Vol. 276, Issue 5, H1755-H1768, May 1999
Cardiac microvascular endothelial cells express 
-smooth
muscle actin and show low NOS III activity
Hiroshi
Ando,
Thomas
Kubin,
Wolfgang
Schaper, and
Jutta
Schaper
Department of Experimental Cardiology, Max Planck Institute, D-61231
Bad Nauheim, Germany
 |
ABSTRACT |
We established a culture system of porcine
coronary microvascular endothelial cells (MVEC) with high cellular
yield and purity >98%. Endothelial origin was confirmed by
immunostaining, immunoblotting and fluorescence-activated cell sorter
(FACS) analysis using low-density lipoprotein uptake, CD31, von
Willebrand factor, and the lectin Dolichos
biflorus agglutinin. MVEC were positive
for 
-smooth muscle actin in culture and in myocardium, as confirmed
by FACS. Of the primary MVEC, ~30% expressed nitric oxide synthase
(NOS) III in numbers decreasing from the first passage (6 ± 1%) to
the second passage (4 ± 1%; P < 0.001 vs. primary isolates), whereas ~100% of aortic endothelial
cells (AEC) expressed NOS III. In AEC, NOS III activity (pmol
citrulline · mg
protein
1 · min1)
was 80 ± 10 and was nearly abolished in the absence of calcium (5 ± 1, P < 0.001). In primary
MVEC, however, NOS III activity in the presence and absence of calcium
was 20 ± 4 and 25 ± 5, respectively. We conclude that cardiac
MVEC, in contrast to AEC, contain -smooth muscle actin, show
low-grade NOS III activity, and provide a suitable in vitro system for
the study of endothelial pathophysiology.
nitric oxide synthase; pig heart; aortic endothelium
| |
INTRODUCTION |
NUMEROUS STUDIES have clearly demonstrated
that the endothelial cells (EC) of the coronary microvasculature play
important roles in diverse physiological and pathological situations
such as angiogenesis (17), cardiac hypertrophy accompanied by an increase in capillary density (42), and myocardial ischemia characterized by an increased vascularity in healing infarct scars and
by the stimulation of coronary collateral development (27). There is
growing evidence that locally acting peptides and/or autacoids released
by coronary microvascular endothelium act on cardiac muscle downstream
or subjacent to the affected region (25, 35). An interaction between
microvascular EC and cardiomyocytes has been shown to be important in
physiological and pathological conditions (5). Furthermore, it has been
suggested that functional and anatomic alterations in coronary
microvasculature play an important role in the development of ischemic
heart disease in patients (14). Thus coronary microvascular EC (MVEC)
are increasingly attracting attention in the field of cardiac
physiology and biology.
Endothelium exhibits a significant heterogeneity; in structure,
function, antigenic composition, metabolic properties, and response to
growth factors, they differ from organ to organ, and there are
differences between macrovascular and microvascular EC (13, 24).
Therefore, a well-characterized MVEC preparation of a target organ is
needed. MVEC have been successfully isolated from rat (30, 33), rabbit
(41), guinea pig (29), and, recently, human (30) heart. However, human
cardiac EC may not provide an appropriate in vitro model to investigate
biologically relevant functions of MVEC because they have been obtained
from diseased hearts in patients treated with various drugs before transplantation (26). Because porcine hearts resemble human hearts
anatomically as well as hemodynamically and because the coronary
morphology of humans without coronary disease is similar to that of
pigs (19, 20), an in vitro system of porcine MVEC might be useful for
the study of EC function.
In this report, we describe methods for the isolation and primary
culture of MVEC obtained from porcine myocardium. These primary
cultures have been extensively characterized to validate their
endothelial phenotype and microvascular origin.
| |
MATERIALS AND METHODS |
Isolation and culture of porcine cardiac MVEC.
German Landrace pigs (~30 kg) were anesthetized with pentobarbital
sodium. After thoracotomy, pigs were heparinized and killed by an
overdose of pentobarbital and 30% KCl. Hearts were rapidly excised and
washed in ice-cold calcium-free PBS. The distal left anterior
descending coronary artery was cannulated, and the surrounding tissue
was removed. The surface of an ~20-g myocardial segment was rinsed
with 70% ethanol, extensively washed in ice-cold PBS, and mounted on a
modified Langendorff perfusion system. The segment was then perfused
with oxygenized calcium-free perfusion buffer for 10 min followed by a
perfusion with the same buffer containing 0.08% collagenase, 0.01%
DNase (Boehringer Mannheim), 0.22% BSA (fatty acid free, Sigma), and
40 mM CaCl2 for 30 min at
37°C. After removal of the epicardial and endocardial surfaces, the remaining myocardial tissue was minced and further digested in the same
enzyme solution containing 1.2% BSA for 10 min. Dissociated cells were
filtered through a 100-µm nylon mesh followed by centrifugation at 25 g for 3 min. The supernatant was
incubated in the presence of 0.02% trypsin (Sigma), 160 mM
CaCl2, and 2% BSA for 30 min at
37°C and then centrifuged at 250 g
for 10 min. The pellet was suspended in medium 199 (Sigma), 10% FCS
(Sigma), 10% newborn calf serum (NCS; Sigma), and antibiotics. MVEC
were seeded on fibronectin (5 g/cm2; Sigma)-coated plastic
dishes (Falcon). After a 4-h attachment period in the incubator (5%
CO2-95%
O2, 37°C), the cells were washed twice and cultured in medium 199.
Isolation of EC from porcine aorta.
Porcine aortic EC (AEC) were obtained by scraping the internal surface
of the aorta excised from freshly slaughtered German Landrace
pigs. The cells were then incubated in calcium-free PBS containing
0.1% collagenase for 15 min at 37°C. After centrifugation at 250 g for 10 min, the cell pellet was
washed twice with medium 199 and, after seeding, was cultured under the
same conditions as described in Isolation and culture
of porcine cardiac MVEC. AEC used in the
present study were from the fourth to tenth passages and were studied
at confluence.
Immunostaining
of cultured cells.
All immunostaining was done with AEC and MVEC. Human dermal fibroblasts
(PromoCell, Heidelberg, Germany) were used as controls to test the
specificity of the antibodies. Cultured EC or fibroblasts were washed
three times with PBS and fixed in PBS containing 4% paraformaldehyde
(Merck, Darmstadt, Germany) for 15 min at room temperature (RT). Cells
were washed and permeabilized with PBS containing 0.01% Triton
X-100. After being blocked in 1% BSA (BSA-C, Aurion),
cell samples were incubated for 2 h at RT with primary antibody diluted
in PBS containing 0.001% Triton X-100 (see Table 1).
Cells were then exposed to biotin-labeled anti-mouse or -rabbit antibodies (1:100; Dianova) for 1 h, followed by streptavidin-Cy2 incubation (1:200; Biotrend) for a further 30 min. The fluorescence was
examined and photographed with a Leica fluorescence microscope or
confocal microscope (Leica).
The uptake of acetylated low-density lipoprotein (LDL) labeled with
1,1-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine perchlorate (Dil-Ac-LDL; Paesel and Lorei, Frankfurt am Main, Germany)
was tested as described by Voyta et al. (45). Cells were incubated for
4 h with 5 mg/ml of Dil-Ac-LDL in EC medium, rinsed three times with
PBS, and viewed under a fluorescence microscope.
Immunohistological staining of porcine myocardium.
Cryosections from porcine aorta and myocardium were stained with the
following primary antibodies used at dilutions as described in
Immunostaining of cultured cells:
anti-von Willebrand factor (VWF) monoclonal antibody (MAb), anti-nitric
oxide (NO) synthase (NOS) III MAb, anti-CD31, and anti--smooth
muscle actin MAb. Signals were detected by exposure to biotin-labeled
anti-mouse or -rabbit antibody, followed by streptavidin-Cy2, as
mentioned above. Sections were evaluated with the confocal microscope.
Double immunostaining of cultured EC and tissue.
Cells and tissue sections for this purpose were prepared as described
in Isolation and culture of porcine cardiac
MVEC and Isolation of EC from porcine
aorta. The preparations were blocked in PBS containing
10% FCS and then incubated with anti--smooth muscle actin MAb
(1:100) in PBS with 1% FCS for 1 h at 37°C. After being rinsed
three times in PBS (10 min each), they were exposed to FITC-labeled
anti-mouse secondary antibody (1:200). The preparations were then
incubated with either anti-VWF MAb (1:50) or anti-CD31 (1:50) for 90 min at RT. They were incubated for 1 h at RT with biotin-labeled
anti-rabbit secondary antibodies (1:100), followed by streptavidin-Cy3
(1:200) for 30 min at RT.
Flow cytometry.
AEC, MVEC, and fibroblasts were grown to confluence and removed from
culture dishes by short exposure to 0.05% trypsin and 0.6 mM EDTA in
PBS at 37°C. Cells were taken up in ice-cold PBS containing 10%
FCS and then fixed in 100% ethanol at 
20°C for 3 min. After
being washed twice in PBS with 10% FCS, cell samples were resuspended
in 50 ml of PBS containing 1% BSA, 0.1% NP-40, and 0.1%
NaN3 followed by incubation at RT
for 60 min with the following antibodies: anti--smooth muscle actin
MAb (40 mg/ml), anti-NOS III MAb (10 mg/ml), and anti-CD31 (40 mg/ml).
Isotype-matched mouse IgG (Sigma) was used as control in the case of
-smooth muscle actin and NOS III staining. Cell preparations were
washed three times for 10 min in PBS with 1% BSA, 0.1% NP-40, and
0.1% NaN3. FITC-labeled
F(ab')2 fragments of donkey
anti-mouse IgG (for -smooth muscle actin and NOS III staining) or
goat anti-rabbit IgG (for anti-CD31 staining) were used as secondary
antibodies at 10 mg/ml, and cells were kept at RT for 60 min. After
being washed, cell preparations were resuspended in PBS with 1% BSA and 0.1% NaN3 at a density of
~6 × 105 cells/ml and were
immediately analyzed by flow cytometry (FACScalibur, Becton-Dickinson).
Dil-Ac-LDL uptake capacity was detected by incubation of cells with 5 mg/ml of Dil-Ac-LDL for 4 h. After being washed with medium 199, cells
were trypsinized (0.05%) in PBS. Cell preparations were washed three
times in PBS with 10% FCS followed by resuspension and flow cytometric
analysis as described above.
Immunoblotting.
Confluent cultures were washed twice in ice-cold PBS and lysed in RIPA
buffer (10 mM Tris · HCl, pH 7.4, 150 mM NaCl, 1%
sodium deoxycholate, 1% Triton X-100, 0.1% SDS, 1 mM sodium
orthovanadate, and 100 mg/ml phenylmethylsulfonyl fluoride) left on ice
for 20 min. Cell lysates were centrifuged at 1,500 g for 10 min at 4°C. From the
resulting supernatant 15 mg of protein were separated by 7% SDS-PAGE.
NOS III and VWF were detected by the above-mentioned antibodies and
visualized by enhanced chemiluminescence (ECL, Amersham). Molecular
weight marker proteins were used as standards for SDS-PAGE (Bio-Rad).
Determination of NOS activity.
Confluent AEC and MVEC from
150-cm2 dishes were washed three
times with MgCl2- and
CaCl2-free ice-cold PBS and
scraped with a rubber policeman in PBS containing 1 mM EDTA. Cell
samples were spun at 500 g for 5 min.
The pellet was resuspended in 100 ml of homogenization buffer (in mM:
25 Tris · HCl, pH 7.4, 1 EDTA, and 1 EGTA) containing
the following protease inhibitors (in mM): 1 pepstatin A, 2 leupeptin,
and 1 phenylmethylsulfonyl fluoride. Cell suspensions were then
flash-frozen in liquid nitrogen and stored at 80°C until
use. After homogenization on ice, the homogenates were centrifuged at
1,500 g for 15 min at 4°C, and the
supernatant (500-600 mg total protein) was used for measurement of
NOS activity. Protein concentration was determined by the Bradford
method (Bio-Rad).
NOS activity was quantified in the presence or absence of calcium by
measuring the conversion of
L-[3H]arginine
(0.5 mCi, specific activity 63 Ci/mmol; Amersham) to L-[3H]citrulline,
according to the manufacturer's instructions (Stratagene). Parallel
reactions were analyzed in the presence of 1 mM
N
-nitro-L-arginine methyl
ester-HCl (Calbiochem).
Statistical analysis.
Data are expressed as means ± SE. Comparisons of data between
different groups were made by analysis of variance, and a
Scheffé's post hoc test was used when differences were
indicated. Differences were considered statistically significant at a
value of P < 0.05.
| |
RESULTS |
Isolation and characterization of porcine MVEC.
Primary isolates of MVEC grew steadily in medium 199 supplemented with
10% FCS and 10% NCS. Cells attached within 2 h, and well-spread
colonies were present within 2 days. The cultures approached confluence
after 4 days when plated at a density of 1.3 × 104
cells/cm2. Their doubling time was
~26 h.
Figure 1 shows the phenotypic difference between MVEC
and AEC at confluence. MVEC demonstrated a swirling growth pattern and appeared as elongated cells (Fig.
1A), quite different from the cobblestone appearance characteristic of EC derived from large vessels.
MVEC did not change their phenotype up to at least the third passage.
Porcine AEC cultured under the same conditions as MVEC (see
MATERIALS AND METHODS) demonstrated
the typical endothelial morphology of closely apposed polygonal cells
forming a contact-inhibited monolayer (Fig.
1B).
View larger version (113K):
[in this window]
[in a new window]
|
Fig. 1.
Phase-contrast micrographs of microvascular endothelial cells (MVEC)
and aortic endothelial cells (AEC). AEC demonstrate a characteristic
"cobblestone" appearance (B).
MVEC show a more elongated phenotype
(A).
|
|
Both MVEC and AEC exhibited intense granular cytoplasmic staining after
4-h incubation in Dil-Ac-LDL-supplemented media (Fig. 2). The degree of uptake and the staining pattern of
Dil-Ac-LDL were homogeneous in MVEC (Fig.
2b) and in polygonal AEC (Fig. 2a).
View larger version (41K):
[in this window]
[in a new window]
|
Fig. 2.
Immunofluorescence micrographs of MVEC
(a) and AEC
(b) loaded with low-density
lipoprotein labeled with
1,1'-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine
perchlorate (Dil-Ac-LDL) for 4 h. Both types of endothelial cells (EC)
show uptake, but it was more intense in MVEC.
|
|
MVEC and AEC stained for vimentin (Fig. 3,
a and
d), whereas they lacked desmin and
tropomyosin (Fig. 3, b,
c, e,
and f). Cytokeratin and myosin heavy
chain were also negative in both types of EC (data not shown).
View larger version (47K):
[in this window]
[in a new window]
|
Fig. 3.
Intermediate filaments and tropomyosin in MVEC
(a-c) and AEC
(d-f). Both types of EC show a
fine filamentous network of vimentin
(a and
d). Desmin
(b and
e) and tropomyosin
(c and
f) were negative. Nuclei are red.
|
|
Porcine MVEC demonstrated an intense granular perinuclear staining with
anti-VWF MAb (Fig.
4a). In contrast, a
diffuse cytoplasmic staining with VWF MAb was observed in AEC (Fig.
4d). MVEC were also stained with the
lectin Dolichos biflorus agglutinin. A
cell membrane-associated staining pattern was clearly evident in MVEC as well as in AEC (Fig. 4, b and
e). Platelet EC adhesion molecule-1 (PECAM-1, CD31) was positive for both types of EC (Fig. 4,
c and f; Table 2).
View larger version (61K):
[in this window]
[in a new window]
|
Fig. 4.
Staining with von Willebrand factor (VWF) in MVEC
(a) and AEC
(d), lectin
Dolichos biflorus agglutinin
(b and
e), and CD31
(c and
f). See text for details.
|
|
Because the antibodies for VWF and CD31 were developed against human
antigens, we examined their cross-reactivity in porcine EC in vivo.
Capillary endothelium in porcine myocardium was clearly stained with
anti-human VWF MAb and CD31 (Fig. 5,
a and
b, respectively).
View larger version (64K):
[in this window]
[in a new window]
|
Fig. 5.
Immunohistochemical staining of myocardium and aorta. Cryosections of
myocardium are stained with anti-VWF
(a), anti-CD31
(b), and anti-nitric oxide synthase
(NOS) III antibodies (c).
a: Endothelium of microvessels is
positive for VWF. Note granular labeling pattern in endothelium of
microvessels. b and
c: Endothelium of microvessels is also
positive for both anti-CD31 and anti-NOS III.
d: Negative control.
e: Endothelium of aorta is positive
for anti-NOS III.
|
|
Flow cytometric (fluorescence-activated cell sorter;
FACS) analysis was performed to determine the homogeneity of MVEC by loading the cells with Dil-Ac-LDL and anti-CD31. Compared with unloaded
cells, there was a distinct rightward shift in the FACS profile in both
MVEC and AEC, revealing a high homogeneity of cell preparations from
both types of EC (Fig. 6). Ninety-nine percent of MVEC
and AEC were Dil-Ac-LDL positive (Table 2). It should be noted that the
mean channel fluorescence of AEC was higher than that of MVEC (Fig. 6,
Table 2). Fibroblasts also took up Dil-Ac-LDL, but the mean channel
fluorescence of the loaded cells was less than that of AEC and MVEC.
Furthermore, >99% of AEC and MVEC were positively stained with
anti-CD31. Fibroblasts were negative for CD31 (Fig. 7).
The mean channel fluorescence of MVEC and AEC was identical.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 6.
Flow cytometric analysis of AEC and MVEC loaded with Dil-Ac-LDL (dark
traces) for 4 h at 37°C. Fluorescence was compared with unlabeled
cells (light traces) in a fluorescence-activated cell sorter (FACS).
A: in AEC, >99% of EC were
positively labeled. b: in MVEC, FACS
analysis reveals a distinct rightward shift also of fluorescence
intensity, but intensity of fluorescence is less than that of AEC.
C: fibroblasts.
|
|
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 7.
Flow cytometric analysis of AEC and MVEC stained with an anti-CD31
antibody. Relative fluorescence of stained cells (dark traces) in
comparison to background fluorescence (light traces) of control cells
(stained only with 2nd FITC-labeled antibody). Both AEC
(A) and MVEC
(B) are stained with antibody.
Fluorescence intensity of both types of EC was comparable.
C: fibroblasts.
|
|
-Smooth muscle actin and EC of porcine heart and
aorta.
Immunostaining clearly demonstrated that cultured MVEC stained for
-smooth muscle actin in a stress fiber-like pattern (Fig. 8, a and
b). Neither AEC nor fibroblasts
contained filamentous -smooth muscle actin structures (Fig. 8,
c and
d). MVEC grown on different
substrates, i.e., fibronectin, gelatin, laminin, or collagen
I1 invariably expressed
-smooth muscle actin in an identical pattern.
View larger version (74K):
[in this window]
[in a new window]
|
Fig. 8.
Immunostaining of MVEC and AEC with anti--smooth muscle actin
monoclonal antibody (MAb). Stress fibers consist of -smooth muscle
actin only in cultured MVEC (a and
b). Both AEC
(c) and fibroblasts
(d) are negative.
|
|
In the aorta and coronary arteries (up to arteriolar levels, ~30 mm
in diameter), -smooth muscle actin was present only in smooth muscle
cells underlying EC monolayers but not in endothelium (Fig.
9, a-e). Only
MVEC stained positively for -smooth muscle actin (Fig.
9f). FACS analysis showed that
~98% of MVEC in vitro were positive for -smooth muscle actin,
whereas AEC lacked this type of actin (Fig. 10, Table
2). This result suggests that isolated and cultured MVEC are primarily
of microvascular origin.
View larger version (52K):
[in this window]
[in a new window]
|
Fig. 9.
Immunohistochemical staining of myocardium and aorta with
anti--smooth muscle actin MAb. AEC
(a) and large to small coronary
arteries (b-e) were negative
for -smooth muscle actin. Endothelium of microvessels, e.g.,
capillary endothelium, was positive for -smooth muscle actin
(f).
|
|
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 10.
Flow cytometric analysis of AEC and MVEC stained with anti--smooth
muscle actin MAb. Fluorescence of stained cells (dark traces) in
comparison to fluorescence (light traces) of control cells exposed to
isotype-matched antibody. AEC were negative for -smooth muscle actin
(A). More than 97% of MVEC were
positive (B).
|
|
Double-immunofluorescence staining clearly demonstrated that EC of the
cardiac microvasculature were positively stained for both CD31 and
-smooth muscle actin in vivo (Fig. 11,
a-c). Double immunostaining of
cultured MVEC also showed that -smooth muscle actin-containing
stress fibers were definitely present in isolated and cultured MVEC
(Fig. 11, d-f). This evidence
confirmed that MVEC of the porcine heart possess -smooth muscle
actin in vivo as well as in vitro.
View larger version (75K):
[in this window]
[in a new window]
|
Fig. 11.
Double immunostaining. Myocardium
(a-c) and cultured MVEC
(d-f) were double immunolabeled
for -smooth muscle actin (a and
d) and endothelial markers CD31
(b) and VWF
(c).
c and
f: Merged images of
a and
b and
d and
e, respectively, with yellow marking
regions in which signals overlap.
|
|
NOS III and porcine microvascular EC.
NOS III (endothelial NOS) was present in the microvasculature of
myocardium and the EC in aorta (Fig. 5,
c and
e). All AEC showed characteristic
staining for NOS III in the plasmalemma and the Golgi complex (Fig.
12f). Figure 12,
a-e, demonstrates the staining
pattern for NOS III observed in MVEC primary isolates. MVEC forming
monolayers did not express NOS III at subconfluence, but overgrowing EC
demonstrated its characteristic staining pattern (Fig.
12b). Interestingly, MVEC forming
monolayers began to express perinuclear NOS III after they became
confluent.
View larger version (70K):
[in this window]
[in a new window]
|
Fig. 12.
Immunostaining of MVEC stained with anti-NOS III MAb. Primary isolates
of subconfluent (a-c) and
confluent (d and
e) MVEC were stained with anti-NOS
III antibody. EC forming a monolayer are negative for NOS III. MVEC
growing over monolayer were strongly positive for NOS III. At
confluence, monolayer-forming MVEC are weakly positive for NOS III
(e).
f: AEC are weakly positive for NOS
III.
|
|
Figure 13 demonstrates that 99% of AEC expressed NOS
III at confluence, with a peak fluorescence intensity of ~50
arbitrary units, whereas fibroblasts did not express NOS III. In
isolated MVEC, two different cell populations were evident in terms of NOS III expression, 1) the MVEC
population in which ~90% of MVEC were included and
2) the small cell population with
peak fluorescence intensity of ~50 arbitrary units. Mean fluorescence
intensity of the first MVEC population was greater than that of
nonstained cells (P < 0.0001), but a
significant overlap was apparent. Fluorescence intensity of the first
cell population was decreased during culture, and the MVEC after the
first and second passages did not express NOS III at confluence. The
mean channel fluorescence and the number of cells expressing NOS III
calculated by FACS analysis are summarized in Table 3.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 13.
Flow cytometric analysis of AEC and MVEC stained with an anti-NOS III
antibody. Fluorescence of stained cells (black area) in comparison to
that of control cells exposed to an isotype-matched antibody instead of
anti-NOS III. Confluent MVEC of primary isolates
(A), after 1st passage
(B), and after 2nd passage
(C) were used.
D: AEC.
E: fibroblasts.
|
|
With MAb directed against human NOS III, denaturing PAGE analysis
revealed that whole cell extracts of both AEC and MVEC contained a
protein that was consistent in size (~135 kDa) with NOS III (Fig.
14A). NOS III was
more abundant in AEC than in MVEC homogenates, confirming the results
obtained by FACS analysis and immunostaining. Passage did not affect
the amount of NOS III in confluent AEC (Fig.
14B). In cultured MVEC, however, NOS
III expression was decreased. Both types of cultured EC consistently
expressed VWF, indicating that these cells did not lose their
endothelial character during culture.
View larger version (92K):
[in this window]
[in a new window]
|
Fig. 14.
A: Western blot analysis shows difference in amount of NOS III from
different cell types. Lane 1, AEC;
lane 2, MVEC; lane
3, smooth muscle cells; lane
4, fibroblasts; lane
5, human umbilical vein EC.
B: Western blot analysis shows effect
of passage on expression of NOS III
(top) and VWF
(bottom) in AEC and MVEC.
Lane 1, AEC (passage
4); lane 2, AEC
(passage 10); lane
3, MVEC (primary isolates); lane
4, MVEC (passage 1);
lane 5, MVEC (passage
2). Units for nos. on left are kDa. See text for
details.
|
|
We then assessed NOS activity by measuring conversion of
L-[3H]arginine
to
L-[3H]citrulline
in MVEC and AEC homogenates (Fig. 15). NOS activity of
confluent AEC was four times higher than that of confluent primary
isolates of MVEC. Passage did not affect NOS activity of AEC, but that
of MVEC decreased during subculture. Removal of calcium from the
reaction mixture abolished NOS activity in AEC lysates. However, NOS
activity in MVEC was not affected by the removal of calcium from the
reaction mixture. Furthermore, the NOS activity in MVEC homogenates was
consistently higher in the absence of calcium than in the presence of
calcium.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 15.
Activity of NO synthase in AEC and MVEC at different passages. Cell
homogenates were assayed by
L-[3H]arginine/L-[3H]citrulline
conversion assay in presence or absence of calcium. Data are means ± SD of triplicates from 5 separate experiments. 0, Primary
isolate; 1 and 2, passages 1 and
2.
|
|
| |
DISCUSSION |
Here we describe for the first time the efficient isolation and
characterization of MVEC from porcine heart. The high yield (~2 × 107 MVEC at confluence of
primary isolates) and the high purity (<1% non-EC) of this cell
preparation provide sufficient amounts of cells for functional studies.
A large body of work concerning EC physiology and the molecular
mechanism related to EC dysfunction has been performed mainly using
large vessel-derived EC [e.g., human umbilical vein EC,
AEC], but studies examining the difference between macrovascular
and microvascular EC are still limited.
The present report shows several differences between the two types of
EC: the phenotype, the rate of Dil-Ac-LDL uptake, the presence of
-smooth muscle actin filaments, the expression of NOS III and the
calcium dependence of the activity of NOS III.
Methodological considerations.
We used an initial Langendorff perfusion with collagenase-containing
buffer for the purpose of tissue digestion, which may cause some
contamination of the final cell population with EC from larger
epicardial and penetrating coronary arteries. On the other hand, it
could be anticipated that the majority of the cells in the present
study (97%) are of microvascular origin because of the predominance of
microvessels over larger blood vessels in the heart (7, 33).
Contamination from non-EC, however, is a perpetual problem, but in the
present study FACS analysis clearly demonstrated that the MVEC in
culture were extremely homogeneous, confirming data by Nishida et al.
(30) of rat cardiac MVEC but not of cardiac MVEC from diseased human hearts.
The rate of uptake of Dil-Ac-LDL was different between AEC and MVEC,
whereas mean fluorescence intensity of CD31 was identical in both types
of EC. Previously, it has been reported that EC from different organs
differ in their ability to take up Dil-Ac-LDL (34) and that EC from
some organs are unable to incorporate Dil-Ac-LDL (28, 38). The results
presented here may, therefore, indicate a certain heterogeneity of EC
function between AEC and MVEC.
CD31 (PECAM-1, EndoCAM) is one of the most reliable endothelial markers
characteristically expressed at the cell borders of EC in culture. In
the present study, however, both AEC and MVEC demonstrated cytoplasmic
granular staining that may represent the Golgi apparatus. This
particular staining pattern disappeared when the cells were maintained
in culture after confluence, but localization of CD31 at cell-cell
contacts was never observed (data not shown). Recent studies from our
group (40) showing that CD31 is localized at both the luminal and
abluminal sides of vascular endothelium are in good agreement with the
present observation.
VWF is a multimeric glycoprotein essential for hemostasis, but it has
been also used as an EC marker. Porcine cardiac MVEC demonstrated the
granular cytoplasmic staining typical for anti-VWF. This is in contrast
to the diffuse cytoplasmic localization observed in AEC. This may be
explained by the fact that EC from different segments of the
vasculature exhibit distinct functional properties with regard to VWF
synthesis, storage, and polarity of secretion (37). In the porcine
system, VWF is expressed at different levels in cultured EC from
different vessels and also in endothelium of tissue sections from
different organs (15, 36, 46). In fetal pig, EC from the aorta
displayed highest fluorescence intensity whereas EC from myocardium
showed low binding after labeling with anti-VWF antibody (34). Diffuse
cytoplasmic staining for VWF has also been demonstrated in EC from
different organs and species, i.e., canine jugular vein EC (16) and rat
and human cardiac MVEC (30), which confirms our findings.
-Smooth muscle actin in EC.
-Smooth muscle actin has been considered a reliable marker
distinguishing arterial smooth muscle from EC. The conclusion that EC
can express -smooth muscle actin only under particular culture
conditions (22) has been drawn from studies employing transforming
growth factor-
1 (TGF-1) treatment (2, 22), whereas untreated
cardiac EC did not express a smooth muscle actin (22, 33). Under
physiological conditions EC have been reported to express exclusively b
and g-actin. However, in the present study the use of different
substrates did not affect -smooth muscle actin expression in MVEC.
Furthermore, we demonstrated that cardiac MVEC express a smooth muscle
actin in vitro as well as in vivo by double immunolabeling. Of note is
the finding that -smooth muscle actin was present only in EC of the
microvasculature but neither in small and large coronary artery
endothelium nor in aortic endothelium. We therefore suggest for the
first time that -smooth muscle actin can thus be used as a marker
for distinguishing between coronary arterial EC and MVEC in myocardium.
The physiological significance of the presence of -smooth muscle
actin in MVEC is unclear. It has been recognized that -smooth muscle
actin contributes to the contraction and relaxation of cells. Our novel
finding in the present study might suggest another functional role of
-smooth muscle actin. EC have been shown to exert tension forces by
cellular traction when they are cultured on basement membrane matrix,
generating linear distortion that interconnects EC on the matrix (9,
44). EC elongate along the tension lines and form cell-cell contacts.
Therefore, the presence of -smooth muscle actin in MVEC may account
for their angiogenetic ability. Further investigations are necessary to explore the functional role of -smooth muscle actin in cardiac MVEC.
NOS III and microvascular EC.
Different types of cells express NOS III (ecNOS), i.e., EC, neurons,
and cardiac myocytes (10). On the other hand, Balligand et al. (6)
reported that cultured rat cardiac MVEC had no detectable constitutive
NOS activity, although they reported that sustained NOS III expression
can be maintained under particular culture conditions (21).
In the present study, we demonstrate that all MVEC in myocardium
express NOS III. AEC constitutively expressed NOS III, and the number
of passage did not affect NOS III expression and its activity. In MVEC,
however, the isolation procedure and the ensuing culture greatly
affected NOS III expression. The present result might suggest that a
modulating mechanism of the microenvironment of MVEC in vivo would be
necessary for cardiac MVEC to express NOS III. Interestingly, Kelly et
al. (21) reported that sustained NOS III expression can be maintained
in rat cardiac MVEC if seeded at high density. Because we seeded
cardiac MVEC at relatively high density and it took only 4 days for
primary isolates to become confluent, the high purity and high yield of
MVEC in the present report may also account for the sustained NOS III
expression in MVEC.
Andries et al. (1) investigated the distribution of NOS III in cardiac
endothelium using immunofluorescence and en face confocal microscopy of
rat heart and showed differences in subcellular ecNOS distribution. We
conclude from the present work that MVEC can express NOS III to the
same level as large vessel-derived EC.
Experiments in cultured EC have demonstrated that NOS III expression
can be modulated by several factors, i.e., shear stress (43) TGF-1
(18), protein kinase C (32), tumor necrosis factor- (48), oxygen
(31), and the proliferative state (3). Inoue et al. (18) reported that
TGF-1 upregulates NOS III at the transcriptional and protein levels.
Adult cardiomyocytes have been reported to express and secrete TGF-1
in vivo and in vitro (39, 47). TGF-1 produced by cardiomyocytes
might diffuse and reach adjacent MVEC, modulating NOS III expression.
This hypothesis is feasible because cardiac myocytes can control
vascular tone of the adjacent -smooth muscle actin-containing MVEC
and eventually control local blood flow by this paracrine pathway.
Another novel finding in the present study was that the activation of
NOS III in MVEC was calcium independent, whereas NOS III activity in
AEC was completely abolished in the absence of calcium. It
has been suggested that there are two different mechanisms of NO
production in response to shear stress in cultured EC, i.e., a
calcium-dependent and an independent NO production (4, 11, 23). Only
calcium-independent NO release was dependent on shear stress magnitude
(23). In addition, Ayajiki et al. (4) showed that shear stress-induced
NO production is evoked principally through a calcium-independent
mechanism in intact rabbit iliac artery.
Furthermore, Fleming et al. (12) reported that NOS III activity in
caveolae prepared from AEC was completely calcium independent. Caveolae
are plasmalemmal microdomains that have been implicated in several
physiological cell functions, i.e., the transcytosis of micromolecules,
the uptake of small molecules by phagocytosis, and the
compartmentalization of signaling molecules. Recently, it was suggested
that localization of NOS III in caveolae is likely to optimize NOS III
activation and extracellular release of NO. Palmitoylation of NOS III
appears to be necessary for its targeting to caveolae. Furthermore,
microvascular endothelium is abundant in caveolae, compared with large
vessel-derived EC (1, 8, 24). Taken together, it is possible that the
function of NOS III in MVEC is closely associated with its presence in
caveolae and its activation is calcium independent. This hypothesis and our present findings may suggest a distinct physiological role of NOS
III in microvascular endothelium in the heart that is different from
that of large vessel-derived EC. Further investigations are needed to
clarify this attractive hypothesis.
In conclusion, we present in this report successful isolation with a
high cellular yield and purity and establishment of the culture of MVEC
from porcine hearts. Porcine MVEC have unique endothelial
characteristics in terms of -smooth muscle actin and NOS III
expression. MVEC will provide a useful tool to investigate physiological and pathophysiological microvascular EC functions in the heart.
| |
ACKNOWLEDGEMENTS |
The authors thank Brigitte Matzke for excellent technical
assistance, Gunther Schuster for computer reproduction of the
immunohistochemical illustrations, and Gerhard Stämmler for
preparation of the manuscript.
| |
FOOTNOTES |
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement"
in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Address for reprint requests and other correspondence: J. Schaper,
Dept. of Experimental Cardiology, Max Planck Institute,
Benekestr. 2, D-61231 Bad Nauheim, Germany (E-mail:
j.schaper{at}kerckhoff.mpg.de).
Received 14 September 1998; accepted in final form 19 January
1999.
| |
REFERENCES |
1.
Andries, L. J.,
D. L. Brutsaert,
and
S. U. Sys.
Nonuniformity of endothelial constitutive nitric oxide synthase distribution in cardiac endothelium.
Circ. Res.
82:
195-203,
1998[Abstract/Free Full Text].
2.
Arciniegas, E.,
A. B. Sutton,
T. D. Allen,
and
S. A. M.
Transforming growth factor beta 1 promotes the differentiation of endothelial cells into smooth muscle-like cells in vitro.
J. Cell Sci.
103:
521-529,
1992[Abstract].
3.
Arnal, J. F.,
J. Yamin,
S. Dockery,
and
D. G. Harrison.
Regulation of endothelial nitric oxide synthase mRNA, protein, and activity during cell growth.
Am. J. Physiol.
267 (Cell Physiol. 36):
C1381-C1388,
1994[Abstract/Free Full Text].
4.
Ayajiki, K.,
M. Kindermann,
M. Hecker,
I. Fleming,
and
R. Busse.
Intracellular pH and tyrosine phosphorylation but not calcium determine shear stress-induced nitric oxide production in native endothelial cells.
Circ. Res.
78:
750-758,
1996[Abstract/Free Full Text].
5.
Balligand, J.,
R. Kelly,
and
T. Smith.
Cardiac endothelium and tissue growth.
Prog. Cardiovasc. Dis.
39:
351-360,
1997[Medline].
6.
Balligand, J. L.,
D. Ungureanu-Longrois,
W. W. Simmons,
L. Kobik,
C. J. Löwenstein,
S. Lamus,
R. A. Kelly,
T. W. Smith,
and
T. Michel.
Induction of NO synthase in rat cardiac microvascular endothelial cells by IL-1 and IFN-
.
Am. J. Physiol.
268 (Heart Circ. Physiol. 37):
H1293-H1303,
1995[Abstract/Free Full Text].
7.
Bassingthwaighte, J. B.,
T. Yipintsoi,
and
R. B. Harvey.
Microvasculature of the dog left ventricular myocardium.
Microvasc. Res.
7:
229-249,
1974[Medline].
8.
Belloni, P. N.,
D. H. Cardney,
and
G. L. Nicolson.
Organ derived microvessel endothelial cells exhibit differential responsiveness to thrombin and other growth factors.
Microvasc. Res.
43:
20-45,
1992[Medline].
9.
Davis, G. E.,
and
C. W. Camarillo.
Regulation of endothelial cell morphogenesis by integrins, mechanical forces, and matrix guidance pathways.
Exp. Cell Res.
216:
113-123,
1995[Medline].
10.
Feron, O.,
L. Belhassen,
L. Kobzik,
T. W. Smith,
R. A. Kelly,
and
T. Michel.
Endothelial nitric oxide synthase targeting to caveolae: specific interactions with caveolin isoforms in cardiac myocytes and endothelial cells.
J. Biol. Chem.
271:
22810-22814,
1996[Abstract/Free Full Text].
11.
Fleming, I.,
J. Bauersachs,
and
R. Busse.
Calcium-dependent and calcium-independent activation of the endothelial NO synthase.
J. Vasc. Res.
34:
165-174,
1997[Medline].
12.
Fleming, I.,
B. Fisslthaler,
T. Benzing,
and
R. Busse.
The calcium-independent activation of the endothelial NO synthase (NOS III) is associated with its dephosphorylation and subsequent redistribution within the caveolae.
Vasc. Biol.
34:
165-174,
1997.
13.
Garlanda, C.,
and
E. Dejana.
Heterogeneity of endothelial cells. Specific markers.
Arterioscler. Thromb. Vasc. Biol.
17:
1193-1202,
1997[Abstract/Free Full Text].
14.
Gersh, B.,
E. Braunwald,
and
J. Rutherford.
Chronic coronary artery disease.
In: Heart Disease, edited by E. Braunwald. Philadelphia, PA: Saunders, 1997, p. 1289-1365.
15.
Giddings, J. C.,
A. L. Jarvis,
and
A. L. Bloom.
Differential localization and synthesis of porcine factor VIII related antigen (VIIIR: Ag) in vascular endothelium and in endothelial cells in culture.
Thromb. Res.
29:
299-312,
1983[Medline].
16.
Gorfien, S.,
A. Spector,
D. Deluca,
and
S. Weiss.
Growth and physiological functions of vascular endothelial cells in a new serum-free medium.
Exp. Cell Res.
206:
291-301,
1993[Medline].
17.
Hudlicka, O.
What makes blood vessels grow?
J. Physiol. (Lond.)
444:
1-24,
1991[Free Full Text].
18.
Inoue, N.,
R. C. Venema,
H. S. Sayegh,
Y. Ohara,
T. J. Murphy,
and
D. G. Harrison.
Molecular regulation of the bovine endothelial cell nitric oxide synthase by transforming growth factor-1.
Arterioscler. Thromb. Vasc. Biol.
15:
1255-1261,
1995[Abstract/Free Full Text].
19.
Kassab, G. S.,
and
Y. C. B. Fung.
Topology and dimensions of pig coronary capillary network.
Am. J. Physiol.
267 (Heart Circ. Physiol. 36):
H319-H325,
1994[Abstract/Free Full Text].
20.
Kassab, G. S.,
C. A. Rider,
N. J. Tang,
and
Y. B. Fung.
Morphometry of pig coronary arterial trees.
Am. J. Physiol.
265 (Heart Circ. Physiol. 34):
H350-H365,
1993[Abstract/Free Full Text].
21.
Kelly, R. A.,
J. L. Balligand,
and
T. W. Smith.
Nitric oxide and cardiac function.
Circ. Res.
79:
363-380,
1996[Free Full Text].
22.
Kocher, O.,
and
J. A. Madri.
Modulation of actin mRNAs in cultured vascular cells by matrix components and TGF-1.
In Vitro Cell. Dev. Biol.
25:
424-434,
1989[Medline].
23.
Kuchan, M. J.,
and
J. A. Frangos.
Role of calcium and calmodulin in flow-induced nitric oxide production in endothelial cells.
Am. J. Physiol.
266 (Cell Physiol. 35):
C628-C636,
1994[Abstract/Free Full Text].
24.
Kumar, S.,
D. West,
and
A. Ager.
Heterogeneity in endothelial cells from large vessels and microvessels.
Differentiation
36:
57-70,
1987[Medline].
25.
Li, K.,
J. Rouleau,
J. Andries,
and
D. Brutsaert.
Effect of dysfunctional vascular endothelium on myocardial performance in isolated papillary muscle.
Circ. Res.
72:
768-777,
1993[Abstract/Free Full Text].
26.
Marijianowski, M. M. H.,
M. van Laar,
J. Bras,
and
A. E. Becker.
Chronic congestive heart failure is associated with a phenotypic shift of intramyocardial endothelial cells.
Circulation
92:
1494-1498,
1995[Abstract/Free Full Text].
27.
Mohri, M.,
and
W. Schaper.
Angiogenesis in porcine heart with microembolization.
In: Collateral Circulation; Heart, Brain, Kidney, Limbs, edited by W. Schaper,
and J. Schaper. Amsterdam: Kluwer Academic, 1992.
28.
Moyer, C. F.,
P. A. Dennis,
G. Majno,
and
I. Joris.
Venular endothelium in vitro: isolation and characterization.
In Vitro Cell. Dev. Biol.
l24:
359-368,
1988[Medline].
29.
Nees, S.,
A. L. Gerbes,
and
E. Gerlach.
Isolation, identification, and continuous culture of coronary endothelial cells from guinea pig hearts.
Eur. J. Cell Biol.
24:
287-297,
1981[Medline].
30.
Nishida, M.,
W. W. Carley,
M. E. Gerritsen,
Ellingsen,
R. A. Kelly,
and
T. W. Smith.
Isolation and characterization of human and rat cardiac microvascular endothelial cells.
Am. J. Physiol.
264 (Heart Circ. Physiol. 33):
H639-H652,
1993[Abstract/Free Full Text].
31.
North, A. J.,
K. S. Lau,
T. S. Brannon,
L. C. Wu,
L. B. Wells,
Z. German,
and
P. W. Shaul.
Oxygen upregulates nitric oxide synthase gene expression in ovine fetal pulmonary artery endothelial cells.
Am. J. Physiol.
270 (Lung Cell. Mol. Physiol. 14):
L643-L649,
1996[Abstract/Free Full Text].
32.
Ohara, Y.,
H. S. Sayegh,
J. J. Yamin,
and
D. G. Harrison.
Regulation of endothelial constitutive nitric oxide synthase by protein kinase C.
Hypertension
25:
415-420,
1995[Abstract/Free Full Text].
33.
Piper, H. M.,
R. Spahr,
S. Mertens,
A. Kreutzfeldt,
and
H. Watanabe.
Microvascular endothelial cells from heart.
In: Cell Culture Techniques in Heart and Vessel Research, edited by H. M. Piper. Berlin: Springer, 1990, p. 158-177.
34.
Plendl, J.,
C. Neumuller,
A. Vollmar,
R. Auerbach,
and
F. Sinowatz.
Isolation and characterization of endothelial cells from different organs of fetal pigs.
Anat. Embryol. (Berl.)
194:
445-456,
1996[Medline].
35.
Ramaciotti, C.,
A. Sharkey,
G. McClellan,
and
S. Winegrad.
Endothelial cells regulate cardiac contractility.
Proc. Natl. Acad. Sci. USA
90:
4033-4036,
1993.
36.
Rand, J. H.,
L. Badimon,
R. E. Gordon,
R. R. Uson,
and
V. Fuster.
Distribution of von Willebrand factor in porcine intima varies with blood vessel type and location.
Arteriosclerosis
7:
287-291,
1987[Abstract/Free Full Text].
37.
Ruggeri, Z. M.,
and
J. Ware.
von Willebrand factor.
FASEB J.
7:
308-316,
1993[Abstract].
38.
Rupnick, M. A.,
A. Carey,
and
S. K. Williams.
Phenotypic diversity in cultured cerebral microvascular endothelial cells.
In Vitro Cell. Dev. Biol.
24:
435-444,
1988[Medline].
39.
Schlüter, K. D.,
X. J. Zhou,
and
H. M. Piper.
Induction of hypertrophic responsiveness to isoproterenol by TGF- in adult rat cardiomyocytes.
Am. J. Physiol.
269 (Cell Physiol. 38):
C1311-C1316,
1995[Abstract/Free Full Text].
40.
Scholz, D.,
and
J. Schaper.
Platelet/endothelial cell adhesion molecule-1 (PECAM-1) is localized over the entire plasma membrane of endothelial cells.
Cell Tissue Res.
290:
623-631,
1997[Medline].
41.
Simionescu, M.,
and
N. Simionescu.
Isolation and characterization of endothelial cells from the heart microvasculature.
Microvasc. Res.
16:
426-452,
1978[Medline].
42.
Tomanek, R.
Response of the coronary vasculature to myocardial hypertrophy.
J. Am. Coll. Cardiol.
15:
528-533,
1990[Abstract].
43.
Uematsu, M.,
Y. Ohara,
J. P. Navas,
K. Nishida,
T. J. Murphy,
R. W. Alexander,
R. M. Nerem,
and
D. G. Harrison.
Regulation of endothelial cell nitric oxide synthase mRNA expression by shear stress.
Am. J. Physiol.
269 (Cell Physiol. 38):
C1371-C1378,
1995[Abstract/Free Full Text].
44.
Vernon, R. B.,
J. C. Angello,
M. L. Iruela-Arispe,
T. F. Lane,
and
E. H. Saga.
Reorganization of basement membrane matrices by cellular traction promotes the formation of cellular networks in vitro.
Lab. Invest.
66:
536-547,
1992[Medline].
45.
Voyta, J. C.,
D. P. Via,
C. E. Butterfield,
and
B. R. Zetter.
Identification and isolation of endothelial cells based on their increased uptake of acetylated-low density lipoprotein.
Cell Biol
99:
2034-2040,
1984.
46.
Wu, Q. Y.,
L. Drouet,
J. L. Carrier,
C. Rothschild,
M. Berard,
C. Rouault,
J. P. Caen,
and
D. Meyer.
Differential distribution of von Willebrand factor in endothelial cells.
Arteriosclerosis
7:
47-54,
1987[Abstract].
47.
Wünsch, M.,
H. S. Sharma,
T. Markert,
S. Bernotat-Danielowski,
R. J. Schott,
P. Kremer,
N. Bleese,
and
W. Schaper.
In situ localization of transforming growth factor 1 in porcine heart: enhanced expression after chronic coronary artery constriction.
J. Mol. Cell. Cardiol.
23:
1051-1062,
1991[Medline].
48.
Yoshizumi, M.,
M. A. Perrella,
J. C. Burnett,
and
M. E. Lee.
Tumor necrosis factor downregulates an endothelial nitric oxide synthase mRNA by shortening its half-life.
Circ. Res.
73:
205-209,
1993[Abstract].
Am J Physiol Heart Circ Physiol 276(5):H1755-H1768
0002-9513/99 $5.00
Copyright © 1999 the American Physiological Society
TOP
ABSTRACT
INTRODUCTION
