Sie sind auf Seite 1von 6

Am J Physiol Heart Circ Physiol 292: H2607H2612, 2007.

First published January 19, 2007; doi:10.1152/ajpheart.01107.2006.

Estrogen modulates TNF--induced inflammatory responses in rat aortic


smooth muscle cells through estrogen receptor- activation
Dongqi Xing,* Wenguang Feng,* Andrew P. Miller,1 Nathaniel M. Weathington,2
Yiu-Fai Chen,1 Lea Novak,3 J. Edwin Blalock,2 and Suzanne Oparil1,2
1
Vascular Biology and Hypertension Program, Division of Cardiovascular Disease,
Department of Medicine, 2Department of Physiology and Biophysics, and 3Department
of Anatomic Pathology, University of Alabama at Birmingham, Birmingham, Alabama
Submitted 10 October 2006; accepted in final form 19 January 2007

Xing D, Feng W, Miller AP, Weathington NM, Chen YF, restenotic vessels (18, 22, 28), and is associated with activation

Downloaded from http://ajpheart.physiology.org/ by 10.220.33.2 on September 12, 2017


Novak L, Blalock JE, Oparil S. Estrogen modulates TNF-- of a variety of cell types, including adipocytes and fibroblasts,
induced inflammatory responses in rat aortic smooth muscle cells in adventitial tissues (20, 28). Since vascular smooth muscle
through estrogen receptor- activation. Am J Physiol Heart Circ cells (VSMCs) have been shown to produce adhesion mole-
Physiol 292: H2607H2612, 2007. First published January 19,
cules (3) and chemokines (34) when stimulated with cytokines
2007; doi:10.1152/ajpheart.01107.2006.We have previously
shown that 17-estradiol (E2) attenuates responses to endoluminal and because our earlier studies (21) demonstrated that activa-
injury of the rat carotid artery, at least in part, by decreasing inflam- tion and migration of adventitial fibroblasts can be stimulated
matory mediator expression and neutrophil infiltration into the injured by media conditioned by VSMCs, we hypothesized that
vessel, with a major effect on the neutrophil-specific chemokine VSMCs are activated early following endoluminal injury, re-
cytokine-induced neutrophil chemoattractant (CINC)-2. Current leasing inflammatory mediators that reach the periadventitial
studies tested the hypothesis that activated rat aortic smooth muscle space to recruit leukocytes and are the chief effector cells for
cells (RASMCs) express these same inflammatory mediators and initiation of the early inflammatory response.
induce neutrophil migration in vitro and that E2 inhibits these pro- Previous studies in our laboratory have characterized an
cesses by an estrogen receptor (ER)-dependent mechanism. Quiescent early estrogen receptor (ER)-dependent anti-inflammatory ef-
RASMCs treated with E2, the ER-selective agonist propyl pyrazole fect for 17-estradiol (E2) in this model of vascular injury (2,
triol (PPT), the ER-selective agonist diarylpropiolnitrile (DPN), or
vehicle for 24 h were stimulated with tumor necrosis factor (TNF)-
25, 40). We have shown that systemic E2 administration
and processed for real-time RT-PCR, ELISA, or chemotaxis assays attenuates both expression of inflammatory mediators and
6 h later. TNF- stimulated and E2 attenuated mRNA expression of infiltration of leukocytes into balloon-injured carotid arteries of
inflammatory mediators, including P-selectin, intercellular adhesion ovariectomized (OVX) rats at a very early time point postin-
molecule (ICAM)-1, vascular cell adhesion molecule (VCAM)-1, jury (25, 40). In this model, we demonstrated a particularly
monocyte chemoattractant protein (MCP)-1, and CINC-2. DPN dose robust effect for E2 in modulating neutrophil chemotaxis via
dependently attenuated TNF--induced mRNA expression of CINC- attenuating expression of cytokine-induced neutrophil che-
2, whereas PPT had no effect. The anti-inflammatory effects of DPN moattractant (CINC)-2, a member of the cysteine-x-cysteine
and E2 were blocked by the nonselective ER-inhibitor ICI-182,780. (CXC) chemokine family and a potent chemoattractant for
ELISA confirmed the TNF--induced increase and E2-induced inhi- neutrophils in vitro and in vivo (23, 24). Based on our hypoth-
bition of CINC-2 protein secretion. TNF- treatment of RASMCs esis that VSMCs are the target cells that initiate the injury
produced a twofold increase in neutrophil chemotactic activity of
conditioned media; E2 and DPN treatment markedly inhibited this
response, the present in vitro study was designed to define
effect. E2 inhibits activated RASMC proinflammatory mediator ex- cellular and molecular mechanisms mediating estrogenic anti-
pression and neutrophil chemotactic activity through an ER-depen- inflammatory effects. We tested the estrogen receptor subtype
dent mechanism. (ER or ER) dependence of the anti-inflammatory effect of
E2 by employing two recently developed selective ER and
inflammation; vascular smooth muscle cell; arteries ER agonists. Based on published observations that 1) tumor
necrosis factor (TNF)- acts as the bodys fire alarm (11)
INFLAMMATION plays an important role in the pathogenesis of and induces the rapid recruitment of leukocytes from the circula-
many forms of vascular disease, including atherosclerosis and tion in response to many forms of stress, 2) it is elaborated from
the response to acute vascular injury (14, 37). Balloon injury of diseased bypass grafts and atherosclerotic arteries (8), and 3) our
arteries has been shown to elicit accumulation of inflammatory own finding that TNF- expression is dramatically upregulated in
cells, specifically granulocytes (neutrophils) and monocyte/ balloon-injured rat carotid arteries and not affected by E2 admin-
macrophages, in the adventitia surrounding the injury site istration (25), we used TNF- as the inflammatory stimulus in the
within hours after the insult (17, 28, 40). The appearance of current study. Since our previous in vivo studies demonstrated
inflammatory cells is predated by expression of inflammatory robust estrogenic attenuation of injury-induced CINC-2 expres-
mediators, including adhesion molecules and chemokines, in sion and of neutrophil infiltration into injured arteries (25, 40), we
acutely injured arteries (25) as well as in atherosclerotic and focused on neutrophil activation and chemotaxis in the current in
vitro study.

* D. Xing and W. Feng contributed equally to this study.


Address for reprint requests and other correspondence: D. Xing, 1014 The costs of publication of this article were defrayed in part by the payment
Zeigler Research Bldg., Univ. of Alabama at Birmingham, UAB Station, of page charges. The article must therefore be hereby marked advertisement
Birmingham, AL 35294 (E-mail: dqxing@uab.edu). in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

http://www.ajpheart.org 0363-6135/07 $8.00 Copyright 2007 the American Physiological Society H2607
H2608 ESTROGEN AND INFLAMMATION

MATERIALS AND METHODS with 150 l of conditioned medium of vehicle control or TNF-
E2-, TNF- DPN-, or TNF- PPT-treated cells. To test the
Cell culture. Primary cultures of rat aortic smooth muscle cells CINC-2 dependence of neutrophil chemoattractant activity of condi-
(RASMCs) were derived from 10-wk-old female Sprague-Dawley tioned media from TNF--treated cells, anti-CINC-2/ antibody
rats (Charles River), as previously described (31). Cells were cultured (R&D System, Minneapolis, MN) at a final concentration of 5 g/ml
in complete medium containing phenol red-free DMEM (GIBCO) or PBS (control) was incubated with conditioned media for 30 min at
supplemented with 10% (vol/vol) FBS, 4 mmol/l L-glutamine, 100 room temperature; 150 l of the conditioned media-antibody solution
U/ml penicillin, and 100 g/ml streptomycin. Cells were identified as were added to the bottom wells of the chamber as described. A
smooth muscle cells (SMCs) by their characteristic morphology and polyvinylpyrrolidone-free polycarbonate filter plate with 3-m pores
positive immunostaining for -smooth muscle actin (-SMA, clone was placed over the samples, and 100 l of the dHL-60 cell suspen-
1A4, DAKO). All experiments were performed using early passage sion (2 106 cells/ml) were placed into the upper wells. The
(5) near-confluent (95%) cultures that were serum starved for 24 h chambers were incubated in humidified air with 5% CO2 at 37C for
before treatment. 90 min. The upper portion was then removed, and four photomicro-
Initial studies tested whether TNF- could stimulate expression of graphs (200) per well were digitally recorded using an Olympus
relevant inflammatory chemokines and cytokines in RASMCs and IX70 microscope and Perkin-Elmer Ultraview image capture equip-

Downloaded from http://ajpheart.physiology.org/ by 10.220.33.2 on September 12, 2017


whether pretreatment with E2 could inhibit these inflammatory re- ment. Cell counts were made from these images. For ease of com-
sponses. In these studies, cells were pretreated with E2 (107 M) or parison of results between experiments, data were standardized to a
vehicle (ethanol at a final concentration 0.01%) for 24 h and then chemotactic index with cell migration to conditioned media of vehi-
incubated with TNF- (0.12.5 ng/ml) for an additional 6 h. cle-treated RASMCs as a baseline (25).
Subsequent studies assessed the ER subtype dependence of the E2 Statistical analysis. Data are expressed as means SE. Statistical
effect. Cells were pretreated with the selective ER agonist diaryl- analysis was performed with one-way ANOVA or Students t-test, as
propiolnitrile (DPN) (1010-107 M) (Tocris Cookson, Ellisville, appropriate. Values of P 0.05 were considered significant.
MO) or the selective ER agonist propyl pyrazole triol (PPT) (1010-
106 M) (OBITER Research, Urbana, IL) for 24 h, and then incubated RESULTS
with 1 ng/ml TNF- for an additional 6 h. Expression of the neutro-
phil-specific chemokine CINC-2 was assessed in these studies. To Real-time quantitative RT-PCR analysis showed that TNF-
confirm ER dependence of the DPN and E2 effects, subgroups of cells (0.12.5 ng/ml) dose dependently stimulated expression of
from the above experiments were exposed to the nonselective ER chemokines (CINC-2 and MCP-1) and adhesion molecules
antagonist ICI-182,780 (106 M) for 2 h before the E2 (107 M) or [intercellular adhesion molecule (ICAM)-1, vascular cell ad-
DPN (107 M) pretreatment. For CINC-2 ELISA and chemotaxis hesion molecule (VCAM)-1, and P-selectin] (results for
assays, cells were pretreated with E2 (107 M) or vehicle for 24 h, CINC-2 shown in Fig. 1). A TNF- concentration of 1 ng/ml
followed by TNF- (1 ng/ml) treatment for an additional 6 h. was chosen for further experiments because it was the mini-
Conditioned media were collected to assess CINC-2 protein concen- mum effective dose for all mediators. All mediators were
tration and ability to stimulate neutrophil migration. expressed at low or undetectable levels in unstimulated vehi-
Real-time quantitative RT-PCR analysis of inflammatory media-
tors. As described previously (25), total RNA was extracted from cells
cle-treated RASMCs (Fig. 2). Pretreatment with E2 (107 M)
using TRIzol (Invitrogen, Carlsbad, CA), treated with DNAase I to significantly inhibited expression of CINC-2, MCP-1,
remove genomic DNA, and then purified by using an RNA purifica- ICAM-1, VCAM-1, and P-selectin in cells treated with TNF-
tion kit (Invitrogen). The protein- and DNA-free RNA was reverse (1 ng/ml, Fig. 2). Inhibition of these chemokines and adhesion
transcribed to cDNA using the SYBR Green RT-PCR kit (Applied molecules expression was abolished by the nonselective ER
Biosystems, Foster City, CA) and specific primers (described in Ref.
25). cDNA was amplified by PCR in the iCycler for 40 cycles, and
relative RNA levels were calculated using the iCycler software and a
standard equation (Applied Biosystems). Unknowns were normalized
using 18S rRNA and then standardized to the mRNA level of
vehicle-treated RASMCs, except for CINC-2, monocyte chemoat-
tractant protein (MCP-1) and P-selectin, which were standardized to
the mRNA level of TNF--treated RASMCs because their mRNA
was undetectable in the vehicle group.
Measurement of CINC-2 protein concentration. The CINC-2
concentration in conditioned media of vehicle control and TNF-
E2-treated cells was assayed with the rat GRO/CINC-2 ELISA kit
(IBL, Japan). Briefly, conditioned media (4 ml) were collected and
concentrated 10-fold using Centricon concentrators (Millipore, Bil-
lerica, MA). Samples were processed according to the manufacturers
instructions. Samples were measured in duplicate, and the CINC-2
concentration was calculated from the standard curve.
Chemotaxis assays. Human myeloid leukemia HL-60 cells (ATCC,
Manassas, VA) were maintained in Iscoves modified medium
(ATCC) supplemented with 10% fetal calf serum, 50 g/ml strepto- Fig. 1. Tumor necrosis factor (TNF)- dose dependently increases mRNA
mycin, 2 U/ml penicillin, and 2 mM L-glutamine. For differentiation, expression of cytokine-induced neutrophil chemoattractant (CINC)-2 in rat
cells (3 105/ml) were incubated in the presence of 1.3% (vol/vol) aortic smooth muscle cells (RASMCs). Cells were grown to subconfluence
(95%) in six-well plates, deprived of serum for 24 h, and then treated with
DMSO for 46 days (Newburger PE) before assessment of their TNF- (0.52.5 ng/ml) for an additional 6 h. Data, expressed as means SE,
chemotactic responsiveness. are from real-time quantitative RT-PCR assays and are normalized by 18S
Neutrophil chemotactic activity was assayed in a 96-well modified RNA and then standardized to the mean mRNA level of the RASMCs treated
Boyden chamber (Millipore, Billerica, MA) using differentiated with TNF- at the 0.5 ng/ml dose. UD, undetectable; *P 0.05 vs. vehicle-
HL-60 (dHL-60) cells. The bottom wells of the chamber were filled treated RASMCs.

AJP-Heart Circ Physiol VOL 292 JUNE 2007 www.ajpheart.org


ESTROGEN AND INFLAMMATION H2609

Downloaded from http://ajpheart.physiology.org/ by 10.220.33.2 on September 12, 2017


Fig. 2. 17-Estradiol (E2, 107 M) inhibits TNF--induced mRNA expression of chemokines (left) and adhesion molecules (right) in RASMCs, and this effect
is abolished by ICI-182,780. Cells were grown to subconfluence (95%) in six-well plates, deprived of serum for 24 h, pretreated with 107 M E2 or vehicle
for 24 h, and then treated with TNF- (1 ng/ml) for an additional 6 h. ICI-182,780 (106 M) was given to cells at 2 h before E2 treatment in some experiments.
Data, expressed as means SE, are from real-time quantitative RT-PCR assays and are normalized by 18S RNA. Data for P-selectin, CINC-2, and monocyte
chemoattractant protein (MCP)-1 are standardized to the mean mRNA level of the TNF--treated RASMCs. Data for intercellular adhesion molecule (ICAM)-1
and vascular cell adhesion molecule (VCAM)-1 are standardized to the mean mRNA level of the vehicle-treated RASMCs. *P 0.05 vs. respective
vehicle-treated RASMCs; #P 0.05 vs. respective TNF--treated RASMCs.

antagonist ICI-182,780 (106 M), indicating ER dependence protein concentration in conditioned media from vehicle-
(Fig. 2). treated RASMCs. TNF- increased CINC-2 protein levels
The selective ER- agonist DPN dose dependently inhibited 2.7-fold compared with vehicle-treated control cells, and E2
TNF- (1 ng/ml)-induced expression of CINC-2 over the pretreatment completely abolished the TNF--induced in-
concentration range 1010 to 107 M (Fig. 3). This effect was crease in CINC-2 expression.
blocked by treatment with ICI-182,780 (106 M), confirming To assess the functional significance of the TNF- and E2-
its ER dependence (Fig. 3). The selective ER- agonist PPT, or DPN-induced alterations in CINC-2 expression, the ability
over the dose range of 1010-106 M, had no effect on of conditioned media from cells subjected to these treatments
CINC-2 expression in TNF- (1 ng/ml)-treated cells (Fig. 4), to stimulate migration of dHL-60 cells was assayed. Chemo-
suggesting that the anti-inflammatory effects of E2 in tactic activity was normalized to conditioned media from
RASMCs are selectively mediated by ER and not ER. vehicle-treated RASMCs as described in MATERIALS AND METH-
CINC-2 protein concentration was quantified in condi- ODS. There was no significant difference in the chemotactic
tioned media from RASMCs using a double sandwich ELISA activity of conditioned media derived from vehicle-treated
technique (Fig. 5). E2 had no significant effect on CINC-2 versus E2-, DPN-, and PPT-treated cells. TNF- treatment
induced a marked (2.2-fold) increase in dHL-60 cell migration,
and E2 pretreatment of RASMCs significantly attenuated the
TNF--induced effect on dHL-60 cell migration (by 48%, P
0.01) (Fig. 6). DPN, but not PPT, mimicked the inhibitory
effect of E2 on TNF- treatment-induced dHL-60 cell migra-

Fig. 3. The ER-selective agonist diarylpropiolnitrile (DPN) inhibits TNF--


induced CINC-2 mRNA expression in RASMCs, and this effect is abolished
by ICI-182,780. Cells were grown to subconfluence (95%) in six-well plates, Fig. 4. ER-selective agonist propyl pyrazole triol (PPT) has no effect on
deprived of serum for 24 h, pretreated with DPN (1010-107 M) or vehicle for TNF--induced CINC-2 mRNA expression in RASMCs. Cells were grown
24 h, and then treated with TNF- (1 ng/ml) for an additional 6 h. ICI-182,780 to subconfluence (95%) in six-well plates, deprived of serum for 24 h,
(106 M) was given to cells at 2 h before DPN treatment in some experiments. pretreated with PPT (1010-106 M) or vehicle for 24 h, and then treated with
Data are from real-time RT-PCR assays and normalized by 18S RNA and then TNF- (1 ng/ml) for an additional 6 h. Data are from real-time RT-PCR assays
standardized to the mean mRNA level of the TNF--treated RASMCs. Results and normalized by 18S RNA and then standardized to the mean mRNA level
are means SE. *P 0.05 vs. vehicle-treated RASMCs; #P 0.05 vs. of the TNF--treated RASMCs. Results are means SE. *P 0.05 vs.
TNF--treated RASMCs. vehicle-treated RASMCs.

AJP-Heart Circ Physiol VOL 292 JUNE 2007 www.ajpheart.org


H2610 ESTROGEN AND INFLAMMATION

effects on other neotrophil functions, including adhesion mol-


ecule expression, intracellular calcium influx, and phagocytosis
(27, 32). Upregulation of CINC-2 has been correlated with
neutrophil infiltration in many disease models, including acute
right ventricular failure following pulmonary embolism (35)
and ischemia-reperfusion injury of the liver (41). The present
study reveals marked estrogenic modulation of CINC mRNA
and protein expression in TNF--stimulated RASMCs, consis-
tent with our previous in vivo reports of attenuated CINC-2
expression and neutrophil infiltration in injured arteries with
E2 treatment. To our knowledge, this report is the first in vitro
study of CINC or CXCR2 ligand regulation by E2 in
Fig. 5. E2 inhibits TNF--induced CINC-2 protein expression in RASMCs. RASMCs.
Cells were grown to subconfluence (95%) in six-well plates, deprived of The finding that antibodies to CINC-2 did not completely

Downloaded from http://ajpheart.physiology.org/ by 10.220.33.2 on September 12, 2017


serum for 24 h, pretreated with 107 M E2 or vehicle for 24 h, and then treated ablate the chemotactic avtivity of supernates from TNF--
with TNF- (1 ng/ml) for an additional 6 h, and conditioned media were treated RASMCs was not unexpected because there are un-
collected. Data, expressed as means SE, are from a double sandwich ELISA
assay. *P 0.05 vs. vehicle-treated RASMCs; #P 0.05 vs. TNF--treated
doubtedly other neutrophil chemoattractants in such super-
RASMCs. nates. These might induce leukotriene B4 (9) and the recently
described tripeptide, PGP (38). It is, however, particularly
interesting that about two-thirds of the chemotactic activity of
tion (by 69%, P 0.01). To test the CINC-2 dependence of supernates from TNF--stimulated RASMCs is apparently due
neutrophil chemoattractant activity of conditioned media from to CINC-2 and that a similar amount of chemotactic activity is
TNF--treated cells, conditioned media from separate groups inhibited by E2 (48%) or DPN (69%). This suggests that the
of cells were preincubated with selective anti-CINC-2 anti- effects of estrogen on neutrophil chemotaxis are likely due to
body. Antibody treatment neutralized chemotactic activity of the ability of estrogen to inhibit CINC-2 production.
conditioned media by 63%, verifying its CINC-2 dependence. In the current study, we employed the selective ER agonist
DISCUSSION DPN and the selective ER agonist PPT to define the relevant
ER subtypes involved in the anti-inflammatory effects of E2 on
This in vitro study demonstrates that isolated RASMCs RASMCs. Because of its selective binding affinity, DPN is 30-
express high levels of proinflammatory mediators, including to 70-fold more potent as an agonist of ER than ER (4).
the neutrophil- and monocyte-selective chemokines CINC-2 Supporting its anti-inflammatory effects, DPN has been shown
and MCP-1, when activated by TNF- and that estrogen to suppress IL-6 promoter activity in human endothelial cancer
inhibits this process via and ER-dependent mechanism. Me- (HEC-1) cells in vitro to the same extent as E2 (15). In a male
dia conditioned by activated SMCs directed the migration of rat model of trauma-hemorrhage, DPN, and not PPT, protected
neutrophils, and this effect was partially inhibited by pretreat- against lung injury, with associated reductions of inducible
ment with anti-CINC-2 antibody, indicating that the CINC-2 nitric oxide synthase and IL-6 expression (42), and restored
secreted into the conditioned media was functional. Treatment cardiac function, with associated upregulation of heat shock
with an ER-selective agonist dose dependently inhibited
CINC-2 expression and reduced the chemotactic activity of
media conditioned by TNF--treated SMCs. Together, these
data support the concept that arterial SMCs are important
sources of proinflammatory mediators in the setting of acute
vascular injury, triggering a robust inflammatory response that
is amplified by recruitment of neutrophils and monocytes into
the adventitial domain of the injured vessel and by activation
and migration of adventitial fibroblasts into the neointima
(25, 40).
The CXC chemokines are powerful mediators of neutrophil
recruitment. A representative member of the CXC chemokine
family is IL-8, which is a major chemoattractant for neutro-
phils in humans (29, 39). In rats, no homologue of IL-8 has
been identified, and the CXC chemokines that recruit neutro-
phils are termed CINC (1, 26). Four CXC chemokines, includ-
Fig. 6. Neutrophil chemotactic activity of conditioned media from TNF--
ing CINC (or CINC-1), CINC-2, CINC-2, and MIP-2 (or stimulated RASMCs is blocked by E2 and DPN. Cells were grown to
CINC-3), have been identified in rats (10, 27, 32). CINC-2 has subconfluence (95%) in 75-mm2 flasks, deprived of serum for 24 h, and
two splicing isoforms, CINC-2 and CINC-2 (33), which pretreated with 107 M E2, 107 M DPN, 107 M PPT, or vehicle, respec-
differ in the three amino acids at the COOH terminus (27). tively, for 24 h followed by TNF- (1 ng/ml) treatment for an additional 6 h.
Both CINC-2 and CINC-2 are chemotactic for neutrophils, Conditioned media were collected. Selective anti-CINC-2 antibody (5 g/ml)
partially neutralized neutrophil chemotactic activity of conditioned media from
with an effect at a 10 nM concentration (80 ng/ml) (27). CINCs TNF--treated cells. Results are means SE normalized by vehicle-treated
are structurally related to one another and share many func- RASMCs. *P 0.05 vs. vehicle-treated RASMCs; #P 0.05 vs. TNF--
tions. They have an ability to attract neotrophils and have treated RASMCs.

AJP-Heart Circ Physiol VOL 292 JUNE 2007 www.ajpheart.org


ESTROGEN AND INFLAMMATION H2611
proteins (43). A further study in this model demonstrated that no effect. In contrast, downregulation of ER expression in
DPN, in subcutaneous doses of 5 g/kg administered during porcine aortic endothelial cells inhibited E2-induced p38 and
resuscitation, inhibited myeloperoxidase activity and ELISA- p42/44 mitogen-activated protein kinase activation, whereas
measured levels of CINC-1, CINC-3, and ICAM-1 in the lung, downregulation of ER had no effect. These observations are
suggesting an important role in regulation of neutrophil infil- relevant to our in vivo model of balloon injury of the rat carotid
tration after acute injury (44). artery, because the injured area is denuded of endothelium for
A related compound ERB-041 has 200-fold increased bind- several weeks and the early injury response is driven by
ing activity for ER over ER and is orally active (16). activated SMCs and infiltrating leukocytes. Thus modulatory
ERB-041 in oral doses as low as 1 mgkg1 day1 has been effects of ER activation on SMC-initiated inflammatory re-
shown to reverse chronic diarrhea and ameliorate colonic sponses likely play an important role in inhibiting early inflam-
lesions in a transgenic rat model of inflammatory bowel disease matory changes in the setting of endoluminal vascular injury.
(HLA-B27) and shown to reduce joint inflammation in the Limitations of the current study include use of ER agonists
Lewis rat adjuvant-arthritis model (16). In the latter model, to delineate ER subtype-specific effects in stimulated VSMCs.
mRNA expression profiling in the spleen, lymph nodes, and These agents, while the best available, are not pure ER or

Downloaded from http://ajpheart.physiology.org/ by 10.220.33.2 on September 12, 2017


liver, and global analysis of the plasma proteome revealed ER subtype agonists. While we recognize that perturbation of
disease-related alterations in a large number of genes and ER and ER signaling with pharmacological ER antagonists
proteins related to immune responses that were completely or and/or siRNA strategies might theoretically provide superior
partially reversed by ERB-041 administration (12). When approaches, no good selective ER and ER antagonists are
taken together, these findings support the current study and available, and small interfering RNA technology has not been
suggest the possibility that ER mediates the anti-inflamma- successful in modulating E2 signaling in SMCs in our hands.
tory effects of estrogen in some tissues. We have not tested whether ER activation, as studied here, is
PPT is 400 times more potent as an agonist for ER than sufficient to attenuate neointima formation in animals sub-
for ER (4). PPT in doses of 0.315 mgkg1 day1 admin- jected to endoluminal injury in vivo. We are aware that other
istered subcutaneously has been shown to evoke a number of ER-modulated processes, e.g., reendothelialization, contrib-
physiologically relevant E2-induced tissue responses (uterine ute to the injury response at later time points. Further study is
weight gain, prevention of OVX-induced body weight gain and needed to elucidate the ER subtype-specific effects of E2 in
loss of bone mineral density, reduction in plasma cholesterol modulating injury responses in arteries.
levels, increases in brain progesterone receptor mRNA levels, Taken together, our in vivo and in vitro results suggest that
and prevention of experimentally induced hot flushes) in rats medial VSMCs are an important source for the proinflamma-
(24). In the vasculature, PPT, and not DPN, has been shown to tory mediators that trigger the response to acute endoluminal
mediate estrogenic vasodilatory responses at physiological arterial injury. VSMCs of injured arteries are initial targets for
doses: Acute administration of PPT (1013-107 M) to pre- the anti-inflammatory actions of E2. In VSMCs, ER appears
contracted aortic rings from intact female rats dose depen- to be the dominated ER subtype that contributes to the anti-
dently induced an ER-dependent vasodilatory response equal inflammatory effects of E2.
to that of E2, whereas DPN had no acute effect on vasomo-
tion (4). GRANTS
Studies with knockout mice support important ER-medi- This work was supported, in part, by National Heart, Lung, and Blood
ated protective effects on vascular injury (19, 30). Pare et al. Institute Grants HL-07457, HL-64614, HL-75211, HL-68806; by American
(30) showed that complete ER knockout mice (ERKOST) lost Heart Association Grant AHA-0425455B; and by a Vascular Biology Working
the protective effects of E2 in a wire injury model. In this Group Fellows Award.
study, the primary outcomes of change in medial area, proteo-
glycan deposition, and VSMC proliferation (by bromode- REFERENCES
oxyuridine staining) were not altered by E2 administration in 1. Bacon K, Baggiolini M, Broxmeyer H, Horuk R, Lindley I, Mantovani
the ERKOST animals. In contrast, E2 did protect against wire A, Maysushima K, Murphy P, Nomiyama H, Oppenheim J, Rot A,
Schall T, Tsang M, Thorpe R, Van Damme J, Wadhwa M, Yoshie O,
injury-induced increases in vascular medial area and prolifer- Zlotnik A, Zoon K, the IUIS/WHO Subcommittee on Chemokine
ation of VSMCs in ER knockout mice (18). ER, but not Nomenclature. Chemokine/chemokine receptor nomenclature. J Inter-
ER, has also been shown to mediate the reendothelialization feron Cytokine Res 22: 10671068, 2002.
effect of E2 after electric carotid injury in mice with targeted 2. Bakir S, Mori T, Durand J, Chen YF, Thompson JA, Oparil S.
disruptions of ER genes (6). As we have previously demon- Estrogen-induced vasoprotection is estrogen receptor dependent: evidence
from the balloon-injured rat carotid artery model. Circulation 101: 2342
strated, E2 restores endothelial cell function after denudation in 2344, 2000.
the balloon injury model of the rat carotid artery as well (38), 3. Bishop-Bailey D, Burke-Gaffney A, Hellewell PG, Pepper JR, Mitchell
but this is a late event that occurs 2 4 wk postinjury (6). JA. Cyclo-oxygenase-2 regulates inducible ICAM-1 and VCAM-1 ex-
In vitro studies support ER subtype and cell type-specific pression in human vascular smooth muscle cells. Biochem Biophys Res
Commun 249: 44 47, 1998.
modulatory effects of E2 on vascular cell function (13). Selec- 4. Bolego C, Cignarella A, Sanvito P, Pelosi V, Pellegatta F, Puglisi L,
tive ER and ER mRNA anti-sense oligomers were used to Pinna C. The acute estrogenic dilation of rat aorta is mediated solely by
examine the ER subtype dependence of E2-induced inhibition selective estrogen receptor- agonists and is abolished by estrogen depri-
of PDGF-BB-induced p38 and p42/44 mitogen-activated pro- vation. J Pharmacol Exp Ther 313: 12031208, 2005.
tein kinase phosphorylation, migration, and proliferation in 5. Bolego C, Vegeto E, Pinna C, Maggi A, Cignarella A. Selective agonists
of estrogen receptor isoforms: new perspectives for cardiovascular dis-
porcine SMCs, and endothelial cells. The inhibitory effects of ease. Arterioscler Thromb Vasc Biol 26: 21922199, 2006.
E2 on porcine SMCs were abrogated by downregulation of 6. Brouchet L, Krust A, Dupont S, Chambon P, Bayard F, Arnal JF.
ER protein expression, whereas downregulation of ER had Estradiol accelerates reendothelialization in mouse carotid artery through

AJP-Heart Circ Physiol VOL 292 JUNE 2007 www.ajpheart.org


H2612 ESTROGEN AND INFLAMMATION

estrogen receptor-alpha but not estrogen receptor-beta. Circulation 103: neutrophil chemotaxis in injured arteries. Circulation 110: 1664 1669,
423 428, 2001. 2004.
7. Chen YF, Oparil S. Effects of sex steroids in vascular injury. In: 26. Modi WS, Yoshimura T. Isolation of novel GRO genes and a phyloge-
Endocrinology of Cardiovascular Function, edited by Levin ER and netic analysis of the CXC chemokine subfamily in mammals. Mol Biol
Nadler JL. New York: Kluwer Academic, 1998, p. 4559. Evol 16: 180 193, 1999.
8. Christiansen JF, Hartwig D, Bechtel JF, Kluter H, Sievers H, Schon- 27. Nakagawa H, Komorita N, Shibata F, Ikesue A, Konishi K, Fujioka
beck U, Bartels C. Diseased vein grafts express elevated inflammatory M, Kato H. Identification of cytokine-induced neutrophil chemoattrac-
cytokine levels compared with atherosclerotic coronary arteries. Ann tants (CINC), rat GRO/CINC-2 and CINC-2, produced by granulation
Thorac Surg 77: 15751579, 2004. tissue in culture: purification, complete amino acid sequences and char-
9. Dahinden CA, Clancy RM, Hugli TE. Stereospecificity of leukotriene acterization. Biochem J 301: 545550, 1994.
B4 and structure-function relationships for chemotaxis of human neutro- 28. Okamoto E, Couse T, De Leon H, Vinten-Johansen J, Goodman RB,
phils. J Immunol 133: 14771482, 1984. Scott NA, Wilcox JN. Perivascular inflammation after balloon angio-
10. Driscoll KE, Hassenbein DG, Howard BW, Isfort RJ, Cody D, Tindal plasty of porcine coronary arteries. Circulation 104: 2228 2235, 2001.
MH, Succhanek M, Carter JM. Cloning, expression, and functional 29. Oppenheim J, Zachariae C, Mukaida N, Matushima K. Properties of
characterization of rat MIP-2: a neutrophil chemoattractant and epithelial the novel pro-inflammatory supergene interkrine cytokine family. Annu
cell mitogen. J Leukoc Biol 58: 359 364, 1995. Rev Immunol 9: 617 648, 1991.
11. Feldmann M, Pusey CD. Is there a role for TNF- in anti-neutrophil 30. Pare G, Krust A, Karas RH, Dupont S, Aronovitz M, Chambon P,

Downloaded from http://ajpheart.physiology.org/ by 10.220.33.2 on September 12, 2017


cytoplasmic antibody-associated vasculitis? Lessons from other chronic Mendelsohn ME. Estrogen receptor-alpha mediates the protective effects
inflammatory diseases. J Am Soc Nephrol 17: 12431252, 2006. of estrogen against vascular injury. Circ Res 90: 10871092, 2002.
12. Follettie MT, Pinard M, Keith JC, Wang L, Chelsky D, Hayward C, 31. Ross R. Growth of smooth muscle cell in culture and formation of elastic
Kearney P, Thibault P, Paramithiotis E, Dorner AJ, Harris HA. Organ fibers. J Cell Biol 50: 172186, 1971.
messenger ribonucleic acid and plasma proteome changes in the adjuvant- 32. Shibata F, Konishi K, Kato H, Komorita N, Al-Mokdad M, Fujioka
induced arthritis model: responses to disease induction and therapy with M, Nakagawa H. Recombinant production and biological properties of rat
the estrogen receptor- selective agonist ERB-041. Endocrinology 147: cytokines-induced neutrophil chemoattractants, GRO/CINC-2 , CINC-2
714 723, 2006. and CINC-3. Eur J Biochem 231: 306 311, 1995.
13. Geraldes P, Sirois MG, Tanguay JF. Specific contribution of estrogen 33. Shibata F, Konishi K, Nakagawa H. Gene structure, cDNA cloning, and
receptors on mitogen-activated protein kinase pathways and vascular cell expression of the rat cytokine-induced neutrophil chemoattractant-2
activation. Circ Res: 399 405, 2003. (CINC-2) gene. Cytokine 10: 169 174, 1998.
14. Hansson GK. Inflammation, atherosclerosis, and coronary artery disease. 34. Wang JM, Sica A, Peri G, Walter S, Padura IM, Libby P, Ceska M,
N Engl J Med 352: 16851695, 2005. Lindley I, Colotta F, Mantovani A. Expression of monocyte chemotactic
15. Harrington WR, Sheng S, Barnett DH, Petz LN, Katzenellenbogen
protein and interleukin-8 by cytokine-activated human vascular smooth
JA, Katzenellenbogen BS. Activities of estrogen receptor alpha- and
muscle cells. Arterioscler Thromb 11: 1166 1174, 1991.
beta-selective ligands at diverse estrogen responsive gene sites mediating
35. Watts JA, Zagorski J, Gellar MA, Stevinson BG, Kline JA. Cardiac
transactivation or transrepression. Mol Cell Endocrinol 206: 1322, 2003.
inflammation contributes to right ventricular dysfunction following exper-
16. Harris HA, Albert LM, Leathurby Y, Malamas MS, Mewshaw RE,
imental pulmonary embolism in rats. J Mol Cell Cardiol 41: 296 307,
Miller CP, Kharode YP, Marzolf J, Komm BS, Winneker RC, Frail
2006.
DE, Henderson RA, Zhu Y, Keith JC Jr. Evaluation of an estrogen
36. Weathington NM, van Houwelingen AH, Noerager BD, Jackson PL,
receptor-beta agonist in animal models of human disease. Endocrinology
Kraneveld AD, Galin FS, Folkerts G, Nijkamp FP, Blalock JE. A
144: 4241 4249, 2003.
17. Horvath C, Welt FG, Nedelman M, Rao P, Rogers C. Targeting CCR2 novel peptide CXCR ligand derived from extracellular matrix degradation
or CD18 inhibits experimental in-stent restenosis in primates: inhibitory during airway inflammation. Nat Med 12: 317323, 2006.
potential depends on type of injury and leukocytes targeted. Circ Res 90: 37. Welt FG, Rogers C. Inflammation and restenosis in the stent era.
488 494, 2002. Arterioscler Thromb Vasc Biol 22: 1769 1776, 2002.
18. Inoue S, Egashira K, Ni W, Kitamoto S, Usui M, Otani K, Ishibashi 38. White CR, Shelton J, Chen SJ, Darley-Usmar V, Allen L, Nabors C,
M, Hiasa K, Nishida K, Takeshita A. Anti-monocyte chemoattractant Sanders PW, Chen YF, Oparil S. Estrogen restores endothelial cell
protein-1 gene therapy limits progression and destabilization of estab- function in an experimental model of vascular injury. Circulation: 1624
lished atherosclerosis in apolipoprotein E-knockout mice. Circulation 106: 1630, 1997.
2700 2706, 2002. 39. Wirthlin MR, Hussain MZ. Clinical and light microscopic observations
19. Karas RH, Hodgin JB, Kwoun M, Krege JH, Aronovitz M, Mackey of gingivitis and early ligature-induced periodontitis in the cynomolgus
W, Gustafsson JA, Korach KS, Smithies O, Mendelsohn ME. Estrogen monkey. J Periodontol 63: 533539, 1992.
inhibits the vascular injury response in estrogen receptor beta-deficient 40. Xing D, Miller A, Novak L, Rocha R, Chen YF, Oparil S. Estradiol and
mice. Proc Natl Acad Sci USA 96: 1513315136, 1999. progestins differentially modulate leukocyte infiltration after vascular
20. Li G, Chen SJ, Oparil S, Chen YF, Thompson JA. Direct in vivo injury. Circulation 109: 234 241, 2004.
evidence demonstrating neointimal migration of adventitial fibroblasts 41. Yonezawa K, Yamamoto Y, Yamamoto H, Ishikawa Y, Uchinami H,
after balloon injury of rat carotid arteries. Circulation 101: 13621365, Taura K, Nakajima A, Yamaoka Y. Suppression of tumor necrosis
2000. factor-alpha production and neutrophil infiltration during ische-
21. Li G, Chen YF, Greene GL, Oparil S, Thompson JA. Estrogen inhibits mia-reperfusion injury of the liver after heat shock preconditioning.
vascular smooth muscle cell-dependent adventitial fibroblast migration in J Hepatol 35: 619 627, 2001.
vitro. Circulation 100: 1639 1645, 1999. 42. Yu HP, Hsieh YC, Suzuki T, Choudhry MA, Schwacha Chaudry IH.
22. Lucas AD, Bursill C, Guzik TJ, Sadowski J, Channon KM, Greaves Salutory effects of estrogen receptor- agonist on lung injury after
DR. Smooth muscle cells in human atherosclerotic plaques express the trauma-hemorrhage. Am J Physiol Lung Cell Mol Physiol 290: L1004
fractalkine receptor CX3CR1 and undergo chemotaxis to the CX3C L1009, 2006.
chemokine fractalkine (CX3CL1). Circulation 108: 2498 2504, 2003. 43. Yu HP, Shimizu T, Choudhry MA, Hsieh YC, Suzuki T, Bland KI,
23. Luster AD. Chemokines: chemotactic cytokines that mediate inflamma- Chaudry IH. Mechanism of cardioprotection following trauma-hemor-
tion. N Eng J Med 338: 436 445, 1998. rhagic shock by a selective estrogen receptor- agonist: up-regulation of
24. Meyers MJ, Sun J, Carlson KE, Marriner GA, Katzenellenbogen BS, cardiac heat shock factor-1 and heat shock proteins. J Mol Cell Cardiol 40:
Katzenellenbogen JA. Estrogen receptor-beta potency-selective ligands: 185194, 2006.
structure-activity relationship studies of diarylpropionitriles and their 44. Yu HP, Shimizu T, Hsieh YC, Suzuki T, Choudhry MA, Schwacha
acetylene and polar analogues. J Med Chem 44: 4230 4251, 2001. MG, Chaudry IH. Tissue-specific expression of estrogen receptors and
25. Miller AP, Feng W, Xing D, Weathington NM, Blalock JE, Chen YF, their role in the regulation of neutrophil infiltration in various organs
Oparil S. Estrogen modulates inflammatory mediator expression and following trauma-hemorrhage. J Leukoc Biol 79: 963970, 2006.

AJP-Heart Circ Physiol VOL 292 JUNE 2007 www.ajpheart.org