagy is not dependent on a single domain.

(A) Schematic for the NICD1 domains.

(B) Western blotting for FLAG tagged NICD1 and its deletion mutants.

(C) H1299 cells were transfected with FALG tagged NICD1 or its deletion mutants, and then treated with CQ (50 μM overnight). Western blotting was performed for the nuclear and cytoplasmic lysates from each of the transfectants. Note: “N” indicates nuclear, and “C” indicates cytoplasmic.

Notably, a recent report has shown that p62 binds to the ubiquitinated PEST domain to promote NICD1 degradation by autophagy (14). Although our above results suggest that the PEST domain alone is dispensable for NICD1 degradation, we confirm that NICD1 did bind to p62 and LC3 (Fig.6A, B). Knockdown of p62 completely abrogated the binding between NICD1 and LC3 (Fig.6C), indicating that such binding is p62-dependent. Knockdown of p62 restored NICD1 expression (Fig.6D), demonstrating that p62 promotes NICD1 degradation. More interestingly, immunoprecipitation (IP)-Western blotting results show that p62 binds to almost all NICD1 deletion mutants (Fig.6E), indicating that p62 binds to multiple sites on NICD1 independently, thereby allowing NICD1 to be degraded in the absence of any individual domain. On the contrary, knockdown of ATG7 (Fig.6F) or p62 (Fig.6G) did not affect the endogenous NICD2 expression, suggesting that p62-mediated selective autophagy differentially regulates the degradation of NICD1 and NICD2.

Fig. 6. p62 binds to multiple sites on NICD1 to regulate its degradation.

(A) IP-Western blotting for H1299 cells transfected with FLAG-NICD1 and HA-p62.

(B) IP-Western blotting for H1299 cells transfected with FLAG-NICD1 and GFP-LC3.

(C) IP-Western blotting for H1299 cells transfected with FLAG-NICD1 and GFP-LC3, and treated with or without CQ (50 μM overnight).

(D) Western blotting for HCC827 cells transfected with NT or p62 siRNAs.

(E) H1299 cells were transfected with HA-p62 and FLAG-tagged NICD1 or its deletion mutants as indicated. The IP-Western blotting was performed for each of the transfectants.

(F, G) Western blotting for HCC827 cells transfected with NT and ATG7 siRNAs (F) or p62 siRNAs (G).

Our results show that the ATGs and LC3, which execute autophagy, exist in the nucleus (Fig.1D, Fig.2D–E, Fig.4A–B, and Fig.5C). Moreover, because LC3 binds to NICD1 to regulate its degradation, the presence of LC3 in the nucleus (Fig.1D, Fig.4A–B, and Fig.5C) suggests that a potential autophagy-related activity in the nucleus may be involved in regulating the tanslocation of NICD1 to cytoplasm for degradation. In support of this idea, immunofluorescence staining results show that LC3 and the lysosomal protein LAMP1 co-localized in both cytoplasm and nucleus (Fig.7A). Notably, LC3 did not co-localize with the early endosomal marker EEA1 (Fig.7B), demonstrating that the co-localization staining of LC3 and LAMP1 is specific. On the other hand, our transmission electron microscope results show that the CQ treatment significantly induced the accumulation of lysosomes and autolysosomes in the cytoplasm but not in the nucleus (Fig.8A–D), confirming that autophagy occurs in the cytoplasm but not the nucleus. Collectively, our results suggest a possibility that the process for selective autophagic degradation of NICD1 may be initiated in the nucleus through the binding with LC3, which co-localized with the lysosomal protein LAMP1 and subsequently led NICD1 to the cytoplasmic autolysosomes for degradation. Therefore, there is a possibility that this process promotes the translocation of NICD1 from the nucleus to the cytoplasm, and the inhibition of autophagy in the cytoplasm in turn contributes to the accumulation of NICD1 in the nucleus.

Fig. 7. LC3 and LAMP1 co-localize in both cytoplasm and nucleus.

(A) Confocal microscopy images for HCC827 cells treated with chloroquine (50 μM overnight) and stained for LC3 (green), LAMP1 (red), and DAPI (blue, staining of the nuclei).

(B) Confocal microscopy images for HCC827 cells treated with chloroquine (50 μM overnight) and stained for LC3 (green), EEA1 (red), and DAPI (blue, staining of the nuclei).

Fig. 8. Chloroquine treatment induces the accumulation of lysosomes and autolysosomes in the cytoplasm but not the nucleus.

(A) TEM image for a non-treated HCC827 cells.

(B) TEM image for a HCC827 cell treated with chloroquine (50 μM overnight).

(C and D) The nuclear (C) and cytoplasmic (D) TEM mage areas from (B) as indicated by red arrows.

DISCUSSION

Previous studies have shown that the PEST domain is critical for the proteasomal degradation of NICD1. For instance, it has been shown that CDK8 can be recruited to the nuclear NICD1 transcription complex and phosphorylates NICD1 on the PEST domain, and the hyperphosphorylation of PEST domain promotes the FBW7-mediated ubiquitination, resulting in the disassembly of the NICD1 transcription complex and eventually the degradation of NICD1 (10). A recent study has shown that the PEST domain is also important for the degradation of NICD1 by autophagy (14). It was shown that MEKK1 phosphorylates the PEST domain on T2512 to facilitate FBW7-mediated ubiquitination. Subsequently, the ubiquitinated NICD1 can be recognized by p62 and delivered to autophagosomes for degradation (14). However, our results show that complete removal of the PEST domain does not affect the degradation of NICD1 by autophagy, suggesting that the PEST domain alone may only have a limited role in this process. Instead, we found that p62 binds to multiple sites on NICD1. It is highly likely that these sites are functionally redundant yet independently in mediating the binding of p62 and the regulation of NICD1 degradation, thereby allowing p62 to tightly control the clearance of NICD1. The exact locations of these p62-binding sites are unclear. Since p62 binds to ubiquitin-like proteins, these sites may be the ones that can be ubiquitinized. Notably, besides the PEST domain, other NICD1 domains can also be ubiquitinated (19), raising a possibility that p62 may non-selectively bind to these ubiquitinated domains to regulate NICD1 degradation.

Although autophagy is considered as a cytoplasmic process, the components of autophagy machinery, especially the ATGs, have been detected in the nucleus (1, 20). The exact roles for these nuclear autophagy related proteins are not well understood, but there has been evidence that they may be critical for the initiation of the selective autophagic degradation of nuclear contents, which is subsequently completed in the cytoplasm. For instance, a recent report has shown that LC3 is present in the nucleus, directly binds to Lamin B1, and promotes Lamin B1 degradation in response to RAS activation (20). In agreement, our results show evidence that the degradation of NICD1 may be similarly initiated in the nucleus through the interaction with LC3, which in turn promotes the translocation of NICD1 to cytoplasm and leads NICD1 to the autolysosomes for degradation. Therefore, the inhibition of autophagy induces the nuclear accumulation of NICD1 and enhanced NICD1 transcriptional activity. Intriguingly, we show evidence that autophagy regulates the nuclear abundance of NICD1, but not other NICDs, providing a novel way to specifically manipulating NICD1 transcriptional activities through the modulation of autophagy. Such an approach may be especially useful for developing strategies that treat NOTCH1-related diseases without affecting the transcription functions of other NOTCHs.

MATERIALS AND METHODS

Cell culture and transfection

Human NSCLC cancer cell lines HCC827 and H1299 were from ATCC (mycoplasma-free and authenticated) and cultured in RPMI1640 supplemented with 10% FBS (GIBCO). Lipofectamine 2000 and RNAiMAX (both from Invitrogen) were used for DNA or siRNA transfection, respectively, as described in the manufacturer’s datasets.

Antibodies and reagents

Rabbit monoclonal antibodies for Tubulin (2125S), Lamin A/C (4777S), p62 (8025S), LC3 (12741S), NICD1 (4147S), HA (3724S), ATG3 (3415S), ATG5 (12994S), ATG7 (8558S), ATG16L1 (8089S), Notch1 (3608S), LC3A/B (Alexa Fluor 488 Conjugate) (13082S), LAMP1 (9091S) and EEA1 (3288S) were purchased from Cell Signaling. Mouse monoclonal antibody for FLAG (F1804) was purchased from Sigma. Goat polyclonal antibody for GFP (sc9996) was purchased from Santa Cruz. SiRNAs against ATG3, ATG5, ATG7, ATG16L1, and p62 were purchased from Origene (3 distinct siRNA sequences for each gene). All other chemicals and reagents were purchased from Sigma unless otherwise indicated.

Mutagenesis and plasmid constructs

Mutagenesis experiments were performed using the Q5 Site-Directed Mutagenesis Kit (NEB) as described in the manufacturer’s dataset. The mouse NTM1 and deletion mutants for NICD1 were cloned into pCMV10 using the following primers:

• NTM1: GCTCTAGAGAGGCCGTGCAGAGTGAGAC (forward) & CGGGATCCTTACTTGAAGGCCTCCGGAATGC (reverse);

• ΔTAD: AGCTCGGCAGCCAATGGG (forward) & GAGGGAGTCCACAGGCGAC (reverse);

• ΔRAM: ACACCGCCTCAGGGGGAG (forward) & GCGCTTGCGGGACAGCAG (reverse);

• ΔPEST: GGGGCCCAGCGGTTGTAC (forward) & TAAGGATCCCGGGTGGCATC (reverse);

• ΔANK: CTGGTGCGCAGCCCACAG (forward) & CTCCAGGTCTTCGTCTCCCC (reverse); ΔNCR: TCACCCCATGGCTACTTG (forward) & GTTGTACTCATCCAAAAGCC (reverse);

• ΔANK-NCR: CTGGTGCGCAGCCCACAG (forward) & GTTGTACTCATCCAAAAGCC (reverse).

Preparation of nuclear and cytoplasmic lysates

The nuclear and cytoplasmic lysates were prepared using the nuclear extract kit from Active Motif (#40010) as described in the manufacturer’s mannual. Briefly, the sub-confluent cells were washed in ice-cold PBS plus phosphatase inhibitors, scraped from the culture dish, and centrifuged. The cell pellets were then lysed in hypotonic buffer (provided by the kit) and centrifuged to collect the supernatants, which are the cytoplasmic lysates. The pellets were further lysed in nuclear lysis buffer (provided by the kit) and centrifuged. The supernatants are the nuclear lysates.

Western blotting

Whole cell protein extraction was performed by lysing cell monolayers in a RIPA lysis buffer containing 1 mM PMSF, 1 mM Na₃VO₄ and protease inhibitor cocktail (Santa Cruz Biotechnology). The nuclear and cytoplasmic protein was extracted using Nuclear Extract Kit (Active Motif). The lysates were quantified for protein concentrations with BCA protein assay reagents (Pierce Biotechnology). Protein lysates (10–20μg) were mixed with 6X sample loading buffer and boiled at 95°C for 10 minutes. For Western blotting, proteins were electrophoresed in SDS-polyacrylamide gel and transferred onto PVDF membranes using Trans-Blot Turbo Transfer System (BIO-RAD). The membranes were blocked then incubated overnight at 4°C with the primary antibody diluted in blocking buffer. After being washed with TBST, the membrane was further incubated with HRP-conjugated secondary antibodies for 1 hour, followed by washing with TBST buffer. Visualization was done using ECL substrates (Pierce Biotechnology).

Immunoprecipitation

48 hours after transfection, the cells were lysed in lysis buffer (Cell signaling technology) at 4°C. Protein lysate (150ug) were subjected to immunoprecipitation incubating with specific antibodies (1ug) and protein G magnetic beads (Cell signaling technology) at 4°C on the rotator. After overnight incubation, the beads were subsequently washed three times in ice-cold PBS, and proteins that remained bound to the beads were eluted by boiling in 2× protein sample buffer. The samples were separated by SDS-PAGE, and visualized by immunoblotting.

ChIP assay

ChIP assay was performed using the ChIP-IT Express Enzymatic Kit (Active Motif #53009) as described in the manufacturer’s manual. Briefly, adherent cells were cross-linked in 1% formaldehyde. After enzymatic digestion, lysates were immunoprecipitated using desired antibodies or control normal IgG. DNA was then eluted, and qPCR was performed to amplify the promoter elements.

Promoter reporter assay

Briefly, the ERBB3 promoter reporters and control vector plasmids were transfected into H1299 cells using Lipofectamine 2000 (Invitrogen). 48 hours after transfection, cells were lysed and luciferase activity was measured using a Dual-Luciferase Reporter Assay System (Promega E1910).

Immunofluorescence staining

For cell immunofluorescence staining, HCC827 cells were seeded in sterilized glass coverslips, which placed in six-well plate. After remove the medium, the cells were fixed in 4% paraformaldehyde for 10 min at room temperature, washed with cold PBS twice, blocked with 5% normal goat serum with 0.3% Triton X-100 in PBS for 60 min at room temperature and then incubated with appropriate primary antibodies overnight at 4°C (dilution 1:200). Following PBS washing, the cells were incubated with Alexa Fluor 594-conjugated secondary antibody (Thermo Fisher Scientific, dilution 1:500). Finally, all cells were mounted and counterstained with ProLong Gold antifade reagent with DAPI (Thermo Fisher Scientific). The stainings were observed under a Zeiss LSM-780 confocal microscope as indicated.

Transmission Electron Microscopy

HCC827 cells cultured on 6-well plates were treated with or without 50 μM chloroquine overnight and sent to the Mayo Clinic electron microscope core facility for tansmission electron microscopy (TEM) sample processing. The TEM micrographs were acquired with a JEOL 1400 electron microscope (JEOL Ltd, Tokyo, Japan).

Statistics

statistical significance was determined using student’s t-test as indicated in the figures, and a difference with p<0.05 is considered as statistically significant in all cases unless otherwise indicated.

Abbreviations

NOTCH1

Notch homolog 1, translocation-associated (Drosophila)

NICD

NOTCH intracellular domain

LC3

microtubule-associated protein 1A/1B-light chain 3

ATG

autophagy related gene

SQSTM1

Sequestosome-1

NTM1

NOTCH1 transmembrane domain

CDK8

cyclin-dependent kinase 8

PEST

proline (P), glutamic acid (E), serine (S), and threonine (T) rich domain

SCF

Skp, Cullin, F-box containing complex

CQ

chloroquine

RBPJ

recombination signal sequence-binding protein Jκ

ERBB3

erythroblastic oncogene B 3

HES1

Hairy and Enhancer of Split 1

RAM

Rbp-associated molecule domain

ANK

Ankyrin repeats domain

NCR

cysteine response region

TAD

transactivation domain

IP

immunoprecipitation

LAMP1

lysosomal-associated membrane protein 1

EEA1

early endosome antigen 1

FBW7

F-box/WD repeat-containing protein 7

MEKK1

mitogen-activated protein kinase kinase kinase 1

RAS

rat sarcoma viral oncogene

ChIP

chromatin immunoprecipitation.