(Hypertension. 2001;37:1465.)
© 2001 American Heart Association, Inc.
Scientific Contributions |
From INSERM U337 (H.B., E.S., M.S.) and INSERM U465 (G.D.), Faculty Broussais-Hotel Dieu; and CNRS, Marie Lannelongue Hospital (H.B., E.S., C.R.-M., J.-F.R.), Paris, France.
Correspondence to Michel Safar, Hôpital Broussais, Service de Médecine Interne 1.0, 96 rue Didot, 75014 Paris, France.
| Abstract |
|---|
|
|
|---|
Key Words: extracellular matrix collagen fibronectin vitronectin angiotensin II muscle, smooth, vascular calcium
| Introduction |
|---|
|
|
|---|
Angiotensin (Ang) II is involved in a number of pathophysiological processes, including cell proliferation and hypertrophy and production of ECM elements. Although the area of Ang II signaling is an area of intensive research, less effort has been devoted to studying the regulation of Ang IIreceptor signaling by the ECM. In this study, we examined the modulation of Ang IIinduced Ca2+ release from internal stores and Ca2+ influx by ECM components in SHR and normotensive Wistar-Kyoto rats (WKY).
| Materials |
|---|
|
|
|---|
The investigation was conducted under guidelines established by the Guide for the Care and Use of Laboratory Animals (NIH publication No. 93-23, revised 1985). VSMCs were isolated from aorta by enzymatic digestion and cultured in Dulbeccos modified Eagles medium (DMEM, Eurobio) as described previously.16 17 Cells were cultured on glass coverslips coated with either collagen type I (10 µg/mL, Sigma), collagen type IV (7 µg/mL, Sigma), vitronectin (0.1 µg/mL, Sigma), fibronectin (3 µg/mL, Sigma), or Matrigel (1:10 dilution, Harbor Bio-Products). Cells at confluence between passages 3 and 9 were incubated in 0.5% FCS medium for 48 hours before the experiments. Cells were seeded at a high density of 1.5x105/mL in culture flasks and 1.5x105/mL in 12-mm wells.
Cell Ca2+
Measurements
Ang IIinduced
[Ca2+]i variations
(1 µmol/L) were assessed at single-cell level by fluorescence
imagery as described
previously.17 Cells loaded
with fura 2 and superfused with Na+-HEPES
solution ([in mmol/L] NaCl 140, KCl 4.5,
MgSO4 0.8,
KH2PO4 0.8,
CaCl2 1.0, glucose 5.6, and HEPES 5.6). The
ratio of the emitted light (520 nm) at each of the excitation
wavelengths (350 and 380 nm) was plotted against time for each single
cell. Because calibration procedures are prone to
errors,18 no attempt was
made to assess absolute free
[Ca2+]i
concentrations. Therefore, qualitative changes in
[Ca2+]i are
represented by changes in the ratio of the emitted
fluorescence.19
Assessing the Effect of Ang II on Cell
Ca2+
Exposure of VSMCs to Ang II induced a transient
increase in [Ca2+]i
(Figure 1A), with a maximal response observed in both strains
for concentrations of Ang II >0.5 µmol/L, as previously
reported.17 20
The number of cells responding to Ang II increases with ligand
concentration. Thus, at
10-7 mol/L,
the number of responding cells varies from batch to batch and
rarely exceeded 60%. In contrast, at
10-6
mol/L, the number of responding cells is >90%. Thus,
pharmacological concentrations of Ang II (1 µmol/L) were used
throughout the study. In such conditions, variations in
Ca2+ release from internal stores and
Ca2+ influx can be properly assessed.
[Ca2+]i transient
was characterized by (1) the amplitude of
[Ca2+]i release
(peak minus basal values;
Figure 1A, a and b); (2) the slope of
[Ca2+]i increase
(Figure 1, a and b); and (3) the total
[Ca2+]i mobilized
(area under transient curve;
Figure 1A, a through c). No difference in the Ang
IIinduced [Ca2+]i
increase could be observed in cells between passages 3 and 9, in accord
with previous
studies.17 21
|
To assess Ca2+ release from internal stores and Ca2+ influx, cells were superfused with a Ca2+-free Na+-HEPES medium and exposed to Ang II for 120 seconds, and then 1 mmol/L CaCl2 was added to the Ca2+-free medium (Figure 1B).17 21 The [Ca2+]i transient was characterized by its amplitude, slope, and total [Ca2+]i released, as described above.
Effect of Thapsigargin on Cell
Ca2+
Several
studies22 23 24
have shown that Ca2+ store depletion,
independently of agonist stimulation of inositol 1,4,5-tris-phosphate
(IP3), activates a
Ca2+ entry mechanism. This can be achieved
by the addition of thapsigargin, an agent that inhibits the
Ca2+-ATPase. Thapsigargin (3 µmol/L,
Sigma) was added to VSMCs in the absence of
Ca2+ in the external medium, and
Ca2+ influx was estimated from the initial
rate of [Ca2+]i
increase on reintroduction of external
Ca2+.
Cell Cycle Analysis
Cells were fixed in ice-cooled 70% ethanol,
harvested by centrifugation, washed once with PBS, and
resuspended in 500 µL of PBS containing Rnase (200 µg/mL) and
phosphatidylinositol (50 µg/mL), then incubated at 37°C for 30
minutes.
Analyses were performed with a Coulter Elite-ESP flow cytometer (Beckman-Coulter) using a 15-mW air-cooled argon-ion laser tuned at 488 nm. Propidium iodide fluorescence was collected through a 620-nm band-pass filter and displayed on a linear scale.
Cells were analyzed at a rate of 100 to 200 cells/second; doublets were eliminated on the basis of DNA peak versus DNA area signals. Each analysis was performed on at least 10 000 cells after doublet discrimination.
Immunochemistry
Indirect immunofluorescence was
performed on cultured VSMCs fixed in 4% formaldehyde for 10 minutes
and then permeabilized with 0.2% Triton X-100 at room
temperature. Cells were washed twice with PBS (pH 7.4) and incubated in
the presence of 5% BSA (20 minutes, room temperature). Cells were left
overnight at 4°C in the presence of monoclonal antibodies for
paxillin (1/400, Sigma-Aldrich) and followed by addition of
biotinylated horse anti-mouse IgG antibodies (1/30, Vector
Laboratories) in PBS containing rat serum (1/200) and streptavidin
Texas red (1/30, Amersham) in PBS. After a final wash in PBS,
coverslips were mounted in mounting medium (Fluoprep,
Merieux).
Data Analysis
Comparison of mean values was performed by ANOVA with
multiple testing according to the Bonferroni method, and by Students
unpaired t test when
appropriate. The results were expressed as percentage of the control
value (the fluorescence ratio units of treated cells per the
fluorescence ratio units of untreated
cells).
| Results |
|---|
|
|
|---|
|
|
|
Collagen I
When cells from SHR were grown on collagen I, no effect
could be observed on Ang IIinduced Ca2+
increase in the presence of external Ca2+
(compared with cells grown on uncoated glass coverslips). This was
associated with a significant decrease in
Ca2+ release from internal stores and with
an increase in Ca2+ influx. Conversely, in
WKY, a significant increase in the Ang IIinduced
Ca2+ mobilization in the presence of
external Ca2+ could be observed. This was
linked to an increase in Ca2+ release from
internal stores, whereas no effect could be observed on
Ca2+ influx.
Collagen IV
Ang IIinduced Ca2+
increase in the presence of external Ca2+
was not modified when cells from SHR were grown on collagen IV. In the
absence of external Ca2+, a decrease in
[Ca2+]i release was
observed and compensated for by an increase in
Ca2+ influx. In contrast, in WKY, no effect
could be observed on Ca2+ transport
processes, in the presence or absence of external
Ca2+ or on Ca2+
influx.
Fibronectin
In SHR, fibronectin induced a decrease in Ang
IIinduced Ca2+ mobilization, linked to a
decrease in Ca2+ release from internal
stores, with no change in Ca2+ influx. In
contrast, in WKY, no effect on the amplitude or slope could be observed
in the presence of Ca2+, whereas total
Ca2+ mobilized was decreased. This was not
related to a decrease in Ca2+ release from
internal stores but could be related either to a decrease in
Ca2+ influx or to an increase in
Ca2+ recapture or extrusion
mechanisms.
Vitronectin
A significant inhibition in the Ang IIinduced
Ca2+ increase in the presence of external
Ca2+ was observed when cells from SHR were
grown on vitronectin. This was consequent to a decrease in
Ca2+ released from internal stores, whereas
Ca2+ influx was not altered. In WKY, a
decrease in Ang II effect was observed in the presence of external
Ca2+. This is related to a decrease in
Ca2+ influx, whereas
Ca2+ release from internal stores was not
modified.
Matrigel
In the SHR, Matrigel induced a significant decrease in
Ang IIinduced Ca2+ mobilization, related
to a decrease in Ca2+ release from internal
stores, despite an increase in Ca2+ influx.
This could be related to the Matrigel-induced increase in basal
Ca2+ observed in this strain
(Table 1). In contrast, in WKY, a significant increase in
Ca2+ transient in the presence of external
Ca2+ was observed. This was associated with
a significant stimulation of Ca2+ influx,
whereas release from Ca2+ stores remained
unchanged.
Comparison of the Ang IIInduced
Ca2+ Response in SHR and WKY
We next compared Ang IIinduced
Ca2+ increase observed in the presence of
external Ca2+ in cells from SHR and WKY
grown on the same ECM component
(Figure 2). SHR cells grown on either collagen I, collagen
IV, vitronectin, fibronectin, or uncoated glass coverslips
showed significant increases in the amplitude of the Ang IIinduced
Ca2+ increase compared with those of WKY. In
contrast, no difference between the 2 strains could be observed when
cells were grown on Matrigel. The total amount of
Ca2+ released showed a significant increase
in SHR compared with that in WKY when cells were grown on uncoated
glass coverslips, fibronectin, or vitronectin, but not on
collagen I, collagen IV, or Matrigel.
|
Effect of Thapsigargin
Thapsigargin (3 µmol/L) in the absence of
Ca2+ in the external medium induced a
transient response in cells from both WKY and SHR, followed by a
decrease in cell Ca2+, which is probably
related to Ca2+ extrusion from the cell
(Figure 3). The reintroduction of
Ca2+ to the medium induced a
Ca2+ influx of a magnitude like that
observed after Ang II (WKY, 0.552±0.017 ratio unit/min, n=58; SHR,
1.002±0.041 ratio unit/min, n=62; SHR versus WKY,
P=0.01). Incubation of VSMCs
with thapsigargin for 5 minutes abolished the response to subsequent
infusion of Ang II in both strains, as previously reported by Cortes et
al20 (data not shown).
Addition of ionomycin did not elicit any increase in cell
Ca2+, suggesting that thapsigargin
completely emptied intracellular Ca2+ stores
(data not shown).
|
We have assessed the effect of ECM components on thapsigargin-induced Ca2+ influx only in conditions in which these components modified the Ang IIinduced Ca2+ influx. Thus, in SHR, no effect on the influx induced by thapsigargin was observed with collagen I (102±5% of control values, P=NS), whereas collagen IV increased the influx (235±15%, P<0.0001) and Matrigel decreased it (50±3%, P<0.0001). In WKY, fibronectin, vitronectin, and Matrigel increased thapsigargin-induced Ca2+ influx (165±9%, 131±5%, and 143±6% of control values, respectively, P<0.0001 for each).
Cell Cycle Analysis
Cells were grown on either collagen type I or IV,
fibronectin, vitronectin, or Matrigel, and the number of
cells in each phase of cell
cycle29 was compared with
the number of cells grown on uncoated glass coverslips
(Table 4). In SHR, no effect could be observed on the number
of cells in each phase of cell cycle for cells grown on different ECM
proteins, except for the cells grown on Matrigel, which had an increase
in the percentage of cells in
pre-G0/G1 phase.
Conversely, in WKY, different ECM proteins have an effect on cell
cycle, except for vitronectin, for which no effect could be
observed. Collagen I and collagen IV induced a decrease in the
percentage of cells in S phase, whereas fibronectin induced an increase
in the G2/M phase. Matrigel induced an increase in the percentage of
cells in the pre-G0/G1
and G0/G1 phases and a
decrease in the S and G2/M
phases.
|
Paxillin Organization in VSMCs Grown on
Different ECM Proteins
Immunofluorescent micrographs of VSMCs stained
with antibodies against paxillin showed that the distribution of
paxillin is different in SHR compared with that in WKY; furthermore,
ECM proteins modified this distribution.
The paxillin-staining profile in WKY cells grown on glass coverslips appears as dots at the cell boundaries (Figure 4A), whereas in SHR the staining profile is more dense and appears as thicker dotted lines originating at the boundaries or fine dots in the center of the cells (Figure 4B). This density further increased in cells from both strains grown on collagen I (Figure 4C and 4D), with a diffuse distribution in WKY cells (Figure 4C). In contrast, in cells grown on Matrigel, paxillin appears in a beltlike fashion at the periphery of the cell, with a higher density in SHR (Figure 4F) than WKY (Figure 4E).
|
| Discussion |
|---|
|
|
|---|
In contrast, fibronectin and vitronectin induced a decrease in Ang IIinduced Ca2+ influx in WKY, whereas no effect could be observed in SHR. Conversely, collagen I and collagen IV induced an increase in this influx in SHR but not in WKY, whereas Matrigel increased the influx in both strains.
A large body of experimental data has accumulated in the past 15 years that indirectly suggests that intracellular calcium regulation is defective in the VSMCs of SHR compared with the VSMCs of WKY. In most of these studies, the effect of agonist on Ca2+ mechanisms was assessed in cells grown on either glass coverslips or culture flasks. The present findings show that the amplitude of Ang IIinduced Ca2+ increase is significantly higher in SHR than WKY when cells are cultured on either collagen I, collagen IV, fibronectin, or vitronectin, in accordance with previous observations obtained on glass coverslips.20 25 26 27 28 30 31 32 No difference could be observed, however, when cells were grown on Matrigel. These results raise the important question of the choice of artificial system to mimic in vivo conditions for data to be interpreted in a way that is meaningful and applicable to intact vessels. This, added to the use of high concentrations of Ang II to adequately assess Ca2+ release or influx, calls for caution in the extrapolation of results obtained in cultured cells to VSMC properties in vivo.
In VSMCs, binding of Ang II to AT1 receptors is known to release [Ca2+]i from internal stores and to increase Ca2+ influx, thus resulting in transient [Ca2+]i increase. The signaling pathways involved in Ca2+ release from internal stores have been characterized and involve IP3 binding to its receptor on the endoplasmic reticulum. Functional studies33 also provide evidence for ryanodine-sensitive [Ca2+]i pools and a cross talk between these 2 pools. Agonist-stimulated release of [Ca2+]i from the intracellular stores is accompanied by repletion of the store by Ca2+ influx from the extracellular space.34 It is now acknowledged that the action of Ang II in VSMCs involves both voltage-operating channels and voltage-independent channels.35 36 The relative contributions of these 2 pathways to either muscle contraction or Ca2+ influx depends on the smooth muscle type and the experimental conditions.34 37 38 We have previously reported that in these cells, Ca2+ influx was mediated by voltage-independent channels, because nifedipine was without effect on the influx with no participation of Na+-Ca2+ exchanger.21 Furthermore, Ca2+ store depletion by thapsigargin, an agent that inhibits the Ca2+-ATPase, activates Ca2+ entry, as observed in this study and others.22 23 24 This pathway, called capacitative Ca2+ entry, was reported in different cell types,39 including VSMCs.40 41 42 It was insensitive to nifedipine42 and contributed to smooth muscle contraction (reviewed by Gibson et al43 ).
Previous studies have shown that ECM modulates cell
Ca2+ release from internal
stores44 45 and
the opening of Ca2+
channels.9 10 46 47
Cell surface receptors for ECM are frequently members of the integrin
family, transmembrane glycoproteins containing an
- and
a ß-subunit. Integrin-mediated signaling involves tyrosine kinase
activity associated with focal adhesions. Currently, over 20 different
ß-combinations are known, each of which exhibits a different
ligand-binding profile. Integrins were found to be abnormal in SHR,
with a decrease in
5ß1-integrins in the
young SHR arteries and an increase in
5ß1-integrins in adult
SHR.14 After binding to ECM,
integrins bind to the cytoskeleton and promote its reorganization. The
actin network has been proposed to play a role in this regulatory
process, and its reorganization could be responsible for the effects
observed in this study. In this regard, we have previously
reported17 that actin plays
a major role in the regulation of Ang IIinduced
Ca2+ release from internal stores in SHR but
not in WKY.
The present results show that growing cells from SHR on different ECM proteins induced a significant decrease in Ang IIinduced Ca2+ release from internal stores, thus suggesting in this strain a cross talk in the signaling pathways of G proteincoupled receptors and those elicited by adhesion to ECM proteins. This would possibly result in a decrease in 1,4,5-IP3 release or a decrease in RYR receptors. In contrast, in WKY, no such interaction could be observed except for collagen I, which increases Ang IIinduced Ca2+ release. One working hypothesis to explain these results could be that cytoskeletal organization induced by binding of SHR cells to ECM proteins, or WKY cells to collagen I, regulates phosphoinositide metabolism or the spatial relationship between phospholipase C and IP3 receptors. In this regard, Kraus-Friedmann33 proposed that the interaction of IP3 with its receptor could induce a conformational change in the cytoskeleton, sensed by the ryanodine-binding Ca2+ channel and resulting in its opening. An alternative mechanism could be that actin organization regulated the activation of IP3 receptors in the endoplasmic reticulum, thus impairing Ca2+ release from storage pools48 49 (see discussion in Samain et al17 ). The present data do not allow differentiation between these mechanisms. In contrast, the stimulation of Ca2+ release from internal stores in WKY cells observed on collagen I could be related to an increase in the number of cells in G2/M phase. In this regard, several studies50 51 stressed the role of Ca2+ ions in mitogenesis and that IP3 receptors may be responsible for the majority of Ca2+ release during cell proliferation. The fact that Matrigel induced a significant decrease in the number of cells in the S and G2/M phases without affecting Ca2+ release in this strain, however, argues against this hypothesis.
In contrast, actin filaments are important for function of ligand-activated calcium-permeable channels. We have previously shown that Ang IIinduced Ca2+ influx was inhibited by the disorganization of actin filaments in SHR, whereas no effect could be observed in WKY. The present results showed that fibronectin and vitronectin induced a decrease in Ang IIinduced Ca2+ influx in WKY, whereas no effect could be observed in SHR. Conversely, collagen I and collagen IV induced an increase in this influx in SHR but not in WKY. Furthermore, this study showed that the effect of thapsigargin and Ang II on Ca2+ influx were not similar in SHR cells cultured on collagen I or Matrigel or in WKY cells cultured on fibronectin or vitronectin. This suggests a difference in the modulation of the 1,4,5-IP3 signaling pathway and the Ca2+-induced Ca2+ release by these proteins and could be related to the organization profile of focal adhesion as shown in results of this study obtained in cells grown on collagen I and Matrigel.
Several data clearly indicate that integrins can signal other intracellular events in addition to cytoskeletal organization. Evidence exists for effects of integrin-ligand engagement on phospholipid metabolism, tyrosine phosphorylation,52 and MAP kinases5 53 that would in turn affect G-protein signaling to Ca2+ stores. In this context, it is possible that the intracellular events triggered via these adhesion receptors interact and synergize with those triggered by G proteincoupled receptors. The interaction of paxillin, a constituent of focal adhesions, with focal adhesion kinase (FAK) might coordinate the signals from focal complexes to the cytoplasm or the cytoskeleton. In addition to FAK, paxillin also associates with vinculin, C-terminal SRc kinase, and the ß1-subunit of integrin (this latter was shown to be reduced in VSMCs from young SHR). Paxillin functions as an adapter protein involved in forming multiprotein complexes. In addition, Ang II (10-6 mol/L) was shown to elevate protein tyrosine phosphorylation levels of paxillin, and this is accompanied by a reorganization of the cytoskeletal network, including the formation of new focal adhesions and actin stress fibers.54 55 The results presented in this study suggest a difference in the organization pattern of paxillin between SHR and WKY and also between cells grown on different ECM proteins. Clearly, further studies are necessary to assess the mechanism by which ECM components control the Ang IIinduced Ca2+ mechanisms.
Conclusions
VSMCs are known to synthesize their own matrix and to
secrete growth factors such as transforming growth factor, a potent
activator of ECM genes. Although the initial ECM substrate
is undoubtedly modified by the cells themselves during the time course
of these experiments, including deposition of endogenously
synthesized matrix components, this approach has enabled us to show
that the initial matrix protein seen by the VSMCs plays an important
role in the regulation of Ang IIinduced
Ca2+ movements.
Calcium is an important second messenger that mediates a large number of signal transduction processes, regulating arterial tone and synthesis of ECM. The understanding of these synergies should provide insight into issues such as development of hypertrophy of large vessels as well as questions concerning the compliance and distensibility of arteries during development of hypertension.
Received August 15, 2000; first decision September 7, 2000; accepted December 27, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
L. H. Romer, K. G. Birukov, and J. G.N. Garcia Focal Adhesions: Paradigm for a Signaling Nexus Circ. Res., March 17, 2006; 98(5): 606 - 616. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. K. Rohra, T. Yamakuni, K.-I. Furukawa, N. Ishii, T. Shinkawa, T. Isobe, and Y. Ohizumi Stimulated Tyrosine Phosphorylation of Phosphatidylinositol 3-Kinase Causes Acidic pH-Induced Contraction in Spontaneously Hypertensive Rat Aorta J. Pharmacol. Exp. Ther., December 1, 2002; 303(3): 1255 - 1264. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Safar, P. Chamiot-Clerc, G. Dagher, and J. F. Renaud Pulse Pressure, Endothelium Function, and Arterial Stiffness in Spontaneously Hypertensive Rats Hypertension, December 1, 2001; 38(6): 1416 - 1421. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2001 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |