Skip Navigation

Cardiovascular Research 2000 46(3):511-522; doi:10.1016/S0008-6363(00)00041-9
© 2000 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
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 Bierhuizen, M. F.A.
Right arrow Articles by Jongsma, H. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bierhuizen, M. F.A.
Right arrow Articles by Jongsma, H. J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 2000, European Society of Cardiology

Characterization of the rat connexin40 promoter: two Sp1/Sp3 binding sites contribute to transcriptional activation

Marti F.A. Bierhuizen*, Shirley C.M. van Amersfoorth, W.Antoinette Groenewegen, Saskia Vliex and Habo J. Jongsma

University Medical Center Utrecht, Department of Medical Physiology, P.O. Box 80043, 3508 TA Utrecht, The Netherlands

* Corresponding author. Tel.: +31-30-253-8418; fax: +31-30-253-9036 m.f.a.bierhuizen{at}med.uu.nl

Received 30 September 1999; accepted 19 January 2000


   Abstract
Top
 Abstract
1 Introduction
 2 Methods
 3 Results
 4 Discussion
 References
 
Objectives: The gap junction protein connexin40 (Cx40) is differentially expressed during embryonic development and in adult tissues, for which the molecular basis is unknown. In order to elucidate the molecular mechanisms controlling Cx40 expression, we set out to map and characterize its promoter. Methods: The transcriptional activity of individual rat Cx40 (rCx40)-derived promoter fragments fused to the luciferase reporter gene was determined by transfection/reporter assays in Cx40-expressing (A7r5, rat smooth muscle embryonic thoracic aorta cells, and BWEM, v-myc transformed rat fetal cardiomyocytes) and Cx40-nonexpressing cells (N2A, mouse neuroblastoma cells). The nature of DNA–protein interactions was investigated by a combination of standard electrophoretic-mobility-shift assays (EMSA) and EMSA/antibody supershift assays. Results: Quantification of luciferase activity in cell lysates revealed that a 235-base-pair fragment, in between map positions –150 and +85 relative to the transcription initiation site, is able to provide for a significant level of transcription in both Cx40-expressing (A7r5, BWEM) and -nonexpressing (N2A) cells. These results indicate that this region contains the basal promoter but is not sufficient to completely determine the endogenous Cx40-expression pattern within these cell types. In search for the responsible transcriptional regulatory element(s), additional segments of the (–150, +85) region were deleted and the remaining fragments were tested for transcriptional activity. These studies established that the regions in between map positions (–96, –71) and (+58, +85) contribute to promoter activity. EMSA with these regions revealed that predominantly two DNA–protein complexes are formed upon incubation with either A7r5, BWEM or N2A nuclear extracts, which could be both inhibited by including excess oligonucleotide containing the Sp1 consensus binding site in the binding reaction. Purified recombinant human Sp1 provided also for a shift in the EMSA using these promoter regions as target fragments. When the DNA–protein complexes formed with nuclear extract were subsequently incubated with either an anti-Sp1 or an anti-Sp3 antibody clear supershifts in the EMSA were obtained, indicating Sp1 and Sp3 binding to both the (–98, –64) and (+53, +87) regions. The introduction of mutations within the core sequence of the putative Sp1/Sp3 binding sites present in these regulatory elements reduced the level of transcriptional activity and abrogated Sp1/Sp3 binding to these sites. Conclusion: The results indicate that at least two Sp1/Sp3 binding sites in the rCx40 promoter contribute to the transcriptional activation of its gene in cultured cells.

KEYWORDS Gap junctions; Gene expression


   1 Introduction
Top
 Abstract
 1 Introduction
2 Methods
 3 Results
 4 Discussion
 References
 
Clusters of intercellular channels (gap junctions) allow adjacent cells to electrically and chemically communicate through the exchange of ions or small substances (such as nutrients or regulatory molecules). A complete gap junction channel interconnects the cytoplasms of two neighboring cells through an aqueous pore formed by the association of two hemichannels (or connexons), each of which is contributed by one of the participating cells. Gap junction proteins (or connexins, Cx) constitute the protein molecules of the gap junction channel, six of which are assembled into each connexon [1,2]. The total number of different vertebrate connexin genes is expected to exceed twenty, fourteen of which have been identified in rodents [3]. Extensive amino acid sequence analysis predicts common structural features for each connexin, consisting of e.g. a well-conserved cytoplasmic N-terminal part, four transmembrane regions, two extracellular domains and a variable cytoplasmic loop and C-terminus [1,2]. The overall genomic organization of connexin genes appears to be rather simple, consisting of two exons with most connexins having the complete coding region on exon2 and the 5'-untranslated region divided by an intron over exons 1 and –2. A new subgroup, however, has recently been defined of which the gene structure is characterized by the presence of an intron within the coding region [4].

Connexins differ in their functional (gating) properties and their developmental- and cell type-specific expression patterns. Some gap junction proteins are widely distributed among cells and tissues (e.g. Cx43), whereas others display a much more restricted expression profile (e.g. Cx33) [1]. Cx40 RNA has been detected in several adult tissues, including heart, lung, ovary and kidney, but not in e.g. brain, spleen or liver [5,6]. In the adult heart of most species, Cx40 expression is restricted to the endothelial layer of blood vessels, the myocytes of atria and the ventricular conduction system, except for the rat where Cx40 has not been detected in atrial myocytes [7]. Differential expression of Cx40 is also observed during embryonic development in the rat heart. Cx40 expression first becomes detectable at embryonic day 13, increases in the fetal period and subsequently decreases towards birth [8]. Altogether, these cell type-specific and developmental expression patterns suggest that Cx40 expression is under tight spatiotemporal control.

The abundant presence of Cx40 in myocytes of the ventricular conduction system has been correlated with the fast electrical conduction properties of this cardiac region [7,9]. Alterations in the expression level and distribution pattern of this gap junction protein may give rise to abnormalities in action potential formation and conduction. Such alterations have been documented for Cx40 in persistent atrial fibrillation [10] and hypertrophy [11,12]. Cardiac conduction abnormalities were observed in knock-out mice lacking Cx40 which were correlated with a higher incidence of arrhythmias [13,14]. In order to begin to understand the molecular mechanisms that control the expression of Cx40 under (patho)physiological circumstances, the transcriptional regulation of the rCx40 gene is addressed. To that purpose, our laboratory [15] as well as others [16] have elucidated its genomic organization in rat and mouse, respectively. Both studies revealed that the Cx40 gene fits the overall gene structure profile for all characterized rodent connexin genes studied to date, i.e. a two-exon gene with the complete coding region on exon2. In this report, we extend these studies by characterizing the rCx40 promoter and providing experimental evidence that two Sp1/Sp3 binding sites contribute to its gene activation in cultured cells.


   2 Methods
Top
 Abstract
 1 Introduction
 2 Methods
3 Results
 4 Discussion
 References
 
2.1 Cell culture
The rat thoracic aorta smooth muscle cell line A7r5 [17] and mouse neuroblastoma N2A cells [18] were grown in Dulbecco's modified Eagle's medium (DMEM; Boehringer Ingelheim, Heidelberg, Germany) supplemented with 10% fetal bovine serum (FBS) and 2 mM of L-glutamine (Boehringer Ingelheim). The rat v-myc transformed fetal cardiomyocyte cell line BWEM [19] was cultured in DMEM–nutrient mix F12 (1:1, Life Technologies, Rockville, MD, USA) supplemented with 5% FBS. These cultures were maintained at 37°C and 5% CO2–95% air in a humidified atmosphere. Penicillin (50 units/ml) and streptomycin (50 µg/ml; both from Boehringer Ingelheim) were routinely included in all culture media.

2.2 Plasmid construction
The pGL3-Basic vector (Promega, Madison, WI, USA), for which Firefly luciferase expression in eukaryotic cells is dependent on the insertion and proper orientation of a functional promoter, served as basis for the generation of chimeric rCx40 promoter/luciferase constructs. Various 5'-upstream regions of the rCx40 gene were obtained by the polymerase chain reaction (PCR) using synthetic oligonucleotides (Pharmacia, Uppsala, Sweden) chosen on the basis of the known nucleotide sequence [15] and either A7r5 genomic DNA or a suitable cloned genomic fragment as template. Site-directed mutations in the rCx40 5' flanking sequence were introduced in a similar manner by employing either 5'-sense or 3'-antisense primers containing dinucleotide substitutions relative to the native nucleotide sequence. For cloning purposes the 5'-sense primers contained an additional XhoI restriction site, whereas all 3'-antisense primers contained a HindIII site. Amplified DNA fragments were subsequently cloned into the XhoI/HindIII sites of the pGL3-Basic vector by standard cloning procedures [20]. The integrity of each construct was confirmed by BigDye-cycle sequencing analysis in two directions on an ABI Prism 310 Genetic Analyzer (Perkin Elmer, Foster City, CA, USA). Four of the thus obtained reporter constructs are schematically represented in Fig. 1. The 5'-sense primer sequences used for the generation of these constructs were as follows (in 5' to 3' direction; numbering relative to the transcription initiation site as determined earlier [15]): TTTCTCGAGTTCCCTGGCCATTGCAAATGG (map positions –1125 to –1105), TTTCTCGAGAGGGTTTGTTGTGTGGATAGA (–455 to –435), TTTCTCGAGACAGCTAGCAAGTGCATGGAC (–296 to –276), and TTTCTCGAGTGAGAAGAAGGGAGGAGAGAG (–175 to –155); the 3'-antisense primer used in all four instances was 5'-TTTAAGCTTCTGCTTCCTTTCCTCCGCCCT-3' (+65 to +85).


Figure 1
View larger version (10K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Schematic representation of the 5'-upstream region of the rCx40 gene (top) and the deduced rCx40 promoter (PCx40)/firefly luciferase (LUC) chimeric constructs (bottom). The two exons are indicated by boxes E1 and E2, which are separated from each other by a large fragment of intervening sequence (intron). The complete rCx40 coding sequence is localized on E2, as is visualized here by the relative position of the protein start codon (ATG). The filled-in portion of the exon boxes represents the 5'-untranslated region as present in the mRNA. Nucleotides are numbered relative to the position of the transcription initiation site as determined earlier [15].

 
All pGL3-based constructs containing different fragments of the 5'-upstream region of rCx40 are named according to the numbered positions of these fragments relative to the transcription initiation site (see e.g. Fig. 1). Site-directed mutations were only introduced in the region of the rCx40 gene in between map positions (–96,+85) and the constructs containing such substitutions are indicated with the prefix MT and named after the substituted positions. The plasmid MT (–84, –83), for example, contains a dinucleotide substitution in the (–96, +85) region at positions –84 and –83 to drive expression of the Firefly luciferase gene.

The pRL-CMV control reporter vector, containing the Renilla luciferase gene under control of the cytomegalovirus immediate-early enhancer/promoter, was obtained from Promega.

2.3 Transfection of cells and luciferase assays
For transient expression analysis of luciferase reporter activity in A7r5 and N2A cells, 2 or 4·105 cells, respectively, were plated in 60-mm diameter culture dishes and transfected by the standard calcium phosphate precipitation technique at 37°C and 5% CO2 [21] using 2 µg of experimental construct and 0.5 µg of pRL-CMV reporter vector. Since BWEM cells proved refractory to standard calcium phosphate transfection, 2·105 cells were transfected on 60-mm diameter dishes with 6.4 µg experimental and 1.6 µg reporter plasmid by a modified procedure at 35°C and 3% CO2 [22]. Cell lysates were prepared after a 24–48 h expression period and assayed for luciferase activity on a Lumat LB 9507 luminometer (EG&G Berthold, Bad WildBad, Germany) using the dual-luciferase reporter assay system (Promega) essentially as described by the manufacturer. The Renilla luciferase activity measurements were used to normalize for differences in transfection efficiency in between individual transfections.

For each construct to be tested, two different plasmid preparations were used to assay for luciferase activity in a total of at least four independent transfection experiments and the results are expressed as mean±SEM.

2.4 Electrophoretic-mobility-shift assays (EMSA)
Microscale preparations of nuclear extracts derived from A7r5, BWEM and N2A cells were obtained as described by Li et al. [23]. Genomic sequences corresponding to specific parts in the 5'-flanking region of the rCx40 gene (e.g. the regions from –98 to –64 and from +53 to +87) were prepared by annealing the appropriate complementary synthetic oligonucleotides and labeling them with [{gamma}-32P]ATP and T4 kinase [20]. Approximately 50 000 cpm of these 32P-endlabeled probes were incubated with nuclear extract (15 µg protein per reaction) or recombinant human Sp1 (Promega; 1.5 footprinting unit per reaction) in a 21-µl reaction mixture containing 27 mM HEPES, pH 7.5, 1.25 mM MgCl2, 1.2 mM DTT, 0.01% NP-40, 70 mM KCl, 4.2% glycerol, 0.1 mM EDTA, and 1 µg poly(dI–dC) for 30–60 min on ice. In competition experiments, a 20- to 100-fold molar excess of double-stranded and non-radioactive self oligonucleotide, mutated versions of the self oligonucleotide or commercially obtained oligonucleotides (Promega) containing transcription factor binding motifs was added just before the start of the reaction as well. The nomenclature and sequences of the oligonucleotides used in this way are summarized in Table 1. For gel supershift experiments, 200 ng of a rabbit anti-Sp1 polyclonal or 2 µg of a goat anti-Sp3 polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA) was added at the end of the binding reaction as described above and incubation was continued for 30 min at 4°C. As a negative control for the Sp1 supershift assay, 200 ng of a rabbit anti-desmin polyclonal antibody (Dako, Carpinteria, CA, USA) was used similarly.


View this table:
[in this window]
[in a new window]

  		Table 1 Oligonucleotides used in the electrophoretic-mobility-shift assays either as radioactive target or non-radioactive competitor sequence

 
All reactions as described above were analyzed on a 4% non-denaturing polyacrylamide gel run in 25 mM Tris, pH 8.2, 0.5 mM EDTA, 25 mM boric acid, at 200 V at room temperature for 2 h. After electrophoresis, the gel was dried under vacuum at 80°C and exposed to X-ray film.


   3 Results
Top
 Abstract
 1 Introduction
 2 Methods
 3 Results
4 Discussion
 References
 
3.1 rCx40 (–175, +85) gene fragment contains the transcriptional regulatory region.
The genomic organization of the rCx40 gene has recently been elucidated and shown to consist of a noncoding exon of 85 bp (E1 in Fig. 1, top), a relatively large intervening sequence of presumably over 5.5 kb, and a second exon (E2 in Fig. 1, top) containing the complete protein coding sequence [15]. In order to map the transcriptional regulatory region, four different parts of the 5'-flanking region were placed in front of the Firefly luciferase reporter gene in the promoterless pGL3-Basic vector. The constructs differed in the length of the 5'-upstream sequence present, the largest containing 1125 bps and the smallest containing 175 bps of 5'-flanking sequence. The 3'-end of each construct, however, was similar and contained the end of exon1 (map position +85) as a common 3'-endpoint (see Fig. 1, bottom, for a schematic overview of these constructs). We have used two Cx40-expressing cell lines (A7r5, BWEM) and one -nonexpressing cell line (N2A) to analyze the transcriptional activity of these chimeric reporter constructs. The nature of the cell lines, with respect to Cx40 expression, was checked by RT-PCR and immunohistochemical analysis (results not shown); Cx40 RNA and protein were easily detected in both A7r5 and BWEM cells but not in N2A cells, the highest protein level being detected in A7r5 cells. The reporter constructs schematically represented in Fig. 1 were transfected into all three cell types and after a 24–48 h expression period cell lysates were analyzed for luciferase activity. As is shown in Fig. 2, full promoter activity was obtained in all three cell types even with the smallest construct tested, i.e. p(–175,+85), indicating that the rCx40 promoter is contained within the (–175,+85) region. The overall promoter activity in the three cell lines decreased in a similar direction as the level of endogenous Cx40 protein expression, i.e. A7r5>BWEM>N2A, with the levels of luciferase activity reaching maximally 72-fold the level obtained with the promoterless pGL3-Basic vector in A7r5, 17-fold in BWEM, and 8-fold in N2A cells.


Figure 2
View larger version (11K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Mapping of the transcriptional regulatory region of the rCx40 gene by luciferase expression analysis. The reporter constructs as indicated in Fig. 1 were tested for transcriptional activity in A7r5, BWEM and N2A cells as described in Methods. The luciferase activity obtained for the pGL3-Basic vector, in which a functional promoter was absent, was set at 1.0 and all other luciferase activities were expressed as fold of the values obtained for pGL3-Basic.

 
3.2 Multiple DNA elements are involved in rCx40 promoter activity
In order to further characterize functional motifs in the rCx40 promoter, a series of constructs was generated in which small parts of the (–175,+85) region were serially deleted at either the 5'- or 3'-end. These truncated promoter fragments were tested for transcriptional activity in A7r5, BWEM, and N2A cells as described before and the results are schematically summarized in Fig. 3. Irrespective of the cell line used for the reporter assay, no difference in relative luciferase activity was detected upon comparison of the p(–175,+85) and p(–150,+85) constructs, thus further confining the promoter to the (–150,+85) region. A reduction in transcriptional activity in all three cell types was obtained when comparing the results for the p(–150,+85) and p(–96,+85) constructs, but since the results varied with the cell line used no direct correlation between deletion of either the (–150,–120) or the (–120,–96) region and the decrease in promoter activity could be noticed. In contrast, the stepwise removal of the sequences in between map positions (–96,–71), (–71,–50), (+58,+85) and (+21,+58) correlated well with successive reductions in relative luciferase activity in all three cell types. These results suggest the involvement of multiple transcription factor binding sites in transcriptional activation of the rCx40 gene.


Figure 3
View larger version (16K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Precise analysis of rCx40 transcriptional regulatory sequences involved in promoter activity. Reporter constructs containing further 5'- and 3'-deletions relative to p(–175,+85) were assayed for transcriptional activity essentially as described in the legend of Fig. 2.

 
Nucleotide comparison of the rCx40 (–175,+85) region with the corresponding sequences from mouse or human origin revealed a relative high degree of homology: the rat and mouse sequences are for 94%, and the rat and human sequences for 72% identical to each other (Fig. 4). The rCx40 (–175,+85) region was searched for the presence of potential transcription factor binding sites with the MATINSPECTOR V2.2 search program (core similarity 0.75; matrix similarity 0.85) using the TRANSFAC transcription factor database [24] and numerous potential binding sites were identified. Most strikingly, five potential binding sites for the transcription factor Sp1 and/or Sp3 (i.e. GC-box elements) [25–27] were identified, which are indicated in Fig. 4 by a thick line underneath the putative core sequence of the element. Four of these elements were highly conserved in between the rat, mouse and human sequence, at least three of which were localized in regions of the rat promoter which upon deletion provided for a marked decrease in relative luciferase activity in all three cell lines tested. The involvement of two of these regions in transcriptional activation, i.e. the regions in between positions (–96,–71) and (+58,+85) (both regions are boxed in Fig. 4), was studied in more detail.


Figure 4
View larger version (26K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Nucleotide sequence comparison of the rat, mouse and human Cx40 5'-upstream region. The nucleotide sequence of the rCx40 gene (top) is numbered relative to the transcription initiation site [15]. The 5'-end of human Cx40 mRNA was isolated by 5'-RACE and the putative promoter of the human Cx40 gene was obtained by long distance PCR essentially described as before [15]. The mouse sequence is obtained from [16]. The nucleotides which are identical between either mouse or human and rat Cx40 are indicated by dashes, whereas insertions in one particular sequence which are absent in the others are indicated by gaps in the latter. The transcriptional regulatory sequences under investigation in this study are boxed. The core sequences of potential GC-box elements are underlined.

 
3.3 EMSA/antibody supershift assays with oligonucleotides corresponding to the (–98,–64) and (+53,+87) regions demonstrate Sp1 and Sp3 binding
Since the reporter assays indicated that removal of the DNA sequences from –96 to –71 and from +58 to +85 resulted in marked decreases in transcriptional activity and analysis of these sequences revealed the presence of putative GC-box elements, we were interested whether Sp1 and/or Sp3 could bind to either one of these regions. Therefore, double-stranded synthetic oligonucleotides encompassing the regions from –98 to –64 and from +53 to +87 were radiolabeled, incubated with nuclear extracts prepared from A7r5, BWEM or N2A cells, and finally analyzed by non-denaturing polyacrylamide gel electrophoresis. As is shown in Fig. 5 for the (–98,–64) region, the results obtained with A7r5 (Fig. 5A), BWEM (Fig. 5B), and N2A (Fig. 5C) nuclear extracts were identical. Whereas in the absence of nuclear extract only the free oligonucleotide probe was visible, in the presence of nuclear extract several slower migrating DNA–protein complexes appeared, two of which were predominant (indicated here with C1 and C2). When the binding reaction was performed in the presence of a 100-fold molar excess of the nonlabeled fragment itself (self in Fig. 5) or a commercial oligonucleotide containing a Sp1 consensus binding site, the complexes C1 and C2 were both competed away in contrast to one of the minor complexes which therefore appeared to be nonspecific. However, when the binding reactions were performed in the presence of a 100-fold molar excess of commercial oligonucleotides containing either an AP1 or a NF-{kappa}B consensus binding site competition was almost absent. Very similar results were obtained when the (–98,–64) region was replaced by the (+53,+87) region as radioactive target sequence, except that the presence of a nonlabeled NF-{kappa}B oligonucleotide in the binding reaction appeared to be more competitive (results not shown). The above-mentioned results suggested the involvement of Sp1 and/or Sp3 in complex formation with both the (–98,–64) and (+53,+87) regions.


Figure 5
View larger version (24K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 EMSA of rCx40 promoter DNA after incubation with nuclear extracts in the presence or absence of specific DNA competitors. A 32P-endlabeled rCx40 specific DNA fragment encompassing nucleotides –98 to –64 was incubated with 15 µg of nuclear extract prepared from A7r5 (A), BWEM (B), or N2A cells (C) in the presence or absence of a 100-fold molar excess of nonlabeled competitor DNA. As competitors either the (–98,–64) rCx40 fragment itself (self) or commercial double-stranded oligonucleotides were used containing binding motifs for the transcription factors Sp1, AP1 or NF-{kappa}B as indicated above the figures. The intense band in each lane at the bottom of the figure represents free probe. The two shifted bands in the upper half of each figure indicated with C1 and C2 represent formed DNA–protein complexes.

 
Therefore, binding reactions between radiolabeled oligonucleotides and nuclear extracts were performed in the presence of anti-Sp1 antibody or, as control, anti-desmin antibody as well. As is shown in Fig. 6A and C, a supershifted complex band was obtained when the incubation was performed in the presence of anti-Sp1 but not anti-desmin antibody which appeared to be primarily derived from the original complex C1. Irrespective of the origin of the nuclear extract or of the rCx40-specific region used in the binding assay, a similar bandshift profile was obtained in each comparative instance (compare e.g. Fig. 6A with Fig. 6C, or within one figure the comparative lanes for the individual nuclear extracts). In Fig. 6B and D, the bandshift and supershift assays for both rCx40 fragments were performed with commercially obtained recombinant human Sp1 transcription factor and compared with the corresponding experiments with A7r5 nuclear extract. The binding of purified recombinant Sp1 transcription factor to both regions was evident from the appearance of a rather heterogeneous shifted band, which migration behavior was further slowed down upon subsequent incubation with the anti-Sp1 antibody. Altogether these results indicated that indeed Sp1 was able to bind to both the (–98,–64) and (+53,+87) region.


Figure 6
View larger version (33K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 Analysis of Sp1 transcription factor binding to the rCx40 promoter.32P-endlabeled rCx40 specific DNA fragments encompassing nucleotides –98 to –64 (A and B) and +53 to +87 (C and D) were incubated with 15 µg of nuclear extract prepared from A7r5 (A7r5 or A above the figure in the ‘extract’ line), BWEM or N2A cells or with 1.5 footprinting unit of recombinant human Sp1 (S above figure in the ‘extract’ line) as indicated. In some instances the reaction mixtures were subsequently incubated with a rabbit anti-Sp1 or, as a negative control, a rabbit anti-desmin polyclonal antibody (indicated with S or D, respectively, in the ‘antibody’ line above the figure). All samples were analyzed as described in Methods. Only shifted bands are shown; the complexes C1 and C2 are indicated.

 
To investigate whether Sp3 could also bind to the same DNA fragments, especially since the DNA elements recognized by both Sp1 and Sp3 appear to be identical [27], the antibody supershift assays were repeated with A7r5 nuclear extract and antibodies directed against Sp1 or Sp3 (Fig. 7). As mentioned before, anti-Sp1 antibody caused a change in the migration behavior of primarily the shifted complex C1. When anti-Sp3 antibody alone was added to the binding reaction also a supershift was observed which appeared to be primarily derived from complex C2. The appliance of both anti-Sp1 and -Sp3 antibodies at the same time, however, revealed that also the remaining part of complex C1 was supershifted, which corresponds with EMSA patterns for Sp1/Sp3 binding reported before (see e.g. Ref. [27]). Two additional (minor) DNA–protein complexes were noticed in Figs. 6 and 7Go, of which the fastest migrating complex appears to be nonspecific (see above); the nature of the other complex is presently unknown but might be due to binding of other Sp-related proteins with overlapping DNA-binding specificities [27]. Nevertheless, these results indicate that, besides Sp1, Sp3 can also bind to both rCx40-derived promoter fragments.


Figure 7
View larger version (21K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7 Analysis of Sp1 and Sp3 transcription factor binding to the rCx40 promoter.32P-endlabeled rCx40 specific DNA fragments encompassing nucleotides –98 to –64 (left figure) and +53 to +87 (right figure) were incubated with 15 µg of A7r5 nuclear extract as indicated above the figure. Anti-Sp1 and/or -Sp3 antibodies were subsequently added as indicated above the figure and incubation was continued. All samples were analyzed as described in Methods. Only shifted bands are shown; the complexes C1 and C2 are indicated.

 
3.4 Deletion or mutation of the Sp1/Sp3 binding sites interferes with transcriptional activity and abolishes Sp1/Sp3 binding
To further confirm that the two Sp1/Sp3 binding sites contribute to transcriptional activation, reporter assays were carried out in A7r5 cells with constructs in which the luciferase gene was driven by rCx40-derived promoter fragments that either lacked both sites completely or contained specific dinucleotide substitutions in these sites. The plasmid p(–96,+85), which still possesses both binding sites, served as a positive reference (see Fig. 8A for its schematic structure) for these experiments whereas the promoterless pGL3-Basic vector was included to relate all of the obtained results to. The schematic structures of all constructs tested and the results of their reporter assays are summarized in Fig. 8B. Removal of both the DNA sequences between map positions –96 to –71 and between +58 and +85 [plasmid p(–71,+58)] resulted in a 75% reduction of transcriptional activity. Introduction of a dinucleotide substitution at positions –91 and –90 [MT(–91,–90)] or at positions +82 and +83 [MT(+82,+83)], which positions are both clearly separated from the core of the GC-box element, hardly had any effect on promoter activity. In contrast, dinucleotide substitutions within the GC-box core sequence either at positions –84 and –83 [MT(–84,–83)] or at positions +68 and +69 [MT(+68,+69)] decreased transcriptional activity with approximately 40% and 89%, respectively. Mutations in both GC-box core sequences at the same time [MT(–84,–83)(+68,+69)] provided for a similar reduction of promoter activity as did MT(+68,+69) alone, suggesting that the Sp1/Sp3 binding site located in exon1 is most critical for promoter activity of both sites studied.


Figure 8
View larger version (9K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 8 Effect of specific mutations in and around the putative Sp1/Sp3 binding sites on transcriptional activity. (A) Schematic representation of the rCx40 promoter region under investigation. The core sequences of putative Sp1/Sp3 binding sites are boxed and the dinucleotides targeted for mutation are marked with asterisks. (B) Schematic structure and nomenclature of reporter constructs containing dinucleotide mutations in or around the Sp1/Sp3 binding sites. A horizontal line represents the original sequence whereas the mutated nucleotides newly introduced are displayed; the positions of these mutations are indicated in between parentheses after the prefix MT. All reporter constructs were tested for transcriptional activity in A7r5 cells essentially as described in the legend of Fig. 2.

 
The effect of the above-mentioned dinucleotide substitutions on Sp1/Sp3 binding was further investigated by EMSA. Double-stranded synthetic oligonucleotides corresponding to the regions from –98 to –64 and from +53 to +87, with or without the aforementioned dinucleotide substitutions, served as radioactive target sequences in an incubation with A7r5 nuclear extract; the resulting samples were further analyzed by non-denaturing gel electrophoresis as before. The nomenclature and sequences of the primers used are summarized in Table 1. As is shown in Fig. 9A for the (–98,–64) and in Fig. 9C for the (+53,+87) region, mutations within the core sequence itself (Mt1 and Mt3, respectively) almost completely abolished the binding of Sp1/Sp3 (as visualized by the absence of DNA–protein complexes C1 and C2) whereas mutations outside the core (Mt2 and Mt4, respectively) had no effect. For comparison, the original (–98,–64) and (+53,+87) regions (denoted with wt in each case) were applied as well. Conversely, the competition ability of 20- to 100-fold molar excesses of the corresponding nonlabeled oligonucleotides was tested in bandshift experiments where either the (–98,–64) (Fig. 9B) or (+53,+87) region (Fig. 9D) served as radioactive target sequence. In both cases the competitors containing mutations in the core sequence of the binding site (Mt1 and Mt3, respectively) were not able to compete for Sp1/Sp3 binding, whereas the native (wt) sequences and those containing mutations outside the core sequence (Mt2 and Mt4, respectively) did compete.


Figure 9
View larger version (29K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 9 EMSA of rCx40 promoter DNA using synthetic oligonucleotides with specific mutations in and around the putative Sp1/Sp3 binding sites. Synthetic oligonucleotides corresponding to the rCx40 promoter (–98,–64) (A) and (+53,+87) (C) regions or mutated representatives thereof were endlabeled with 32P, incubated with 15 µg of A7r5 nuclear extract and analyzed by non-denaturing polyacrylamide gel electrophoresis. Alternatively, 32P-endlabeled rCx40 specific DNA fragments encompassing nucleotides –98 to –64 (B) and +53 to +87 (D) were incubated with 15 µg of A7r5 nuclear extract in the presence of 20-, 50- or 100-fold molar excess of nonlabeled synthetic oligonucleotides as indicated above the figure. The nomenclature and sequences of the oligonucleotides applied are represented in Table 1. Only shifted bands are shown; the complexes C1 and C2 are indicated.

 
Altogether, these results demonstrate that mutations within the core sequence of the Sp1/Sp3 binding sites interfere with transcriptional activity of the rCx40 promoter and with Sp1/Sp3 binding to these sites.


   4 Discussion
Top
 Abstract
 1 Introduction
 2 Methods
 3 Results
 4 Discussion
References
 
In the present study, the promoter of the rCx40 gene has been mapped and characterized as a prelude to the elucidation of the molecular mechanisms that determine the cell type-specific and developmental expression pattern of its gene product. Initially the transcriptional activity of a series of rCx40 promoter/reporter constructs was analyzed in two Cx40-expressing (A7r5, BWEM) and one -nonexpressing cell line (N2A). The results demonstrated that a 235 base pair region (in between map positions –150 and +85) was essential for full promoter activity in all three cell lines and that consecutive deletions of this region either at the 5'- or 3'-end gradually decreased the transcriptional activity of the remaining promoter fragment. Removal of the sequences from –96 to –71 and from +58 to +85 correlated with a marked reduction of promoter activity in all three cell lines, suggesting the presence of important regulatory elements within these regions. Both regions proved to be highly conserved in the corresponding rat, mouse and human promoter and appeared to contain a GC-box like core element, which forms a potential binding site for Sp family transcription factors. These results prompted us to investigate whether these GC-box like elements were involved in the transcriptional activation of the rCx40 gene. For that reason, a series of EMSA experiments was performed using oligonucleotide probes representing the (–98,–64) and the (+53,+87) regions and nuclear extracts derived from A7r5, BWEM, and N2A cells. Irrespective of the source of the nuclear extract used, with both oligonucleotide probes predominantly two DNA–protein complexes (referred to as C1 and C2) were obtained. A combination of competition studies and antibody supershift experiments then revealed that both Sp1 and Sp3 were able to form complexes with the rCx40-derived (–98,–64) and (+53,+87) regions, thus accounting for the occurrence of C1 and C2. This was further substantiated by the observation that purified recombinant human Sp1 was also able to bind to both rCx40 derived fragments, although the mobility shift obtained was rather heterogeneous and did not co-migrate with the C1 or C2 bands. However, since the purified Sp1 used was from human origin and the nuclear extracts studied were derived from either rat or mouse cells, this discrepancy in migration behavior may be explained by species-related differences in glycosylation and/or phosphorylation of the Sp1 protein [28,29]. The contribution of both Sp1/Sp3 binding sites to transcriptional activation of the rCx40 minimal promoter was confirmed by reporter assays in A7r5 cells with promoter/reporter constructs containing mutations in these particular GC-box like elements. Furthermore, bandshift experiments with the rCx40 (–98,–64) and (+53,+87) regions containing the corresponding mutations in the GC-box like elements revealed that Sp1/Sp3 binding is dependent on the integrity of these elements. Based on the combined results, we conclude that these two Sp1/Sp3 binding sites contribute to transcriptional activation of the rCx40 gene.

A search for potential transcription factor binding sites within the (–175,+85) rCx40 promoter suggested the presence of a total of five GC-box like elements (Fig. 4), four of which appear to be highly conserved between rat, mouse and human. As evidenced in this study, two of these conserved elements (localized within the (–98,–64) and (+53,+87) regions) are involved in transcriptional regulation of rCx40. The core sequences of the GC-box like elements in these regions are composed of GGGTGG and GGGCGG, respectively, although the latter core sequence is replaced by GGGAGG in the corresponding mouse and human sequence. The consensus core sequence of a GC-box element is composed of the hexanucleotides GGGCGG or CCGCCC [25,26], but examples exist of functional GC-box like elements with degenerate core sequences composed of e.g. GGGAGG [30] and GGGTGG [30,31]. In this regard it is interesting to note that the C1 and C2 complex formation in a EMSA is not impaired when the rCx40 (+53,+87) region is replaced by the corresponding mouse or human region (results not shown). This is especially important since our mutational analysis of this Sp1/Sp3 binding site in the rat promoter suggested that this site is functionally more relevant than the one located in the (–98,–64) region. The other two conserved GC-box like elements in the Cx40 promoter, occurring at positions (–113,–104), and (–57,–52), are localized in regions that are more or less implicated in transcriptional regulation by our luciferase reporter assays (Fig. 3) and possess GGGAGG core sequences. The intriguing possibility thus exists that rCx40 promoter activity is mainly determined by the cumulative contribution of distinct GC-box like elements. However, EMSA/antibody supershift experiments and mutational analysis of these sites are clearly required to support that hypothesis. Although the fifth GC-box like element (located in between positions –16 and –11) contains a consensus GC-box core sequence in the rat and mouse, it was absent in the corresponding human sequence.

The level of transcriptional activity obtained with the rCx40 promoter in A7r5, BWEM and N2A cells correlated well with endogenous Cx40 protein levels in these cells, suggesting the presence of cell type-specific determinants within this promoter region. For several reasons it appears unlikely that the rCx40 promoter studied here is sufficient to provide for the cell-type specific expression pattern observed in vivo. First, a relatively low but significant promoter activity was observed in N2A cells, which do not express either Cx40 RNA or protein, suggesting the existence of additional mechanisms/determinants for suppression of this transcriptional activity. Second, precise analysis of the rCx40 regulatory sequences by promoter/reporter assays in A7r5, BWEM and N2A cells (Fig. 2 and 3Go) resulted in almost identical profiles in the Cx40-expressing and -nonexpressing cells, thus supporting the absence of any cell type-specific regulatory elements. Third, two (and potentially more) Sp1/Sp3 binding sites are implicated in rCx40 transcriptional regulation. Sp1 is ubiquitously expressed in many mammalian cell types and believed to provide for a basal level of transcription for a great number of genes, especially housekeeping genes, and therefore Sp1 involvement by itself is unlikely to account for differential expression of Cx40 [27]. Nevertheless, at this point the presence of some cell type-specific determinants within the (–175,+85) region can not be excluded; to resolve this issue a more systematic study is required. Such a study, for instance, will have to include extensive testing of the presented but also larger rCx40 promoter fragments in other Cx40-expressing and -nonexpressing cell types, in particular cells of primary origin. The nature and involvement of each transcription factor binding site implicated in transcriptional regulation of Cx40 by this study needs to be investigated. The potential role of Sp3, which is like Sp1 also found in multiple mammalian cell types, has to be defined more precisely especially since Sp3 has been reported to act as transcriptional activator but also as a repressor of Sp1-mediated activation [27,32]. It is possible, for instance, that the differences observed in transcriptional activity of the rCx40 promoter in A7r5, BWEM and N2A cells are correlated with the level of Sp3 expression in those cells. Ultimately, the results obtained with primary cells and cell lines will have to be tested in an in vivo experimental model system as well, e.g. through the generation of transgenic mice carrying a reporter gene driven by Cx40-promoter regions of particular interest.

Time for primary review 26 days.


   Acknowledgements
 
We are grateful to Dr. Gary L. Engelmann (Loyola University Medical Center, Maywood, IL, USA) for providing the BWEM cells and to Dr. F.E.J. Coenjaerts for helpful discussions. This study was supported by the Netherlands Heart Foundation, grant number M96.001.


   References
Top
 Abstract
 1 Introduction
 2 Methods
 3 Results
 4 Discussion
 References
 

  1. Bruzzone R., White T.W., Paul D.L. Connections with connexins: the molecular basis of direct intercellular signaling. Eur J Biochem (1996) 238:1–27.[Web of Science][Medline]
  2. Goodenough D.A., Goliger J.A., Paul D.L. Connexins, connexons, and intercellular communication. Ann Rev Biochem (1996) 65:475–502.[CrossRef][Web of Science][Medline]
  3. White T.W., Paul D.L. Genetic diseases and gene knockouts reveal diverse connexin functions. Ann Rev Physiol (1999) 61:283–310.[CrossRef][Web of Science][Medline]
  4. O’Brien J., Bruzzone R., White T.W., Al-Ubaidi M.R., Ripps H. Cloning and expression of two related connexins from the perch retina define a distinct subgroup of the connexin family. J Neurosci (1998) 18:7625–7637.[Abstract/Free Full Text]
  5. Haefliger J.A., Bruzzone R., Jenkins N.A., Gilbert D.J., Copeland N.G., Paul D.L. Four novel members of the connexin family of gap junction proteins. Molecular cloning, expression, and chromosome mapping. J Biol Chem (1992) 267:2057–2064.[Abstract/Free Full Text]
  6. Beyer E.C., Reed K.E., Westphale E.M., Kanter H.L., Larson D.M. Molecular cloning and expression of rat connexin40, a gap junction protein expressed in vascular smooth muscle. J Memb Biol (1992) 127:69–76.[Web of Science][Medline]
  7. Gros D.B., Jongsma H.J. Connexins in mammalian heart function. BioEssays (1996) 18:719–730.[CrossRef][Web of Science][Medline]
  8. Van Kempen M.J.A., Vermeulen J.L.M., Moorman A.F.M., Gros D., Paul D.L., Lamers W.H. Developmental changes of connexin40 and connexin43 mRNA distribution patterns in the rat heart. Cardiovasc Res (1996) 32:886–900.[Abstract/Free Full Text]
  9. Davis L.M., Kanter H.L., Beyer E.C., Saffitz J.E. Distinct gap junction protein phenotypes in cardiac tissues with disparate conduction properties. J Am Coll Cardiol (1994) 24:1124–1132.[Abstract]
  10. Van der Velden H.M.W., Van Kempen M.J.A., Wijffels M.C.E.F., et al. Altered pattern of connexin40 distribution in persistent atrial fibrillation in the goat. J Cardiovasc Electrophysiol (1998) 9:596–607.[Web of Science][Medline]
  11. Bastide B., Neyses L., Ganten D., Paul M., Willecke K., Traub O. Gap junction protein connexin40 is preferentially expressed in vascular endothelium and conductive bundles of rat myocardium and is increased under hypertensive conditions. Circ Res (1993) 73:1138–1149.[Abstract/Free Full Text]
  12. Montgomery M.O., Jiao Y., Phillips S.J., et al. Alterations in sheep fetal right ventricular tissue with induced hemodynamic pressure overload. Basic Res Cardiol (1998) 93:192–200.[CrossRef][Web of Science][Medline]
  13. Simon A.M., Goodenough D.A., Paul D.L. Mice lacking connexin40 have cardiac conduction abnormalities characteristic of atrioventricular block and bundle branch block. Curr Biol (1998) 8:295–298.[CrossRef][Web of Science][Medline]
  14. Kirchhoff S., Nelles E., Hagendorff A., Krüger O., Traub O., Willecke K. Reduced cardiac conduction velocity and predisposition to arrhythmia's in connexin40-deficient mice. Curr Biol (1998) 8:299–302.[CrossRef][Web of Science][Medline]
  15. Groenewegen W.A., Van Veen T.A.B., Van der Velden H.M.W., Jongsma H.J. Genomic organization of the rat connexin40 gene: identical transcription start sites in heart and lung. Cardiovasc Res (1998) 38:463–471.[Abstract/Free Full Text]
  16. Seul K.H., Tadros P.N., Beyer E.C. Mouse connexin40: gene structure and promoter analysis. Genomics (1997) 46:120–126.[CrossRef][Web of Science][Medline]
  17. Kimes B.W., Brandt B.L. Characterization of two putative smooth muscle cell lines from rat thoracic aorta. Exp Cell Res (1976) 98:349–366.[CrossRef][Web of Science][Medline]
  18. Olmsted J.B., Carlson K., Klebe R., Ruddle F., Rosenbaum J. Isolation of microtubule protein from cultured mouse neuroblastoma cells. Proc Natl Acad Sci USA (1970) 65:129–136.[Abstract/Free Full Text]
  19. Engelmann G.L., Birchenall-Roberts M.C., Ruscetti F.W., Samarel A.M. Formation of fetal rat cardiac cell clones by retroviral transformation: retention of select myocyte characteristics. J Mol Cell Cardiol (1993) 25:197–213.[CrossRef][Web of Science][Medline]
  20. Sambrook J., Fritsch E.F., Maniatis T. Molecular cloning. A laboratory manual. (1989) Cold Spring Harbor: Cold Spring Harbor Laboratory Press.
  21. Graham F.L., Van der Eb A.J. A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology (1973) 52:456–467.[CrossRef][Web of Science][Medline]
  22. Henderson S.A., Spencer M., Sen A., Kumar C., Siddiqui M.A.Q., Chien K.R. Structure, organization, and expression of the rat cardiac myosin light chain-2 gene. Identification of a 250-base pair fragment which confers cardiac-specific expression. J Biol Chem (1989) 264:18142–18148.[Abstract/Free Full Text]
  23. Li Y., Ross J., Scheppler J.A., Franza B.R. An in vitro transcription analysis of early responses of the human immunodeficiency virus type 1 long terminal repeat to different transcriptional activators. Mol Cell Biol (1991) 11:1883–1893.[Abstract/Free Full Text]
  24. Heinemeyer T., Wingender E., Reuter I., et al. Databases on transcriptional regulation: TRANSFAC, TRRD and COMPEL. Nucl Acids Res (1998) 26:362–367.[Abstract/Free Full Text]
  25. Bucher P. Weight matrix descriptions of four eukaryotic RNA polymerase II promoter elements derived from 502 unrelated promoter sequences. J Mol Biol (1990) 212:563–578.[CrossRef][Web of Science][Medline]
  26. Kadonaga J.T., Jones K.A., Tjian R. Promoter-specific activation of RNA polymerase II transcription by Sp1. Trends Biochem Sci (1986) 11:20–23.[Medline]
  27. Lania L., Majello B., De Luca P. Transcriptional regulation by the Sp family proteins. Int J Biochem Cell Biol (1997) 29:1313–1323.[CrossRef][Web of Science][Medline]
  28. Jackson S.P., Tjian R. O-Glycosylation of eukaryotic transcription factors: implications for mechanisms of transcriptional regulation. Cell (1988) 55:125–133.[CrossRef][Web of Science][Medline]
  29. Jackson S.P., MacDonald J.J., Lees-Miller S., Tjian R. GC box binding induces phosphorylation of Sp1 by a DNA-dependent protein kinase. Cell (1990) 63:155–165.[CrossRef][Web of Science][Medline]
  30. Schanke J.T., Durning M., Johnson K.J., Bennett L.K., Golos T.G. SP1/SP3-binding sites and adjacent elements contribute to basal and cyclic adenosine 3',5'-monophosphate-stimulated transcriptional activation of the rhesus growth hormone-variant gene in trophoblasts. Mol Endocrin (1998) 12:405–417.[Abstract/Free Full Text]
  31. Kudo S., Fukuda M. Transcriptional activation of human leukosialin (CD43) gene by Sp1 through binding to a GGGTGG motif. Eur J Biochem (1994) 223:319–327.[Web of Science][Medline]
  32. Hagen G., Müller S., Beato M., Suske G. Sp1-mediated transcriptional activation is repressed by Sp3. EMBO J (1994) 13:3843–3851.[Web of Science][Medline]

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


This article has been cited by other articles:


Home page
 Am. J. Physiol. Cell Physiol.Home page
A. Makino, O. Platoshyn, J. Suarez, J. X.-J. Yuan, and W. H. Dillmann
Downregulation of connexin40 is associated with coronary endothelial cell dysfunction in streptozotocin-induced diabetic mice
Am J Physiol Cell Physiol, July 1, 2008; 295(1): C221 - C230.
[Abstract] [Full Text] [PDF]

Home page
 Cardiovasc ResHome page
V. L.F. Linhares, N. A.S. Almeida, D. C. Menezes, D. A. Elliott, D. Lai, E. C. Beyer, A. C. Campos de Carvalho, and M. W. Costa
Transcriptional regulation of the murine Connexin40 promoter by cardiac factors Nkx2-5, GATA4 and Tbx5
Cardiovasc Res, December 1, 2004; 64(3): 402 - 411.
[Abstract] [Full Text] [PDF]

Home page
 Cardiovasc ResHome page
B. E.J Teunissen and M. F.A Bierhuizen
Transcriptional control of myocardial connexins
Cardiovasc Res, May 1, 2004; 62(2): 246 - 255.
[Abstract] [Full Text] [PDF]

Home page
 Circ. Res.Home page
W. A. Groenewegen, M. Firouzi, C. R. Bezzina, S. Vliex, I. M. van Langen, L. Sandkuijl, J. P.P. Smits, M. Hulsbeek, M. B. Rook, H. J. Jongsma, et al.
A Cardiac Sodium Channel Mutation Cosegregates With a Rare Connexin40 Genotype in Familial Atrial Standstill
Circ. Res., January 10, 2003; 92(1): 14 - 22.
[Abstract] [Full Text] [PDF]

Home page
 Arterioscler. Thromb. Vasc. Bio.Home page
S. Ausoni and S. Sartore
Cell Lineages and Tissue Boundaries in Cardiac Arterial and Venous Poles : Developmental Patterns, Animal Models, and Implications for Congenital Vascular Diseases
Arterioscler Thromb Vasc Biol, March 1, 2001; 21(3): 312 - 320.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
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 Bierhuizen, M. F.A.
Right arrow Articles by Jongsma, H. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bierhuizen, M. F.A.
Right arrow Articles by Jongsma, H. J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?