Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Molecular mimicry wikipedia , lookup
Polyclonal B cell response wikipedia , lookup
Adaptive immune system wikipedia , lookup
Lymphopoiesis wikipedia , lookup
Immunosuppressive drug wikipedia , lookup
Cancer immunotherapy wikipedia , lookup
Innate immune system wikipedia , lookup
From www.bloodjournal.org by guest on June 15, 2017. For personal use only. Regulation of Granulocyte-Macrophage Colony-Stimulating Factor and E-Selectin Expression in Endothelial Cells by Cyclosporin A and the T-cell Transcription Factor NFAT By Gillian W. Cockerill, Andrew G. Bert, Gregory R. Ryan, Jennifer R. Gamble, Mathew A. Vadas, and Peter N. Cockerill Nuclear factor ofactivated T cells (NFAT)was originally describedasaT-cell-specifictranscriptionfactor that supported the activation of cytokine gene expression and mediated the immunoregulatory effects of cyclosporin A ICsA). As we observed that activated endothelial cellsalsoexpressed NFAT, we tested the antiinflammatory properties of CsAin endothelialcells. Significantly,CsA completely suppressedthe induction ofNFAT in endothelialcells and inhibited the activity of granulocyte-macrophage colony-stimulating factor (GM-CSF) gene regulatory elements that use NFAT by 60%. CsA similarly mediated a reduction of up to 65% in GM-CSF mRNA and protein expression in activated endothelial cells.CsAalsosuppressedE-selectin, but not vascularcelladhesion molecule-l (VCAM-1) expression in endothelial cells, even thoughthe E-selectin promoter is activated by NF-KB rather than NFAT. Hence, induction of cell surface expression of this leukocyte adhesion molecule by tumor necrosis factor ITNF1-a was reduced by 40% in the presence of CsA, and this was reflected by a 29% decrease in neutrophil adhesion. The effects ofCsA on endothelial cells were also detected at the chromatin structure level,as DNase1 hypersensitivesites within both the GM-CSF enhancer andthe E-selectinpromoter were suppressedby CSA. This represents the first repot? of NFAT in endothelial cells and suggests mechanisms by which CsA could function as an antiinflammatory agent. 0 1995 by The American Societyof Hematology. T hematopoietic compartment where immune effector cells originate and tissues where primary immune responses occur, and endothelial cells play a direct role in governing the traffic of leukocytes across the e n d o t h e l i ~ m . On ~ ~ .activa~~ tion, endothelial cells produce numerous cytokines, including GM-CSF, which activate immune response effector cells such as neutrophils, monocytes, basophils, and eosinophils.2,222,25Furthermore, GM-CSF may be one of the mediators of acute and chronic inflammatory conditions such as rheumatoid arthritis, asthma, and GM-CSF is typically induced in endothelial cells by proinflammatory cytokines such as tumor necrosis factor-a (TNF-a)and IL1 and its expression is regulated at both transcriptional and posttranscriptional le~els.2~ Agents that activate the protein kinase C and Ca2+pathways in endothelial cells would also be expected to induce GM-CSF expression. In addition to producing cytokines, activated endothelial cells also express the cell surface adhesion molecules Eselectin, P-selectin, vascular cell adhesion molecule-l (VCAM- l), and intercellular adhesion molecule-l (ICAMl), which mediate recruitment and transmigration of leukocytes across the e n d ~ t h e l i u m . ’ ~ E-selectin ~ ~ ~ ” ~ (previously termed ELAM-l) is a cell surface glycoprotein found exclu- HE REGULATION of cytokine gene expression is cen- tral to the regulation of the immune response.’,2The T cell has been the most intensively studied model system for investigating cytokine gene regulation, as a wide variety of cytokines are induced on activation of the T-cell receptor.’ The transcription factor nuclear factor of activated T cells (NFAT)’.3-7plays a key role in the activation of many of these genes and is itself induced on T-cell receptor activation. The induction of cytokines such as interleukin-2 (IL-2) and granulocyte-macrophage colony-stimulating factor (GMCSF) in T cells can be suppressed by cyclosporin A (CsA), and these effects are largely caused by the ability of CsA to inhibit activation of NFAT.’,5*6,s,9 Furthermore, the immunosuppressive actions of CsA have been widely attributed to its ability to suppress T-cell expression of cytokines such as IL-2, and hence T-cell activation and proliferation. NFAT, initially thought to be a T-cell specific factor, typically consists of a complex of AP-1 family proteins and NFATp or NFATc, which are respectively induced via protein kinase C and Ca2+signals originating from T-cell receptor activation.*-” NFATp and NFATc are related proteins4.” that are expressed in the cytoplasm of T cells and translocate to the nucleus in response to the Ca2’ activation of the CsAinhibitible phosphatase calcine~rin.~.’~.’~’~ Hence, NFATp and NFATc are major targets for the actions of CsA in T cells. The genes for NFATp and NFATc have recently been c l ~ n e d ~ . ’and ~ , ’ it~ is now believed that NFATc is largely restricted to lymphoid tissues, while NFATp expression is more widespread. NFAT contributes to the activation of the GM-CSF locus in T cells by interacting with four strong binding sites in an upstream and two weak binding sites in the GMCSF However, GM-CSF is also expressed at sites of inflammation in many cell types other than T cells.z~22 This raises the possibility that induction of GM-CSF expression in these cell types could also require NFAT. Consequently, the GM-CSF locus may also be a target for CsA in cells such as activated endothelial cells that express GMCSF. The endothelial cell represents the interface between the Blood, Vol 86, No 7 (October l ) , 1995: pp 2689-2698 From the Division of Human Immunology, Hanson Centre For Cancer Research, lastitute for Medical and Veterinary Science, Adelaide, Australia. Submitted December 27, 1994; accepted June 2, 1995. Supported by the National Health and Medical Research Council of Australia and the Anti-Cancer Foundation of the Universities of South Australia. Address reprint requests to Peter N. Cockerill, Division of H m n Immunology, HansonCentrefor Cancer Research, Institute for Medical and Veterinary Science, PO Box 14, Rundle Mall, Frome Rd, Adelaide 5000, Australia. The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. section 1734 solely to indicate this fact. 0 I995 by The American Society of Hematology. 0ooS-497I/95/8607-0018$3.00/0 2689 From www.bloodjournal.org by guest on June 15, 2017. For personal use only. 2690 COCKERILL ET AL sively on endothelial cells that interacts with receptors on a subset of leukocytes that include neutrophils, eosinophils, monocytes, and a subset of memory T cells.24E-selectin is involved in the earliest stages of leukocyte recruitment as it mediates the initial tethering and rolling of leukocytes across the end~thelium.'~ Acute and chronic inflammatory conditions are typically maintained by the activation of both cytokines and adhesion molecules that function in concert to orchestrate the interactions between numerous participating cell types. Elevated expression of E-selectin, for example, is associated with inflammatory conditions such as a ~ t h m a , rheumatoid ~' arthritis,32 and psoriasis.33While CsA is more commonly used as an immunosuppressive agent to prevent organ transplant rejection, it has also been found to be effective in the treatment of a variety of chronic inflammatory and autoimmune disorders such as psoriasis," asthma,27ulcerative ~olitis,'~ and rheumatoid arthriti~.~'While some of the benefits of these treatments can be attributed to actions against T cells, there may be additional direct effects on either the activation or trafficking of cell types such as neutrophils, eosinophils, and monocytes. To gain a better understanding of the involvement of the endothelium in inflammation, we have investigated the regulation of the GM-CSF and E-selectin genes in endothelial cells. For this purpose we have used passagedhuman umbilical vein endothelial cells (HUVEs) and the recently derived endothelial cell line C11STH.36We initially cloned Cl lSTH cells from a spontaneous transformant that arose in a culture of HUVEs and showed that they have a repertoire of adhesion molecules and TNF-a receptors similar to that found in normal H U V E S . ~ ~ In this study we found that, as in T cells, the GM-CSF enhancer was required for efficient activation of the GMCSF promoter in endothelial cells. Interestingly, these effects appeared to be mediated in part by the transcription factor NFAT. Furthermore, NFAT and GM-CSF induction in endothelial cells was suppressed by CsA. The induction of Eselectin in endothelial cells and acquired ability to support neutrophil adhesion were also inhibited by CsA, suggesting a wider role for this drug in combating inflammation. In the course of these studies we established that the endothelial cell line C1 lSTH36is a valuable model system for studying inflammatory events in the endothelium. MATERIALS AND METHODS Reagents. The phorbol ester phorbol 12-myristate 13-acetate (PMA) was purchased from Sigma (St Louis, MO), the calcium ionophore A23 187 (I) was purchased from Boehringer (Mannheim, Germany), forskolin was purchased from Calbiochem (San Diego, CA), cyclosporin A (CsA) was a gift from Sandoz (Basel, Switzerland), DNase1 was purchased from Worthington (Freehold, NJ), and TNF-a and interferon-? (IFN-y) were provided by Genetech (San Francisco, CA). PMA was prepared as a I-mg/mL stock in dimethyl sulfoxide (DMSO), A23187 as a 20-mmoUL stock in DMSO, forskolin as a IO-mmoVL stock in DMSO, and CsA as a l-mmoVL stock in ethanol. Oligonucleotides. Oligonucleotide duplexes used as probes and competitors had the following sequences, with complementary single stranded regions at each end shown in lower case for the upper strand only: GM430, gatcTCACACATCTTTCTCATGGAAAGATGA; IL-2 NFAT, gatcCGAAAGGAGGAAAAACTGI7-l- CATACAGAAG;X,TCGCCAATGAGCTCCCGGGTCGACTGCAGAAGCTTC; IgK NF-KB, AACAGAGGGGACTTTCCGAGGCCATCT; E-selectin NF-KB, aattCGTTTTTGGATGCCATGGGGATTTCCTCTTTACTGGATGTG. The GM430 sequence is a highaffinity NFATp/c binding site located in the GM-CSF enhan~er.'.~ Plasmids. The luciferase reporter gene plasmid pGMluc3' had a 655-bp fragment of the human GM-CSF promoter (-627 to +28) upstream of the luciferase gene. The plasmid pGMEluc has a 716bp BglII fragment of the human GM-CSF enhancer placed upstream of the GM-CSF promoter in the BglII site of pGMluc. Cell culture. C1 lSTH cells and HUVEs were cultured as previously described." Antibodies. The antibody R59 was raised against recombinant truncated NFATp prepared from pNFATpXS.'' The antibody 67.5 was raised against peptide 72 of purified mouse NFATp." These antibodies were a gift from A. Rao" and cross-react between human and mouse NFATp. The oct-l antibody was a gift from R. Stunn (Centre for Molecular Biology and Biotechnology, Brisbane, Australia), Gel electrophoretic mobility shift assays. Gel shift assays of nuclear extracts were essentially as previously described* and used 0.2 ngof "P-labeled probe, 5 pg of nuclear protein, and 2 pg of poly(d1-dC) in a 15-pL vol. Nuclear extracts were prepared as previously described' from unstimulated cells and cells stimulated for 2 to 3 hours with 20 ng/mL PMA and 2 pmol/L A23187 (I) in the presence and absence of 0.1 pmoVL CsA. CsA was added 10 minutes before the addition of PMA and A23 187. Assays of recombinant proteins differed in that theyused 200 ng poly(d1-dC), 4 mmoVL dithiothreitol and 0.3 mg/rnL bovine serum albumin together with 0.5 ng NFATp, 10 ng NF-KB p50, or 0.07 pL NF-KB p65. The NFATp was a 293-amino acid truncated derivative of NFATp containing the DNA-binding domain and was prepared from the plasmidpQE-31#1 (a gift from A. Rao), which is derived from pNFATpXS,'* by expression in Escherichia coli and chromatography on Quiagen Niz+ NTA agarose according to manufacturer's instructions. NF-KB p50 was purchased from Promega (Madison, WI), while the NF-KBp65 used here was a gift of purified recombinant truncated protein (amino acids 1-313) supplied by S. Gerondakis (Walter and Eliza Hall Institute, Melbourne, Australia).zR GM-CSF enzyme-linked immunosorbent assay. GM-CSF protein concentrations in cell culture supernatants were determined according to manufacturer's instructions using a human GM-CSF enzyme-linked immunosorbent assay (ELISA) kit supplied by R & D Systems (Minneapolis, MN). Supernatants were harvested from cultures containing 2 X IO5 cells seeded in 24-well trays with 400 pL medium. Cultures were initially incubated for 24 hours and then activated for 16 hours as indicated. GM-CSF mRNA assays. GM-CSF mRNA levels were assayed as previously described' after stimulation of cells for 8 hours with 20 ng/mL PMAand 2 pmol/L A23187 (I) in the presence and absence of 0.1 pnol/L CsA. Briefly,RNAwas prepared by the guanidine lysis method and assayed by RNase protection assay. Each hybridization reaction contained probes for both the GM-CSF and the constitutively expressed glyceraldehyde-3-phosphatedehydrogenase (GAPDH) genes, and the levels of GM-CSF expression were expressed relative to GAPDH. DNase I hypersensitive site analysis. DNase I hypersensitive sites (DH) sites in C1 lSTH cells, HUVEs, Jurkat cells, and peripheral blood T lymphocytes were assayed as described*after stimulating the cells for 4 to 6 hours with 20 ng/mL PMA and 2 pmol/L A23187 (I) or 100 U/mL TNF-a in the presence and absence of 0.1 pmoVL CsA. Filters were first hybridized with the GM-CSF probe? From www.bloodjournal.org by guest on June 15, 2017. For personal use only. GM-CSF AND E-SELECTINGENEEXPRESSION then stripped by washing in 50% formamide at 65"C, before rehybridizing with the E-selectin probe. The E-selectin probe was a 1.7kb BgnYEcoRI fragment normally located 1.3 to 3 kb downstream of the human E-selectin gene transcription start site, and was a gift from T. Collins (Brigham and Women's Hospital, Boston, MA). T lymphocytes were freshly purified from peripheral blood by Ficoll to remove centrifugationtoisolatemononuclearcells,elutriation monocytes, and nylon wool chromatography to remove B cells. Transfection assays. Luciferase reporter gene activities were determined as previously describedby transfecting 20 pg of each plasmid construct into 1 X lo6 C1 lSTH cells by the diethylaminoethyl (DEAE) dextran procedure36 and measuring luciferase expression" after stimulation with 2 pmoK A23 187 (I) and 20 ng/mL PMA for 9 hours in the presence and absence of 0.1 pmoK CsA. Neutrophiladhesion assays. Neutrophiladhesionassayswere Briefly, confluent monolayers performed as previously de~cribed.~~ of CllSTH cellswereincubated with 100 UlmL TNF-a in the presence or absenceof 1 pmoK CsA.After 4 hours' activation, "Cr-labeled neutrophils (9 X lo5 cpm)wereaddedandincubated atroomtemperature(approximately 22T) for 15 minutes.After severalwashes with medium,neutrophiladherencewasestimated by measuring bound 5'Cr. E-selectin and VCAM-I expression. Cellsurfaceexpressionof E-selectin and VCAM-I on C l ISTH cells was determined by flow cytometric (FACS) analysis using specific antibodies for E-selectin and VCAM-l, as previously described.36 Before analysis, cells were stimulated for 4 hours with different combinations 100 U/mL I F N 7 , 100 UlmL TNF-a, 20 ng/mL PMA, 2 pmovL A23187 (I), and 10 pmom forskolin. RESULTS Activated endothelial cells express NFAT. CsA is a potent immunosuppressive drug known to function in T cells by inhibiting induction of the transcription factor NFAT. Although there have been no previous reports of NFAT in endothelial cells, CsA is used as an antiinflammatory agent, and some of its actions could be mediated by suppression of factors in the endothelium. Therefore, we have examined whether NFAT can be induced in endothelial cells by pathways known to activate NFAT in T cells. To this end, we stimulated C1 lSTH cells and passaged HUVEs with a combination of the phorbol ester PMA and the calcium ionophore A23187. These signals mimic the phospholipase C driven activation of protein kinase C and increase in cytosolic Ca2+ and induce the AP-I and NFATpk components of NFAT, respectively. We performed gel electrophoretic mobility shift assays of nuclear extracts prepared from activated HUVEs and C1 ISTH cells using the GM-CSF enhancer GM430 sequence as a probe for NFATpk binding (lefthand panel in Fig 1) and the human IL-2 gene distal NFAT site as a probe for NFAT (righthand panel in Fig 1). The IL-2 site served as the prototypic NFAT site most commonly used to assay for the presence of NFAT in T cells as it cooperatively binds AP- 1 and NFATpIc toform NFAT complexe~.~ The GM430 site is a high-affinity NFATp site that is able to associate strongly with the NFATp/c component of NFAT independently of AP- 1.9 Significantly, nuclear extracts from HUVEs and C l ISTH cells both formed inducible NFATpk-like complexes with the GM430 element and NFAT-like complexes with the IL- 269 1 2 NFAT site that comigrated with Jurkat T-cell complexes in gel mobility shift assays (Fig 1). ASin T cells, CsA completely inhibited the induction of these endothelial cell NFAT and NFATpk-like complexes (Fig 1). Furthermore, complex formation with both probes was completely inhibited when either the IL-2 or GM430 NFAT elements were included as DNA competitors, but only partially inhibited when an unrelated competitor (X) was used. NFATp appeared to be the major protein contributing to NFATpk-like complexes with the GM430 probe in C 11STH and HUVE extracts as antibodies raised against either recombinant NFATp or toan NFATp peptide specifically inhibited formation of or super-shifted NFATp-like complexes as efficiently as they did withJurkat cell extracts (Fig 2). Addition of a 5- to 20-fold greater amount of oct-l antiserum had little effect on NFATp complex formation with the endothelial cell nuclear extracts. As the 67.1 NFATp antibody was raised against a peptide not conserved in NFATc, there appeared to be little or no NFATc present in either the endothelial cell or the Jurkat cell extracts employed in this study. NFATp induction in endothelial cells may require a Ca2+ signal not provided by proinflammatory cytokines, as TNFa did not induce NFATp (data not shown). Incontrast, TNFa was a more efficient inducer of NF-KB in endothelial cells than PMA and A23187 (data not shown). GM-CSF expression is regulated by CsA in endothelial cells. If NFATp regulates gene expression in endothelial cells then GM-CSF gene activation via Ca2+ pathways in endothelial cells should also be CsA-sensitive. To explore this possibility, we examined GM-CSF synthesis in four independent lines of passaged HUVEs. Asin T cells, the level of GM-CSF secretion by HUVEs was increased 100fold by the combined actions of the phorbol ester PMA and the calcium ionophore A23 187 (P", Fig 3). For each of the four lines of activated HUVEs, there was a decrease in GM-CSF expression in the presence of 0.1 pmoUL CsA, with an average reduction of 50%. As we anticipated that CsA would inhibit GM-CSF at the transcriptional level, we also performed RNase protection assays of GM-CSF mRNA in HUVEs and CllSTH cells. Just as we observed for GM-CSF protein synthesis, GMCSF mRNAwas strongly induced in both HUVEs and CllSTH cells byPMA and A23187 (Fig 4). GM-CSF mRNA induction required calcium ionophore as well PMA, as sixfold less GM-CSF mRNA was detected in the presence of PMA alone. In the presence of 0.1 pmoVL CsA the induction of GM-CSF mRNA was reduced by 65% in HUVEs and 56% in C1 lSTH cells (Fig 4).This inhibition appeared to be specific, as we observed no effect of CsA on K-8 expression in endothelial cells (data not shown). CsA also appeared to have little effect on induction of GM-CSF expression by pathways that do not use Ca2+.Hence, CsA had no significant influence on induction of GM-CSF mRNA or protein expression in endothelial cells activated with TNFa (data not shown). The GM-CSF enhancer exists asa DNaseI hypersensitive site in endothelial cells. In T cells, the GM-CSF enhancer exists as an inducible CsA-sensitive DNaseI hypersensitive (DH) site that spans at least two NFAT site^.^'^ To identify From www.bloodjournal.org by guest on June 15, 2017. For personal use only. COCKERILL ET AL 2692 I HUVE pmbe GM430 NFATp c-ll m I pmbeL 2 NFAT C-l 1 STH a role for this enhancer in the GM-CSF locus in endothelial cells, we examined DH sites upstream of the GM-CSF gene. These analyses were performed both in Cl ISTH cells (Fig 5A) and in passaged HUVEs (Fig 5B). We located two DH sites that coincided with the proximal promoter and the previously defined enhancer 3 kb upstream of the gene. The DH site located within the promoter ap- c-l1 sm U Fig 1. Gel electrophetic mobility shift assays of NFATp and NFAT in endothelial cellnuclear extracts. Nuclear extractswere prepared from passaged HUVEs, CllSTH cells, and Jurkat cells that were either unstimulated (nil) or stimulated 2for t o 3 hours with 20 ng/mL PMA and 2 pmol/L A23187 in the presence (PMAIIICsA) or absence (PMA/I) of 0.1 pmol/L CsA. Bands migrating below theNFATp and NFAT complexes appear t o represent nonspecific constitutive factors. Where indicated, assays of stimulated CllSTH cell nuclear extract binding t o the GM430 NFATp and IL-2 NFAT probes also include 25 ng duplex oligonucleotides corresponding t o the GM430 NFATp site, the IL-2 NFAT site, or an unrelated sequence (X) as competitor. Each assay used 5 p g protein, and the asterisk indicates one lane where a shorter autoradiographic exposure was included t o assist identification of the NFATp complex in the Jurkat extract. .d Jurkat Tcells - HUVE 1 + NFATp Fig 2. Gel electrophoretic mobility shift assays of nuclear extracts binding t o the GM430 NFATp probe in the presence of NFATp antibodies. Assays used 1 p g stimulated Jurkat cellnuclear extract or 5 p g stimulated CllSTH or HUVE nuclear extract. Extracts were assayed either with no added anti-sera (nil) or after preincubation for 20 minutes on ice with 0.2 p L of preimmune serum, 0.2 p L R59 antisera raised against recombinant NFATp, 0.05 p L 67.1 antisera raised against an NFATp peptide, or 1p L of an antisera raised against oct-l. The 67.1-antisera produced super-shifted complexes characteristic ofspecific NFATp binding and does not cross-react with NFATc. peared as a constitutive site in both Cl I STH cells (Fig 5A) and HUVEs (Fig 5B). Althoughnot detected in Jurkat T cells, this site also appeared as a DH site in activated T lymphocytes (Fig 5B). The DH site within the enhancer appeared as a weak site before activation and was strongly induced in the presence of the coactivators PMA andA23 I87 in both C1 ISTH cells and HUVEs. The DH site existed as a broad region 2.8 to 3.0 kb upstream of the GM-CSF gene, thus resembling the DH site induced in T cells. Significantly, induction of this DH site was reduced by about 70% in the presence of 0.1 pmol/L CsA, which had no effect on the promoter DH site (Fig 5A). We also examined endothelial cells activated by TNF-a, but observed that this cytokine was not a significant inducer ofany DH sites within the GM-CSF locus (data not shown). The GM-CSF enhancercontributes to GM-CSF activation in endothelial cells. To determine whether the upstream enhancer was required for efficient GM-CSF promoter activation in endothelial cells, the enhancer was coupled to the GM-CSF promoter in a luciferase reporter gene plasmid. On transfection into C1 ISTH cells and stimulation with the coactivators PMA and A23187, the enhancer supported a fivefold greater induction of luciferase gene expression than was obtained with a luciferase gene driven by the GM-CSF promoter alone (Fig 6). Furthermore, this induction was suppressed by about 60% in the presence of CsA. CsA inhibits neutrophil adhesion to activated endothelial cells. Having found that CsA could suppress cytokine gene expression, we next examined whether CsA could influence neutrophil adhesion to activated endothelial cells. Neutrophils become involved at the earliest stages during the progression of inflammation and are also one of the cell types activated by GM-CSF. In the presence of TNF-a, we observed an approximately sixfold increase in the adherence of neutrophils to a confluent monolayer of C1 ISTH cells (Table l ) . In the presence of CsA, we observed a 29% decrease (P = .0006) in the TNFa-induced neutrophil adherence. From www.bloodjournal.org by guest on June 15, 2017. For personal use only. 2693 GM-CSF AND E-SELECTINGENEEXPRESSION T NIL PI PIC Fig 3. Suppression of GM-CSF protein synthesisin HUVEs b y CsA. Levels of GM-CSF secretion by cultured HUVEs were measured by ELISA. Four independent lines of HUVEs were either not stimulated (NIL) or stimulated for 16 hours with 20 nglmL PMA and 2 p m o l l L A23187 in the presence (PIC) or absence(PI) of 0.1 pmol/L CsA. Shown above are the mean expression levels obtained with the four lines of HUVEs together with error bars indicating the SEM. There was a statistically significant meandecrease of 320 pglmL GM-CSF (SEM = 132, P < .01)in the four lines of HUVEs in the presence of CsA. E-selectin expression is suppressed by CsA in endothelial cells. To determine which leukocyte adhesion molecules on endothelial cells might be suppressed by CsA, we examined the regulation of E-selectin and VCAM-I expression in C1 lSTH cells. E-selectin can interact with receptors on neutrophils, eosinophils, monocytes, and subsets of T cells and is an important regulator of both acute and chronic inflammation.23.24 VCAM-l plays a major role in the adherence of monocytes and lymphocytes to endothelial cells.23 We first showed that E-selectin was induced in C1 lSTH cells in the expected fashion by either TNF-a or a combination of PMA and A23 187 (PMAII) (Fig 7). IFN-y alone was unable to induce E-selectin expression, but greatly enhanced its activation by TNF-a. We also observed that the CAMP inducer forskolin was a potent inhibitor of TNF-a-inducible E-selectin expression in C l lSTH cells, as previously observed by others in HUVES.’~ On inclusion of increasing amounts of CsA, we observed up to a 40% decrease in the cell surface expression of Eselectin expression in C l l STH cells activated by either TNFa, TNF-a plus I F N , or PMA and A23 187 (Fig 7). CsA reduced the expression obtained with PMA and A23187 to a level equivalent to that obtained with PMA alone. Furthermore, activation of E-selectin expression in the presence of PMA alone was essentially resistant to the effects of CsA. Interestingly, CsA also did not suppress the component of TNF-a-inducible E-selectin expression that was resistant to forskolin. The effects of CsA on E-selectin expression appeared to be specific, as CsA had no effect on the induction of VCAM-l expression by any of the above agents (data not shown). A CsA-sensitive DNaseI hyper-sensitivesite exists in the E-selectin promoter. To search for regulatory elements that might account for the CsA-sensitivity of E-selectin gene expression, we screened the E-selectin locus for CsA-sensitive DH sites. For this purpose we reprobed the filter used in Fig 5 with an intragenic E-selectin probe. DNaseI digestion of the E-selectin chromosomal locus produced additional 3.0- and 6.5-kb subfragments of the 9kb EcoRI fragment, indicating that two DH sites existed upstream of the E-selectin gene in activated endothelial cells (Fig 8). One DH site near the origin of transcription was induced by either TNF-a or a combination ofPMA and A23187 in both Cl ISTH cells (Fig 8A) and HUVEs (Fig 8B) and has also recently been detected byothers:’ The second DH site, located 3.5kb upstream, appeared as a constitutive DH site. Significantly, the inducible but not the constitutive site, was suppressed by CsA (Fig 8A), suggesting that CsA-sensitive transcriptional activation ele- c11-sTH HUVE ” U U U” U” M 0- \ Y 4- GM-CSF 4-w -GAPDH 0.Fig 4. Suppression of GM-CSF mRNA expression in endothelial cells by CsA. RNA was isolated from HUVEs and CllSTH cells that were either unstimulated (Nil),stimulated for 8 hours with 20 n g l mL PMA (PMA), or stimulated for 8 hours with 20 nglmL PMA and 2 pmollL A23187 in the presence (PMAlIICsA) or absence IPMAlI) of 0.1 pmollL CsA. GM-CSF mRNA levels were assayed by RNase protection assay, using GAPDH as an internal control. From www.bloodjournal.org by guest on June 15, 2017. For personal use only. COCKERILLET AL 2694 A B rj GM-CSF PROMOTER + ENHANCER U \ z a + + z \ a 111, n b L 3 zE U X l X - E - Probe Enhancer GM-CSF Fig 5. Mapping ofDH sites from an EcoRl site 6.8 kb upstream of the GM-CSF gene in endothelial cells. Cells were either unstimulated (Nil) or stimulated for 4 t o 6 hours with 20 nglmL PMA and 2 p m o l l L A23187 in the presence (PIIlCsA) or absence (PIU of 0.1 pmolIL CsA. (A) shows DH sites in CllSTH cells. Note that thesesites are not detected in the absence of DNase1 digestion (control DNA). P represents a DH site in the proximal region of the promoter, whereas E represents a DH site located in the center of the enhancer 2.8 t o 3.0 kb upstream of the transcription startsite. ments existed in the proximal promoter. Both DH sites appeared to be endothelial cell specific as they did not appear in activated Jurkat cells or T cells (Fig 8B) or in a wide range of other cell types examined (data not shown). CsA may suppress NF-KB activity in the E-selectin promoter. We examined the sequence of the E-selectin promoter seeking transcription factor binding sites that might be involved in DH site formation and be suppressed by CsA. Previous studies of E-selectin gene regulation have implicated an array of three NF-KB sites in the promoter as being crucial for activation of E-selectin expression,"0d' and these sites may be the most likely targets for the actions of CsA in this locus. Indeed, there is some evidence that NFKB p50 and c-re1 expression is inhibitible by either CsA or FK506T" which might account for the 40% inhibition of E-selectin expression that occurs in the presence of CsA. Nevertheless, there is also evidence obtained with T cells that NFAT-like factors can in some cases associate with NF&-like regulatory regions." To determine whether NFAT or NF-KB proteins are more likely to account for the CsA-sensitivity of E-selectin expression, we examined binding of recombinant NFATp, NF-KB GM-CSF PROMOTER T NIL PMA PMA I I CsA NIL PMA PMA I I CsA Fig 6. Activationof GM-CSF enhancerlpromoter luciferase reporter gene plasmids in endothelial cells. Luciferase gene plasmids containing either the GM-CSF promoter (pGMlucl or the GM-CSF enhancer linked to the promoter (pGMEluc) were transfected into C l lSTH cells and stimulated for 9 hours with 20 nglmL PMA and 2 pmol/L A23187 in the presence (PMAlIlCsA) or absence (PMAII) of 0.1 pmollL CsA. Luciferase activities wereexpressed as a percentage of thevalues obtained with stimulated pGME and represent the mean of nine experiments. Error bars represent the standard deviation. p50, and NF-KB p65 to the best characterized NF-KB site in the E-selectin promoter (PD14'), and to three control NFATp and NF-KB binding sites (Fig 9). This particular E-selectin NF-KB site was the most proximal of the three NF-KB sites in the promoter and the one found to be the most essential for E-selectin promoter activity. As this element encompasses the sequence AGAGGAAA, it is also the E-selectin NF-KB site that most resembles binding sites for NFATP.'.~ As anticipated from earlier studies of T-cell nuclear extracts,"both the IL-2 NFAT site and the IgK NF-KB site had the capacity to associate withall three proteins. This confirmed the view that NF-KB and NFATp have the poten- Table 1. Suppression of Neutrophil Adhesion to CllSTHCells bv CsA Treatment Nil T N L F ITNF-dCsA Adhesion (cpm x 10") 117 2 7 (n = 7) 795 2 33 (n = 14) 565 2 49 (n = 9) Results are presented as mean 2 SEM. The 29% decreasein neutrophil adhesion in the presence of CsA was statistically significant as a Student's t-test gave P = .0006. From www.bloodjournal.org by guest on June 15, 2017. For personal use only. GENE GM-CSF AND E-SELECTIN 2695 Fig 7. CsA suppresses induction of E-selectin expression in endothelial cells. E-selectin expression was inducedin CllSTH cells as indicated above in the presence of increasing concentrations of CsA, which was added 10 minutes before stimulation of thecells. See Materials and Methods for activationconditions. CsA@W 0 0 0 0.01 0.1 1 tial to cross-compete for binding sites. However, the E-selectin NF-KB site exhibited the weakest relative affinity for NFATp, while efficiently binding both p50 and p65. Hence, although NFATp did have limited potential to bind the Eselectin NF-KB site, it appeared less likely than NF-KB to play a major role in its regulation. In contrast, very little 0 0.01 0.1 1 0 0.01 0.1 1 0 0 . 0 1 0.1 1 00.01 0.1 1 probe, which is a high-affinity NFATp site, and the dominant NFAT site in the GM-CSF enhancer.' DISCUSSION To gainnew insights into processes that regulate inflammation, we focused on the endothelium, which orchestrates NF-KB p50 or p65 binding was detected with the GM430 IG2 NF'AT B A Sa I 3 + + + + & a + + \om Igk E-selectin GM430 NF"kB NF-kB NFATp m 0 aa O A + + vzaa " A 6.5 . 6.5+ + 3.0 + 3.0+ DNase1 -6.0 -3.5 E-selectin gene Y DN$? 0 BgUI EcoRI +1.3 c - +3.0 probe Fig 8. Mapping of DH sites upstream of the E-selectin gene in endothelial cells from anEcoRl site 3.0 k b upstream of the tranxrip tion startsite. Cells were either unstimulated(Nil1 or stimulated for 4 t o 6 hours with 20 ng/mL PMA and 2 pmollL A23187 in the presence (PlIICsA) orabsence (PlI) of0.1 pmollL CsA or with 100 UlmL TNFQ. (AIshows DH sites in CllSTH cells. Notethatthese sites are not detected in the absence of DNase1 digestion (control DNA). The inducible DH site in the proximal region of the promoter appeared as a 3.0-kb fragment, whereas an upstream constitutive DH site appeared as a 6.5-kb fragment. The membranes used in (AI and (B) are the same as those in Fig 3. Fig 9. Gelelectrophoreticmobility shift assays of recombinant NFATp and NF-KB. The indicated probes were incubated with 0.5 ng NFATp, 10 ng NF-wBp50, or 0.07 p L NF-KB p65. The probes corresponded t o t h eIL-2 promoter distal NFAT site," the K lg gene NF-KB site." the E-selectin promoter proximal NF-KBsite,u and the GM430 high-affinity NFATp the from site GM-CSF enhancer.' From www.bloodjournal.org by guest on June 15, 2017. For personal use only. 2696 many early inflammatory events, and CsA, which is in widespread use as an immunosuppressive and antiinflammatory drug. Two important classes of molecules that are activated during the early stages of inflammation are the cytokines and adhesion molecules that direct the activation and trafficking of leukocytes that enter inflammatory lesions. We have concentrated on GM-CSF and E-selectin as potential targets for antiinflammatory actions of CsA in endothelial cells, and these are two molecules that might function in concert on neutrophils and eosinophils. Surprisingly, we discovered that immunoregulatory mechanisms initially thought to be restricted to lymphoid cells also operated in endothelial cells. Hence, the T-cell transcription factor NFAT was induced in endothelial cells by pathways that in T cells are triggered by T-cell receptor and phospholipase C activation. CsA, an immunosuppressant previously thought to function primarily by inhibiting induction of NFAT and cytokine gene expression in T cells, also inhibited activation of NFAT in endothelial cells. The endothelial cell NFAT complexes were found to contain the protein NFATp, which has recently been reported to be expressed in both hematopoietic and nonhematopoietic cells," rather than NFATc, which is thought to be lymphoid-specific. Although there are few previous reports suggesting functions for NFAT outside of the T-cell lineage, a recent study has implicated NFAT as a factor required for TNF-a gene expression in B cells.45Other studies have demonstrated that ligation of surface immunoglobulin results in CsA-sensitive induction of NFAT and activation of NFAT-responsive reporter genes in B cell^.^,^' GM-CSF is a proinflammatory cytokine having the potential to activate neutrophils, monocytes, and eosinophils. Of all the cytokines that can be expressed by endothelial cells, GM-CSF is perhaps the only one strongly inhibited by CsA in T cells. Therefore, we have explored the potential of CsA to act as an antiinflammatory agent working to suppress activation of myeloid cells by GM-CSF. We observed that CsA suppressed the Caz+-mediated activation of GM-CSF protein expression in endothelial cells by 50% and GM-CSF mRNA expression by approximately 60%. To determine the molecular basis for the regulation of GM-CSF expression in endothelial cells, we examined the functions of regulatory elements within the GM-CSF locus. Previous studies in T cells suggested that GM-CSF gene activation is mediated to a large extent by binding of NFAT to the upstream enhan~er,'.~and binding of NFAT, AP-1, and NF-KB to the promoter.''-z0 In this study we found that the upstream enhancer was also required for efficient activation of the GM-CSF promoter in transfected endothelial cells, and that the activity of the promoter and enhancer was suppressed twofold in the presence of CsA. As seen previously in T cells, stimulation with PMA and A23187 led to chromatin structure changes within the enhancer that were suppressed by CsA. These observations may suggest a direct role for NFAT in the induction of the GM-CSF enhancer in endothelial cells. In contrast to T cells, endothelial cells appeared to have only a partial dependence on NFAT for induction of GMCSF expression. Although CsA completely suppressed in- COCKERILL ET AL duction of NFAT, it only partially suppressed the GM-CSF promoter and enhancer and GM-CSF mRNAandprotein expression in endothelial cells. In T cells, CsA essentially eliminates GM-CSF gene expression.' Furthermore, TNF-a was a potent inducer of NF-KB and GM-CSF expression in endothelial cells, but was a poor activator of NFAT and an inefficient inducer of the DH site in the enhancer (data not shown). Therefore, it is likely that TNF-a induces GM-CSF expression byusing factors that activate the promoter in the absence of NFAT. Interestingly, the GM-CSF promoter existed as a constitutive DH site in endothelial cells (Fig 5), but not in T cells' (data not shown). Consequently, the GMCSF promoter may exist in a primed state in cultured endothelial cells and have a reduced dependence on the enhancer for its activation. We also found some evidence suggesting that CsA can inhibit endothelial cell functions by suppressing factors other than NFAT. E-selectin expression was inhibited by about 40% in the presence of CsA by a process unlikely to directly involve NFAT. The E-selectin promoter is activated principally by an array of NF-KB site^,^.^' which represented poor binding sites for NFATp. Furthermore, the suppression of E-selectin by CsA was not limited to activation pathways involving Ca". CsA reduced E-selectin expression by the same extent regardless of whether TNF-(Uor PMNA23187 was used to induce expression. This was in marked contrast to the GM-CSF gene, which was suppressed by CsA when activated by PMNA23187, but not when activated by TNFa. CsA also suppressed induction of the DH site that forms over the array of NF-KB sites in the E-selectin promoter, suggesting that CsA may regulate members of the NF-KB family in endothelial cells. This suggestion is supported by reports that CsA and FK506 can partially suppress two NFKB family members in lymphoid cells.43*"Indeed, the NFKB p105/p50 gene promoter4'mayitself be regulated by CsA, as it includes the sequence TGGACCGCATGACTCTA that closely resembles the NFAT consensus sequence.' Alternatively, CsA may suppress E-selectin expression at the posttranscriptional level, thus accounting for the ability of CsA to suppress distinct pathways tothe same extent. However, E-selectin induction via PMA was not significantly influencedby CsA, arguing against a posttranscriptional mechanism. The evidence accumulated in this study suggests that the ability of CsA to act as an antiinflammatory drug maybe partly caused by direct effects on the endothelium. Although immune and inflammatory disorders clearly involve complex cytokine networks operating within a confederacy of immune and nonimmune cell types, there is a significant role that the endothelium plays in their progression. By suppressing GM-CSF and E-selectin expression in endothelial cells, CsA could reduce the extent to which neutrophils and eosinophils are recruited at sites of inflammation. In this study, the effective dose of CsA (0.1 p r n o a ) waswell within the range of the plasma concentration of approximately 0.3pmol/L CsA that is usually maintained therapeutically. Hence, this study may, in part, account for the reductionin neutrophil and eosinophil infiltration observed on treatment of disorders such as p s o r i a ~ i s ~and ~ * ~asthma27s4' ' with CsA. From www.bloodjournal.org by guest on June 15, 2017. For personal use only. GM-CSF AND E-SELECTINGENEEXPRESSION 2697 ACKNOWLEDGMENT 17. O’Keefe SI, Tamura J, Kincaid RL, Tocci M, O’Neill EA: FK-506- and CsA-sensitive activation of the interleukin-2 promoter by calcineurin. Nature 357:692, 1992 18. Masuda ES, Tokumitsu H, Tsubio A, Shlomai J, Hung P, AraiK, Arai N: The granulocyte-macrophage colony-stimulating factor promoter cis-acting element CLEO mediates induction signals in T cells and is recognised by factors related to API and NFAT. Mol Cell Biol 13:7399, 1993 19. Tsubio A, Muramatsu M, Tsutsumi A, Arai K, Arai N: Calcineurin activates transcription from the GM-CSF promoter in synergy with either protein kinase C or NF-KBIAP-I in T cells. Biochem Biophys Res Commun 199:1064, 1994 20. Jenkins F, Cockerill PN, Bohman D, Shannon MF: Multiple signals are required for function of the human granulocyte-macrophage colony-stimulating factor gene promoter in T cells. J Irnmunol 155:1240, 1995 21. McCaffrey PG, Jain J, Jamieson C, Sen R, Rao A: A T cell nuclear factor resembling NF-AT binds to an NF-KB site andto the conserved lymphokine promoter sequence “cytokine-l.” J Biol Chem 267:1864, 1992 22. Gasson JC: Molecular physiology of granulocyte-macrophage colony-stimulating factor. Blood 77: 1131, 1991 23. Bevilacqua MP: Endothelial-leukocyte adhesion molecules. Annu Rev Immunol 11:767, 1993 24. Springer TA: Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell 76:301, 1994 25. Lopez AF, Williamson DJ, Gamble JR, Begley CG, Harlan JM, Klebanoff SJ, Waltersdorph A, Wong G, Clark SC, Vadas MA: Recombinant human granulocyte-macrophage colony-stimulating factor (rH GM-CSF) stimulates in vitro mature human neutrophil and eosinophil function, surface receptor expression and survival. J Clin Invest 78: 1220, 1986 26. Arend WP, Dayer JM: Cytokines and cytokine inhibitors or antagonists in rheumatoid arthritis. Arthritis Rheum 33:305, 1990 27. Calderon E, Lockey RF, Bukantz SC, Coffey RG, Ledford DK: Is there a role for cyclosporine in asthma? J Allergy Clin Immunol 89:629, 1992 28. Wong RL, Winslow CM, Cooper KD: The mechanisms of action of cyclosporin A in the treatment of psoriasis. Irnmunol Today 14:69, 1993 29. Kaushansky K: Control of granulocyte-macrophage colonystimulating factor production in normal endothelial cells by positive and negative regulatory elements. J Immunol 143:2525, 1989 30. Luscinskas F W , Ding H, Lichtman AH: P-selectin and vascular cell adhesion molecule 1 mediate rolling and arrest, respectively, of CD4+ T lymphocytes on tumour necrosis factor a-activated vascular endothelium under flow. J Exp Med 181:1 179, 1995 31. Montefort S, Gratziou C, Goulding D, Polosa R, Haskard DO, Howarth PH, Holgate ST, Carroll M P Bronchial biopsy evidence for leukocyte infiltration and upregulation and leukocyte-endothelial cell adhesion molecules 6 hours local allergen challenge of sensitised asthmatic airways. J Clin Invest 93:1411, 1994 32. Koch AE, Burrows A E , Haines GK, Carlos TM, Harlan JM: Immunolocalisation of endothelial and leukocyte adhesion molecules in rheumatoid and osteoarthritic synovial tissues. Lab Invest 64:313, 1991 33. Wakita H, Takigawa M: E-selectin and vascular cell adhesion molecule-l are critical for initial trafficking of helper/inducer/memory T cells in psoriatic lesions. Arch Dermatol 130:457, 1994 34. Sartor RB: Cyclosporine therapy for inflammatory bowel disease. N Engl J Med 330:1897, 1994 35. Wells G, Tugwell P: Cyclosporin A in rheumatoid arthritis: Overview of efficacy. Br J Rheumatol 3251, 1993 36. Cockerill GW, Meyer G, Noack L, Vadas MA, Gamble JR: We are indebted to A. Rao for providing NFATp anti-sera and DNA clones for NFATp, Y. Khew-Goodall for the E-selectin NFKB probe, and T. Collins for providing E-selectin DNA clones. We thank R. Himes, F. Jenkins, and F. Shannon for the luciferase reporter gene plasmids. We thank L. Noack for technical assistance with neutrophil adhesion assays and B. Stein for assistance with statistical analyses. We thank D. Gillis, Y. Khew-Goodall, and M.F. Shannon for helpful advice and critical reading of the manuscript. REFERENCES 1. Masuda ES, Naito Y, Arai K, Ami N: Expression of lymphokine genes in T cells. The Immunologist 1:198, 1993 2. Nicola NA: Hemopoietic cell growth factors and their receptors. Annu Rev Biochem 58:45, 1989 3. Mattila PS, Ullman KS, Fiering S , Emmel EA, McCutcheon M, Crabtree GR, Herzenberg LA: The actions of cyclosporin A and FK506 suggest a novel step in the activation of T lymphocytes. EMBO J 9:4425, 1990 4. Nolan GP: NF-AT-AP-1 and Rel-bZIP: hybrid vigor and binding under the influence. Cell 77:795, 1994 5. Rao A: NF-ATP: A transcription factor required for the coordinate induction of several cytokine genes. Immunol Today 15:274, 1994 6. Schreiber SL, Crabtree G R The mechanism of action of cyclosporin A and FK506. Immunol Today 13:136, 1992 7. Shaw JP, Utz PJ, Durand DB, Toole JJ, Emmel EA, Crabtree GR: Identification of a putative regulator of early T cell activation genes. Science 241:202, 1988 8. Cockerill PN, Shannon MF, Bert AG, Ryan GR, Vadas MA: The granulocyte-macrophage colony-stimulating factorhterleukin 3 locus is regulated by an inducible cyclosporin A sensitive enhancer. Roc Natl Acad Sci USA 90:2466, 1993 9. Cockerill PN, Bert AG, Jenkins F, RyanGR, Shannon MF, Vadas MA: Human granulocyte-macrophage colony-stimulating factor enhancer function is associated with cooperative interactions between AP-l and NFATpk. Mol Cell Biol 15:2071, 1995 10. Jain J, McCaffrey PG, Miner Z, Kerpola TK, Lambert TN, Verdine GL, Curran T, RaoA: The T-cell transcription factor NFATp is a substrate for calcineurin and interacts with Fos and Jun. Nature 365:352, 1993 11. Jain J, McCaffrey PG, Valge-Archer VE, Rao A: Nuclear factor of activated T cells contains Fos and Jun. Nature 356:801, 1992 12. McCaffrey PG, Luo C, Kerppola TK, Jain J, Badalian TM, HoAM, Burgeon E, Lane WS, Lambert JS, Curran T, Verdine GL, Rao A, Hogan PG: Isolation of the cyclosporin-sensitive T cell transcription factor NFATp. Science 262:750, 1993 13. Northrop JP, Ho SN. Chen L, Thomas DJ, Timmerman LA, Nolan GP, Admon A, Crabtree GR: NF-AT components define a family of transcription factors targeted in T-cell activation. Nature 369:497, 1994 14. Northrop JP, Ullman KS, Crabtree GR: Characterization of the nuclear and cytoplasmic components of the lymphoid-specific nuclear factor of activated T cells (NF-AT) complex. J Biol Chem 268:2917, 1993 15. Boise LH, Pehyniak B, Mao X,June CH, Wang CY, Lindsten T, Bravo R, Kovary R, Leiden JM, Thompson JB: The NFAT-l DNA binding complex in activated T cells contains Fra-l and JunB. Mol Cell Biol 13:1911, 1993 16. Clipstone NA, Crabtree GR: Identification of calcineurin as a key signalling enzyme in T-lymphocyte activation. Nature 357:695, I992 From www.bloodjournal.org by guest on June 15, 2017. For personal use only. 2698 Characterisation of a spontaneously transformed human endothelial cell line. Lab Invest 71:497, 1994 37. Himes SR. Coles LS, Katsikeros R, Lang RK, Shannon MF: HTLV-I tax activation of the GM-CSF and G-CSF promoters requires the interaction of NF-KB with other transcription factor families. Oncogene 8:3189, 1993 38. Dunn SM, Coles LS, Lang RK, Gerondakis S, Vadas MA, Shannon MF: Requirement for NF-KB p65 and NF-interleukin-6 binding elements in the tumour necrosis factor response region of the granulocyte colony-stimulating factor promoter. Blood 83:2469, 1994 39. Pober JS, Slowik MR, DeLuca LG, Ritchie AJ: Elevated cyclic AMP inhibits endothelial cell synthesis and expression of TNF-induced endothelial leukocyte adhesion molecule-1, andvascular cell adhesion molecule-l, but not intercellular adhesion molecule1. J Immunol 150:.5114, 1993 40. Schindler U, Baichwal VR: Three NF-KBbinding sites in the human E-selectin gene required for maximal tumour necrosis factor alpha-induced expression. Mol Cell Biol 14:5820, 1994 41. Lewis H, Kaszubska W, DeLarmarter JF, Whelan J: Cooperativity between two NF-KB complexes, mediated by high mobility group protein I(Y), is essential for cytokine induced expression of the E-selectin promoter. Mol Cell Biol 145701, 1994 42. Whitely MZ, Thanos D, Read MA, Maniatis T, Collins T: A COCKERILL ET AL strikmg similarity inthe organisation of the E-selectin andbeta interferon gene promoters. Mol Cell Biol 145464, 1994 43. McCaffrey PG, Kim PK, Valge-Archer VE, Sen R, Rao A: Cyclosporin A sensitivity of the NF-KB site of the IL2Ra promoter in untransformed murine T cells. Nucleic Acids Res 22:2134, 1994 44. Venkataraman L, Burakoff SJ, Sen R: FK.506 inhibits antigen receptor-mediated induction of c-re1 in B and T lymphoid cells. J Exp Med 181:1091, 1995 45. Goldfeld AE, Tsai E, Kincaid R, Beelshaw PJ, Schreiber SL, Strominger JL, Rao A: Calcineurin mediates human tumour necrosis factor a gene induction in stimulated T and B cells. J Exp Med 180:763, 1994 46. Choi MSK, Brines RD, Holman MJ, Klaus GGB: Induction of NF-AT in normal B lymphocytes by anti-immunoglobulin or CD40 ligand in conjunction with IL-4. Immunity 1:179, 1994 47. Venkataraman L, Francis DA, Wang Z, Liu J, Rothstein TL, Sen R: Cyclosporin-A sensitive induction of NF-AT in murine B cells. Immunity 1:189, 1994 48. Ten RM, Paya CV, Israel N, Le Bail 0, Mattei MG, Virilizier JL, Kourilsky P, Israel A: The characterization of the promoter of the gene encoding the p50 subunit of NF-KB indicates that it participates in its own regulation. EMBO J 11:195, 1993 49. Noms AA, Jackson DM, Eady RP: Protective effects of cyclophosphamide, cyclosporin A and FK506 against antigen-induced lung eosinophilia in guinea pigs. Clin Exp Immunol 89:347, 1992 From www.bloodjournal.org by guest on June 15, 2017. For personal use only. 1995 86: 2689-2698 Regulation of granulocyte-macrophage colony-stimulating factor and E- selectin expression in endothelial cells by cyclosporin A and the Tcell transcription factor NFAT GW Cockerill, AG Bert, GR Ryan, JR Gamble, MA Vadas and PN Cockerill Updated information and services can be found at: http://www.bloodjournal.org/content/86/7/2689.full.html Articles on similar topics can be found in the following Blood collections Information about reproducing this article in parts or in its entirety may be found online at: http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests Information about ordering reprints may be found online at: http://www.bloodjournal.org/site/misc/rights.xhtml#reprints Information about subscriptions and ASH membership may be found online at: http://www.bloodjournal.org/site/subscriptions/index.xhtml Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036. Copyright 2011 by The American Society of Hematology; all rights reserved.