Eotaxin/CCL11 Suppresses IL-8/CXCL8 Secretion from Human Dermal Microvascular Endothelial Cells

This information is current as Sara S. Cheng, Nicholas W. Lukacs and Steven L. Kunkel of September 25, 2021. J Immunol 2002; 168:2887-2894; ; doi: 10.4049/jimmunol.168.6.2887 http://www.jimmunol.org/content/168/6/2887 Downloaded from

References This article cites 60 articles, 27 of which you can access for free at: http://www.jimmunol.org/content/168/6/2887.full#ref-list-1

Why The JI? Submit online. http://www.jimmunol.org/

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on September 25, 2021

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2002 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Eotaxin/CCL11 Suppresses IL-8/CXCL8 Secretion from Human Dermal Microvascular Endothelial Cells1

Sara S. Cheng,* Nicholas W. Lukacs,† and Steven L. Kunkel2*† The CC chemokine eotaxin/CCL11 is known to bind to the receptor CCR3 on eosinophils and Th2-type lymphocytes. In this study, we demonstrate that CCR3 is expressed on a subpopulation of primary human dermal microvascular endothelial cells and is up-regulated by TNF-␣. We found that incubation of human dermal microvascular endothelial cells with recombinant eotaxin/ CCL11 suppresses TNF-␣-induced production of the neutrophil-specific chemokine IL-8/CXCL8. The eotaxin/CCL11-suppressive effect on endothelial cells was not seen on IL-1␤-induced IL-8/CXCL8 release. Eotaxin/CCL11 showed no effect on TNF-␣-induced up-regulation of growth-related oncogene-␣ or IFN-␥-inducible protein-10, two other CXC chemokines tested, and did not affect production of the CC chemokines monocyte chemoattractant protein-1/CCL2 and RANTES/CCL5, or the adhesion molecules ICAM-1 and E-selectin. These results suggest that eotaxin/CXCL11 is not effecting a general suppression of TNF-␣R levels or signal transduction. Suppression of IL-8/CXCL8 was abrogated in the presence of anti-CCR3 mAb, pertussis toxin, and wort- Downloaded from mannin, indicating it was mediated by the CCR3 receptor, Gi proteins, and phosphatidylinositol 3-kinase signaling. Eotaxin/ CCL11 decreased steady state levels of IL-8/CXCL8 mRNA in TNF-␣-stimulated cells, an effect mediated in part by an accel- eration of IL-8 mRNA decay. Eotaxin/CCL11 may down-regulate production of the neutrophil chemoattractant IL-8/CXCL8 by endothelial cells in vivo, acting as a negative regulator of neutrophil recruitment. This may play an important biological role in the prevention of overzealous inflammatory responses, aiding in the resolution of acute inflammation or transition from neutro- philic to mononuclear/eosinophilic inflammation. The Journal of Immunology, 2002, 168: 2887–2894. http://www.jimmunol.org/

hemokines are a large family of small, basic peptides that eases possessing an eosinophilic component, such as allergic have mainly been characterized as leukocyte chemoat- asthma (20, 21), chronic sinusitis (22), and allergic rhinitis (23), C tractants. They are divided into four subfamilies based on and is thought to be a key player in the pathogenesis of these the number and spacing of conserved cysteines within the amino conditions. In addition, eotaxin/CCL11 mRNA is up-regulated in terminus of each peptide, with each subfamily having a limited the lesions of patients with inflammatory bowel disease (24) and specificity for different leukocyte subsets. Chemokines from the within lymphomas from patients with Hodgkin’s disease (25), sug- CXC family, in which the first two conserved cysteines are sepa- gesting that eotaxin/CCL11 may play a role in these diseases as well. rated by a nonconserved amino acid, are chemoattractants for poly- Early reports characterizing the tissue expression patterns of by guest on September 25, 2021 morphonuclear phagocytes, some T lymphocyte subsets, and NK eotaxin/CCL11 in human, guinea pig, and mouse tissues demon- cells (1). Members of the CC family, in which the cysteines are strate that eotaxin/CCL11 mRNA is constitutively expressed in a adjacent to one another, have a broader spectrum of action that wide array of tissues, including the gut mucosa, lung, heart, testes, includes monocytes, eosinophils, basophils, NK cells, and T lym- and endometrium (15, 26Ð28). Relatively few chemokines are ex- phocytes (2). Lymphotactin/XCL1 and fractalkine/CX3CL1 are pressed in a constitutive fashion. The constitutive expression of the sole members of the C and CX3C families, respectively. Lym- eotaxin/CCL11 in a wide variety of tissues, often in the absence of photactin is chemotactic for T lymphocytes (3), while fractalkine a significant eosinophil infiltrate, suggests that it may play a role in acts on both lymphoid cells and neutrophils (4). While chemokines maintaining homeostasis in these tissues. Eotaxin/CCL11 exerts its are widely recognized as important chemotactic factors, some of chemotactic activity primarily through the chemokine receptor these proteins have other varied functions, including immunoregu- CCR3, a seven-transmembrane receptor coupled to heterotrimeric lation (5Ð7), lymphocyte activation (8, 9), embryonic development G proteins. The CCR3 has been found on human brain endothelial (10, 11), and angiogenesis (12Ð14). cells (29, 30), and recent studies indicate it may be involved in Eotaxin/CCL11 is a member of the CC chemokine family that angiogenesis (31). To further study CCR3 function in endothelial has potent chemotactic activity for eosinophils (15, 16), basophils cells in vitro, we investigated whether primary cultures of endothelial (17), mast cells (18), and Th2-type lymphocytes (19). Eotaxin/ cells express CCR3, and we uncovered a novel regulatory role for eotaxin and CCR3 on endothelial cell chemokine production. CCL11 protein is up-regulated in a variety of inflammatory dis- Materials and Methods *Graduate Program in Cellular and Molecular Biology and †Department of Pathology, Cytokines and other reagents University of Michigan Medical Center, Ann Arbor, MI 48109 Recombinant human IL-1␤, TNF-␣, eotaxin/CCL11, RANTES/CCL5, and Received for publication September 21, 2001. Accepted for publication January monocyte chemoattractant protein-1 (MCP-1)3/CCL2 were purchased from 7, 2002. R&D Systems (Minneapolis, MN). Abs against E-selectin, ICAM-1, IL- The costs of publication of this article were defrayed in part by the payment of page 8/CXCL8, MCP-1/CCL2, RANTES/CCL5, and CCR3 were purchased charges. This article must therefore be hereby marked advertisement in accordance from R&D Systems. Actinomycin D, pertussis toxin, and wortmannin with 18 U.S.C. Section 1734 solely to indicate this fact. were purchased from Sigma-Aldrich (St. Louis, MO) and stored at a 1 This work was supported by National Institutes of Health Grants HL35276 (to S.L.K.), NIGMS T32GM0786 (to S.S.C.), and GM07315 (to S.S.C.). 2 Address correspondence and reprint requests to Dr. Steven L. Kunkel, Department 3 Abbreviations used in this paper: MCP-1, monocyte chemoattractant protein-1; of Pathology, University of Michigan Medical Center, 1301 Catherine Street, Ann Gro␣, growth-related oncogene-␣; HDMEC, human dermal microvascular endothe- Arbor, MI 48109-0602. E-mail address: [email protected] lial cell; IP-10, IFN-␥-inducible protein-10; PI3K, phosphatidylinositol 3-kinase.

Copyright © 2002 by The American Association of Immunologists 0022-1767/02/$02.00 2888 EOTAXIN REGULATES IL-8 PRODUCTION IN ECs

concentration of 2.5 mg/ml, 100 ng/␮l, and 10 mM, respectively, in ICAM-1 expression occurred after 24 h. At appropriate time points, culture DMSO at Ϫ20¡C. medium was removed and the adherent monolayers were washed three times with 200 ␮l PBS containing Ca2ϩ/Mg2ϩ and 0.1% BSA (PBS/BSA). Cell culture A total of 100 ␮l saturating concentration of biotinylated primary Ab Human dermal microvascular endothelial cells (HDMEC) were obtained in against E-selectin, ICAM-1, or goat IgG (diluted in endothelial cell growth medium) was added. Plates were incubated for 45 min at 37¡C. Primary Ab single donor ampules from Clonetics. Cells were passaged by trypsiniza- ␮ ␮ tion and seeding at 1 ϫ 105 cells/ml in endothelial cell growth medium-2 was removed by washing twice with 200 l PBS/BSA. A total of 100 l (Clonetics, Walkersville, MD) medium on plates or flasks (Costar, Corn- 1/5000 dilution of streptavidin-HRP (BD PharMingen) diluted in PBS/ BSA was added, and plates were incubated for 30 min at 37¡C. Plates were ing, NY) coated with 2% gelatin (Sigma-Aldrich). Cells were used at pas- ␮ ␮ sages 2Ð6 for all experiments. washed twice with 200 l PBS/BSA, and 100 l o-phenylenediamine sub- strate (DAKO, Glostup, Denmark) was added to wells. Plates were allowed RNase protection assay to develop at room temperature for 2Ð20 min. Fifty microliters of 3 M sulfuric acid were added to stop reaction, and chromophore development Total RNA was isolated using TRIzol reagent (Life Technologies, Rock- was determined by measuring OD490 using microplate reader. OD readings ville, MD), according to manufacturer’s instructions, and used in the stan- from samples stained with goat IgG were consistently indistinguishable dard BD PharMingen (San Diego, CA) RNase protection protocol, as fol- from readings taken from unstained samples, indicating no nonspecific lows. The multiprobe template sets hCR5 (containing DNA templates for binding of the Ab was occurring. CCR1, CCR3, CCR4, CCR5, CCR8, CCR2aϩb, CCR2a, CCR2b, L32, and GAPDH) and hCR6 (containing DNA templates for CXCR1, CXCR2, RNA isolation and cDNA synthesis CXCR3, CXCR4, CXCR5, CX3CR1, L32, and GAPDH) were purchased from BD PharMingen. This template set was used to synthesize For analysis of chemokine mRNA, total RNA was isolated using TRIzol [␣-32P]UTP (Amersham, Little Chalfont, Buckinghamshire, U.K.)-labeled reagent (Life Technologies), according to manufacturer’s instructions. To-

probes in the presence of a GACU pool using a T7 RNA polymerase. tal RNA was used to synthesize first-strand cDNA using standard tech- Downloaded from Ϫ Probes were hybridized overnight with 10Ð15 ␮g target RNA, followed by niques. cDNA was stored at 20¡C until RT-PCR analysis. RNase digestion and proteinase K treatment. Samples were chloroform extracted, ethanol precipitated in the presence of ammonium acetate, and Real-time RT-PCR analysis loaded on an acrylamide-urea sequencing gel made in TBE (0.53 M Tris- Relative levels of chemokine mRNA were analyzed using semiquantitative borate/EDTA) buffer. After electrophoresis, the gel was adsorbed to filter pa- real-time PCR analysis. Predeveloped probes for detection of human IL- per, dried under vacuum, and exposed to film (X-OMAT; Kodak, Rochester, 8/CXCL8 and the housekeeping gene cyclophilin were purchased from PE NY) with intensifying screen at Ϫ70¡C. Alternatively, the dried gel blot was Ϫ Applied Biosystems (Foster City, CA) and stored at 70¡C. Each reaction http://www.jimmunol.org/ exposed to a phosphorscreen for phosphor imagery analysis using the Quantity mixture contained primer/probe sets for detecting the amplification of the One software application (Bio-Rad, Richmond, CA). The intensity of each target gene of interest as well as the internal control cyclophilin, to control band was normalized to the intensity of the housekeeping gene L32. for intersample variation in the amount of cDNA added. Each reaction ␮ Flow cytometry analysis of chemokine receptor expression mixture was prepared by combining the following components: 12.5 l TaqMan 2ϫ PCR Mix (PE Applied Biosystems), 1 ␮l target probe, 1 ␮l Single cell suspensions were prepared by harvesting adherent cells with control probe, 0.5 ␮l prepared cDNA, and 10 ␮l sterile water. Samples cold EDTA (20 mM), then pelleting and washing three times with flow were amplified on a PRISM 7700 Sequence Detection System (PE Applied buffer (Dulbecco’s PBS plus 0.5% BSA). Cells were resuspended in 25 ␮l Biosystems) under the following conditions: 2 min at 50¡C, 10 min at flow buffer, mixed with 10 ␮l FITC-conjugated rat anti-human CCR3 Ab 95¡C, followed by 40 cycles of 15 s at 95¡C, 1 min at 60¡C. The amount (50 ␮g/ml; R&D Systems) or appropriate isotype control Ab, and incu- of a particular cDNA species within a sample is related logarithmically to by guest on September 25, 2021 bated at 4¡C for 30 min. Cells were washed three times in flow buffer, fixed the ct number, which indicates the first cycle number at which amplification for 10 min in 2% paraformaldehyde, and stored in flow buffer at 4¡C until is detectable. Calculation of the relative amount of each cDNA species was FACS analysis. Fluorescent cytometric analysis was performed on a done according to standard manufacturer’s protocol. In brief, the amplifi- Coulter (Palo Alto, CA) EPICs XL cytometer, with at least 5000 cells cation of target genes in stimulated cells was calculated by first normaliz- being counted per sample. ing to the amplification of cyclophilin and then expressing these normal- ized values as a fold increase over the value obtained with unstimulated ELISA analysis of chemokine protein in cell supernatants control cells. Human growth-related oncogene-␣ (Gro␣)/CXCL1, IFN-␥-inducible pro- Statistical analysis tein-10 (IP-10)/CXCL10, and stromal cell-derived factor-1␣/CXCL12 pro- tein levels were measured using Quantikine Immunoassay kits (R&D Sys- Results are expressed as mean Ϯ SEM of n experiments, unless otherwise tems), according to manufacturer’s instructions. Human IL-8/CXCL8, stated. One-way ANOVA followed by a Bonferroni posttest between se- MCP-1/CCL2, and RANTES/CCL5 protein levels were determined in lected samples was performed using GraphPad Prism version 3.0a (Graph- 50-␮l samples from cell supernatants using a standard sandwich ELISA Pad Software, San Diego, CA) for Macintosh (Apple Computer, Cupertino, technique previously described in detail (5). In brief, Nunc-immuno ELISA CA). Results were deemed significant if p Ͻ 0.05. plates (Fisher Scientific, Springfield, NJ) were coated with the appropriate polyclonal capture Ab (R&D Systems), washed, and blocked with 2% Results BSA/PBS. Plates were then washed and cell supernatants were added. Af- CCR3 is expressed on the surface of HDMEC ter incubation at 37¡C for 1 h, plates were washed and the appropriate biotinylated polyclonal detection Ab was added. After a 45-min incubation, To assess whether unstimulated HDMEC express CCR3 pro- plates were washed and streptavidin-peroxidase (1/5000 dilution; Bio-Rad) tein, we performed flow cytometry analysis on single cell sus- was added for 30 min. Plates were thoroughly washed, a chromagen sub- pensions of HDMEC from five different donors that had been strate solution was added, and OD readings at 490 nm were obtained using an ELISA plate scanner. Recombinant human chemokines were used to grown to 70Ð90% confluence (Fig. 1). A significant subpopu- generate the standard curves from which the concentrations present in the lation of HDMEC showed positivity when stained with a mAb samples were derived. The limit of ELISA detection for each cytokine was against CCR3 (Fig. 1). When results from five donors were Ͼ consistently 50 pg/ml. Each ELISA was screened to ensure the specificity pooled, 35.6 Ϯ 3.6% stained positively with an anti-CCR3 Ab of each Ab used. Due to donor variability in basal and stimulated levels of Ϯ Ͻ chemokine secretion, some ELISA results are expressed as percentage of vs 7.4 2% with control Ab ( p 0.001), indicating these the maximal amount produced upon stimulation with TNF-␣ (5 ng/ml) (see results were statistically significant. figures). To study cytokine regulation of surface levels of CCR3, HDMEC ELISA analysis of adhesion molecule on cell surfaces monolayers were grown to confluence and stimulated with the proinflammatory cytokines TNF-␣ or IL-1␤ (5 ng/ml; 24 h). CCR3 Adhesion molecule expression on the surface of HDMEC was assayed protein expression was then analyzed by flow cytometry (Fig. 2A). using a modified ELISA protocol. Briefly, confluent cells were stimulated ϭ ␣ with cytokines for 4 (E-selectin) or 24 (ICAM-1) h. These time points were In all donors tested (n 6), TNF- stimulation increased the num- chosen because initial experiments in the lab established that peak expres- ber of CCR3-positive cells (Fig. 2A). However, parallel cultures sion of E-selectin occurred at 4 h after cytokine stimulation, while peak stimulated with IL-1␤ showed no up-regulation in surface levels of The Journal of Immunology 2889

FIGURE 1. CCR3 is expressed on a subpopulation of proliferating HDMEC. HDMEC were cultured to 60Ð80% confluence, stained with a mAb specific for human CCR3 or an isotype-matched control Ab, and analyzed by flow cytometry. Representative histogram showing data ob- tained from cells stained with anti-CCR3 (open histogram) or control (filled histogram) mAbs. Results are representative of experiments done with five different HDMEC donors.

CCR3, with donors showing no change or decreases in CCR3 sur- Downloaded from face levels (Fig. 2A). The basis of this variability is unclear, as cultures were all treated at the same growth stage and with freshly prepared aliquots of IL-1␤. These data demonstrate that TNF-␣ stimulation consistently increases the frequency of CCR3-positive cells, while IL-1␤ does not.

To determine whether cytokine-induced changes in CCR3 pro- http://www.jimmunol.org/ tein were mirrored by similar changes in mRNA expression, we performed RNase protection analysis on RNA samples taken from HDMEC stimulated with IL-1␤, TNF-␣, or medium alone (5 ng/ ml; 8 h) (Fig. 2B). When RNA probes for CC chemokine receptors were used, CCR2, CCR3, CCR4, and CCR8 mRNA were present in unstimulated HDMEC, while CCR1 and CCR5 mRNA were undetectable. Unexpectedly, cytokine stimulation did not affect the expression patterns of any of these chemokine receptor mRNAs. CXCR4 was the only CXC receptor message detected in unstimu- by guest on September 25, 2021 lated HDMEC, and as previously demonstrated, was down-regu- lated by both IL-1␤ and TNF-␣ (32) (gel not shown). Quantitative results were obtained by normalizing the band intensities of CCR3 FIGURE 2. A, TNF-␣ increases CCR3 surface expression on HDMEC. or CXCR4 to that of the housekeeping gene L32; pooled results Confluent HDMEC were stimulated with TNF-␣ or IL-1␤ (5 ng/ml; 24 h), from three donors are depicted in Fig. 2C. stained with anti-CCR3 or control Ab, and analyzed by flow cytometry. Representative histograms showing data obtained from cells stained with Recombinant eotaxin/CCL11 reduces IL-8/CXCL8 release anti-CCR3 (open histogram) or control (filled histogram) mAbs. Results ␣ induced by TNF- are representative of experiments done with five different HDMEC donors. After we had established the presence of CCR3 protein on the B, Cytokine stimulation has no effect on the expression of CC chemokine ␮ surface of HDMEC, we then investigated whether interaction of receptors in HDMEC. Total RNA (10 g) was prepared from HDMEC ␣ ␤ CCR3 with its major ligand, eotaxin, affected parameters of endo- stimulated with TNF- or IL-1 (5 ng/ml; 8 h) and analyzed by RNase protection assay using the template set hCR5 (see Materials and Methods). thelial cell activation. Because both TNF-␣ and IL-1␤ are potent Data shown are representative of three individual experiments using RNA stimulators of chemokine production by endothelial cells, we preparations from different donors. C, Quantitative analysis of CCR3 and chose to test whether eotaxin/CCL11 was involved in modulating CXCR4 mRNA expression as measured by RNase protection assay, ex- production of a CXC family chemokine, IL-8/CXCL8. As has been pressed as the ratio of each target band intensity to L32 band intensity. Data shown previously, we found that 48-h stimulation of HDMEC with shown are the mean Ϯ SEM of data obtained from three different donors. .p Ͻ 0.01 ,ء -TNF-␣ and IL-1␤ (5 ng/ml) strongly up-regulated IL-8/CXCL8 pro duction, as measured by ELISA analysis of cell-free supernatants (Fig. 3). Surprisingly, the addition of recombinant eotaxin/CCL11 (20 ng/ml) during the period of cytokine stimulation partially sup- ␣ pressed TNF-␣ induction of IL-8/CXCL8 production (Fig. 3). Recombinant eotaxin/CCL11 does not suppress TNF- induction Eotaxin/CCL11 suppression of TNF-␣-induced IL-8/CXCL8 was of other chemokines and adhesion molecules concentration dependent, with a suppressive effect being found at Endothelial cells elaborate a number of other CXC-type chemo- eotaxin/CCL11 concentrations from 20 to 200 ng/ml (Fig. 3A). kines in addition to IL-8/CXCL8. When cell-free supernatants Eotaxin had no effect on IL-1␤-induced chemokine levels (Fig. from the same experiment were assayed in parallel for IL-8/ 3A), which is consistent with the variable effects of IL-1␤ on CXCL8 content as well as two other CXC chemokines, Gro␣/ CCR3 expression in these cells. The magnitude of suppression CXCL1 and IP-10/CXCL10, eotaxin suppressed the TNF-␣-in- varied with the concentration of TNF-␣ stimulation, with the great- duced secretion of IL-8/CXCL8, but not the latter two chemokines est suppression being obtained at a TNF-␣ concentration of 0.5 (Fig. 4). An additional CXC chemokine, stromal cell-derived fac- ng/ml (Fig. 3B). tor-1␣, was not detectable under any of the stimulatory conditions 2890 EOTAXIN REGULATES IL-8 PRODUCTION IN ECs

glutamine-leucine-arginine (ELR)-positive CXC chemokines IL- 8/CXCL8 rather than a broad effect on TNF-␣R levels or signal- ing. We also used an MTT assay to demonstrate that eotaxin had no effect on cellular viability (data not shown), a result consistent with the lack of effect on molecules other than IL-8/CXCL8.

Anti-CCR3 Ab, wortmannin, and pertussis toxin block the effect of eotaxin/CCL11 on IL-8/CXCL8 secretion Eotaxin/CCL11 suppression of IL-8/CXCL8 production was blocked by a neutralizing mAb against CCR3, indicating that the Downloaded from

FIGURE 3. Recombinant eotaxin suppresses IL-8/CXCL8 release from http://www.jimmunol.org/ TNF-␣-stimulated HDMEC. Significant variation was found in basal (0.02Ð3.98 ng/ml) and TNF-␣-elicited (2.16Ð84.14 ng/ml) levels of IL-8 from cells from different donors. To pool results from different experi- ments, the data were normalized by setting the IL-8 level induced by 5 ng/ml TNF-␣ in each experiment as 1 and then calculating the fraction with respect to 1 for each condition within that experiment (Fraction of maxi- mum). These normalized values from each experiment were then pooled and shown as the mean Ϯ SEM of four to six experiments done in tripli- cate. A, Cells were stimulated with TNF-␣ or IL-1␤ (5 ng/ml) with or without eotaxin (2Ð200 ng/ml) for 48 h and analyzed for immunoreactive by guest on September 25, 2021 IL-8/CXCL8 content by ELISA. B, Cells were stimulated with TNF-␣ (0.05Ð5 ng/ml) with or without eotaxin (20 ng/ml) for 48 h. Cell-free su- pernatants were collected and analyzed for immunoreactive IL-8/CXCL8 .p Ͻ 0.05 ,ء .content by ELISA tested. TNF-␣ is also a strong inducer of the CC chemokines MCP- 1/CCL2 and RANTES/CCL5, as well as the adhesion molecules E-selectin and ICAM-1. Recombinant eotaxin/CCL11 had no ef- fect on TNF-␣ up-regulation of these molecules (Fig. 5), suggest- ing that eotaxin/CCL11 is exerting a very specific effect on the

FIGURE 5. Eotaxin/CCL11 does not affect TNF-␣-induced MCP-1/ CCL2 and RANTES/CCL5 release, or E-selectin and ICAM-1 expression. A, Cells were stimulated with TNF-␣ (0.05Ð5 ng/ml) with or without eotaxin (20 ng/ml) for 48 h. Cell-free supernatants were collected and analyzed for immunoreactive IL-8/CXCL8, MCP-1/CCL2, and RANTES/ CCL5 content by ELISA. Results shown are the means Ϯ SE of data from triplicate wells of one experiment and are representative of data obtained in two separate experiments. For each chemokine, the results were normal- ized by setting the level induced by 5 ng/ml TNF-␣ as 100% and calcu- lating all other values in the experiment as a fraction of this value. B, Cells FIGURE 4. Eotaxin/CCL11 does not affect TNF-␣-induced Gro␣/ were stimulated with TNF-␣ (5 ng/ml) with or without eotaxin (2Ð200 CXCL1 or IP-10/CXCL10. Cells were stimulated with TNF-␣ or IL-1␤ (5 ng/ml) for 4 h and analyzed for cell surface expression of E-selectin using ng/ml) with or without eotaxin (2Ð200 ng/ml) for 48 h and analyzed in a modified ELISA format. C, Cells were stimulated with TNF-␣ (5 ng/ml) parallel for immunoreactive IL-8/CXCL8, Gro␣/CXCL1, or IP-10/ with or without eotaxin (2Ð200 ng/ml) for 24 h and analyzed for cell sur- .p Ͻ 0.05 ,ء .CXCL10 content by ELISA. face expression of ICAM-1 using a modified ELISA format The Journal of Immunology 2891 suppressive effect of eotaxin/CCL11 is mediated via binding to CCR3 (Fig. 6A). Eotaxin suppression of IL-8/CXCL8 production was also abrogated in the presence of pertussis toxin and wort- mannin (Fig. 6B), indicating a dependence on Gi proteins and phosphatidylinositol 3-kinase (PI3K). This is consistent with pre- vious reports demonstrating these mediators are important in CCR3 signaling in eosinophils (33).

Recombinant eotaxin/CCL11 reduces steady state levels of IL-8/ CXCL8 mRNA via effects on mRNA stability Because we found that eotaxin/CCL11 was exerting a specific ef- fect on IL-8/CXCL8 protein production, we investigated whether it was reducing steady state levels of IL-8 mRNA. Confluent cells were stimulated with TNF-␣ (5 ng/ml) with or without eotaxin FIGURE 7. Eotaxin decreases steady-state levels of TNF-␣-induced (20Ð200 ng/ml), and total RNA was harvested at 48 h. This RNA IL-8 mRNA. Cells were stimulated with TNF-␣ (5 ng/ml) with or without was used to synthesize cDNA, which was analyzed for IL-8 ex- eotaxin (20Ð200 ng/ml) for 48 h, at which time total RNA extracts were pression using real-time PCR (TaqMan) analysis. IL-8/CXCL8 prepared and cDNA was generated. Steady state levels of IL-8 mRNA were mRNA was elevated in TNF-␣-stimulated samples ϳ5.5-fold over by real-time RT-PCR analysis. Results are expressed as fold increase in control levels (Fig. 7). This elevation was suppressed in samples IL-8 mRNA levels over cells grown in medium alone. Please refer to Ma- Downloaded from terials and Methods for further details on the types of calculations in- stimulated with TNF-␣ and eotaxin, with a concentration-depen- volved. Data shown are from one experiment done in triplicate and are dent effect being found at eotaxin concentrations of 20 and 200 representative of three experiments done with cells from different donors. ng/ml. Eotaxin did not affect the steady state levels of IL-1␤-in- duced IL-8 mRNA (data not shown), which is consistent with its lack of effect on IL-1␤-induced IL-8 release. IL-8/CXCL8 production is controlled at the level of transcrip- 8/CXCL8 release was likely to be posttranscriptional. To test http://www.jimmunol.org/ tion and mRNA stability (34, 35). Because eotaxin did not affect whether eotaxin/CCL11 affects mRNA stability, we stimulated cytokine induction of adhesion molecules or other chemokines, we HDMEC with TNF-␣ (5 ng/ml) with or without eotaxin/CCL11 reasoned that the mechanism of eotaxin/CCL11 suppression of IL- (100 ng/ml) for 24 h. At this time, actinomycin D (5 ␮g/ml) was added to inhibit transcription. Total RNA was harvested at appro- priate time points after actinomycin D addition, cDNA was syn- thesized, and samples were assayed for steady state levels of IL- 8/CXCL8 mRNA using real-time PCR analysis. Samples ␣

stimulated with TNF- (5 ng/ml) alone showed steady degradation by guest on September 25, 2021 of IL-8 mRNA, with ϳ40% remaining at 4 h after actinomycin D addition (Fig. 8). The rate of degradation was significantly in- creased in samples stimulated with TNF-␣ and eotaxin (100 ng/

FIGURE 6. Eotaxin suppression of IL-8 release is blocked by neutral- ization of CCR3, pertussis toxin, and wortmannin. After the indicated treat- ments, cell-free supernatants were collected and analyzed for immunore- active IL-8/CXCL8 content by ELISA. Results shown are the mean Ϯ FIGURE 8. Eotaxin decreases TNF-␣-induced IL-8 mRNA stability. SEM of six separate experiments using cells from two different donors. A, Cells were stimulated with TNF-␣ (5 ng/ml) with or without eotaxin (100 Cells were stimulated with TNF-␣ (0.5 ng/ml) with or without eotaxin (20 ng/ml) for 24 h, at which time actinomycin D (5 ␮g/ml) was added to stop ng/ml) for 48 h in the presence of an anti-CCR3 mAb (10 ␮g/ml) or control de novo transcription. Total RNA extracts were collected at appropriate Ab. B, Cells were stimulated with TNF-␣ (0.5 ng/ml) with or without time points after actinomycin D addition. The amount of IL-8 mRNA re- eotaxin (20 ng/ml) for 48 h in the presence of pertussis toxin (1 ng/ml), maining was assessed by real-time RT-PCR analysis. Data are expressed as p Ͻ 0.05 (significant sup- percentage of mRNA remaining after actinomycin D addition. The results ,ء .wortmannin (100 ␮M), or DMSO control pression of IL-8 release compared with TNF-␣-induced release without are expressed as mean Ϯ SEM of three separate experiments using cells .p Ͻ 0.01 ,ء .eotaxin). from different donors 2892 EOTAXIN REGULATES IL-8 PRODUCTION IN ECs ml), with levels dropping to ϳ40% within 30 min after actinomy- neutralizing Ab against CCR3, indicating that eotaxin/CCL11 was cin D addition. IL-8 mRNA levels did not drop significantly below exerting its effects via CCR3. In addition, treatment with pertussis 40% under either treatment condition. toxin and wortmannin also blocked the suppressive effect of

eotaxin/CCL11, indicating the involvement of Gi proteins and Discussion PI3K. Previous reports have indicated the activity of these medi- Endothelial cells act as physical and functional barriers between ators in CCR3-mediated chemotaxis and respiratory burst in eo- the vascular space and the tissue. Rather than being passive by- sinophils (33). This study now demonstrates that a novel link exists standers to leukocyte transmigration, endothelial cells are impor- between a Gi-, PI3K-dependent pathway and chemokine produc- tant participants in the development and maintenance of inflam- tion in HDMEC. mation (36). Upon activation by soluble mediators or physical The mechanism by which eotaxin/CCL11 suppresses production trauma, they express a number of leukocyte chemoattractants and of IL-8/CXCL8 appears to be via an eotaxin/CCL11-induced ac- adhesion molecules that work together to cause leukocyte adhesion celeration of IL-8/CXCL8 mRNA decay. Previous reports have and diapadesis into tissues (37). While endothelial cell activation indicated that cellular control of IL-8/CXCL8 production is largely is part of a typical immunoinflammatory response, inappropriate or regulated at the posttranscriptional level (35, 53, 54). Experimental overzealous activation can lead to tissue damage by infiltrating evidence suggests that binding of certain proteins to AU-rich re- leukocytes. Thus, a better understanding of the factors regulating gions in the 3Ј-untranslated regions of chemokine mRNAs mod- endothelial cell activation may provide insights into possible an- ulates the stability of these mRNAs (55). At present, it is not tiinflammatory therapies. known, however, whether eotaxin regulates the expression of any Downloaded from A number of chemokines are active on endothelial cells (13, 14, known AU-binding factors. 32, 38Ð40). Several groups have documented expression of CXC- Disparate results between TNF-␣- and IL-1␤-stimulated cells type receptors by these cells (12, 32, 39, 41, 42). CXCR1, CXCR2, were observed in that eotaxin/CCL11 did not affect the production and CXCR4 function as angiogenic receptors (32, 39), while of IL-8/CXCL8 by IL-1␤-stimulated endothelial cells nor the deg- CXCR3 has an angiostatic function (43Ð45). In addition, the CC- radation rate of IL-1␤-stimulated IL-8/CXCL8 mRNA. We hy- type receptors CCR2 (46) and CCR3 (31) have also been docu- pothesize that this is related to the specific mechanism of eotaxin mented on endothelial cells. The expression of CXCR4 is down- action. Eotaxin most likely activates an mRNA-destabilizing path- http://www.jimmunol.org/ regulated by TNF-␣ (32) and up-regulated by basic fibroblast way in endothelial cells. It has already been shown that this path- factor and vascular endothelial growth factor (42), while CCR2 way is countered in IL-1␤-stimulated cells, but not in TNF-␣- is up-regulated by IL-1␤ (47), suggesting that regulation of stimulated cells, in that IL-1␤, but not TNF-␣, activates a signaling these endothelial cell chemokine receptors by cytokines or pathway that leads to protein-dependent stabilization of an IL-8 growth factors may be important in various disease or physio- homolog, the murine chemokine KC (49). The rate of mRNA deg- logical processes. radation is also not the main determinant of the basal “setpoint” of To study the regulation and function of CCR3 on endothelial IL-8 release in unstimulated endothelial cells, which is controlled cells in vitro, we first established the expression of this receptor on

at the level of IL-8 basal promotor activity (56). This mechanism by guest on September 25, 2021 HDMEC. We found that a subpopulation of proliferating HDMEC of eotaxin action would explain its lack of effect on IL-8 secretion has significant levels of surface-expressed CCR3, as assayed by from IL-1␤-stimulated and resting endothelial cells. flow cytometry. CCR3 was only expressed on a subpopulation of Other factors with antiinflammatory effects on vascular inflam- proliferating HDMEC. This finding differs slightly from results ␤ obtained by Salcedo et al. (31), which demonstrate low levels of matory responses include TGF- , IL-1R antagonist, and IL-10 CCR3 expression across the entire cell population in human mi- (57). The mechanism of action of each of these antiinflammatory crovascular endothelial cells. The basis for these differences is un- cytokines results in inhibition of a broad array of activation pa- clear, but is likely to involve differences in source tissue, culture rameters, including multiple chemokines and adhesion molecules. conditions, or harvesting techniques. Interestingly, a recent report Unlike these cytokines, eotaxin/CCL11 specifically inhibits the shows that the CXC-type chemokine receptor CXCR3 exhibits a production of the CXC chemokine IL-8/CXCL8 but does not affect cell cycle-specific expression on endothelial cells, which explains expression of other chemokines or adhesion molecules. The bio- why its expression is detectable on a small subpopulation of cells logical significance of this specificity is unknown; however, we from proliferating cultures (43). A similar cell cycle-restricted ex- speculate that eotaxin may play a role in specifically dampening pression pattern may exist for CCR3 on HDMEC. neutrophil influx via down-regulation of IL-8/CXCL8 in vivo. IL- The level of surface-expressed CCR3 protein on HDMEC could 8/CXCL8 is generally regarded as the most robust and specific be up-regulated by TNF-␣, a cytokine that has potent proinflam- neutrophil chemoattractant known. In contrast, IP-10/CXCL10 is matory effects on these cells. Surprisingly, IL-1␤ did not have known as a lymphocyte chemoattractant (58), while MCP-1/CCL2 consistent effects on surface levels of CCR3. While TNF-␣ and and RANTES/CCL5 attract primarily mononuclear cells. There- IL-1␤ are generally quite similar in the signaling molecules they fore, a specific reduction in IL-8/CXCL8 might affect neutrophil use (47, 48), these cytokines also activate nonoverlapping signal- influx without affecting the trafficking of other leukocyte subtypes. ing pathways (49Ð51). TNF-␣ up-regulation of CCR3 protein ex- It is interesting to speculate whether eotaxin may be a key effector pression is probably dependent on a TNF-␣-specific pathway that of the switch from granulocytic to mononuclear inflammation ob- is not activated by IL-1␤. Because TNF-␣ did not affect CCR3 served in vivo during the time course of acute infection. mRNA levels, up-regulation of CCR3 surface protein by TNF-␣ is Surprisingly, eotaxin/CCL11 did not suppress release of Gro␣/ most likely dependent on posttranscriptional mechanisms such as CXCL1, another CXC chemokine that acts as a neutrophil che- decreased receptor internalization or altered receptor trafficking. moattractant in vitro. However, a recent report indicates that acti- Posttranscriptional control of chemokine receptor levels has al- vated neutrophils do not respond to Gro␣ (59), although they give ready been demonstrated for other receptor types (52). robust responses to IL-8/CXCL8. Therefore, IL-8/CXCL8 may be We next demonstrated that eotaxin/CCL11 suppressed the pro- more relevant to neutrophil influx in vivo. duction of IL-8/CXCL8 by TNF-␣-stimulated endothelial cells. A recent report supports a regulatory role for endogenous Suppression of IL-8/CXCL8 was abrogated in the presence of a eotaxin/CCL11 in vivo. Guo et al. (60) have demonstrated that The Journal of Immunology 2893 endogenous eotaxin/CCL11 is expressed during acute lung inflam- 15. Rothenberg, M. E., A. D. Luster, and P. Leder. 1995. Murine eotaxin: an eosin- mation and negatively regulates neutrophil influx into the lungs. ophil chemoattractant inducible in endothelial cells and in interleukin 4-induced tumor suppression. Proc. Natl. Acad. Sci. USA 92:8960. Furthermore, this study shows that eotaxin/CCL11-induced de- 16. Garcia-Zepeda, E. A., M. E. Rothenberg, R. T. Ownbey, J. Celestin, P. Leder, and creases in neutrophil content correlate with decreased lung dam- A. D. Luster. 1996. Human eotaxin is a specific chemoattractant for eosinophil cells and provides a new mechanism to explain tissue eosinophilia. Nat. Med. age. This study also demonstrated that recombinant eotaxin spe- 2:449. cifically decreased levels of the neutrophil-specific CXC 17. Uguccioni, M., C. R. Mackay, B. Ochensberger, P. Loetscher, S. Rhis, chemokines macrophage-inflammatory protein 2 and cytokine-in- G. J. LaRosa, P. Rao, P. D. Ponath, M. Baggiolini, and C. A. Dahinden. 1997. High expression of the chemokine receptor CCR3 in human blood basophils: role duced neutrophil chemoattractant in stimulated rat alveolar mac- in activation by eotaxin, MCP-4, and other chemokines. J. Clin. Invest. 100:1137. rophages. Our studies suggest that the mechanism of the effect of 18. Romagnani, P., A. De Paulis, C. Beltrame, F. Annunziato, V. Dente, E. Maggi, eotaxin/CCL11 may also involve altering levels of endothelial S. Romagnani, and G. Marone. 1999. Tryptase-chymase double-positive human mast cells express the eotaxin receptor CCR3 and are attracted by CCR3-binding cell-derived IL-8/CXCL8. Taken together, these two studies sug- chemokines. Am. J. Pathol. 155:1195. gest that eotaxin/CCL11 can affect neutrophil trafficking in vivo 19. Sallusto, F., C. R. Mackay, and A. Lanzavecchia. 1997. Selective expression of via effects on several different cell types. the eotaxin receptor CCR3 by human T helper 2 cells. Science 277:2005. 20. Ganzalo, J. A., G. Q. Jia, V. Aguirre, D. Friend, A. J. Coyle, N. A. Jenkins, We have demonstrated a novel link between CCR3 chemokine G. S. Lin, H. Katz, A. Lichtman, N. Copeland, et al. 1996. Mouse eotaxin ex- receptor activation and a pertussis toxin-sensitive, PI3K-dependent pression parallels eosinophil accumulation during lung allergic inflammation but it is not restricted to a Th2-type response. Immunity 4:1. pathway that leads to an acceleration in IL-8/CXCL8 mRNA de- 21. Campbell, E. M., S. L. Kunkel, R. M. Strieter, and N. W. Lukacs. 1998. Temporal cay and a decrease in IL-8/CXCL8 protein release. Endothelial- role of chemokines in a murine model of cockroach allergen-induced airway derived IL-8/CXCL8 is most likely a key effector of neutrophil hyperreactivity and eosinophilia. J. Immunol. 161:7047.

22. Minshall, E. M., L. Cameron, F. Lavigne, D. Y. Leung, D. Hamilos, Downloaded from elicitation in vivo. Eotaxin may act as an important negative reg- E. A. Garcia-Zepada, M. Rothenberg, A. D. Luster, and Q. Hamid. 1997. Eotaxin ulator that, in summation with proinflammatory factors, regulates mRNA and protein expression in chronic sinusitis and allergen-induced nasal the net output of IL-8/CXCL8 in vivo. Eotaxin/CCL11 down-reg- responses in seasonal allergic rhinitis. Am. J. Respir. Cell Mol. Biol. 17:683. 23. Baraniuk, J. N. 1997. Pathogenesis of allergic rhinitis. J. Allergy Clin. Immunol. ulation of the neutrophil chemoattractant IL-8/CXCL8 by endothelial 99:S763. cells in vivo may play an important biological role in protecting tis- 24. Chen, W., B. Paulus, D. Shu, Wilson, and V. Chadwick. 2001. Increased serum sues from overexposure to toxic neutrophil-derived substances, levels of eotaxin in patients with inflammatory bowel disease. Scand. J. Gastroenterol. 36:515. aiding in the resolution of acute inflammation, or transition from 25. Teruya-Feldstein, J., E. S. Jaffe, P. R. Burd, D. W. Kingma, J. E. Setsuda, and http://www.jimmunol.org/ neutrophilic to mononuclear/eosinophilic inflammation. G. Tosato. 1999. Differential chemokine expression in tissues involved by Hodgkin’s disease: direct correlation of eotaxin expression and tissue eosino- philia. Blood 93:2463. Acknowledgments 26. Rothenberg, M. E., A. D. Luster, C. M. Lilly, J. M. Drazen, and P. Leder. 1995. Constitutive and allergen-induced expression of eotaxin mRNA in the guinea pig We thank Kate Blease for helpful discussions. We also thank Holly lung. J. Exp. Med. 181:1211. Evanoff, Pam Lincoln, and Mary Lutzke for technical support. 27. Kitaura, M., T. Nakajima, T. Imai, S. Harada, C. Combadiere, H. L. Tiffany, P. M. Murphy, and O. Yoshie. 1996. Molecular cloning of human eotaxin, an eosinophil-selective CC chemokine, and identification of a specific eosinophil References eotaxin receptor, CC chemokine receptor 3. J. Biol. Chem. 271:7725. 1. Baggiolini, M. 1998. Chemokines and leukocyte traffic. Nature 392:565. 28. Hornung, D., K. Dohrn, K. Sotlar, R. R. Greb, D. Wallwiener, L. Kiesel, and 2. Premack, B., and T. Schall. 1996. Chemokine receptors: gateways to inflamma- R. N. Taylor. 2000. Localization in tissues and secretion of eotaxin by cells from by guest on September 25, 2021 tion and infection. Nat. Med. 2:1174. normal endometrium and endometriosis. J. Clin. Endocrinol. Metab. 85:2604. 3. Kelner, G. S., J. Kennedy, K. B. Bacon, S. Kleyensteuber, D. A. Largaespada, 29. Berger, O., X. Gan, C. Gujuluva, A. R. Burns, G. Sulur, M. Stins, D. Way, N. A. Jenkins, N. G. Copeland, J. F. Bazan, K. W. Moore, T. J. Schall, et al. 1994. M. Witte, M. Weinand, J. Said, et al. 1999. CXC and CC chemokine receptors on Lymphotactin: a cytokine that represents a new class of chemokine. Science coronary and brain endothelia. Mol. Med. 5:795. 266:1395. 30. Lavi, E., D. L. Kolson, A. M. Ulrich, L. Fu, and F. Gonzalez-Scarano. 1998. 4. Bazan, J. F., K. B. Bacon, G. Hardiman, W. Wang, K. Soo, D. Rossi, Chemokine receptors in the human brain and their relationship to HIV infection. D. R. Greaves, A. Zlotnik, and T. J. Schall. 1997. A new class of membrane- J. Neurovirol. 4:301. bound chemokine with a CX3C motif. Nature 385:640. 31. Salcedo, R., H. A. Young, M. L. Ponce, J. M. Ward, H. K. Kleinman, 5. Blease, K., B. Mehrad, N. W. Lukacs, S. L. Kunkel, T. J. Standiford, and W. J. Murphy, and J. J. Oppenheim. 2001. Eotaxin (CCL11) induces in vivo C. M. Hogaboam. 2001. Antifungal and airway remodeling roles for murine angiogenic responses by human CCR3ϩ endothelial cells. J. Immunol. 166:7571. monocyte chemoattractant protein-1/CCL2 during pulmonary exposure to As- 32. Feil, C., and H. G. Augustin. 1998. Endothelial cells differentially express func- perigillus fumigatus conidia. J. Immunol. 166:1832. tional CXC-chemokine receptor-4 (CXCR-4/fusin) under the control of autocrine 6. Zisman, D. A., S. L. Kunkel, R. M. Strieter, W. C. Tsai, K. Bucknell, activity and exogenous cytokines. Biochem. Biophys. Res. Commun. 247:38. J. Wilkowski, and T. J. Standiford. 1997. MCP-1 protects mice in lethal endo- 33. Elsner, J., R. Hochstetter, D. Kimmig, and A. Kapp. 1996. Human eotaxin rep- toxemia. J. Clin. Invest. 99:2832. resents a potent activator of the respiratory burst of human eosinophils. Eur. 7. Karpus, W. J., N. W. Lukacs, K. J. Kennedy, W. S. Smith, S. D. Hurst, and J. Immunol. 26:1919. T. A. Barrett. 1997. Differential CC chemokine-induced enhancement of Th cell 34. Lloyd, A. R., A. Biragyn, J. A. Johnston, D. D. Taub, L. Xu, D. Michiel, cytokine production. J. Immunol. 158:4129. H. Sprenger, J. J. Oppenheim, and D. J. Kelvin. 1995. Granulocyte-colony stim- 8. Taub, D. D., J. R. Ortaldo, S. M. Turcovski-Corrales, M. L. Key, D. L. Longo, ulating factor and lipopolysaccharide regulate the expression of interleukin 8 and W. J. Murphy. 1996. ␤ Chemokines costimulate lymphocyte cytolysis, pro- receptors on polymorphonuclear leukocytes. J. Biol. Chem. 270:28188. liferation, and lymphokine production. J. Leukocyte Biol. 59:81. 35. Chang, M. M., M. Juarez, D. M. Hyde, and R. Wu. 2001. Mechanism of dexa- 9. Nanki, T., and P. E. Lipsky. 2000. Cutting edge: stromal cell-derived factor-1 is methasone-mediated interleukin-8 gene suppression in cultured airway epithelial a costimulator for CD4ϩ T cell activation. J. Immunol. 164:5010. cells. Am. J. Physiol. 280:L107. 10. McGrath, K. E., A. D. Koniski, K. M. Maltby, J. K. McGann, and J. Palis. 1999. 36. Ebnet, K., and D. Vestweber. 1999. Molecular mechanisms that control leukocyte Embryonic expression and function of the chemokine SDF-1 and its receptor, extravasation: the selectins and the chemokines. Histochem. Cell Biol. 112:1. CXCR4. Dev. Biol. 213:442. 37. Smart, S. J., and T. B. Casale. 1994. TNF-␣-induced transendothelial neutrophil 11. Nagasawa, T., K. Tachibana, and T. Kishimoto. 1998. A novel CXC chemokine migration is IL-8 dependent. Am. J. Physiol. 266:L238. PBSF/SDF-1 and its receptor CXCR4: their functions in development, hemato- 38. Gupta, S. K., and J. P. Singh. 1994. Inhibition of endothelial cell proliferation by poiesis and HIV infection. Semin. Immunol. 10:179. platelet factor-4 involves a unique action on S phase progression. J. Cell Biol. 12. Addison, C. L., T. O. Daniel, M. D. Burdick, H. Liu, J. E. Ehlert, Y. Y. Xue, 127:1121. L. Buechi, A. Walz, A. Richmond, and R. M. Strieter. 2000. The CXC chemokine 39. Gupta, S. K., P. G. Lysko, K. Pillarisetti, E. Ohlstein, and J. M. Stadel. 1998. receptor 2, CXCR2, is the putative receptor for ELRϩ CXC chemokine-induced Chemokine receptors in human endothelial cells: functional expression of angiogenic activity. J. Immunol. 165:5269. CXCR4 and its transcriptional regulation by inflammatory cytokines. J. Biol. 13. Arenberg, D. A., S. L. Kunkel, P. J. Polverini, S. B. Morris, M. D. Burdick, Chem. 273:4282. M. C. Glass, D. T. Taub, M. D. Iannettoni, R. I. Whyte, and R. M. Strieter. 1996. 40. Maione, T. E., G. S. Gray, J. Petro, A. J. Hunt, A. L. Donner, S. I. Bauer, Interferon-␥-inducible protein 10 (IP-10) is an angiostatic factor that inhibits H. F. Carson, and R. J. Sharpe. 1990. Inhibition of angiogenesis by recombinant human non-small cell lung cancer (NSCLC) tumorigenesis and spontaneous me- human platelet factor-4 and related peptides. Science 247:77. tastases. J. Exp. Med. 184:981. 41. Salcedo, R., J. H. Resau, D. Halverson, E. A. Hudson, M. Dambach, D. Powell, 14. Koch, A. E., P. J. Polverini, S. L. Kunkel, L. A. Harlow, L. A. DiPietro, K. Wasserman, and J. J. Oppenheim. 2000. Differential expression and respon- V. M. Elner, S. G. Elner, and R. M. Strieter. 1992. Interleukin-8 as a macrophage- siveness of chemokine receptors (CXCR1Ð3) by human microvascular endothe- derived mediator of angiogenesis. Science 258:1798. lial cells and umbilical vein endothelial cells. FASEB J. 14:2055. 2894 EOTAXIN REGULATES IL-8 PRODUCTION IN ECs

42. Salcedo, R., K. Wasserman, H. A. Young, M. C. Grimm, O. M. Howard, 51. Huang, Z. F., J. B. Massey, and D. P. Via. 2000. Differential regulation of cy- M. R. Anver, H. K. Kleinman, W. J. Murphy, and J. J. Oppenheim. 1999. Vas- clooxygenase-2 (COX-2) mRNA stability by interleukin-1␤ (IL-1␤) and tumor cular endothelial growth factor and basic fibroblast growth factor induce expres- necrosis factor-␣ (TNF-␣) in human in vitro differentiated macrophages. Bio- sion of CXCR4 on human endothelial cells: in vivo neovascularization induced chem. Pharmacol. 59:187. by stromal-derived factor-1␣. Am. J. Pathol. 154:1125. 52. Xu, L., R. Rahimpour, L. Ran, C. Kong, A. Biragyn, J. Andrews, M. Devries, 43. Romagnani, P., F. Annunziato, L. Lasagni, E. Lazzeri, C. Beltrame, J. M. Wang, and D. J. Kelvin. 1997. Regulation of CCR2 chemokine receptor M. Francalanci, M. Uguccioni, G. Galli, L. Cosmi, L. Maurenzig, et al. 2001. Cell mRNA stability. J. Leukocyte Biol. 62:653. cycle-dependent expression of CXC chemokine receptor 3 by endothelial cells 53. Kasama, T., R. M. Strieter, N. W. Lukacs, M. D. Burdick, and S. L. Kunkel. mediates angiostatic activity. J. Clin. Invest. 107:53. 1994. Regulation of neutrophil-derived chemokine expression by IL-10. J. Im- 44. Belperio, J. A., M. P. Keane, D. A. Arenberg, C. L. Addison, J. E. Ehlert, munol. 152:3559. M. D. Burdick, and R. M. Strieter. 2000. CXC chemokines in angiogenesis. J. Leukocyte Biol. 68:1. 54. Blease, K., Y. Chen, P. G. Hellewell, and A. Burke-Gaffney. 1999. Lipoteichoic 45. Strieter, R. M., S. L. Kunkel, D. A. Arenberg, M. D. Burdick, and P. J. Polverini. acid inhibits lipopolysaccharide-induced adhesion molecule expression and IL-8 1995. Interferon ␥-inducible protein 10 (IP-10), a member of the C-X-C chemo- release in human lung microvascular endothelial cells. J. Immunol. 163:6139. kine family, is an inhibitor of angiogenesis. Biochem. Biophys. Res. Commun. 55. Carter, B. Z., and J. S. Malter. 1991. Regulation of mRNA stability and its 210:51. relevance to disease. Lab. Invest. 65:610. 46. Weber, K. S., P. J. Nelson, H. J. Grone, and C. Weber. 1999. Expression of CCR2 56. Wen, X., and G. D. Wu. 2001. Evidence for epigenetic mechanisms that silence by endothelial cells: implications for MCP-1 mediated wound injury repair and in both basal and immune-stimulated transcription of the IL-8 gene. J. Immunol. vivo inflammatory activation of endothelium. Arterioscler. Thromb. Vasc. Biol. 166:7290. 19:2085. 57. Tedgui, A., and Z. Mallat. 2001. Anti-inflammatory mechanisms in the vascular 47. Auron, P. E. 1998. The interleukin 1 receptor: ligand interactions and signal wall. Circ. Res. 88:877. transduction. Cytokine Growth Factor Rev. 9:221. 58. Romagnani, P., F. Annunziato, E. Lazzeri, L. Cosmi, C. Beltrame, L. Lasagni, 48. Wallach, D., E. E. Varfolomeev, N. L. Malinin, Y. V. Goltsev, A. V. Kovalenko, G. Galli, M. Francalanci, R. Manetti, F. Marra, et al. 2001. Interferon-inducible and M. P. Boldin. 1999. Tumor necrosis factor receptor and Fas signaling mech- protein 10, monokine induced by interferon ␥, and interferon-inducible T-cell ␣ anisms. Annu. Rev. Immunol. 17:331. chemoattractant are produced by thymic epithelial cells and attract T-cell receptor Downloaded from 49. Tebo, J. M., S. Datta, R. Kishore, M. Kolosov, J. A. Major, Y. Ohmori, and (TCR) ␣␤ϩ CD8ϩ single-positive T cells, TCR␥␦ϩ T cells, and natural killer- T. A. Hamilton. 2000. Interleukin-1-mediated stabilization of mouse KC mRNA type cells in human thymus. Blood 97:601. depends on sequences in both 5Ј- and 3Ј-untranslated regions. J. Biol. Chem. 275:12987. 59. Patel, L., S. J. Charlton, J. K. Chambers, and C. H. Macphee. 2001. Expression 50. Zen, K., A. Karsan, A. Stempien-Otero, E. Yee, J. Tupper, X. Li, T. Eunson, and functional analysis of chemokine receptors in human peripheral blood leu- M. A. Kay, C. B. Wilson, R. K. Winn, and J. M. Harlan. 1999. NF-␬B activation kocyte populations. Cytokine 14:27. is required for human endothelial survival during exposure to tumor necrosis 60. Guo, R. F., A. B. Lentsch, R. L. Warner, M. Huber-Lang, J. V. Sarma, T. Hlaing, factor-␣ but not to interleukin-1␤ or lipopolysaccharide. J. Biol. Chem. 274: M. M. Shi, N. W. Lukacs, and P. A. Ward. 2001. Regulatory effects of eotaxin

28808. on acute lung inflammatory injury. J. Immunol. 166:5208. http://www.jimmunol.org/ by guest on September 25, 2021