Molecular diversity of cardiac endothelial cells in vitro and in vivo

Jan Hendrickx1, Kris Doggen1, Ellen O. Weinberg2, Pascale Van Tongelen1, Paul Fransen1 and Gilles W. De Keulenaer1

1 Laboratory for Physiology, Department of Pharmacology, University of Antwerp, Antwerp, Belgium
2 Cardiovascular Research, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
In addition to a number of common features, cardiovascular endothelium displays structural, functional, and genetic differences according to its position in the cardiovascular tree. In the heart, endocardial and cardiac microvascular endothelia (CMVE) interact directly with surrounding cardiomyocytes, whereas the endothelium within blood vessels interacts with smooth muscle cells. In this study, we investigated whether cardiac endothelial cells were distinct from aortic endothelial (AE) cells at the transcriptional level. Using Affymetrix microarray technology and subsequent real-time PCR analyses for validation, we identified sets of genes with marked preferential expression in cultured endocardial endothelium (EE) compared with cultured AE and vice versa. Among the genes preferentially expressed in EE, some were also expressed in cultured CMVE. Immunohistochemical staining of cardiac and aortic tissue revealed that the endothelial genetic diversity observed in culture reflects, in part, a physiological diversity existing in vivo. The identification of a set of genes preferentially expressed in EE provides new insights in the functional adaptations of this endothelial subtype to its intracavitary localization and to its role in the control of ventricular performance.

heart; gene expression; microarray analysis; endothelium; cardiomyocyte


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
THE ENDOTHELIAL CELL LAYER lining the vasculature plays a crucial role in the control of vascular tone, hemostasis, inflammation, tissue growth, capillary exchange, and angiogenesis. Integrity of the endothelium is essential for cardiovascular homeostasis, and dysfunction underlies cardiovascular and other diseases such as acute coronary syndrome and atherosclerosis. In addition to a number of common features, the endothelium displays structural and functional differences related to biomechanical and biochemical stimuli, according to its position in the cardiovascular system. One of the most prominent biomechanical differences is the spatial variation of shear stress along the cardiovascular tree. At arterial bifurcations such as the aorta and in the cardiac ventricles, blood flow shows a turbulent or oscillatory flow pattern with flow reversals and low shear stress amplitudes. In contrast, in straight parts of the vascular tree, blood flow is laminar, and amplitudes are dependent on the position in the vascular tree (e.g., arterial vs. venous). Shear stress has important effects on endothelial cell phenotype, function, and gene expression, and some types of shear stress are associated with endothelial pro-atherosclerotic responses and atherosclerotic lesions (4, 10, 14). Aside from biomechanical factors, endothelial diversity may also be explained by biochemical stimuli via local intercellular cross talk with neighboring cells through the release of diffusible mediators, which is tissue specific. In resistance vessels, the endothelium functions in communication with subjacent smooth muscle cells in the control of vascular resistance and regional blood flow, whereas in capillaries the endothelium interacts with surrounding tissue-specific cells. Hence, endothelial function may be part of a specialized organ function.

Endothelial diversity is most striking where the endothelium has developed unique functional adaptations required for a specific organ function. The endothelium in the corneal microcirculation, for example, creates transendothelial ion fluxes that underlie the corneal fluid pump, which is essential to counterbalance fluid leak into the cornea through the corneal endothelium (3). Similarly, the endothelial layer of the blood-brain barrier forms a tight interface between blood and neuronal tissue, with active transport systems that create a milieu essential for the function of the underlying neuronal cells.

Endothelial diversity is an essential feature of the cardiovascular system, but its origin and mechanisms are incompletely understood. Whether endothelial diversity is an "intrinsic" property, independent of its location within the organ, cellular cross talk, and hemodynamic factors, is not yet clear. Recent studies have shown that endothelial diversity is not only reflected at the structural and functional level but also at the transcriptional level. Previous studies using microarray analysis, have shown that endothelial cells originating from distinct cardiovascular locations have unique expression patterns (8, 15, 16, 17). Based on the transcriptional profile of 53 endothelial cell cultures originating from different locations, Chi et al. (8) divided endothelial cells into several groups, according to their anatomical origin (large vessel vs. microvascular endothelium, arterial vs. venous endothelium, and microvascular endothelium from different tissues). Kallmann et al. (17) compared the expression profile of human cerebral endothelial cells and human umbilical cord vascular endothelial cells and identified 35 genes specific for brain endothelium, which included genes for the control of growth and neurotrophy, neuroprotection, angiogenesis, and immunoregulation.

In the heart, both the endocardial endothelium (EE) and cardiac microvascular endothelium (CMVE) share the property of close anatomical and functional interaction with cardiomyocytes (5). In the developing heart, the EE is responsible for myocardial trabeculation, cushion formation, and formation of Purkinje fibers (6, 13, 20). In the mature heart, EE and CMVE modulate cardiac muscle performance and growth (1, 6). Although EE and CMVE have similar functions in the adult heart, they have a different embryological origin. Furthermore, whereas the CMVE has a rather leaky structure (5, 7), EE cells show extensive intercellular overlap and a large number of gap junctions. This latter property suggests that the EE has a barrier function with properties comparable with the blood-brain barrier. Indeed, it has been shown that the EE influences the subendothelial myocardial ionic environment (11, 12).

In the present study, we studied the transcriptional profile of cardiac endothelial cells to test whether cardiac endothelial functional differentiation was reflected at the level of gene expression. We performed a single screening DNA microarray comparison between cultured EE and aortic endothelial (AE) cells. This experiment allowed us to identify a number of differentially expressed genes revealing that EE and AE cells express different genes when grown under identical culture conditions in the absence of the physiological differences present in vivo. These results further support a specific role for EE cells in the cardiovascular system and may provide novel underlying mechanistic insights.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Cell Culture
Aortic endothelium.
Adult rats were killed by inhalation of CO2, followed by cardiectomy. The thoracic aorta was removed, rinsed with PBS, and dissected to expose the luminal surface. AE cells were isolated by incubation with collagenase solution (1 mg/ml collagenase II and IV) at 37°C for 1 h, followed by centrifugation at 1,600 rpm for 10 min. After washing with PBS, the AE cells were resuspended in DMEM culture medium (Cambrex) containing 1% (wt/vol) glucose, 10% FBS (Invitrogen), and 0.5% streptomycin/penicillin solution (10,000 U/ml, Invitrogen). Cell cultures were kept in a humidified incubator with 5% CO2 at 37°C. Culture medium was replaced every 2 to 3 days.

Endocardial endothelium.
After dissection of the heart, the left and right ventricle were dissected, rinsed with PBS, and incubated in DMEM (Cambrex) containing 2 mg/ml collagenase and 0.5% streptomycin/penicillin solution for 20 min in a humidified incubator with 5% CO2 at 37°C. EE cells were gently scraped, rinsed with PBS, and collected by centrifugation at 1,600 rpm for 10 min. Cells were further processed as described above.

Cardiac microvascular endothelium.
After removal of the EE and removal of the epicardial coronaries, the left and right ventricles were cut into small pieces and incubated in 25 ml HBSS-Ca containing 0.08% (wt/vol) collagenase at 37°C for 30 min. After shearing by repeated pipetting, 25 ml trypsin solution (0.05%, wt/vol; Invitrogen) was added, and the fragments were incubated at 37°C for 30 min, then sheared by pipetting. The CMVE cells were subsequently collected by centrifugation at 700 rpm for 5 min and processed as described above.

Purity of the cell cultures was analyzed by staining with 15 µg Alexa Fluor 594 AcLDL (Molecular Probes) overnight at 37°C. This demonstrated a greater than 95% purity of AE, EE, and CMVE cells (Fig 1).



View larger version (60K):
[in this window]
[in a new window]
 
Fig. 1. Staining of aortic endothelium (AE), endocardial endothelium (EE), and cardiac microvascular endothelium (CMVE) with Alexa 594-labeled AcLDL, demonstrating the purity of endothelial cell cultures.

 
RNA Isolation
Endothelial cells were harvested by trypsinization. RNA was isolated using TRIzol reagent (Invitrogen). RNA concentrations were determined by measuring the absorbance at 260 nm.

Affymetrix Microarray Analysis
Affymetrix oligonucleotide rat genome array GeneChips (RG-U34A) were used. This microarray GeneChip contains 8,800 transcripts, of which ~5,350 represented known genes and 3,450 represented expressed sequences tags (ESTs) (~25% of the rat transcriptome, which contains ~12,400 known genes and ESTs). Each gene or EST is represented on the GeneChip by 16 probe pairs. Each probe pair consists of a 25-bp oligonucleotide sequence that is a perfect match to the gene of interest and a 25-bp oligonucleotide containing a single base mismatch at bp 13. Affymetrix GeneChip analysis was performed according to recommended procedures. Briefly, total RNA was reverse-transcribed to double-stranded cDNA (Life Technologies), and biotinylated cRNA was generated by an in vitro transcription (Enzo). Labeled cRNA was purified (Qiagen), fragmented by alkaline treatment, and hybridized to a GeneChip array overnight at 45°C. The array was washed, stained with streptavidin phycoerythrin, and scanned. Each array was scanned twice, and an average intensity for each probe pair was generated. Data were analyzed using Affymetrix Microarray Suite (MAS 4.01) to assess quality of RNA and hybridization.

PCR and RT-PCR
PCR was performed with the ReadyMix RedTaq reaction mix (Sigma) in a reaction mixture containing 400 nM of both primers (Table 1) and 0.5 µl of template. After a 5-min denaturation at 92°C, 30–35 cycles of PCR were performed consisting of a denaturation step at 92°C for 30 s, a primer annealing step at 55°C for 30 s, and a primer extension step at 72°C for 1 min, with a final extension step of 8 min.


View this table:
[in this window]
[in a new window]
 
Table 1. PCR primers used for hybridization probe construction

 
RT-PCR was performed using the SuperScript One-Step RT-PCR System (Invitrogen) in a 25-µl reaction volume containing 0.5 µl total RNA (10–100 ng) as described previously (11).

Northern Blot Analysis
Probe construction and labeling.
Hybridization probes were generated by RT-PCR on RNA isolated from rat EE or AE. PCR fragments were analyzed by gel electrophoresis, excised from the gel, and purified using the QIAquick Gel Extraction Kit (Qiagen).

Approximately 25 ng of DNA probe was labeled in a 50-µl reaction volume containing 5 µl [{alpha}-32P]dCTP (3,000 Ci/mmol) using the Prime-It random primer labeling kit (Stratagene). Labeled probes were purified using the Chroma Spin+ TE-100 purification columns (Clontech).

RNA electrophoresis and Northern transfer.
Five micrograms of RNA was loaded on a 1.5% agarose gel containing 6.3% formaldehyde. Electrophoresis was performed in 1x MOPS buffer at a constant voltage of 120 V during 2 h. Following electrophoresis, RNA was transferred to a Nytran SuPerCharge nylon membrane (Schleicher and Schuell) with the TurboBlotter transfer system (Schleicher and Schuell). Subsequently, RNA was cross-linked with a CL-1000 ultraviolet cross-linker (UVP Inc.).

Hybridization.
Membranes were prehybridized in 10 ml of QuickHyb solution (Stratagene) for 30 min at 68°C. Hybridization was performed for 1 h at 68°C in QuickHyb solution containing the denatured labeled hybridization probe. Following hybridization, the membranes were washed at 50°C in wash buffers with increasing stringency, ranging from 2x SSC, 0.05% SDS to 0.1x SSC, 0.05% SDS, until background signals disappeared. Signals were visualized by autoradiography with Kodak Biomax MS film.

Real-Time Quantitative RT-PCR
Primers for real-time quantitative RT-PCR for all transcripts except macrophage metalloelastase (MME) (Table 2) were designed with the Primer Express software (Applied Biosystems). Reverse transcription was performed with the TaqMan reverse transcription reagents (Applied Biosystems) in a 50-µl reaction volume containing 5 µl total RNA solution (~1 µg), using 2.5 µM random hexamers. Samples were diluted to 250 µl, and 5 µl was used for real-time PCR in a 25-µl reaction volume containing 12.5 µl SYBR Green PCR Reaction Mix (Applied Biosystems) and 750 nM of both PCR primers. After an initial incubation for 2 min at 50°C and 10 min at 95°C, 45 PCR cycles were carried out consisting of a denaturation step of 15 s at 95°C and a primer annealing and elongation step of 1 min at 60°C, on an ABI Prism 7700 sequence detection system. Expression was normalized to GAPDH expression.


View this table:
[in this window]
[in a new window]
 
Table 2. PCR primers used for real time RT-PCR

 
Real-time PCR for MME was performed using the rat MME TaqMan Assay-on-Demand (Applied Biosystems) with the TaqMan One-Step RT-PCR Master Mix (Applied Biosystems) according to the manufacturer’s instruction.

Immunofluorescent Staining
Cryosections were stained as published previously (11). In summary, following fixation with acetone for 10 min, cryosections were successively incubated with blocking solution [either goat or donkey IgG (Jackson ImmunoResearch), 0.2% BSA in PBS] for 30 min at room temperature, primary antibody (VWF, PECAM, DCN, ATPB1, GATA-GT2) for 24 h at 4°C, and secondary antibody [Alexa 488- or 594-labeled goat or donkey anti-IgG (Molecular Probes)] for 3 h at 37°C, and finally treated with SlowFade (Molecular Probes).


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Identification of Differentially Expressed Genes by Microarray Analysis and Real-Time RT-PCR
To identify EE-specific genes, transcriptional profiles of rat EE and AE were compared using a single Affymetrix DNA microarray analysis. EE and AE cells were isolated from two rats, pooled (EE and AE kept separated), and cultured until second passage, under identical culture conditions. This single array experiment revealed 299 genes that were preferentially expressed in EE (2-fold or greater difference), of which 100 showed a 10-fold or greater overexpression, and 201 genes that were preferentially expressed in AE, 38 of which showed a 10-fold or greater overexpression. The complete data set has been submitted to the Gene Expression Omnibus database(http://www.ncbi.nlm.nih.gov/geo; accession nos. GSM25148 and GSM25149, series GSE1478).

In subsequent validation experiments, about one-third of the genes with 10-fold or greater differential expression between EE and AE were chosen for validation by Northern blot analysis and/or real-time quantitative RT-PCR. For validation experiments, four additional and independent parallel cultures of AE and EE cells identical to those used for microarray analysis were analyzed. Figure 2 shows the results of Northern blot and real-time PCR (means + SE, n=4) of three genes preferentially expressed in AE, i.e., decorin (DCN, accession no. XM_216883), WDNM1 (accession no. X13309), and vasopressin receptor 1a (VPR1a, accession no. U39450) and three genes preferentially expressed in EE, i.e., sortilin (SORT, accession no. XM_215675), MME (accession no. X98517), and lysozyme (LYZ, accession no. L12459). Validations of other genes were performed with quantitative RT-PCR. Tables 3 and 4 show the results of the validation of 30 genes with pronounced preferential expression in EE (Table 3) and 10 genes with pronounced preferential expression in AE (Table 4), identified by the microarray analysis. Respectively, 24 and 6 of these genes indeed showed preferential expression in either cell type (Tables 3 and 4, Fig. 3).



View larger version (25K):
[in this window]
[in a new window]
 
Fig. 2. Representative Northern blot and real-time quantitative PCR results of decorin (DCN), vasopressin receptor 1a (VPR1a), and WDNM1, preferentially expressed in AE (top), and sortilin (SORT), macrophage metalloelastase (MME), and lysozyme (LYZ), preferentially expressed in EE (bottom). Quantitative RT-PCR results represent relative gene expression in AE vs. EE (means + SE of 4 independent EE and AE cell cultures).

 

View this table:
[in this window]
[in a new window]
 
Table 3. Genes upregulated in endocardial endothelium

 

View this table:
[in this window]
[in a new window]
 
Table 4. Genes upregulated in aortic endothelium

 


View larger version (39K):
[in this window]
[in a new window]
 
Fig. 3. Quantitative RT-PCR of 24 genes preferentially expressed in EE (AC) and 6 genes preferentially expressed in AE (D). Graphs show relative gene expression in AE vs. EE (means + SE of 4 independent EE and AE cell cultures).

 
Control Experiments
Two additional control experiments validated the experimental procedures. First, the expression of two genes with identical expression in AE and EE from the microarray analysis [sodium bicarbonate cotransporter 1 (NBC1, accession no. AF004017) and endothelin converting enzyme 1 (ECE1, accession no. D29683)] was assessed by real-time RT-PCR. Real-time RT-PCR confirmed similar expression levels for both NBC1 (P=0.589, n=4) and ECE1 (P=0.089, n=4) (data not shown). Second, the expression levels of two genes with pronounced preferential expression in EE or AE, respectively, GATA-GT2 DNA binding protein (accession no. L22761) and WDNM1, were determined in four parallel EE or AE cell cultures, respectively, at different cell culture passages. These experiments showed that the expression levels of these genes did not show significant differences up to passage 2, the passage used in our screening and validation experiments (P=0.115 for GATA-GT2 in EE and P=0.568 for WDNM1 in AE; data not shown).

In Vivo Validation
To test the validity of our findings in vivo, cryosections of rat heart and aorta were analyzed by immunofluorescent staining. Figure 4A shows double staining of aorta and heart with an antibody against DCN (red) and von Willebrand factor (VWF), an endothelium-specific marker (green). DCN, which was preferentially expressed in AE at the mRNA level in vitro, was abundantly expressed in AE in vivo but was absent in EE. Figure 4, B and C, respectively, show positive staining of heart and aorta for the ß1-subunit of the sodium-potassium ATPase (ATPB1, accession no. NM_013113) and GATA-GT2, two genes found to be preferentially expressed in EE at the mRNA level in vitro. Consistent with the mRNA data, both proteins were expressed in EE in vivo, but were absent in AE. These results demonstrate that the differential expression of mRNA between EE and AE in vitro reflects differences in vivo at the protein level.



View larger version (47K):
[in this window]
[in a new window]
 
Fig. 4. Double staining of cryosections of rat aorta (top) and heart (bottom). Sections were stained with an antibody against the protein of interest, labeled with Alexa 594 (red), and an antibody against an endothelial-specific marker, labeled with Alexa 488 (green). A: staining with antibodies against DCN (red) and von Willebrand factor (VWF, green), clearly showing a double signal in AE (blue arrows), whereas in EE only the green signal of VWF is present. B: staining with antibodies against ATPB1 (red) and PECAM (green); the AE only shows staining for PECAM in contrast to staining for both PECAM and ATPB1 in EE (blue arrows). C: staining with antibodies against GATA-GT2 (red) and PECAM (green); the GATA-GT2 DNA binding protein is clearly present in the EE cell nucleus (blue arrows) but absent in AE.

 
Left vs. Right Ventricular EE
The EE cultures used in above experiments were derived from a mixture of endothelial cells from rat right and left ventricle. As EE cells from right and left ventricle originate from different physiological environments, their expression profile may also differ. To test this, the expression levels of 10 genes with high differential expression in EE were determined in 3 parallel EE cultures from right and left ventricle from the same animal. As shown in Fig. 5, there were no significant differences in expression levels for any of the genes tested, suggesting that none of these genes was specific for either left or right ventricle.



View larger version (30K):
[in this window]
[in a new window]
 
Fig. 5. Relative gene expression of 10 genes preferentially expressed in EE, in left vs. right ventricular EE. Graphs show the means + SE of 3 independent EE cell cultures from right (RV) and left ventricle (LV), respectively.

 
Expression of Differentially Expressed Genes in CMVE
CMVE and EE cells share the property of close anatomical and functional interaction with cardiomyocytes, but their embryological origin is different. The expression levels of the identified genes with differential expression in EE or AE were compared in CMVE, EE, and AE after isolation from the same rat and culture to passage 2 as parallel triplets (Fig. 6). These comparisons allowed classification of endothelial genes in three groups. The first group was expressed preferentially in EE. The mRNA abundance of these genes was markedly lower in AE and CMVE, and these genes were classified as "endocardial endothelial genes." The second group was expressed preferentially in cardiac endothelium (EE and CMVE), and these genes were classified as "cardiac endothelial genes." Finally, the mRNA abundance of the third group was high in AE and lower in EE and CMVE. These genes were classified as "aortic endothelial genes." Strikingly, none of the genes with preferential expression in AE showed an equally high expression level in CMVE, and hence, a group of "vascular endothelial genes" could not be identified.



View larger version (36K):
[in this window]
[in a new window]
 
Fig. 6. Relative gene expression of 24 genes preferentially expressed in EE and 6 genes preferentially expressed in AE in microvascular vs. AE and EE. Graphs show means + SE of 3 independent AE, MVE, and EE cell cultures. Based on their relative expression, the genes can be divided in endocardial endothelial genes (A), cardiac endothelial genes (B), and aortic endothelial genes (C).

 

    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Consistent with previously described morphological and functional differentiation of EE (5), here we describe diversity of cardiac and aortic endothelial cells at the transcriptional level. A set of genes with differential expression in EE when compared with AE or CMVE was described, and for a subset of these genes, differential expression at the protein level was also observed in vivo. The characteristic gene expression profile of EE described here may therefore be independent of extrinsic and physiological environmental factors such as ventricular blood flow or an in situ position at the myocardial border. Instead, this profile may be linked to an intrinsic cell-specific program that is preserved in culture.

EE cultures used for microarray analysis and validation experiments consisted of mixtures of EE cells isolated from left and right ventricles. Further analysis of 10 genes highly expressed in this left/right EE mixture, but nearly absent in AE, showed that the expression of these genes was similar in separate cultures of left and right ventricular EE, consistent with the concept that an intrinsic genetic "endocardial" program may exist in these endothelial cells. Interestingly, this "endocardial" program was only partly preserved in CMVE. Since CMVE is embryologically different from EE but shares an anatomical proximity and functional interaction with cardiomyocytes, this observation suggests that genes commonly expressed in EE and CMVE including GATA-GT2, oxidized low-density lipoprotein receptor 1 (OLR1), apolipoprotein E (APOE), creatine kinase brain type (CKB), parathyroid hormone receptor (PTHR), and UDP glucosyltransferase 1 (UGT1A7) participate in the functional interaction with cardiac myocytes.

Interpretation of the results at an individual gene level may provide novel mechanistic and functional insights into cardiac endothelial physiology. For example, three of the genes with preferential expression in EE encode for ion channels or ion channel related genes. These include an atypical voltage-gated Na+ channel (SCL11/SCN6a), ATPB1 (the ß1-subunit of the Na+-K+-ATPase), and GATA-GT2, which is the transcription factor for the H+-K+-ATPase (18). Previous observations from our laboratory suggest that the EE may act as an electrophysiological barrier that affects subendocardial ion concentrations (11, 12). In this process, a role for the Na+-K+-ATPase has already been implicated (11). The precise role of the ß1-subunit in the function of the Na+-K+-ATPase is not clear, but selective expression of one of the ß-subunit isoforms may reflect specific electrophysiological features of EE. Similarly, although the function of the atypical voltage-gated Na channel is not known, its expression in EE may unmask novel aspects of endothelial membrane ion transport. This also accounts for the H+-K+-ATPase, whose only known function is the control of gastric acid production. Preliminary studies in our laboratory, however, suggest that H+-K+-ATPase also plays a role in an endocardial transendothelial ion transport. H+-K+-ATPase activity may also be linked to the function of the sodium bicarbonate exchanger (NBC) (21), which was recently shown to be important for the regulation of subendocardial intracellular pH (12).

Another subset of genes differentially expressed between EE and AE relates to angiogenesis: macrophage metalloelastase (MME or MMP12) and decorin (DCN). This observation may indicate that both cell types have different angiogenetic properties. Gene activity of MME, the activator of angiostatin (9), was high in EE but very low in AE. mRNA for DCN, on the other hand, a gene expressed in sprouting endothelial cells and essential for angiogenesis (23), was nearly undetectable in EE but expressed at high levels in AE and CMVE. These results may suggest that the endocardial phenotype of the endothelial cell includes downscaled angiogenetic properties. Further experiments are required to test this hypothesis.

Finally, some of the genes preferentially expressed in EE are involved in embryonic cardiac development, cell growth, or proliferation. These include {alpha}-actinin-associated LIM protein (LIM), which plays an essential role in right ventricular development (19), transforming growth factor ß (TGF-ß) (22), and A5D3 protein (2). This observation is in agreement with the observation that the EE participates in growth responses of the embryonic as well as of the mature heart (6, 13, 20).

In conclusion, in the present study, we have shown molecular diversity of cardiac vs. aortic endothelial cells in vitro and in vivo. This diversity was preserved when the endothelial cells were removed from their specific biomechanical and biochemical environment. From these observations, we hypothesize the existence of an intrinsic genetic program for EE cells, microvascular cells, and AE cells. This program may be related to the modulatory role of the endocardium on cardiac development and function and may reveal specific functions of this endothelial phenotype.


    GRANTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
This work was supported by the Belgian Science Policy (project no. IAP-P5/02), the Fund for Scientific Research, Flanders (FWO Vlaanderen, project no. G.046.03), and by a Bekales Research Grant (Liechtenstein) in the field of cardiology. Jan Hendrickx is a postdoctoral fellow of the Fund for Scientific Research, Flanders (FWO Vlaanderen).


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

Address for reprint requests and other correspondence: G. W. De Keulenaer, Laboratory for Human Physiology and Pathophysiology, Dept. of Pharmacology, Univ. of Antwerp, Groenenborgerlaan 171, 2020 Antwerp, Belgium (E-mail: gilles.dekeulenaer{at}ua.ac.be).


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Balligand JL, Kelly RA, and Smith TW. Cardiac endothelium and tissue growth. Prog Cardiovasc Dis 39: 351–360, 1997.[ISI][Medline]
  2. Blomberg LA, Chan WY, Clerch LB, and Massaro D. Molecular cloning and characterization of a novel gene upregulated early during postnatal rat lung development. Biochim Biophys Acta 1574: 391–398, 2002.[ISI][Medline]
  3. Bonanno JA and Giasson C. Intracellular pH regulation in fresh and cultured bovine corneal endothelium. II. Na+:HCO3 cotransport and Cl/HCO3 exchange. Invest Ophthalmol Vis Sci 33: 3068–3079, 1992.[Abstract]
  4. Borang S, Andersson T, Thelin A, Odeberg J, and Lundeberg J. Vascular gene expression in atherosclerotic plaque-prone regions analyzed by representational difference analysis. Pathobiology 71: 107–114, 2004.[ISI][Medline]
  5. Brutsaert DL, Fransen P, Andries LJ, De Keulenaer GW, and Sys SU. Cardiac endothelium and myocardial function. Cardiovasc Res 38: 281–290, 1998.[CrossRef][ISI][Medline]
  6. Brutsaert DL. Cardiac endothelial-myocardial signaling: its role in cardiac growth, contractile performance, and rhythmicity. Physiol Rev 83: 59–115, 2003; doi:10.1152/physrev.00017.2002.[Abstract/Free Full Text]
  7. Bundgaard M. The three-dimensional organization of tight junctions in a capillary endothelium revealed by serial-section electron microscopy. J Ultrastruct Res 88: 1–17, 1984.[ISI][Medline]
  8. Chi JT, Chang HY, Haraldsen G, Jahnsen FL, Troyanskaya OG, Chang DS, Wang Z, Rockson SG, van de Rijn M, Botstein D, and Brown PO. Endothelial cell diversity revealed by global expression profiling. Proc Natl Acad Sci USA 100: 10623–10628, 2003.[Abstract/Free Full Text]
  9. Cornelius LA, Nehring LC, Harding E, Bolanowski M, Welgus HG, Kobayashi DK, Pierce RA, and Shapiro SD. Matrix metalloproteinases generate angiostatin: effects on neovascularization. J Immunol 161: 6845–6852, 1998.[Abstract/Free Full Text]
  10. De Keulenaer GW, Chappell DC, Ishizaka N, Nerem RM, Alexander RW, and Griendling KK. Oscillatory and steady laminar shear stress differentially affect human endothelial redox state: role of a superoxide-producing NADH oxidase. Circ Res 82: 1094–1101, 1998.[Abstract/Free Full Text]
  11. Fransen P, Hendrickx J, Brutsaert DL, and Sys SU. Distribution and role of Na+/K+ ATPase in endocardial endothelium. Cardiovasc Res 52: 487–499, 2001.[CrossRef][ISI][Medline]
  12. Fransen P, Lamberts R, Hendrickx J, and De Keulenaer GW. Endocardial endothelium modulates pHi in cardiac muscle: role of transendothelial HCO3 transport. Cardiovasc Res 63: 700–708, 2004.[CrossRef][ISI][Medline]
  13. Gassmann M, Casagranda F, Orioli D, Simon H, Lai C, Klein R, and Lemke G. Aberrant neural and cardiac development in mice lacking the ErbB4 neuregulin receptor. Nature 378: 390–394, 1995.[CrossRef][ISI][Medline]
  14. Glagov S, Zarins C, Giddens DP, and Ku DN. Hemodynamics and atherosclerosis. Insights and perspectives gained from studies of human arteries. Arch Pathol Lab Med 112: 1018–1031, 1988.[ISI][Medline]
  15. Ho M, Yang E, Matcuk G, Deng D, Sampas N, Tsalenko A, Tabibiazar R, Zhang Y, Chen M, Talbi S, Ho YD, Wang J, Tsao PS, Ben-Dor A, Yakhini Z, Bruhn L, and Quertermous T. Identification of endothelial cell genes by combined database mining and microarray analysis. Physiol Genomics 13: 249–262, 2003. First published March 18, 2003; doi:10.1152/physiolgenomics.00186.2002.[Abstract/Free Full Text]
  16. Huminiecki L and Bicknell R. In silico cloning of novel endothelial-specific genes. Genome Res 10: 1796–1806, 2000.[Abstract/Free Full Text]
  17. Kallmann BA, Wagner S, Hummel V, Buttmann M, Bayas A, Tonn JC, and Rieckmann P. Characteristic gene expression profile of primary human cerebral endothelial cells. FASEB J 16: 589–591, 2002.[Abstract/Free Full Text]
  18. Maeda M, Kubo K, Nishi T, and Futai M. Roles of gastric GATA DNA-binding proteins. J Exp Biol 199: 513–20, 1996.[Abstract/Free Full Text]
  19. Pashmforoush M, Pomies P, Peterson KL, Kubalak S, Ross J Jr, Hefti A, Aebi U, Beckerle MC, and Chien KR. Adult mice deficient in actinin-associated LIM-domain protein reveal a developmental pathway for right ventricular cardiomyopathy. Nat Med 7: 591–597, 2001.[CrossRef][ISI][Medline]
  20. Puri MC, Partanen J, Rossant J, and Bernstein A. Interaction of the TEK and TIE receptor tyrosine kinases during cardiovascular development. Development 126: 4569–4580, 1999.[Abstract/Free Full Text]
  21. Romero MF, Fulton CM, and Boron WF. The SLC4 family of HCO3 transporters. Pflügers Arch 447: 495–509, 2004.[CrossRef][ISI][Medline]
  22. Saltis J, Agrotis A, and Bobik A. Regulation and interactions of transforming growth factor-beta with cardiovascular cells: implications for development and disease. Clin Exp Pharmacol Physiol 23: 193–200 1996.[ISI][Medline]
  23. Schonherr E, O’Connell BC, Schittny J, Robenek H, Fastermann D, Fisher LW, Plenz G, Vischer P, Young MF, and Kresse H. Paracrine or virus-mediated induction of decorin expression by endothelial cells contributes to tube formation and prevention of apoptosis in collagen lattices. Eur J Cell Biol 78: 44–55, 1999.[ISI][Medline]