Title: Chemokine Receptors in Human Endothelial Cells
Abstract: Chemokines play an important role in the regulation of endothelial cell (EC) function, including proliferation, migration and differentiation during angiogenesis, and re-endothelialization after injury. In this study, reverse transcriptase-polymerase chain reaction was used to reveal expression of various CXC and CC chemokine receptors in human umbilical vein EC. Northern analysis showed that CXCR4 was selectively expressed in vascular EC, but not in smooth muscle cells. Compared with other chemokines, stromal cell-derived factor-1α (SDF-1α), the known CXCR4 ligand, was an efficacious chemoattractant for EC, causing the migration of ∼40% input cells with an EC50 of 10–20 nm. Of the chemokines tested, only SDF-1α induced a rapid, though variable mobilization of intracellular Ca2+in EC. Experiments with actinomycin D demonstrated that CXCR4 transcripts were short-lived, indicating a rapid mRNA turnover. Interferon-γ (IFN-γ) caused a pronounced down-regulation of CXCR4 mRNA in a concentration- and time-dependent manner. In a striking functional correlation, IFN-γ treatment also attenuated the chemotactic response of EC to SDF-1α. IL-1β, tumor necrosis factor-α, and lipopolysaccharide produced a time course-dependent biphasic effect on CXCR4 transcription. Expression of CXCR4 in EC is significant, more so as it and several CC chemokine receptors have been shown to serve as fusion co-receptors along with CD4 during human immunodeficiency virus infection. Taken together, these findings provide evidence of chemokine receptor expression in EC and offer an explanation for the action of chemokines like SDF-1α on the vascular endothelium. Chemokines play an important role in the regulation of endothelial cell (EC) function, including proliferation, migration and differentiation during angiogenesis, and re-endothelialization after injury. In this study, reverse transcriptase-polymerase chain reaction was used to reveal expression of various CXC and CC chemokine receptors in human umbilical vein EC. Northern analysis showed that CXCR4 was selectively expressed in vascular EC, but not in smooth muscle cells. Compared with other chemokines, stromal cell-derived factor-1α (SDF-1α), the known CXCR4 ligand, was an efficacious chemoattractant for EC, causing the migration of ∼40% input cells with an EC50 of 10–20 nm. Of the chemokines tested, only SDF-1α induced a rapid, though variable mobilization of intracellular Ca2+in EC. Experiments with actinomycin D demonstrated that CXCR4 transcripts were short-lived, indicating a rapid mRNA turnover. Interferon-γ (IFN-γ) caused a pronounced down-regulation of CXCR4 mRNA in a concentration- and time-dependent manner. In a striking functional correlation, IFN-γ treatment also attenuated the chemotactic response of EC to SDF-1α. IL-1β, tumor necrosis factor-α, and lipopolysaccharide produced a time course-dependent biphasic effect on CXCR4 transcription. Expression of CXCR4 in EC is significant, more so as it and several CC chemokine receptors have been shown to serve as fusion co-receptors along with CD4 during human immunodeficiency virus infection. Taken together, these findings provide evidence of chemokine receptor expression in EC and offer an explanation for the action of chemokines like SDF-1α on the vascular endothelium. The vascular endothelium is strategically located to play a prominent sensory and effector cell role in the maintenance of hemostasis, and during the vascular response to inflammation, infection, and injury (1Pober J.S. Cotran R.S. Transplantation. 1990; 50: 537-544Crossref PubMed Scopus (715) Google Scholar, 2Mantovani A. Bussolino F. Dejana E. FASEB J. 1992; 6: 2591-2599Crossref PubMed Scopus (628) Google Scholar). The endothelium is also integrally associated with angiogenesis (3Maciag T. Burgess W.H. Ryan U.S. Endothelial Cells. II. CRC Press, Inc., Boca Raton, FL1988: 3-10Google Scholar) and cardiovascular disorders such as atherosclerosis and restenosis (4Gibbons G.H. Dzau V.J. N. Engl. J. Med. 1994; 330: 1431-1438Crossref PubMed Scopus (1351) Google Scholar). Endothelial cells (EC) 1The abbreviations used are: EC, endothelial cell(s); CCR, CC chemokine receptor; CXCR, CXC chemokine receptor; FACS, fluorescence-activated cell sorter; FBHEC, fetal bovine heart endothelial cell(s); HUVEC, human umbilical vein endothelial cell(s); HCAEC, human coronary artery endothelial cell(s); HBMEC, human brain microvascular endothelial cell(s); IFN, interferon; LPS, lipopolysaccharide; TNF, tumor necrosis factor; PCR, polymerase chain reaction; RT-PCR, reverse transcription PCR; IL, interleukin; HIV, human immunodeficiency virus; PBS, phosphate-buffered saline; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; RANTES, regulated on activation normal T cell expressed and secreted. interact with various inflammatory cells, as well as platelets and smooth muscle cells via a variety of chemotactic factors such as chemokines and their receptors (5Ben-Baruch A. Michiel D.F. Oppenheim J.J. J. Biol. Chem. 1995; 270: 11703-11706Abstract Full Text Full Text PDF PubMed Scopus (368) Google Scholar, 6Strieter R.M. Kunkel S.L. Showell H.J. Remick D.G. Phan S.H. Ward P.A. Marks R.M. Science. 1989; 243: 1467-1469Crossref PubMed Scopus (633) Google Scholar). Chemokines are classified into at least two groups, which differ with respect to the organization of the dicysteine motif present at the NH2 terminus. The α-chemokines, characterized by the CXC motif include PF-4, IL-8, γIP-10 and SDF-1. The β-chemokines, characterized by the CC motif include MCP-1, MIP-1α and 1β, and RANTES (5Ben-Baruch A. Michiel D.F. Oppenheim J.J. J. Biol. Chem. 1995; 270: 11703-11706Abstract Full Text Full Text PDF PubMed Scopus (368) Google Scholar, 7Stoeckle M.Y. Barker K.A. New Biol. 1990; 2: 313-323PubMed Google Scholar, 8Shirozu M. Nakano T. Inazawa J. Tashiro K. Tada H. Shinohara T. Honjo T. Genomics. 1995; 28: 495-500Crossref PubMed Scopus (536) Google Scholar). Chemokines mediate their specific effect on target cells through two related subfamilies of G-protein coupled receptors. To date, several CXC and CC functional human chemokine receptors have been discovered (9Lee J. Horuk R. Rice G.C. Bennett G.L. Camerato T. Wood W.I. J. Biol. Chem. 1992; 267: 16283-16287Abstract Full Text PDF PubMed Google Scholar, 10Loetscher M. Geiser T. O'Reilly T. Zwahlen R. Baggiolini M. Moser B. J. Biol. Chem. 1994; 269: 232-237Abstract Full Text PDF PubMed Google Scholar, 11Loetscher M. Gerber B. Loetscher P. Jones S.A. Piali L. Clark-Lewis I. Baggiolini M. Moser B. J. Exp. Med. 1996; 184: 963-969Crossref PubMed Scopus (1063) Google Scholar, 12Gao J.-L. Kuhns D.B. Tiffany H.L. McDermott D. Li X. Francke U. Murphy P.M. J. Exp. Med. 1993; 177: 1421-1427Crossref PubMed Scopus (339) Google Scholar, 13Charo I.F. Myers S.J. Herman A. Franci C. Connolly A.J. Coughlin S.R. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 2752-2756Crossref PubMed Scopus (650) Google Scholar, 14Combadiere C. Ahuja S.K. Murphy P.M. J. Biol. Chem. 1995; 270: 16491-16494Abstract Full Text Full Text PDF PubMed Scopus (217) Google Scholar, 15Power C.A. Meyer A. Nemeth K. Bacon K.B. Hoogewerf A.J. Proudfoot A.E.I. Wells T.N.C. J. Biol. Chem. 1995; 270: 19495-19500Abstract Full Text Full Text PDF PubMed Scopus (254) Google Scholar, 16Samson M. Labbe O. Mollereau C. Vassart G. Parmentier M. Biochemistry. 1996; 35: 3362-3367Crossref PubMed Scopus (588) Google Scholar). In line with their well defined role as mediators of diapedesis, the chemokine receptors have been primarily localized on neutrophils, monocytes, lymphocytes, and eosinophils (5Ben-Baruch A. Michiel D.F. Oppenheim J.J. J. Biol. Chem. 1995; 270: 11703-11706Abstract Full Text Full Text PDF PubMed Scopus (368) Google Scholar). However, little is known about other distinct functions of these cytokines and their interaction with non-hematopoietic cells. Three lines of evidence indicate that human EC also express the genes for chemokine receptors and thus play an active and important role as target cells for chemokine function. First, the proliferation, migration, and differentiation of vascular EC, during angiogenesis, is modulated by chemokines, apparently via specific receptors. Thus, IL-8 is an inducer of angiogenesis (17Koch A.E. Polverini P.J. Kunkel S.L. Harlow L.A. DiPietro L.A. Elner V.M. Elner S.G. Strieter R.M. Science. 1992; 258: 1798-1801Crossref PubMed Scopus (1904) Google Scholar), whereas PF-4 (18Maione T.E. Gray G.S. Petro J. Hunt A.J. Donner A.L. Bauer S.I. Carson H.F. Sharpe R.J. Science. 1990; 247: 77-79Crossref PubMed Scopus (627) Google Scholar, 19Gupta S.K. Singh J.P. J. Cell Biol. 1994; 127: 1121-1127Crossref PubMed Scopus (108) Google Scholar, 20Gupta S.K. Hassel T. Singh J.P. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 7799-7803Crossref PubMed Scopus (145) Google Scholar), Gro-β (21Cao Y. Chen C. Weatherbee J.A. Tsang M. Folkman J. J. Exp. Med. 1995; 182: 2069-2077Crossref PubMed Scopus (124) Google Scholar), and γIP-10 (22Strieter R.M. Kunkel S.L. Arenberg D.A. Burdick M.D. Polverini P.J. Biochem. Biophys. Res. Commun. 1995; 210: 51-57Crossref PubMed Scopus (267) Google Scholar) are inhibitors of EC proliferation and angiogenesis. Second, it has been suggested that leukocyte adhesion to the endothelium and transmigration require that chemotactic factors be immobilized on the EC surface (23Rot A. Immunol. Today. 1992; 13: 291-294Abstract Full Text PDF PubMed Scopus (413) Google Scholar, 24Tanaka Y. Adams D.H. Shaw S. Immunol. Today. 1993; 14: 111-114Abstract Full Text PDF PubMed Scopus (382) Google Scholar). This idea is necessitated due to the obvious conceptual difficulty in generating a chemotactic gradient of soluble chemokines under conditions of blood flow. Although chemokines can bind cell surface proteoglycans (24Tanaka Y. Adams D.H. Shaw S. Immunol. Today. 1993; 14: 111-114Abstract Full Text PDF PubMed Scopus (382) Google Scholar, 25Luster A.D. Greenberg S.M. Leder P. J. Exp. Med. 1995; 182: 219-231Crossref PubMed Scopus (419) Google Scholar), vascular endothelium may still require expression of receptors that are capable of immobilizing chemokines to generate a specific haptotactic gradient. Third, recent studies have shown that CXCR4 (26Feng Y. Broder C.C. Kennedy P.E. Berger E.A. Science. 1996; 272: 872-877Crossref PubMed Scopus (3636) Google Scholar) and several CC chemokine receptors like CCR2b, CCR3, and CCR5 (27Dragic T. Litwin V. Allaway G.P. Martin S.R. Huang Y. Nagashima K.A. Cayanan C. Maddon P.J. Koup R.A. Moore J.P. Paxton W.A. Nature. 1996; 381: 667-673Crossref PubMed Scopus (2816) Google Scholar, 28Alkhatib G. Combadiere C. Broder C.C. Feng Y. Kennedy P.E. Murphy P.M. Berger E.A. Science. 1996; 272: 1955-1958Crossref PubMed Scopus (2446) Google Scholar, 29Doranz B.J. Rucker J. Yi Y. Smyth R.J. Samson M. Peiper S.C. Parmentier M. Collman R.G. Doms R.W. Cell. 1996; 85: 1149-1158Abstract Full Text Full Text PDF PubMed Scopus (1686) Google Scholar, 30Choe H. Farzan M. Sun Y. Sullivan N. Rollins B. Ponath P.D. Wu L. Mackay C.R. Gregory L. Newman W. Gerard N. Gerard C. Sodroski J. Cell. 1996; 85: 1135-1148Abstract Full Text Full Text PDF PubMed Scopus (2093) Google Scholar) serve as co-factors in association with CD4 to permit HIV infection. This also raises the possibility that the HIV susceptibility of EC in a CD4-independent manner (31Stefan A.-M. Lafon M.-E. Gendrault J.-L. Schweitzer C. Royer C. Jaeck D. Arnaud J.-P. Schmitt M.-P. Aubertin A.-M. Kirn A. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 1582-1586Crossref PubMed Scopus (99) Google Scholar, 32Corbeil J. Evans L.A. McQueen P.W. Vasak E. Edward P.D. Richman D.D. Penny R. Cooper D.A. Immunol. Cell Biol. 1995; 73: 140-145Crossref PubMed Google Scholar) may be due to their expression of CXCR4 and other chemokine co-receptors. Indeed, evidence for this hypothesis was provided in a recent study (33Endres M.J. Clapham P.R. Marsh M. Ahuja M. Turner J.D. McKnight A. Thomas J.F. Stobenau-Haggarty B. Choe S. Vance P.J. Wells T.N.C. Power C.A. Sutterwala S.S. Doms R.W. Landau N.R. Hoxie J.A. Cell. 1996; 87: 745-756Abstract Full Text Full Text PDF PubMed Scopus (587) Google Scholar), which showed that CXCR4 could function as an alternative receptor for isolates of HIV-2 in the absence of CD4. Therefore, to gain a better understanding of the role of chemokines, we examined the repertoire of chemokine receptor mRNAs expressed by EC. Furthermore, the functional expression and transcriptional regulation of CXCR4 receptor in EC was studied in detail, and its biological implications are discussed. Recombinant human IFN-γ, TNF-α, IL-1β, basic fibroblast growth factor, and transforming growth factor-β were purchased from Genzyme (Cambridge, MA). Bacterial LPS, actinomycin D, and Me2SO were from Sigma. SDF-1α was obtained from Gryphon Sciences (South San Francisco, CA), and other chemokines were from R&D Systems (Minneapolis, MN). Primary cultures of HUVEC, human brain microvascular endothelial cells (HBMEC), and human coronary artery endothelial cells (HCAEC) were purchased from Cell Systems (Kirkland, WA) and maintained in their proprietary CS-C complete medium without antibiotics, in tissue culture flasks coated with 0.1% gelatin (Sigma). Fetal bovine heart endothelial cells (FBHEC) were obtained from ATCC (CRL1395) and cultured in Dulbecco's modified Eagle's medium containing 10% fetal calf serum, 2 mm glutamine, and 20 ng/ml basic fibroblast growth factor. Cells were passaged at confluence and used within the first seven passages. Based on the published sequence of human chemokine receptors, the following pair of consensus degenerate 20-mer primers were synthesized from the ends of the third and seventh transmembrane domains of chemokine receptors.CKF:5′TAYCTSGCYATYGTSCAYGC3′CKR:5′AARGCRTARATSAYKGGRTT3′ The symbols follow the IUB/GCG convention (Y = C/T, S = C/G, R = A/G, and K = G/T). Total cellular RNA was isolated from 107 early passage HUVEC and HCAEC by the single extraction Tri-reagent procedure (Molecular Research Center, Inc. Cincinnati, OH), according to the manufacturer's protocol and stored dissolved in Formazol at −80 °C. PCR amplification of total RNA was done with the GeneAmp RNA PCR kit (Perkin-Elmer) as described previously (34Gupta S.K. Singh J.P. Gene (Amst.). 1993; 124: 287-290Crossref PubMed Scopus (6) Google Scholar). Two μg of total RNA was reverse-transcribed with the “downstream” antisense oligomer, CK-R. The “upstream” oligomer CK-F, was added directly to the reaction tubes along with the PCR “reaction mix” and subjected to 35 cycles of amplification. The PCR products were analyzed on agarose gels and subcloned directly into the PCRII TA vector (Invitrogen). Plasmid DNA from individual colonies were analyzed by restriction digestion and sequencing. Total RNA (10 μg/lane) was fractionated on 1% agarose-formaldehyde gels, transferred to a nylon membrane (Amersham Corp.), and covalently linked with a UV cross-linker (Stratagene Inc., La Jolla, CA). For Northern analysis, 515-base pair cDNA probes of CXCR1, CXCR2, CXCR3, CXCR4, CCR1, CCR2, and CCR3 were used. The GAPDH gene probe (CLONTECH) was used to normalize RNA sample differences in each lane. The probes were labeled with [α-32P]dCTP using a random-prime labeling kit (Promega Corp., Madison, WI) and hybridized overnight at 42 °C in 6 × SSC buffer (1 × SSC = 150 mm NaCl, 15 mm sodium citrate), 0.1% sodium dodecyl sulfate, 5 × Denhardt's solution, 50% formamide, and 100 μg/ml denatured salmon sperm DNA. Membranes were washed with a final stringency of 0.2 × SSC at 60 °C, and analyzed with a phosphorimager (Molecular Devices, Inc.) after exposure at room temperature for 3–5 days. Densitometry was used for quantitative analysis. Cell surface expression of CXCR4 receptor was analyzed as described previously (33Endres M.J. Clapham P.R. Marsh M. Ahuja M. Turner J.D. McKnight A. Thomas J.F. Stobenau-Haggarty B. Choe S. Vance P.J. Wells T.N.C. Power C.A. Sutterwala S.S. Doms R.W. Landau N.R. Hoxie J.A. Cell. 1996; 87: 745-756Abstract Full Text Full Text PDF PubMed Scopus (587) Google Scholar, 35LaBranche C.C. Sauter M.M. Haggarty B.S. Vance P.J. Romano J. Hart T.K. Bugelski P.J. Hoxie J.A. J. Virol. 1994; 68: 5509-5522Crossref PubMed Google Scholar). Briefly, 5 × 105 HUVEC were permeabilized in the presence of 0.2% Triton X-100/PBS for 2 min, and then resuspended in ice-cold PBS, 0.1% bovine serum albumin. Cells were incubated on ice for 30 min with the primary 12G5 antibody (35LaBranche C.C. Sauter M.M. Haggarty B.S. Vance P.J. Romano J. Hart T.K. Bugelski P.J. Hoxie J.A. J. Virol. 1994; 68: 5509-5522Crossref PubMed Google Scholar) or a control anti-PECAM antibody (R&D Systems) of the same subclass. Cells were then washed twice with ice-cold PBS, 0.1% bovine serum albumin and labeled with a second-stage fluorescein isothiocyanate-conjugated goat anti-mouse IgG (Tago Laboratories). FACS analysis was done with a FACScan flow cytometer (Becton Dickinson). For measurements of intracellular calcium [Ca2+]i, EC were loaded with 2 μm fura-2/AM (Molecular Probes, Eugene, OR), rinsed with 1 mm EDTA in Dulbecco's PBS, and resuspended into Krebs-Ringer-Henseleit buffer, pH 7.4, containing 0.1% gelatin. Cells (1 × 106/ml) were stored on ice and diluted for use 1:1 with fresh Krebs-Ringer-Henseleit buffer at 37 °C. Fluorescence of fura-2 in cells was measured with a dual channel fluorometer as described previously (36Lysko P.G. Webb C.L. Yue T.-L. Gu J.-L. Feuerstein G. Stroke. 1994; 25: 2476-2482Crossref PubMed Scopus (99) Google Scholar). Chemokines were added from concentrated stocks in water. To establish the integrity of EC, we also measured [Ca2+]i stimulated by thrombin. HUVEC migration assay was performed using 5 × 105 cells/well (in CS-C medium) in the top chamber of a 6.5-mm diameter, 8-μm pore polycarbonate Transwell culture insert (Costar, Cambridge, MA) as reported previously (37Bleul C.C. Fuhlbrigge R.C. Casanovas J.M. Aiuti A. Springer T.A. J. Exp. Med. 1996; 184: 1101-1109Crossref PubMed Scopus (1285) Google Scholar). Incubation was carried out at 37 °C in 5% CO2 for 20 h. After incubation, migrated cells in the lower chamber were counted with a ZM Coulter counter (Coulter Diagnostics, Hialeah, FL). Percent migration was calculated based on the total initial input cells per well. To explore the expression of chemokine receptor transcripts in human EC, total cellular RNA from HUVEC was amplified by RT-PCR with the consensus region primers (see “Experimental Procedures”). An expected 515-base pair cDNA band was amplified and subcloned to generate a cDNA plasmid library enriched for chemokine receptor clones. A total of 110 out of the 250 isolated clones were randomly sequenced and analyzed for their sequence distribution. CXCR4, representing 45% of the sequenced clones was the most prevalent chemokine receptor, followed by clones with identity to CCR3 (10%), the eotaxin receptor. Also present were clones having inserts with CXCR1, CCR1, and CCR2 sequences. These data provide evidence that vascular EC have the ability to express mRNA for several chemokine receptors. The results are also consistent with previous reports where CXCR2 expression was detected in HUVEC by RT-PCR (38Schonbeck U. Brandt E. Petersen F. Flad H.-D. Loppnow H. J. Immunol. 1995; 154: 2375-2383PubMed Google Scholar), and specific binding of IL-8 and RANTES was observed on the endothelium of postcapillary venules and veins in human skin by using an in situ binding assay (39Rot A. Hub E. Middleton J. Pons F. Rabeck C. Thierer K. Wintle J. Wolff B. Zsak M. Dukor P. J. Leukocyte Biol. 1996; 59: 39-44Crossref PubMed Scopus (116) Google Scholar). Steady state expression of chemokine receptors in vascular EC was studied by Northern blot analysis of total RNA. Fig.1 A (arrow) shows that both HUVEC and HCAEC express similar amounts of an expected 1.8-kilobase size mRNA after hybridization with the CXCR4 cDNA probe. In fact, these results also suggest that CXCR4 is the most abundant chemokine receptor expressed in vascular EC, as identical Northern blots with EC RNA did not hybridize with 515-base pair CXCR1, CXCR2, CXCR3, CCR1, CCR2, and CCR3 cDNA probes (data not shown). It is conceivable that EC primarily express these chemokine receptors at a low level, and the binding of chemokines with cell surface proteoglycans facilitates their interaction with the specific receptors expressed in low copy numbers. In this context, it is important to note that several chemokines like IL-8, Gro-β, γIP-10, and PF-4 directly modulate EC proliferation or migration (17Koch A.E. Polverini P.J. Kunkel S.L. Harlow L.A. DiPietro L.A. Elner V.M. Elner S.G. Strieter R.M. Science. 1992; 258: 1798-1801Crossref PubMed Scopus (1904) Google Scholar, 18Maione T.E. Gray G.S. Petro J. Hunt A.J. Donner A.L. Bauer S.I. Carson H.F. Sharpe R.J. Science. 1990; 247: 77-79Crossref PubMed Scopus (627) Google Scholar, 19Gupta S.K. Singh J.P. J. Cell Biol. 1994; 127: 1121-1127Crossref PubMed Scopus (108) Google Scholar, 20Gupta S.K. Hassel T. Singh J.P. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 7799-7803Crossref PubMed Scopus (145) Google Scholar, 21Cao Y. Chen C. Weatherbee J.A. Tsang M. Folkman J. J. Exp. Med. 1995; 182: 2069-2077Crossref PubMed Scopus (124) Google Scholar, 22Strieter R.M. Kunkel S.L. Arenberg D.A. Burdick M.D. Polverini P.J. Biochem. Biophys. Res. Commun. 1995; 210: 51-57Crossref PubMed Scopus (267) Google Scholar), presumably in a receptor-mediated interaction. CXCR4 transcripts are well expressed in many non-hematopoietic vascular tissues like heart, brain, lung, and colon (40Federsppiel B. Melhado I.G. Duncan A.M.V. Delaney A. Schappert K. Clark-Lewis I. Jirik F.R. Genomics. 1993; 16: 707-712Crossref PubMed Scopus (162) Google Scholar); however, at the cellular level, we found this expression was selective for EC, as indicated by the failure of total RNA from human pulmonary artery smooth muscle cells to hybridize with the CXCR4 cDNA probe (Fig.1 B, lane 1). To obtain an initial insight into the regulation of CXCR4 in EC during inflammation, we treated the HUVEC with various mediators and measured its steady state mRNA levels after normalization against the GAPDH cDNA probe. Fig. 1(B and C) shows that IFN-γ and, to a lesser extent, TNF-α caused a decrease in CXCR4 mRNA levels after 24 h of treatment. IL-1β and LPS caused a significant induction, whereas no effect was observed after treatment with transforming growth factor-β, γIP-10, and Me2SO. The transcription inhibitor actinomycin D caused an almost complete abrogation of CXCR4 message in the same time period. The cell surface expression of CXCR4 was evaluated by FACS analysis of HUVEC by using the specific anti-CXCR4 monoclonal antibody 12G5 (35LaBranche C.C. Sauter M.M. Haggarty B.S. Vance P.J. Romano J. Hart T.K. Bugelski P.J. Hoxie J.A. J. Virol. 1994; 68: 5509-5522Crossref PubMed Google Scholar). As demonstrated in Fig.2, there was a shift in the fluorescence intensity of cells after treatment with 12G5, clearly indicating that mRNA expression of CXCR4 is translated into surface expression of the receptor on HUVEC. To help understand the kinetics of inflammation-mediated transcriptional regulation of CXCR4, we used actinomycin D to determine the half-life of its mRNA. As indicated by the selective degradation of existing mRNA upon addition of actinomycin D to EC cultures (Fig.3), CXCR4 mRNA has a short half-life of around 2 h and is probably subject to a rapid turnover. This is noteworthy, as such rapid turnover of CXCR4 may allow the EC to respond promptly during conditions of infection and inflammatory stress. In addition we also observed that actinomycin D had the unexpected effect of sharply increasing the steady state levels of CXCR4 mRNA after a short term exposure of only 15–30 min. Many cytokines and cytokine receptors, including CXCR4, have A-U-rich elements in their untranslated regions, which serve as targeting motifs for transcript degradation by specific RNases (41Shaw G. Kamen R. Cell. 1986; 46: 659-667Abstract Full Text PDF PubMed Scopus (3123) Google Scholar). It is possible that, in addition to its action as a transcriptional inhibitor, actinomycin D also has the unique and immediate effect of imparting stability to existing transcripts of mRNA undergoing rapid turnover. Unlike other inducible chemokines, SDF-1α, which is the known ligand for CXCR4 (42Bleul C.C. Farzan M. Choe H. Parolin C. Clark-Lewis I. Sodroski J. Springer T. Nature. 1996; 382: 829-833Crossref PubMed Scopus (1752) Google Scholar, 43Oberlin E. Amara A. Bachelerie F. Bessia C. Virelizier J.-L. Seisdedos F.A. Schwartz O. Heard J.-M. Lewis I.C. Legler D.F. Loetscher M. Baggiolini M. Moser B. Nature. 1996; 382: 833-835Crossref PubMed Scopus (1483) Google Scholar), is constitutively expressed in numerous tissues (8Shirozu M. Nakano T. Inazawa J. Tashiro K. Tada H. Shinohara T. Honjo T. Genomics. 1995; 28: 495-500Crossref PubMed Scopus (536) Google Scholar); therefore, its biological action is likely to be regulated at the level of CXCR4 receptor expression. We examined the kinetics of cytokine modulation of CXCR4 mRNA expression in EC, and Northern blots were done to study the effects of IL-1β, IFN-γ, TNF-α, and LPS at different time intervals and concentration ranges. These mediators are known to be simultaneously up-regulated during inflammation and the pathogenesis of vascular diseases like atherosclerosis and restenosis (44Schwartz S.M. deBlois D. O'Brien E.R.M. Circ. Res. 1995; 77: 445-465Crossref PubMed Scopus (898) Google Scholar), and exhibited distinct effects on the expression of CXCR4 in EC in the initial studies (Fig. 1). As shown in Fig. 4, treatment of HUVEC with IFN-γ (103 units/ml) caused a rapid and sustained decrease in steady state levels of CXCR4 mRNA, which reached its maximum within 3 h after treatment and continued to exert its inhibitory effect up to 24 h thereafter. Furthermore, in HUVEC treated with IFN-γ (103 units/ml) for 24 h, there was a marked reduction in the half-life of CXCR4 mRNA from ∼2 h to about 15 min. Nuclear run-on experiments did not reveal any effect of IFN-γ on the rate of synthesis of CXCR4 mRNA in HUVEC (data not shown), thereby indicating that its inhibitory effect is caused at the level of CXCR4 mRNA stability. In contrast to IFN-γ, mediators like TNF-α, IL-1β, and LPS had a distinctly more complex and unique time-dependent biphasic effect on CXCR4 expression. This effect was characterized by an immediate decrease, followed by a subsequent reversal and increase in the steady state levels of CXCR4 mRNA despite continuous exposure of EC to the cytokines. The mechanism behind this biphasic mode of transcriptional regulation is unclear at present, although the most likely explanation is that the extended exposure of EC to TNF-α, IL-1β, and LPS imparts stability to newly synthesized CXCR4 transcripts that are otherwise subject to a rapid degradation. In comparison, LPS has been shown to cause a reduction in mRNA levels of CCR2, CCR1, and CCR5 (45Sica A. Saccani A. Borsatti A. Power C.A. Wells T.N. Luini W. Polentarutti N. Sozzani S. Mantovani A. J. Exp. Med. 1997; 185: 969-974Crossref PubMed Scopus (257) Google Scholar). Our studies also show that, among the inflammatory mediators studied, IFN-γ is the dominant effector of CXCR4 transcription in EC. Even low concentrations of IFN-γ (10 units/ml, Fig.5), which had no effect on the expression of CCR2 (46Tangirala R.K. Murao K. Quehenberger O. J. Biol. Chem. 1997; 272: 8050-8056Abstract Full Text Full Text PDF PubMed Scopus (97) Google Scholar), were sufficient to achieve significant inhibition of CXCR4 mRNA. Moreover, when IFN-γ (100 units/ml) was added along with TNF-α, IL-1β, or LPS to EC cultures for 24 h, it continued to exert a down-regulatory effect on transcription (Fig. 5). In the same experiment, a combination of TNF-α (10 ng) with either IL-1β or LPS did not have a synergistic effect on CXCR4 expression. This is in marked contrast to their effect on CCR2 expression, which was completely abolished (46Tangirala R.K. Murao K. Quehenberger O. J. Biol. Chem. 1997; 272: 8050-8056Abstract Full Text Full Text PDF PubMed Scopus (97) Google Scholar), and CXCR1 expression, which was significantly down-regulated by a combination of TNF-α and LPS (47Lloyd A.R. Biragyn A. Johnston J.A. Taub D.D. Xu L. Michiel D. Sprenger H. Oppenheim J.J. Kelvin D.J. J. Biol. Chem. 1995; 270: 28188-28192Abstract Full Text Full Text PDF PubMed Scopus (111) Google Scholar). Thus, CXCR4 mRNA expression in EC is regulated in a unique pattern that has not been observed for other chemokine receptors such as CXCR1 and CCR2. To determine whether EC express a functional CXCR4 receptor, our subsequent studies used SDF-1α along with several other chemokines to assess their ability to induce changes in intracellular levels of Ca2+ and cause migration. As shown in Fig. 6, SDF-1α induced a rapid elevation of [Ca2+]i in various EC types, with maximal response at a concentration of 100 nm. In contrast, other chemokines like γ-IP10, IL-8, PF-4, MIP-1α, MCP-1, eotaxin, and RANTES had no effect on EC (data not shown). Since the Ca2+ flux induced by SDF-1α in primary cultures of HUVEC and HBMEC was small (50–70 nm) and characteristically variable, we used FBHEC, an established EC line, to calculate the EC50 of SDF-1α-mediated response. As evident in Fig. 6 (inset), SDF-1α induced a robust Ca2+ flux (up to 1 μm) with an EC50 of ∼2 nm in the case of FBHEC. We next studied the chemotactic response of EC to SDF-1α. SDF-1α induced a pronounced migration of ∼40% of input EC in a concentration-related manner with an EC50 of 10–20 nm (Fig. 7 A). It is intriguing to observe the high percentage of EC that migrated in response to SDF-1α, even though EC have limited migratory capability in comparison with neutrophils and monocytes. We also noticed that in contrast with other EC chemo-attractants like vitronectin (data not shown), the chemotactic response to SDF-1α was kinetically robust, and a majority of the migrated cells entered the lower chamber without adhering to the Transwell filter. In addition, checkerboard analysis also indicated that the migratory response of EC to SDF-1α was chemotactic rather than being chemokinetic (data not shown). It is important to note that, in our experiments, other chemokines like γ-IP10, IL-8, MIP-1α, MCP-1, eotaxin, and RANTES had no effect on EC chemotaxis. The lack of EC chemotaxis in response to IL-8 is noteworthy, especially in view of previous data (17Koch A.E. Polverini P.J. Kunkel S.L. Harlow L.A. DiPietro L.A. Elner V.M. Elner S.G. Strieter R.M. Science. 1992; 258: 1798-1801Crossref PubMed Scopus (1904) Google Scholar), and may be attributed to the heterogeneity that is known to exist among preparations of HUVEC cultures (48Watson C.A. Camera-Benson L. Palmer-Crocker R. Pober J.S. Science. 1995; 268: 447-448Crossref PubMed Scopus (49) Google Scholar, 49Koch A.E. Polverini P.J. Kunkel S.L. Harlow L.A. DiPietro L.A. Elner V.M. Elner S.G. Strieter R.M. Science. 1995; 268: 448Crossref PubMed Google Scholar). Taken together, these observations have obvious biological significance and may imply a role for SDF-1α in re-endothelialization after injury, an event that requires the directed migration of EC. Since CXCR4 expression is sharply down-regulated by IFN-γ, the ability of EC to migrate in response to SDF-1α was studied to examine the functional consequences of altered gene transcription. Treatment of EC for 24 h with IFN-γ (103 units/ml) produced a significant decrease (>60%) in the number of EC migrating in response to a SDF-1α gradient (Fig. 7 B). This correlates well with the transcriptional down-regulation of CXCR4 observed after treatment with IFN-γ. The data from the present study suggest that chemokines and their receptors, especially SDF-1α and CXCR4, may play an important role in the etiology of the EC response during vascular disease, inflammation, and infection. Constitutively expressed SDF-1α and CXCR4 may also be involved in the basal recruitment and diapedesis of monocytes and T-lymphocytes that is observed in the early fatty streaks in infant children (50Stary H.C. Atherosclerosis. 1987; 64: 91-108Abstract Full Text PDF PubMed Scopus (178) Google Scholar), and in animal models of atherogenesis (51Gerrity R. Am. J. Pathol. 1981; 103: 181-190PubMed Google Scholar). It is unlikely that inducible chemokines like MCP-1, which are usually not expressed in normal arteries (52Nelken N.A. Coughlin S.R. Gordon D. Wilcox J.N. J. Clin. Invest. 1991; 88: 1121-1127Crossref PubMed Google Scholar), are responsible for the initial recruitment of these cells. Apart from this, CXCR4 is a co-receptor for the infection of T-lymphotropic strains of HIV-1 in CD4-positive susceptible cells (26Feng Y. Broder C.C. Kennedy P.E. Berger E.A. Science. 1996; 272: 872-877Crossref PubMed Scopus (3636) Google Scholar). It is noteworthy that many cell lines that do not express CD4, including EC, are also susceptible to infection with various strains of HIV (32Corbeil J. Evans L.A. McQueen P.W. Vasak E. Edward P.D. Richman D.D. Penny R. Cooper D.A. Immunol. Cell Biol. 1995; 73: 140-145Crossref PubMed Google Scholar, 53Moses A.V. Bloom F.E. Pauza C.D. Nelson J.A. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 10474-10478Crossref PubMed Scopus (200) Google Scholar) by a CD4-independent mechanism. Although the chemokine receptor preferences of these HIV-1 strains have not been studied yet, it may involve the use of CXCR4 as their primary receptor. Indeed, some isolates of HIV-2 have been shown to use CXCR4 as an alternative receptor to infect CD4-negative B and T lymphocyte lines like Daudi, Nalm6, and BC7 (33Endres M.J. Clapham P.R. Marsh M. Ahuja M. Turner J.D. McKnight A. Thomas J.F. Stobenau-Haggarty B. Choe S. Vance P.J. Wells T.N.C. Power C.A. Sutterwala S.S. Doms R.W. Landau N.R. Hoxie J.A. Cell. 1996; 87: 745-756Abstract Full Text Full Text PDF PubMed Scopus (587) Google Scholar). It is thus no coincidence that IFN-γ, which down-regulates CXCR4 expression in EC, should be clinically tested as a prophylactic in advanced HIV infections and disease progression (54Murray H.W. Intensive Care Med. 1996; 22: S456-S461Crossref PubMed Google Scholar), a situation that coincides with the appearance of CXCR4-dependent lymphotropic strains of HIV in infected individuals. Further studies, aided by the development of anti-CXCR4 antibodies and antagonists will be needed to more clearly understand the pathophysiological role of CXCR4. We thank Dr. J. Hoxie at the University of Pennsylvania (Philadelphia, PA) for the gift of anti-CXCR4 monoclonal antibody 12G5.