GTC
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS
 QUICK SEARCH:   [advanced]


     


Genes to Cells (2004) 9, 457-469. doi:10.1111/j.1356-9597.2004.00735.x
© 2004 Blackwell Publishing or its licensors

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hori, O.
Right arrow Articles by Ogawa, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hori, O.
Right arrow Articles by Ogawa, S.

Role of Herp in the endoplasmic reticulum stress response

Osamu Hori1,4,*, Fusae Ichinoda1, Atsushi Yamaguchi2,4, Takashi Tamatani1, Manabu Taniguchi2,4, Yoshihisa Koyama2,4, Taiichi Katayama2,4, Masaya Tohyama2,4, David M. Stern3, Kentaro Ozawa1,4, Yasuko Kitao1 and Satoshi Ogawa1,4

1 Department of Neuroanatomy (Anatomy III), Kanazawa University Graduate School of Medicine, Kanazawa City, Ishikawa 920-8640, Japan
2 Department of Anatomy and Neuroscience, Osaka University Graduate School of Medicine, Suita City, Osaka, 565-0871, Japan
3 Medical College of Georgia, Augusta, Georgia 30912, USA
4 CREST, JST (Japan Science and Technology), Kawaguchi, Saitama 332-0012, Japan


    Abstract
 Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Application of differential display to cultured rat astrocytes allowed cloning of Herp cDNA. Although Herp was strongly induced by endoplasmic reticulum (ER) stress, it decayed rapidly consequent to proteasome-mediated degradation. To investigate the role of this molecule in terms of the stress response, Herp knockout cells were developed using F9 embryonic carcinoma cells. F9 Herp null cells were more vulnerable to ER stress compared with F9 wild-type cells. In the early period of ER stress (0–8 h after tunicamycin treatment), Herp null cells displayed enhanced ER stress signalling and stabilization of an endogenous ERAD substrate, compared with wild-type cells. In the intermediate period (8–20 h after tunicamycin treatment), Herp null cells displayed reduced ER stress signalling, whereas in the late period (20–40 h after tunicamycin treatment), Herp null cells manifested irreversible cellular changes that lead to apoptotic cell death. Transfection analysis revealed that the N-terminal region, including the ubiquitin-like domain of Herp, was required for the survival of F9 cells under ER stress. These results indicate that Herp is a short-lived Ub-like protein improving the balance of folding capacity and protein loads in the ER and plays crucial roles for the ER stress resistance in F9 cells.


    Introduction
 Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Membrane and secretory proteins achieve correct folding and oligomerization in the endoplasmic reticulum (ER). Once cells are exposed to stresses such as glucose starvation, inhibition of protein modification, disturbance of Ca2+ homeostasis and oxygen deprivation, unfolded proteins accumulate in the ER (ER stress). Under such conditions, eukaryotic cells normally respond utilizing at least three different mechanisms: transcriptional induction, translational attenuation and degradation (ERAD; ER-associated degradation) (Mori 2000). The transcriptional induction results from the activation of an intracellular signalling pathway from the ER to the nucleus, known as the unfolded protein response (UPR). UPR is transmitted through the ER-resident membrane proteins Ire1{alpha}/ß (Tirasophon et al. 1998; Wang et al. 1998) and ATF6 (Yoshida et al. 1998), and UPR target genes include molecular chaperones, folding catalysts in the ER and ERAD molecules. If protein loads in the ER exceed its folding capacity or dome defects in the ER stress response exist, cells tend to die, typically, with apoptotic features (Harding et al. 2001).

We cloned a cDNA encoding a ubiquitin-like protein from cultured rat astrocytes exposed to hypoxia and submitted this to the DNA Data Bank of Japan under the name, Stress-associated Ubiquitin (Ub)-like Protein (SUP) (accession number: AB033771). However, there have already been several reports concerning this molecule with different names; (Nomura et al. 1994), methyl methanesulphonate (MMS)-inducible fragment1 (Mif1: van Laar et al. 2000, 2002) and Herp (Kokame et al. 2000, 2001; Sai et al. 2002). Because of the amount of work in the literature under the name Herp, we have employed this name for the current paper.

Herp is an ER-resident membrane protein which has a ubiquitin (Ub)-like domain at its N-terminus. Because of its membrane topology, both the N and C terminus of Herp face the cytosol. It seemed unlikely that Herp acts directly as a molecular chaperone for proteins in the ER lumen (Kokame et al. 2000). Instead, it has been postulated that Herp may function for ERAD (van Laar et al. 2002).

To investigate the role of this molecule in terms of the stress response, we targeted the Herp gene in F9 embryonic carcinoma cells. F9 Herp null cells were more vulnerable to ER stress compared with F9 wild-type cells. The ER stress-induced death in F9 Herp null cells was associated with the stabilization of an endogenous ERAD substrate, aberrant ER stress signalling, structural changes in the ER and caspase activation. The N-terminal region, including the ubiquitin-like domain of Herp, was required for the ER stress resistance in F9 cells.


    Results
 Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Identification of Herp

Cultured rat astrocytes were exposed to hypoxia for 20 h and a differentially expressed amplicon of 528 bp was identified. Northern analysis, using this cDNA as a probe, and total RNA harvested from hypoxic astrocytes, confirmed selective up-regulation, compared with normoxia and led us to clone the full-length cDNA. A rat brain cDNA library was screened and a 1.86-kb cDNA was cloned (GENBANK/EMBL/DDBJ accession number: AB033771). The cDNA encoded a protein of 391 amino acids, which was later found to be identical to (Nomura et al. 1994), Mif1 (van Laar et al. 2000) and Herp (Kokame et al. 2000).

Expression and stability of Herp under stress conditions

Consistent with previous reports, expression of Herp protein was up-regulated in response to ER stress in a variety of cell lines including HeLa cells (Fig. 1A), 293T cells (data not shown) and F9 cells (Fig. 2D), or primary astrocytes (data not shown). The specificity of the anti-Herp antibody was confirmed by pre-absorption of the antibody preparation with the peptide used as an immunogen; appearance of the band corresponding to the molecular weight (Mr) of Herp was prevented (data not shown). The existence of the Ub-like domain at the N-terminus of Herp led us to investigate the stability of Herp protein. After treating HeLa cells (Fig. 1B) or 293T cells (data not shown) with tunicamycin (2 µg/mL), pulse-chase analysis demonstrated more rapid degradation of Herp (half-life:T1/2 < 4 h) than ER molecular chaperones, such as GRP (glucose-regulated protein)-78 and GRP94 (T1/2 > 8 h). Addition of the proteasome inhibitor lactacystin slowed degradation of Herp (Fig. 1B, La). To extend the study of Herp expression in response to stress in vivo, the expression of Herp mRNA was assessed in the rat brain after MCA occlusion. Increased Herp transcripts were observed mainly in the peri-ischaemic penumbral region by in situ hybridization (Fig. 1CI), although neuronal cells displayed the highest levels of transcripts (Fig. 1CII).



View larger version (27K):
[in this window]
[in a new window]
 
Figure 1  Expression and stability of Herp under stress conditions. (A) Expression of endogenous Herp. HeLa cells were treated with tunicamycin (Tm; 2 µg/mL) or thapsigargin (Tg; 0.3 µM) for 8 h and protein extracts (30 µg) were subjected to Western blotting with anti-Herp or anti-ß-actin antibody. The asterisk indicates a non-specific band (see text). Migration of simultaneously run molecular weight standards is indicated on the far left in kDa. (B) Stability of endogenous Herp, GRP78 and GRP94. HeLa cells were pre-treated with tunicamycin (2 µg/mL) for 8 h and a pulse-chase study was performed in the presence of tunicamycin. In some cases, lactacystin (La; 5 µM) was added when pulse labelling began. Protein extracts were immunoprecipitated with anti-Herp or with KDEL antibody and separated by SDS–PAGE. (C) Expression of Herp transcripts was studied in brain slices after MCA occlusion (8 h) by in situ hybridization with riboprobes derived from Herp cDNA as described in the text.

 


View larger version (30K):
[in this window]
[in a new window]
 
Figure 2  Development of mouse Herp-knockout cells. (A) Structure of the mouse Herp gene (top), targeting vectors (middle) and expected disrupted alleles (bottom). Recombination with targeting constructs, including neomycin (neo), puromycin (puro) or Zeocin (Zeo) resistance gene was performed consecutively in this order. Solid arrows indicate PCR primers for screening recombinants. Thick bars represent probes for Southern blotting. (B) Southern blotting of Herp-targeted F9 cells. Kpn1-digested genomic DNA from F9 wild-type cells (F9 Wt) and cells after single (F9 S), double (F9 D) and triple (F9 T) recombination events were hybridized with a 5' probe (left) and 3' probe (right), respectively. Bands representing migration of the wild-type allele are indicated on the far right (wild) and migration of simultaneously run molecular weight standards is indicated on the far left in kb. (C, D) The expression of Herp mRNA (C) and protein (D) in F9 wild-type cells and F9 Herp null cells. Both cell types (Herp deficient and wild-type) were treated with inducers of ER stress, as described in the legend to Figure 1(A), and subjected to Northern blotting (C) or Western blotting (D).

 
Development of Herp-knockout cells in F9 embryonic carcinoma cells

The mouse Herp gene was disrupted by homologous recombination in F9 embryonic carcinoma cells (Fig. 2A). Because F9 cells have three alleles for the Herp gene for unknown reasons, we needed to perform three homologous recombinations to eliminate all wild-type alleles of Herp (Fig. 2B). The Herp-knockout F9 cells lost expression of Herp mRNA (Fig. 2C) and protein (Fig. 2D).

ER stress-induced cell death: effect of Herp gene deletion

To analyse the effect of Herp gene disruption on the cellular stress response, F9 wild-type and Herp null cells were treated with ER stress-inducers, tunicamycin (Tm) or thapsigargin (Tg) for 40 h. In related studies, other stress inducers, staurosporine (St) or H2O2, were incubated with cultures for 12 h. In each case, cell viability was measured by MTT assay (Fig. 3A). F9 Herp null cells showed significantly lower viability, compared with F9 wild-type cells, after tunicamycin treatment (Fig. 3AI). A similar trend was observed following thapsigargin treatment, although the effect was less striking (Fig. 3AII). Consistent with these observations, the cellular response to two other inducers of ER stress, calcium ionophore A23187 [GenBank] or 2-deoxyglucose, also depended on the presence of Herp genes (data not shown). In contrast, deletion of Herp alleles had no effect on cell viability following exposure to staurosporine (Fig. 3AIII) or H2O2 (Fig. 3AIV). To distinguish ER stress-induced growth arrest from ER stress-induced cell death, LIVE/DEAD cell toxicity assay was performed after tunicamycin (1 µg/mL; Fig. 3B) or thapsigargin (0.3 µM; data not shown) treatment for 36 h. Nulear staining of dead cells, typically with fragmented structures, was readily apparent in cultures of Herp null cells (Fig. 3BII), while the majority of F9 wild-type cells were intact (Fig. 3BI). Treatment of Herp null cells with cycloheximide (1 µg/mL), a general protein synthesis inhibitor that decreases protein loading in the ER, almost completely rescued these cells from ER stress-induced cell death (Fig. 2BIII). In our system, 1 µg/mL of cycloheximide blocks approximately 75% of protein synthesis (Hori et al. 1994). In contrast, treatment of Herp null cells with lactacystin (1 µM), a potent proteasome inhibitor, accelerated ER stress-induced cell death (Fig. 3BIV). No nuclear staining of dead cells was observed in F9 Herp null cells under normal conditions (data not shown).



View larger version (69K):
[in this window]
[in a new window]
 
Figure 3  The viability of F9 wild-type and Herp null cells in response to stress. (A) MTT assay. F9 wild-type (open bars) or Herp null cells (closed bars) were treated with tunicamycin for 40 h (I), thapsigargin for 40 h (II), staurosporine for 12 h (III) or H2O2 for 12 h (IV) at the indicated concentrations and subjected to MTT assay. Values shown are mean ± SD of three experiments. *P < 0.01, **P < 0.005. (B) LIVE/DEAD cell toxicity assay. F9 wild-type and Herp null cells were treated with 1 µg/mL tunicamycin for 36 h in the absence (I, II) or presence of 1 µg/mL cycloheximide (Cx; III) or 1 µM lacatcystin (La; IV) and LIVE/DEAD cell toxicity assay was performed. Dead cell labelling (nucleus; upper row) and live cell labelling (cytosol; lower row) are demonstrated. Results representative of three experiments are shown.

 
To evaluate ER stress-induced death in F9 Herp null cells in more detail, ultrastructural and biochemical studies were performed. Electron microscopy analysis revealed that disruption of the Herp gene did not affect the structure of the intracellular organelles (nucleus, Golgi, ER, mitochondria, cell membrane, etc.) under basal/non-stimulated conditions (Fig. 4A,E). After 24 h of tunicamycin treatment (1 µg/mL), F9 Herp-deficient cells displayed structural changes in the ER, including accumulation or aggregation of the ER in a juxta-nuclear region (Fig. 4B,F). While the detailed mechanism remains to be studied, protein aggregation in the ER or disruption of the cytoskeleton system could be involved in these changes. Two hundred cells were evaluated to obtain a ratio of F9 Herp null cells with these features in the ER to total Herp null cell population. After 24 and 36 h of tunicamycin (1 µg/mL) treatment, 30 and 96% of Herp null cells showed significant structural changes in the ER, respectively. In contrast, only 2.5 and 5% of F9 wild-type cells, respectively, had similar changes in the ER after tunicamycin treatment. Chromatin condensation and nuclear fragmentation were observed 36 h after treatment with tunicamycin (1 µg/mL; Fig. 4C) and membranous structures were completely destroyed at 40 h post-treatment (Fig. 4D) in F9 Herp null cells.



View larger version (141K):
[in this window]
[in a new window]
 
Figure 4  Electron microscopic analysis of F9 Herp null cells under tunicamycin treatment. Electron microscopic analysis was performed as described in the text. (E, F) Higher magnification views of the area indicated in (A, B), respectively. Scale bar: 2 µm.

 
As caspase activation, including that of ER-resident caspase 12, was reported to be involved in ER stress-induced cell death (Nakagawa et al. 2000), the activities of caspase 3, 2, 6 and 8 were measured with p-nitroanilide (pNA)-conjugated synthetic peptides. When F9 Herp null cells were treated with tunicamycin (Tm) or thapsigargin (Tg) for 24 h, the activities of caspase 3 and caspase 2 increased dramatically compared with wild-type F9 cells (Fig. 5AI,II). In contrast, the activity of caspase 6 increased to a lesser extent and the activity of caspase 8 remained at a low, basal level on exposure to tunicamycin (Fig. 5AIII,IV). Immunostaining of the cells with an antibody selective for activated caspase 3 confirmed the activation of caspase 3 in Herp null cells (data not shown). Western blotting was also performed to detect possible proteolytic cleavage of caspase 9 and caspase 12 (Fig. 5B). Herp null cells, but not F9 wild-type cells, treated with tunicamycin for 24 h, caused the cleavage of both molecules, although the amount of cleaved caspase 9 was much less than after staurosporine (St) treatment. Treatment of cells with ER stress inducers for less than 20 h did not activate any caspases, regardless of the presence of Herp gene (data not shown).



View larger version (27K):
[in this window]
[in a new window]
 
Figure 5  Caspase activation in F9 wild-type and Herp null cells subjected to ER stress. (A) Measurement of caspase activities using pNA-conjugated synthetic peptides. F9 wild-type (open bars) and Herp null cells (closed bars) were treated with tunicamycin (Tm) or thapsigargin (Tg) for 24 h, and caspase activities were measured. Values shown represent the mean ± SD of three experiments. (B) Measurement of caspase activities by Western blotting. F9 wild-type and Herp null cells were treated with tunicamycin (Tm) for 24 h or staurosporine (St) for 6 h, and, after cell lysis, they were subjected to Western blotting with anti-caspase 9 (I) and anti-caspase 12 (II) antibodies. Migration of simultaneously run molecular weight standards is indicated on the far left in kDa.

 
ER stress signalling and ERAD in F9 wild-type and Herp null cells

To assess the proximal events underlying ER stress-induced death in F9 Herp null cells, ER stress signalling was analysed in F9 wild-type and Herp null cells. When cultures were exposed to tunicamycin or thapsigargin for 6 h, GRP78 and CHOP transcripts, which are downstream targets of UPR and translation suppression, respectively, showed a somewhat greater degree of up-regulation in F9 Herp null cells than in F9 wild-type cells (Fig. 6A). The time-course studies in which F9 cells were treated with tunicamycin (2 µg/mL) revealed that, early in the incubation period (5–8 h), enhanced expressions of GRP78 and CHOP (Zinszner et al. 1998) mRNA were more readily detected in F9 Herp null cells compared with F9 wild-type cells. In contrast, later in the tunicamycin-treated period (12–21 h), the expressions of GRP78 and CHOP transcripts were evident in F9 wild-type cells compared with F9 Herp null cells (Fig. 6B). At the protein level, increased expressions (Fig. 6C) or activation (Fig. 6D) of the downstream targets of the ER stress signalling pathways, such as GRP 78, GRP 94, XBP1 (Yoshida et al. 2001; Calfon et al. 2002), CHOP and JNK (Urano et al. 2000), were detected at higher levels in F9 Herp null cells than F9 wild-type cells, following exposure to ER stress-inducers for 8 h.



View larger version (35K):
[in this window]
[in a new window]
 
Figure 6  ER stress signalling in F9 wild-type and F9 Herp null cells. (A, B) Expression of GRP78 and CHOP mRNA in F9 wild-type (lanes 1–6) and Herp null cells (lanes 7–12). Cells were treated with tunicamycin (Tm) or thapsigargin (Tg) for 6 h (A) or for the indicated times (B) and total RNA (10 µg) was used for Northern blotting. Quantification of relative intensities for CHOP mRNA (lower panels) is shown as the ratio of the relative intensity of CHOP band to that of ß-actin band. (C) Expression of GRP78, GRP94, CHOP and XBP1 proteins in F9 wild-type cells (lanes 1–5) and F9 Herp null cells (lanes 6–10). Cells were treated with the indicated reagents for 8 h and subjected to Western blotting with anti-KDEL, anti-XBP1, anti-CHOP or anti-ß-actin antibody. (D) JNK activity in F9 wild-type cells (lanes 1–6) and F9 Herp null cells (lanes 7–12). Cells were treated with tunicamycin (1–2 µg/mL), thapsigargin (0.15–0.3 µM) for 8 h or with H2O2 (200 µM) for 1 h as control and JNK activity was measured based on the phosphorylation of c-jun. The expression of JNK1 protein was also assessed by Western blotting with anti-JNK antibody. (E) Expression of connexin 43. Cells were metabolically labelled with 35S-methionine for 4 h and cell lysates were immunoprecipitated with anti-connexin 43 antibody. The asterisk indicates a non-specific band around 35 kDa. (F) Stability of connexin 43. Pulse-chase study was performed as described in the text and cell lysates were immunoprecipitated with anti-connexin 43 antibody. The asterisk indicates a non-specific band as in (E, G, H). Effect of terminating ER stress at the indicated times on the cell viability (G) and caspase 3 activation (H). ER stress was quenched at 8, 18 and 32 h after treating cells with tunicamycin (1 µg/mL for MTT assay and 2 µg/mL for caspase 3 activation). Where indicated, cultures were washed in drug-free medium and then maintained for the indicated times in control/unsupplemented medium. MTT assay and measurement of caspase 3 activation was performed as described above. Values shown are mean ± SD of three experiments.

 
To study the effect of Herp deletion on ERAD, the expression and stability of connexin 43, an endogenous substrate of ERAD (VanSlyke & Musil 2002), was assessed in F9 wild-type and Herp null cells. The expression of connexin 43 was observed in F9 wild-type cells and also, to a slightly greater extent, in Herp null cells (corresponding to Mr of 42 kDa) when cells were labelled with 35S-methonine for 3 h (Fig. 7A). The pulse-chase analysis revealed that connexin 43 was rapidly degraded in F9 wild-type cells (the approximately half-life is 3.5 h), while it was more stable in Herp null cells (Fig. 7B,C). Treatment of cells with lactacystin (1 µM) during the chase periods recovered the level of connexin 43 as previously described (data not shown, VanSlyke & Musil 2002).



View larger version (26K):
[in this window]
[in a new window]
 
Figure 7  Rescuing F9 Herp null cells by ectopic expression of Herp cDNA. (A, B) Expressions of Herp protein in F9 wild-type cells, F9 Herp null cells and F9 Herp null cells transfected with FL-Herp or dN-Herp cDNA. Cells were cultured in medium containing tunicamycin (Tm: 2 µg/mL) or medium alone (C) for 8 h and whole cell extracts were subjected to Western blotting with anti-Herp antibody, anti-FLAG antibody or ß-actin antibody. Migration of simultaneously run molecular weight standards is indicated on the far left in kDa in (B). (C, D) Effects of FL- and dN-Herp over-expression in the cell viability (C) and caspase 3 activation (D). Cells were treated with tunicamycin (1 µg/mL) for 40 h and MTT assay was performed as in Figure 3. Caspase 3 activity was also measured after treating cells with tunicamycin (2 µg/mL) for 24 h as in Figure 5.

 
To determine the critical time for the ER stress-induced death in F9 Herp null cells, tunicamycin treatment was terminated at 8, 18 or 32 h and downstream events were monitored. When tunicamycin-containing medium was washed away after an 8-h treatment period, downstream events such as cell death (MTT assay, Fig. 6E), caspase 3 activation (Fig. 6F) and morphological changes of the ER (data not shown) were all prevented. However, quenching ER stress at later times (18 and 32 h) did not successfully rescue the cells under tunicamycin-treated conditions, suggesting a critical effect of the Herp gene deletion during the 8–18 h interval of ER stress (Fig. 6E,F).

Ectopic expression of Herp cDNA rescues F9 Herp null cells from enhanced vulnerability to ER stress

To be certain that disruption of the Herp gene was responsible for the susceptibility of Herp null cells to ER stress, rescue experiments were performed by over-expressing Herp cDNAs. In order to accomplish this, full-length (FL-Herp) or N-terminus truncated (dN-Herp) rat Herp cDNA in Hpch(+) was constructed and transfected into F9 Herp null cells. Hpch(+) was a plasmid with Herp promoter and hygromycin resistance gene, and generated as described in Experimental procedures. Two clones stably expressing full-length Herp protein (FL-Herp 1, 2) and three clones stably expressing N-terminus truncated Herp protein (dN-Herp 1, 2, 3) were obtained after selecting cells with hygromycin. The levels of expression of Herp antigen in control and tunicamycin-treated conditions, assessed by Western blotting with anti-Herp or anti-FLAG antibody, were compared with those of F9 wild-type cells (Wt) (Fig. 7A,B). As anti-Herp antibody recognizes the Ub-like domain of Herp, dN-Herp protein was only detected by anti-FLAG antibody. Subcellular fractionation studies followed by Western blotting with anti-Herp or with anti-FLAG antibody revealed that both endogenous and ectopically expressed Herp were localized in the ER and, probably to a lesser extent, in Golgi apparatus (data not shown). When cells were exposed to tunicamycin for the indicated times, FL-Herp, but not dN-Herp, rescued F9 Herp null cells with respect to the levels of Herp antigen (Fig. 7A,B), and in terms of cell viability (Fig. 7C), and caspase 3 activation (Fig. 7D).


    Discussion
 Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Herp belongs to the group of type-2 Ub-like proteins (Tanaka et al. 1998) or Ub-domain proteins (UDPs: Jentsch & Pyrowolakis 2000), in which the Ub-like domain does not possess Gly–Gly residues at its C-terminus and would not undergo cleavage by ubiquitin hydrolases. Although there are no common features of type-2 Ub-like proteins/UDPs reported, their roles as proteasome adaptors were postulated (Jentsch & Pyrowolakis 2000). In the present study, we first demonstrated that Herp was strongly induced by ER stress in vitro and by brain ischaemia in vivo, but, unlike GRP78 or GRP94, decays rapidly by the proteasome pathway (Fig. 1), suggesting the possible linkage of Herp to ERAD.

Targeted disruption of the Herp gene caused F9 embryonic carcinoma cells to become vulnerable to ER stress. The response of F9 Herp null cells to tunicamycin treatment can be divided into three phases, in order to contrast them with F9 wild-type cells. The first or early period (0–8 h after tunicamycin treatment) included enhanced ER stress signalling and stabilization of an ERAD substrate, in F9 Herp null cells compared with F9 wild-type cells (Fig. 6A,C,D,E,F). If tunicamycin was removed during this period, F9 Herp null cells were rescued and downstream events were prevented (Fig. 6G,H). As we also found that cycloheximide treatment prevented the ER stress-induced death in F9 Herp null cells (Fig. 3BIII), while lactacystin treatment accelerated it (Fig. 3BIV), it is likely that unfolded proteins accumulate to a greater degree in F9 Herp null cells, than in wild-type cells, and F9 Herp null cells compensate by increasing folding capacities in the ER in this early period. The second or intermediate stage (8–20 h after tunicamycin treatment) was characterized by reduced ER stress signalling. Transcripts of both GRP78 and CHOP were reduced in F9 Herp null cells during this period (Fig. 6B). In this context, reduced expression of chaperones has been linked to ER stress-induced cell death (Katayama et al. 1999). Once F9 Herp null cells passed into the later or final period (20–40 h after tunicamycin treatment), apoptotic cell death occurred: caspase activation (Fig. 5), structural alteration of the ER (Fig. 4B,F) and chromatin condensation/nuclear fragmentation (Fig. 4C,D). These death-linked events were irreversible, as removal of tunicamycin at this later time did not rescue the cells.

Transfection of Herp cDNAs into F9 Herp null cells revealed that the N-terminal region, including the Ub-like domain of Herp, was required for the ER stress resistance in F9 cells. This also raises the possibility that Herp may function for ERAD and recruit the proteasome complex to the ER with its Ub-like domain, as recently hypothesized (van Laar et al. 2002).

However, Herp has another aspect as an ATF6-dependent gene. Recent reports have attempted to classify ER stress-related genes. One group of such stress-related genes consists of ER-resident molecular chaperones and folding enzymes directly regulated by ATF6 (Okada et al. 2002; Yoshida et al. 2003). EDEM, a gene primarily involved in ERAD (Hosokawa et al. 2001), has been reported to be induced by ER stress, but its up-regulation occurred after those of folding factors whose expressions were under control of ATF6 (Yoshida et al. 2003). Our preliminary results revealed that EDEM was induced within 8 h after tunicamycin treatment in F9 cells (determined by Northern blotting using EDEM cDNA: a gift from Dr N. Hosokawa, Kyoto University), while Herp and GRP78 were both induced within 4 h after tunicamycin treatment, suggesting that Herp may function for some folding processes and enhances general folding capacities in the ER. In this context, it was recently reported that Herp binds to full-length presenilins and enhances production of amyloid beta peptide (Sai et al. 2002). As over-expression of Herp did not alter the steady-state levels of either full-length presenilins or their N-terminal fragments, it is not likely that Herp enhances proteolysis of presenilins.

While further studies will be required to disclose the entire function of Herp, our results emphasize the important role of Herp in the ER stress response.


    Experimental procedures
 Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Cell culture and stress conditions

Astrocytes were isolated from the cerebral cortex of E18 Wistar rat embryos and cultured in MEM with 10% FCS as described before (Yamaguchi et al. 1999). 293T cells (a gift from Dr K. Imaizumi, Nara Institute of Science and Technology), HeLa cells and F9 embryonic carcinoma cells (a gift from Drs K. Ohishi and T. Kinoshita, Osaka University) were maintained in DMEM with 10% FCS. Hypoxic stress was created by incubating cells in a chamber (Coy Laboratory Products, Ann Arbor, MI) as previously described (Hori et al. 2002). Other stresses were induced by treating cells with tunicamycin (0.5–2 µg/mL; Sigma, St Louis, MO), thapsigargin (0.1–0.3 µM; Sigma), A23187 [GenBank] (1 µM; Sigma), 2-deoxyglucose (25 mM; Sigma), staurosporine (0.5–2 µM; Sigma), H2O2 (Nakalai Tesque, Kyoto, Japan), cycloheximide (1–2 µg/mL; Sigma) or lactacystin (1–5 µM; Calbiochem, La Jolla, CA) for the indicated times. In some experiments, stress was terminated by washing cells with drug(s)-free medium three times and cells were further incubated for the indicated times.

Differential display and cloning of rat Herp cDNA

Differential display was performed as described before (Yamaguchi et al. 1999). Differential expression of candidate genes in hypoxia vs. normoxia was confirmed by Northern blotting using 32P-radiolabelled cDNAs as probes. Several ER stress-inducible genes including GRP78, GRP94 and calreticulin were obtained, and one of the cDNA fragments from unknown genes was used to screen an adult cDNA library (lambda ZapII cDNA library, Stratagene, La Jolla, CA).

Northern blot analysis

Total RNA (10 µg), isolated from cultured astrocytes, HeLa cells and F9 cells was separated on agarose/formaldehyde (1%) gels and transferred onto nylon membranes. cDNA fragments for probes were generated as follows. Herp cDNA was obtained as described above. Rat GRP78 and ß-actin cDNAs were cloned by PCR with specific primers. CHOP cDNA was kindly provided by Dr D. Ron (New York University, USA). Each fragment was labelled with 32P-dCTP by the random hexamer procedure (Yamaguchi et al. 1999).

Plasmid construction, transfection and generation of antibodies

Rat Herp cDNA encoding the complete open reading frame was amplified by PCR using primers tagged with FLAG epitope at the N-terminus and cloned into Hpch(+). Hpch(+) vector was generated by replacing CMV promoter of pcDNA3.1(+) hygro (Invitrogen) with proximal 517 bp of mouse Herp promoter which includes both ERSE and ERSE II. Rat Herp cDNA lacking N-terminus (residues 1–97) was also developed by PCR using FLAG tagged primers at the N-terminus and cloned into Hpch(+). All constructs were sequenced prior to transfection studies. Transfection was performed into F9 cells by electroporation (25 µg DNA/107 cells). To obtain antibodies reactive with Herp, a peptide derived from its Ub-like domain with the sequence KSPNQRHRDLELSGDRG (residues 15–31) was synthesized and conjugated to keyhole limpet haemocyanin. Rabbits were immunized by conventional methods and, once high titre antibodies were obtained, the antisera were purified by protein G column (Invitrogen).

Cell lysis, subcellular fractionation and Western blotting

Cultured astrocytes, HeLa cells, F9 cells or 293T cells (5 x 106 cells) were lysed in buffer containing 10 mM Tris-HCl, 150 mM NaCl, 1 mM EDTA, 1% NP-40, 0.1% SDS, 0.2% deoxycholate, 1 mM PMSF, 1 µg/mL aprotinin, 1 µg/mL leupeptin and 1 µg/mL pepstatin. Western blotting was then performed using anti-Herp, anti-FLAG (Sigma), anti-KDEL (Stressgen Biotechnologies Corp., Victoria, BC, Canada), anti-CHOP (Santa Cruz Biotechnology, Santa Cruz, CA), anti-caspase 9 (Cell Signalling Technology, Beverly, MA), anti-caspase 12 or anti-XBP1 antibody (the latter two antibodies were gifts from Dr D. Ron, New York University). Sites of primary antibody binding were determined by alkaline phosphatase-conjugated secondary antibodies.

Metabolic labelling and pulse-chase analysis

HeLa cells, 293T cells (5 x 106 cells/condition) or F9 cells either with or without exposure to tunicamycin for the indicated times, were labelled with 35S-methionine (200 µCi/mL; Amersham Pharmacia Biotech, Piscataway, NJ) for 30 min in methionine-free DMEM with 10% dialysed FCS and chased for the indicated times (up to 16 h) in regular medium. In some experiments, cells were continuously labelled for 3 h with 35S-methionine. Cell extracts (see above) were immunoprecipitated with anti-Herp, anti-KDEL or anti-connexin 43 antibody (Santa Cruz Biotechnology) for 12 h at 4 °C and subjected to SDS–PAGE followed by autoradiography.

In situ hybridization

Unilateral middle cerebral artery (MCA) occlusion was performed in male Sprague-Dawley rats (250 g) as previously described (Yamaguchi et al. 1999). After 8-h ischaemia, the rats were killed and brains were frozen at –80 °C. Serial coronal sections were cut and the distribution of Herp mRNA was examined by in situ hybridization using previously described techniques (Yamaguchi et al. 1999). In brief, sense and anti-sense riboprobes for Herp were in vitro transcribed from the rat Herp cDNA inserted into the pGEM T vector. After linearizing the vector with NcoI (for the sense probe) or SpeI (for the anti-sense probe), reaction mixtures were incubated with 35S-UTP (NEG-039H, Dupont NEN; Wilmington, DE) and SP6 or T7 RNA polymerase (Promega, Madison, WI). Brain sections were then hybridized with either sense or anti-sense probes and washed, dried and subjected to autoradiography. Two days later, films were developed and brain images were examined. For some sections, slides were covered with photographic emulsion (Kodak, NY) for 2 weeks and were then developed and analysed by dark-field microscopy.

Establishment of Herp-knockout F9 embryonic carcinoma cells

Herp null F9 cells were prepared as previously described (Ohishi et al. 2000). Briefly, a targeting vector was constructed in pPNT (a gift from Dr Tybulewicz, MRC National Institute for Medical Research, London, UK) by replacing exon 1 of the mouse Herp gene derived from 129Sv/J library (Incyte Genomics, St Louis, MO) with PGK-neo, PGK-puro or PGK-zeo cassette (Fig. 3A). F9 cells were electroporated with targeting plasmids linearized with NotI or SwaI (25 µg DNA/107 cells) and selected with appropriate drugs. Concentrations of G418 (Sigma), puromycin (Sigma) and Zeocin (Invitrogen) were 400, 2 and 500 µg/mL, respectively. Recombinants were screened by PCR and confirmed by Southern blotting with 5'- and 3'-probes after digesting DNA with Kpn1 (Fig. 3B). For rescue experiments, F9 Herp null cells were electroporated with rat Herp cDNA/Hpch(+) and selected with hygromycin (Sigma) at 350 µg/mL.

Measurement of cell viability, caspase activities and JNK activity

After cells were treated with the indicated reagents, cell viability was measured by 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H tetrazolium bromide (MTT) assay (Chemicon, Temecula, CA) or with LIVE/DEAD cell toxicity kit (Molecular Probes, Eugene, OR). Caspase activities were measured by substrate cleavage (caspase 2, 3, 6 and 8-like activities; Chemicon), by Western blotting (caspase 9 and 12) or by immunostaining with activated-caspase 3 antibody (Promega). The substrates used were DEVD-p-nitroanilide (pNA; caspase 3), VDVAD-pNA (caspase 2), VEID-pNA (caspase 6) and IETD-pNA (caspase 8). JNK activity was measured using SAPK/JNK Assay kit (Cell Signalling Technology).

Electron microscopic analysis

Electron microscopic analysis was performed as described before (Miyoshi et al. 2000). Briefly, F9 wild-type cells or Herp null cells were treated with tunicamycin (1 µg/mL) for the indicated times and fixed in PBS containing 4% paraformaldehyde and 2% glutaraldehyde. They were post fixed with OsO4 at 25 °C for 30 min, dehydrated in graded ethanol solutions and embeded in Quetol 812 (Nisshin EM Co.). Sections (80 nm) were cut, stained with uranyl acetate (10% in 50% ethanol) and lead citrate and then examined with a Hitachi H-7100 electron microscope.

Laser densitometric analysis and statistical analysis

Laser densitometric analysis was performed to standardize the results of Western and Northern blotting with NIH Image software. Statistical analysis was performed with Student's t-test.


    Acknowledgements
 
We are grateful to Dr D. Ron, New York University, for valuable discussion and provision of anti-XBP1 antibody, anti-caspase 12 antibody and the mouse CHOP cDNA. We thank Dr K. Ohishi, Osaka University, for providing F9 embryonic carcinoma cells and suggestions for developing Herp null cells and Dr K. Imaizumi, Nara Institute of Science and Technology, for providing 293T cells. We also thank Dr K. Mori, Dr K. Nagata and Dr N. Hosokawa, Kyoto University, for providing EDEM cDNA and for valuable advice.


    Footnotes
 
Communicated by: Keiji Tanaka

* Correspondence: Email: osamuh{at}nanat.m.kanazawa-u.ac.jp


    References
 Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Calfon, M., Zeng, H., Urano, F., et al. (2002) IRE1 couples endoplasmic reticulum load to secretory capacity by processing the XBP-1 mRNA. Nature 415, 92–96.[CrossRef][Medline]

Harding, H.P., Zeng, H., Zhang, Y., et al. (2001) Diabetes mellitus and exocrine pancreatic dysfunction in perk–/– mice reveals a role for translational control in secretory cell survival. Mol. Cell 7, 1153–1163.[CrossRef][Medline]

Hori, O., Matsumoto, M., Maeda, Y., et al. (1994) Metabolic and biosynthetic alteration in cultured astrocytes exposed to hypoxia/reoxygenation. J. Neurochem. 62, 1489–1495.[Medline]

Hori, O., Ichinoda, F., Tamatani, T., et al. (2002) Transmission of cell stress from endoplasmic reticulum to mitochondria: enhanced expression of Lon protease. J. Cell Biol. 157, 1151–1160.[Abstract/Free Full Text]

Hosokawa, N., Wada, I., Hasegawa, K., et al. (2001) A novel ER {alpha}-mannosidase-like protein accelerates ER-associated degradation. EMBO Report 2, 415–422.[CrossRef][Medline]

Jentsch, S. & Pyrowolakis, G. (2000) Ubiquitin and its kin: how close are the family ties? Trends Cell Biol. 10, 335–342.[CrossRef][Medline]

Katayama, T., Imaizumi, K., Sato, N., et al. (1999) Presenilin-1 mutations downregulate the signalling pathway of the unfolded-protein response. Nat. Cell Biol. 1, 479–485.[CrossRef][Medline]

Kokame, K., Agarwala, K.L., Kato, H. & Miyata, T. (2000) Herp, a new ubiquitin-like membrane protein induced by endoplasmic reticulum stress. J. Biol. Chem. 275, 32846–32853.[Abstract/Free Full Text]

Kokame, K., Kato, K. & Miyata, T. (2001) Identification of ERSE-II, a new cis-acting element responsible for the ATF6-dependent mammalian unfolded protein response. J. Biol. Chem. 276, 9199–9205.[Abstract/Free Full Text]

van Laar, T., Schouten, T., Hoogervorst, E., van Eck, M., van der Eb, A.J. & Terleth, C. (2000) The novel MMS-inducible gene Mif1/KIAA0025 is a target of the unfolded protein response pathway. FEBS Lett. 469, 123–131.[CrossRef][Medline]

van Laar, T., van der Eb, A.J. & Terleth, C. (2002) A role for Rad23 proteins in 26S proteasome-dependent protein degradation? Mutat. Res. 499, 53–61.[Medline]

Miyoshi, K., Katayama, T., Imaizumi, K., et al. (2000) Characterization of mouse Ire1{alpha}: cloning, mRNA localization in the brain and functional analysis in a neuronal cell line. Mol. Brain Res. 85, 68–76.[Medline]

Mori, K. (2000) Tripartite management of unfolded proteins in the endoplasmic reticulum. Cell 101, 451–454.[CrossRef][Medline]

Nakagawa, T., Zhu, H., Morishima, N., et al. (2000) Caspase-12 mediates endoplasmic reticulum-specific apoptosis and cytotoxity by amyloid-ß. Nature 403, 98–103.

Nomura, N., Miyajima, N., Sazuka, T., et al. (1994) Prediction of the coding sequences of unidentified human genes. I. The coding sequence of 40 new genes (KIAA0001-KIAA0040) deduced by analysis of randomly sampled cDNA clones from human immature myeloid cell line KG-1. DNA Res. 1, 27–35.[Abstract/Free Full Text]

Ohishi, K., Inoue, N., Maeda, Y., Takeda, J., Riezman, H. & Kinoshita, T. (2000) Gaa1p and Gpi8p are components of a glycosylphosphatidylinositol (GPI) transamidase that mediates attachment of GPI to proteins. Mol. Biol. Cell 11, 1523–1533.[Abstract/Free Full Text]

Okada, T., Yoshida, H., Akazawa, R., Negishi, M. & Mori, K. (2002) Distinct roles of activating transcription factor 6 (ATF6) and double-stranded RNA-activated protein kinase-like endoplasmic reticulum kinase (PERK) in transcription during the mammalian unfolded protein response. Biochem. J. 366, 585–594.[CrossRef][Medline]

Sai, X., Kawamura, Y., Kokame, K., et al. (2002) Endoplasmic reticulum stress-inducible protein, Herp, enhanced presenilin-mediated generation of amyloid ß-protein. J. Biol. Chem. 277, 12915–12920.[Abstract/Free Full Text]

Tanaka, K., Suzuki, T. & Chiba, T. (1998) The ligation system for ubiquitin and ubiquitin-like proteins. Mol. Cells 8, 503–512.

Tirasophon, W., Welihinda, A.A. & Kaufman, R.J. (1998) A stress response pathway from the endoplasmic reticulum to the nucleus requires a novel bifunctional protein kinase/endoribonuclease (Ire1p) in mammalian cells. Genes Dev. 12, 1812–1824.[Abstract/Free Full Text]

Urano, F., Wang, X.Z., Bertolotti, A., et al. (2000) Coupling of stress in the ER to activation of JNK protein kinase IRE1. Science 287, 664–666.[Abstract/Free Full Text]

VanSlyke, J.K. & Musil, L.S. (2002) Dislocation and degradation from the ER are regulated by cytosolic stress. J. Cell Biol. 157, 381–394.[Abstract/Free Full Text]

Wang, X.Z., Harding, H.P., Zhang, Y., Jolicoeur, E.M., Kuroda, M. & Ron, D. (1998) Cloning of mammalian Ire1 reveals diversity in the ER stress responses. EMBO J. 17, 5708–5717.[CrossRef][Medline]

Yamaguchi, A., Hori, O., Stern, M.D., Hartmann, E., Ogawa, S. & Tohyama, M. (1999) Stress-associated endoplasmic reticulum protein 1 (SERP)/ribosome-associated membrane protein 4 (RAMP4) stabilizes membrane proteins during stress and facilitates subsequent glycosylation. J. Cell Biol. 147, 1195–1204.[Abstract/Free Full Text]

Yoshida, H., Haze, K., Yanagi, H., Yura, T. & Mori, K. (1998) Identification of the cis-acting endoplasmic reticulum stress response element responsible for transcriptional induction of mammalian glucose-regulated proteins. J. Biol. Chem. 273, 33741–33749.

Yoshida, H., Matsui, T., Yamamoto, A., Okada, T. & Mori, K. (2001) XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factor. Cell 107, 881–891.

Yoshida, H., Matsui, T., Hosokawa, N., Kaufman, R.J., Nagata, K. & Mori, K. (2003) A time-dependent phase shift in the mammalian unfolded protein response. Dev. Cell 4, 265–271.[CrossRef][Medline]

Zinszner, H., Kuroda, M., Wang, X.Z., et al. (1998) CHOP is implicated in programmed cell death in response to impaired function of the endoplasmic reticulum. Genes Dev. 12, 982–995.[Abstract/Free Full Text]

Received: 4 January 2004
Accepted: 11 February 2004




This article has been cited by other articles:


Home page
Hum Mol GenetHome page
G. Liang, J. Yang, Z. Wang, Q. Li, Y. Tang, and X.-Z. Chen
Polycystin-2 down-regulates cell proliferation via promoting PERK-dependent phosphorylation of eIF2{alpha}
Hum. Mol. Genet., October 15, 2008; 17(20): 3254 - 3262.
[Abstract] [Full Text] [PDF]


Home page
Am. J. Physiol. Cell Physiol.Home page
K. Takano, Y. Kitao, Y. Tabata, H. Miura, K. Sato, K. Takuma, K. Yamada, S. Hibino, T. Choshi, M. Iinuma, et al.
A dibenzoylmethane derivative protects dopaminergic neurons against both oxidative stress and endoplasmic reticulum stress
Am J Physiol Cell Physiol, December 1, 2007; 293(6): C1884 - C1894.
[Abstract] [Full Text] [PDF]


Home page
The OncologistHome page
F. Lefranc, V. Facchini, and R. Kiss
Proautophagic Drugs: A Novel Means to Combat Apoptosis-Resistant Cancers, with a Special Emphasis on Glioblastomas
Oncologist, December 1, 2007; 12(12): 1395 - 1403.
[Abstract] [Full Text] [PDF]


Home page
J. Virol.Home page
G. A. Versteeg, P. S. van de Nes, P. J. Bredenbeek, and W. J. M. Spaan
The Coronavirus Spike Protein Induces Endoplasmic Reticulum Stress and Upregulation of Intracellular Chemokine mRNA Concentrations
J. Virol., October 15, 2007; 81(20): 10981 - 10990.
[Abstract] [Full Text] [PDF]


Home page
J. Biol. Chem.Home page
M. Renna, M. G. Caporaso, S. Bonatti, R. J. Kaufman, and P. Remondelli
Regulation of ERGIC-53 Gene Transcription in Response to Endoplasmic Reticulum Stress
J. Biol. Chem., August 3, 2007; 282(31): 22499 - 22512.
[Abstract] [Full Text] [PDF]


Home page
Am. J. Physiol. Cell Physiol.Home page
Y. Tabata, K. Takano, T. Ito, M. Iinuma, T. Yoshimoto, H. Miura, Y. Kitao, S. Ogawa, and O. Hori
Vaticanol B, a resveratrol tetramer, regulates endoplasmic reticulum stress and inflammation
Am J Physiol Cell Physiol, July 1, 2007; 293(1): C411 - C418.
[Abstract] [Full Text] [PDF]


Home page
JCBHome page
S. Tuvia, D. Taglicht, O. Erez, I. Alroy, I. Alchanati, V. Bicoviski, M. Dori-Bachash, D. Ben-Avraham, and Y. Reiss
The ubiquitin E3 ligase POSH regulates calcium homeostasis through spatial control of Herp
J. Cell Biol., April 9, 2007; 177(1): 51 - 61.
[Abstract] [Full Text] [PDF]


Home page
Am. J. Physiol. Cell Physiol.Home page
K. Takano, Y. Tabata, Y. Kitao, R. Murakami, H. Suzuki, M. Yamada, M. Iinuma, Y. Yoneda, S. Ogawa, and O. Hori
Methoxyflavones protect cells against endoplasmic reticulum stress and neurotoxin
Am J Physiol Cell Physiol, January 1, 2007; 292(1): C353 - C361.
[Abstract] [Full Text] [PDF]


Home page
Mol. Cell. Biol.Home page
G. Liang, T. E. Audas, Y. Li, G. P. Cockram, J. D. Dean, A. C. Martyn, K. Kokame, and R. Lu
Luman/CREB3 Induces Transcription of the Endoplasmic Reticulum (ER) Stress Response Protein Herp through an ER Stress Response Element
Mol. Cell. Biol., November 1, 2006; 26(21): 7999 - 8010.
[Abstract] [Full Text] [PDF]


Home page
Mol. Biol. CellHome page
C. Wojcik, M. Rowicka, A. Kudlicki, D. Nowis, E. McConnell, M. Kujawa, and G. N. DeMartino
Valosin-containing Protein (p97) Is a Regulator of Endoplasmic Reticulum Stress and of the Degradation of N-End Rule and Ubiquitin-Fusion Degradation Pathway Substrates in Mammalian Cells
Mol. Biol. Cell, November 1, 2006; 17(11): 4606 - 4618.
[Abstract] [Full Text] [PDF]


Home page
Mol. Cell. Biol.Home page
O. Hori, M. Miyazaki, T. Tamatani, K. Ozawa, K. Takano, M. Okabe, M. Ikawa, E. Hartmann, P. Mai, D. M. Stern, et al.
Deletion of SERP1/RAMP4, a Component of the Endoplasmic Reticulum (ER) Translocation Sites, Leads to ER Stress.
Mol. Cell. Biol., June 1, 2006; 26(11): 4257 - 4267.
[Abstract] [Full Text] [PDF]


This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hori, O.
Right arrow Articles by Ogawa, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hori, O.
Right arrow Articles by Ogawa, S.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS