Presenilin-2 regulates the degradation of RBP-Jk protein through p38 mitogen-activated protein kinase

Notch is a vitally important signaling receptor, which modulates cell fate determination and pattern formation in a number of ways during the development of both invertebrate and vertebrate species (Artavanis-Tsakonas and Simpson, 1991; Duffy and Perrimon, 1996; Weinmaster, 1998; Artavanis-Tsakonas et al., 1999; Robey, 1999; Neumann, 2001; Wang and Sternberg, 2001; Whitford et al., 2002; Lai, 2004). Mammals express four Notch genes (NOTCH1–NOTCH4) and five ligands for Notch from two conserved families, Jagged (JAG1 and JAG2) and Delta (DLL1, DLL3 and DLL4) in different combinations in most cell types. After ligand binding, Notch proteins are activated by a series of cleavages that release its intracellular domain (Notch-IC), followed by its nuclear translocation (Kopan and Ilagan, 2009). The nuclear translocation of Notch-IC results in the transcriptional activation of genes of the HES family [Hes/E(spl) family] and the HEY family (Hesr/Hey family) through the interaction of Notch-IC with RBP-Jk [CBF1 or RBP-Jk in vertebrates, Suppressor of Hairless (SuH) in Drosophila melanogaster, Lag-1 in Caenorhabditis elegans] and through mastermind-like (MAML) (Weinmaster, 1997; Kopan and Ilagan, 2009). In the absence of Notch-IC, RBP-Jk functions as a transcriptional repressor because of its ability to bind to transcriptional co-repressors and histone deacetylase-1. Notch-IC translocates to the nucleus, where it interacts with the RBP-Jk and displaces co-repressor complexes and recruits co-activators, thus turning RBP-Jk into a transcriptional activator (Kao et al., 1998).

Presenilin proteins are integral components of a multiprotein protease complex, referred to as γ-secretase, which is responsible for the intramembranous cleavage of the amyloid precursor protein (APP), the Notch receptors, Delta, Jagged, E-cadherin, p75 neurotrophin receptor, low density lipoprotein-related protein, nectin-1α, CD44 and Erb-b4, resulting in the release of intracellular signaling fragments (Koo and Kopan, 2004). These intracellular fragments of presenilin substrates are translocated into the nucleus, yet their functional roles in the cellular system are still not apparent. Presenilins are stably associated with the γ-secretase complex along with three other proteins: nicastrin, Pen-2 and Aph-1, which collectively constitute the functional γ-secretase (Edbauer et al., 2003). Presenilin-1 (PS1)-knockout mice have an embryonic lethal phenotype with similarities to that of the Notch-knockout mouse, as well as dramatically reduced amyloid-beta (Aβ) peptide levels (Shen et al., 1997; Palacino et al., 2000). By way of contrast, PS2-knockout mice have no apparent phenotype or change in levels of Aβ peptide (Herreman et al., 1999). These observations indicate that, under normal circumstances, γ-secretase activity resulting in the cleavage of substrates is associated predominantly with PS1, rather than PS2 complexes.

Recent studies have confirmed the role of ubiquitylation in the regulation of Notch signaling (Lai, 2002). Several research groups have demonstrated that the F-box containing protein Sel10/Fbw7 mediates Notch ubiquitylation in the nucleus, and targets it for proteasome-dependent degradation (Wu et al., 2001; Gupta-Rossi et al., 2001; Hubbard et al., 1997; O’Neil et al., 2007). We demonstrated that ILK, SGK1 and Jagged-1 intracellular domain (JICD) downregulate the protein stability of Notch1-IC through the ubiquitin-proteasome pathway by means of Fbw7 (Mo et al., 2007; Mo et al., 2011; Kim et al., 2011). Thus far, little has been definitively determined regarding the regulation of RBP-Jk protein stability. In our recent report, we proposed that degradation mediated by the amyloid precursor protein intracellular domain (AICD), might be involved in the preferential degradation of RBP-Jk by the lysosomal pathway. Despite this observation, the precise mechanisms underlying the degradation of the RBP-Jk protein remain to be accurately delineated.

Therefore, we evaluated the degrdation of RBP-Jk using proteasome and lysosome inhibitors. In this study, we determined that RBP-Jk degradation involves both the proteasome and the lysosome. We also demonstrated that PS2 modulates the protein stability of RBP-Jk by the p38 mitogen-activated protein kinase ∂ (MAPK13; referred to here as p38 MAPK) signaling pathway. Furthermore, we showed that the phosphorylation of RBP-Jk by p38 MAPK induces RBP-Jk protein degradation and ubiquitylation. Collectively, our findings indicate that p38 MAPK functions as a regulator of RBP-Jk protein stability by RBP-Jk phosphorylation.

Despite the many studies of RBP-Jk thus far conducted, the mechanism underlying the degradation of the protein remains to be completely understood. To determine whether the RBP-Jk protein is degraded by the proteasome, proteasome inhibitors were used to treat Myc–RBP-Jk-transfected HEK293 cells. These cells were then treated with ALLN (25 μM), MG132 (5 μM) and lactacystin (10 μM) for 4 hours (Fig. 1A). Myc–RBP-Jk protein was detected with the anti-Myc antibody 9E10. As shown in Fig. 1A, similarly to lactacystin treatment, ALLN and MG132 significantly increased RBP-Jk levels relative to those observed in control cells treated with vehicle solution only. Protein levels were normalized to β-actin. Myc–RBP-Jk-transfected HEK293 cells were then treated with different doses of MG132 (0–10 μM) for 4 hours or at 5 μM for different periods of time (0–6 hours). MG132 treatment markedly increased RBP-Jk protein levels in a dose-dependent (Fig. 1B) and time-dependent manner (Fig. 1C). In order to determine whether treatment with proteasome inhibitors also induces an increase in endogenous RBP-Jk proteins, HEK293 and C2C12 cells were treated with the proteasome inhibitor MG132 at different dosages for 4 hours (0–10 μM) (Fig. 1D,E). Endogenous RBP-Jk protein was detected with rabbit anti-RBP-Jk antibody. MG132 markedly increased endogenous RBP-Jk protein levels in HEK293 and C2C12 cells.

The half-life of RBP-Jk was determined using cycloheximide, a protein translation inhibitor. Cycloheximide interacts with the translocase enzyme and blocks protein synthesis in eukaryotic cells. After cycloheximide treatment, the level of RBP-Jk protein dropped gradually, and approximately half of the protein was degraded after 2 hours compared with levels in control cells (Fig. 1F,G). Upon treatment with MG132, the level of RBP-Jk protein dropped slightly and approximately half of the protein was degraded after 8 hours in the presence of MG132 (Fig. 1F,G). Taken together, these results clearly demonstrate that treatment with proteasome inhibitor significantly increases RBP-Jk protein levels, thereby suggesting that the degradation of RBP-Jk protein is indeed mediated by the proteasome pathway.

In an effort to determine whether RBP-Jk protein is lysosomally degraded, lysosome inhibitors were administered to Myc–RBP-Jk-transfected HEK293 cells. The cells were treated with chloroquine (100 μM) and NH₄Cl (50 mM) for 6 hours. Myc–RBP-Jk protein was then detected with the anti-Myc antibody 9E10. Chloroquine and NH₄Cl significantly increased RBP-Jk levels in comparison with levels in the vehicle control cells (Fig. 2A). Protein levels were normalized to β-actin. The cells were subsequently treated with different doses of NH₄Cl for 10 hours (0–50 mM) or at 50 mM for different periods of time (0–10 hours). NH₄Cl treatment affected a marked increase in RBP-Jk protein levels in a dose-dependent (Fig. 2B) and time-dependent manner (Fig. 2C). To determine whether treatment with lysosome inhibitors also induces an increase in endogenous RBP-Jk proteins, HEK293 and C2C12 cells were treated with the lysosome inhibitor NH₄Cl at different doses for 10 hours (Fig. 2D,E). NH₄Cl markedly increased endogenous RBP-Jk protein levels in HEK293 and C2C12 cells.

To determine the half-life of RBP-Jk in the presence of lysosome inhibitor, Myc–RBP-Jk-transfected HEK293 cells were treated with NH₄Cl and cycloheximide. After cycloheximide treatment, the level of RBP-Jk protein declined gradually, with approximately half of the protein degraded after 2.5 hours relative to the level in control cells (Fig. 2F,G). Upon NH₄Cl treatment, the level of RBP-Jk protein changed moderately; approximately half of the protein was degraded after 6 hours in the presence of NH₄Cl (Fig. 2F,G). Taken together, these results clearly demonstrate that treatment with lysosome inhibitor moderately increases RBP-Jk protein levels, thereby suggesting that degradation of RBP-Jk is also mediated by the lysosome pathway.

To determine the effects of the absence of PS1 and PS2 on RBP-Jk expression, immunoblot analyses were conducted using wild-type PS1 and PS2 cells and double-knockout cells. The half-life of endogenous RBP-Jk was determined using cycloheximide, a protein translation inhibitor. PS1 and PS2 double-knockout and wild-type control cells were exposed to cycloheximide, and the amount of RBP-Jk was analyzed using immunoblot. As a result, the steady-state level and the half-life of endogenous RBP-Jk were higher in the double-knockout cells than in wild-type cells (Fig. 3A,B). We subsequently assessed the involvement of PS1 or PS2 in RBP-Jk protein degradation using PS1-knockout and PS2-knockout cells. The steady-state level and the half-life of endogenous RBP-Jk were higher in PS2-knockout cells than in PS1-knockout cells (Fig. 3C,D). We next investigated the highly specific γ-secretase inhibitor DAPT (Dovey et al., 2001) for its capacity to block RBP-Jk protein degradation and half-life. With this objective, wild-type PS1 and PS2 cells were treated with or without DAPT along with cycloheximide (Fig. 3E). As shown in Fig. 3E, endogenous RBP-Jk levels were increased markedly in cells treated with DAPT alone and in those with DAPT and cycloheximide. We also attempted to confirm the effect of PS2 on the protein level of RBP-Jk using PS1 and PS2 double-knockout cells. We measured protein levels of RBP-Jk in the presence of wild-type PS2 or an inactive dominant-negative mutant of PS2 [PS2(D366A)] (Fig. 3F). As shown in Fig. 3F, endogenous levels of RBP-Jk protein were markedly decreased in cells transfected with wild-type PS2. However, levels were substantially increased in cells expressing the PS2(D366A) mutant. Furthermore, we attempted to confirm the effect of PS1 and PS2 on the half-life of endogenous RBP-Jk using PS1 and PS2 double-knockout cells. We measured levels of RBP-Jk in the presence of the PS1 or PS2 using cycloheximide, a protein translation inhibitor. The steady-state level and the half-life of endogenous RBP-Jk were higher in PS1-transfected cells than in PS2-transfected cells (Fig. 3G,H). These results indicate that PS2 has a critical role in the regulation of RBP-Jk protein degradation and half-life and show that the stability of the endogenous RBP-Jk protein is downregulated by PS2.

PS1-knockout cells showed degradation of RBP-Jk, followed by periods during which we inhibited further protein synthesis by applying cycloheximide (CHX). In the presence of CHX alone, RBP-Jk degraded over a 4 hour observation period (Fig. 4A). By contrast, when CHX and MG132 were applied together, RBP-Jk levels were unchanged over 6 hours (Fig. 4B,D). Degradation of RBP-Jk was inhibited significantly by proteasome inhibition. Collectively, these results clearly show that treatment with proteasome inhibitor significantly increases RBP-Jk protein levels, thus suggesting that degradation of RBP-Jk is mediated by the proteasome pathway. We then attempted to determine whether lysosomal inhibitors exerted any detectable effects on the degradation of RBP-Jk. Our studies showed that lysosomal inhibitors moderately regulate the steady-state levels of RBP-Jk proteins in PS1-knockout cells (Fig. 4C,D). These results demonstrated that the stability of the RBP-Jk protein is downregulated by PS2 through proteasome- and lysosome-dependent pathways.

A previous report has demonstrated that PS2 functions as a signaling molecule upstream of the p38 MAPK pathway (Sun et al., 2001). To determine whether treatment with p38 MAPK inhibitors increases the stability of endogenous RBP-Jk proteins, PS1-knockout cells were treated with the p38 MAPK inhibitor SB203580 (Fig. 4E,G). SB203580 markedly increased endogenous RBP-Jk protein stability in PS1-knockout cells. Additionally, to determine whether treatment with GSK-3β inhibitors increases the protein stability of endogenous RBP-Jk proteins, PS1-knockout cells were treated with the GSK-3β inhibitor LiCl. LiCl did not increase endogenous RBP-Jk protein stability in PS1-knockout cells. The results show that the downregulated level of RBP-Jk protein was not restored to a sufficient degree by the inhibition of GSK-3β (Fig. 4F,G). These results demonstrate that the negative regulation of the stability of RBP-Jk by PS2 occurs in a GSK-3β-independent pathway. Our results show that the downregulated RBP-Jk protein is restored by treatment with a p38-MAPK-specific inhibitor in PS1-knockout cells, thereby indicating that PS2-mediated degradation of RBP-Jk occurs in a manner that is dependent on p38 MAPK signaling.

We investigated whether the presenilins affect p38 MAPK activity in MEF cells by examining the endogenous activity of p38 MAPK in wild-type and presenilin-knockout cells. Endogenous p38 MAPK activity at the basal state was higher in wild-type PS1 and PS2 cells and PS1-knockout MEF cells than it was in PS2-knockout MEF cells in normal serum conditions (Fig. 5A). The basal activity of p38 MAPK activity in wild-type and PS1-knockout MEF cells was substantially suppressed in the presence of the p38 MAPK inhibitor SB203580 (Fig. 5A). This suggests that PS2 might influence the basal activity of endogenous p38 MAPK. We transfected the HEK293 cells with Myc–RBP-Jk and HA–p38-MAPK, and the quantity of remaining RBP-Jk was evaluated after various periods of cycloheximide treatment. We determined the protein stability of RBP-Jk in HEK293 cells by cycloheximide treatment with or without p38 MAPK. After cycloheximide treatment, the level of RBP-Jk protein declined gradually, with approximately half of the protein being degraded after 4 hours without p38 MAPK (Fig. 5B,C). Upon cycloheximide treatment, the level of RBP-Jk protein declined rapidly, with approximately half of the protein being degraded after 1.5 hours in the presence of p38 MAPK (Fig. 5B,C). This result demonstrates that RBP-Jk is rapidly turned over in the presence of p38 MAPK. To determine whether the degradation of RBP-Jk proteins is mediated by the proteasome pathway, the proteasome inhibitor MG132 was administered to cells expressing RBP-Jk and p38 MAPK. The cells were treated with proteasome inhibitor for 6 hours and the RBP-Jk protein was detected in an immunoblot assay. The results of our studies showed that the proteasome inhibitor significantly increased RBP-Jk levels (Fig. 5D). The level of RBP-Jk protein was reduced in the presence of p38 MAPK, but was significantly restored by MG132 treatment (Fig. 5D).

We then attempted to determine whether lysosomal inhibitors exerted any detectable effects on RBP-Jk degradation. We transfected HEK293 cells with Myc–RBP-Jk and HA–p38-MAPK, and the quantity of RBP-Jk remaining was analyzed after various periods of treatment with NH₄Cl. Our findings revealed that lysosomal inhibitor also increased the steady-state levels of RBP-Jk proteins (Fig. 5E). These results showed that the stability of the RBP-Jk protein was downregulated by p38 MAPK by a proteasome- and lysosome-dependent pathway.

Because our results suggested that RBP-Jk is a target of p38 MAPK, we subsequently attempted to determine whether these two proteins interact physically in intact cells. In in vitro binding studies, purified GST and GST–RBP-Jk proteins were immobilized onto GSH–agarose. HA–p38-MAPK-expressing cell lysates were incubated either with GST or GST–RBP-Jk immobilized onto GSH–agarose. The interaction between GST–RBP-Jk and p38 MAPK was detected on bead complexes (Fig. 6A). We then evaluated the physical interactions occurring between RBP-Jk and p38 MAPK. HEK293 cells were co-transfected with vectors encoding Myc–RBP-Jk and HA–p38-MAPK, and were then subjected to co-immunoprecipitation analysis (Fig. 6B). Immunoblot analysis using the anti-Myc antibody of the anti-HA immunoprecipitates from the transfected cells revealed that p38 MAPK did indeed physically associate with RBP-Jk in the cells.

We next conducted an in vitro kinase assay with HA–p38-MAPK and purified GST–RBP-Jk. HA–p38-MAPK immunocomplexes prepared from HEK293 cells catalyzed the phosphorylation of purified recombinant GST–RBP-Jk (Fig. 6C). p38 MAPK preferentially phosphorylates substrate protein serine and threonine residues that lie in Pro-x-(Ser/Thr)-Pro or S/T-P motifs (Alvarez et al., 1991; Court et al., 2004). The results of in silico studies have shown that RBP-Jk harbors a possible conserved threonine residue in vertebrates; this threonine residue is accessible and is located immediately after the DNA binding domain. Furthermore, using site-directed mutagenesis, we determined that the replacement of Thr339 of RBP-Jk with alanine effected a reduction in the in vitro phosphorylation of the recombinant protein by HA–p38-MAPK immunoprecipitates (Fig. 6C). p38 MAPK phosphorylated GST–RBP-Jk(WT) but did not phosphorylate GST-RBP-Jk(T339A). Additionally, HEK293 cells were transfected with expression vectors encoding Myc–RBP-Jk(WT), Myc–RBP-Jk(T339A) and HA–p38-MAPK. After 48 hours of transfection, the cell lysates were subjected to immunoblotting with an antibodies against phosphorylated Ser and Thr, or the HA tag (Fig. 6D). p38 MAPK phosphorylated RBP-Jk(WT) but did not phosphorylate RBP-Jk(T339A).

We then attempted to characterize the involvement of phosphorylation in the physical association between p38 MAPK and RBP-Jk by co-immunoprecipitation. HEK293 cells were cotransfected with vectors coding for HA–p38-MAPK, Myc–RBP-Jk and Myc–RBP-Jk(T339A) and were then subjected to co-immunoprecipitation analysis. RBP-Jk and p38 MAPK were co-immunoprecipitated, but when co-transfected with RBP-Jk(T339A), the band of RBP-Jk that interacted with p38 MAPK disappeared (Fig. 6E). These results demonstrate that the phosphorylation of RBP-Jk by p38 MAPK plays a pivotal role in its ability to bind with p38 MAPK. Moreover, we ascertained that RBP-Jk(T339A) is resistant to p38-MAPK-induced degradation, which implies that the p38-MAPK-induced phosphorylation of RBP-Jk is crucially important for the degradation of the RBP-Jk protein (Fig. 6F). We then attempted to characterize the involvement of phosphorylation in the polyubiquitylation of RBP-Jk by p38 MAPK. HEK293 cells were co-transfected with vectors coding for wild-type Myc–RBP-Jk, Myc–RBP-Jk(T339A) mutant and HA–ubiquitin, and were then subjected to ubiquitylation analysis. The results of immunoblot analysis of the anti-HA immunoprecipitates from the transfected cells with an anti-Myc antibody demonstrated that p38 MAPK facilitated the ubiquitylation of RBP-Jk and that the ubiquitylation of RBP-Jk(T339A) prevented this (Fig. 6G). We attempted to characterize the involvement of T339 phosphorylation in the degradation of RBP-Jk. PS1-knockout cells were transfected with vectors coding for Myc–RBP-Jk and Myc–RBP-Jk(T339A) and were then subjected to immunoblotting with antibodies against phosphorylated Ser or Thr. RBP-Jk protein levels and phosphorylation of RBP-Jk were increased by the addition of proteasome inhibitor in a dose-dependent manner (Fig. 6H). Furthermore, our findings revealed that lysosomal inhibitor also increased the steady-state levels of RBP-Jk proteins and phosphorylated RBP-Jk proteins (Fig. 6I). However, we did not see phosphorylation of the RBP-Jk(T339A) mutant. These results demonstrate that the phosphorylation of RBP-Jk by p38 MAPK is crucial to its ability to degrade RBP-Jk. RBP-Jk is a direct substrate of p38 MAPK, and that the phosphorylation of RBP-Jk by p38 MAPK is one of the key factors in the regulation of RBP-Jk protein stability.

RBP-Jk functions as a transcription repressor when it is not associated with Notch1-IC. When associated with the Notch1-IC protein, RBP-Jk functions as a transcription activator complex that activates the transcription of Notch1 target genes, including Enhancer of split [E(spl)] complex genes and the mammalian homologues of the Hairy and E(spl) genes, HES1 and HES5 (Jennings et al., 1994; Abu-Issa and Cavicchi, 1996; de Celis et al., 1996; Ligoxygakis et al., 1998; Ohtsuka et al., 1999; Jouve et al., 2000). The results of one of our recent studies suggested that the proteasomal and lysosomal degradation of RBP-Jk proteins could occur in intact cells and plays a crucial role in Notch1 signaling (Kim et al., 2011). However, the precise mechanism by which RBP-Jk protein degradation occurs remains to be clearly defined. In this study, we demonstrate that RBP-Jk degradation involves both the proteasomal and lysosomal pathways. PS2 modulates the protein stability of RBP-Jk through the p38 MAPK signaling pathway and the phosphorylation of RBP-Jk by p38 MAPK induces RBP-Jk protein degradation and ubiquitylation.

Protein degradation is generally mediated through two major systems: the proteasome and the lysosome. These different proteolysis pathways can be differentiated by their sensitivity to proteasome- or lysosome-specific inhibitors. Proteasomal proteolysis can be inhibited by ALLN, MG132 and lactacystin. MG132 can reversibly block all activity of the 26S proteasome (Rock et al., 1994). ALLN inhibits neutral cysteine proteases and the proteasome (Sasaki et al., 1990; Debiasi et al., 1999). Lactacystin is a microbial metabolite isolated from Streptomyces, which is now used broadly as a selective inhibitor of the 20S proteasome (Lee and Goldberg, 1998). We determined that lactacystin, as well as other peptide-based reversible proteasome inhibitors, including MG132 and ALLN, markedly reduced exogenous RBP-Jk degradation in RBP-Jk ectopically expressed cells, and endogenous RBP-Jk protein in HEK293 and C2C12 cells. Our data demonstrate that RBP-Jk is ubiquitylated, and also clearly show that blockage of the ubiquitin–proteasome pathway inhibits RBP-Jk degradation. Degradation of proteins by the lysosome can be inhibited by NH₄Cl and chloroquine. NH₄Cl is a very effective inhibitor of lysosomal function and inhibits the function of lysosomal proteases by an induced increase in intralysosomal pH (Ohkuma and Poole, 1978; Dean et al., 1984). Chloroquine is another lysosomotrophic agent, and is commonly used to inhibit lysosome function (Welman and Peters, 1977). Chloroquine is a weak base, which can disrupt lysosomal function by blocking acidification. Our data also show that RBP-Jk degradation is mediated by the lysosome. Lysosome inhibition significantly slows down turnover of RBP-Jk. Our data strongly indicate that these two proteolytic pathways are involved in RBP-Jk degradation.

Recent studies have emphasized the role of ubiquitylation in the regulation of Notch signaling (Lai, 2002). Five known E3 ligases regulate Notch itself, Notch ligands, or known Notch antagonists. LNX is a positive regulator of Notch signaling that is responsible for the degradation of Numb, a membrane-associated protein that inhibits the function of the Notch receptor (Nie et al., 2002). Neuralized (neur) and Mind bomb (mib) promote the monoubiquitylation and endocytosis of Delta (Deblandre et al., 2001; Lai et al., 2001; Pavlopoulos et al., 2001; Itoh et al., 2003). Itch associates with the intracellular domain of Notch through its WW domains and promotes the ubiquitylation of Notch1-IC through its HECT domain (Qiu et al., 2000). Finally, Notch1-IC is also degraded in the nucleus by the ubiquitin–proteasome system with Fbw7, an E3 ligase for the ubiquitylation of Notch1-IC (Oberg et al., 2001; Wu et al., 2001; Lai, 2002; Minella and Clurman, 2005; Mo et al., 2007). Ubiquitylation and degradation might actually act on all Notch1 signaling components, rather than separately on its individual components.

The transcriptional activation of downstream target genes by Notch1-IC depends on the association of Notch1-IC with RBP-Jk within the nucleus. To determine whether Notch1-IC-deficient conditions also modulate RBP-Jk protein levels through proteasomal and lysosomal pathways, we constructed PS1 and PS2 double- and single-knockout cells. Our results show that disruption of the PS2 gene significantly reduced RBP-Jk protein turnover. However, knockout of the PS1 gene does not affect RBP-Jk protein turnover, thereby indicating that RBP-Jk protein degradation is both PS2 dependent and PS1 independent. The results of a previous report demonstrated that PS2 functions as a signaling molecule upstream of the p38 MAPK pathway (Sun et al., 2001). To determine whether p38 MAPK modulates RBP-Jk protein levels, we treated PS1-knockout cells with a specific p38 MAPK inhibitor. Collectively, our findings show that the stability of the RBP-Jk protein was downregulated by p38 MAPK by proteasome- and lysosome-dependent pathways. p38 MAPK preferentially phosphorylates serine and threonine residues that lie within P-x-S/T-P or S/T-P motifs (Kobayashi et al., 1999). RBP-Jk harbors one conserved consensus site for phosphorylation by p38 MAPK and is phosphorylated by p38 MAPK both in vitro and in vivo. The transcriptional activation of Notch1 signaling plays a crucial role in the regulation of diverse cell fate determinations. RBP-Jk protein performs a central role in Notch1 signaling for the induction of Notch1 target genes. Therefore, the protein turnover of RBP-Jk is critically relevant to the regulation of Notch1 signaling. Taking this into consideration, the mechanism underlying the downregulation of RBP-Jk proteins by p38 MAPK holds promise in controlling developmental programs or reducing the functions of Notch1 proteins (Fig. 7). To achieve this, the precise mechanisms underlying the functions of p38 MAPK and other kinase(s) in the regulation of RBP-Jk protein turnover should be investigated intensively in the future.

In summary, our results show that RBP-Jk degradation involves both the proteasome and the lysosome pathways. We also demonstrate that PS2 modulates the protein stability of RBP-Jk through the p38 MAPK signaling pathway. Furthermore, we show that phosphorylation of RBP-Jk by p38 MAPK induces RBP-Jk protein degradation and ubiquitylation. Henceforth, the findings of this study might begin to shed some light onto what might be a signal crosstalk mechanism of RBP-Jk and p38 MAPK, or might point to the existence of phosphorylation-dependent regulation of RBP-Jk.

HEK293 (Human embryonic kidney 293) cells, C2C12 (Mouse myoblast), PS1/2^+/+, PS1/2^–/–, PS1^–/– and PS2^–/– MEF cells were maintained at 37°C in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin, in a humidified incubator with an atmosphere containing 5% CO₂. For transient transfection, cells were grown to ∼80% confluence. The cultured cells were transiently transfected by the calcium phosphate method or with Lipofectamine (Invitrogen) (Kim et al., 2007).

For immunoprecipitation and immunoblotting, antibodies against HA (12CA5 1:1000), Myc (9E10, 1:1000) and FLAG (Sigma, 1:1000) were used. RBP-Jk was detected with the RBP-Jk (H-50) rabbit polyclonal antibody (Santa Cruz, 1:1000), phosphorylated RBP-Jk with antibodies against phosphorylated Ser and Thr (Upstate, 1:1000). For detection of β-actin (C-4), the β-actin mouse monoclonal antibody was used (Santa Cruz, 1:5000). Phosphorylated p38 was detected with the antibody against phosphorylated Ser and Thr (Upstate, 1:1000).

Cells were treated with the proteasomal inhibitors ALLN (25 μM), MG132 (5 μM) or Lactacystin (10 μM) or the lysosomal inhibitors Chloroquine (100 μM) or NH₄Cl (50 mM). MG132 was used at 0, 2.5, 5 and 10 μM for 4 hours for the dosage assay, and at 10 μM for 0, 4 and 6 hours for the time-course assay of the proteasomal inhibitor. NH₄Cl was used at 0, 10, 20 and 50 mM for 10 hours for the dosage, and at 50 mM for the 0, 6, 8 and 10 hours for the time-course assay of the lysosomal inhibitor. Cycloheximide (CHX) was used at 200 μM at 0, 2, 4, 6 and 8 hours for the time-course assay of the translational inhibitor (Mo et al., 2011).

The recombinant GST–RBP-Jk protein was expressed in E. coli BL21 strain using the pGEX system. The GST fusion protein was then purified with GSH–agarose beads (Sigma) in accordance with the manufacturer’s instructions. An equal amount of GST or GST–RBP-Jk fusion protein was incubated with the lysates of the HEK293 cells, which had been transfected for 4 hours with combinations of expression vectors at 4°C, with rotation. After incubation, the beads were washed three times in ice-cold phosphate-buffered saline (PBS, pH 7.4) and boiled with 20 μl of Laemmli sample buffer. The precipitates were then separated by SDS-PAGE, and the pull-down proteins were detected by immunoblotting with specific antibodies (Mo et al., 2007).

The cells were transfected with pCS4 Myc–RBP-Jk or the ubiquitin plasmid. Cells were lysed 48 hours post transfection in 1 ml of RIPA buffer for 30 minutes at 4°C. After 20 minutes of centrifugation at 12,000 g, the supernatants were subjected to immunoprecipitation with appropriate antibodies coupled to Protein-A–agarose beads. The resultant immunoprecipitates were washed three times in phosphate-buffered saline (PBS, pH 7.4). The samples were then treated with 5× protein sample buffer and boiled before immunoblotting. Western blotting was then conducted with the indicated antibodies (Mo et al., 2008).

The site-directed mutagenesis of cDNA encoding RBP-Jk was conducted using a QuikChange kit (Stratagene), and the mutagenic primer was T339A (5′-CCTTGCCCCAGTCgCTCCTGTGCCTGT-3′) (mismatch with the RBP-Jk cDNA template is indicated by lowercase letter). The mutations were confirmed by automatic DNA sequencing (Kim et al., 2008).

The cultured cells were harvested and lysed in buffer A containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM PMSF, 2 μg/ml of leupeptin, 2 μg/ml of aprotinin, 25 mM glycerophosphate, 0.1 mM sodium orthovanadate, 1 mM sodium fluoride, 1% NP-40, 0.5% deoxycholate and 0.1% SDS for 30 minutes at 4°C. The cell lysates were then subjected to 20 minutes of centrifugation at 12,000 g at 4°C. The soluble fraction was incubated for 3 hours with appropriate antibodies against the indicated protein kinase at 4°C. The immunocomplexes were then coupled to Protein-A–agarose for 1 hour at 4°C, after which they were pelleted by centrifugation. The immunopellets were rinsed three times in lysis buffer and then twice with 20 mM HEPES (pH 7.4). The immunocomplex kinase assays were conducted by incubation of the immunopellets for 30 minutes at 30°C with 2 μg of substrate protein in 20 μl of reaction buffer containing 0.2 mM sodium orthovanadate, 10 mM MgCl₂, 2 μCi [³²P]ATP and 20 mM HEPES (pH 7.4). The phosphorylated substrates were separated by SDS-PAGE and quantified with a Fuji BAS 2500 PhosphoImager (Mo et al., 2008). The GST fusion proteins to be used as substrates were expressed in E. coli using pGEX-4T (Pharmacia) and purified with GSH–Sepharose, as previously described. The protein concentrations were determined using the Bradford method (Shimadzu).

We would like to thank T. Honjo (Kyoto University, Japan) for the RBP-Jk-related constructs, Jiahuai Han (The Scripps Research Institute, La Jolla, CA) for providing the p38 MAPK constructs and Jin Tae Hong (Chungbuk National University, South Korea) for providing PS1 and PS2 constructs.