Microvascular Research 68 (2004) 273 – 285 www.elsevier.com/locate/ymvre
Tissue-specific effect of estradiol on endothelial cell-dependent lymphocyte recruitment Hedwig S. Murphya,b,*, Quan Suna, Brian A. Murphyc, RuRan Mod, Jirong Huod, Jun Chend, Stephen W. Chensuea,b, Matthew Adamsd, Bruce C. Richardsond, Raymond Yungd,e a
Department of Pathology, University of Michigan, Ann Arbor, Michigan 48109, United States b Pathology and Laboratory Medicine, Ann Arbor, Michigan 48109, United States c John Carroll University, Cleveland, Ohio 44118, United States d Department of Internal Medicine, University of Michigan, Ann Arbor, Michigan 48109, United States e GRECC, Veterans Administration Ann Arbor Health Service, Ann Arbor, Michigan 48109, United States Received 1 October 2003 Available online 3 August 2004
Abstract Estrogen profoundly affects onset and severity of many immune-mediated diseases. In our murine model of drug-induced autoimmunity, female-specific, estrogen-dependent increase in splenic lymphocyte homing was directly implicated in increased disease severity. The present study evaluated the effect of estradiol on microvascular endothelial cells from the spleen compared to endothelial cells from the dermis, which has no disease manifestation in this model. Estradiol increased spleen endothelial cell estrogen receptor (ER) alpha 2.9-fold and decreased estrogen receptor beta 2.1-fold while decreasing both receptors on dermal cells. Estradiol enhanced adhesion of D10 cells to spleen but not dermal endothelial cells 1.53-fold ( P b 0.001), an increase that was inhibited by antibodies to VCAM-1 and ICAM-1, and by the estrogen receptor antagonists tamoxifen and ICI 182,780. Estradiol induced greater VCAM-1 expression on spleen than dermal endothelial cells ( P b 0.05). Estradiol increased spleen endothelial cell estrogen receptor alpha 2.9-fold and decreased estrogen receptor beta 2.1-fold while decreasing both receptors on the dermal cells. Estrogen specifically and preferentially promoted spleen chemokine protein expression for MCP-1 and MCP-3, while having no effect on dermal protein expression for these chemokines. Estradiol-mediated effects on splenic chemokines were abrogated by tamoxifen and ICI 182,780. The gender-specific increase in lymphocyte homing to spleen may be attributable, at least in part, to tissue-specific estrogen-mediated effects on microvascular endothelial cells. D 2004 Elsevier Inc. All rights reserved. Keywords: Endothelial cells; Estrogen; Adhesion molecules; Chemokines; Autoimmunity
Introduction Hormone-mediated gender differences in immune function are evident in the onset and severity of a wide range of immune-mediated diseases (Whitacre et al., 1999). Lymphocyte-mediated autoimmune diseases are more prevalent in females as compared to male; females have higher * Corresponding author. Department of Pathology, University of Michigan, 5240 Medical Science I, 1301 Catherine, Ann Arbor, Michigan 48109. Fax: +1 734 761 5037. E-mail address:
[email protected] (H.S. Murphy). 0026-2862/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.mvr.2004.06.004
immunoglobulin levels and stronger immune responses following infection or immunization; female gender is associated with increased risk for the development of autoimmunity (Whitacre et al., 1999). Increasing evidence implicates sex steroid hormones, especially estrogen, as mediators of these differences. Manifestation of many autoimmune diseases correlates with the time frame of increased estrogen in women. For example, SLE primarily affects women in the reproductive years with flares during pregnancy and with oral contraceptive use (Ahmed and Verthelyi, 1993; Ahmed et al., 1985; Da Silva, 1995; Fox, 1995; Friedman and Waksman, 1997; Lahita, 1993; Wilder,
274
H.S. Murphy et al. / Microvascular Research 68 (2004) 273–285
1998). While sex hormones alone are not responsible for the development of autoimmune diseases, their influence on the immune system appears to result in enhanced immune responsiveness leading to increased disease severity (Da Silva, 1995). In some animal models of systemic lupus erythematosus, female gender and sex hormones alter disease development (Aronica et al., 2000; Dhaher et al., 2000; Talal, 1989). Our group has developed a model of druginduced lupus to identify novel gender-specific immune mechanisms (Yung et al., 1995, 1997). In this model, D10 cells, a cloned Th2 line derived from AKR mice, are made autoreactive by treatment with DNA methylation inhibitors including 5-azacytidine, procainamide, and hydralazine, then injected into un-irradiated syngeneic recipients. Similar to the NZB/NZW and MRL/lpr lupus mice, female mice receiving the drug-treated autoreactive T cells developed a more severe disease than males. Castration of female mice before the adoptive transfer protocol resulted in a milder disease, with less autoantibody production and absence of renal disease. Significantly, more D10 cells accumulated in the female spleens and this selective retention also decreased following oophorectomy, implicating a role for sex hormones in homing and disease severity. Splenectomy prevents the development of autoimmunity, indicating that the spleen is essential to disease development and that the increased disease severity seen in female mice is due in part to quantitatively more autoreactive T cells accumulating in the spleen (Yung et al., 1997). The spleen, a major secondary lymphoid organ participating in the immune response to blood borne antigens, has been implicated in the progression of human and experimental animal immune diseases, including rheumatoid arthritis, diabetes mellitus, and systemic lupus erythematosus. These findings prompted an evaluation of estrogen effects on resident cells of the spleen, specifically the microvascular endothelial cells, which play an essential role in recruitment of cells during the development of immune and inflammatory responses. Estrogen exerts its biological action on cells expressing the high-affinity estrogen receptors (ERs). Two genetically distinct receptors ERa and ERh, identified in human, rat, and mouse cells, bind 17h-estradiol with similar affinity (Kuiper et al., 1997) and are expressed on vascular endothelial cells (Couse and Korach, 1999; Geraldes et al., 1997; Russell et al., 2000a,b). Estrogen effects on vascular endothelium have been examined primarily in large vessels in the context of atherogenesis and in vitro after short-term exposure to estrogen (Caulin-Glaser et al., 1996; Cid et al., 1994; Seli et al., 2001). In those cells, estrogen induces nitric oxide release from eNOS, activation of guanylate cyclase, enhances TNFa-stimulated Eselectin and VCAM-1 expression (Russell et al., 2000a,b; Simoncini and Genazzani, 2000; Simoncini et al., 2002; Zhang et al., 2002) Endothelial cells from tissue micro-
vasculature have a significantly different function than endothelial cells derived from large vessels and may have differing responses to hormone exposure. In the microvasculature, endothelial cells actively recruit immune and inflammatory cells to peripheral tissues and in primary or secondary lymphoid tissue; the recruitment of lymphocytes is central to the immune response. There is currently little information regarding microvascular endothelial cells in the spleen, the effect of hormones on these cells, or the role of hormone-mediated endothelial cell responses during disease development. Our previous studies have demonstrated tissue specificity of endothelial cell responses (Murphy et al., 1998), suggesting that hormonal modulation of these responses may be tissue-specific as well. To understand how estrogen might affect endothelial cell function to ultimately modulate lymphocyte recruitment to the spleen in our model of drug-induced autoimmunity, we exposed endothelial cells to estradiol for 5 days and evaluated adhesion molecule expression, adhesion of lymphocytes, chemokine expression, chemotaxis of lymphocytes, and ER expression. Microvascular endothelial cells from the spleen were compared to endothelial cells from the dermis, which has no manifestations during druginduced disease. Our results demonstrate that long-term estradiol exposure upregulated ERa and downregulated ERh expression, and selectively enhanced adhesion molecule expression, adhesion of D10 cells, and chemokine expression in spleen endothelial cells.
Materials and methods Animals For isolation and culture of endothelial cells, female, 4week-old AKR mice were obtained from Jackson Laboratories, Bar Harbour, ME. For the homing experiments, 6- to 8-week-old oophorectomized mice were purchased from Jackson Laboratory. The mice were allowed to recover for at least 4 weeks after the surgery. h-Estradiol (60-day release, 0.36 mg/pellet achieving blood level of 150–200 pg/ml), progesterone (60-day release, 25 mg/ pellet achieving blood level of 15–20 ng/ml), or placebo pellets (all from Innovative Research of America, Sarasota, FL) were implanted under the skin on the lateral side of the neck of the animals. Experiments were done on the mice 2–3 weeks after the pellets had been implanted. All animal procedures were performed according to the Guide for the Care and Use of Laboratory Animals (NIH, 1996). D10 cell culture D10 cells, a conalbumin reactive Th2 line derived from the AKR mice, were obtained from the American Type
H.S. Murphy et al. / Microvascular Research 68 (2004) 273–285
Tissue Culture Collection (Rockville, MD) and cultured as previously described (Yung et al., 1995, 1997). The cells were maintained by weekly restimulation with irradiated splenocytes from AKR/J mice and conalbumin (100 Ag/ ml) and were used at least 4 days after the challenging. In vivo homing of D10 cells D10 cell splenic homing in oophorectomized mice supplemented with h-estradiol, progesterone, or placebo pellets was determined as we have described (Yung et al., 1997). Briefly, D10 cells were labeled with 5-chloromethylfluorescein diacetate (CMFDA) (Molecular Probes, Eugene, OR) using the protocol provided by the manufacturer. CMFDA-labeled cells (5 106) were injected intravenously via the tail vein of the hormone-treated or placebo-treated mice (total 10 mice in each treatment group). The mice were killed 24 h later and the CMFDAlabeled D10 cells were detected by staining splenocytes with phycoerythrin (PE)-conjugated anti-CD4 (PharMingen, San Diego, CA). The percentage of CMFDA-positive CD4 cells were then enumerated using a Coulter ELITE flow cytometer. Endothelial cell culture Microvascular endothelial cells were isolated from spleens of 28-day-old female mice using methods we have previously established for isolation of microvascular endothelial cells (Murphy et al., 1998). Briefly, strips of peripheral spleen were removed, minced, and incubated in 1% gelatin-coated 25 cm2 tissue culture flasks in growth media consisting of phenol red-free RPMI-1640 (GIBCO, Invitrogen Corp, Auckland, NZ), supplemented with 20% fetal calf serum (Hyclone, Logan, UT), penicillin– streptomycin, and endothelial cell growth supplement. Individual cells developed into colonies of endothelial cells with few scattered dendritic cells. Cells were further purified by negative selection with antibodies to CD11b (BD Biosciences, Bedford, MA) and Dec205 (Serotec, Raleigh, NC), and magnetic beads bearing appropriate immunoglobulin. Microvascular endothelial cells were isolated from mouse ear dermis as we have previously described (Murphy et al., 1999). Ears were removed, split into two pieces, and incubated in 5 mg/ml dispase II for 45 min. The microvasculature was removed and the individual endothelial cells were released into plating medium using the blunt end of a scalpel. For both cell types, endothelial cells were maintained in culture for 10 days, followed by passage to plates appropriate to each assay. Cultures consisted of a single uniform cell population, with cobblestone morphology and good uptake of fluorescent-labeled acetylated low density lipoprotein (1,1V-diotadecyl-3,3,3V,3V-tetramethylindo-carbocyanine perchlorate, DiI-Ac-LDL). Cells were uniformly positive with anti-CD31 (PECAM-1) by flow cytometry and immunostaining. Compared to human serum,
275
endothelial cells had significant angiotensin-converting enzyme activity. For all experimental studies, cells were used at 80–90% confluence after one passage (P1). 17h-Estradiol (Sigma Corp, St Louis, MO) was added for the final 5 days of culture in the presence and absence of ER antagonists. ICI 182,780 (Fulvestrant, Faslodex) was obtained as an investigative compound from AstraZeneca, Inc. Tamoxifen was from ICN Biomedicals Inc. (Aurora, OH). Estrogen receptor expression Endothelial cells were grown to confluence in 100-mm plates. Cell lysates were subjected to SDS-PAGE on 10% acrylamide gels and transferred to PVDF membranes. Blots were probed with anti-ERa or anti ERh antibodies (Santa Cruz Biotech, Santa Cruz, CA). Scanning densitometry was performed and analyzed using an MCID imaging system from Imaging Research. Quantitation of chemokine protein Chemokines were measured in supernatant media by specific ELISA using commercially available reagents (R&D Systems, Minneapolis, MN) as we have previously described (Shang et al., 2002). Briefly, Nunc immunoELISA plates were coated with the appropriate cytokine capture antibody at a dilution of 1 Ag/ml of coating buffer for 16 h at 48C. Excess capture antibody was washed away and each plate blocked for 90 min with 2% bovine serum albumin in phosphate-buffered saline (BSA–PBS) at 378C. Each plate was washed with PBS–Tween-20 and samples added to wells in duplicate for 1 h at 378C. Recombinant murine cytokine standard curves were used to calculate cytokine concentrations. Plates were washed and biotinylated polyclonal rabbit anti cytokine antibody (3.5 Ag/ml) was added followed by streptavidin–peroxidase for 30 min. Each plate was washed, chromagen substrate added, and plates were read on an ELISA plate scanner at 492 nm. Chemokine levels were normalized to final cell number for each data point. Quantitation of chemokine mRNA Poly (A) pure mRNA was isolated from cells using a Poly (A) pure mRNA isolation kit (Ambion, Austin TX) following manufacturers instructions as we have previously described (Shang et al., 2002). Approximately, 1 Ag of mRNA was reverse-transcribed in a 20-Al reaction using a Reverse Transcription System kit (Promega, Madison, WI). Triplicate reactions were performed for each data point. The products were subjected to real-time PCR using a Taqman 7000 light cycler (Applied Biosystems, Foster City, CA). GAPDH was used as an endogenous reference. The generating PCR products of the target gene and GAPDH were monitored simultaneously in real-time with a fluorescence amplification factor measured for each gene
276
H.S. Murphy et al. / Microvascular Research 68 (2004) 273–285
relative to GAPDH. Data were expressed as arbitrary units (AU) calculated from the fluorescence amplification factor as measured by the real-time PCR fluorescent detection unit. The original gene copy number is related to fluorescence of the generated signal and expressed as an arbitrary unit (AU) AU ¼ F E 1 I 2n F is an arbitrary conversion constant, E 1 is an amplification efficiency constant (approximately = 1 for manufacturer’s real-time primer sets), I is the fluorescent intensity reading, and n is the amplification cycle number. Hence, Co constitutes an arbitrary measure of copy number that is related to the fluorescent product and inversely related to cycle number. Taqman predeveloped reaction kits (Perkin–Elmer) were used. Oligonucleotides for PCR primers were obtained from Operon Technologies (Alameda, CA) and Taqman probes were purchased from Applied Biosystems. In all cases, Taqman Universal PCR Master Mix (Perkin–Elmer) was used and the thermal cycling condition was programmed according to the manufacturer’s instructions. Before being used in the actual mRNA expression analysis, each primerprobe set was pretested with an undiluted positive control sample in 1:4 and 1:16 dilutions. Water served as a negative control.
Flow cytometry Adhesion molecule expression was evaluated by direct immunofluorescence staining of whole cells. After brief trypsinization and washing, resuspended cells were incubated in FITC-labeled or PE-labeled antibody for 30 min on ice. After washing, cells were resuspended in staining buffer (DMEM plus 0.1% FCS plus 0.1% sodium azide). Cells were gated on typical forward and side scatter profiles and analyzed on a Becton-Dickenson Flow Cytometer using Cell Quest software. Static adhesion assay Adhesion of D10 cells to spleen microvascular endothelial cells was assessed in a standard adhesion assay. D10 cells were prepared as previously described (Yung et al., 1995). Na251CrO4-labeled D10 cells were added to 24-well monolayers of endothelial cells previously exposed to estradiol (for 5 days) and/or additional stimuli. After incubation and washing, lymphocytes were lysed and percentage of bound lymphocytes calculated as % lymphocytes bound ¼
cpm in lysate 100 cpm in original lymphocyte suspension
Chemotaxis assay Statistics In vitro dual-chamber chemotaxis assays were performed as we have described (Mo et al., 2003). D10 cells (4 105) in 100 Al of RPMI 1640 medium supplemented with 0.5% BSA were placed in Transwell Clear culture inserts with 5-Am pores (Corning-Costar, Cambridge, MA). The inserts were then placed in a 24-well tissue culture plate (Corning-Costar) containing 600 ml of supernatant from splenic and dermal endothelial cells cultured with or without h-estradiol treatment (2.5 ng/ml final concentration). Two hours later, the cells from the top and bottom chambers were harvested and counted with a Beckman Coulter counter (Fullerton, CA). D10 chemokine receptors D10 cell chemokine receptor expression profile was determined by ribonuclease protection assay as before using a modification of the manufacturer’s protocol (Mo et al., 2003). GACU nucleotide pool and [a-32P]UTP, RNasin, T7 RNA polymerase were added to the multiprobe template set mCR-5 (CCR1-5) and mCR-6 (CXCR2, 4 and 5) (BD Biosciences). Five micrograms of total RNA was used for hybridization. The protected probes were then allowed to be resolved by electrophoresis using a 5% acrylamide gel, exposed to a phosphor screen, and quantified by a PhosphoImager using Image Quaint Software (Molecular Dynamics).
Where appropriate, a Student’s t test was used to compare control with experimental groups. Statistical significance was defined as P b 0.05.
Results Effect of hormones on in vivo homing of lymphocytes to spleen In our previous report, female gonadal hormones were implicated in recruitment of D10 cells to the spleen during the development of autoimmunity (Yung et al., 1997). To confirm a role for estrogen in this recruitment, oophorectomized mice were supplemented with estrogen or progesterone then injected with CMFDA-labeled D10 cells. The 17h-estradiol-treated mice had a 2.0-fold increase ( P = 0.043) in splenic homing versus control and a 2.5-fold increase ( P = 0.032) versus progesterone (Fig. 1), indicating that estrogen is the hormone primarily responsible for the increased splenic homing. Endothelial cells express estrogen receptors ERa and ERh function as transcription factors mediating estrogen-induced gene expression. Dermal and spleen
H.S. Murphy et al. / Microvascular Research 68 (2004) 273–285
277
Estradiol increases adhesion of D10 cells to spleen endothelial cells
Fig. 1. Estrogen increases in vivo homing of D10 cells to spleen. Relative accumulation of D10 cells in oophorectomized mice supplemented with hormones. Data are the mean F SEM of 10/group (2 independent experiments using 5 mice/group). *P = 0.043, estrogen-treated compared to control, and P = 0.032 estrogen-treated compared to progesteronetreated.
endothelial cells expressed both the a and h forms of the estrogen receptor, evaluated at the protein level by Western blot analysis (Fig. 2). In spleen endothelial cells, estradiol (2.5 ng/ml) increased expression of ERa 2.9-fold while decreasing expression of ERh 2.1-fold. Estradiol had less effect on dermal cells, decreasing both a and h forms by 1.1- and 1.06-fold, respectively.
Fig. 2. Estrogen receptor expression. Western blot analysis of ERa and ERh expression in (A) dermal and (B) spleen endothelial cells cultured in the presence and absence of estradiol (2.5 ng/ml) for 5 days. Graphs represent densitometry measurements of the immunoblots. Data from a single representative experiment. Duplicate experiments yielded similar data.
To evaluate the effect of estradiol on lymphocyte– endothelial cell interactions, adhesion of cloned Th2 cells (D10 cells) to spleen and dermal endothelial cells in vitro was determined after exposure of endothelial cells to 17hestradiol (0–2.5 ng/ml) for 5 days, followed by incubation with D10 cells (cultured in the absence of estradiol). Estradiol increased adhesion of D10 cells to spleen endothelial cells in a concentration dependent manner (1.53-fold, P b 0.001, adhesion of spleen cells exposed to 2.5 ng/ml estradiol compared to cells in the absence of estradiol), while having little effect on adhesion to dermal endothelial cells (Fig. 3). Adhesion was inhibited by antibodies to ICAM-1 and VCAM-1 ( P b 0.001 for all values) with an enhanced inhibitory effect when the antibodies were used together. ER antagonists inhibit estradiol-mediated adhesion To demonstrate that estradiol exerted its effects on adhesion via interaction with ER, endothelial cells were exposed to estradiol supplemented with ER antagonists
Fig. 3. Effect of estradiol on adhesion of D10 cells to endothelial cells. Dermal (A) and spleen (B) endothelial cells exposed to estradiol (0–25 ng/ ml) for 5 days. Adhesion of D10 cells to endothelial cells was in the presence and absence of antibodies to ICAM-1, VCAM-1, and ICAM-1 plus VCAM-1. Data are normalized values and represent mean F SEM of three data points from one experiment. Duplicate experiments yielded similar results. *P b 0.001, adhesion of spleen cells exposed to estradiol compared to spleen endothelial cells in the absence of estradiol.
278
H.S. Murphy et al. / Microvascular Research 68 (2004) 273–285
tamoxifen (106) or ICI 182,780 (106) before incubation with D10 cells (cultured in the absence of estradiol) (Fig. 4). Estradiol-induced adhesion was completely abrogated by the ER antagonists ( P b 0.001, for all values comparing data in the presence and absence of inhibitors), indicating that estradiol was competitively inhibited by tamoxifen or ICI 182,780 and that estradiol was acting via ER to promote adhesion. Adhesion molecule expression increases with estradiol Reciprocal expression of adhesion molecules on endothelial cells and circulating lymphocytes is responsible for adhesive interactions between these cells and is critical to recruitment to tissues. Expression of adhesion molecules was evaluated by flow cytometry of endothelial cells incubated in 17h-estradiol (0–25 ng/ml) for 5 days without additional stimuli. Spleen and dermal microvascular endothelial cells expressed VCAM-1, ICAM-1, and Eselectin. In the absence of estradiol but under the same culture conditions, a twofold higher percentage of spleen endothelial cells were positive for VCAM-1 expression ( P b 0.05, Fig. 5), with higher mean fluorescence intensity as compared to dermal endothelial cells and isotype control (Fig. 6). In the presence of increasing concentrations of estradiol, both spleen and dermal endothelial cells displayed increased surface expression of VCAM-1, evident as an increase in positive cells (Fig. 5) as well as an increase in the mean fluorescence intensity (Fig. 6A, B). However, VCAM-1 expression was consistently higher on the spleen as compared to dermal endothelial cells and the effect of estradiol was greater on spleen as compared to dermal endothelial cells ( P b 0.05 for all values comparing dermal to spleen VCAM-1 expression). Estradiol had no effect on expression of E-selectin or ICAM-1 on either dermal or spleen endothelial cells (data not shown).
Fig. 4. Inhibition of estradiol-mediated adhesion. Adhesion of D10 cells to spleen endothelial cells exposed to estradiol (0–2.5 ng/ml) for 5 days in the presence and absence of tamoxifen (1 106) or ICI 182,780 (1 106). Data are the mean F SEM of three data points. *P b 0.001 comparing data in the presence and absence of inhibitor.
Fig. 5. Estradiol increases VCAM-1-positive cells. Relative ratio of percent spleen and dermal EC expressing VCAM-1 after exposure to increasing concentrations of estradiol, 0–25 ng/ml for 5 days, evaluated by flow cytometry. Data are from three separate experiments and are the mean F SEM of the relative ratios. *P b 0.05 increase in spleen EC compared to increase in dermal EC VCAM-1 positive cells.
Fig. 6. Estradiol increases VCAM-1 expression. VCAM-1 expression on dermal (A) and spleen (B) endothelial cells exposed to estradiol (0–25 ng/ ml) for 5 days was determined by flow cytometry. Fluorescence intensity of cells expressing VCAM-1 after exposure to increasing concentrations of estradiol, compared to isotype control. Data are from a single representative experiment. Three separate experiments yielded similar data.
H.S. Murphy et al. / Microvascular Research 68 (2004) 273–285
Fig. 7. Chemokine receptor expression on D10 cells. CC and CXC receptor expression on D10 cells determined by RPA. Data are representative of duplicate experiments.
Effect of estradiol on D10 Cells Expression of reciprocal adhesion molecules on D10 cells was evaluated in cells cultured in the presence and absence of estradiol (0.5–50 ng/ml for 24 h). Estradiol had no effect on D10 expression of P-selectin, E-selectin, CD49d, CD11a, or CD43 (data not shown). Altered expression of chemokines by splenic endothelial cells, or chemokine receptor expression on T cells, could also contribute to gender-specific T-cell trafficking. D10 chemokine receptor profile was therefore examined by RPA. D10 cells were found to express CCR1, CCR4, CXCR4, and to a less extent CXCR 2 (Fig. 7). Exposure of D10 cells to estradiol (0.025–25 ng/ml for 24 h) increased RNA transcript for CCR1 2.49-fold ( P = 0.012) and CCR4 2.47fold ( P = 0.048), using ribonuclease protection assays. These increases have been confirmed at the protein level using immunoblotting for CCR1 and flow cytometry for CCR4 (data not shown).
279
chemotactic protein-3 (MCP-3/CCL7), macrophage inhibitory protein-2 (MIP-2/CXCL1), and macrophage inhibitory protein-1a (MIP-1a/CCL3), and no evidence of RANTES (CCL5), cytokine-induced neutrophil chemoattractant (KC/ CXCL1), macrophage inhibitory protein-1h (MIP-1h, CCL4), Thymus and activation-regulated chemokine (TARC/CCL17), or secondary lymphoid chemokine (SLC/ CCL21) (data not shown). After exposure of dermal and spleen endothelial cells to estradiol for 5 days, cell-free supernatant media and RNA were collected. Estradiol alone (2.5 ng/ml), in the absence of additional stimuli, substantially and significantly increased spleen endothelial cell RNA transcripts for MCP-1 (11.27-fold increase) and MCP3 (13.48-fold increase), and to a lesser degree MIP-2 (4.3fold increase), as determined by real-time RT-PCR (Fig. 8). There was little effect on KC (b1.0 AU) and no effect on SLC, RANTES, MIP-1a, MIP-1h, MDC, or TARC (data not shown). Estradiol had a greater effect on MCP-1 and MCP-3 RNA in spleen endothelial cells than in dermal endothelial cells. Estradiol was effective at promoting chemokine protein generation. In the spleen but not the dermal endothelial cells, MCP-1 and MCP-3 protein were generated in substantial quantity and were increased by estradiol in a
Estradiol enhances chemokine generation Estradiol-mediated increase in chemokine receptor expression on D10 cells suggested a role for chemokines in enhanced splenic homing in the murine model. Vascular endothelial cells have a limited but important chemokine repertoire. To determine if estrogen-enhanced chemokine expression might play a role in the differential recruitment of D10 cells to the spleen in preference to other tissues such as the dermis, we evaluated the effect of estradiol on chemokine expression, determining protein and mRNA transcripts for CC and CXC chemokines. In the absence of additional cytokine stimulation, endothelial cells, both dermal and spleen, expressed low levels of protein of monocyte chemotactic protein-1 (MCP-1/CCL2), monocyte
Fig. 8. Estradiol increases endothelial cell chemokine RNA transcripts. (A) Dermal and (B) spleen endothelial cells were exposed to estradiol (0–25 ng/ ml) for 5 days, and chemokine RNA determined by real time RT-PCR. *P b 0.05, cells in the presence of estradiol compared to cells in the absence of estradiol. Data are from a single representative experiment. Duplicate experiments yielded similar data.
280
H.S. Murphy et al. / Microvascular Research 68 (2004) 273–285
exposed to estradiol supplemented with tamoxifen or ICI 182,780. We examined inhibition of MCP-1, which was generated in the greatest quantity, in spleen endothelial cells, which had the strongest response to estradiol. When present with estradiol, tamoxifen and ICI 182,780 each significantly reduced spleen endothelial cell MCP-1 RNA expression (*P b 0.05, cells in the presence of estrogen without inhibitor compared to cells in the presence of estrogen and inhibitor, Fig. 10A) as well as chemokine protein generation (*P b 0.05, cells in the presence of estrogen without inhibitor compared to cells in the presence of estrogen and inhibitor, Fig. 10B), which was reduced to levels equivalent to that seen in cells cultured in the absence of estradiol. These data indicated that estradiol was competitively inhibited by the two ER antagonists.
Fig. 9. Estradiol increases spleen endothelial cell chemokine protein. Dermal (—5—) and spleen (—n—) endothelial cells were exposed to estradiol (0–25 ng/ml) for 5 days, and chemokine protein determined by ELISA. Data are from a single representative experiment. Three separate experiments yielded similar data. *P b 0.05, data in the presence of estradiol compared to data in the absence of estradiol.
concentration-dependent manner (MCP-1, 1.6-fold increase, MCP-3, 1.4-fold increase after exposure to estradiol 2.5 ng/ ml for 5 days) (Fig. 9). Estradiol had a greater effect on spleen chemokines than dermal chemokines ( P b 0.00001 for MCP-1 and MCP-3, comparing increase in response to estradiol in spleen to dermal endothelial cells), correlating well with RNA expression. KC, RANTES, MIP-1a, SLC, or TARC were not generated in the presence of estradiol alone, requiring additional cytokine stimulation for generation of these chemokines (data not shown). ER antagonists inhibit estradiol-mediated chemokine expression To confirm that estradiol enhanced chemokine production via ligand interaction with ER, endothelial cells were
Fig. 10. Inhibition of estradiol effect on MCP-1. MCP-1 expression by spleen endothelial cells exposed to estradiol (0–2.5 ng/ml) for 5 days in the presence and absence of tamoxifen (105, 106) or ICI 182,780 (105, 106). Data are the mean F SEM of three data points. *P b 0.05, cells in the presence of estrogen without inhibitor compared to cells in the presence of estrogen and inhibitor.
H.S. Murphy et al. / Microvascular Research 68 (2004) 273–285
Fig. 11. Chemotaxis of D10 cells. D10 cells, in the absence of estradiol, transmigrate to cell-free supernatant media from dermal and spleen endothelial cells cultured in the presence and absence of 2.5 ng/ml estradiol. Data represent the transmigrated cells as percent of total cells and are from a representative experiment. Three experiments yielded similar data. *P b 0.005, cells in the presence of estradiol compared to cells in the absence of estradiol.
Chemotaxis To determine if chemokine expression correlated functionally with altered chemotactic responses, D10 cells were exposed to cell-free supernatant media from spleen and dermal endothelial cells cultured in the presence and absence of estradiol. Estradiol increased D10 cell transmigration to both dermal and spleen endothelial cell supernatant media (Fig. 11) ( P b 0.01, dermal or spleen endothelial cells in the presence of estradiol compared to cells in the absence of estradiol). However, in spite of the fact that estradiol clearly enhanced MCP-1, MCP-3, and MIP-2 production by spleen endothelial cells to a greater extent than dermal endothelial cells, there was no significant difference between chemotaxis to spleen as compared to dermal media, suggesting that dermal endothelial cells generate additional estrogen-sensitive chemokines.
Discussion Human disease and experimental animal models evidence a sex bias in lymphocyte-mediated autoimmune disease. In our model of autoimmunity, lymphocyte accumulation in the spleen is increased in female estrogen-sufficient mice and decreased in male and Oophorectomized female mice (Yung et al., 1997). Oophorectomized mice supplemented with estrogen but not progesterone demonstrated increased lymphocyte recruitment to the spleen. To determine if estrogen-dependent lymphocyte recruitment was regulated by hormone effects on splenic vascular endothelial cells, we developed in vitro assays of splenic microvascular endothelial cells and exposed these
281
cells to estradiol in a relatively long-term culture (5 days). Our data demonstrating estradiol regulation of endothelial cell ER suggested that endothelial cell function might be directly influenced by sex hormones. Estradiol selectively increased expression of adhesion molecules, specifically VCAM-1, resulting in enhanced adhesion of D10 cells, and regulated chemokine expression, functions that were inhibited by estrogen receptor antagonists. In contrast, estradiol had significantly less effect on dermal endothelial cell adhesion molecule expression, adhesive function, and chemokine expression. Previous studies have generated conflicting data regarding the effect of estradiol on vascular endothelial cell function. Those studies examined endothelium derived from large vessels including human umbilical vein (Aziz and Wakefield, 1996; Caulin-Glaser et al., 1996, 1997; Cid et al., 1994), saphenous vein (Simoncini and Genazzani, 2000; Simoncini et al., 1999, 2000), and aorta (Nathan et al., 1999; Simoncini and Genazzani, 2000; Simoncini et al., 2000). Divergent results among those studies may be a reflection of inherent differences in the species and the vasculature studied, as well as variations in the hormone concentrations and length of time of exposure to estradiol. Estrogen is now well-known to have biphasic dose effects. In the context of atherogenesis, estrogenic effects have been found on adhesion molecules and cytokines (Caulin-Glaser et al., 1996, 1998; Cid et al., 1994; Seli et al., 2002). Endothelial cells derived from tissue microvasculature, however, have a different function than endothelial cells from large vessels. The latter have a primary role in vasoregulation as well as pathological processes such as atherogenesis, whereas in the microvasculature, endothelial cells recruit immune and inflammatory cells to peripheral tissues and to lymphoid tissues during the immune response. Our data indicate that estradiol treatment of endothelial cells increased adhesion of D10 cells. This increase occurred in the absence of additional cytokine stimulation of endothelial cells and was inhibited by ER antagonists. Further, since surface expression of VCAM-1 was increased in estradiol-treated endothelial cells, and estradiol-enhanced adhesion was blocked by antibodies to VCAM-1, increased expression of VCAM-1 was largely responsible for the hormone enhancement of adhesion. The cloned Th2 cells (D10 cells) used in our model expressed CCR1, CCR4, CCR7, CCR8, CXCR2, and CXCR4, and of these, CCR1, CCR4, and to a lesser degree CXCR2 were upregulated by exposure to estradiol. Mouse spleen endothelial cells expressed RANTES, MCP3, and MIP-1a, ligands for CCR1; TARC and MDC, ligands for CCR4; MIP-2, the ligand for CXCR2, as well as MCP-1. Of these, MCP-1, MCP-3, and to a lesser extent MIP-2 were regulated by estradiol. Significant MCP-1 and MCP-3 were generated by the spleen endothelial cells, with an enhanced production of both chemokines, evident at the level of both protein secretion and messenger ribonucleic acid (mRNA) synthesis, in response to physiological
282
H.S. Murphy et al. / Microvascular Research 68 (2004) 273–285
concentrations of estradiol, an effect that was less evident in dermal endothelial cells. While D10 cells do not express CCR2, the primary receptor for MCP-1, D10 cells do express CCR4, which reportedly binds to MCP-1 with high affinity (Power et al., 1995), and they demonstrate increased expression of this latter receptor on exposure to estradiol. CCR4 has also been reported to bind to MCP-1. This chemokine is not only one of the most prominent secreted by EC, but is also affected by estradiol in both in vivo and in vitro studies and has been implicated in autoimmune diseases. In addition to an association with macrophage recruitment to atherosclerotic lesions, MCP-1 has been implicated in the development of rheumatoid arthritis, experimental autoimmune encephalitis, multiple sclerosis, and systemic lupus erythematosus, especially development of lupus nephritis (Cho et al., 2002; Wada et al., 1999). However, in MCP-1 / mice, there is an absence of a T-cell trafficking defect (Gu et al., 2000), suggesting that the influence of MCP-1 on T-cell immunity may be equally important. MCP-1 overexpression is associated with defects in cell-mediated immunity and mice deficient in MCP-1 fail to mount a Th2 response (Gu et al., 2000). By increasing IL-4 production by T cells and decreasing IFNg and IL-12, as well as by inducing suppressive cytokines to indirectly affect the Th1 response, MCP-1 is a critical factor in the development of Th1/Th2 responses (Chensue et al., 1996; Karpus et al., 1997, 1998). In our murine model, MCP-1 may therefore be involved not only in recruitment of T cells but may indirectly affect T cells by stimulating production of other chemokines, attracting other immune cells that might in turn release chemoattractants for T cells, or by stimulating cytokine production to modulate Th1/Th2 responses. Estradiol modulation of MCP-1 has been reported in a variety of cells and tissues including endometrial stromal cells and fibroblasts, human coronary artery EC, rat thoracic aorta, rat ovary, and in serum of patients taking hormone supplements (Akoum et al., 2000; Pervin et al., 1998; Seli et al., 2001) (Koh et al., 2001). Our data, however, suggest that estrogenic effects on the vasculature may differ depending on the tissue or organ. MCP-3 was expressed in relatively lesser quantities than MCP-1, but was equally affected by estradiol treatment, which enhanced protein and RNA expression and was inhibited by tamoxifen and ICI 182,780. MCP-3 is a potent chemokine capable of activating most types of leukocytes, and although generally produced in small quantities, it is thought to play an important role in normal homeostasis as well as pathology (Menten et al., 2001; Wong et al., 2002). MCP-3 binds to CCR1, 2, and 3 and may be involved not only in recruitment of CCR1 expressing T cells, but also of CCR2 expressing dendritic cells from the circulation (Moser and Loetscher, 2001). MCP-3 is expressed in multiple sclerosis lesions, is associated with Th1 but not Th2 pancreatic infiltrates in diabetes, and in the rat ovary MCP-3 is amplified by
estradiol (Bradley et al., 1999; McManus et al., 1998; Menten et al., 2001). Estradiol modulated ERa and ERh expression, and the functional effects of estradiol were inhibited by estrogen receptor antagonists, demonstrating that the estrogen effects in our data were due to hormone acting via the ER. Estrogen exerts its biological action on cells via binding to the specific high-affinity receptors, ERa and ERh. Although both receptors regulate cell function via their ligand and DNA binding domains and have genomic and nongenomic functional effects, their activation may lead to distinct and even contrasting biological activities (Evans et al., 2002; Geraldes et al., 1997; Tremblay et al., 1997). Tissue-specific expression of ER variants has been described in mouse (Kos et al., 2000) and in humans (Flouriot et al., 2000). In cultured ovine endothelial cells, exposure to estradiol for short time periods (2 h) downregulated ERa followed by an increase in expression with continued exposure (6 h) to hormone (Ihionkhan et al., 2002). Persistently elevated ERa levels under long-term estrogen treatment indicate a role for the hormone in maintaining enhanced endothelial ERa expression. In our study, incubation of splenic endothelial cells in estradiol for 5 days suppressed ERh expression while increasing ERa expression. Ihionkhan et al. (2002) reported that in contrast to the upregulation in ERa after long-term estradiol administration, the expression of ERh was downregulated and 7a-thio-phenyl-17h-estradiol reportedly functions as an agonist of ERa and an antagonist of ERh in umbilical vein endothelial cells (Evans et al., 2002). These findings complement our data and are consistent with reports in which ERa and ERh signal in opposite ways when complexed with estradiol (Paech et al., 1997). The nonsteroidal estrogen antagonist tamoxifen has been noted to have both agonistic and antagonistic effects, depending on concentrations, conditions, and cells evaluated. Thus, tamoxifen may have some estrogen-like agonist activity (Cutertre and Smith, 2000) while the steroidal estrogen antagonist ICI-182,780, devoid of agonist activity, specifically downregulates the estrogen receptor (Howell et al., 2000; Jones, 2003). At the concentrations used in our studies, both compounds antagonized the effects of estradiol. The preferential effect of estradiol on spleen endothelial cells as compared to dermal endothelial cells may explain, at least in part, the greater recruitment of D10 cells to spleen compared to dermis in the presence of estrogen in vivo, supporting our hypothesis that estradiol influences progression of disease in our model via effects on the vascular endothelium. In drug-induced lupus, enhanced and selective recruitment of T cells to the spleen in the presence of estrogen (Yung et al., 1997) may be attributable to an enhancement of the concerted action of chemokines and adhesion molecules. Quantitative and qualitative variations in chemokines control the specificity of lymphocyte homing, so that the vascular microenviron-
H.S. Murphy et al. / Microvascular Research 68 (2004) 273–285
ment consists of a chemoattractant gradient, which is tissue-specific as well as triggering the arrest of different cell subsets. Within lymphoid tissue, chemokines direct lymphocytes to the appropriate T- or B-cell areas, thereby facilitating the interactions between dendritic cells/T cells and B/T cells (Luther and Cyster, 2001). Chemokines modulate lymphoid infiltration in some experimental autoimmune diseases including autoimmune thyroiditis, experimental autoimmune encephalomyelitis, rheumatoid arthritis, and diabetes (Cameron et al., 2000; Goulvestre et al., 2002; Kaneko et al., 1999; Karpus and Ransohoff, 1998; Thornton et al., 1999). Elevated chemokines have been found in serum of patients with autoimmune diseases and several chemokines have been noted to be specifically associated with lupus including RANTES, MCP-1, MIP1a, IP-10, IL-8, and GRO-alpha (Kaneko et al., 1999; Noris et al., 1995; Wada et al., 1999). Distinct alterations in chemokine secretion may have importance in the triggering or maintenance of the different pathophysiological mechanisms and clinical aspects characterizing immune disorders. Little is known regarding the role of sex hormones in modulating autoimmune disease-associated chemokines. However, human endometrial leukocyte recruitment increases with administration of oral estrogen and subsequent increase in circulating levels of estrogen (Deloia et al., 2002), and in rat ovary, cyclic variation of hormone is related to inflammatory cell infiltration (Wong et al., 2002), suggesting hormonal regulation of inflammatory cell recruitment. Finally, there are clear differences in splenic lymphocyte recruitment in male and female animals in vivo. Our data suggest that these differences, at least in part, are hormone-mediated, although further studies will address whether this reflects hormone concentrations alone or differences between male and female cells in ER-mediated signaling. There is little data regarding the effect of estrogen on spleen endothelial cell function. Our murine model of autoimmunity implicates estradiol in modulation of the immune response, suggesting that recruitment of immune cells to the spleen is enhanced in the presence of this hormone, contributing to disease severity. The data in this study provide evidence for estrogen modulation of microvascular endothelial cell functions, which may at least in part explain the gender-specific increased lymphocyte homing to spleen. Our observations also define tissuespecific differences in endothelial cell sensitivity to sex hormones leading to the suggestion that in vivo, endogenous or exogenously administered estrogen may have differential, either beneficial or detrimental, effects with implications for therapeutic hormone manipulation.
Acknowledgments This work was supported by the National Institutes of Health grant #AI42753, KO8AR01977, HL61577, and the
283
American Federation for Aging Research Paul Beeson Physician Faculty Scholar Award. B.A. Murphy was supported by The Gina Finzi Memorial Student Fellowship, Lupus Foundation of America, Inc.
References Ahmed, S.A., Verthelyi, D., 1993. Antibodies to cardiolipin in normal C57BL/6J mice: induction by estrogen but not dihydrotestosterone. J. Autoimmun. 6, 265 – 279. Ahmed, S., Penhale, W., et al., 1985. Sex hormones, immune response, and autoimmune diseases. Am. J. Pathol. 121, 531 – 551. Akoum, A., Jolicoeur, C., et al., 2000. Estradiol amplifies interleukin-1induced monocyte chemotactic protein-1 expression by ectopic endometrial cells of women with endometriosis. J. Clin. Endocrinol. Metab. 85 (2), 896 – 904. Aronica, S.M., Dozier, A., et al., 2000. Altered bone marrow cell sensitivity in the lupus-prone NZB/W mouse: regulation of CFU-GM colony formation by estrogen, tamoxifen and thrombopoietin. Lupus 9, 271 – 277. Aziz, K.E., Wakefield, D., 1996. Modulation of endothelial cell expression of ICAM-1, E-selectin, and VCAM-1 by beta-estradiol, progesterone, and dexamethasone. Cell. Immunol. 167 (1), 79 – 85. Bradley, L.M., Asensio, V.C., et al., 1999. Islet-specific Th1, but not Th2, cells secrete multiple chemokines and promote rapid induction of autoimmune diabetes. J. Immunol. 162, 2511 – 2520. Cameron, M.J., Arreaza, G.A., et al., 2000. Differential expression of CC chemokines and CCR5 receptor in the pancreas is associated with progression to type 1 diabetes. J. Immunol. 165, 1102 – 1110. Caulin-Glaser, T., Watson, C.A., et al., 1996. Effects of 17 beta-estradiol on cytokine-induced endothelial cell adhesion molecule expression. J. Clin. Invest. 98 (1), 36 – 42. Caulin-Glaser, T., Garcia-Cardena, G., et al., 1997. 17 beta-estradiol regulation of human endothelial cell basal nitric oxide release, independent of cytosolic Ca2+ mobilization. Circ. Res. 81 (5), 885 – 892. Caulin-Glaser, C., Sessa, W.C., et al., 1998. Modulation of cellular adhesion molecules and endothelial nitric oxide synthase by estrogen. In: Rubanyi, G.M., Kauffman, R. (Eds.), Estrogen and the Vessel Wall. Harwood Academic, Amsterdam. Chensue, S.W., Warmington, K.S., et al., 1996. Role of monocyte chemoattractant protein-1 (MCP-1) in Th1 (Mycobacterial) and Th2 (Schistosomal) antigen-induced granuloma formation. J. Immunol. 157, 4602 – 4608. Cho, C., Cho, M.L., et al., 2002. Anitphospholipid antibodies induce monocyte chemoattractant protein-1 in endothelial cells. J. Immunol. 168, 4209 – 4215. Cid, M.C., Kleinman, H.K., et al., 1994. Estradiol enhances leukocyte binding to tumor necrosis factor (TNF)-stimulated endothelial cells via an increase in TNF-induced adhesion molecules E-selectin, intercellular adhesion molecule type 1, and vascular cell adhesion molecule type 1. J. Clin. Invest. 93 (1), 17 – 25. Couse, J.F., Korach, K.S., 1999. Estrogen receptor null mice: what have we learned and where will they lead us? Endocr. Rev. 20, 358 – 417. Cutertre, M., Smith, C.L., 2000. Molecular mechanisms of selective estrogen receptro modulator (SERM) action. J. Pharmacol. Exp. Ther. 295, 431 – 437. Da Silva, J.A., 1995. Sex hormones, glucocorticoids and autoimmunity: facts and hypotheses. Ann. Rheum. Dis. 54 (1), 6 – 16. Deloia, J.A., Stewart-Akers, A., et al., 2002. Effect of exogenous estrogen on uterine leukocyte recruitment. Fertil. Steril. 77, 548 – 554. Dhaher, Y.Y., Chan, K., et al., 2000. Impaired estrogen priming of progesterone receptors in uterus of MRL/MP-lpr/lpr mice, a model of systemic lupus eruthematosus (SLE). Int. J. Immunopharmacol. 22, 537 – 545.
284
H.S. Murphy et al. / Microvascular Research 68 (2004) 273–285
Evans, M.J., Harris, H.A., et al., 2002. Estrogen receptors a and b have similar activities in multiple endothelial cell pathways. Endocrinology 143, 3785 – 3795. Flouriot, G., Brand, H., et al., 2000. Identification of a new isoform of the human estrogen receptor- (hER-) that is encoded by distinct transcripts and that is able to repress hER-activation function. EMBO J. 19, 4688 – 4700. Fox, H.S., 1995. Sex steroids and the immune system. Ciba Found. Symp. 191, 203 – 211. (discussion 211-7). Friedman, A., Waksman, Y., 1997. Sex hormones and autoimmunity. Isr. J. Med. Sci. 33 (4), 254 – 257. Geraldes, P., Sirois, M.G., et al., 1997. Specific contribution of estrogen receptors on mitogen-activated protein kinase pathways and vascular cell activation. Circ. Res. 93, 399 – 405. Goulvestre, C., Batteux, F., et al., 2002. Chemokines modulate experimental autoimmune thyroiditis through attraction of autoreactive or regulatory T cells. Eur. J. Immunol. 32, 3435 – 3442. Gu, L., Tseng, S., et al., 2000. Control of TH2 polarization by the chemokine monocyte chemoattractant protein-1. Nature 404, 407 – 411. Howell, A., Osbonre, C.K., et al., 2000. ICI 182,780 (FaslodexTM). Development of a novel, bPureQ antiestrogen. Cancer Immunol. Immunother. 89, 817 – 825. Ihionkhan, C.E., Chambliss, K.L., et al., 2002. Estrogen causes dynamic alterations in endothelial estrogen receptor expression. Circ. Res. 91, 814 – 820. Jones, S.E., 2003. Fulvestrant: an estrogen receptor antagonist that downregulates the estrogen receptor. Semin. Oncol. 30, 14 – 20. Kaneko, H., Ogasawara, H., et al., 1999. Circulating levels of betachemokines in systemic lupus erythematosus. J. Rheumatol. 26, 568 – 573. Karpus, W.J., Ransohoff, R.M., 1998. Chemokine regulation of experimental autoimmune encephalitis: temporal and spatial expression patterns govern disease pathogenesis. J. Immunol. 161, 2667 – 2671. Karpus, W.J., Lukacs, N.W., et al., 1997. Differential CC chemokineinduced enhancement of T helper cell cytokine production. J. Immunol. 158, 4129. Karpus, W.J., Kennedy, K.J., et al., 1998. Monocyte chemotactic protein1 regulates oral tolerance induction by inhibition of T helper cell 1-related cytokines. J. Exp. Med. 187, 733 – 741. Koh, K.K., Son, J.W., et al., 2001. Effect o hormone replacement therapy on nitric oxide bioactivity and monocyte chemoattractant protein-1 levels. Int. J. Cardiol. 81, 43 – 50. Kos, M., O’Brien, S., et al., 2000. Tissue-specific expression of multiple mRNA variants of the mouse estrogen receptor gene. FEBS Lett. 477, 15 – 20. Kuiper, G., Carlsson, B., et al., 1997. Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors a and h. Endocrinology 138, 863 – 870. Lahita, R.G., 1993. Sex hormones as immunomodulators of disease. Ann. N. Y. Acad. Sci. 685, 278 – 287. Luther, S.A., Cyster, J.G., 2001. Chemokines as regulators of T cell differentiation. Nat. Immunol. 2, 102 – 107. McManus, C., Berman, J.W., et al., 1998. MCP-1, MCP-2 and MCP-3 expression in multiple sclerosis lesions: an immunohistochemical and in situ hybridization study. J. Neuroimmunol. 86, 20 – 29. Menten, P., Wuyts, A., et al., 2001. Monocyte chemotactic protein-3. Eur. Cytokine Network 12, 554 – 560. Mo, R.R., Chen, J., et al., 2003. T cell chemokine receptor expression in aging. J. Immunol. 170, 895 – 899. Moser, B., Loetscher, P., 2001. Lymphocyte traffic control by chemokines. Nat. Immunol. 2, 123 – 128. Murphy, H.S., Bakopoulos, N., et al., 1998. Heterogeneity of vascular endothelial cells: differences in susceptibility to neutrophil-mediated injury. Microvasc. Res. 56 (3), 203 – 211.
Murphy, H.S., Warner, R.L., et al., 1999. Endothelial cell determinants of susceptibility to neutrophil-mediated killing. Shock 12 (2), 111 – 117. Nathan, L., Pervie, S., et al., 1999. Estradiol inhibits leukocyte adhesion and transendothelial migration in rabbits in vivo: possible mechanism for gender differences in atherosclerosis. Circ. Res. 85, 377 – 385. Noris, M., Bernasconi, S., et al., 1995. Monocyte chemoattractant protein-1 is excreted in excessive amounts in the urine of patients with lupus nephritis. Lab. Invest. 73, 804 – 809. Paech, K., Webb, P., et al., 1997. Differential ligand activation of estrogen receptors ERa and ERh at AP1 sites. Science 277, 1508 – 1510. Pervin, S., Singh, R., et al., 1998. Estradiol suppresses MCP-1 expression In vivo: implications for atherosclerosis. Arterioscler., Thromb., Vasc. Biol. 18 (10), 1575 – 1582. Power, C.A., Meyer, A., et al., 1995. Molecular cloning and functional expression of a novel CC chemokine receptro CDNA fr om a human basophilic cell line. J. Biol. Chem. 270, 19495 – 19500. Russell, K.S., Haynes, M.P., et al., 2000. Estrogen stimulates heat shock protein 90 binding to endothelial nitric oxide synthase in human vascular endothelial cells. Effects on calcium sensitivity and NO release. J. Biol. Chem. 275 (7), 5026 – 5030. Russell, K.S., Haynes, M.P., et al., 2000. Human vascular endothelial cells contain membrane binding sites for estradiol, which mediate rapid intracellular signaling. Proc. Natl. Acad. Sci. U. S. A. 97 (11), 5930 – 5935. Seli, E., Selam, B., et al., 2001. Estradiol regulates monocyte chemotactic protein-1 in human coronary artery smooth muscle cells: a mechanism for its antiatherogenic effect. Menopause 8 (4), 296 – 301. Seli, E., Pehlivan, T., et al., 2002. Estradiol down-regulates MCP-1 expression in human coronary artery endothelial cells. Fertil. Steril. 77, 542 – 547. Shang, X.Z., Chiu, B.-C., 2002. Eosinophil recruitment in type-2 (Schistosomal antigen-induced) hypersensitivity pulmonary granulomas: source and contribution of monocyte chemotactic protein-3 (CCL7). Am. J. Pathol. 161, 257 – 266. Simoncini, T., Genazzani, A.R., 2000. Direct vascular effects of estrogens and selective estrogen receptor modulators. Curr. Opin. Obstet. Gynecol. 12, 181 – 187. Simoncini, T., De Caterina, R., et al., 1999. Selective estrogen receptor modulators: different actions on vascular cell adhesion molecule-1 (VCAM-1) expression in human endothelial cells. J. Clin. Endocrinol. Metab. 84 (2), 815 – 818. Simoncini, T., Maffei, S., et al., 2000. Estrogens and glucocorticoids inhibit endothelial vascular cell adhesion molecule-1 expression by different transcriptional mechanisms. Circ. Res. 87 (1), 19 – 25. Simoncini, T., Genazzani, A.R., et al., 2002. Nongenomic mechanisms of endothelial nitric oxide synthase activation by the selective estrogen receptor modulator raloxifene. Circulation, 1368 – 1373. Talal, N., 1989. Autoimmunity and sex revisited. Clin. Immunol. Immunopathol. 53 (3), 355 – 357. Thornton, S., Duwel, L.E., et al., 1999. Association of the course of collagen-induced arthritis with distinct patterns of cytokine and chemokine messenger RNA expression. Arthritis Rheum., 1109 – 1118. Tremblay, G.B., Tremblay, A., et al., 1997. Cloning, chromosomal localization and functional analysis of the murine estrogen receptor B. Mol. Endocrinol. 11, 353 – 365. Wada, T., Furuichi, K., et al., 1999. MIP-1alpha and MCP-1 contribute to crescents and interstitial lesions in human crescentic glomerulonephritis. Kidney Int. 56, 995 – 1003. Whitacre, C.C., Reingold, S.C., et al., 1999. A gender gap in autoimmunity. Science 283, 1277 – 1278. Wilder, R.L., 1998. Hormones, pregnancy, and autoimmune diseases. Ann. N. Y. Acad. Sci. 840, 45 – 50.
H.S. Murphy et al. / Microvascular Research 68 (2004) 273–285 Wong, K.H., Negishi, H., et al., 2002. Expression, hormonal regulation, and cyclic variation of chemokines in the rat ovary: key determinants of the intraovarian residence of representatives of the white blood cell series. Endocrinology 143 (3), 784 – 791. Yung, R.L., Quddus, J., et al., 1995. Mechanism of drug-induced lupus. I. Cloned Th2 cells modified with DNA methylation inhibitors in vitro cause autoimmunity in vivo. J. Immunol. 154 (6), 3025 – 3035.
285
Yung, R., Williams, R., et al., 1997. Mechanisms of drug-induced lupus. III. Sex-specific differences in T cell homing may explain increased disease severity in female mice. Arthritis Rheum. 40 (7), 1334 – 1343. Zhang, X., Wang, L.Y., et al., 2002. Effects of testosterone and 17-betaestradiol on TNF-alpha-induced E-selectin and VCAM-1 expression in endothelial cell. Analysis of the underlying receptor pathways. Life Sci. 71, 15 – 29.
