Zyxin Mediation of Stretch-Induced Gene Expression in Human Endothelial CellsNovelty and Significance
This article has a correction. Please see:
Rationale: Prolonged exposure to enhanced stretch, such as in hypertension, triggers endothelial dysfunction, a hallmark of pathological vascular remodeling processes. Despite its clinical relevance, little is known about stretch-induced gene expression in endothelial cells.
Objective: Here, we have characterized a new stretch-inducible signaling pathway and the subsequent changes in endothelial gene expression in response to stretch.
Methods and Results: Using human primary endothelial cells, we observed that the protein zyxin translocates from focal adhesions to the nucleus solely in response to stretch. There, it orchestrates complex changes in gene expression by interacting with a novel cis-acting element found in all zyxin-regulated genes analyzed so far. By way of DNA microarray pathway analyses, stretch-induced changes in endothelial cell gene expression were systematically explored, revealing that zyxin mainly regulates proinflammatory pathways.
Conclusions: Stretch appears to be an important factor in the development of endothelial dysfunction with zyxin as a potential therapeutic target to interfere with these early changes in endothelial cell phenotype.
Hypertension is a major predisposing factor for endothelial dysfunction,1 causing vascular remodeling and atherosclerosis. However, the mechanism(s) by which a prolonged increase in blood pressure, and thus stretch, translates into a change in endothelial cell (EC) phenotype remains elusive.
Fluid shear stress, the viscous drag of the flowing blood, is the second hemodynamic force sensed by ECs and generally considered to be antiatherosclerotic. In contrast to straight arterial segments, at bifurcations (cyclic) stretch dominates over shear stress, and it is mainly here where endothelial dysfunction and concomitantly atherosclerosis develops. Although there is some insight into shear stress–induced gene expression,2,3 rather little is known about stretch-induced signaling,4 with no systematic analysis of endothelial cell pathways affected by stretch-induced alterations in gene expression available to date. Here, we have comprehensively explored such transcriptional changes in human cultured ECs and in addition have identified the focal adhesion protein zyxin to be critically involved therein.
An expanded Methods section is available in the Online Data Supplement at http://circres.ahajournals.org.
Chiefly, primary ECs isolated from human umbilical cord veins were exposed to cyclic stretch (0% to 10% extension at 0.5 Hz) for the indicated periods by using a Flexercell FX-3000 strain unit. Protein and mRNA expression were analyzed by Western blot, immunofluorescence, and real-time RT-PCR analyses. DNA microarray analyses were performed using the human HG-U133 Plus 2.0 chip (Affymetrix) followed by Gene Set Enrichment Analysis (GSEA) using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG). The role of zyxin in endothelial gene expression was characterized by chromatin immunoprecipitation (ChIP), electrophoretic mobility-shift analysis (EMSA), use of an appropriate decoy oligodeoxynucleotide (ODN), and short interfering (si)RNA-mediated knockdown of the protein. All data are expressed as mean±SEM; statistical calculations were performed using unpaired Student's t test or repeated-measures ANOVA, followed by a Tukey–Kramer post hoc analysis as required.
Cyclic Stretch Causes Zyxin to Redistribute to the Nucleus, Where It Controls Specific Changes in Gene Expression
In quiescent ECs, zyxin localizes to focal adhesions and stress fibers. After exposure to cyclic stretch, although partly remaining in the focal adhesions, zyxin migrates to the nucleus where, according to immunofluorescence (Figure 1A) and Western blot analysis (Figure 1B), it partially colocalizes with the nuclear protein Ku80. Given that downregulation of zyxin affects stretch-induced gene expression in rat vascular smooth muscle cells,5 we used a siRNA-based approach in the human cultured ECs that resulted in a robust (80%) decrease in zyxin abundance (Online Figure I). Testing of 3 genes known to be stretch-sensitive revealed that the chemokines interleukin (IL)-8 and CXCL1 but not the B-type endothelin receptor (ETB-R) are controlled by zyxin (Figure 2A). Moreover, IL-8 expression, for example, was dependent on both the frequency and magnitude of stretch (Online Figure II) and, in addition to venous ECs, both stretch-induced nuclear translocation of zyxin and gene expression also occur in arterial ECs (Online Figures III and IV).
To get the full picture of these stretch-induced changes in gene expression and the role of zyxin therein, whole genome arrays were performed with 3 individual batches of primary ECs. Apart from comparing quiescent and stretched control ECs also cells without or with zyxin knockdown were analyzed side by side. Only gene products not affected by the transfection itself were included in the analysis. Both, pathway analysis (GSEA) based on KEGG and GO (Online Table II) and analyses of single genes revealed a prominent stretch response that was virtually identical in all 3 experiments. A total of 592 (with a P≤0.02) genes turned out to be stretch-sensitive of which 402 (67.9%) were regulated by zyxin (Figure 2B). This outcome was corroborated by reanalysis of 9 randomly chosen gene products using real time RT-PCR (Figure 2C). Moreover, zyxin seemed to be involved in the regulation of rather distinct pathways such as suppression of apoptosis or chemokine release (Online Table II).
Mechanism of Zyxin-Induced Gene Expression
Zyxin appears to associate with the promoter of the human thrombomodulin gene at a stretch of pyrimidines (PyPu box; position −453 to −481; M. Cattaruzza, unpublished observation, 2005). On analyzing the promoter regions of genes identified to be stretch-sensitive, comparable motifs close to the transcription start site were found solely in the 13 stretch-sensitive genes controlled by zyxin (Online Figure V) but not in the 11 zyxin-independent genes. To test whether zyxin indeed associates with this motif, ChIP assays were performed for 9 stretch-sensitive genes, revealing that zyxin in fact interacts with the promoter region of these genes. Genes that are expressed independently of zyxin did not emerge in this assay (Figure 3A).
To corroborate this finding, the cultured ECs were exposed to a decoy ODN mimicking the PyPu box in the human prepro-endothelin-1 (ET-1) gene (edn1; 5′-GCACTTCCTTCCTTTTCCCGAA-3′; position −163 to −136). Unlike the scrambled control ODN, the decoy ODN virtually abolished the stretch-induced expression of IL-8, as well as that of prepro-ET-1, another well-known stretch-sensitive gene (Figure 3B). EMSA using a probe with a sequence identical to that of the decoy ODN finally confirmed that in human cultured ECs exposed to cyclic stretch a nuclear protein–DNA complex forms that according to supershift analysis contains zyxin (Figure 3C).
Although stretch-induced gene expression in human cultured ECs has been analyzed by DNA microarrays before,6,–,8 the present data to our knowledge provide the first comprehensive pathway analysis. We further demonstrate that in ECs the cytoskeletal protein zyxin acts as a transducer of stretch into the nucleus, where it orchestrates the expression of a large subset of stretch-sensitive genes through a novel DNA-response element.
Prolonged hypertension, ie, a long-term increase in circumferential stretch, promotes endothelial dysfunction, which is characterized by an aberrant expression of proinflammatory genes. Our findings substantiate the assumption that ECs respond to such a supraphysiological increase in stretch with the activation of pathways known to be pivotal for the development of atherosclerosis.9,10
In addition, we show that zyxin, a protein normally associated with focal adhesions and stress fibers, plays a major role in stretch-induced endothelial gene expression. Apart from potential effects on the structural and functional integrity of the EC on its loss, zyxin directly interacts with the promoter region of most stretch-sensitive genes. Its binding motif does not resemble a typical consensus sequence but rather a particular base composition, ie, a stretch of pyrimidines. A preliminary “manual” screen of several stretch-sensitive genes revealed that it is only present in the proximal promoter regions of zyxin-regulated genes. This, however, does not exclude the possibility that similar elements may stochastically exist in unrelated regions of the genome as frequently is the case for the binding motifs of conventional transcription factors.
Almost all zyxin-dependent genes can be attributed to a set of well-defined pathways that, on the one hand, inhibit apoptosis and proliferation and strengthen cell–matrix interactions but, on the other hand, contribute to a broad range of proinflammatory responses (Online Figure VI). Such an orchestrated change in the endothelial transcriptome suggests a role for zyxin in vascular remodeling processes, especially when considering that ECs largely govern the phenotype of vascular smooth muscle cells11 and regulate leukocyte migration into the vessel wall.12,13 Although potentially preventing EC damage, this signaling pathway may thus facilitate endothelial dysfunction and consequently both hypertension-induced arterial remodeling and atherogenesis. Future studies must reveal whether the prevention of zyxin-mediated gene expression in ECs in fact constitutes a meaningful therapeutic strategy to combat these vascular complications.
Sources of Funding
This work was supported by a grant from the Deutsche Forschungsgemeinschaft (CA 262/1-3).
We thank Dr Mary Beckerle for providing the zyxin antibodies. We also thank Danijela Heide and Renate Cattaruzza for excellent technical assistance.
In June 2010, the average time from submission to first decision for all original research papers submitted to Circulation Research was 14.5 days.
Non-standard Abbreviations and Acronyms
- chromatin immunoprecipitation
- endothelial cell
- electrophoretic mobility-shift assay
- Gene Ontology
- Gene Set Enrichment Analysis
- Kyoto Encyclopedia of Genes and Genomes
- PyPu box
- pyrimidine-purine box (zyxin-binding DNA motif)
- short interfering RNA
- Received August 27, 2009.
- Revision received July 7, 2010.
- Revision received July 25, 2010.
- Accepted July 28, 2010.
- © 2010 American Heart Association, Inc.
- Ramkhelawon B,
- Vilar J,
- Rivas D,
- Mees B,
- de Crom R,
- Tedgui A,
- Lehoux S
- AbouAlaiwi WA,
- Takahashi M,
- Mell BR,
- Jones TJ,
- Ratnam S,
- Kolb RJ,
- Nauli SM
- Thodeti CK,
- Matthews B,
- Ravi A,
- Mammoto A,
- Ghosh K,
- Bracha AL,
- Ingber DE
- Cattaruzza M,
- Lattrich C,
- Hecker M
- Birukov KG,
- Jacobson JR,
- Flores AA,
- Ye SQ,
- Birukova AA,
- Verin AD,
- Garcia JG
- Shin HY,
- Smith ML,
- Toy KJ,
- Williams PM,
- Bizios R,
- Gerritsen ME
- Frye SR,
- Yee A,
- Eskin SG,
- Guerra R,
- Cong X,
- McIntire LV
- Cai H,
- Harrison DG
- Weber C
Novelty and Significance
What Is Known?
Supraphysiological cyclic stretch causes a pathological phenotype shift in endothelial cells (ECs).
This phenotypic alteration, referred to as endothelial dysfunction, is critically implicated in the onset and progression of hypertension-induced arterial remodeling and atherosclerosis.
What New Information Does This Article Contribute?
The cytoskeletal protein zyxin is a pivotal mediator of stretch-induced endothelial signaling.
Zyxin also acts as a transcription factor orchestrating the expression of up to 67% of all stretch-induced genes.
Exposure of ECs to an inadequate increase in blood pressure, and hence stretch, provokes a phenotype shift referred to as endothelial dysfunction, which is a key step in the development of atherosclerosis and related vascular remodeling processes. Despite its clinical impact, neither the primary stretch-induced signaling pathways nor the resulting changes in the endothelial transcriptome have as yet been analyzed in detail. Here, we provide first experimental evidence that the cytoskeletal protein zyxin mediates the bulk of stretch-induced endothelial gene expression, acting both as a mechanotransducer and transcription factor. Because zyxin activation mainly accounts for proinflammatory gene expression, the selective blockade of its activity may provide a promising therapeutic concept to target atherosclerosis and related vascular complications.