Skip Navigation


Glycobiology Advance Access originally published online on August 29, 2008
Glycobiology 2008 18(12):1028-1035; doi:10.1093/glycob/cwn082
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
18/12/1028    most recent
cwn082v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
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 Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Okuda, T.
Right arrow Articles by Nakayama, K.-i.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Okuda, T.
Right arrow Articles by Nakayama, K.-i.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2008. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org

Identification and characterization of the human Gb3/CD77 synthase gene promoter

Tetsuya Okuda and Ken-ichi Nakayama1

Glycolipids Function Analysis Team, Health Technology Research Center, National institute of Advanced Industrial Science and Technology (AIST), 2217-14 Hayashi, Takamatsu, Kagawa 761-0395, Japan

1 To whom correspondence should be addressed: Tel: +81-87-869-3593; Fax: +81-87-869-3593; e-mail: k-nakayama{at}aist.go.jp

Received on April 4, 2008; revised on August 6, 2008; accepted on August 25, 2008


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Material and methods
 Funding
 Conflict of interest statement
 References
 
Hemolytic uremic syndrome (HUS) is triggered by verotoxin (VT) produced by the Escherichia coli O157 strain. Several studies have demonstrated that VT induces endothelial cell (EC) death via the VT receptor globotriaosylceramide (Gb3/CD77) leading to this symptom. Inflammatory mediators which are produced as a result of E. coli O157 infection, increase the expression level of Gb3 in EC. Therefore increased expression of Gb3 is considered as a progression step for HUS. The increased expression of Gb3 is due to the transcriptional upregulation of Gb3/CD77 synthase gene (Gb3S, also known as {alpha}1,4-galactosyltransferase gene), the mechanism of which still remains unknown. To understand the transcriptional machinery and to elucidate the onset mechanism of HUS, we cloned and characterized the human Gb3S promoter. A modified 5'-RACE was used to determine the transcriptional initiation site, which revealed the presence of a TATA-less GC-rich sequence in the proximal region. Promoter activity measured using a luciferase assay demonstrated that the GC-rich sequence is necessary for the basal transcriptional activity, and two silencer elements located 5'-upstream of this GC-rich region regulated the transcriptional level. Furthermore, we found that the GC-rich sequence contained three potential Sp1 binding sites and that all three Sp1 binding elements acted as positive regulators. Since Sp1 is an inducer of several genes in the presence of the inflammatory cytokines in EC, our results suggest that the transcriptional regulation of the Gb3S gene by Sp1 might affect the VT sensitivity of EC and HUS progression.

Key words: Escherichia coli O157 / glycosphingolipid / glycosyltransferase / hemolytic uremic syndrome / promoter


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Material and methods
 Funding
 Conflict of interest statement
 References
 
Diversity of glycosphingolipids is mainly due to the presence of various structures of sugar chains. In the mammalian cells, a unique {alpha}1,4-galactose structure is found only in the globotriaosylceramide (Gb3/CD77) or its derivatives (Hakomori et al. 1971Go), but not in any other glycans. Synthesis of Gb3 is mediated by the Gb3/CD77 synthase gene (Keusch et al. 2000Go; Kojima et al. 2000Go; Steffensen et al. 2000Go), which codes for the enzyme lactosylceramide {alpha}1,4-galactosyltransferase (EC 2.4.1.228 [EC] ). In human, Gb3 is found in several tissues, especially abundantly in the kidney (Makita 1964Go). Gb3 is known as the Pk blood group antigen on erythrocyte (Marcus et al. 1981Go) and is also known as the CD77 antigen, which was previously characterized as Burkitt lymphoma-associated antigen or germinal center B-cell differentiation antigen (Wiels et al. 1981Go). Its role as the receptor for Verotoxin (VT, also called Shiga-like toxin), produced by Escherichia coli O157 strain, is also well characterized (Jacewicz et al. 1986Go; Lingwood et al. 1987Go). VT specifically binds to Gb3 via its B-subunit and induces host cell death. The effect of VT on endothelial cell (EC) leads to circulatory disturbances and tissue damages and induces the onset of hemolytic uremic syndrome (HUS) (Noris and Remuzzi 2005Go). HUS is associated with acute renal failure, thrombocytopenia, and hemolytic anemia. Consciousness disorder is also known as a HUS-related symptom and is a major risk factor leading to the death of a patient. Previously, we created a Gb3S null mutant mice and demonstrated that the VT-induced tissue damage completely depended on the expression of its receptor Gb3 in vivo and also that the main target of VT was the Gb3 on the endothelial cells (Okuda et al. 2006Go). Gb3 is heterogeneously expressed on the endothelial cell surface, and its expression is increased by inflammatory mediators, such as LPS and TNF-{alpha} (van de Kar et al. 1992Go; Obrig et al. 1993Go). This inflammatory mediator-induced increase in the Gb3 expression is due to the transcriptional upregulation of the Gb3S gene (Stricklett et al. 2005Go; Okuda et al. 2006Go), which has been considered as a reason for the progression of HUS. However, the molecular basis of this mechanism still remains unknown. In this report, we identified and characterized the human Gb3S promoter and revealed that the transcriptional factor Sp1 (Dynan and Tjian 1983Go) is a key factor that regulated the Gb3S expression. The 5'-flanking region of the Gb3S gene was located at chromosome 22q13, and this region included two silencer elements and a promoter element containing three Sp1 binding sites. Furthermore, we used site-directed mutagenesis and gel mobility shift assay to assess the role of Sp1 in Gb3S promoter activity. Our results showed that the Sp1 binding sites contributed to the cell-specific expression of the Gb3S gene, indicating that the Sp1 activity is important for the Gb3S transcription.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Material and methods
 Funding
 Conflict of interest statement
 References
 
Identification of the 5'-flanking region of human Gb3/CD77 synthase gene
We used the human cervix carcinoma HeLa cells as the host for our experiments because thin-layer chromatography (TLC) analysis of glycolipids revealed that Gb3 and its derivatives were abundant in these cells (Figure 1A). In contrast, acidic glycolipids were hardly detected (Figure 1B), suggesting that they are minor components in these cells. We next performed RNA ligase-mediated 5'-RACE to amplify the 5'-end of the Gb3S cDNA from the HeLa cell mRNA (Figure 2A). DNA sequence analysis of the amplified product identified the single transcriptional start site of the Gb3S mRNA (Figure 2B). Neighboring 40 bp from this transcriptional start site included the entire 5'-end sequence of the Gb3S full-length cDNA, which was cloned from various organs and was reported in the Genbank, suggesting that this region is a major transcriptional initiation site of the human Gb3S gene. According to the NCBI Genome Blast search result (GenBank accession number AL049757 [GenBank] ), the Gb3S gene is derived from three exons (Figure 2C) and is located at chromosome 22q13. The genomic organization of the human Gb3S was also previously published (Iwamura et al. 2003Go). We amplified the 5'-flanking region of the Gb3S gene from the HeLa cell DNA by PCR, and the entire DNA sequence of the 5'-flanking region is shown in Figure 3. The proximal region of the transcriptional initiation site includes four GC boxes but lacks typical basal elements such as TATA and CAAT boxes. In addition, there were four places where the DNA sequence of the HeLa clone differed from the existing human genome sequence found in NCBI. The clone from the HeLa cells contained an insertion of a G residue at position 95 upstream (position –95) from the transcription initiation site and a mutation of A–G was found at position –151. In addition, compared to the existing information in the human genome database, the DNA sequence of the HeLa clone lacked three T bases in a poly(T) containing stretch (position –1331 to –1356) and revealed two additional A bases in a poly(A) containing stretch (position –1394 to –1422). The mutation found at position –151 was previously reported as a SNP related to the P1/P2 phenotype of the P antigen system (Iwamura et al. 2003Go), suggesting that the genotype of HeLa cell is equivalent to the P2 phenotype found in Japanese individuals. Next, we searched for potential transcriptional factor binding sites in the 5'-flanking region using an algorithm (MATCHTM, http://www.gene-regulation.com) and a command: "cut-off selection for the matrix group to minimize false positives." Based on this analysis, we identified consensus binding sequences for MZF1, AML1a, E47, Oct-6, and Sp1 in the 5'-flanking region of the Gb3S gene as shown in Figure 3.


Figure 1
View larger version (71K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 TLC analysis of glycolipids from HeLa cells. (A) TLC of neutral glycolipids from HeLa cells was visualized by orcinol-H2SO4. Lane 1, standard neutral glycolipids and lane 2, glycolipids from HeLa cells. Lane 3 shows TLC immunostaining of lane 2 with the anti-Gb3 antibody (38.13). (B) TLC of acidic glycolipids from HeLa cells (lane 3) was visualized by resorcinol. The bovine brain ganglioside (lane 1) and GM3 (lane 2) were used as standard markers. These experiments were performed with a solvent system consisting of chloroform/methanol/water (60:35:8).

 

Figure 2
View larger version (36K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Identification of the transcriptional initiation site of the Gb3S gene. (A) RNA ligase-mediated 5'-RACE. Total RNA was prepared from HeLa cells and the 5'-capping structure was replaced with a RNA oligonucleotide by using a series of enzymatic reactions, which was subsequently used for dual PCR amplification with Gb3S gene-specific primers. First PCR was performed using the GRP-1 and GSP-1 primers and second PCR was performed using the GRP-2 and GSP-2 primers. The amplified product was electrophoresed on 1.5% agarose gel and visualized by EtBr staining: lane 1, 100 bp standard DNA marker and lane 2, the second PCR product. (B) The nucleotide sequence of the second PCR product. Capital fonts and bold fonts represent the sequence of the second PCR product and the RNA oligonucleotide sequence, respectively. Arrows indicate the sequences of the corresponding primers, arrowheads indicate the splice junctions, and a solid black box indicates the ATG start codon. (C) Genomic organization of the Gb3S gene. Boxes represent exons and the solid bar indicates the coding region. Respective positions of the GSP-1 and GSP-2 primers used in this experiment are shown using long arrows.

 

Figure 3
View larger version (74K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Nucleotide sequence and structural organization of the Gb3S gene promoter region. Putative binding sites for the transcriptional factors are underlined. The transcriptional initiation site is shown in italic. Solid black boxes, GC box; the unfilled box, exon-1.

 
Characterization of regulatory elements of Gb3S gene promoter
To determine the apparent promoter activity of the 5'-flanking region of Gb3S, the entire fragment (–1893 bp to +84 bp) was cloned into the luciferase reporter vector pML, the resultant plasmid (pML–1893/+84) was transfected into the HeLa cells, and a luciferase assay was performed. As shown in Figure 4, the luciferase activity in HeLa cells transfected with pML–1893/+84 plasmid was much higher than in the cells transfected with the control pML plasmid (empty vector), suggesting that this 5'-flanking region indeed contained promoter activity. To determine the role of the putative regulatory elements that were identified in this 5'-flanking fragment, we generated several 5'- and 3'-deletion constructs and analyzed their promoter activities. First, we noted that the promoter activities associated with the deletion plasmids pML–845/+84 and pML–133/+84 were two times higher than the plasmid constructs containing longer upstream regions, suggesting that the –1207 bp to –846 bp and –222 bp to –134 bp regions contain silencer elements. Among all the constructs tested, the pML–133/+84 construct showed the highest promoter activity. Deletion of the region from –133 bp to –60 bp ablated the promoter activity of the pML–133/+84, whereas deletion of the region from –60 bp to +84 bp retained, albeit to a lesser extent, the promoter activity of the pML–133/+84. These results suggest that the region between –133 bp and –60 bp contains one or more elements positively regulating the Gb3S transcription.


Figure 4
View larger version (8K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Determination of promoter elements in the 5'-flanking region of the Gb3S gene. DNA fragment containing the entire –1893 bp to +84 bp (–1893/+84) region of the Gb3S gene and deletion fragments containing various portions of the 5-flanking region (schematically shown on the left-hand side of the figure) were cloned into the pMetLuc luciferase reporter vector (pML) and each of the resultant plasmid was transiently transfected into the HeLa cells. The pSEAP-control was used for cotransfection in each case as internal control for transfection. Luciferase and alkaline phosphatase activities in each transfected cells were measured as described in Material and methods. For each transfection, the luciferase activity is first normalized with respect to the alkaline phosphatase activity (internal control) and then the relative activity was determined from the ratio of the normalized activity and the activity in cells transfected with the empty pML vector. Error bars: mean ± SE, n = 4, *P < 0.001.

 
Sp1 is a positive transregulator of Gb3S promoter
As shown in Figure 3, there are three putative consensus Sp1 binding sites in the –133 bp to –60 bp region. To examine whether Sp1 acts on this region as a transactivator, we altered all three putative Sp1 binding sites (designated as Sp1a, Sp1b, and Sp1c in Figure 5A), one at a time, by mutagenesis and examined the effect of each mutation on the promoter activity. Thus, the following point mutations were introduced to individually alter each Sp1 consensus sequence in the pML–133/+84: Sp1a, C-109A; Sp1b, C-104A; and Sp1c, G-86T (Figure 5A). The promoter activities of the Sp1a C-109A, Sp1b C-104A, and Sp1c G-86T mutants were approximately 75%, 75%, and 65%, respectively, of that of the wild type (Figure 5B). We further confirmed the role of these consensus Sp1 binding sites on the promoter activity by deletion analysis. Thus, deletion of the Sp1c site (construct {Delta}–90/–81) reduced the promoter activity in similar manner as the Sp1c G-86T site-specific mutant, whereas deletion of both Sp1a and Sp1b sites (construct –99/+84) very significantly reduced the promoter activity (Figure 5B). Together, these results suggest that all three Sp1 binding sites affect the Gb3S promoter activity. In order to demonstrate that the Sp1 protein regulates the Gb3S promoter activity, we examined the effects of several Sp1 inhibitors on the promoter activity. Mithramycin A is known to act on the DNA sequence of the consensus Sp1 binding site and inhibits the promoter activity (Ray et al. 1989Go; Blume et al. 1991Go). On the other hand, streptozotocin (STZ) and 2-deoxy-D-glucose (2-DG) are known to inhibit the transcriptional activity of Sp1 by increasing the O-linkage of N-acetylglucosamine to Sp1 activation domain (Yang et al. 2001Go; Kang et al. 2003Go). As expected, mithramycin A, STZ, and 2-DG inhibited the Gb3S promoter activity in a dose-dependent manner (Figure 6). Next, we confirmed direct interaction between the Sp1 and three consensus Sp1 binding sites using the gel mobility shift assay. For this purpose, the HeLa cell nuclear extract was incubated with either the biotinylated oligonucleotide probe-1, which included both the Sp1a and Sp1b binding site sequences, or the biotinylated oligonucleotide probe-2, which included the Sp1c binding site sequence, and subsequently each mixture was analyzed by polyacrylamide gel electrophoresis (PAGE) (Figure 7). Migration of the oligonucleotide probe was retarded (Figure 7, lanes 2 and 5, both panels indicated by arrowheads) as compared to the samples where the assay mixture was incubated with either excess of unlabeled Sp1 consensus oligonucleotide (Figure 7, lane 3, both panels) or with anti-Sp1 antibody (Figure 7, lane 6, both panels). The retarded band was observed even when the nuclear extract was preabsorbed with a mutant oligonucleotide probe carrying point mutation in the Sp1 consensus sequence to eliminate any nonspecific binding (Figure 7, lane 4, both panels). The intensity of the retarded band was reduced by pretreatment of the mutant probes. This result should be due to the remaining Sp1 binding capability of mutant probes because Sp1 consensus sequences were slightly retained and excessively added to absorb the nonspecific background completely. These results indicate that Sp1 binds specifically to these consensus binding sites on the Gb3S promoter and positively regulates the promoter activity.


Figure 5
View larger version (25K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 Role of putative Sp1 binding sites on the promoter activity. Panel A, top: positions of three putative Sp1 consensus binding sites (named here as Sp1a, Sp1b, and Sp1c) in the –133/+84 construct (top of the panel) and bottom: point mutations (C-109A, C-104A, and G-86T; marked using solid black box) created by site-directed mutagenesis to individually alter each one of these three Sp1 binding sites. Panel B, luciferase activity expressed by the three Sp1 site-directed mutants and three deletion mutants ({Delta}–90/–81, –99/+84, –59/+84) lacking partly or completely the three Sp1 binding sites. The deletion mutant {Delta}–90/–81 lacks the bases –90 to –81 in the corresponding Sp1c site, deletion mutant –99/+84 lacks both Sp1a and Sp1b sites, and deletion mutant –59/+84 lacks all three Sp1 sites. Relative luciferase activity expressed by each mutant is expressed as a percentage of the activity expressed by the –133/+84 construct. Error bars: mean ± SE, n = 4.

 

Figure 6
View larger version (11K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 Effect of various Sp1 inhibitors on the promoter activity. pML–133/+84 plasmid was transiently transfected into the HeLa cells and the transfected cells were then treated with indicated concentrations of a given Sp1 inhibitor (mithramycin A, STZ or 2-DG) for 16 h. Following the treatment with the inhibitor, luciferase activity in the culture medium was measured as described in Material and methods. Results shown represent the relative luciferase activity (as percentage of the untreated control). Error bars: mean ± SE, n = 2.

 

Figure 7
View larger version (65K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7 Binding of Sp1 to its consensus binding sites in the Gb3S promoter. HeLa cell nuclear extracts were incubated with a biotinylated oligonucleotide probe (probe-1) encompassing both Sp1a and Sp1b sites (position –117 bp to –96 bp, left panel) or a biotinylated oligonucleotide probe (probe-2) encompassing the Sp1c site (position –95 bp to –74 bp, right panel), and the mixture was subsequently analyzed by a gel shift assay (lanes 2–6) as described in Material and methods. To absorb the nonspecific background, the nuclear extract was preincubated with the 1.75 pmol of unlabeled probes carrying point mutations in the Sp1 consensus sequences (Mut-probe), and then used for the gel shift assay as described above (lane 4). Competition assay experiments using a nonlabeled Sp1-consensus oligonucleotide (Competitor, lane 3) or an anti-Sp1 antibody (lane 6) were performed as described in Material and methods. After electroblotting these samples on nylon membranes, the biotinylated probes were detected with the avidin–biotin–HRP complex.

 
Correlation between the activation of the Gb3S promoter and Gb3 expression in cells
The properties of the identified Gb3S promoter were further characterized in Gb3-positive (HeLa, NCCIT) and Gb3-negative (SK-MEL-28) cells. Reverse transcription polymerase chain reaction (RT-PCR) analysis showed that the Gb3S transcripts were barely expressed in the SK-MEL-28 cell (Figure 8B, upper panel), which is consistent with the absence of Gb3 in these cells, as its synthesis is controlled by the Gb3S gene product. The construct pML–1893/+84 showed the high level of promoter activity in both the Gb3-positive HeLa and NCCIT cells but quite low in SK-MEL-28 cells, which was reduced to the background level (Figure 8A). Even the construct pML–133/+84 that lacked the silencer domains showed weak promoter activity in Gb3-negative cells (Figure 8A). These results suggest that the –133 bp to –60 bp region is essential for the Gb3S promoter activity. Thus, we expected that the expression of Gb3 would be determined by the expression level of Sp1 in the cells. Although the expression level of the Sp1 mRNA was similar in all three cells (Figure 8B, middle panel), the relative expression of the Sp1 protein in the nuclear extracts of the Gb3-positive cells was to some extent higher than that of the Gb3-negative cells. A transcriptional regulator KLF4, which is known as a synergistic coactivator of Sp1 (Brembeck and Rustgi 2000Go; Higaki et al. 2002Go), was also found to be expressed at a higher level in the Gb3-positive cells than in the Gb3-negative cells. Taken together, these results suggest that the expression of Gb3 in the cells depends on the promoter activity of the –133 bp to –60 bp region of the Gb3S gene and its expression is regulated by Sp1 and some other factors.


Figure 8
View larger version (28K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 8 The promoter activity and mRNA levels of the Gb3S gene in Gb3-positive and Gb3-negative cells. (A) Gb3-positive (HeLa, open bar; NCCIT, gray bar) and Gb3-negative (SK-MEL-28, closed bar) cells were cotransfected with pSEAP-control and one of the following plasmids: pML–1893/+84, pML–133/+84, or pML–59/+84. Luciferase activity in the culture medium of the transfected cells was measured as described before. The relative luciferase activities (RLA) for a given cell line transfected with the indicated plasmids are shown after normalization with respect to the secreted alkaline phosphatase activity (internal transfection control). (B) RT-PCR analysis of the Gb3S (upper panel) and Sp1 (middle panel) mRNA expression in HeLa (lane 1), NCCIT (lane 2), and SK-MEL-28 (lane 3) cells. Expression of GAPDH mRNA in these cells was monitored as a control (lower panel). (C) Western blot analysis of the nuclear proteins in HeLa (lane 1), NCCIT (lane 2), and SK-MEL-28 (lane 3) cells. Nuclear extracts prepared from HeLa, NCCIT, and SK-MEL-28 cells (3 µg of protein) were separated by SDS–PAGE and subsequently analyzed by Western blot using the anti-Sp1 antibody (upper panel) or the anti-KLF4 antibody (lower panel) as described in Material and methods.

 

   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Material and methods
 Funding
 Conflict of interest statement
 References
 
This study described the structure of the Gb3S promoter and offered information on mechanisms regulating this promoter activity. Here, we identified a single transcription initiation site of the Gb3S gene from the HeLa cell mRNA. The neighboring bases from this transcriptional start site included the entire 5'-end sequence of the Gb3S full-length cDNAs, cloned from various organs and reported in the Genbank, suggesting that this region is a major transcriptional initiation site of the human Gb3S gene. Our analysis of the 5'-flanking region of the Gb3S gene led to the localization of the promoter element and also suggested that this promoter acts as the principal promoter for the Gb3S gene in human tissues.

The Gb3S promoter lacks the TATA and CCAAT boxes that were commonly found in the mammalian glycolipid glycosyltransferase genes (Ichikawa et al. 1998Go; Kim et al. 2002Go; Zeng et al. 2003Go; Sato and Furukawa2004Go). Instead, we found a highly GC-rich sequence in the proximal promoter, which is controlled by the transcriptional factor Sp1. Similar promoter structure was found in the glucosylceramide synthase gene (Ichikawa et al. 1998Go) and also in the β-1,4-galactosyltransferase V gene (Sato and Furukawa 2004Go), which is known to synthesize highly branched N-glycans (Sato and Furukawa 2004Go). Importance of the β-1,4-galactosyltransferase gene in lactosylceramide synthesis has also been reported (Kolmakova and Chatterjee 2005Go). These genes, which are involved in the Gb3 synthesis pathway, have similar promoter structure and are regulated by Sp1. Thus, the transcription factor Sp1 plays an important role in the Gb3 synthesis.

We found a correlation between the Gb3S promoter activity and Gb3 expression in our test cells. Since the Sp1 protein seems to be mainly involved in regulating this promoter activity, we believe that activation or silencing of the Gb3S promoter by Sp1 and its cofactors may play a part in the cell-specific expression of Gb3S. In the present study, we found silencer elements in two places of the 5'-flanking region of the Gb3S promoter (Figure 4). Since these silencer elements are unlikely to influence the cell-specific expression of Gb3S, they may constantly control the promoter activity at low levels in any cells. A SNP related to P2 phenotype of the blood group P antigen system was also found in the 5'-flanking region of Gb3S. This SNP correlated with the decrease of transcriptional levels of Gb3S (Iwamura et al. 2003Go); however, we were unable to find any function of this SNP in the Gb3S promoter activity (Figure 4). The strength of basic transcription activity of the Gb3S promoter through the Sp1 or Sp1 consensus sequence seemed to be important for the cell-specific expression of Gb3S (results shown in Figure 8A). Although the relative expression level of the Sp1 mRNA was similar in both Gb3-positive and Gb3-negative cells, the relative expression level of the Sp1 protein was higher in the Gb3-positive cells than in the Gb3-negative cells. It was previously reported that posttranslational modifications of Sp1, such as glycosylation (Zhang et al. 2003Go) or sumoylation (Spengler and Brattain 2006Go), regulate degradation of this protein in cells. These modifications may contribute to the cell-specific activation of Gb3S promoter by Sp1. On the other hand, certain transcriptional factors that are known to interact with Sp1 might also play a role in regulating the cell-specific activation of the target gene promoter. In this regard, it is noteworthy that the transcription factor KLF4, which was abundantly expressed in Gb3-positive cells (Figure 8C), was previously shown to modulate the Sp1 activity by interacting with it. For example, KLF4 was shown to act as a synergistic coactivator of Sp1 on the keratin 19 gene (Brembeck and Rustgi 2000Go) or laminin gamma1 chain gene promoter (Higaki et al. 2002Go), but was also shown to inhibit the activity of other promoters by direct interaction with the Sp1 protein (Zhang et al. 1998Go) or via the Sp1 binding domain on other gene promoter (Shie et al. 2000Go). In the case of the Gb3S promoter, KLF4 might act as a coactivator.

In EC, Sp1 is known to play a role as a gene inducer under inflammatory condition. Hamanaka et al. (1992)Go showed that the expression level of the LDL receptor gene in EC was enhanced by TNF-{alpha} even though there was no NF-{kappa}B motif on the promoter of this gene and that the induction of the LDL receptor gene by TNF-{alpha} was mediated via Sp1. Similar results were also obtained with the VEGF receptor 2 gene (Giraudo et al. 1998Go). These results suggest that under inflammatory condition Sp1 is a key factor regulating gene expression in EC. However, the mechanism for the regulation of the Gb3S promoter in EC still remains unknown. Further progress in this field would be required to elucidate how Gb3S regulates the level of Gb3 in EC and to determine how to develop prophylactic treatment for HUS based on the outcome.


   Material and methods
Top
 Abstract
 Introduction
 Results
 Discussion
 Material and methods
Funding
 Conflict of interest statement
 References
 
Cells culture
HeLa cell and SK-MEL-28 cell were obtained from RIKEN Cell Bank (Tsukuba, Japan) and maintained in Dulbecco's modified Eagle's minimal essential medium (DMEM) supplemented with 10% fetal bovine serum. NCCIT human teratocarcinoma cells were purchased from American Type Culture Collection (Manassas, VA), which were cultured in the RPMI-1640 medium supplemented with 10% fetal bovine serum and 2 mM L-glutamine. All cells were cultured at 37°C in a humidified atmosphere containing 5% CO2.

Glycolipid extraction, TLC, and TLC immunostaining
Glycolipid extraction and TLC immunostaining were performed as described (Furukawa et al. 1985Go). Briefly, total lipids from 1 x 108 cells were sequentially extracted with the 2:1, 1:1, and 1:2 mixtures (v/v), respectively, of chloroform and methanol. Contaminating phospholipids were hydrolyzed with mild alkaline treatment, and the glycolipids were subsequently purified by using Sep-Pack C18 cartridge (Waters, Milford, MA). Acidic and neutral glycolipids were separated by column chromatography using DEAE Sephadex A-25 (SIGMA, St. Louis, MO) and Iatrobeads 6RS-8060 (Mitsubishi Kagaku Iatron, Tokyo, Japan), respectively. Purified glycolipids were analyzed on HPTLC plates (Merck, Darmstadt, Germany) with a solvent system consisting of chloroform/methanol/water (60:35:8, v/v/v). Glycolipids were visualized by resorcinol (for acidic glycolipids) or orcinol-H2SO4 (for neutral glycolipids). TLC immnostaining was performed using an anti-Gb3 monoclonal antibody 38.13 (Immunotech, Marseille, France), and antibody binding was detected using an ABC kitTM (Vector Laboratories, Burlingame, CA) and SuperSignalTM West Dura Extended Duration Substrate (Pierce, Rockford, IL).

Identification of the 5'-end of the Gb3S mRNA
The 5'-end of Gb3S mRNA was determined by RNA ligase-mediated 5'-RACE. This experiment was carried out using a GeneRacerTM Kit (Invitrogen, Carlsbad, CA) following the manufacturer's instructions. Briefly, the 5'-cap structure of HeLa cell mRNAs was replaced with a RNA oligonucleotide (5'-CGA CUG GAG CAC GAG GAC ACU GAC AUG GAC UGA AGG AGU AGA AA-3') by using a series of enzyme reactions. After reverse transcribing with oligo (dT) primer, the Gb3S cDNA was amplified by dual PCR using Gb3S gene-specific primers. First round of PCR was carried out using the GeneRacerTM 5' Primer (GRP-1) and GSP-1 (5'-TGG CGG GCC CCT CAC AAG TAC A-3') reverse primer. Second round of PCR was performed using the GeneRacerTM 5' Nested Primer and GSP-2 (5'-GAA GCC GAT GAT GAA CAG GGT GCA GA-3') primer. Corresponding positions of the primers on the Gb3S gene are depicted in Figure 2.

Construction of plasmids
The 5'-flanking region of the Gb3S gene was isolated by PCR using the HeLa cell genomic DNA as a template and the following amplification primers: 5'-TGA GTC GAC TCA GCT CTT GGA GGG GCA ACA-3' and 5'-GCG CGC ACA AAT GTC GCC TCC AGA ACA-3'. To measure promoter activity of the 5'-flanking region of the Gb3S gene, the amplified DNA was digested with the restriction enzymes SalI and BamHI, recognition sites for which were internally located in the fragment, and the digested DNA was subcloned into the pMet-Luciferase (pML) reporter vector (Clontech, Mountain View, CA). This DNA insert consisted of –1893 bp to +84 bp region of the 5'-flanking region of the Gb3S gene (shown in Figure 3); we therefore designated the plasmid as pML–1893/+84. Plasmid constructs containing the deletion fragments –1893/–456 and –462/+84 were created by digesting the pML–1893/+84 plasmid with HindIII/EcoRI and EcoRI/BamHI, respectively, and subcloning the respective DNA fragments into the pML reporter vector. Other deletion fragments were generated by PCR using the pML–1893/+84 plasmid DNA as a template. A common 3'-primer, 5'-GCT AAG CTT AGC ACG CGT AAG AGC TCG GTA-3', and the following 5'-primers were used for the generating the following deletion fragments: for the –1207/+84 fragment, 5'-GAG TTT CAC CAT GTT GGC CA-3'; for the –845/+84 fragment, 5'-GGG CAC CCA TTG AGT GCC A; for the –657/+84 fragment, 5'-GAC ATG CTG GAA ACA TGG CA-3'; for the –222/+84 fragment, 5'-GGA CTG GGG ACT GTC CGC A; for the –133/+84 fragment, 5'-ACC CGC AGG GTA GGT CGG GA-3'; for the –99/+84 fragment, 5'-CGG CGC TGA CCC CGC CCC GGG CCG GA-3'; and for the –59/+84 fragment; 5'-TCC CGC CGG GCC CCA GGC ACT-3'. Since these PCR products contained the full-length sequence of the pML plasmid, the 5'- and 3'-ends were ligated using T4 DNA ligase to obtain the following deletion plasmids: pML–1207/+84, pML–845/+84, pML–657/+84, pML–222/+84, pML–133/+84, pML–99/+84, and pML–59/+84. Deletion plasmids pML–133/–60 and pML{Delta}–90/–81 were also prepared following the similar protocol except for using the pML–133/+84 plasmid DNA as a template and the following combination of primers: for the –133/–60 fragment, 5'-GGA TCC ACC GGT CGC CAC CAT-3' and 5'-GGC AGC ACC GCC CCT-3' and for the {Delta}–90/–81 fragment, 5'-GGC CGG AGG GGC GGT GCT GCC T-3' and 5'-TCA GCG CCGCCC CGC CCC GCC CGT CCC GAC CTA-3'. Site-directed mutagenesis of the putative Sp1 sites in pML–133/+84 (shown in Figure 5) was performed by PCR using the following primer sets: for C-109A mutagenesis, 5'-AGG TCG GGA CGG GAG GGG CGG GGC GGC GCT-3' and 5'-AGC GCC GCC CCG CCC CTC CCG TCC CGA CCT-3'; for C-104A mutagenesis, 5'-GTC GGG ACG GGC GGG GAG GGG CGG CGC TGA-3' and 5'-TCA GCG CCG CCC CTC CCC GCC CGT CCC GAC-3'; and for G-86T mutagenesis, 5'-GCT GAC CCC TCC CCG GGC CGG A-3' and 5'-TCC GGC CCG GGG AGG GGT CAG C-3'.

All PCR experiments were performed using the PrimeSTAR HS DNA polymerase (Takara Bio, Shiga, Japan). The nucleotide sequence of the DNA insert in each plasmid construct was verified by sequencing.

Transfection and luciferase assay
0.4 µg of each plasmid was transfected into 2 x 105 cells (HeLa, NCCIT and SK-MEL-28) with lipofectamine 2000 (Invitrogen) according to manufacturer's instructions and then incubated for 16 h (HeLa) or 48 h (NCCIT and SK-MEL-28). To provide an internal control, we used 0.2 µg of the pSEAP2-control vector (Clontech), which contained the secreted form of the human placental alkaline phosphatase (SEAP) under the SV40 promoter, for cotransfection. Since the reporter proteins (luciferase and SEAP) were secreted in the culture media, 50 µL of the culture medium from each transfected cells was used for measuring SEAP and luciferase activities. The luciferase activity was measured using the Ready-To-GlowTM Secreted Luciferase Reporter Assay kit (Clontech) and Luminescencer JNRII (ATTO, Tokyo, Japan). The SEAP activity was measured using the Great EscAPeTM SEAP Chemiluminescence Detection kit (Clontech). The Sp1 inhibitors, mithramycin A, 2-deoxy-D-glucose, and streptozotocin (SIGMA), were diluted with a culture medium and added to the cells after transfection, and the cells were incubated for 16 h.

Gel shift assay
The biotin-labeled double-stranded oligonucleotide probes (oligonucleotides for probe-1: 5'-GGG ACG GGC GGG GCG GGG CGG C-3' and 5'-GCC GCC CCG CCC CGC CCG TCC C-3'; oligonucleotides for probe-2: 5'-GCT GAC CCC GCC CCG GGC CGG A-3' and 5'-TCC GGC CCG GGG CGG GGT CAG C-3') were prepared by an enzyme reaction using Terminal Deoxynucleotidyl Transferase (Promega, Madison, WI) and Biotin-14-dCTP (Invitorogen). The HeLa Nuclear Extract (HeLaScribeTM, Promega) was incubated with the 75 fmol of probe for 20 min at room temperature in the presence of a gel shift binding buffer (4% glycerol, 1 mM MgCl2, 0.5 mM EDTA, 0.5 mM DTT, 50 mM NaCl, 0.05 mg/mL poly(dI-dC)poly(dI-dC), 10 mM Tris–HCl, pH 7.5). Reaction mixtures were separated on native PAGE prepared using 6% NovexTM DNA Retardation Gels (Invitrogen), and the separated products were transferred from the gel onto a Hybond N+ membrane (GE healthcare, UK) by electrophoresis at 400 mV constant voltage for 30 min. The biotin-labeled probes were detected using the ABC kitTM (Vector Laboratories) and SuperSignalTM West Dura Extended Duration Substrate (Pierce). Unlabeled double-stranded oligonucleotide probes carrying point mutations in the Sp1 consensus sequences (oligonucleotides for probe-1: 5'-GGG ACG GGA GGG GAG GGG CGG C-3' and 5'-GCC GCC CCT CCC CTC CCG TCC C-3'; oligonucleotides for probe-2, 5'-GCT GAC CCC TCC CCG GGC CGG A-3' and 5'-TCC GGC CCG GGG AGG GGT CAG C-3') were used for the absorption assays. For competition assays, the HeLa nuclear extract was incubated with 1.75 pmol of nonlabeled Sp1 consensus oligonucleotide (Promega) or 1 µg anti-Sp1 antibody (PEP2, Santa Cruz Biotechnology, CA) for 60 min at 4°C in the presence of the binding buffer before the addition of the biotin-labeled probes.

Western blot analysis
Nuclear proteins were extracted from HeLa, NCCIT, and SK-MEL-28 cell by using NE-PERTM Nuclear and Cytoplasmic Extraction Regents (Pierce) according to manufacturer's instructions. Nuclear proteins (3 µg) were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) (SuperSepTM 5–20% Gel, Wako, Osaka, Japan) and the separated proteins were transferred from the gel onto the Immobilon-P PVDF membranes (Millipore Corp., Bedford, MA) by electrophoresis at 90mV constant voltage for 1 h. After blotting, the membranes were incubated either with the anti-Sp1 antibody (PEP2, Santa Cruz Biotechnology) or anti-KLF4 antibody (H-180, Santa Cruz Biotechnology), and bound antibody was detected using the ABC kitTM (Vector Laboratories) and SuperSignalTM West Dura Extended Duration Substrate (Pierce).


   Funding
Top
 Abstract
 Introduction
 Results
 Discussion
 Material and methods
 Funding
Conflict of interest statement
 References
 
National Institute of Advanced Industrial Science and Technology.


   Conflict of interest statement
Top
 Abstract
 Introduction
 Results
 Discussion
 Material and methods
 Funding
 Conflict of interest statement
References
 
None declared.


   Abbreviations
 
EC, endothelial cell; 2-DG, 2-deoxy-D-glucose; DMEM, Dulbecco's modified Eagle's minimal essential medium; Gb3, globotriaosylceramide (Gal{alpha}1,4Galβ1,4Glc–Ceramide); GM3, NeuAc{alpha}2,3Galβ1,4Glc–ceramide; HUS, hemolytic uremic syndrome; LDL, low-density lipoprotein; LPS, lipopolysaccharide; RT-PCR, real-time polymerase chain reaction; SDS-PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis; SNP, single nucleotide polymorphism; STZ, streptozotocin; TLC, thin-layer chromatography; TNF, tumor necrosis factor; VT, verotoxin; VEGF, vascular endothelial growth factor


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Material and methods
 Funding
 Conflict of interest statement
 References
 
Blume SW, Snyder RC, Ray R, Thomas S, Koller CA, Miller DM. Mithramycin inhibits SP1 binding and selectively inhibits transcriptional activity of the dihydrofolate reductase gene in vitro and in vivo. J Clin Invest. (1991) 88:1613–1621.[Web of Science][Medline]

Brembeck FH, Rustgi AK. The tissue-dependent keratin 19 gene transcription is regulated by GKLF/KLF4 and Sp1. J Biol Chem. (2000) 275:28230–28239.[Abstract/Free Full Text]

Dynan WS, Tjian R. The promoter-specific transcription factor Sp1 binds to upstream sequences in the SV40 early promoter. Cell. (1983) 35:79–87.[CrossRef][Web of Science][Medline]

Furukawa K, Clausen H, Hakomori S, Sakamoto J, Look K, Lundblad A, Mattes MJ, Lloyd KO. Analysis of the specificity of five murine anti-blood group A monoclonal antibodies, including one that identifies type 3 and type 4 A determinants. Biochemistry. (1985) 24:7820–7826.

Giraudo E, Primo L, Audero E, Gerber HP, Koolwijk P, Soker S, Klagsbrun M, Ferrara N, Bussolino F. Tumor necrosis factor-alpha regulates expression of vascular endothelial growth factor receptor-2 and of its co-receptor neuropilin-1 in human vascular endothelial cells. J Biol Chem. (1998) 273:22128–22135.[Abstract/Free Full Text]

Hakomori SI, Siddiqui B, Li YT, Li SC, Hellerqvist CG. Anomeric structures of globoside and ceramide trihexoside of human erythrocytes and hamster fibroblasts. J Biol Chem. (1971) 246:2271–2277.[Abstract/Free Full Text]

Hamanaka R, Kohno K, Seguchi T, Okamura K, Morimoto A, Ono M, Ogata J, Kuwano M. Induction of low density lipoprotein receptor and a transcription factor SP-1 by tumor necrosis factor in human microvascular endothelial cells. J Biol Chem. (1992) 267:13160–13165.[Abstract/Free Full Text]

Higaki Y, Schullery D, Kawata Y, Shnyreva M, Abrass C, Bomsztyk K. Synergistic activation of the rat laminin gamma1 chain promoter by the gut-enriched Kruppel-like factor (GKLF/KLF4) and Sp1. Nucleic Acids Res. (2002) 30:2270–2279.[Abstract/Free Full Text]

Ichikawa S, Ozawa K, Hirabayashi Y. Molecular cloning and characterization of the mouse ceramide glucosyltransferase gene. Biochem Biophys Res Commun. (1998) 253:707–711.[CrossRef][Medline]

Iwamura K, Furukawa K, Uchikawa M, Sojka BN, Kojima Y, Wiels J, Shiku H, Urano T, Furukawa K. The blood group P1 synthase gene is identical to the Gb3/CD77 synthase gene. A clue to the solution of the P1/P2/p puzzle. J Biol Chem. (2003) 278:44429–44438.[Abstract/Free Full Text]

Jacewicz M, Clausen H, Nudelman E, Donohue-Rolfe A, Keusch GT. Pathogenesis of shigella diarrhea: XI. Isolation of a shigella toxin-binding glycolipid from rabbit jejunum and HeLa cells and its identification as globotriaosylceramide. J Exp Med. (1986) 163:1391–1404.[Abstract/Free Full Text]

Kang HT, Ju JW, Cho JW, Hwang ES. Down-regulation of Sp1 activity through modulation of O-glycosylation by treatment with a low glucose mimetic, 2-deoxyglucose. J Biol Chem. (2003) 278:51223–51231.[Abstract/Free Full Text]

Keusch JJ, Manzella SM, Nyame KA, Cummings RD, Baenziger JU. Cloning of Gb3 synthase, the key enzyme in globo-series glycosphingolipid synthesis, predicts a family of alpha 1, 4-glycosyltransferases conserved in plants, insects, and mammals. J Biol Chem. (2000) 275:25315–25321.[Abstract/Free Full Text]

Kim SW, Lee SH, Kim KS, Kim CH, Choo YK, Lee YC. Isolation and characterization of the promoter region of the human GM3 synthase gene. Biochim Biophys Acta. (2002) 1578:84–89.[Medline]

Kojima Y, Fukumoto S, Furukawa K, Okajima T, Wiels J, Yokoyama K, Suzuki Y, Urano T, Ohta M, Furukawa K. Molecular cloning of globotriaosylceramide/CD77 synthase, a glycosyltransferase that initiates the synthesis of globo series glycosphingolipids. J Biol Chem. (2000) 275:15152–15156.[Abstract/Free Full Text]

Kolmakova A, Chatterjee S. Platelet derived growth factor recruits lactosylceramide to induce cell proliferation in UDP Gal:GlcCer: Beta1 -> 4Galactosyltransferase (GalT-V) mutant Chinese hamster ovary cells. Glycoconj J. (2005) 22:401–407.[CrossRef][Medline]

Lingwood CA, Law H, Richardson S, Petric M, Brunton JL, De Grandis S, Karmali M. Glycolipid binding of purified and recombinant Escherichia coli produced verotoxin in vitro. J Biol Chem. (1987) 262:8834–8839.[Abstract/Free Full Text]

Marcus DM, Kundu SK, Suzuki A. The P blood group system: Recent progress in immunochemistry and genetics. Semin Hematol. (1981) 18:63–71.[Web of Science][Medline]

Makita A. Biochemistry of organ glycolipids: II. Isolation of human kidney glycolipids. J Biochem. (1964) 55:269–276.[Free Full Text]

Noris M, Remuzzi G. Hemolytic uremic syndrome. J Am Soc Nephrol. (2005) 16:1035–1050.[Free Full Text]

Obrig TG, Louise CB, Lingwood CA, Boyd B, Barley-Maloney L, Daniel TO. Endothelial heterogeneity in Shiga toxin receptors and responses. J Biol Chem. (1993) 268:15484–15488.[Abstract/Free Full Text]

Okuda T, Tokuda N, Numata S, Ito M, Ohta M, Kawamura K, Wiels J, Urano T, Tajima O, Furukawa K, et al. Targeted disruption of Gb3/CD77 synthase gene resulted in the complete deletion of globo-series glycosphingolipids and loss of sensitivity to verotoxins. J Biol Chem. (2006) 281:10230–10235.[Abstract/Free Full Text]

Ray R, Snyder RC, Thomas S, Koller CA, Miller DM. Mithramycin blocks protein binding and function of the SV40 early promoter. J Clin Invest. (1989) 83:2003–2007.[Web of Science][Medline]

Sato T, Furukawa K. Transcriptional regulation of the human beta-1,4-galactosyltransferase V gene in cancer cells: Essential role of transcription factor Sp1. J Biol Chem. (2004) 279:39574–39583.[Abstract/Free Full Text]

Shie JL, Chen ZY, Fu M, Pestell RG, Tseng CC. Gut-enriched Krüppel-like factor represses cyclin D1 promoter activity through Sp1 motif. Nucleic Acids Res. (2000) 28:2969–2976.[Abstract/Free Full Text]

Spengler ML, Brattain MG. Sumoylation inhibits cleavage of Sp1 N-terminal negative regulatory domain and inhibits Sp1-dependent transcription. J Biol Chem. (2006) 281:5567–5574.[Abstract/Free Full Text]

Steffensen R, Carlier K, Wiels J, Levery SB, Stroud M, Cedergren B, Nilsson Sojka B, Bennett EP, Jersild C, Clausen H. Cloning and expression of the histo-blood group Pk UDP-galactose: Ga1beta-4G1cbeta1-cer alpha1, 4-galactosyltransferase. Molecular genetic basis of the p phenotype. J Biol Chem. (2000) 275:16723–16729.[Abstract/Free Full Text]

Stricklett PK, Hughes AK, Kohan DE. Inhibition of p38 mitogen-activated protein kinase ameliorates cytokine up-regulated shigatoxin-1 toxicity in human brain microvascular endothelial cells. J Infect Dis. (2005) 191:461–471.[CrossRef][Medline]

van de Kar NC, Monnens LA, Karmali MA, van Hinsbergh VW. Tumor necrosis factor and interleukin-1 induce expression of the verocytotoxin receptor globotriaosylceramide on human endothelial cells: Implications for the pathogenesis of the hemolytic uremic syndrome. Blood. (1992) 80:2755–2764.[Abstract/Free Full Text]

Wiels J, Fellous M, Tursz T. Monoclonal antibody against a Burkitt lymphoma-associated antigen. Proc Natl Acad Sci USA. (1981) 78:6485–6488.[Abstract/Free Full Text]

Yang X, Su K, Roos MD, Chang Q, Paterson AJ, Kudlow JE. O-Linkage of N-acetylglucosamine to Sp1 activation domain inhibits its transcriptional capability. Proc Natl Acad Sci USA. (2001) 98:6611–6616.[Abstract/Free Full Text]

Zeng G, Gao L, Xia T, Tencomnao T, Yu RK. Characterization of the 5'-flanking fragment of the human GM3-synthase gene. Biochim Biophys Acta. (2003) 1625:30–1635.[Medline]

Zhang F, Su K, Yang X, Bowe DB, Paterson AJ, Kudlow JE. O-GlcNAc modification is an endogenous inhibitor of the proteasome. Cell. (2003) 115:715–725.[CrossRef][Web of Science][Medline]

Zhang W, Shields JM, Sogawa K, Fujii-Kuriyama Y, Yang VW. The gut-enriched Krüppel-like factor suppresses the activity of the CYP1A1 promoter in an Sp1-dependent fashion. J Biol Chem. (1998) 273:17917–17925.[Abstract/Free Full Text]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
18/12/1028    most recent
cwn082v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
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 Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Okuda, T.
Right arrow Articles by Nakayama, K.-i.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Okuda, T.
Right arrow Articles by Nakayama, K.-i.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?