Amyloid-β (Aβ) neurotoxicity is believed to contribute to the pathogenesis of Alzheimer's disease (AD). Previously we found that E2-25K/Hip-2, an E2 ubiquitin-conjugating enzyme, mediates Aβ neurotoxicity. Here, we report that E2-25K/Hip-2 modulates caspase-12 activity via the ubiquitin/proteasome system. Levels of endoplasmic reticulum (ER)–resident caspase-12 are strongly up-regulated in the brains of AD model mice, where the enzyme colocalizes with E2-25K/Hip-2. Aβ increases expression of E2-25K/Hip-2, which then stabilizes caspase-12 protein by inhibiting proteasome activity. This increase in E2-25K/Hip-2 also induces proteolytic activation of caspase-12 through its ability to induce calpainlike activity. Knockdown of E2-25K/Hip-2 expression suppresses neuronal cell death triggered by ER stress, and thus caspase-12 is required for the E2-25K/Hip-2–mediated cell death. Finally, we find that E2-25K/Hip-2–deficient cortical neurons are resistant to Aβ toxicity and to the induction of ER stress and caspase-12 expression by Aβ. E2-25K/Hip-2 is thus an essential upstream regulator of the expression and activation of caspase-12 in ER stress–mediated Aβ neurotoxicity.

Amyloid-β (Aβ) is thought to be the principle cause of the pathogenesis and progression of the neuron and memory loss associated with Alzheimer's disease (AD; for review see Barnham et al., 2006). Substantial evidence suggests that Aβ-induced neuronal apoptosis is central to the progression of AD (Abad et al., 2006; Biswas et al., 2007). This has prompted investigation of the potential for caspase family cystein proteases, which are activated during apoptosis and execute various apoptotic processes to serve as therapeutic targets in the treatment of neurodegenerative diseases (Sanges et al., 2006; Emamaullee et al., 2007). Among the 14 caspase isoforms known to be expressed in mammals, the ER-resident caspase-12 was initially identified as a mediator of Aβ neurotoxicity, and its conserved pathway, termed “ER stress,” has been implicated in the pathogenesis of such neurodegenerative ailments as Huntington's disease, prion disease, and AD (Nakagawa et al., 2000; Kouroku et al., 2002; Hetz et al., 2003, 2007; Hoozemans et al., 2007). However, the functional role of caspase-12 in humans remains to be further clarified. Human caspase-12 has no role in apoptotic pathways (Saleh et al., 2004) and it is still controversial that human caspase-12 acts as a functional counterpart of mouse caspase-12. Interestingly, the amino acid sequence of mouse caspase-12 has a 61% identity with human caspase-4, which is involved in apoptosis induced by ER stress (Hitomi et al., 2004) in the interleukin-1–converting enzyme homologous region. Thus, a fuller understanding of the mechanisms involved in regulating ER stress, caspase-12 activity, and their roles in Aβ neurotoxicity would be highly desirable.

The ubiquitin/proteasome system (UPS) is involved in many biological pathways, including regulation of the cell cycle and modulation of the degradation of short-lived and regulatory proteins (for review see Rubinsztein, 2006). Protein degradation by the UPS proceeds through two successive steps: (1) ubiquitination, i.e., conjugation of ubiquitin to a target protein via the sequential actions of ubiquitin-activating (E1), ubiquitin-conjugating (E2), and ubiquitin-ligating (E3) enzymes; and (2) degradation, i.e., recognition of the Lys48 polyubiquitin chain by 26S proteasome and degradation of the target protein with generation of free ubiquitin by ubiquitin-recycling enzymes. Malfunction of the UPS leading to accumulation of aggregation-prone proteins is thought to be involved in some neurodegenerative diseases including AD (Al-Ramahi et al. 2006; Kristiansen et al., 2007; for review see Rubinsztein, 2006). Indeed, it was proposed that E2-25K/Hip-2, an E2 ubiquitin-conjugating enzyme, acts as an essential mediator of Aβ neurotoxicity by promoting the inhibition of proteasome (Song et al., 2003; for review see Song and Jung, 2004). Furthermore, other evidence has shown that the inhibition of proteasome activity promotes ER stress (Nishitoh et al., 2002), suggesting that altered regulation of proteasome activity and ER stress may be associated with Aβ neurotoxicity. However, a critical mediator coordinating ER stress and caspase-12 activity in Aβ neurotoxicity remains unknown. Here we show that E2-25K/Hip-2 regulates the activation of caspase-12 and ER stress responses during Aβ neurotoxicity, and that E2-25K/Hip-2–deficient cortical neurons cultured from E2-25K/Hip-2 knockout mice lack Aβ-induced ER stress responses, including accumulation of caspase-12, and are resistant to Aβ toxicity.

Regulation of caspase-12 expression by E2-25K/Hip-2 during Aβ toxicity

To assess the relationship between E2-25K/Hip-2 and caspase-12, we initially performed an immunohistochemical analysis of the brains of Tg2576 mice (Hsiao et al., 1996). We found that, as with E2-25K/Hip-2 (Song et al., 2003), there was much greater expression of caspase-12 in the hippocampal region of the Tg2576 mouse brain than in their wild-type littermates (Fig. 1 A). Interestingly, the E2-25K/Hip-2 immunoreactivity colocalized with that of caspase-12 around amyloid plaques, but not with that of caspase-2, which is also known to mediate Aβ neurotoxicity (Troy et al., 2000). In addition, Western blotting showed that levels of both caspase-12 and E2-25K/Hip-2 were increased in the brains of 6- and 9-mo-old double transgenic mice expressing Swedish mutant Aβ precursor protein (APPswe) and exon 9–deleted presenilin 1 (PS1dE9; Fig. 1, B and C). These results imply there is concerted regulation of caspase-12 and E2-25K/Hip-2 in the brains of these AD model mice.

We then examined the expression and activation of caspase-12 in rat B103 neuroblastoma cells stably expressing the antisense E2-25K/Hip-2 cDNA (B103/E2-25K-AS cells; Fig. 2 A, left). As compared with control cells, mixed populations of B103/E2-25K-AS cells showed reduced expression of caspase-12 but no changes in the expression of caspase-2 and -8 (Fig. 2 A, right). Treatment with Aβ induced the accumulation and proteolytic processing of caspase-12 in control cells, but these effects were substantially reduced in B103/E2-25K-AS cells. Conversely, ectopic expression of E2-25K/Hip-2 efficiently induced proteolytic activation of caspase-12 in B103 cells (Fig. 2 B). Collectively, these results suggest that Aβ-induced accumulation and the proteolytic activation of caspase-12 protein may be mediated by E2-25K/Hip-2. RT-PCR analysis showed that the basal levels of caspase-12 mRNA in B103/E2-25K-AS cells were also somewhat lower than in control cells (Fig. 2 C, left) and that caspase-12 mRNA was apparently induced by Aβ in control cells but not in B103/E2-25K-AS #1 cells (Fig. 2 C, right). When reporter assays were performed with pGL3-3.0 (5′ flanking region + 1' intron + 2' exon) and pGL3-0.8 (1' intron + 2' exon; Oubrahim et al., 2005), the promoter activity of caspase-12 was weak in B103 cells (twofold higher than control) compared with NIH 3T3 cells (sevenfold higher than control; Oubrahim et al., 2005) and the promoter activity of caspase-12 in pGL3-0.8 was slightly increased by E2-25K/Hip-2 (Fig. S1). However, a quantitative comparison of our Western and RT-PCR analyses revealed that the reduction in caspase-12 protein was greater than that in caspase-12 mRNA in B103/E2-25K-AS cells (Fig. 2 C, left), suggesting that E2-25K/Hip-2 regulates caspase-12 at both the mRNA and protein levels.

Caspase-12 is a downstream mediator of E2-25K/Hip-2 in ER stress–induced cell death

Because caspase-12 is known to play an essential role in the apoptosis induced by ER stress, we examined the involvement of E2-25K/Hip-2 in that process. Treating cells with thapsigargin or tunicamycin, which cause ER stress, induced cell death in >80% of control B103 cells but in a much smaller percentage of B103/E2-25K-AS cells (Fig. 2 D, left). Our Western analyses showed that thapsigargin and tunicamycin induced proteolytic activation of caspase-12 in control cells (Fig. 2 D, right), but levels of both procaspase-12 and activated caspase-12 were reduced in B103/E2-25K-AS cells. In contrast, GADD153/CHOP, an ER stress marker (Rao et al., 2002), was induced equally in both cell lines. E2-25K/Hip-2 thus appears to contribute to ER stress–induced cell death and the activation of caspase-12, but not the expression of at least some ER stress markers.

Given our finding that the expression and activation of caspase-12 is regulated by E2-25K/Hip-2, we were eager to know whether caspase-12 is a downstream mediator of E2-25K/Hip-2 in the cell death (Fig. 3). As compared with the relatively weak proapoptotic activities of E2-25K/Hip-2 (32%) and caspase-12 (26%) alone, coexpressed E2-25K/Hip-2 and caspase-12 acted synergistically to efficiently induce apoptosis (76%) in B103 cells (Fig. 3 A). In addition, ectopic expression of a caspase-12 active site mutant (Cys298→Ser) significantly suppressed apoptosis triggered by E2-25K/Hip-2. Also, caspase-12 knockout mouse embryonic fibroblasts (MEFs; C12 [−/−]) were resistant to cell death induced by the overexpression of E2-25K/Hip-2 but not to cell death induced by caspase-8, whereas wild-type and caspase-11 knockout MEFs (C11 [−/−]) remained sensitive to E2-25K/Hip-2 (Fig. 3 B). This suggests caspase-12 is a critical mediator of E2-25K/Hip-2–mediated cell death.

E2-25K/Hip-2–mediated stabilization of caspase-12 protein via UPS

We next investigated how E2-25K/Hip-2 induces the accumulation of caspase-12 protein. Although we found that both caspase-12 mRNA and protein were regulated by E2-25K/Hip-2, we mainly focused on the stabilization of the protein because the regulation at the protein level was more pronounced and the higher molecular mass forms of caspase-12 protein were found to accumulate in the brains of 9-mo-old APPswe/PS1dE9 double transgenic mice (Fig. S2). E2-25K/Hip-2 is an E2 ubiquitin-conjugating enzyme capable of inhibiting proteasome activity (Song et al., 2003). We therefore hypothesized that caspase-12 may be regulated by protein degradation via proteasome. Western analysis showed that in contrast to caspase-3 and -8, there was substantial dose-dependant accumulation of caspase-12 protein in B103 cells incubated with the proteasome inhibitor MG132, but not with the lysosomal inhibitor NH4Cl (Fig. 4 A and Fig. S3). In contrast, the level of caspase-12 mRNA was unchanged in cells exposed to MG132 (Fig. 4 A). Immunoprecipitation analysis using antiubiquitin antibody confirmed that the accumulated forms of caspase-12 protein in cells exposed to MG132 were ubiquitinated (Fig. 4 B). In addition, an exogenous caspase-12–GFP fusion protein was stabilized in HEK293 cells incubated with MG132 (Fig. 4 C). Also, the level of caspase-12–GFP fusion protein was increased by the coexpression with wild-type E2-25K/Hip-2, but not with E2-25K/Hip-2 Ser86→Tyr (S86Y) or Cys92→Ser (C92S) mutant (Fig. 4 C), which lacked the capacity to inhibit proteasome function (Song et al., 2003). These results suggest that the degradation of caspase-12 protein is regulated by UPS and that E2-25K/Hip-2 stabilizes caspase-12 protein by inhibiting proteasome activity.

We used in vitro ubiquitination assays to further characterize E2-25K/Hip-2–dependent regulation of caspase-12 protein. Incubating ubiquitin with purified GST–E2-25K/Hip-2 in the presence of E1, ATP, and an ATP regeneration system led to the appearance of polyubiquitin (Fig. 4 D, left), which confirmed that the purified GST-E2-25K/Hip-2 protein is active and able to generate polyubiquitin. GST and GST-fused E2D (GST-E2D), another E2 ubiquitin-conjugating enzyme, served as negative controls. Incubating B103 cell extract supplemented with ubiquitin, E1, ATP, and an ATP regeneration system with either the purified GST–E2-25K/Hip-2 or MG132 led to the accumulation of caspase-12 protein, but not caspase-3 (Fig. 4 D, middle and right) or IκBα (not depicted). The stability of caspase-12 protein thus appears to reflect the ability of E2-25K/Hip-2 to generate polyubiquitin, which likely interferes with the degradation of caspase-12 through the regulation of UPS. We then determined the subregion of caspase-12 required for MG132-induced accumulation by using full length, prodomain-deleted mutant (ΔPro), prodomain (Pro), or the large subunit (p20) of caspase-12. Compared with Pro, ΔPro and p20 fragments are highly stabilized in the cells exposed to MG132 (10- and 7-fold each; Fig. S4).

Modulation of ER stress responses and calpainlike activity by E2-25K/Hip-2

Caspase-12 can be activated by calpain during ER stress (Nakagawa et al., 2000; Nakagawa and Yuan, 2000), which prompted us to test whether E2-25K/Hip-2 is also involved in the regulation of calpain activity. Treatment with Aβ induced the accumulation of the ER stress marker GADD153/CHOP in control B103 cells, but not in B103/E2-25K-AS cells (Fig. 5 A). In addition, a reporter assay using grp78 promoter luciferase showed that the overexpression of E2-25K/Hip-2, but not caspase-12, increased luciferase activity about threefold (Fig. 5 B), indicating that E2-25K/Hip-2 is required for the induction of ER stress markers by Aβ.

Using fluorogenic enzymatic assays, we consistently found that Aβ induced calpainlike activity in control cells (Fig. 5 C), but not in B103/E2-25K-AS cells (#1 and #3). In addition, calpainlike activity was also induced by the ectopic expression of E2-25K/Hip-2, but not by the expression of E2-25K/Hip-2–Δtail mutant (Fig. 5 D), which lacked the ability to inhibit proteasome activity (Song et al., 2003). Furthermore, E2-25K/Hip-2–induced cell death was significantly attenuated by incubating the cells with calpeptin or z-LLY, calpainlike protease inhibitors (Fig. 5 E), or by the overexpression of calpastatin, an endogenous inhibitor of calpain, with the reduction in the numbers of calpain activity-positive cells (Fig. 5 F). E2-25K/Hip-2 thus appears to mediate the induction of calpainlike activity by Aβ, which may lead to activation of caspase-12.

As oxidative stress is another potent regulator of ER stress and a downstream mediator of Aβ, we also investigated the regulation of E2-25K/Hip-2 and caspase-12 expression. We found that treating B103 cells with ascorbic acid, an antioxidant, but not Bapta-AM, a calcium chelator, reduced Aβ-induced expression of E2-25K/Hip-2 and neurotoxicity (Fig. 6 A, top). Also, treatment with H2O2 induced the accumulation of E2-25K/Hip-2 (Fig. 6 A, bottom) and caspase-12 (Fig. 6 C). In addition, H2O2-induced accumulation and activation of caspase-12 were suppressed in B103/E2-25K/Hip-2-AS cells (Fig. 6 C), and there was a concomitant inhibition of cell death (Fig. 6 B). Further, H2O2-induced accumulation of GADD153 observed in control B103 cells was also suppressed in E2-25K/Hip-2-AS cells (Fig. 6 C). These results suggest that E2-25K/Hip-2 is required for cell death triggered by H2O2-derived reactive oxygen species (ROS), which are also generated by Aβ, further supporting our proposal that E2-25K/Hip-2 mediates ER stress responses.

E2-25K/Hip-2–deficient cortical neurons are resistant to Aβ neurotoxicity

To evaluate the role of E2-25K/Hip-2 in the regulation of caspase-12 and ER stress during Aβ neurotoxicity in vivo, we used a general protocol from BayGenomics to generate E2-25K/Hip-2–deficient mice from embryonic stem cells in which a gene trap was inserted into the E2-25K/Hip-2 gene (Fig. S5). The genotypes of the gene-trap insertion and depletion of E2-25K/Hip-2 expression were confirmed using genomic DNA-PCR, RT-PCR, and Western analyses (Fig. 7 A). We found that the levels of caspase-12 protein were significantly lower in the brains of 7-d-old E2-25K/Hip-2 knockout (−/−) mice than wild-type (+/+) mice (Fig. 7 A), and that the levels of caspase-12 mRNA were reduced in the brains of 9-mo-old E2-25K/Hip-2–deficient mice (Fig. 7 B). Expression of both caspase-12 mRNA and protein thus appears to be regulated by E2-25K/Hip-2 in mice.

Bearing that in mind, we prepared in vitro cultures of primary cortical neurons from E2-25K/Hip-2 wild-type (+/+), heterozygotic (+/−), and homozygotic (−/−) embryos at embryonic day 15 and examined their sensitivity to Aβ neurotoxicity. We found that E2-25K/Hip-2 (−/−) neurons were completely resistant to Aβ, whereas E2-25K/Hip-2 (+/−) neurons showed partial resistance (Fig. 7 C). In addition, immunocytochemical analysis revealed that caspase-12 and GRP78 proteins, which showed marked accumulation in E2-25K/Hip-2 (+/+) cortical neurons exposed to Aβ, were not induced by Aβ in E2-25K/Hip-2 (−/−) neurons (Fig. 7 D), which confirms that E2-25K/Hip-2 is indeed an essential regulator of caspase-12 and ER stress during Aβ neurotoxicity.

The neurotoxicity of Aβ is a fundamental contributor to the pathogenesis and progression of AD (for review see Barnham et al., 2006). To better understand Aβ neurotoxicity, we examined the functional interactions of E2-25K/Hip-2, caspase-12, and ER stress, and we report here that E2-25K/Hip-2 regulates the accumulation and activation of caspase-12 in vitro and in vivo as a necessary step leading to Aβ-induced neuronal cell death. It is also clear that E2-25K/Hip-2 plays a central role in ER stress–associated cell death triggered by Aβ and H2O2: E2-25K/Hip-2 mediates the induction of two ER stress markers, GADD153/CHOP and GRP78, as well as the activation of calpain (Fig. 8). In particular, experiments performed with E2-25K/Hip-2 knockout cortical neurons confirmed the essential role of E2-25K/Hip-2 in Aβ-induced accumulation and activation of caspase-12 for neurotoxicity. Interestingly, the accumulation of caspase-12 protein occurred via proteasome inhibition by E2-25K/Hip-2, which also ubiquitinates UBB+1, causing the inhibition of proteasomeal activity in a dose-dependant way (for review see Song and Jung, 2004; van Tijn et al., 2007). Though the accumulation of caspase-12 protein through proteasome inhibition is strongly plausible, it is still possible that E2-25K/Hip-2 may also accumulate caspase-12 protein by generating K63-linked ubiquitin chain (for review see Lim et al., 2006). In contrast, the accumulation of caspase-12 mRNA was not observed in cells exposed to MG132, though the promoter activity of caspase-12 (pGL3-08) was weakly induced by coexpression of E2-25K/Hip-2. Thus, the regulation of caspase-12 mRNA by E2-25K/Hip-2 is separated from its proteasome inhibitory activity.

Although caspase-12 is a proximal caspase and is an important mediator of apoptosis triggered by ER stress, we believe its accumulation may not, itself, be sufficient to induce cell death. When we overexpressed caspase-12 in various cell types, including fibroblasts and neuronal cells, it was much less effective in inducing cell death than other proximal caspases, such as caspase-8, -9, and -10 (unpublished data). To be effective, apparently, caspase-12 needs to interact with one or more activators. In our study, E2-25K/Hip-2 was able to induce both calpainlike activity and efficient proteolytic processing of caspase-12. One possible explanation is that the inhibition of proteasome activity by E2-25K/Hip-2 leads to an accumulation of misfolded proteins within cells, which in turn induces ER stress, including the activation of calpain. Consistent with that idea, it has been shown that the inhibition of proteasome activity by aggregation-prone proteins or proteasome inhibitors does indeed induce ER stress (Kouroku et al., 2002; Nishitoh et al., 2002; Nawrocki et al., 2005). Still, the details of the molecular pathway downstream of E2-25K/Hip-2 leading to induction of calpainlike activity remain unclear.

Along with those of an earlier paper showing that calpain directly activates caspase-12 (Nakagawa et al., 2000), our present findings suggest that calpainlike activity is required for proteolytic activation of caspase-12. This means that E2-25K/Hip-2 regulates both the accumulation of caspase-12 and its proteolytic activation during Aβ neurotoxicity. In that sense, calpain inhibitors suppressed Aβ neurotoxicity. Similar accumulation and activation of caspase-12 was observed in neuronal cells exposed to oxidative stress. However, calpain inhibitors still partially suppressed E2-25K/Hip-2–induced cell death, indicating that an additional regulator (e.g., Jun N-terminal kinase) also functions as a downstream mediator of E2-25K/Hip-2 during cell death (Song et al., 2003).

We found that treating cultured cells with Aβ led to E2-25K/Hip-2–dependent up-regulation of two ER stress markers, GADD153/CHOP and GRP78, as well as caspase-12, suggesting E2-25K/Hip-2 is an upstream regulator of ER stress in Aβ neurotoxicity (Fig. 8). If so, the observed accumulation of caspase-12 in the brains of APP transgenic mice might be induced by the increased expression of E2-25K/Hip-2, which would be expected to further increase neuronal susceptibility to cell death or stress. We believe that E2-25K/Hip-2 may induce ER stress via generating unanchored polyubiquitin, which is able to inhibit proteasome activity in vitro like polyubiquitinated UBB+1 (Song et al., 2003; van Tijn et al., 2007; for review see Song and Jung, 2004). That said, it is noteworthy that the contributions of E2-25K/Hip-2 to the induction of GADD153/CHOP in cells exposed to various ER stress signals differ. For instance, E2-25K/Hip-2 is required for the induction of GADD153/CHOP by Aβ and ROS, but not by tunicamycin or thapsigargin, implying that E2-25K/Hip-2 might be more specific to ER stress response caused by Aβ than by tunicamycin or thapsigargin. Thus, a possible model is that E2-25K/Hip-2 may be situated upstream of intracellular calcium destabilization or at the accumulation of unfolded proteins found in the case of Aβ neurotoxicity (Fig. 8; Nakagawa et al., 2000; Ferreiro et al., 2006).

It has been proposed that oxidative stress contributes to the progression of pathological processes in neurons, including AD (Manton et al., 2004; for review see Huber et al., 2006). We found that E2-25K/Hip-2 accumulated in B103 cells exposed to ROS and that antioxidants, which suppress Aβ neurotoxicity (Hensley et al., 1994; Bruce et al., 1996), inhibited the accumulation of E2-25K/Hip-2. The regulation of E2-25K/Hip-2 expression may thus be involved in mediating oxidative damage underlying neuronal pathology. Recently, E2-25K/Hip-2 was proposed to be involved in the aggregation of polyglutamine-expanded Huntingtin with proteasome inhibitory activity (de Pril et al., 2007). Thus, it would be interesting to know whether E2-25K/Hip-2 is associated with the pathogenesis of other neurodegenerative ailments, including Huntington's and Parkinson's disease, because malfunction of the UPS, its diminished activity, and generation of ROS are common phenomena in neurodegenerative diseases (for review see Song and Jung, 2004; Halliwell, 2006).

Aβ-induced, E2-25K/Hip-2–dependent regulation of caspase-12 and ER stress was confirmed in E2-25K/Hip-2–deficient neurons. Despite the important role of E2-25K/Hip-2 in the regulation of ER stress, E2-25K/Hip-2–deficient (−/−) mice were well bred and healthy until the age of 9 mo. The absence of any detectable defects during the development of these mice suggests that the function of E2-25K/Hip-2 might be compensated for during that period by other E2s, like Mdm2, which interacts with several E2s to ubiquitinate tumor suppressor p53 (Saville et al., 2004). Alternatively, E2-25K/Hip-2 may be mainly associated with neuronal pathogenesis. Although a human homologue of caspase-12 is in debate, our identification of E2-25K/Hip-2 as an upstream regulator of ER stress suggests that E2-25K/Hip-2 could serve as a plausible therapeutic target for the treatment of AD.

Proteasome and calpain inhibitor

Proteasome inhibitor MG132 (Sigma-Aldrich), calpain inhibitors z-LLY (EMD), and calpeptin (EMD) were purchased.

Plasmid construction and RT-PCR

pE2-25K/Hip-2 and its mutants (Δtail, deletion of tail region; S86Y, Ser→Tyr; C92S substitution, Cys→Ser) were described previously (Song et al., 2003). pCaspase-12–GFP and its active site mutant (C298S) have also been described previously (Nakagawa et al., 2000). E2-25K/Hip-2 cDNA was subcloned into pcDNA3 in an antisense orientation (pAS-E2-25K/Hip-2). To coexpress with calpastatin, E2-25K/Hip-2 was cloned into the EcoRI and BamHI sites of pDsReD-C2 (Clontech Laboratories, Inc.) using synthetic oligonucleotides mRed-Hip-2-5′ (5′-GCGAATTCTATGGCCAACATCGCGGTG-3′) and mRed-Hip-2-3′ (5′-GCGGATCCTCAGTTACTCAGAAGCAA-3′). Human calpastatin cDNA cloned in pCMV-SPORT6 vector was purchased from Korea Research Institute of Bioscience and Biotechnology.

Total RNA was purified and reverse transcribed as described previously (Song et al., 2003). Levels of E2-25K/Hip-2, caspase-12, β-actin, and glyceraldehyde-3-phosphate dehydrogenase mRNA were analyzed using PCR with gene-specific synthetic oligonucleotides.

Cell culture, DNA transfection, and assessment of cell death

Wild-type MEFs, caspase-11 (−/−) MEFs, and caspase-12 (−/−) MEFs were provided by J. Yuan (Harvard Medical School, Boston, MA). MEFs, B103 cells (rat neuroblastoma), and HEK293 cells (human embryonic kidney cells) were cultured in DME supplemented with 10% (vol/vol) fetal bovine serum. Cells were transfected using Lipofectamine Reagent (Invitrogen) according to the manufacturer's protocol, after which their viability was assessed based on the morphology of GFP-positive cells viewed under a fluorescence microscope (DMRBE; Leica), trypan blue exclusion assays, and live/dead cell assays (Invitrogen).

Preparation and treatment with Aβ1-42 peptide

Commercially available Aβ1-42 was purchased from Sigma-Aldrich and dissolved to a concentration of 500 μM in phosphate-buffered saline. The Aβ1-42 stock solution was incubated for 1 wk at 4°C and divided into small aliquots for storage at −70°C. Cells were incubated with Aβ1-42 or other cell death–inducing drugs including tunicamycin (Sigma-Aldrich) and thapsigargin (Sigma-Aldrich) in serum-free DME.

Luciferase and β-galactosidase (β-gal) assays

B103 cells were cotransfected with Grp78-luciferase reporter plasmid, cytomegalovirus β-gal, and effecter plasmids. After 32 h, the cells were harvested and the luciferase activities in the cell extracts were determined using a Luciferase assay system (Promega). To measure β-gal activity, cell extracts were mixed with equal amounts of β-gal buffer (2×) containing 200 mM sodium phosphate, pH 7.3, 2 mM MgCl2, 100 mM β-mercaptoethanol, and 1.33 mg/ml O-Nitrophenyl-β-d-galactopyranoside, and incubated for 1 h at 37°C. The absorbance at 420 nm was then measured using a microplate reader (Bio-Rad Laboratories).

Enrichment of antisense E2-25K/Hip-2 cDNA-transfected cells

B103 cells were transfected with pAS-E2-25K/Hip-2 or pcDNA3 for 24 h and incubated with 800 μg/ml G418 sulfate (Invitrogen) for 10 d to generate a stable mixed cell population. Single-cell clones (E2-25K-AS #1 and #3) were isolated using standard cell cloning methods and examined for expression of E2-25K/Hip-2 using Western analysis.

Transgenic mice

Tg2576 transgenic mice (13–22 mo of age) were used in our study. Each expressed human APP 695 containing the double K670N/M671L mutation (huAPP695.K67ON/M671L) found in a large Swedish family with early onset AD (Hsiao et al., 1996). Double transgenic mice expressing APPswe and PS1dE9 exon 9–deleted PS1dE9 (The Jackson Laboratory) were purchased and bred to generate transgenic mice and wild-type littermates.

Antibody generation, Western blotting, and immunocytochemistry

Generation of anti–E2-25K/Hip-2 and anticaspase-8 antibodies was described previously (Kim et al., 2002; Song et al., 2003). Anti–α-tubulin (Sigma-Aldrich), anticaspase-2 (Santa Cruz Biotechnology, Inc.), anticaspase-3 (Biomeda), anticaspase-12 (Santa Cruz Biotechnology, Inc.), and anti-GFP (Santa Cruz Biotechnology, Inc.) antibodies were purchased. Cells were lyzed in sampling buffer (10% glycerol, 2% SDS, 62.5 mM Tris-HCl, and 2% β-mercaptoethanol, pH 6.8), after which aliquots of lysate containing equal amounts of protein were subjected to SDS-PAGE and Western analysis as described previously (Kim et al., 2002). Primary neurons were analyzed with immunocytochemistry using anti-GRP78 (dilution 1:200; Santa Cruz Biotechnology, Inc.) and anticaspase-12 (dilution 1:50) primary antibodies and Alexa Fluor 488– or 555–conjugated secondary antibodies (1:1,000; Invitrogen) as described previously (Song et al., 2003).

Histology and immunohistochemistry

12-μm-thick frozen brain sections, which were taken from the human Swedish mutant amyloid precursor protein (huAPP695.K670N/M671L) transgenic Tg2576 and their littermate mice at the age of 21 mo, were fixed with 4% PFA and blocked with 3% normal serum and 0.3% Triton X-100 in phosphate-buffered saline, pH 7.4. After coincubation with rabbit anti–E2-25K/Hip-2 antibody (dilution, 1:200) and rat anticaspase-12 antibody (1:25), the sections were reacted with Alexa Fluor 488– or 555–conjugated secondary antibody (1:1,000). For the double fluorescent immunostaining with rabbit anti–E2-25K/Hip-2 antibody and rabbit anticaspase-2 antibody (1:100), we treated the primary antibodies with Zenon Alexa Fluor rabbit IgG labeling kit (Invitrogen) before the reaction with the tissue sections. All stained samples were mounted using a mounting medium for fluorescence (VECTASHIELD; Vector Laboratories).


Immunohistochemical studies were photographed under a fluorescence microscope (Eclipse 80i with Plan Apo VC 100×/1.40 Oil WD 0.13. lens; Nikon) using the manufacturer's acquisition system (DS-Fi1/DS-U2 digital camera and NIS-Elements F 2.20 program). Fluorescence (IX71S1F-3 with two objective lenses [LCPlanFl, 40×/0.60 Ph2 and 20×/0.45 Ph1]; Olympus) and confocal microscopes (UltraVIEW ERS FRET-EH with an oil immersion objective lens [DIC-H, 100×/1.40]; PerkinElmer] were used for analysis of the immunostained samples. Images were captured using each manufacturer's acquisition system (DP20 digital camera and DP ver. manager program [Olympus]; UltraVIEW LCI [PerkinElmer]). All stained samples were examined at room temperature and Photoshop (Adobe) was used to make combinatory figures (color contrast, medium).

In vitro degradation assay

Various GST-fused proteins cloned into pGEX4T-1 (GST–E2-25K/Hip-2 [provided by S. Kang, Korea University, Seoul, Korea] and GST-E2D) were expressed in DH5α cells and purified using glutathione–Sepharose 4B (GE Healthcare). For in vitro degradation assays, B103 cells (5 × 107) were harvested and lyzed in tris-based buffer (10 mM Tris-HCl, 0.05% NP-40, 150 mM NaCl, 3 mM MgCl2, 0.5 mM PMSF, 1 mM EDTA, 1 mM ATP, and 1 mM DTT). Cell extract was incubated for 2 h at 37°C in the presence or absence of GST-fused proteins preincubated with purified ubiquitin (Sigma-Aldrich), ubiquitin-activating enzyme E1 (Macrogen), and 10 mM ATP. Protein degradation was then analyzed by Western blotting.

Measurement of calpainlike protease activity

The rhodamine 110-based substrate, (t-BOC-Leu-Met)2-R110, and chloromethylcoumarin-based substrate, CMAC-(t-BOC-Leu-Met; Invitrogen), were used as a calpain substrate to assess calpainlike protease activity in intact cells. When (t-BOC-Leu-Met)2-R110 was used, B103 cells were sedimented and resuspended in Hepes-buffered saline (5 mM Hepes and 0.15 NaCl, pH 7.35) containing 2 mM EDTA (Hepes-buffered saline–EDTA) to a density of ∼0.5−107 cells/ml. Cells diluted by 100× were incubated with 10 μM of substrate for 20 min at 37°C. After washing twice with HBSS-EDTA, the fluorescence emitted by the cells was measured using a fluorescence microplate reader (FL-600; BioTek Instruments, Inc.). In the case of CMAC-(t-BOC-Leu-Met), B103 cells were incubated with 1 μM of substrate for 30 min and examined under a fluorescence microscope using a Hoescht filter.

Generation of E2-25K/Hip-2–deficient mice

E2-25K/Hip-2 gene-trapped embryonic stem cells (XK109) were provided by BayGenomics of the International Gene Trap Consortium. Mice heterozygotic for E2-25K/Hip-2 were generated by following the protocol provided by BayGenomics. Insertion of the gene-trap vector was confirmed by Southern blotting using NEBlot Phototope (New England Biolabs, Inc.) with a β-gal probe recommended by BayGenomics. Genotypes were analyzed using genomic PCR with synthetic oligonucleotides for β-gal (β-gal-5′, 5′-TTATCGATGAGCGTGGTGGTTATGC-3′; β-gal-3′, 5′-GCGCGTACATCGGGCAAATAATATC-3′) and β-actin. The expression of E2-25K/Hip-2 was examined using RT-PCR with synthetic oligonucleotides derived from E2-25K/Hip-2 (E2-25K/Hip-2-5′ and E2-25K/Hip-2-3′) and β-gal. The absence of E2-25K/Hip-2 expression in homozygous mice was confirmed by Western analysis of the proteins extracted from the brains of 5-d-old mice.

Primary culture of cortical neurons

E2-25K/Hip-2 heterozygous mice were interbred to generate homozygous, heterozygous, and wild-type embryos. Cortical tissues from embryonic day 14.5 embryonic brains were dissociated by incubation with 0.01% trypsin-EDTA (Invitrogen) and plated on culture dishes coated with poly-l-lysine (0.01% in 100 mM borate buffer, pH 8.5). Details of the procedures were described previously (Song et al., 2003). Cortical neurons from each embryo were plated separately and their genotypes were examined by Western blotting and genomic PCR. Neurons were cultured for 3 d before treatment with Aβ in the serum-free minimum essential medium.

Online supplemental material

In Fig. S1 the reporter assay showed that the promoter activity of caspase-12 in pGL3-0.8 was slightly increased by E2-25K/Hip-2. Western analysis showed that the higher molecular mass forms of caspase-12 protein accumulated in the brains of 9-mo-old APPswe/PS1dE9 double transgenic mice (Fig. S2), and that caspase-12 protein accumulated in B103 cells incubated with the proteasome inhibitor MG132 (Fig. S3). Further, in Fig. S4 DPro and p20 fragments of caspase-12 were highly stabilized in the cells exposed to MG132 (10- and sevenfold each). Fig. S5 shows the genomic locus of the gene-trap insertion in E2-25K/Hip-2–deficient mice.

Supplemental materials and methods

Lysosomal inhibitor.

Lysosome inhibitor ammonium chloride (NH4Cl; Sigma-Aldrich) was purchased.

Plasmid construction.

Serial deletions of caspase-12 were generated by PCR using synthetic oligonucleotides (C12full 5′KpnI [5′-GGTACCCGCATGGCGGCCAGGAGGACACAT-3′], C12full 3′KpnI [5′-GGTACCCGCATTCCCGGGAAAAAGGTAGAA-3′], C12Δpro 5′HindIII [5′-AAGCTTCGCATGTGTCCACGTGATCAGT-3′], C12Δpro 3′HindIII [5′-AAGCTTCGCATTCCCGGGAAAAAGGTAGAA-3′], C12pro 5′HindIII [5′-AAGCTTCGCATGGCGGCCAGGAGGACACAT-3′], C12pro 3′HindIII [5′-AAGCTTCAGTGTATCTTGGACTTCTGA-3′], and C12p20 3′KpnI [5′-GGTACCCGCATCAGCAGTGGCTATCCC-3′]), after which the PCR products were subcloned into pcDNA3-HA. Caspase-12 promoter constructs (pGL3-3.0 and pGL3-0.8) were previously described (Oubrahim et al., 2005).

S. Song and T.-I. Kam were supported in part by the BK21 program. This work was supported by grants from the Brain Research Center of the 21st Century Frontier Research Program, by Functional Cellulomics funded by the Korea Science and Engineering Foundation of the Korean government, by Basic Science program of Korea Research Foundation (Y.K. Jung), and by a grant of the Korea Health 21 R&D Project, Ministry of Health and Welfare, Korea (A040147).

Abad, M.A., M. Enguita, N. DeGregorio-Rocasolano, I. Ferrer, and R. Trullas.
. Neuronal pentraxin 1 contributes to the neuronal damage evoked by amyloid-beta and is overexpressed in dystrophic neurites in Alzheimer's brain.
J. Neurosci.
Al-Ramahi, I., Y.C. Lam, H.K. Chen, B. de Gouyon, M. Zhang, A.M. Perez, J. Branco, M. de Haro, C. Patterson, H.Y. Zoghbi, and J. Botas.
. CHIP protects from the neurotoxicity of expanded and wild-type ataxin-1 and promotes their ubiquitination and degradation.
J. Biol. Chem.
Barnham, K.J., R. Cappai, K. Beyreuther, C.L. Masters, and A.F. Hill.
. Delineating common molecular mechanisms in Alzheimer's and prion diseases.
Trends Biochem. Sci.
Biswas, S.C., Y. Shi, J.P. Vonsattel, C.L. Leung, C.M. Troy, and L.A. Greene.
. Bim is elevated in Alzheimer's disease neurons and is required for beta-amyloid-induced neuronal apoptosis.
J. Neurosci.
Bruce, A.J., B. Malfroy, and M. Baudry.
. beta-Amyloid toxicity in organotypic hippocampal cultures: protection by EUK-8, a synthetic catalytic free radical scavenger.
Proc. Natl. Acad. Sci. USA.
de Pril, R., D.F. Fischer, R.A. Roos, and F.W. van Leeuwen.
. Ubiquitin-conjugating enzyme E2-25K increases aggregate formation and cell death in polyglutamine diseases.
Mol. Cell. Neurosci.
Emamaullee, J.A., L. Stanton, C. Schur, and A.M. Shapiro.
. Caspase inhibitor therapy enhances marginal mass islet graft survival and preserves long-term function in islet transplantation.
Ferreiro, E., R. Resende, R. Costa, C.R. Oliveira, and C.M. Pereira.
. An endoplasmic-reticulum-specific apoptotic pathway is involved in prion and amyloid-beta peptides neurotoxicity.
Neurobiol. Dis.
Halliwell, B.
. Oxidative stress and neurodegeneration: where are we now?
J. Neurochem.
Hensley, K., J.M. Carney, M.P. Mattson, M. Aksenova, M. Harris, J.F. Wu, R.A. Floyd, and D.A. Butterfield.
. A model for beta-amyloid aggregation and neurotoxicity based on free radical generation by the peptide: relevance to Alzheimer disease.
Proc. Natl. Acad. Sci. USA.
Hetz, C., M. Russelakis-Carneiro, K. Maundrell, J. Castilla, and C. Soto.
. Caspase-12 and endoplasmic reticulum stress mediate neurotoxicity of pathological prion protein.
Hetz, C., J. Castilla, and C. Soto.
. Perturbation of endoplasmic reticulum homeostasis facilitates prion replication.
J. Biol. Chem.
Hitomi, J., T. Katayama, Y. Eguchi, T. Kudo, M. Taniguchi, Y. Koyama, T. Manabe, S. Yamagishi, Y. Bando, K. Imaizumi, et al.
. Involvement of caspase-4 in endoplasmic reticulum stress-induced apoptosis and Aβ-induced cell death.
J. Cell Biol.
Hoozemans, J.J., E.S. van Haastert, P. Eikelenboom, R.A. de Vos, J.M. Rozemuller, and W. Scheper.
. Activation of the unfolded protein response in Parkinson's disease.
Biochem. Biophys. Res. Commun.
Hsiao, K., P. Chapman, S. Nilsen, C. Eckman, Y. Harigaya, S. Younkin, F. Yang, and G. Cole.
. Correlative memory deficits, Abeta elevation, and amyloid plaques in transgenic mice.
Huber, A., G. Stuchbury, A. Burkle, J. Burnell, and G. Munch.
. Neuroprotective therapies for Alzheimer's disease.
Curr. Pharm. Des.
Kim, K.W., B.J. Kim, C.W. Chung, D.G. Jo, I.K. Kim, Y.H. Song, Y.K. Kwon, H.N. Woo, and Y.K. Jung.
. Caspase cleavage product lacking amino-terminus of IkappaBalpha sensitizes resistant cells to TNF-alpha and TRAIL-induced apoptosis.
J. Cell. Biochem.
Kouroku, Y., E. Fujita, A. Jimbo, T. Kikuchi, T. Yamagata, M.Y. Momoi, E. Kominami, K. Kuida, K. Sakamaki, S. Yonehara, and T. Momoi.
. Polyglutamine aggregates stimulate ER stress signals and caspase-12 activation.
Hum. Mol. Genet.
Kristiansen, M., P. Deriziotis, D.E. Dimcheff, G.S. Jackson, H. Ovaa, H. Naumann, A.R. Clarke, F.W. van Leeuwen, V. Menendez-Benito, N.P. Dantuma, et al.
. Disease-associated prion protein oligomers inhibit the 26S proteasome.
Mol. Cell.
Lim, K.L., V.L. Dawson, and T.M. Dawson.
. Parkin-mediated lysine 63-linked polyubiquitination: a link to protein inclusions formation in Parkinson's and other conformational diseases?
Neurobiol. Aging.
Manton, K.G., S. Volovik, and A. Kulminski.
. ROS effects on neurodegeneration in Alzheimer's disease and related disorders: on environmental stresses of ionizing radiation.
Curr. Alzheimer Res.
Nakagawa, T., and J. Yuan.
. Cross-talk between two cysteine protease families: activation of caspase-12 by calpain in apoptosis.
J. Cell Biol.
Nakagawa, T., H. Zhu, N. Morishima, E. Li, J. Xu, B.A. Yankner, and J. Yuan.
. Caspase-12 mediates endoplasmic-reticulum-specific apoptosis and cytotoxicity by amyloid-beta.
Nawrocki, S.T., J.S. Carew, K. Dunner Jr., L.H. Boise, P.J. Chiao, P. Huang, J.L. Abbruzzese, and D.J. McConkey.
. Bortezomib inhibits PKR-like endoplasmic reticulum (ER) kinase and induces apoptosis via ER stress in human pancreatic cancer cells.
Cancer Res.
Nishitoh, H., A. Matsuzawa, K. Tobiume, K. Saegusa, K. Takeda, K. Inoue, S. Hori, A. Kakizuka, and H. Ichijo.
. ASK1 is essential for endoplasmic reticulum stress-induced neuronal cell death triggered by expanded polyglutamine repeats.
Genes Dev.
Oubrahim, H., J. Wang, E.R. Stadtman, and P.B. Chock.
. Molecular cloning and characterization of murine caspase-12 gene promoter.
Proc. Natl. Acad. Sci. USA.
Piotrowski, J., R. Beal, L. Hoffman, K.D. Wilkinson, R.E. Cohen, and C.M. Pickart.
. Inhibition of the 26 S proteasome by polyubiquitin chains synthesized to have defined lengths.
J. Biol. Chem.
Rao, R.V., A. Peel, A. Logvinova, G. del Rio, E. Hermel, T. Yokota, P.C. Goldsmith, L.M. Ellerby, H.M. Ellerby, and D.E. Bredesen.
. Coupling endoplasmic reticulum stress to the cell death program: role of the ER chaperone GRP78.
FEBS Lett.
Rubinsztein, D.C.
. The roles of intracellular protein-degradation pathways in neurodegeneration.
Saleh, M., J.P. Vaillancourt, R.K. Graham, M. Huyck, S.M. Srinivasula, E.S. Alnemri, M.H. Steinberg, V. Nolan, C.T. Baldwin, R.S. Hotchkiss, et al.
. Differential modulation of endotoxin responsiveness by human caspase-12 polymorphisms.
Sanges, D., A. Comitato, R. Tammaro, and V. Marigo.
. Apoptosis in retinal degeneration involves cross-talk between apoptosis-inducing factor (AIF) and caspase-12 and is blocked by calpain inhibitors.
Proc. Natl. Acad. Sci. USA.
Saville, M.K., A. Sparks, D.P. Xirodimas, J. Wardrop, L.F. Stevenson, J.C. Bourdon, Y.L. Woods, and D.P. Lane.
. Regulation of p53 by the ubiquitin-conjugating enzymes UbcH5B/C in vivo.
J. Biol. Chem.
Song, S., and Y.K. Jung.
. Alzheimer's disease meets the ubiquitin-proteasome system.
Trends Mol. Med.
Song, S., S.Y. Kim, Y.M. Hong, D.G. Jo, J.Y. Lee, S.M. Shim, C.W. Chung, S.J. Seo, Y.J. Yoo, J.Y. Koh, et al.
. Essential role of E2-25K/Hip-2 in mediating amyloid-beta neurotoxicity.
Mol. Cell.
Troy, C.M., S.A. Rabacchi, W.J. Friedman, T.F. Frappier, K. Brown, and M.L. Shelanski.
. Caspase-2 mediates neuronal cell death induced by beta-amyloid.
J. Neurosci.
van Tijn, P., F.M. de Vrij, K.G. Schuurman, N.P. Dantuma, D.F. Fischer, F.W. van Leeuwen, and E.M. Hol.
. Dose-dependent inhibition of proteasome activity by a mutant ubiquitin associated with neurodegenerative disease.
J. Cell Sci.

Abbreviations used in this paper: AD, Alzheimer's disease; Aβ, amyloid-β; APP, Aβ precursor protein; β-gal, β-galactosidase; KD, knockdown; MEF, mouse embryonic fibroblast; PS1, presenilin 1; ROS, reactive oxygen species; UPS, ubiquitin/proteasome system.

Supplementary data