Molecular Medicine |
Reduces Aortic Smooth Muscle Differentiation
From the Department of Medicine (C.R.M., M.G.H., M.A.G., C.B.M.), Ohio State University College of Medicine, Columbus; Center for Biostatistics (G.S.P.), Ohio State University, Columbus; and School of Medicine (P.J.G.-C.), University of Miami, Fla.
Correspondence to Clay B. Marsh, The Ohio State University Heart and Lung Research Institute, Division of Pulmonary and Critical Care Medicine, 473 W 12th Ave, Columbus, OH 43210. E-mail Clay.Marsh{at}osumc.edu
| Abstract |
|---|
|
|
|---|
(ER
) by agonists or by growth factors altered SMC function. To determine the effect of growth factors, estrogen, and ER
expression on SMC differentiation, human aortic SMC were cultured in serum-free conditions for 10 days. SMC from men had lower spontaneous expression of ER
and higher levels of the differentiation markers calponin and smooth muscle
-actin than SMC from women. When SMC containing low expression of ER
were transduced with a lentivirus containing ER
, activation of the receptor by ligands or growth factors reduced differentiation markers. Conversely, inhibiting ER
expression by small interfering RNA (siRNA) in cells expressing high levels of ER
enhanced the expression of differentiation markers. ER
expression and activation reduced the phosphorylation of Smad2, a signaling molecule important in differentiation of SMC and initiated cell death through cleavage of caspase-3. We conclude that ER
activation switched SMC to a dedifferentiated phenotype and may contribute to plaque instability.
Key Words: apoptosis cardiovascular disease gene expression nuclear receptors smooth muscle differentiation
| Introduction |
|---|
|
|
|---|
A few studies comparing vascular wall properties and disease presentation of men and women with symptoms of coronary artery disease provide insight into the complicated effects of female hormones and their receptors in vascular cells. Although women with acute coronary syndromes are often free of angiographically visible stenoses, testing of coronary flow reserve demonstrates endothelial and smooth muscle dysfunction.8,9 Younger women who die from coronary artery thrombosis are more likely than men or postmenopausal women to have plaque erosion, rather than rupture of a lipid-rich plaque.10,11 Plaque erosions are characterized by loss of endothelial cells covering a nonocclusive, smooth muscle cell (SMC)- and hyaluronan-rich plaque with few inflammatory cells or type I collagen.12 It is speculated that migration of dedifferentiated SMC and expression of hyaluronan weakens endothelial cell adhesion and predisposes the coronary arteries for thrombotic events.12
After menopause, women experience a dramatic rise in aortic stiffness, which may cause hypertension.13 In those who develop coronary artery disease, the plaques become more numerous with larger lipid cores and thinner fibrous caps marked by calcification.11 It is uncertain how matrix deposition and plaque stability are affected by the lack of estrogen or by growth factors that activate the estrogen receptor (ER).2
Because SMC are responsible for many of the differences in coronary disease noted between men and women, such as microvessel dysfunction, plaque erosion, and matrix deposition, we sought to understand the role of ER
in smooth muscle differentiation in estrogen- or growth factor-rich environments to mimic gender or menopausal effects. In this study, we obtained aortic SMC from male and female donors and determined the effects of ER
expression, estrogen, and growth factors on differentiation, survival, and adherence of these cells.
| Materials and Methods |
|---|
|
|
|---|
-Actin Detection
-actin expression using FACS Calibur flow cytometer (BD Biosciences, San Jose, Calif). Cell populations containing at least 85% positive staining for SM
-actin were used for subsequent studies.
Real-Time PCR for ER
ER
mRNA was analyzed by real-time PCR in SMC from 5 male donors and 5 female donors that were starved for 5 days to allow ER
upregulation. TaqMan primers and probe designed by Primer 3 software (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi14) were synthesized by Applied Biosystems (Foster City, Calif). The following primers were used to detect ER
: forward, 5'-agctcctcctcatcctctcc-3'; reverse, 5'-tctccagcagcaggtcatag-3'; and probe 5'-6FAM-tcaggcacatgagtaacaaaggca-TAMRA-3'. RNA was isolated using NucleoSpin RNA II (BD Clontech, Mountain View, Calif), and cDNA generated using random hexamers (Invitrogen, Carlsbad, Calif). A 111-bp product from ER
was amplified over 40 cycles with 18S RNA as internal control using the ABI PRISM 7700 Sequence Detection System (Applied Biosystems).
Cloning of ER
into EGFP-pLenti6/V5 Plasmid and Transduction
The pLenti6/V5-D-TOPO vector (Invitrogen) was engineered to contain an enhanced green fluorescent protein (EGFP) surrounded by additional restriction sites and designated pLenti-EGFP (generously provided by Mark Wewers, Ohio State University). cDNA for ER
was amplified by PCR from a pBK-CMV/ER
plasmid kindly provided by Robert Brueggemeier (Ohio State University), introducing EcoRI and EcoRV restriction sites. EGFP was removed from pLenti-EGFP by digestion with EcoRI and EcoRV and replaced with ER
to generate pLenti-ER
. pLenti vector lacking EGFP was used as a control. Purified pLenti-ER
or empty vector control (3 µg) were transfected with 2 µg of pMD.G and 10 µg of pCMV
R8.2 helper plasmids (kindly provided by Dr K. Boris-Lawrie, Ohio State University) into HEK293FT cells according to the directions for the ViraPower Lentiviral Expression System. Virus secreted into the media was concentrated (Vivaspin 100 000 MWCO; Vivascience, Germany) and titered in SMC cultures, with blasticidin (2 µg/mL) for selection. SMC were then transduced with the virus for each experiment at approximately 5 multiplicity of infection and incubated overnight in growth media containing 6 µg/mL polybrene.
Transfection of Small Interfering RNA Plasmids
SMC (1x106) were transfected with 10 µg of control or ER
small interfering RNA (siRNA) plasmid (Panomics, Redwood City, Calif) using nucleofection (Amaxa, Gaithersburg, Md). Transfection efficiency was monitored using 2 µg of pmaxGFP plasmid (Amaxa).
SMC Differentiation and Activation
Differentiation experiments were performed on SMC in the following groups, seeded in an 8-well plate as noted: native cells expressing endogenous ER
(7x104 cells per well), cells with low expression of ER
to be transduced with ER
lentivirus (8x104), and cells with high ER
levels transfected with ER
siRNA (1.8x105). After recovery, the cells were starved overnight in phenol red and serum-free basal media (EBM-PRF) (Clonetics/Cambrex) and exposed for 10 days to vehicle control (either 4 µmol/L HCL or 1:400 000 dilution ethanol), 17ß-estradiol (10 nmol/L; Sigma, St Louis, Mo), the ER
agonist propyl pyrazole triol (PPT) (10 nmol/L; Tocris Cookson, Ellisville, Mo), epidermal growth factor (EGF) (10 ng/mL, R&D Systems, Minneapolis, Minn), platelet-derived growth factor-BB (PDGF-BB) (10 ng/mL; R&D Systems), or transforming growth factor-ß1 (TGFß1) (5 ng/mL; R&D Systems) in EBM-PRF. Agonists or vehicle controls were added each day and then cells were lysed in cell lysis buffer (Cell Signaling Technology, Danvers, Mass). Samples of the culture media at the end of the experiment were quantitated for active TGFß1 by ELISA (Quantikine, R&D Systems). Activation studies were performed on SMC stably transduced with pLenti control or ER
(7x104 cells per well or 2x105 cells per 25-cm2 flask), incubated for the times indicated using agonists as listed above, then lysed with cell lysis buffer or CelLytic NuCLEAR Extraction Kit (Sigma). Equal protein amounts (20 to 50 µg) were subjected to Western blot analysis and detected with SuperSignal West Femto Maximum Sensitivity Substrate (Pierce, Rockford, Ill) and the Fluor S-Max system (Bio-Rad, Hercules, Calif). Smooth muscle (SM)
-actin, ß-actin, and calponin antibodies were obtained from Sigma. Antibodies to phospho-ER
and cleaved caspase-3 were from Cell Signaling Technology. Cyclin D1 (clone DCS-6), Erk2 and ER
(HC20) antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, Calif).
ER
Transcriptional Activation
Stably transduced SMC were transfected with an estrogen response element (ERE) reporter construct producing secreted alkaline phosphatase (SEAP) (Clontech/BD Biosciences) using Effectene (Qiagen, Valencia, Calif). The SEAP signal was obtained over 3 days and normalized as a percentage of the maximum signal achieved.
Immunofluorescence for ER
Virally transduced SMC were fixed in 70% ETOH, permeabilized and blocked with 0.05% triton/1% goat serum. Cells were incubated overnight with ER
antibody (Ab-16, Lab Vision-Neomarkers, Fremont, Calif) in 1% goat serum. ER
was detected with Alexa Fluor 568 anti-rabbit secondary antibody (Molecular Probes, Invitrogen) and a DP-11 digital camera connected to an IX-50 inverted microscope with 10x objective (Olympus, Melville, NY).
Cell Density
Phase contrast images were taken using identical settings at day 10 of the differentiation experiments using the DP-11 digital camera and IX-50 inverted microscope with 4x objective (Olympus). Quantity One colony counting software (Bio-Rad) was used to detect live cells (gray) but exclude apoptotic cells (white). Numbers were normalized to vehicle control samples for each cell population or control transfection/transduction cells.
Statistics
Real-time PCR results for ER
expression were analyzed using longitudinal regression over 10 experiments to test the difference in
cycle times, which are normally distributed. Western blot densitometry ratios for contractile proteins in starved or PDGF-stimulated SMC from 6 people were compared using a mixed model regression to account for correlation within cell lines. Densitometry values from the remaining immunoblots were normalized to loading controls and by the vehicle control sample for the control group, compared by 2-factor ANOVA (Stata version 9; StataCorp, College Station, Tex), and pairwise comparisons were adjusted using the Holms method.15
| Results |
|---|
|
|
|---|
in Human Aortic Smooth Muscle
enhances proliferation in transformed cells, we hypothesized that the expression and activation of SMC ER
modulated cell differentiation. The ER
mRNA level, stated as a fold induction above the SMC population containing the lowest level of ER
, was &4.3 times higher on average for female donors than for male donors (P<0.001, Figure 1A). By comparison, serving as a positive control, the ER
level for the breast cancer line MCF7 was &1000 times higher than SMC containing the lowest ER
levels, whereas as a negative control, the colon cancer cell line HT29 had little to no ER
detected by PCR. For subsequent studies, we used SMC from either the female donor with the greatest amount of ER
or the male donor with the lowest ER
expression. We confirmed proportional ER
protein expression in these 2 cell populations (Figure 1B).
|
SMC Containing ER
Failed to Differentiate
Because growth factors activate ER
and TGFß1 causes differentiation of SMC,16,17 SMC expressing the highest ER
levels and the lowest ER
levels were treated with these growth factors as well as ER
ligands (PPT and 17ß-estradiol). As shown, cells with high ER
levels had little expression of the differentiation markers SM
-actin or calponin except in the presence of TGFß1 (Figure 2A). In contrast, cells with lower levels of ER
retained both SM
-actin and calponin in all conditions except when incubated with EGF or PDGF (Figure 2A). We observed that low ER
cells had significantly greater amounts of calponin (P<0.0001 overall) and greater amounts of SM
-actin (P=0.0001 overall) compared with high ER
-expressing cells (Figure 2B). Individual comparisons are as shown in Figure 2B.
|
Because SMC from the low ER
(male) donor had more differentiation markers than the high ER
(female) donor, we further characterized basal differentiation of SMC from other male or female donors (n=3 each). As shown in Figure 2C, vehicle control-stimulated SMC from male donors had high levels of calponin but lost much of this marker on PDGF stimulation, similar to cells in Figure 2A. In contrast, female donor SMC expressed only low levels of calponin in either condition. Consequently, the average calponin ratio was significantly higher for SMC from men than for SMC from women (P=0.0281).
Transduction of ER
Inhibited SMC Differentiation
Because ER
expression correlated with SMC dedifferentiation, we examined whether induced expression of ER
in low ER
-containing cells directly inhibited SMC differentiation. Transduction efficiency was determined by ER
immunofluorescent staining (Figure 3A).Transduction of ER
lowered calponin (P<0.0001) and SM
-actin (P<0.0001) expression overall. ER
-expression reduced SM
-actin in response to 17ß-estradiol, PPT, and TGFß1 treatment and reduced calponin in response to vehicle, 17ß-estradiol, or PPT (Figure 3B and 3C, probability values as shown). In contrast to these cell markers, cyclin D increased on transduction of ER
(P=0.0001 overall).
|
Interruption of ER
by siRNA Augmented TGFß1-Induced Differentiation
Because high native levels of ER
correlated with low levels of SMC differentiation markers, we next reduced ER
expression through siRNA to enhance their differentiation program. High transfection efficiency was obtained using pmaxGFP plasmid DNA in cotransfections (Figure 4A). A reduction in ER
protein was confirmed in the siRNA-transfected cells compared with the empty vector control (Figure 4A).
|
We found that reduced ER
expression led to higher levels of differentiation markers. ER
siRNA upregulated calponin expression overall (P=0.0038), with significant pairwise difference occurring in TGFß1-treated cells (Figure 4B and C). Although ER
siRNA slightly raised SM
-actin levels in cells incubated with TGFß1, the increase was not significant (P=0.1542).
Ligand Activation of ER
Inhibited Smad2 Phosphorylation
We next investigated ligand-dependent ER
activation by 17ß-estradiol and PPT, and the ligand-independent activation by EGF and PDGF. A low level of ER
phosphorylation was observed in the ER
-transduced cells in the vehicle-treated condition, whereas 17ß-estradiol and PPT preferentially phosphorylated ER
S118, and EGF and PDGF phosphorylated ER
S167 (Figure 5A). TGFß1 caused no activation above vehicle control of either serine residue.
|
In contrast to the activation by phosphorylation seen with EGF and PDGF, only 17ß-estradiol and PPT activated transcription of an ERE reporter construct (Figure 5B, P=0.0013 for 17ß-estradiol and P=0.0008 for PPT compared with vehicle). Activation for up to ten days with EGF, PDGF-BB or TGFß1 caused no detectable signal above vehicle control samples (data not shown). As a partial explanation for the transcriptional inactivity of ER
phosphorylated by growth factors, we found that stimulating the cells with 17ß-estradiol, but not EGF, PDGF-BB or TGFß1, for 20 to 60 minutes caused nuclear translocation of ER
(Figure 5B and data not shown).
Because our differentiation analysis suggested that TGFß1 elevated SMC differentiation markers in the presence of ER
but full expression of these markers required lower levels of ER
, we examined whether ER
inhibited TGFß signaling by interfering with Smad activation, as previously described.18 Because 17ß-estradiol or PPT potently reduced SMC differentiation, we determined whether 17ß-estradiol inhibited TGFß1 signaling through Smads. TGFß1 induced the phosphorylation of Smad2 in pLenti or ER
-transduced SMC for up to 60 minutes (Figure 5C). However, Smad2 phosphorylation was reduced if the ER
-transduced cells were preincubated with 17ß-estradiol 30 minutes before activation. The relevance of TGFß1 to the differentiation of SMC was examined by measuring whether the cells spontaneously produced TGFß1 and whether this production correlated to cellular differentiation. Active TGFß1 was detected in the supernatant of pLenti-transduced SMC at 49.0±27.9 pg/mL in the vehicle control condition and 67.2±45.4 and 110.7±42.1 after 17ß-estradiol or PPT incubation, respectively (no significant differences, n=2, mean±SEM). Transduction of ER
in the cells did not alter TGFß1 production suggesting ER
expression altered the response to TGFß1 (45.6±23.1, 49.8±33.6 and 112.0±39.7 for vehicle, 17ß-estradiol, and PPT exposed cells, respectively, n=2).
Ligand Activation of ER
Initiated Apoptosis
Because estrogen inhibits the growth of SMC and causes apoptosis,2,19 we examined the initiation of apoptosis in the presence of ER
agonists. Indeed, SMC transduced to express ER
had evidence of caspase-3 activity when stimulated with 17ß-estradiol or PPT (Figure 6A, P=0.0231 and P=0.0646, respectively; n=2).
|
The effects of ER
expression on cell growth were apparent by cell detachment when ER
-transduced SMC were treated with 17ß-estradiol (Figure 6B, lower right) or PPT (not shown), indicating that cells were undergoing apoptosis. Consequently, fewer ER
-transduced cells were counted after 17ß-estradiol or PPT treatment compared with pLenti-transduced cells (P=0.0006 overall; Figure 6D, i). Consistent with this observation, cells natively expressing high levels of ER
had significantly lower cell densities than the low-ER
cells when treated with 17ß-estradiol or PPT (P=0.0003 overall; Figure 6D, ii). Finally, a small increase in cell density was found overall (P=0.0274) when the high-ER
SMC were transfected with ER
siRNA, although no individual paired comparisons were significant (Figure 6C and 6D, iii).
| Discussion |
|---|
|
|
|---|
in vascular SMC beyond its ability to inhibit growth. To understand differences in SMC status between men and women, we characterized aortic SMC differentiation and ER
expression in these 2 groups. We detected significantly higher levels of ER
in SMC from our female donors compared with SMC from male donors. The inverse was true for differentiation markers, however, as cells from men expressed greater levels of SM
-actin and calponin protein under starved conditions, providing a connection between ER
expression and differentiation.
We analyzed the effect of 2 ER
ligands and 3 growth factors on cell populations containing the lowest and highest levels of ER
. SMC differentiation markers remained high for cells natively expressing low amounts of ER
whether incubated with vehicle, ER
agonists, or TGFß1. However, EGF and PDGF decreased SM
-actin and calponin levels in these cells, similar to published accounts.20 In contrast, cells expressing high native ER
had a low level of SM
-actin and calponin under most conditions except when treated with TGFß1. Similar findings were observed in cells virally transduced with ER
, which resulted in their dedifferentiation. Only TGFß1 could partially overcome the inhibitory effect of ER
. These data indicated that ER
may play a role in causing the low contractile protein levels detected in SMC from women. The ability of ER
to inhibit differentiation was unexpected, because ER
is known to inhibit growth and would be expected to induce differentiation. In contrast, ER
caused an increase in cyclin D1 expression, indicating that growth inhibition did not align with quiescence. To confirm this biological role for ER
, we found that reduction in ER
resulted in greater contractile protein expression, especially in the presence of TGFß1.
Several possible pathways could be involved in the reduced SMC differentiation caused by ER
. Inhibition of cell cycle regulators and activation of proliferation genes such as cyclin D are known to occur in ER
-positive breast cancer cells exposed to 17ß-estradiol.21 Similar changes in SMC could induce a phenotypic switch from differentiated to proliferating or migratory SMC. Alternatively, ER
may inhibit transcription by shunting coactivator proteins such as p300/CBP away from other transcription factors, some of which are necessary for smooth muscle gene expression.2224 ER
activates transcription at estrogen response elements on DNA, but is known to suppress the TGFß1/Smad pathway by binding to and repressing Smad2 and -3, positive regulators of contractile protein transcription in SMC.18,25 In agreement, Smad2 phosphorylation was inhibited by estrogen in ER
-transduced SMC in the current study. Because the SMC released detectable levels of active TGFß1, the ability of ER
to inhibit Smad-regulated differentiation is a likely mechanism of action.
Cytoplasmic signaling pathways activated by ER
including phosphatidylinositol 3-kinase and Akt, growth factor receptor autophosphorylation, mitogen-activated protein kinases (MAPKs), and src kinases can contribute to SMC dedifferentiation.19,2628 A positive-feedback loop also exists in which S118 of ER
is phosphorylated by 17ß-estradiol and MAPK, whereas S167 of ER
is phosphorylated through the Akt pathway.16,29,30 Depending on the stimulus, we saw preferential phosphorylation of ER
epitopes in SMC, indicating that upstream and downstream signaling events likely differed in these cells. Only ER
ligands caused nuclear translocation and transcriptional activity at an ERE.
Many studies show that 17ß-estradiol induces apoptosis through ER
in SMC.2,19 We found that 17ß-estradiol and PPT significantly reduced cell density of native and transduced cells expressing high levels of ER
, whereas inhibition of ER
by siRNA increased cell density.
Our results may explain some differences in coronary events in women and men. ER
activation in an affected coronary artery may cause the dedifferentiation and migration of SMC into the intima, causing microvessel dysfunction. Our observation of apoptosis of SMC after estrogen exposure could partly explain why postmenopausal hormone replacement therapy causes higher rates of myocardial infarction through thinning of collagen and rupture of plaques.
| Acknowledgments |
|---|
Sources of Funding
This work was supported by NIH grants HL63800-05, HL67176-04, HL70294-03 and HL066108-04 (to C.B.M.); and NIH Individual National Research Service Award 5F32 HL09550 and American Heart Association Ohio Affiliate Postdoctoral Fellowship Award 9920597V (to C.R.M.).
Disclosures
None.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
Related Article:
This article has been cited by other articles:
![]() |
A. Benjo, R. E. Thompson, D. Fine, C. W. Hogue, D. Alejo, A. Kaw, G. Gerstenblith, A. Shah, D. E. Berkowitz, and D. Nyhan Pulse Pressure Is an Age-Independent Predictor of Stroke Development After Cardiac Surgery Hypertension, October 1, 2007; 50(4): 630 - 635. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. M Gruen and C. J Pepine Raising awareness of angina in women Heart, March 1, 2007; 93(3): 279 - 280. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. J. Pepine, W. W. Nichols, and D. F. Pauly Estrogen and Different Aspects of Vascular Disease in Women and Men Circ. Res., September 1, 2006; 99(5): 459 - 461. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Research Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2006 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |