Physiol. Genomics AJP: Lung Cellular and Molecular Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Physiol. Genomics 27: 380-390, 2006. First published August 29, 2006; doi:10.1152/physiolgenomics.00145.2006
1094-8341/06 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Table
Right arrow All Versions of this Article:
27/3/380    most recent
00145.2006v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lanahan, A. A.
Right arrow Articles by Simons, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lanahan, A. A.
Right arrow Articles by Simons, M.
Received 7 July 2006; accepted in final form 29 August 2006.
Physiological Genomics 27:380-390 (2006)
1094-8341/06 $8.00 © 2006 American Physiological Society

Synectin-dependent gene expression in endothelial cells

Anthony A. Lanahan 1,*, Thomas W. Chittenden 1,*, Eileen Mulvihill 2, Kimberly Smith 3, Stephen Schwartz 2 and Michael Simons 1

1 Angiogenesis Research Center and Section of Cardiology, Departments of Medicine and Pharmacology and Toxicology, Dartmouth Medical School, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
2 Department of Pathology, University of Washington School of Medicine, Seattle, Washington
3 Department of Pharmacology, University of Washington School of Medicine, Seattle, Washington


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Synectin (GIPC1), a receptor scaffold protein, has been isolated by our laboratory as a syndecan-4 cytoplasmic domain binding partner that regulates important aspects of cell motility (Gao Y, Li M, Chen W, Simons M. J Cell Physiol 184: 373–379, 2000; Tkachenko E, Elfenbein A, Tirziu D, Simons M. Circ Res 98: 1398–1404, 2006). Moreover, synectin plays a major role in arterial morphogenesis and in growth factor signaling in arterial endothelial cells by regulating Rac1 activity (Chittenden TW, Claes F, Lanahan AA, Autiero M, Palac RT, Tkachenko EV, Elfenbein A, Ruiz de Almodovar C, Dedkov E, Tomanek R, Li W, Westmore M, Singh J, Horowitz A, Mulligan-Kehoe MJ, Moodie KL, Zhuang ZW, Carmeliet P, Simons M. Dev Cell 10: 783–795, 2006). The present study was carried out to characterize changes in synectin-dependent gene expression induced by homozygous disruption of the gene in endothelial cells. Using a combination of suppression subtraction hybridization and high throughput microarray technology, we have identified aberrant biological processes of transcriptional regulation in synectin–/– primary endothelial cells including abnormal basal regulation of genes associated with development, cell organization and biogenesis, intracellular tracking, and cell adhesion. Analysis of gene expression following FGF2 treatment demonstrated significant abnormalities in transcription, cytoskeletal organization and biogenesis, and protein modification and transport in synectin–/– compared with synectin+/+ endothelial cells. These results confirm synectin involvement in FGF2-dependent signal transduction and provide insights into synectin-dependent gene expression in the endothelium.

angiogenesis; functional genomics


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
FIBROBLAST GROWTH FACTORS have long been thought to stimulate intracellular signaling by binding to high affinity receptors (FGF receptors 1–4) that belong to the classic transmembrane tyrosine kinase family of receptors (19). More recent studies showed that syndecan-4, a nontyrosine kinase transmembrane protein also functions as an FGF receptor (15, 25). While studies of the syndecan-4 signaling cascade focused on its activation of PKC{alpha}, other possibilities are now emerging as well (10, 21). Particularly intriguing is the role played by synectin that has been isolated as a syndecan-4 binding partner (13).

Synectin is a single PDZ domain containing protein that interacts with a wide variety of plasma membrane and cytoplasmic molecules including, in addition to syndecan-4, G protein regulator of signaling interacting protein, cytoplasmic G{alpha} interacting protein (regulator of G protein signaling) (11), semaphorin 4c (27), Glut-1 glucose transporter (5), neuropilin-1 (6), {alpha}5 and {alpha}6 integrins (23) among others.

We previously demonstrated a PDZ-dependent synectin-syndecan-4 interaction (13), suggesting that synectin may be involved in FGF2 signal transduction since FGF2 signaling in endothelial cells (ECs) requires an intact syndecan-4 PDZ binding domain (15). Moreover, we recently showed that disruption of synectin gene expression in mice results in significant abnormalities in arterial EC migration, proliferation, and in vivo arterial branching morphogenesis resulting from abnormal cellular activation of Rac1 as well as impaired response to FGF2 signaling (8). To further study the molecular role of synectin in FGF2 signaling, we used suppression subtraction hybridization (SSH) to analyze immediate early FGF2-dependent gene expression in primary pulmonary ECs derived from synectin+/+ or –/– mice.

Immediate early genes are important mediators of growth factor responsiveness and represent the initial genomic response to growth factor stimulation. Subsequent waves of gene expression downstream of the initial immediate early gene response serve to amplify the initial response and further exaggerate any initial difference in the genomic response between synectin–/– and wild-type (WT) cells.

Suppression subtraction hybridization is a powerful approach for identifying differences in gene expression between two different mRNA populations (12). It is a complementary approach to gene expression analysis using microarrays that allows for the identification of differential expression of low-abundance RNAs and identification of RNAs not present on an array (7). We have combined SSH with differential screening (DS) and DNA microarrays to identify both basal and FGF2-induced differences in gene expression between synectin WT and knockout (KO) ECs.

Using this combination of approaches we have identified differences in gene expression between synectin+/+ and –/– ECs before and after FGF2 stimulation that provide new insights into synectin function.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Isolation of murine ECs from lung and cardiac tissue.
Primary ECs from adult mice hearts and lungs were isolated as previously described (1) and stimulated with 50 ng/ml FGF2 (30). In brief, the hearts and lungs of four mice were harvested, minced finely with scissors, and then digested in 25 ml of collagenase 0.2% (wt/vol) at 37°C for 45 min. The crude cell preparation was pelleted and resuspended in Dulbecco’s phosphate-buffered saline. The cell suspension was incubated with platelet endothelial cell adhesion molecule-1-coated beads (IgG Dynal beads from Dynal, Great Neck, NY) at room temperature for 10 min. The bead-bound cells were recovered, washed with DMEM-20% FBS, suspended in 12 ml of complete culture medium (DMEM containing 20% fetal calf serum, 100 µg/ml heparin, 100 µg/ml endothelial cell growth factor growth supplement), and then plated in a gelatin-coated 75-cm2 tissue culture flask.

SSH and generation of subtracted libraries.
SSH was performed with the PCR-Select cDNA Subtraction Kit (Clontech). Starting material consisted of 2 µg of poly(A)+ RNA from FGF2-treated (1/2, 1, 2 h)-synectin+/+ and –/– lung ECs. cDNA was synthesized, digested with RsaI, and ligated with two different adapters. Subtracted libraries were generated with adapted synectin+/+ cDNA subtracted against synectin–/– mRNA and adapted synectin–/– cDNA subtracted against synectin–+/+ mRNA. The first round of hybridization was performed at 68°C for 8 h, and the second round of hybridization was incubated at 68°C for 22 h. After two rounds of subtractive hybridization, two rounds of PCR were performed to amplify differentially expressed cDNA fragments. The PCR products were digested with NotI/EagI, cloned into pBluescript(SK), and transformed into Escherichia coli.

Colony PCR and differential screening.
A portion of each SSH library was plated on Luria-Bertani (LB) ampicillin plates with X-GAL, and "white" insert-containing colonies were picked at random for further screening. Individual colonies were picked and gridded onto an LB ampicillin plate, and then the remaining bacteria were used for colony PCR using the M13 forward and reverse primers. PCR conditions were: 95°C/10 min, 40x (94°C/30 s, 64°C/1 min, 72°C/2 min), 72°C/5 min. PCR products were denatured, neutralized, and slot-blotted onto duplicate nitrocellulose filters. Filters were prehybridized overnight at 65°C in 5x SSC, 5x Denhardt solution, 50 mM NaPO4, 10 mM EDTA, 0.2% SDS, and 100 µg/ml sheared sonicated salmon sperm DNA. 32P-dCTP-labeled double-stranded cDNA probes were prepared from synectin–/– and +/+ cDNA by random nonamer labeling (Stratagene). Probes were denatured and added to freshly made hybridization buffer, and the filters were hybridized for 40 h. Filters were washed twice at room temperature with 2x SSC/0.2% SDS, twice at 65°C with 0.5x SSC/0.2% SDS, and exposed to X-ray film. After exposure, clones showing differential hybridization to WT or KO cDNA were expanded for further analysis.

Microarray data normalization and processing and statistical analysis.
PCR products from the subtraction of KO cDNA against an excess of WT cDNA were labeled with Cy3, while PCR products from the subtraction of WT cDNA against an excess of KO cDNA were labeled with Cy5 using random priming and Exo-Klenow kit (Invitrogen). Equimolar quantities of the labeled probes were combined and hybridized by the Specialized Cooperative Centers Program in Reproductive Research Array Facility (http://depts.washington.edu/popctrma/index.shtml) to an "in-house" whole genome mouse cDNA array created by combining the 15 k and 7.4 k mouse cDNA libraries generated at the NIA (http://lgsun.grc.nia.nih.gov/cDNA/). Dye swapping was performed on a second cDNA chip to control for labeling bias. The array chips were created using a Genemachines Omnigrid arrayer with a 16 pen print head. The slides were scanned with a Axon GenePix 4000a scanner, and the images were analyzed using Silicon Genetics GeneSpring software. All slides were normalized using a LOWESS function to address intensity-dependent dye biases. The data from the dye flip experiments were merged, and an Excel spreadsheet report was created including the normalized ratio of synectin –/– vs. synectin+/+, the treatment average for each spot, the control average for each spot, the t-test P value of each synectin –/– vs. synectin+/+ spot, and the annotation. Because of both the limited sample size related to this dataset and the increased differential expression associated with SSH enrichment procedures, a conservative significance level of fourfold differential expression, and P < 0.05 was applied to this dataset. Differentially regulated genes were then classified according to biological process, cellular component, and molecular function with Princeton University Generic GO::TermFinder software (4).

Northern blotting.
Synectin –/– and +/+ ECs were grown to confluence and then starved overnight in 0.5% FBS. After starvation, cells were stimulated with 50 ng/ml FGF2 for various times, and the cells were rinsed with PBS, then pelleted, flash frozen with liquid nitrogen, and stored at –80°C until RNA was isolated. Total RNA was isolated by Polytron disruption with the RNeasy kit (Qiagen). For Northern blot analysis, 10 µg of RNA from each time point were electrophoresed on a 1.4% formaldehyde-agarose gel and then transferred to nitrocellulose. The nitrocellulose filter was prehybridized overnight at 45°C in 5x SSPE (150 mM sodium chloride, 10 mM sodium phosphate, and 1 mM Na 2 EDTA) pH 7.4, 50% formamide, 5x Denhardt solution, 0.1% SDS, 0.1% NaPiPi, and 0.2 mg/ml boiled sonicated salmon sperm DNA. The filter was then hybridized overnight at 45°C with a 32P-labeled cDNA probe derived from the gene indicated in freshly prepared hybridization buffer. The filter was washed twice in 2x SSC/0.2% SDS at room temperature and twice in 1x SSC/0.2% SDS at 60°C before autoradiography.

Quantitative PCR validation.
For quantitative PCR (qPCR) analysis, total EC RNA was isolated, and first-strand cDNA was synthesized as previously described. PCR amplification was carried out using 20 ng of cDNA per reaction with gene-specific TaqMan-based assays (Applied Biosystems) for the following genes on a GeneAmp 5700 sequence detection system (Applied Biosystems): cyclin G1 (CCNG1, NM_009831.2; Mm00438084_m1), immediate early response 3 gene (IER3, NM_133662.1; Mm00519290_g1), Ras and Rab interactor 3 (RIN3, NM_177620.2; Mm00617220_m1), tissue inhibitor of metalloproteinase 3 (TIMP3, NM_011595.1; Mm00441826_m1), and normalized to cytoplasmic B-actin (ACTB, NM_007393.1; Mm00607939_s1). Statistical significance was assessed in R (version 2.1.1) with one-sample t-tests.

Cell adhesion assay.
Synectin–/– and +/+ ECs were grown to confluence, detached with cell dissociation solution, rinsed in serum-free media, and plated at a density of 50,000 cells/well in serum-free media on fibronectin coated 12-well plates. WT and KO cells for each time point were plated in triplicate and allowed to attach at 37°C. After 30 and 60 min, cells were washed twice with PBS and then fixed with 3.7% paraformaldehyde for 10 min at room temperature. Cells were then washed three times with PBS; adherent cells were stained with Coomassie blue for 1 h at room temperature and then rinsed three times with PBS. Adhesion for each cell type and time point was evaluated by counting stained cells in six separate fields under an inverted optical microscope with a x20 objective.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
SSH and DS analysis of differential FGF2 gene regulation in synectin+/+ and synectin–/– ECs.
To identify potential differences in the response of primary ECs from synectin+/+ and synectin–/– mice to FGF2, we performed SSH as shown in the flow diagram (Fig. 1). To identify clones expressed at elevated levels in FGF2-treated synectin+/+ ECs, their cDNA was subtracted against an excess of FGF2 treated synectin–/– poly(A)+ RNA. Conversely, sequences expressed at higher levels in synectin–/– cells were identified by subtracting their cDNA against an excess of synectin+/+ poly(A)+ RNA. Additionally mRNA was isolated from the synectin+/+ and synectin–/– cells following overnight starvation, and similar subtractions to those described above were performed to identify basal differences in gene expression.


Figure 1
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 1. Schematic representation of differential cloning approach used to generate and screen forward/reverse suppression subtraction hybridization (SSH) libraries. WT, wild type; KO, knockout.

 
SSH generated two distinct sets of differential expression libraries: baseline differences between synectin+/+ and –/– cells and FGF2-induced gene expression in both types of ECs. Since the SSH technique is an enrichment for differentially expressed sequences and has a background of nondifferential clones when used alone, it leads to a significant number of false positive clones in the absence of a secondary screen. To avoid this problem, we incorporated a differential screen of the SSH libraries before further analysis (Fig. 1). In this step, colonies were picked at random from both the WT and KO subtracted libraries, and their inserts were isolated by colony PCR using primers flanking the cDNA insert. The PCR products were then slot-blotted in duplicate as described in MATERIALS AND METHODS and were hybridized with cDNA from WT and synectin–/– cells to identify sequences expressed at different levels. Using unsubtracted cDNA further reduced identification of false positive clones, which is sometimes a problem when subtracted PCR products are used as probes. Additionally the use of unsubtracted cDNA for the DS step results in a direct readout of the relative abundance of the different cDNAs and their difference between cells. Thus differences in the strength of hybridization of a given cDNA with probes generated from synectin+/+ and –/– cells directly reflect the differences in expression level between these cells.

Following the DS step, 16% of all the synectin+/+ clones and 24% of all the synectin–/– clones showed a differential signal in our differential screen. This agrees well with the results from a recent study showing that only one-third of the clones in an SSH library represent true differentially expressed RNAs (7). Roughly half of the clones arrayed showed a detectable signal, those showing no signal likely reflect low-abundance mRNAs that cannot be detected with an unsubtracted probe.

DNA sequence analysis of clones identified by DS.
The initial differential screen of 192 clones from the synectin+/+ SSH library (clones expressed in synectin+/+ but not in synectin–/– cells) identified 28 clones that showed a differential signal; similarly screening of 192 clones from the synectin–/– library (clones expressed in synectin–/– but not in synectin+/+ cells) identified 46 clones showing a differential signal. These 74 clones were then sequenced, and their identities were established using online genomic databases (Supplemental Table 1; the online version of this article contains supplemental material). The differentially expressed clones represented a variety of known cellular proteins including nuclear, cytoplasmic, membrane, and secreted proteins in both WT and KO subtracted cDNA populations. In addition, several novel sequences were identified in both the WT- and the synectin–/–-specific SSH libraries. Several of the genes were isolated multiple times, including clones representing the extracellular matrix molecules fibronectin 1 and thrombospondin 1, as well as the putative RNA helicase Ddx3x. Also among the clones identified is the known immediate early gene Mig-6. Mig-6 has previously been shown to be regulated by FGF2 (9), and its identification in this screen points to disruption of the immediate early response to FGF2 in synectin–/– cells.


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

  		Table 1. Generic GO::TermFinder analysis of basal primary microvascular endothelial cells

 
Microarray analysis using SSH probes.
SSH performed as described above resulted in four populations of enriched cDNA species that were pairwise applied to microarrays containing 22,400 cDNAs (basal and FGF2-stimulated: synectin+/+ enriched vs. synectin–/– enriched). The first pair of probes was designed to identify differences in basal gene expression between synectin+/+ an synectin–/– cells. This set contained one cDNA enriched for sequences expressed more highly in synectin–/– cells before stimulation, and the second was enriched for those expressed in unstimulated synectin+/+ cells. Analysis of microarrays as described in MATERIALS AND METHODS show that in the basal state ~1.5% of the 22,400 cDNAs were expressed at higher levels in synectin–/– enriched vs. synectin+/+ enriched cells, and 1.3% were expressed at higher levels in synectin+/+ vs. synectin–/– cells.

Similarly, the second set of probes contained one enriched for sequences more abundant in FGF2-treated synectin KO cells and a second enriched for FGF2 treated WT cells. In the FGF2-treated experiment, 2.5% of the cDNAs were expressed at higher levels in KO vs. WT cells, and 2.5% were expressed at higher levels in synectin+/+ vs. synectin–/– cells.

Gene ontology analysis of microarray results.
We used microarray analysis to assess differential expression between synectin+/+ and –/– cDNA libraries generated from the SSH procedure. Using fourfold difference (P < 0.05) in expression as a cutoff, we identified 523 differentially expressed probes with known gene names at baseline, while the FGF2 stimulation resulted in a probe set comprising 977 differentially expressed transcripts with known gene names. Both data sets were then used for subsequent gene ontology (GO) analysis.

GO analysis (Table 1) shows that after correction for multiple hypothesis testing, there are statistically significant enrichments of transcripts within the basal dataset displaying protein and nucleotide binding as well as enzyme regulatory activities. These differences in molecular function are also observed as significant enrichments of transcripts involved in the biological processes of development, cell organization and biogenesis, cellular localization and protein and biopolymer metabolism, and cell adhesion. Interestingly, synectin null cells have decreased levels of PIK3R1, MAP2K1, FZD4, DVL2, DAAM1, TIAM1, and TIAM2 with concomitantly increased expression levels of APC, ARPC1A, and CDC42EP4 (Table 2), a profile consistent with abnormal regulation of Rac1. Moreover, the apparent coordinated dysregulation of specific genes families, including the kinesins, tubulins, RABs, myosins, and suppressor cytokine signaling genes indicate that abnormalities in transcriptional regulation of cellular organization and biogenesis did not arise by chance (Table 2).


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

  		Table 2. Basal biological process transcripts of the generic GO::TermFinder analysis

 
Finally, significant cellular component terms, which include the cytoplasm, intracellular membrane-bound organelle, endoplasmic reticulum, and extracellular matrix, support appropriate compartmentalization of the observed biological processes and molecular functions within the basal transcript list. A partial tabulation of differential basal transcripts that include these GO terms is shown in Table 2.

GO analysis (Table 3) of the FGF2-induced probe set indicates, after correction for multiple hypothesis testing, that there are a much greater number of statistically significant enrichments of transcripts representing a larger number of aberrant GO terms than in the basal dataset, thus indicating the importance of synectin in FGF2-related processes in ECs. The FGF2-induced aberrant enhancements in protein binding, translation regulator, and enzyme activities are also observed in significant abnormalities in biological process terms from cell cycle to RNA metabolism to protein modification and transport. Synectin null cells show increased levels of ARPC1A and ARPC2 with concomitant decreased levels of PLC{gamma} and JAK2, supporting the basal findings of apparent abnormal regulation of Rac1 as well as abnormal activation of Rac1 in response to FGF2 (8). Moreover, the coordinated dysregulation of the specific gene families identified at baseline, which comprise the kinesins, tubulins, RABs, myosins, and suppressor cytokine signaling genes, are also abnormally regulated after FGF2 stimulation, thus further suggesting that abnormalities in transcriptional regulation of cellular organization and biogenesis as well as transport did not arise by chance (Table 4).


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

  		Table 3. Generic GO::TermFinder analysis of FGF2-induced primary microvascular endothelial cells

 

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

  		Table 4. FGF2 biological process transcripts of the generic GO::TermFinder analysis

 
Lastly, significant FGF2-induced cellular component terms, which comprise the nucleus, cytoplasm, cytosol, mitochondrion, cytoskeleton, Golgi apparatus, and ribosomes, support appropriate compartmentalization of the observed biological processes and molecular functions within FGF2-induced transcript list. A partial tabulation of differential basal transcripts that include these GO terms is shown in Table 4. Complete basal and FGF2-induced GO analyses are included in the supplemental results section.

Validation of differential gene expression.
To confirm differential expression results as determined by SSH and microarray analysis, we performed Northern analysis of expression of several representative genes in synectin+/+ and–/– primary pulmonary endothelial cells at baseline and following FGF2 treatment (Fig. 2). In all cases, Northern blotting confirmed the differential nature of gene expression. In two cases, the levels of gene expression were similar at baseline between synectin+/+ and –/– cells with FGF2 inducing differential response (DEAD/H and thrombospondin-1), while in two cases there were significant baseline differences that were further amplified by FGF2 (Mig-6 and Hook2). Thus in all cases examined, Northern analysis confirmed the results seen by SSH/DS.


Figure 2
View larger version (21K):
[in this window]
[in a new window]

  		Fig. 2. Northern analysis of differentially expressed genes in lung endothelial cells. Cells were starved overnight in 0.5% serum and stimulated with 50 ng/ml FGF2 for the times indicated. Total RNA (10 µg) was electrophoresed, transferred to nitrocellulose, and probed with 32P-labeled cDNA fragment from the gene indicated. Following hybridization the blots were scanned using the Typhoon 9410 Variable Mode Imager (Molecular Dynamics), and RNA levels were quantitated using ImageQuant software and normalized to the amount of 18S rRNA in each lane.

 
Additionally four clones identified as differentially expressed on the microarrays were checked for differential expression by quantitative PCR (Fig. 3). Clones selected for qPCR validation came from both probe sets and represent both basal and FGF2 regulated differentially expressed genes. Rin3, a novel Rab5 guanine exchange factor, was identified by microarray analysis to be expressed at higher basal levels in synectin–/– cells than synectin+/+, and this is confirmed by PCR. Similarly, Ier3 (gly96/IEX-1) was identified by microarray to be expressed at higher levels in synectin–/– cells than synectin+/+ following FGF2 stimulation, and PCR confirms this. Finally, cyclin G1 and TIMP3 were shown by array to be expressed at higher levels in synectin+/+ cells following FGF2 stimulation (confirmed by PCR).


Figure 3
View larger version (9K):
[in this window]
[in a new window]

  		Fig. 3. Quantitative real-time PCR analysis of differentially expressed genes in lung endothelial cells. Cells were starved overnight in 0.5% serum and stimulated with 50 ng/ml FGF2 for the times indicated. "Pooled" refers to sample of RNA harvested 30, 60, 120 min after FGF2 stimulation. Gene expression shown normalized to cytoplasmic B-actin. *P < 0.05.

 
Validation of functional differences in cell adhesion.
GO analysis of the basal microarray results (Tables 1, 2) and array analysis of FGF2-induced gene expression (Supplemental Table 1) identified perturbations in the gene expression of several cell adhesion and extracellular matrix proteins. Differences in the expression levels of three of these genes, fibronectin 1, thrombospondin 1, and TIMP3, were confirmed by either Northern blotting (Fig. 2) or qPCR (Fig. 3). Changes in the levels of these proteins and the others identified would be expected to result in differences in cell adhesion between WT and KO cells. To determine whether the genomic differences identified correlated with a functional change in cell adhesion, we performed adhesion assays using synectin–/– and +/+ ECs. We find that at both 30 and 60 min after plating, synectin –/– cells are significantly more adherent than synectin+/+ cells (Fig. 4, A and B). Thus differences in adherence predicted by our microarray analysis are confirmed by the cell adhesion assay. Presumably alterations in the expression of cell adhesion molecules such as Ncam1 and Icam1 in combination with altered expression of extracellular matrix proteins, such as fibronectin 1, thrombospondin 1, and TIMP3, are responsible for the differences in cell adhesion seen.


Figure 4
View larger version (43K):
[in this window]
[in a new window]

  		Fig. 4. Quantitative analysis of endothelial cell adhesion. A: significant increase in endothelial cell adhesion at both 30 and 60 min in synectin–/– cells. Images were acquired with an inverted optical microscope at x4. B: quantification of relative endothelial cell adhesion at both 30 and 60 min in synectin +/+ and –/– cells. *P < 0.05, **P < 0.01.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
Using SSH, DS, and microarrays, we demonstrated that there are differences in both basal and FGF2-induced gene expression in ECs lacking synectin compared with controls. GO analysis of both basal and FGF2 differences in gene expression shows that the observed alterations fall into a number of cellular pathway categories that are phenotypically altered in synectin–/– ECs such as cell proliferation, migration, and angiogenesis (8). These results also demonstrate that synectin is an integral part of the FGF2 signaling cascade and begin to unravel the role played by this protein in FGF2-dependent signaling.

We chose the SSH/DS approach in combination with microarrays to explore this facet of FGF2 signaling because it is a rapid and proven technique for identifying differences in gene expression. Subtractive hybridization is a powerful technique for comparing two populations of mRNA and identifying clones expressed in one population but not the other. There are a variety of techniques available for the study of global changes in gene expression in a biological system. These include differential display (2), representational difference analysis (26), linker capture subtraction (29), and serial analysis of gene expression (18) among others. All have proven successful in identifying differentially expressed sequences but possess drawbacks. A common limitation to all of these approaches is a relative inability to isolate rare transcripts because the disproportionate concentrations of high- versus low-abundance differentially expressed genes are maintained in the subtraction and/or analysis. This minor fraction is of particular interest because it contains transcripts for many regulatory proteins. In contrast, SSH combines normalization and subtraction in a single procedure.

The normalization step equalizes the abundance of cDNAs, and the subtraction step excludes common sequences between the two populations being analyzed. Thus the subtracted library generated is normalized in terms of abundance of different cDNAs. Rare sequences have been reported to be enriched >1,000-fold in a single round of subtractive hybridization. This increases the probability of isolating low-abundance differentially expressed cDNAs and simplifies analysis of the subtracted library. Employing a differential screen of the SSH libraries generated makes it possible to limit the analysis of clones in the library to only those that are truly differentially expressed. This eliminates the presence of false positive in the population of clones analyzed. Thus SSH coupled with DS allows for the efficient identification of rare differentially expressed mRNAs as well as more abundant mRNAs.

DNA microarrays represent a complementary approach to SSH. One important limitation of the array-based analysis of gene expression is its relative insensitivity to low-abundance transcripts with up to 25–30% of the sequences identified by SSH not identified by microarrays (7). Thus, SSH may be preferable to microarrays for identifying novel, differentially regulated sequences. Furthermore, SSH requires as little as 25 ng of (polyA)+ RNA or 50 ng of total RNA, which is less than the minimum of 10–20 µg of total RNA required for a typical microarray experiment. Together these considerations make SSH a nice complementary approach to arrays.

Combining SSH with whole genome microarray analysis produces a comprehensive approach that allows us to identify differences in gene expression amongst even the lowest-abundance RNAs due to the enhanced sensitivity of the probe generated. Also by comparing basal and FGF2 differences in gene expression, we can identify those differences in gene expression that are a consequence of the loss of synectin expression and those differences that reflect synectin dependent alterations in the responsiveness of cells to FGF2.

Treatment of cells or tissue with FGF2 triggers a program of gene expression that ultimately leads to the changes in cellular phenotype associated with growth factor stimulation. Disruption of this program would then be expected to result in alterations in cellular responses to the growth factor stimulation. We hypothesized that synectin, a single PDZ domain containing protein that has been identified as a cytoplasmic binding partner for an FGF2 receptor syndecan-4, plays an important role in FGF2 signal transduction. Hence, the present study was designed to assess the effect of homozygous disruption of the synectin gene on FGF2-induced gene expression. We have focused on identifying changes in the expression of immediate early genes in response to FGF2 because regulation of this class of genes occurs within 2 h of growth factor stimulation and these changes represent the initial genomic response to extracellular stimuli. Several differences in phenotype seen between synectin+/+ and synectin–/– mice include a profound aberration in arterial morphogenesis and abnormal responses of primary ECs to FGF2 among others. Additionally, ECs isolated from these animals demonstrate a decrease in both proliferation and migration responses to FGF2 stimulation as well as an abnormal pattern of Rac1 activation (8).

We find, using GO analysis of both basal and FGF2-induced gene expression, that synectin silencing produces significant abnormal transcriptional regulation of biological processes relating to development, transcription, cell organization and biogenesis, protein transport, cell adhesion, cell cycle, and cytoskeletal regulation among others. Analysis of differential expression of specific transcripts, comprising the enrichment of these aberrant biological processes, reveals significant abnormal expression of genes involved with Rac1 regulation, thus providing additional evidence of the importance synectin in Rac1-dependent angiogenic processes. Of particular interest to the synectin null phenotype is the abnormal expression of eight RAB family members including a 6.5-fold reduction in RAB7 expression, which could account, in part, for the aberrant localization of activated Rac1 in primary arterial endothelial synectin–/– cells. Another observation of interest is a 46-fold decrease in IQGAP1 gene expression in synectin null ECs (22). This may also account for the apparent suppression of the majority of transcripts involved in cycle cell regulation and provide a plausible explanation for the decreased proliferative ability of synectin null ECs, a process partly regulated by Rac1 and requiring cytoskeletal changes mediated by IQGAP. Moreover, the increased expression of myosin 6 and kinesin family member 1B, proteins shown to directly interact with synectin, provides further evidence of abnormal actin filament and microtubule-based vesicular transport (5).

In the case of cell proliferation, the array results and GO analysis (Tables 3 and 4) identify FGF2-dependent genomic changes that functionally correlate with a reduction in proliferation observed when synectin KO ECs are stimulated with FGF2 (8). Interestingly, the important cell cycle regulators cyclin D2 and G1 are both expressed at higher levels in synectin WT cells following FGF2 stimulation, whereas the growth inhibitory cyclin-dependent kinase inhibitor 2D is expressed at higher levels in KO cells. In the case of cyclin G1, this increase in expression has been confirmed by qPCR. Furthermore, the membrane molecule growth arrest-specific 1 (Gas1) is expressed at elevated levels in synectin–/– relative to synectin+/+ cells. Gas1 blocks entry into the S phase and prevents cycling of normal and transformed cells and is required for normal regulation of FGF10 and FGF8 signaling in the mesenchyme (17). It seems reasonable that an increase in the two growth-associated cyclins in synectin+/+ cells coupled with an increase in the growth inhibitory CDK inhibitor 2D and the growth arrest molecule Gas1 in KO cells in response to FGF2 might underlie the difference in proliferation rates in response to FGF2 reported for synectin+/+ and –/– ECs (8).

The SSH analysis also identified a number of cytoplasmic, membrane, and secreted proteins that play a role in cell migration and adhesion. Array and GO analysis predicted a difference in adhesion between synectin WT and KO ECs that was experimentally validated. One of the membrane proteins altered is E-cadherin, which is more abundant in synectin–/– cells, a finding consistent with increased adhesion of these compared with WT cells. Interestingly, the E-cadherin/catenin system is a target of the FGF/FGFR system, and its loss of function is thought to contribute to progression in cancer by increasing proliferation and migration (28). Icam1 gene is also expressed at higher levels in synectin–/– cells, and the increased levels of this cell adhesion molecule could also result in increased cell adhesion. Secreted proteins expressed at higher levels in synectin+/+ cells include fibronectin, thrombospondin 1, and TIMP3. These proteins act in concert to regulate the extracellular matrix. Furthermore, the basement membrane protein entactin (Nid1), which interacts with integrin receptors to mediate cell adhesion, spreading and motility, is more abundant in synectin–/– than synectin+/+ cells. Differences in expression of any one of these matrix/basement membrane proteins can result in changes in cell adhesion and migration. Changes in the levels of all of these proteins likely contribute to some of the observed differences in cell adhesion, spreading and migration (8) between WT and synectin–/– ECs.

Also among the changes in gene expression we have detected in the present study are differences in the expression of genes encoding nuclear proteins that alter transcription by either directly binding DNA/RNA or modulating the activity of the transcriptional apparatus. The predicted transcriptional regulator Arid2 is more abundant in synectin+/+ than synectin–/– cells. A member of the DEAD/H box family of proteins is also more abundant in synectin+/+ cells. Members of this family of proteins include RNA helicases, which may play a role in maintaining correct expression levels of early growth response (EGR) target genes through their interaction with EGR transcription factors (3). Another transcriptional regulator that is expressed at higher levels in synectin+/+ than in synectin–/– cells is the protein kinase Hipk3, which transduces growth regulatory signals and is involved in regulation of Fas activity (20).

Finally, emerging evidence indicates that synectin is involved in endocytic trafficking through an interaction with myosin 6 (14). Interestingly, basal expression of myosin 6, which directly binds synectin, is upregulated in synectin KO cells. This may represent an attempt at compensation for reduced synectin expression. Additionally, it is interesting to note that one of the mRNAs elevated in synectin–/– cells is the mammalian homolog of Drosophila Hook2. Hook proteins are cytosolic coiled-coil proteins that contain conserved NH2-terminal domains, which attach to microtubules and more divergent COOH-terminal domains, which mediate binding to organelles (16). The Drosophila Hook protein is a component of the endocytic compartment and a negative regulator of endocytic trafficking. It is possible that elevated levels of Hook in synectin–/– cells represent a compensatory effect resulting from the absence of synectin in the endocytic pathway. In fact, as shown by Northern blotting, levels of Hook mRNA are over fivefold higher in synectin–/– than synectin+/+ cells even in the absence of any FGF2 treatment. FGF2 stimulation results in no induction of Hook in synectin+/+ cells, whereas in synectin–/– cells levels increase 20-fold over baseline. This would be consistent with an upregulation of Hook in synectin–/– cells as a compensatory response to the absence of synectin.

In summary, results from this study show that homozygous deletion of the synectin gene in primary ECs leads to substantial changes in gene expression in response to FGF2 stimulation. We conclude, therefore, that synectin plays an important role in FGF2 signal transduction affecting multiple cellular functions. These studies provide a framework for future experiments aimed at determining which of the genomic changes have a causal role in the functional alterations reported in synectin knockout (8).


   GRANTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
Supported in part by National Heart, Lung, and Blood Institute Grants HL-053793 and HL-062289 (M. Simons) and American Heart Association Grant 0525982T (T. W. Chittenden).


   ACKNOWLEDGMENTS
 
Present address for T. W. Chittenden: Functional Genomics & Computational Biology Group, Dana-Farber Cancer Inst., Harvard School of Public Health, Boston, MA (thomas{at}jimmy.harvard.edu).


   FOOTNOTES
 
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).

Address for reprint requests and other correspondence: M. Simons, Section of Cardiology, Dartmouth-Hitchcock Medical Center, 1 Medical Center Dr., Lebanon, NH 03756 (e-mail: michael.simons{at}dartmouth.edu).

* A. A. Lanahan and T. W. Chittenden contributed equally to this work. Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Allport JR, Lim YC, Shipley JM, Senior RM, Shapiro SD, Matsuyoshi N, Vestweber D, Luscinskas FW. Neutrophils from MMP-9- or neutrophil elastase-deficient mice show no defect in transendothelial migration under flow in vitro. J Leukoc Biol 71: 821–828, 2002.[Abstract/Free Full Text]
  2. Bartlett JM. Differential display: a technical overview. Methods Mol Biol 226: 217–224, 2003.[Medline]
  3. Bork P, Koonin EV. An expanding family of helicases within the ’DEAD/H’ superfamily. Nucleic Acids Res 21: 751–752, 1993.[Free Full Text]
  4. Boyle EI, Weng S, Gollub J, Jin H, Botstein D, Cherry JM, Sherlock G. GO::TermFinder–open source software for accessing Gene Ontology information and finding significantly enriched Gene Ontology terms associated with a list of genes. Bioinformatics 20: 3710–3715, 2004.[Abstract/Free Full Text]
  5. Bunn RC, Jensen MA, Reed BC. Protein interactions with the glucose transporter binding protein GLUT1CBP that provide a link between GLUT1 and the cytoskeleton. Mol Biol Cell 10: 819–832, 1999.[Abstract/Free Full Text]
  6. Cai H, Reed RR. Cloning and characterization of neuropilin-1-interacting protein: a PSD-95/Dlg/ZO-1 domain-containing protein that interacts with the cytoplasmic domain of neuropilin-1. J Neurosci 19: 6519–6527, 1999.[Abstract/Free Full Text]
  7. Cao W, Epstein C, Liu H, DeLoughery C, Ge N, Lin J, Diao R, Cao H, Long F, Zhang X, Chen Y, Wright PS, Busch S, Wenck M, Wong K, Saltzman AG, Tang Z, Liu L, Zilberstein A. Comparing gene discovery from Affymetrix GeneChip microarrays and Clontech PCR-select cDNA subtraction: a case study. BMC Genomics 5: 26, 2004.[CrossRef][Medline]
  8. Chittenden TW, Claes F, Lanahan AA, Autiero M, Palac RT, Tkachenko EV, Elfenbein A, Ruiz de Almodovar C, Dedkov E, Tomanek R, Li W, Westmore M, Singh J, Horowitz A, Mulligan-Kehoe MJ, Moodie KL, Zhuang ZW, Carmeliet P, Simons M. Selective regulation of arterial branching morphogenesis by synectin. Dev Cell 10: 783–795, 2006.[CrossRef][Web of Science][Medline]
  9. Chung HA, Hyodo-Miura J, Kitayama A, Terasaka C, Nagamune T, Ueno N. Screening of FGF target genes in Xenopus by microarray: temporal dissection of the signalling pathway using a chemical inhibitor. Genes Cells 9: 749–761, 2004.[Abstract/Free Full Text]
  10. Couchman JR. Syndecans: proteoglycan regulators of cell-surface microdomains? Nat Rev Mol Cell Biol 4: 926–937, 2003.[CrossRef][Web of Science][Medline]
  11. De Vries L, Lou X, Zhao G, Zheng B, Farquhar MG. GIPC, a PDZ domain containing protein, interacts specifically with the C terminus of RGS-GAIP. Proc Natl Acad Sci USA 95: 12340–12345, 1998.[Abstract/Free Full Text]
  12. Diatchenko L, Lau YF, Campbell AP, Chenchik A, Moqadam F, Huang B, Lukyanov S, Lukyanov K, Gurskaya N, Sverdlov ED, Siebert PD. Suppression subtractive hybridization: a method for generating differentially regulated or tissue-specific cDNA probes and libraries. Proc Natl Acad Sci USA 93: 6025–6030, 1996.[Abstract/Free Full Text]
  13. Gao Y, Li M, Chen W, Simons M. Synectin, syndecan-4 cytoplasmic domain binding PDZ protein, inhibits cell migration. J Cell Physiol 184: 373–379, 2000.[CrossRef][Web of Science][Medline]
  14. Hasson T. Myosin VI: two distinct roles in endocytosis. J Cell Sci 116: 3453–3461, 2003.[Abstract/Free Full Text]
  15. Horowitz A, Tkachenko E, Simons M. Fibroblast growth factor-specific modulation of cellular response by syndecan-4. J Cell Biol 157: 715–725, 2002.[Abstract/Free Full Text]
  16. Kramer H, Phistry M. Genetic analysis of hook, a gene required for endocytic trafficking in drosophila. Genetics 151: 675–684, 1999.[Abstract/Free Full Text]
  17. Liu Y, Liu C, Yamada Y, Fan CM. Growth arrest specific gene 1 acts as a region-specific mediator of the Fgf10/Fgf8 regulatory loop in the limb. Development 129: 5289–5300, 2002.[Medline]
  18. Patino WD, Mian OY, Shizukuda Y, Hwang PM. Current and future applications of SAGE to cardiovascular medicine. Trends Cardiovasc Med 13: 163–168, 2003.[CrossRef][Web of Science][Medline]
  19. Powers CJ, McLeskey SW, Wellstein A. Fibroblast growth factors, their receptors and signaling. Endocr Relat Cancer 7: 165–197, 2000.[Abstract]
  20. Rochat-Steiner V, Becker K, Micheau O, Schneider P, Burns K, Tschopp J. FIST/HIPK3: a Fas/FADD-interacting serine/threonine kinase that induces FADD phosphorylation and inhibits fas-mediated Jun NH(2)-terminal kinase activation. J Exp Med 192: 1165–1174, 2000.[Abstract/Free Full Text]
  21. Simons M, Horowitz A. Syndecan-4-mediated signalling. Cell Signal 13: 855–862, 2001.[CrossRef][Web of Science][Medline]
  22. Sun Y, Buki KG, Ettala O, Vaaraniemi JP, Vaananen HK. Possible role of direct Rac1-Rab7 interaction in ruffled border formation of osteoclasts. J Biol Chem 280: 32356–32361, 2005.[Abstract/Free Full Text]
  23. Tani TT, Mercurio AM. PDZ interaction sites in integrin alpha subunits. T14853, TIP/GIPC binds to a type I recognition sequence in alpha 6A/alpha 5 and a novel sequence in alpha 6B. J Biol Chem 276: 36535–36542, 2001.[Abstract/Free Full Text]
  24. Tkachenko E, Elfenbein A, Tirziu D, Simons M. Syndecan-4 clustering induces cell migration in a PDZ-dependent manner. Circ Res 98: 1398–1404, 2006.[Abstract/Free Full Text]
  25. Volk R, Schwartz JJ, Li J, Rosenberg RD, Simons M. The role of syndecan cytoplasmic domain in basic fibroblast growth factor-dependent signal transduction. J Biol Chem 274: 24417–24424, 1999.[Abstract/Free Full Text]
  26. Wallrapp C, Gress TM. Isolation of differentially expressed genes by representational difference analysis. Methods Mol Biol 175: 279–294, 2001.[Medline]
  27. Wang LH, Kalb RG, Strittmatter SM. A PDZ protein regulates the distribution of the transmembrane semaphorin, M-SemF. J Biol Chem 274: 14137–14146, 1999.[Abstract/Free Full Text]
  28. Wheelock MJ, Johnson KR. Cadherins as modulators of cellular phenotype. Annu Rev Cell Dev Biol 19: 207–235, 2003.[CrossRef][Web of Science][Medline]
  29. Yang M, Sytkowski AJ. Cloning differentially expressed genes by linker capture subtraction. Anal Biochem 237: 109–114, 1996.[CrossRef][Web of Science][Medline]
  30. Zhang Y, Li J, Partovian C, Sellke FW, Simons M. Syndecan-4 modulates basic fibroblast growth factor 2 signaling in vivo. Am J Physiol Heart Circ Physiol 284: H2078–H2082, 2003.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Am. Soc. Nephrol.Home page
E. A. Ashley, J. M. Spin, R. Tabibiazar, and T. Quertermous
Frontiers in Nephrology: Genomic Approaches to Understanding the Molecular Basis of Atherosclerosis
J. Am. Soc. Nephrol., November 1, 2007; 18(11): 2853 - 2862.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Table
Right arrow All Versions of this Article:
27/3/380    most recent
00145.2006v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lanahan, A. A.
Right arrow Articles by Simons, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lanahan, A. A.
Right arrow Articles by Simons, M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2006 by the American Physiological Society.