The CysLT1 Ligand Leukotriene D4 Supports α4β1- and α5β1-Mediated Adhesion and Proliferation of CD34 + Hematopoietic Progenitor Cells This information is current as of October 2, 2021. Andreas M. Boehmler, Adriana Drost, Lena Jaggy, Gabriele Seitz, Tina Wiesner, Claudio Denzlinger, Lothar Kanz and Robert Möhle J Immunol 2009; 182:6789-6798; ;

doi: 10.4049/jimmunol.0801525 Downloaded from http://www.jimmunol.org/content/182/11/6789

References This article cites 68 articles, 33 of which you can access for free at:

http://www.jimmunol.org/content/182/11/6789.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists

by guest on October 2, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

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 © 2009 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

␣ ␤ The CysLT1 Ligand Leukotriene D4 Supports 4 1- and ؉ ␤ ␣ 5 1-Mediated Adhesion and Proliferation of CD34 Hematopoietic Progenitor Cells1

Andreas M. Boehmler,* Adriana Drost,* Lena Jaggy,* Gabriele Seitz,* Tina Wiesner,* Claudio Denzlinger,† Lothar Kanz,* and Robert Mo¨hle2*

Cytokines and chemokines control hematopoietic stem and progenitor cell (HPC) proliferation and trafficking. However, the role of nonpeptide mediators in the microenvironment has remained elusive. Particularly CysLT1, a G -coupled receptor recognizing inflammatory mediators of the cysteinyl leukotriene family, is highly expressed in HPCs. We therefore ؉ analyzed the effects of its ligands on human CD34 HPCs. The most potent CysLT1 ligand, LTD4, rapidly and significantly ␣ ␤ ␣ ␤ up-regulated 4 1 and 5 1 integrin-dependent adhesion of both primitive and committed HPC. LTD4-triggered adhesion was Downloaded from inhibited by specific CysLT1 antagonists. The effects of other CysLT1 ligands were weak (LTC4) or absent (LTE4). In serum-free liquid cultures supplemented with various hematopoietic cytokines including IL-3, only LTD4 significantly augmented the expan- sion of HPCs in a dose-dependent manner comparable to that of peptide growth factors. LTC4 and LTE4 were less effective. In ؉ CD34 cell lines and primary HPCs, LTD4 induced phosphorylation of p44/42 ERK/MAPK and focal adhesion kinase-related tyrosine kinase Pyk2, which is linked to integrin activation. Bone marrow stromal cells produced biologically significant amounts of cysteinyl leukotrienes only when hematopoietic cells were absent, suggesting a regulatory feedback mechanism in the hema- topoietic microenvironment. In contrast to antagonists of the homing-related G protein-coupled receptor CXCR4, administration http://www.jimmunol.org/ ؉ of a CysLT1 antagonist failed to induce human CD34 HPC mobilization in vivo. Our results suggest that cysteinyl leukotriene may contribute to HPC retention and proliferation only when cysteinyl leukotriene levels are increased either systemically during inflammation or locally during marrow aplasia. The Journal of Immunology, 2009, 182: 6789–6798.

oming of hematopoietic stem and progenitor cells diators like LTs, which regulate leukocyte migration and function (HPCs)3 to the bone marrow and their subsequent reten- at sites of inflammation, may also affect HPCs, particularly their H tion and survival in the hematopoietic microenviron- trafficking and proliferation in the bone marrow. ment are regulated by particular and peptides (cytokines, Cysteinyl leukotrienes represent important lipid mediators in in- by guest on October 2, 2021 chemokines, and adhesion molecules) (1–3). However, the role of flamed tissues and have recognized roles in respiratory diseases, lipid mediators such as leukotrienes (LTs) in HPC homeostasis has cancer, and cardiovascular, gastrointestinal, skin, and immune dis- remained elusive, in contrast to their well-characterized involve- orders (10). They elicit many of their effects through seven-trans- ment in allergy and inflammation (4–6). In this context, it is worth membrane G protein-coupled receptors (GPCRs) that show struc- mentioning that the hematopoietic microenvironment shares sim- tural similarities to chemokine receptors such as CXCR4, which is ilarities with inflamed tissues and that the homing of HPCs resem- a key receptor involved in HPC homing to the bone marrow (11, bles extravasation of mature leukocytes during inflammation. For 12). To date, two GPCRs recognizing cysteinyl leukotrienes, instance, bone marrow endothelial cells (BMEC) constitutively ex- CysLT1 and CysLT2, have been cloned and characterized (10, 12– press the adhesion molecules E-, P-selectin, and VCAM-1, 14). They show different binding affinities for their ligands, LTD4, which are found on the of other tissues only during LTC , and LTE (15). CysLT binds LTD with a 350-fold greater inflammation (7–9). It is therefore conceivable that nonpeptide me- 4 4 1 4 affinity than it does LTC4 (10). LTE4, which acts as a partial ag-

onist, is the least active cysteinyl leukotriene (12). CysLT2,in † *Department of Medicine II, University of Tu¨bingen, Tu¨bingen, Germany; and De- contrast, binds LTD4 and LTC4 with equal affinity (13, 14). partment of Medicine III, Marienhospital, Stuttgart, Germany CysLT1 couples to pertussis toxin (PTX)-sensitive Gi/o and PTX- Received for publication May 9, 2008. Accepted for publication March 31, 2009. insensitive Gq proteins and activates signaling pathways that mod- The costs of publication of this article were defrayed in part by the payment of page ulate multiple cellular activities, including proliferation and mi- charges. This article must therefore be hereby marked advertisement in accordance gration. The role and function of CysLT is not completely with 18 U.S.C. Section 1734 solely to indicate this fact. 2 understood. This receptor may also activate G signaling path- 1 This work was supported by Deutsche Forschungsgemeinschaft Grant SFB 510/A4 i/o and Deutsche Jose´Carreras Leuka¨miestiftung Grants DJCLS R04/13 and R08/24v. ways, although recent studies suggest that CysLT2 interacts with 2 Address correspondence and reprint requests to Dr. Robert Mo¨hle, Department of CysLT1 on the cell surface, resulting in the formation of het- Medicine II, University of Tu¨bingen, Otfried-Mu¨ller-Strasse 10, 72076 Tu¨bingen, erodimers and eventually in the attenuation of CysLT1-mediated Germany. E-mail address: [email protected] effects (16). 3 Abbreviations used in this paper: HPC, hematopoietic stem and progenitor cell; We have previously shown that CysLT1 is highly expressed in BMEC, bone marrow endothelial cell; CAFC, cobblestone area-forming cell; EIA, ϩ enzyme immunoassay; GPCR, G protein-coupled receptor; 5-LO, 5-lipoxygenase; CD34 progenitor cells (17). In contrast to CXCR4, the estab- LT, leukotriene; MPA, mycophenolic acid; PTX, pertussis toxin; rh, recombinant lished “homing receptor” for HPCs that is expressed much stron- human; SDF, -derived factor. ger in mature mononuclear cells than in CD34ϩ HPCs (18),

Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00 CysLT1 expression appears to be more progenitor specific. Indeed, www.jimmunol.org/cgi/doi/10.4049/jimmunol.0801525 6790 LTD4 SUPPORTS PROGENITOR ADHESION AND PROLIFERATION

even though mature leukocytes express some CysLT1 at the Flow cytometric analysis mRNA and protein level (17), it does not mediate typical func- Expression of glycophorin-A, CD14, CD15, CD34, CD38, CD41, and tions in these cells, e.g., calcium fluxes (19). In contrast, cir- CD45 was evaluated by incubating 2–5 ϫ 105 cells with the respective culating HPCs express high levels of CysLT1 and respond to FITC- or PE-labeled Abs (BD Biosciences) for 20 min at 4°C and analyzed on a FACSCalibur flow cytometer equipped with CellQuestPro software. LTD4 with much more pronounced calcium fluxes than those observed after stromal cell-derived factor (SDF)-1-induced Isotype-matched IgG Abs served as negative controls. A proportion of 1% or fewer false positive events were accepted in the negative control sam- CXCR4 activation (20). ples throughout. However, besides inducing considerable calcium fluxes and che- motaxis in vitro, the function of CysLT receptors and the potential assays source of their ligands in the hematopoietic system in vivo remain to be characterized. Similar to CXCR4, which plays an important To prepare HUVEC monolayers, 1 ϫ 104 cells were seeded on gelatin- role in B cell trafficking and proliferation in addition to its function coated (0.1% in PBS), 96-well, flat-bottom tissue culture plates 48 h before the assay and grown to confluence in endothelial cell growth medium plus as a HPC homing receptor, CysLT1 as well may regulate the func- 10% FCS. Before adding HPCs, the monolayers were stimulated with 10 tion of different cell types, i.e., inflammatory effector cells ng/ml IL-1␤ (Boehringer Mannheim) for 4 h. For the adhesion assays on and HPCs. recombinant human (rh) VCAM-1, 96-well microtiter plates were coated In the present study, we analyzed the effects of cysteinyl leu- for 90 min at 37°C with 50 ␮lofa5␮g/ml solution of carrier protein-free VCAM-1 (R&D Systems) diluted in 0.05 M bicarbonate/carbonate buffer kotrienes on progenitor cell adhesion and proliferation in vitro as (pH 9.6). The plate was washed with PBS to remove unbound protein, and substantial cellular functions involved in the retention, survival, nonspecific binding sites were blocked using a 0.1% (w/v) solution of BSA and differentiation of HPCs in the bone marrow microenvironment in PBS (1 h at 37°C). Before the experiment, wells were washed twice with Downloaded from and identified potential sources of LT production in hematopoietic 200 ␮l/well assay medium. Adhesion experiments on fibronectin were per- tissue. In analogy to studies demonstrating HPC mobilization by formed using BioCoat Cellware human fibronectin 96-well Multiwell plates (BD Biosciences) that were preincubated for 1 h with 100 ␮l/well the CXCR4 antagonist AMD3100 (21), we evaluated the capacity assay medium at 37°C in 5% CO2 before adding HPCs. ϩ of CysLT1 antagonists to induce mobilization. CD34 progenitor cells were washed and resuspended in assay medium and held at 37°C for 3 h under gentle agitation before preincubation at

37°C with or without LTD4 for the times and concentrations indicated. To

block cell adhesion, mAbs were added together with LTD4. In some ex- http://www.jimmunol.org/ Materials and Methods periments, cells were preincubated for 2–5 min with MK-571 (final con- Reagents and Abs centration 2 ␮M), LY17183 (10 ␮M), or PTX (20 ng/ml). Cells (6 ϫ 104 in 100 ␮l) were added in triplicate to the precoated plates, spun down for The LTs C ,D, and E were obtained from Paesel and Lorei or Biomol. ϫ 4 4 4 15sat30 g, allowed to adhere for 5 min at 37°C in 5% CO2 and then Purity and integrity of the lots were assessed by HPLC analysis. Working washed twice with warm assay medium to remove nonadherent cells. solutions were prepared using HPLC-grade methanol (Merck) and were Bound cells were harvested with trypsin/EDTA, washed in RPMI 1640 kept below 0°C and under argon atmosphere at all times to avoid formation plus 10% FCS to inactivate the trypsin, and subsequently quantified and/or of biologically inactive isomers. The compounds MK-571, LY171883, characterized by flow cytometry, counted, and assessed for CFU or trans- PTX, genistein, and PD98059 were obtained from Biomol. ferred into long-term cultures on FBMD-1 stromal feeders (see below). Abs used in the adhesion assays were mAbs that have been shown to For quantification, recovered cells from the adhesion assays were sus- functionally block the respective adhesion molecule (22, 23) VLA-4 pended in PBS containing 0.1% BSA and stained with anti-CD34-PE. by guest on October 2, 2021 (CD49d, clone HP2/1; Beckman Coulter-Immunotech), VLA-5 (CD49e, FITC-conjugated calibration beads (3 ϫ 104) (CaliBRITE; BD Bio- ␤ clone SAM-1; Immunotech), LFA-1 (CD11a, clone TS 1/11; Endogen), 2 sciences) were added, and samples were analyzed by FACS. FITC-beads integrin (CD18, clone TS 1/18; Endogen), or IgG-matched isotype controls were gated as G1 ϭ R1 (where G is gate and R is region), and data ac- (BD Biosciences). All Abs were used at a final concentration of 12.5 quisition was stopped after 5,000 acquired events of beads. The number of ␮g/ml. adherent cells per well was thus calculated from the ratio of HPC to beads multiplied by the number of beads added.

Human cells and cell lines Cytokine-supplemented liquid culture of CD34ϩ cells After informed consent, peripheral cells were obtained from healthy ϩ donors or patients with nonhematological malignancies during G-CSF-in- Isolated CD34 progenitors were seeded into 24-well cell culture plates ϫ 4 duced stem cell mobilization according to the guidelines of the ethical (Costar) at an input concentration of 5 10 cells/ml in X-Vivo 20 con- committee of the University of Tu¨bingen, Tu¨bingen, Germany (project no. taining stem cell factor, IL-3, and FLT3 ligand (100 ng/ml each; all from 268/2003-LP). Mononuclear cells were separated by Ficoll density gradient PeproTech) and the indicated concentrations of leukotrienes, and incubated ϩ centrifugation. Enrichment of CD34 cells was performed using immuno- at 37°C with 5% CO2 in a humidified atmosphere for 1 wk. Experimental magnetic microbeads (MACS system; Miltenyi Biotec) according to the setups with LTs must take into consideration that these compounds are manufacturer’s instructions (purity Ն 96%). Primary bone marrow cells rapidly metabolized and degraded in most systems, a process due to surface were isolated after informed consent from healthy volunteers (project no. enzymes and/or enzymatic active components in the incubation medium 268/2003-BM). Bone marrow CD34ϩ cells were isolated by Ficoll sepa- (4). Therefore, LTs were added at days 0, 2, and 4 to allow stimulation of ration of mononuclear cells and the subsequent enrichment of CD34ϩ cells cells at multiple time points. At the end of the incubation period, cells were by MACS as described above for peripheral blood-derived progenitors. quantified in a hematocytometer. Viability was assessed by trypan blue Adherent stromal cells were isolated and cultured as previously described exclusion. For further characterization of the proliferated cells, aliquots of (17). To deplete stromal cultures of contaminating hematopoietic cells, the the cell samples were labeled for flow cytometry. Additional subsets were layers were treated with three cycles of 5 ␮g/ml mycophenolic acid (MPA; reseeded into methylcellulose clonogenic cultures and cobblestone area- Sigma Aldrich), each cycle consisting of a 3-day period with MPA fol- forming cell (CAFC)-assays as described below. lowed by a 4-day recovery period without MPA (24–26). Stromal cells were then washed twice with PBS and maintained for at least 1 wk in Colony-forming cell assays MPA-free medium (including two complete changes of medium) before using them in experiments. HUVECs (27) were grown in endothelial cell Clonogenic progenitor assays were performed by plating equal numbers of growth medium (PromoCell) plus 10% FCS. Cells were used between the cells (500/ml) either from the starting populations or harvested from the fifth and eighth passage. The human microvascular BMEC line BMEC-1 adhesion/proliferation assays in duplicate with MethoCult H4230 (Stem (28, 29) was cultured in endothelial cell growth medium plus 10% FCS. Cell Technologies). Methylcellulose medium was supplemented with The CD34ϩ CysLT1ϩ human cell lines KG-1a and rhSCF (stem cell factor), rhIL-3, rhIL-6 (20 ng/ml each; all from Pepro- ␣ Kasumi-1 as well as the CysLT1-negative T lymphoblastic cell line Jurkat Tech), rhG-CSF (100 ng/ml; Amgen), and rhEPO (epoetin- , 6 U/ml; were propagated in RPMI 1640 medium with 10% FCS, 100 U/ml peni- Jannsen-Cilag). Colonies were counted after 14 days of culture in a hu- ␮ cillin, and 100 g/ml streptomycin. midified environment at 37°C and 5% CO2. The Journal of Immunology 6791

CAFC assays AAGCTTGTCATCAATGG-3Ј (forward) and 5Ј-GGCAGTGATG Ј 3 GCATGGACTG-3 (reverse; product size 358 bp). A pre-PCR heat step (5 Stromal feeders were prepared by seeding 10 FBMD-1 cells (passages min at 95°C) provided a PCR Hot-Start by partially activating AmpliTaq 16–19) per well into fibronectin-coated 96-well plates (BioCoat Cellware) Gold DNA polymerase (Applied Biosystems). For analysis of 5-LO and (30). Cells were grown to confluence in MyeloCult H 5100 medium (Stem GAPDH expression, 36 and 17 amplification cycles, respectively, were Cell Technologies) supplemented with gentamicin (100 mg/l; Boehringer Ϫ6 run, each consisting of denaturation for 15 s at 95°C, annealing for 30 s at Mannheim) and hydrocortisone (10 M; Sigma Aldrich). During cocul- 58°C, and elongation for 90 s at 72°C, followed by a final extension step ture with HPCs, the medium was additionally enriched with 10 ng/ml IL-3 of 10 min at 72°C. PCR products were separated by horizontal agarose gel and 20 ng/ml G-CSF (CAFC medium). For CAFC analysis, FBMD-1 cells ϩ electrophoresis with subsequent ethidium bromide staining. To detect pos- were overlaid with CD34 cells in a limiting dilution setup, with 12 dilu- sible contamination of genomic DNA in the RNA sample, each reaction tion steps 2-fold apart and 12 replicate wells per dilution for each set of was performed without previous reverse transcription. experiments. To assess the frequency of cobblestone area-initiating cells within the CD34ϩ populations adherent to VCAM-1, cells harvested from Cell lysate preparation and Western blotting the adhesion assays were spun down, resuspended in equal volumes of ϩ CAFC medium, and transferred 1:1 without prior counting, i.e., the content Primary peripheral blood CD34 HPC and cell lines were serum starved ϫ 7 from one well of the adhesion assay was transferred to one well in the row for3hinX-Vivo 20 medium, resuspended at 2 10 /ml in X-Vivo 20, with the highest cell density of the CAFC assay plate. For the cells recov- and incubated with 100 nM LTD4 (final concentration) for the indicated ered from the proliferation experiments, starting input values ranged be- time periods at 37°C. After stimulation, whole cell lysates were prepared in tween 5000 and 8000 cells/well. Once a week half of the medium (50 ␮l) buffer containing 50 mM HEPES (pH 7.5), 10% glycerol, 1% Triton was replaced with fresh CAFC medium. The appearance of cobblestone X-100, 1.5 mM MgCl2, 150 mM NaCl, protease inhibitors (Complete areas, consisting of at least five cells, was evaluated after 5 wk with a phase Mini; Roche) and the phosphatase inhibitors NaF (100 mM), Na4P2O7 (10 contrast illumination on an inverted microscope. CAFC frequencies were mM), and activated Na3VO4 (1 mM; all from Sigma-Aldrich). Equal ␮ calculated as single hit kinetics from the fraction of nonresponding cultures amounts of protein (20 g/lane) were separated on a 10% SDS-polyacryl-

using the Poisson equation as described (30, 31). amide gel and transferred onto nitrocellulose membrane. The blots were Downloaded from probed with phospho-specific polyclonal Abs against p44/42 MAPK 202 204 402 308 Analysis of cysteinyl leukotriene production (Thr /Tyr ), Pyk2 (Tyr ), and AKT (Thr ) or with control Abs against the nonphosphorylated forms, respectively (all from Cell Signaling Stromal cells and confluent layers of BMEC-1 cells were washed with Technology). Bands were visualized with ECL (GE Healthcare). PBS and incubated for 18 h in serum-free medium (X-Vivo 15 without phenol red; BioWhittaker). A 1-ml aliquot of each medium sample was In vivo effects of the specific CysLT1 antagonist montelukast on added to 8 ml of 90% aqueous methanol (Merck) containing 0.5 HPC mobilization.

mmol/L EDTA and 1 mmol/L of the radical scavenger 4-hydroxy- http://www.jimmunol.org/ 2,2,6,6,-tetramethylpiperidine-1-oxyl (pH 5.6; both from Sigma-Al- After informed consent as approved by the ethical committee of the Uni- drich) (32). The methanolic medium suspension was stored at Ϫ80°C versity of Tu¨bingen (project number 147/2006V), blood samples for re- for at least 24 h and then spun down at 5,000 ϫ g for 20 min at Ϫ10°C. search purposes were obtained from otherwise healthy human subjects re- The supernatants were concentrated to dryness in a centrifuge under a ceiving treatment with montelukast (independently of the measurements reduced atmosphere and dissolved in 30% aqueous methanol for HPLC described here) as a prophylaxis of exercise-induced asthma. All subjects analysis. Immediately after collection of the conditioned medium, cells had normal blood cell counts and normal findings from a physical exam- were harvested with trypsin plus EDTA and counted. Equal amounts of ination and were taking no other medication. Patients with allergic asthma or active asthmatic disease or any other abnormal symptoms were ex- nonconditioned assay medium with or without known amounts of LTC4, LTD , and LTE run under identical conditions were used as negative cluded. The study design was adopted from studies with the CXCR4 an- 4 4 tagonist AMD3100 for stem cell mobilization, which demonstrated an control or internal standard, respectively. ϩ by guest on October 2, 2021 To assess the effect of hematopoietic cells on LT production, stromal 8-fold increase in circulating CD34 HPC 4 h after administration of the cultures depleted of hematopoiesis were overlaid or not with 5 ϫ 104 bone antagonist. In our study, venous blood samples were obtained before and at marrow-derived CD34ϩ hematopoietic cells per 12.5-cm2 culture flask in 2, 4, 8, 12, and 24 h after the first oral dose (10 mg) of the specific CysLT1 antagonist montelukast-sodium (Singulair) for analysis of the white blood 5 ml of MyeloCult H 5100 supplemented with gentamicin (100 mg/l) and ϩ hydrocortisone (10Ϫ6 M). Both stromal cells and CD34ϩ progenitors were count and CD34 cell number by flow cytometry. isolated from the same donor source. After 4 wk of culture with a weekly Statistical analysis semidepopulation of nonadherent cells, layers were rinsed twice with PBS and cultured for an additional 18 h in X-Vivo 15 medium without phenol All results are expressed as the mean Ϯ SEM of 3–5 independent exper- red for collection of conditioned medium as described above. iments, except where indicated. Data analysis was performed with Sig- LT profiles were determined by combined use of HPLC and competitive maStat 2.03 software (SPSS). Differences between groups were compared enzyme immunoassay (EIA). HPLC was performed as described (32) using by Student’s t test or paired t test and considered significant at p Յ 0.05. a Hypersil ODS column (4.6 ϫ 100 mm, 5 ␮m particles; Agilent Tech- nologies) with a Hypersil ODS precolumn (4.0 ϫ 20 mm, 5 ␮m particles; Results Agilent) that was run at a flow rate of 1 ml/min on an Agilent 1100 system. ϩ LTD4-induced adhesion of human CD34 progenitors The mobile phase consisted of methanol, water, acetic acid (65:35:0.1 by volume) and 1 mmol/l EDTA (pH 5.6; adjusted with ammonium hydrox- Preincubation with LTD4 rapidly up-regulated adhesion of HPCs ide). Cysteinyl leukotrienes were monitored at 280 nm, and HPLC frac- to endothelium in a time- and dose-dependent manner by up to tions containing LTC4, LTD4, and LTE4 were isolated and collected for ␮ 80%, with a maximum effect in cells treated with 1 M LTD4 for EIA according to the retention times of known standards run under iden- Ϯ tical conditions. EIA analysis of endogenously produced leukotrienes was 15 min (Fig. 1A). Under these conditions, 22.17 2.45% of input performed using cysteinyl leukotriene EIA kits (Cayman Chemical) ac- cells adhered to HUVEC, compared with 12.53 Ϯ 2.08% without cording to the manufacturer’s instructions. The prior separation by HPLC LTD4 preincubation. Increased HPC adhesion was already detect- allowed assessment for the amount of the respective LT metabolites within Ͻ able at LTD4 concentrations of 1 nM (Fig. 1B). In contrast, LTC4 each fraction. weakly up-regulated HPC adhesion to HUVEC at 1 ␮M, whereas RT-PCR analysis addition of LTE4 did not increase adhesion at concentrations up to ␮ ϩ 1 M (Fig. 1C). LTD4-triggered adhesion of CD34 cells to en- Total RNA from BMEC-1 endothelial cells and primary human stroma ␣ dothelium could be reduced by mAbs against 4 integrin, but not samples (treated or not with MPA for three cycles as described above) was ␣ ␤ isolated using an RNeasy mini (Qiagen), including digestion with by L and 2 integrin mAbs, indicating an effect of LTD4 on DNase I. First-strand cDNA was synthesized with the use of murine leu- VLA-4-dependent mechanisms in this system (Fig. 1D). When ϩ kemia virus reverse transcriptase and oligo(dT)16 primers (both Applied CD34 cells pretreated with LTD4 under optimum conditions (1 Biosystems). Subsequent PCR analysis for 5-lipoxygenase (5-LO) and ␮M/15 min) were brought in contact with the VLA-4 ligand GAPDH (used as internal standard) was performed with gene-specific Ͼ primers based on the human receptor sequences as follows: 5-LO, 5Ј-ACT VCAM-1 immobilized on a plastic surface, adhesion was 2-fold GGAAACACGGCAAAAAC-3Ј (forward) and 5Ј-GTGCAGGGGTCT increased compared with that of untreated cells (Fig. 1E). Ј Ј ␣ GTTTTGTT-3 (reverse; product size 501 bp); GAPDH, 5 -CGGG Adhesion could be completely blocked by the 4 integrin mAb 6792 LTD4 SUPPORTS PROGENITOR ADHESION AND PROLIFERATION Downloaded from http://www.jimmunol.org/ FIGURE 1. LT-induced adhesion of HPCs to HUVECs, VCAM-1, and fibronectin. A, Time-dependent adhesion of HPC to HUVEC in the presence of Ͻ ءء Ͻ ء ␮ 1 M LTD4. Adhesion is expressed as relative adhesion (rel.) compared with the respective control (carrier solution only). , p 0.05; , p 0.01 vs ϩ control (Ctr). B, Dose dependence of HPC adhesion to HUVEC monolayers. CD34 cells were incubated with the indicated concentrations of LTD4 for Ͻ ϩ ء 15 min, followed by exposure to activated HUVECs. , p 0.05 vs control (0 M LTD4). C, Adhesion of HPC to HUVEC induced by different LTs. CD34 ϩ Ϫ ␮ cells were incubated in the presence ( ) or absence ( )ofLTD4, LTC4,orLTE4 (all at 1 M) for 15 min, followed by exposure to activated HUVECs. ␤ Ͻ ء , p 0.05 vs, control without leukotrienes. D, Effect of blocking Abs to VLA-4 and 2 integrins or receptor antagonists on LTD4-induced adhesion to ϩ ␮ HUVECs. CD34 cells were incubated with mAb, isotype-matched IgG, CysLT1 antagonists (MK-571, LY17183), or PTX and stimulated with 1 M LTD4 Ͻ ϩ ءء for 15 min before exposure to activated HUVECs. , p 0.01 vs isotype. E, LTD4-induced adhesion of HPC to VCAM-1 and fibronectin. CD34 cells ␮ were exposed to VCAM-1 and fibronectin-coated plates after simulation with LTD4 (1 M/15 min). The specificity of the effects was assessed with MK-571 Ͻ by guest on October 2, 2021 ءءء Ͻ ءء ␣ ␣ and Abs against VLA-4 ( 4) and VLA-5 ( 5). Wells coated with BSA served as control for unspecific binding. , p 0.01; and , p 0.001 vs isotype (Iso.).

Ͻ ␣ ␤ ( p 0.001 vs isotype plus LTD4), demonstrating its specificity. LTD4 not only on VLA-4 ( 4 1 integrin), but also on VLA-5 ␣ ␤ The effect of LTD4 on HPC adhesion to HUVECs and VCAM-1 ( 5 1 integrin)-mediated mechanisms. Both integrins are ex-

could be significantly blocked with the CysLT1 antagonists MK- pressed on HPCs and serve as receptors for fibronectin. Ͻ Ͻ 571 ( p 0.01 vs LTD4 in Fig. 1D and p 0.05 vs isotype plus Ͻ VLA-4-mediated adhesion of both primitive and committed LTD4 in Fig. 1E, respectively) and LY171883 ( p 0.01 vs progenitors was up-regulated by LTD4 LTD4), respectively, demonstrating the involvement of CysLT1 (Fig. 1, D and E). The data also ruled out the potential participation HPCs harvested from the adhesion experiments on VCAM-1

of multidrug resistance protein 1 in LTD4-induced up-regulation of were further characterized. The percentage of the more imma- HPC adhesion, because only MK-571 but not LY171883 recog- ture CD34ϩCD38Ϫ cells among the CD34ϩ progenitors that

nizes multidrug resistance protein 1. PTX only partially blocked adhered to VCAM-1 in response to LTD4 was similar to that of

LTD4-induced adhesion of HPC to HUVEC, indicating that spontaneously adhering cells and tended to be greater than the CysLT1 signals via both PTX-sensitive (i.e., Gi) and PTX-insen- percentage of immature progenitors among the initially isolated ϩ sitive (e.g., Gq) G proteins (Fig. 1D). Adhesion of HPC to fi- CD34 cells before adhesion, but the difference was not statis- ␤ bronectin, an extracellular matrix component serving as a 1 tically significant (Fig. 2A). Analogous results were seen for ligand, was also significantly increased after pretreatment with CD34ϩCD90ϩ cells, a subpopulation that is also enriched for

LTD4 (Fig. 1E), similar to the experiments examining adhesion more primitive progenitors, indicating that the effect of LTD4 to endothelium or immobilized VCAM-1. Again, MK-571 ab- on VLA-4-mediated adhesion was not dependent on progenitor Ͻ rogated the triggering effect of LTD4 ( p 0.05 vs isotype plus maturation. With regard to lineage-committed progenitors, pre- ␣ ␮ LTD4). Adhesion was partially inhibited by Abs against 4 in- incubation of HPCs with LTD4 (1 M/15 min) also resulted in Ͻ ␣ Ͻ tegrin ( p 0.05 vs isotype plus LTD4) and 5 integrin ( p a significant increase in the total numbers of attached BFU-E 0.05 vs isotype plus LTD4) in this model system. Considering (erythroid burst-forming unit) and CFU-GM cells as well as in that the already spontaneous adhesion to fibronectin is mediated the overall number of attached clonogenic cells, whereas the by integrins, complete inhibition would result in adhesion be- amount of adhered CFU-M and CFU-Mix cells was only mod-

low that of an unstimulated control and reach the value of BSA- erately enhanced (Fig. 2B). Similarly, preincubation with LTD4 coated wells, as observed for HPC adhesion to VCAM, which is consistently resulted in up-regulation of CAFC adhesion to a ␣ solely mediated by 4 integrin. The data point out an effect of donor-specific extent (up to 238%; Fig. 2C), which underscored The Journal of Immunology 6793

FIGURE 2. LTD4-induced adhesion of primitive and more committed CD34ϩ HPCs to VCAM-1. A, HPC phenotype in the starting population and after adhesion. More primitive, immature CD34high/CD38low and CD34high/CD90ϩ progenitor cells (rectangular gates) were quantified in the starting population before adhe- sion (left), in spontaneously adhering cells (middle; Me- dium), and in cells adhering after stimulation with 1 ␮M

LTD4 for 15 min (right). Data are expressed as percent- age of gated events. B, Adhesion of lineage-committed Downloaded from HPCs (CFU). CD34ϩ cells were stimulated with 1 ␮M

LTD4 for 15 min before exposure to rhVCAM. Adher- ing cells were harvested and plated in a semisolid col- p Ͻ 0.001 vs ,ءءء ;p Ͻ 0.01 ,ءء ;p Ͻ 0.05 ,ء .ony-assay control, respectively. BFU-E, Erythroid burst-forming unit. C, Adhesion of primitive HPCs (CAFC). Long- term cultures on FBMD-1 stromal feeders were set up http://www.jimmunol.org/ with HPC harvested from adhesion experiments on VCAM-1 by a limiting dilution setup. Data are pre- sented as number of CAFCs per 104 initial cells brought in contact with VCAM-1 (cells from three different do- nors were analyzed). by guest on October 2, 2021

that LTD4 also affected VLA-4-mediated adhesion of more medium, as determined by limiting dilution analysis (Fig. 3E). primitive progenitor cells. As CAFCs represent only a small fraction of the CD34ϩ pro-

ϩ genitors with a distinct proliferation capacity, we also chose the LTD4 up-regulated proliferation of CD34 HPC in vitro 1 ␮M concentration, which was highly active in the adhesion Analysis of cell numbers after ex vivo expansion of CD34ϩ experiments, for analysis of CAFC proliferation. HPC in serum-free, cytokine-supplemented medium for 1 wk in FACS analysis of cells expanded under the above mentioned con- the presence or absence of the indicated concentrations of LTD4 ditions in the presence or absence of 100 nM LTD4 showed no demonstrated a dose-dependent effect on cell proliferation in marked differences in the expression patterns of the differentiation cytokine-supplemented liquid cultures (Fig. 3A). The presence markers CD14, CD15, CD38, CD41, and , demonstrat- of 100 nM LTD4 resulted in a 2-fold increase in the total cell ing an equal proliferative effect of LTD4 on the monocytic, granulo- number (Fig. 3B). In contrast, the presence of LTC4 and LTE4 cytic, megakaryocytic, and erythropoietic lineages (Table I). augmented HPC proliferation only weakly (18 and 12% in- ϩ crease, respectively), which was not statistically significant LTD4 activated major signaling pathways in CD34 cells compared with proliferation without LT addition. Proliferation To understand possible molecular mechanisms involved in the ef- experiments using varying cytokine combinations revealed an fects of LTD4 on progenitor cell adhesion and proliferation, ϩ effect of LTD4 on cell expansion predominantly in samples con- CD34 cells were stimulated with 100 nM LTD4 and analyzed by taining IL-3 (Fig. 3C). After 1 wk of cytokine-supplemented Western blotting for the activation of signaling molecules using liquid culture, lineage-committed and more primitive progeni- phospho-specific Abs. As shown in Fig. 4A, LTD4 rapidly induced ϩ tors were increased in the presence of 100 nM LTD4,asdem- the activation of p44/42 MAPK (ERK1/2) in both CD34 cell onstrated by the expansion of clonogenic progenitors (Fig. 3D) lines as well as in the primary mobilized CD34ϩ HPCs, whereas and the concentration of primitive CAFCs present in the culture LTD4 had no effect on the Jurkat line that does not express 6794 LTD4 SUPPORTS PROGENITOR ADHESION AND PROLIFERATION

Table I. Expression of differentiation antigens after cytokine-supplemented liquid culture of peripheral blood CD34ϩ HPCsa

Control (%) 100 nM LTD4 (%)

CD45ϩ 99.46 Ϯ 0.24 99.52 Ϯ 0.11 CD34ϩ 49.55 Ϯ 2.94 53.57 Ϯ 12.64 CD38ϩ 25.32 Ϯ 3.80 27.86 Ϯ 3.73 CD34ϩ/CD38ϩ 14.85 Ϯ 1.68 16.51 Ϯ 4.54 CD14ϩ 9.00 Ϯ 2.90 9.18 Ϯ 2.79 CD15ϩ 13.27 Ϯ 0.74 10.10 Ϯ 2.32 CD41ϩ 34.62 Ϯ 0.57 35.79 Ϯ 2.08 Gly-Aϩ 12.96 Ϯ 6.31 8.70 Ϯ 2.44

a The percentages of progenitor-associated (CD34ϩ), lineage commitment-related (CD38ϩ), myeloid (CD14ϩ, CD15ϩ), megakaryocytic (CD41ϩ), and erythroid (Gly- Aϩ) Ags were assessed by flow cytometry after 1 wk of liquid culture (mean Ϯ SEM).

ylated after LTD4 treatment (Fig. 4B). Pyk2 represents a homo- logue of the focal adhesion kinase and is involved in the integrin- signaling pathway, particularly in CD34ϩ hematopoietic Downloaded from progenitor cells that do not express the focal adhesion kinase (35).

The effects of LTD4 on p44/42 MAPK and Pyk2 support our ob-

servations that LTD4 triggers not only chemotaxis (17) but also integrin-mediated adhesion and proliferation of CD34ϩ HPCs. In- terestingly, protein kinase B (AKT), a major signal molecule that has been implicated in progenitor cell survival and proliferation http://www.jimmunol.org/

(36), was less affected by LTD4 treatment under the conditions used in our studies (Fig. 4C). LTD4-induced adhesion to HUVECs was significantly blocked by an inhibitor of tyrosine kinases (genistein) and by PD98059, an inhibitor of the MEK/ERK/MAPK Ͻ Ͻ pathway ( p 0.05 and p 0.001 vs LTD4, respectively, shown in Fig. 4D). Expansion of HPCs in the presence of hematopoietic cytokines involves MAPK and various tyrosine kinases. The effect of these inhibitors on HPC expansion was therefore not evaluated, by guest on October 2, 2021 as they completely block any proliferation in the assay FIGURE 3. LT-induced proliferation of CD34ϩ HPC. A, Effect of systems used. ϩ LTD4 on ex vivo expansion of HPC. CD34 cells were cultured in cytokine-supplemented serum-free medium (control (Ctr), ϳ10-fold in- crease after 1 wk) and compared with those additionally supplemented Cysteinyl leukotrienes are produced by bone marrow stroma Ͻ and endothelium ءء Ͻ ء with the indicated concentrations of LTD4. , p 0.05; , p 0.01 vs control. rel., Relative adhesion (compared with control) B, LT-induced ϩ By RT-PCR, stromal and immortalized BMEC1 cells expressed expansion of HPCs. CD34 cells were incubated in cytokine-supple- 5-LO, which is the key enzyme of the leukotriene biosynthesis mented, serum-free medium in the presence (ϩ) or absence (Ϫ)of1 ␮ pathway (4, 37) (Fig. 5A). As measured by HPLC and immu- M LTD4, LTC4,orLTE4, respectively, before the amount of cells was ,p Ͻ 0.05 vs control. C, Effect of cytokines on expansion of noassay (Fig. 5B), biologically significant amounts of LTC4 ,ء .assessed HPCs. The indicated cytokines were added (100 ng/ml each) in the LTD4, and LTE4 were found in the supernatants of confluent layers of these cells (Fig. 5C). Summarizing all three LTs presence or absence of 100 nM LTD4, and the cell number was assessed p Ͻ 0.01 vs control. SCF, Stem cell factor. (LTC4, LTD4, and LTE4), concentrations were as high as ,ءء ;p Ͻ 0.05 ,ء .after 1 wk Ϯ Ϯ D, Effect of LTD4 on HPC expansion of committed progenitors (CFU). 21.35 5.25 nM (endothelium) and 18.27 4.13 nM (stroma) HPCs before and after 1 wk of culture in cytokine-supplemented, se- in the culture medium, respectively. As mentioned above, sig- rum-free medium with or without 100 nM LTD4 were analyzed for their nificant effects on HPC proliferation in vitro were observed clonogenic potential. Plating efficiency was assessed by plating equal already at a final LTD4 concentration of 1 nM, demonstrating numbers of cells in methylcellulose. By multiplying with the corre- the physiological relevance of the data observed. Interestingly, p Ͻ ,ء .sponding total cell number, the CFU expansion was calculated in a conditioned medium of stromal cells not deprived of he- 0.05 vs control. E, Effect of LTD4 on primitive 5-wk CAFCs in cyto- kine-supplemented liquid culture. Long-term cultures on FBMD-1 stro- matopoietic cells by treatment with MPA, almost no cysteinyl mal feeders were set up with cells cultured for 1 wk in cytokine sup- leukotrienes could be detected (Fig. 5C). To confirm that the plemented serum-free medium by a limiting dilution setup (12 replicate absence of hematopoiesis is required for LT synthesis in these p Ͻ 0.05 vs control. cells, pure stromal layers were cocultured with bone marrow ,ء .(wells/condition; 12 dilution steps CD34ϩ cells (derived from the same donor source). Indeed, LT levels were reduced by 89.7–91.3% (Fig. 5D). This suggests LT

CysLT1 and therefore served as a negative control. An effect of metabolism and/or modulation of LT production by bone mar- LTD4 on ERK p44/42 MAPK signaling has been reported in other row hematopoietic cells and supports previous findings that the cells (including differentiated epithelial and hematopoietic cells) quantity and profile of LTs produced both in vitro and in vivo (33, 34), but not in hematopoietic progenitor and stem cells to date. critically depend on cellular interactions and the composition of Additionally, the nonreceptor tyrosine kinase Pyk2 was phosphor- mixed cell populations (4). The Journal of Immunology 6795 Downloaded from

FIGURE 5. Production of cysteinyl leukotrienes by BMEC-1 and stro- mal cells (Stroma). A, Expression of 5-LO mRNA in BMEC-1 and primary stromal cells. RT-PCR was used for this analysis (positive control, http://www.jimmunol.org/ GAPDH; negative control, absence of reverse transcriptase (w/o RT)). Complete elimination of hematopoiesis from stromal layers was achieved by MPA treatment. MW, Molecular marker. B, Experimental setup for the isolation and quantification of culture-derived cysteinyl leukotrienes. Con- fluent layers of primary bone marrow stromal cells or the human bone marrow endothelial cell line BMEC-1 were incubated for 18 h in serum-

free medium. Fractions containing LTC4, LTD4, and LTE4 were isolated by HPLC according to the retention times of known standards and quan-

tified by EIA. C, Production of cysteinyl leukotrienes by human BMEC-1 by guest on October 2, 2021

ϩ and primary human stromal cells. LTC4, LTD4, and LTE4 were quantified FIGURE 4. LTD4-induced signal transduction in CD34 cell lines as described in B. D, Cysteinyl leukotriene production of stromal cells and HPCs. A, Phosphorylation of p44/42 MAPK (P-44/42 MAPK) in ϩ without and with establishment of hematopoiesis in vitro. Bone marrow ϩ cell lines and primary CD34 HPC in response to LTD4. KG1a and (BM)-derived CD34 cells originating from the same donor were added to Kasumi-1 were analyzed by Western blotting after stimulation with a MPA-treated stromal layer and cultured for 4 wk. The amount of cys- LTD4 (100 nM) for up to 20 min. Jurkat cells not expressing CysLT1 ϩ teinyl leukotrienes was analyzed as described in B and compared with the served as a negative control. Primary CD34 cells were analyzed before stromal cells without hematopoiesis. (0) and after stimulation with LTD4 for 1 min. PB, Peripheral blood. B, Phosphorylation of Pyk2 (P-Pyk2) in KG1a cells and in primary CD34ϩ HPCs. The conditions were similar to those in the experiments de- the time points studied. In asthmatic patients, a single dose of scribed in A. C, Phosphorylation of AKT (P-AKT) in KG1a cells and in 10 mg of montelukast causes physiological effects (bronchodi- primary CD34ϩ HPCs. A positive control for phosphorylated AKT was lation) within several hours after oral administration, with the added (Pos.). D, Effect of tyrosine kinase inhibitors on LTD4-induced ϩ ␮ mean peak plasma concentration of the drug being achieved 3 h adherence of HPCs. CD34 cells were incubated with LTD4 (1 M) ␮ ␮ after oral intake (38, 39). Based on this, we conclude that in- and genistein (10 M) or LTD4 and PD98059 (10 M) for 15 min p Ͻ 0.05 vs control. rel., hibition of CysLT1 does not result in HPC mobilization during ,ء .before exposure to activated HUVECs Relative adhesion (compared with control). steady-state hematopoiesis in vivo.

Table II. White blood count (WBC) and CD34ϩ peripheral blood (PB) Effect of a CysLT1 antagonist on human HPC mobilization progenitor cell numbers before and after administration of montelukasta We investigated whether montelukast-sodium (Singulair), a selec- ϩ ␮ tive CysLT1 antagonist used in asthma therapy, could mobilize Time (h) WBC (nl) PB CD34 cells ( l) hematopoietic progenitor cells from bone marrow to peripheral 0 7.76 Ϯ 1.41 3.22 Ϯ 1.31 blood in human subjects with effects similar to those described 2 7.45 Ϯ 1.31 2.84 Ϯ 1.18 for the CXCR4 antagonist AMD3100 (21). In these previous 4 6.73 Ϯ 0.94 2.89 Ϯ 0.84 studies, an 8-fold increase of circulating CD34ϩ cells was ob- 8 7.00 Ϯ 0.68 2.44 Ϯ 0.95 Ϯ Ϯ served 4 h after a single dose of the CXCR4 antagonist, sug- 12 6.94 0.23 3.72 0.59 24 6.01 Ϯ 1.15 2.68 Ϯ 0.31 gesting that HPC mobilization is due to prevention and disrup- tion of homing. Table II demonstrates that administration of a Before (t ϭ 0) and at the indicated time points after the administration of 10 mg of montelukast-sodium, the WBC was measured and the number of CD34ϩ cells was Singulair (10 mg) had no effect on the white blood count and on evaluated by flow cytometry. Values are expressed as mean Ϯ SEM, n ϭ 4; the ϩ the number of CD34 HPCs in the peripheral blood at any of differences were statistically not significant. 6796 LTD4 SUPPORTS PROGENITOR ADHESION AND PROLIFERATION

Discussion LTs have been proposed to stimulate cell proliferation in a num- Homing of hematopoietic progenitor cells to the bone marrow and ber of different in vitro systems, including myeloid CFUs derived mobilization from the marrow into the peripheral blood are mul- from mononuclear cells (51, 52). However, it remained unclear tifactorial processes that are regulated at least in part at the level of whether these effects were directly mediated by HPCs or by mature the bone marrow endothelium (1, 40). The principal mechanisms leukocytes, as unseparated cells were used. In the present study, we clearly showed that LTD4 dose-dependently triggered prolif- involved in transendothelial migration have been elucidated in a ϩ variety of in vitro and in vivo studies, revealing a multistep process eration of CD34 HPCs in the presence of certain cytokines, es- that comprises mainly selectin-mediated rolling interactions on en- pecially IL-3. Due to their lower receptor affinity, the effects of LTC and LTD were substantially weaker. There is a considerable dothelial cells followed by integrin-mediated firm adhesion after 4 4 amount of evidence that points toward interactions between lipid activation of progenitors with suitable stimulants (8, 41–43). mediators and peptides, and above all between LTs and cytokines, Within this system, activation of cells with chemokines like SDF-1 including IL-3. Indeed, IL-3 and other cytokines have been shown leads to increased affinity of receptors of the ␤ and ␤ integrin 1 2 to enhance LT production and receptor expression in vitro in var- subfamilies, with VLA-4 and VLA-5 playing predominant roles in ious cell types (53, 54), including monocytes/macrophages (55) the actual transmigration process and in subsequent HPC engraft- and HL-60 cells differentiated into eosinophils (56). Moreover, ment (2, 22). treatment of patients with IL-3 leads to an increase in the endog- Studies addressing the factors that are able to modulate integrin- enous LT production (54). In reverse, LTs can affect cytokine pro- mediated HPC adhesion and transendothelial migration mainly fo- duction as shown for different interleukins and TNF (57–59). Of cus on the role of the proteins and peptide compounds therein, i.e., note, the stimulating effect of LTs on lineage-committed myeloid cytokines, chemokines, and enzymes. We have shown in previous cells observed in earlier studies was dependent on the presence of Downloaded from studies that the LTD4 receptor CysLT1 is expressed in human GM-CSF (52, 60). The receptors for IL-3 and GM-CSF share a ϩ ϩ CD34 hematopoietic progenitor cells as well as in the CD34 common ␤ subunit (61) and therefore activate similar signal trans- cell line KG1a in a functionally active form and can trigger trans- duction pathways. In our study using separated CD34ϩ cells, the ϩ migration of CD34 cells across bone marrow endothelial layers effects of cysteinyl leukotrienes on hematopoietic progenitors were ␤ in an in vitro system (17). We now demonstrate that the major 1 most likely directly receptor mediated, but stimulation of autocrine integrins VLA-4 and VLA-5 expressed on immature human cytokine production could also contribute to HPC proliferation. http://www.jimmunol.org/ ϩ CD34 cells can be functionally activated by the lipid mediator Production of lipid mediators was observed in the human bone

LTD4, leading to a significant increase in the firm adhesion of HPC marrow in earlier studies (60, 62, 63), but the source of cysteinyl to endothelial monolayers, plastic-bound VCAM-1, and fibronec- leukotrienes and their potential function within the hematopoietic tin. We also show that the effect of the other cysteinyl leukotrienes microenvironment remained unknown. We demonstrated in our

(LTC4 and LTE4) on progenitor adhesion is substantially weaker study the ability of defined bone marrow components (human bone or even absent, which could be explained by their much lower marrow stromal and endothelial cells) to synthesize LTC4, LTD4, affinity for CysLT1. To date, up-regulation of cell adhesion as a and LTE4 in biologically significant amounts. The artificial, se- rum-free culture system in vitro cannot be compared with tissues result of LT-mediated activation of integrins had been demon- by guest on October 2, 2021 strated only for mature inflammatory leukocytes (e.g., neutrophils in vivo. The concentrations and distribution of the cysteinyl leu- and eosinophils) (44–47). kotrienes may therefore be different in the bone marrow microen- vironment. Moreover, in vitro leukotriene production from non- In the present study, the effect of LTD4 on HPC adhesion to endothelial cells was time- and dose-dependent and not restricted myeloid cells could be due to contamination with leukocytes to a particular hematopoietic progenitor lineage, because primitive, donating LTA4. Because particularly the enzyme 5-LO is largely pluripotent progenitors also adhered more avidly after stimulation confined to hematopoietic cells, only a few previous studies report LT biosynthesis in nonhematopoietic (i.e., endothelial and epithe- with the CysLT1 ligand. These observations are consistent with studies addressing the modulatory effect of chemokines like SDF-1 lial) cells (64, 65). Due to the use of a bone marrow-derived en- dothelial cell line to analyze leukotriene biosynthesis, transcellular on integrin-mediated progenitor cell adhesion. There, the effects of synthesis of cysteinyl leukotrienes could be ruled out in our study, SDF-1 are also rapid and transient, and both primitive and com- at least regarding the cell line BMEC-1. Indeed, we could also mitted progenitors respond (29, 48). Considering these data, our demonstrate 5-LO mRNA expression in the BMEC-1 cell line and findings suggest that LTD acts via similar inside-out signaling 4 bone marrow stromal cells from different donors. mechanisms, leading to an alteration in integrin conformation and To minimize transcellular contribution to cysteinyl leukotriene placing the integrin in an activated state. This is confirmed by our production in our stromal cultures, cells were treated with MPA to previous observation that LTD4 acts via G protein-coupled signal ϩ remove hematopoiesis. Treatment of murine and human stromal transduction pathways in human CD34 HPCs (17). The focal layers with MPA has been demonstrated to completely eliminate adhesion kinase-related protein tyrosine kinase Pyk2, which rep- colony-forming hematopoietic capacity (24–26). The presence and resents a key regulator of both inside-out and outside-in signaling survival of mature mast cells or monocytes in our stromal cultures of integrins in myeloid cells (49), was phosphorylated after acti- as sources of LTD4 cannot completely be excluded, but the fact vation of CysLT1 and may play a particular role in LT-induced that the layers were kept in culture for several weeks before anal- HPC adhesion. However, not only inhibition of tyrosine kinases by ysis of leukotriene production renders this possibility rather un- genistein but also specific inhibition of the MEK/ERK/MAPK sig- likely. The observation that the amount of cysteinyl leukotrienes naling pathway reduced LTD4-induced HPC adhesion. The ERK/ was not increased but in contrast drastically reduced in coculture 2-dependent signaling pathways may therefore contribute not only of stroma with hematopoietic progenitors also argues against the to LTD4-mediated proliferation of HPCs in response to cysteinyl remaining hematopoietic cells as a source of LT production. Be- leukotrienes, but also to activation of integrin-dependent adhesion. cause functional primary bone marrow stroma contains a variety of A similar role of ERK1/2 signaling in integrin-dependent adhesion cell types including adipocytes (66), it seems reasonable that sev- has been described in mature hematopoietic cells (e.g., eosino- eral nonhematopoietic cell types might participate in LT produc- phils) (50). tion in these cultures. The Journal of Immunology 6797

The fact that hematopoietic cells down-regulated the amount of 7. Jacobsen, K., J. Kravitz, P. W. Kincade, and D. G. Osmond. 1996. Adhesion LTs present in culture supernatants further underlines a potential receptors on bone marrow stromal cells: in vivo expression of vascular cell ad- hesion molecule-1 by reticular cells and sinusoidal endothelium in normal and role of cysteinyl leukotrienes as paracrine mediators within the ␥-irradiated mice. Blood 87: 73–82. hematopoietic microenvironment. In analogy to the chemokine 8. Schweitzer, K. M., A. M. Dra¨ger, P. V. van der Valk, S. F. Thijsen, A. Zevenbergen, A. P. Theijsmeijer, C. E. van der Schoot, and SDF-1, LTs might be produced by components of the bone marrow M. M. Langenhuijsen. 1996. Constitutive expression of E-selectin and vascular microenvironment, resulting in the adhesion of HPCs to the bone cell adhesion molecule-1 on endothelial cells of hematopoietic tissues. marrow endothelium and the subsequent chemotactic migration Am. J. Pathol. 148: 165–175. 9. Mazo, I. B., and U. H. von Andrian. 1999. Adhesion and homing of blood-borne toward stromal niches, here interacting with the LT metabolism. cells in bone marrow microvessels. J. Leukocyte Biol. 66: 25–32. Lack of hematopoietic activity (bone marrow aplasia) could there- 10. Capra, V., M. D. Thompson, A. Sala, D. E. Cole, G. Folco, and G. E. Rovati. fore result in an increase in the amount of LTs produced by stromal 2007. Cysteinyl-leukotrienes and their receptors in asthma and other inflamma- tory diseases: critical update and emerging trends. Med. Res. Rev. 27: 469–527. and/or endothelial cells within the bone marrow, thus triggering 11. Lynch, K. R., G. P. O’Neill, Q. Liu, D. S. Im, N. Sawyer, K. M. Metters, adhesion, homing, and proliferation of hematopoietic progenitors, N. Coulombe, M. Abramovitz, D. J. Figueroa, Z. Zeng, et al. 1999. Character- which represents a negative feedback mechanism. In addition, in- ization of the human cysteinyl leukotriene CysLT1 receptor. Nature 399: 789–793. flammation is often present during aplasia, which can lead to sys- 12. Sarau, H. M., R. S. Ames, J. Chambers, C. Ellis, N. Elshourbagy, J. J. Foley, temic levels of inflammatory mediators such as cysteinyl leukotri- D. B. Schmidt, R. M. Muccitelli, O. Jenkins, P. R. Murdock, et al. 1999. Iden- tification, molecular cloning, expression, and characterization of a cysteinyl leu- enes and further support activation of the CysLT1 expressed on kotriene receptor. Mol. Pharmacol. 56: 657–663. HPC. During steady-state hematopoiesis, however, the local and 13. Heise, C. E., B. F. O’Dowd, D. J. Figueroa, N. Sawyer, T. Nguyen, D. S. Im, systemic cysteinyl leukotriene level may be low without a signif- R. Stocco, J. N. Bellefeuille, M. Abramovitz, R. Cheng, et al. 2000. Character- ization of the human cysteinyl leukotriene 2 receptor. J. Biol. Chem. 275: icant contribution to progenitor cell function. This could explain 30531–30536. Downloaded from the observation that treatment with a CysLT1 antagonist did not 14. Nothacker, H. P., Z. Wang, Y. Zhu, R. K. Reinscheid, S. H. Lin, and O. Civelli. result in progenitor cell mobilization, in contrast to the CXCR4 2000. Molecular cloning and characterization of a second human cysteinyl leu- kotriene receptor: discovery of a subtype selective agonist. Mol. Pharmacol. 58: antagonist AMD3100, which induced an substantial increase in the 1601–1608. number of circulating CD34ϩ HPCs several hours after adminis- 15. Mellor, E. A., A. Maekawa, K. F. Austen, and J. A. Boyce. 2001. Cysteinyl leukotriene receptor 1 is also a pyrimidinergic receptor and is expressed by hu- tration (21). man mast cells. Proc. Natl. Acad. Sci. USA 98: 7964–7969. LTs are not the only lipid mediators that act via GPCR ex- 16. Jiang, Y., L. A. Borrelli, Y. Kanaoka, B. J. Bacskai, and J. A. Boyce. 2007. pressed on HPC. The fatty acid anandamide, which is the natural CysLT2 receptors interact with CysLT1 receptors and down-modulate cysteinyl http://www.jimmunol.org/ leukotriene dependent mitogenic responses of mast cells. Blood 110: 3263–3270. ligand for the peripheral cannabinoid receptor, has been shown to 17. Bautz, F., C. Denzlinger, L. Kanz, and R. Mohle. 2001. Chemotaxis and trans- act as a growth factor for hematopoietic cells (67). Moreover, in endothelial migration of CD34ϩ hematopoietic progenitor cells induced by the murine bone marrow cultures the migratory behavior of hemato- inflammatory mediator leukotriene D4 are mediated by the 7-transmembrane re- ceptor CysLT1. Blood 97: 3433–3440. poietic progenitors depends on the phospholipid sphingosine 18. Mo¨hle, R., F. Bautz, S. Rafii, M. A. Moore, W. Brugger, and L. Kanz. 1998. The 1-phosphate (S1P) (68). In accordance, we were able to demon- chemokine receptor CXCR-4 is expressed on CD34ϩ hematopoietic progenitors strate expression of S1P receptors (S1PRs) in human HPC and and leukemic cells and mediates transendothelial migration induced by stromal cell-derived factor-1. Blood 91: 4523–4530. found that modulation of S1PR activity leads to an increase in the 19. Kanaoka, Y., and J. A. Boyce. 2004. Cysteinyl leukotrienes and their receptors: bone marrow homing of primitive CD34ϩCD38Ϫ HPCs both in cellular distribution and function in immune and inflammatory responses. J. Im- by guest on October 2, 2021 vitro and in vivo by up-regulation of CXCR4-mediated effects munol. 173: 1503–1510. 20. Xue, X., Z. Cai, G. Seitz, L. Kanz, K. C. Weisel, and R. Mo¨hle. 2007. Differential (69). Taken together, in addition to cytokines and other peptide effects of G protein coupled receptors on hematopoietic progenitor cell growth compounds, cysteinyl leukotrienes and other lipid mediators are depend on their signaling capacities. Ann. NY Acad. Sci. 1106: 180–189. 21. Devine, S. M., R. Vij, M. Rettig, L. Todt, K. McGlauchlen, N. Fisher, H. Devine, likely to play specific roles in the crosstalk between HPC and other D. C. Link, G. Calandra, G. Bridger, et al. 2008. Rapid mobilization of functional components of the bone marrow microenvironment, modulating donor hematopoietic cells without G-CSF using AMD3100, an antagonist of the HPC retention, proliferation, and trafficking. CXCR4/SDF-1 interaction. Blood 112: 990–998. 22. Yahata, T., K. Ando, T. Sato, H. Miyatake, Y. Nakamura, Y. Muguruma, S. Kato, and T. Hotta. 2003. A highly sensitive strategy for SCID-repopulating cell assay Acknowledgments by direct injection of primitive human hematopoietic cells into NOD/SCID mice bone marrow. Blood 101: 2905–2913. We thank Christine Zimmermann for excellent technical assistance as well 23. Mentzer, S. J., M. A. Crimmins, S. J. Burakoff, and D. V. Faller. 1987. Alpha and as Hans-Jo¨rg Bu¨hring and Michael Schumm for carefully reading the beta subunits of the LFA-1 membrane molecule are involved in human mono- manuscript. cyte-endothelial cell adhesion. J. Cell. Physiol. 130: 410–415. 24. Johnson, A., and K. Dorshkind. 1986. Stromal cells in myeloid and lymphoid long-term bone marrow cultures can support multiple hemopoietic lineages and Disclosures modulate their production of hemopoietic growth factors. Blood 68: 1348–1354. The authors have no financial conflict of interest. 25. Dorshkind, K., A. Johnson, L. Collins, G. M. Keller, and R. A. Phillips. 1986. Generation of purified stromal cell cultures that support lymphoid and myeloid precursors. J. Immunol. Methods 89: 37–47. References 26. Zhang, W., G. Knieling, G. Vohwinkel, T. Martinez, R. Kuse, D. K. Hossfeld, 1. Lapidot, T., and I. Petit. 2002. Current understanding of stem cell mobilization: and U. Duhrsen. 1999. Origin of stroma cells in long-term bone marrow cultures the roles of chemokines, proteolytic enzymes, adhesion molecules, cytokines, and from patients with acute myeloid leukemia. Ann. Hematol. 78: 305–314. stromal cells. Exp. Hematol. 30: 973–981. 27. Jaffe, E. A., R. L. Nachman, C. G. Becker, and C. R. Minick. 1973. Culture of 2. Peled, A., O. Kollet, T. Ponomaryov, I. Petit, S. Franitza, V. Grabovsky, human endothelial cells derived from umbilical veins. Identification by morpho- M. M. Slav, A. Nagler, O. Lider, R. Alon, et al. 2000. The chemokine SDF-1 logic and immunologic criteria. J. Clin. Invest. 52: 2745–2756. activates the integrins LFA-1, VLA-4, and VLA-5 on immature human CD34ϩ 28. Candal, F. J., S. Rafii, J. T. Parker, E. W. Ades, B. Ferris, R. L. Nachman, and cells: role in transendothelial/stromal migration and engraftment of NOD/SCID K. L. Kellar. 1996. BMEC-1: a human bone marrow microvascular endothelial mice. Blood 95: 3289–3296. cell line with primary cell characteristics. Microvasc. Res. 52: 221–234. 3. Lataillade, J. J., D. Clay, C. Dupuy, S. Rigal, C. Jasmin, P. Bourin, and 29. Mo¨hle, R., M. A. Moore, R. L. Nachman, and S. Rafii. 1997. Transendothelial ϩ M. C. Bousse-Kerdiles. 2000. Chemokine SDF-1 enhances circulating CD34ϩ migration of CD34 and mature hematopoietic cells: an in vitro study using a cell proliferation in synergy with cytokines: possible role in progenitor survival. human bone marrow endothelial cell line. Blood 89: 72–80. Blood 95: 756–768. 30. Breems, D. A., E. A. Blokland, S. Neben, and R. E. Ploemacher. 1994. Frequency 4. Denzlinger, C. 1996. Biology and pathophysiology of leukotrienes. Crit. Rev. analysis of human primitive haematopoietic stem cell subsets using a cobblestone Oncol. Hematol. 23: 167–223. area forming cell assay. Leukemia 8: 1095–1104. 5. Busse, W. W. 1998. Leukotrienes and inflammation. Am. J. Respir. Crit. Care 31. Ploemacher, R. E., J. P. van der Sluijs, J. S. Voerman, and N. H. Brons. 1989. An Med. 157: S210–S213. in vitro limiting-dilution assay of long-term repopulating hematopoietic stem 6. Christie, P. E., and W. R. Henderson, Jr. 2002. Lipid inflammatory mediators: cells in the mouse. Blood. 74: 2755–2763. leukotrienes, prostaglandins, platelet-activating factor. Clin. Allergy Immunol. 32. Denzlinger, C., C. Haberl, and W. Wilmanns. 1995. Cysteinyl leukotriene pro- 16: 233–254. duction in anaphylactic reactions. Int. Arch. Allergy Immunol. 108: 158–164. 6798 LTD4 SUPPORTS PROGENITOR ADHESION AND PROLIFERATION

33. Nielsen, C. K., R. Massoumi, M. Sonnerlind, and A. Sjolander. 2005. Leukotri- adhesion of human eosinophils caused by ERK1/2-mediated activation of cPLA2. ene D4 activates distinct G-proteins in intestinal epithelial cells to regulate stress J. Leukocyte Biol. 72: 1046–1053. fibre formation and to generate intracellular Ca2ϩ mobilisation and ERK1/2 51. Miller, A. M., R. S. Weiner, and V. A. Ziboh. 1986. Evidence for the role of activation. leukotrienes C4 and D4 as essential intermediates in CSF-stimulated human my- 34. Capra, V., S. Ravasi, M. R. Accomazzo, M. Parenti, and G. E. Rovati. 2004. eloid colony formation. Exp. Hematol. 14: 760–765. CysLT1 signal transduction in differentiated U937 cells involves the activation of 52. Stenke, L., M. Mansour, P. Reizenstein, and J. A. Lindgren. 1993. Stimulation of the small GTP-binding protein Ras. Biochem. Pharmacol. 67: 1569–1577. human myelopoiesis by leukotrienes B4 and C4: interactions with granulocyte- 35. Recher, C., L. Ysebaert, O. Beyne-Rauzy, V. Mansat-De Mas, J. B. Ruidavets, macrophage colony-stimulating factor. Blood 81: 352–356. P. Cariven, C. Demur, B. Payrastre, G. Laurent, and C. Racaud-Sultan. 2004. 53. Murakami, M., K. F. Austen, C. O. Bingham III, D. S. Friend, J. F. Penrose, and Expression of focal adhesion kinase in acute myeloid leukemia is associated with J. P. Arm. 1995. Interleukin-3 regulates development of the 5-lipoxygenase/leu- enhanced blast migration, increased cellularity, and poor prognosis. Cancer Res. kotriene C4 synthase pathway in mouse mast cells. J. Biol. Chem. 270: 64: 3191–3197. 22653–22656. 36. Lee, Y., A. Gotoh, H. J. Kwon, M. You, L. Kohli, C. Mantel, S. Cooper, 54. Denzlinger, C., J. Walther, W. Wilmanns, and H. H. Gerhartz. 1993. Interleu- G. Hangoc, K. Miyazawa, K. Ohyashiki, and H. E. Broxmeyer. 2002. Enhance- kin-3 enhances the endogenous leukotriene production. Blood 81: 2466–2468. ment of intracellular signaling associated with hematopoietic progenitor cell sur- 55. Thivierge, M., J. Stankova, and M. Rola-Pleszczynski. 2001. IL-13 and IL-4 vival in response to SDF-1/CXCL12 in synergy with other cytokines. Blood 99: up-regulate cysteinyl leukotriene 1 receptor expression in human monocytes and 4307–4317. macrophages. J. Immunol. 167: 2855–2860. 37. Funk, C. D. 2001. Prostaglandins and leukotrienes: advances in eicosanoid biol- 56. Thivierge, M., M. Doty, J. Johnson, J. Stankova, and M. Rola-Pleszczynski. ogy. Science 294: 1871–1875. 2000. IL-5 up-regulates cysteinyl leukotriene 1 receptor expression in HL-60 38. Lipworth, B. J. 1999. Leukotriene-receptor antagonists. Lancet 353: 57–62. cells differentiated into eosinophils. J. Immunol. 165: 5221–5226. 39. Balani, S. K., X. Xu, V. Pratha, M. A. Koss, R. D. Amin, C. Dufresne, 57. Mellor, E. A., K. F. Austen, and J. A. Boyce. 2002. Cysteinyl leukotrienes and R. R. Miller, B. H. Arison, G. A. Doss, M. Chiba, et al. 1997. Metabolic profiles uridine diphosphate induce cytokine generation by human mast cells through an of montelukast sodium (Singulair), a potent cysteinyl leukotriene1 receptor an- interleukin 4-regulated pathway that is inhibited by leukotriene receptor antago- tagonist, in human plasma and bile. Drug Metab. Dispos. 25: 1282–1287. nists. J. Exp. Med. 195: 583–592. 40. Mo¨hle, R., F. Bautz, S. Rafii, M. A. Moore, W. Brugger, and L. Kanz. 1999. 58. Rola-Pleszczynski, M., and J. Stankova. 1992. Leukotriene B4 enhances inter- Regulation of transendothelial migration of hematopoietic progenitor cells. Ann. leukin-6 (IL-6) production and IL-6 messenger RNA accumulation in human NY Acad. Sci. 872: 176–185. monocytes in vitro: transcriptional and posttranscriptional mechanisms. Blood. Downloaded from 41. Frenette, P. S., S. Subbarao, I. B. Mazo, U. H. von Andrian, and D. D. Wagner. 80: 1004–1011. 1998. Endothelial and vascular cell adhesion molecule-1 promote he- 59. Gagnon, L., L. G. Filion, C. Dubois, and M. Rola-Pleszczynski. 1989. Leuko- matopoietic progenitor homing to bone marrow. Proc. Natl. Acad. Sci. USA 95: trienes and macrophage activation: augmented cytotoxic activity and enhanced 14423–14428. interleukin 1, tumor necrosis factor and hydrogen peroxide production. Agents 42. Mazo, I. B., J. C. Gutierrez-Ramos, P. S. Frenette, R. O. Hynes, D. D. Wagner, Actions 26: 141–147. and U. H. von Andrian. 1998. Hematopoietic progenitor cell rolling in bone 60. Lindgren, J. A., L. Stenke, M. Mansour, C. Edenius, L. Lauren, marrow microvessels: parallel contributions by endothelial selectins and vascular B. Nasman-Glaser, I. Ericsson, and P. Reizenstein. 1993. Formation and effects

cell adhesion molecule 1. J. Exp. Med. 188: 465–474. of leukotrienes and lipoxins in human bone marrow. J. Lipid Mediat. 6: 313–320. http://www.jimmunol.org/ 43. van der Loo, J. C., X. Xiao, D. McMillin, K. Hashino, I. Kato, and 61. Martinez-Moczygemba, M., and D. P. Huston. 2003. Biology of common ␤ re- D. A. Williams. 1998. VLA-5 is expressed by mouse and human long-term re- ceptor-signaling cytokines: IL-3, IL-5, and GM-CSF. J. Allergy Clin. Immunol. populating hematopoietic cells and mediates adhesion to extracellular matrix pro- 112: 653–665. tein fibronectin. J. Clin. Invest. 102: 1051–1061. 62. Stenke, L., L. Lauren, P. Reizenstein, and J. A. Lindgren. 1987. Leukotriene 44. Friedrich, E. B., A. M. Tager, E. Liu, A. Pettersson, C. Owman, L. Munn, production by fresh human bone marrow cells: evidence of altered lipoxygenase A. D. Luster, and R. E. Gerszten. 2003. Mechanisms of leukotriene B4-triggered activity in chronic myelocytic leukemia. Exp. Hematol. 15: 203–207. monocyte adhesion. Arterioscler. Thromb. Vasc. Biol. 23: 1761–1767. 63. Stenke, L., M. Mansour, C. Edenius, P. Reizenstein, and J. A. Lindgren. 1991. 45. Damtew, B., and P. J. Spagnuolo. 1997. Leukotriene C4 from vascular endothe- Formation and proliferative effects of lipoxins in human bone marrow. Biochem. lium enhances neutrophil adhesiveness. Prostaglandins Leukot. Essent. Fatty Ac- Biophys. Res. Commun. 180: 255–261. ids 56: 111–116. 64. Zhang, Y. Y., J. L. Walker, A. Huang, J. F. Keaney, C. B. Clish, C. N. Serhan, 46. Nagata, M., K. Saito, K. Tsuchiya, and Y. Sakamoto. 2002. Leukotriene D4 and J. Loscalzo. 2002. Expression of 5-lipoxygenase in pulmonary artery endo-

upregulates eosinophil adhesion via the cysteinyl leukotriene 1 receptor. thelial cells. Biochem. J. 361: 267–276. by guest on October 2, 2021 J. Allergy Clin. Immunol. 109: 676–680. 65. Luo, M., S. Lee, and T. G. Brock. 2003. Leukotriene synthesis by epithelial cells. 47. Fregonese, L., M. Silvestri, F. Sabatini, and G. A. Rossi. 2002. Cysteinyl leu- Histol. Histopathol. 18: 587–595. kotrienes induce human eosinophil locomotion and adhesion molecule expression 66. Fuchs, E., T. Tumbar, and G. Guasch. 2004. Socializing with the neighbors: stem via a CysLT1 receptor-mediated mechanism. Clin. Exp. Allergy 32: 745–750. cells and their niche. Cell 116: 769–778. 48. Hidalgo, A., F. Sanz-Rodriguez, J. L. Rodriguez-Fernandez, B. Albella, C. Blaya, 67. Valk, P., S. Verbakel, Y. Vankan, S. Hol, S. Mancham, R. Ploemacher, N. Wright, C. Cabanas, F. Prosper, J. C. Gutierrez-Ramos, and J. Teixido. 2001. A. Mayen, B. Lowenberg, and R. Delwel. 1997. Anandamide, a natural ligand for Chemokine stromal cell-derived factor-1␣ modulates VLA-4 integrin-dependent the peripheral cannabinoid receptor is a novel synergistic growth factor for he- adhesion to fibronectin and VCAM-1 on bone marrow hematopoietic progenitor matopoietic cells. Blood 90: 1448–1457. cells. Exp. Hematol. 29: 345–355. 68. Yanai, N., N. Matsui, T. Furusawa, T. Okubo, and M. Obinata. 2000. Sphin- 49. Miura, Y., Y. Tohyama, T. Hishita, A. Lala, E. De Nardin, Y. Yoshida, gosine-1-phosphate and lysophosphatidic acid trigger invasion of primitive he- H. Yamamura, T. Uchiyama, and K. Tohyama. 2000. Pyk2 and Syk participate in matopoietic cells into stromal cell layers. Blood 96: 139–144. functional activation of granulocytic HL-60 cells in a different manner. Blood 96: 69. Kimura, T., A. M. Boehmler, G. Seitz, S. Kuci, T. Wiesner, V. Brinkmann, 1733–1739. L. Kanz, and R. Mo¨hle. 2004. The sphingosine 1-phosphate receptor agonist 50. Zhu, X., B. Jacobs, E. Boetticher, S. Myou, A. Meliton, H. Sano, FTY720 supports CXCR4-dependent migration and bone marrow homing of hu- A. T. Lambertino, N. M. Munoz, and A. R. Leff. 2002. IL-5-induced integrin man CD34ϩ progenitor cells. Blood. 103: 4478–4486.