The CysLT₁ Ligand Leukotriene D₄ Supports α₄β₁- and α₅β₁-Mediated Adhesion and Proliferation of CD34⁺ Hematopoietic Progenitor Cells

Reagents and Abs

The LTs C₄, D₄, and E₄ were obtained from Paesel and Lorei or Biomol. Purity and integrity of the lots were assessed by HPLC analysis. Working solutions were prepared using HPLC-grade methanol (Merck) and were kept below 0°C and under argon atmosphere at all times to avoid formation of biologically inactive isomers. The compounds MK-571, LY171883, PTX, genistein, and PD98059 were obtained from Biomol.

Abs used in the adhesion assays were mAbs that have been shown to functionally block the respective adhesion molecule (22, 23) VLA-4 (CD49d, clone HP2/1; Beckman Coulter-Immunotech), VLA-5 (CD49e, clone SAM-1; Immunotech), LFA-1 (CD11a, clone TS 1/11; Endogen), β₂ integrin (CD18, clone TS 1/18; Endogen), or IgG-matched isotype controls (BD Biosciences). All Abs were used at a final concentration of 12.5 μg/ml.

Human cells and cell lines

After informed consent, peripheral blood cells were obtained from healthy donors or patients with nonhematological malignancies during G-CSF-induced stem cell mobilization according to the guidelines of the ethical committee of the University of Tübingen, Tübingen, Germany (project no. 268/2003-LP). Mononuclear cells were separated by Ficoll density gradient centrifugation. Enrichment of CD34⁺ cells was performed using immunomagnetic microbeads (MACS system; Miltenyi Biotec) according to the manufacturer’s instructions (purity ≥ 96%). Primary bone marrow cells were isolated after informed consent from healthy volunteers (project no. 268/2003-BM). Bone marrow CD34⁺ cells were isolated by Ficoll separation of mononuclear cells and the subsequent enrichment of CD34⁺ cells by MACS as described above for peripheral blood-derived progenitors. Adherent stromal cells were isolated and cultured as previously described (17). To deplete stromal cultures of contaminating hematopoietic cells, the layers were treated with three cycles of 5 μg/ml mycophenolic acid (MPA; Sigma Aldrich), each cycle consisting of a 3-day period with MPA followed by a 4-day recovery period without MPA (24, 25, 26). Stromal cells were then washed twice with PBS and maintained for at least 1 wk in MPA-free medium (including two complete changes of medium) before using them in experiments. HUVECs (27) were grown in endothelial cell growth medium (PromoCell) plus 10% FCS. Cells were used between the fifth and eighth passage. The human microvascular BMEC line BMEC-1 (28, 29) was cultured in endothelial cell growth medium plus 10% FCS. The CD34⁺ CysLT1⁺ human acute myeloid leukemia cell lines KG-1a and Kasumi-1 as well as the CysLT₁-negative T lymphoblastic cell line Jurkat were propagated in RPMI 1640 medium with 10% FCS, 100 U/ml penicillin, and 100 μg/ml streptomycin.

Flow cytometric analysis

Expression of glycophorin-A, CD14, CD15, CD34, CD38, CD41, and CD45 was evaluated by incubating 2–5 × 10⁵ cells with the respective FITC- or PE-labeled Abs (BD Biosciences) for 20 min at 4°C and analyzed on a FACSCalibur flow cytometer equipped with CellQuestPro software. Isotype-matched IgG Abs served as negative controls. A proportion of 1% or fewer false positive events were accepted in the negative control samples throughout.

Cell adhesion assays

To prepare HUVEC monolayers, 1 × 10⁴ cells were seeded on gelatin-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 10% FCS. Before adding HPCs, the monolayers were stimulated with 10 ng/ml IL-1β (Boehringer Mannheim) for 4 h. For the adhesion assays on recombinant human (rh) VCAM-1, 96-well microtiter plates were coated for 90 min at 37°C with 50 μl of a 5 μg/ml solution of carrier protein-free VCAM-1 (R&D Systems) diluted in 0.05 M bicarbonate/carbonate buffer (pH 9.6). The plate was washed with PBS to remove unbound protein, and nonspecific binding sites were blocked using a 0.1% (w/v) solution of BSA in PBS (1 h at 37°C). Before the experiment, wells were washed twice with 200 μl/well assay medium. Adhesion experiments on fibronectin were performed using BioCoat Cellware human fibronectin 96-well Multiwell plates (BD Biosciences) that were preincubated for 1 h with 100 μl/well assay medium at 37°C in 5% CO₂ before adding HPCs.

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 experiments, cells were preincubated for 2–5 min with MK-571 (final concentration 2 μM), LY17183 (10 μM), or PTX (20 ng/ml). Cells (6 × 10⁴ in 100 μl) were added in triplicate to the precoated plates, spun down for 15 s at 30 × g, allowed to adhere for 5 min at 37°C in 5% CO₂ and then washed twice with warm assay medium to remove nonadherent cells. Bound cells were harvested with trypsin/EDTA, washed in RPMI 1640 plus 10% FCS to inactivate the trypsin, and subsequently quantified and/or characterized by flow cytometry, counted, and assessed for CFU or transferred into long-term cultures on FBMD-1 stromal feeders (see below).

For quantification, recovered cells from the adhesion assays were suspended in PBS containing 0.1% BSA and stained with anti-CD34-PE. FITC-conjugated calibration beads (3 × 10⁴) (CaliBRITE; BD Biosciences) were added, and samples were analyzed by FACS. FITC-beads were gated as G1 = R1 (where G is gate and R is region), and data acquisition was stopped after 5,000 acquired events of beads. The number of adherent cells per well was thus calculated from the ratio of HPC to beads multiplied by the number of beads added.

Cytokine-supplemented liquid culture of CD34⁺ cells

Isolated CD34⁺ progenitors were seeded into 24-well cell culture plates (Costar) at an input concentration of 5 × 10⁴ cells/ml in X-Vivo 20 containing stem cell factor, IL-3, and FLT3 ligand (100 ng/ml each; all from PeproTech) and the indicated concentrations of leukotrienes, and incubated at 37°C with 5% CO₂ in a humidified atmosphere for 1 wk. Experimental setups with LTs must take into consideration that these compounds are rapidly metabolized and degraded in most systems, a process due to surface enzymes and/or enzymatic active components in the incubation medium (4). Therefore, LTs were added at days 0, 2, and 4 to allow stimulation of cells at multiple time points. At the end of the incubation period, cells were quantified in a hematocytometer. Viability was assessed by trypan blue exclusion. For further characterization of the proliferated cells, aliquots of the cell samples were labeled for flow cytometry. Additional subsets were reseeded into methylcellulose clonogenic cultures and cobblestone area-forming cell (CAFC)-assays as described below.

Colony-forming cell assays

Clonogenic progenitor assays were performed by plating equal numbers of cells (500/ml) either from the starting populations or harvested from the adhesion/proliferation assays in duplicate with MethoCult H4230 (Stem Cell Technologies). Methylcellulose medium was supplemented with rhSCF (stem cell factor), rhIL-3, rhIL-6 (20 ng/ml each; all from PeproTech), rhG-CSF (100 ng/ml; Amgen), and rhEPO (epoetin-α, 6 U/ml; Jannsen-Cilag). Colonies were counted after 14 days of culture in a humidified environment at 37°C and 5% CO₂.

CAFC assays

Stromal feeders were prepared by seeding 10³ FBMD-1 cells (passages 16–19) per well into fibronectin-coated 96-well plates (BioCoat Cellware) (30). Cells were grown to confluence in MyeloCult H 5100 medium (Stem Cell Technologies) supplemented with gentamicin (100 mg/l; Boehringer Mannheim) and hydrocortisone (10⁻⁶ M; Sigma Aldrich). During coculture with HPCs, the medium was additionally enriched with 10 ng/ml IL-3 and 20 ng/ml G-CSF (CAFC medium). For CAFC analysis, FBMD-1 cells were overlaid with CD34⁺ cells in a limiting dilution setup, with 12 dilution steps 2-fold apart and 12 replicate wells per dilution for each set of experiments. To assess the frequency of cobblestone area-initiating cells within the CD34⁺ populations adherent to VCAM-1, cells harvested from the adhesion assays were spun down, resuspended in equal volumes of CAFC medium, and transferred 1:1 without prior counting, i.e., the content from one well of the adhesion assay was transferred to one well in the row with the highest cell density of the CAFC assay plate. For the cells recovered from the proliferation experiments, starting input values ranged between 5000 and 8000 cells/well. Once a week half of the medium (50 μl) was replaced with fresh CAFC medium. The appearance of cobblestone areas, consisting of at least five cells, was evaluated after 5 wk with a phase contrast illumination on an inverted microscope. CAFC frequencies were calculated as single hit kinetics from the fraction of nonresponding cultures using the Poisson equation as described (30, 31).

Analysis of cysteinyl leukotriene production

Stromal cells and confluent layers of BMEC-1 cells were washed with 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 added to 8 ml of 90% aqueous methanol (Merck) containing 0.5 mmol/L EDTA and 1 mmol/L of the radical scavenger 4-hydroxy-2,2,6,6,-tetramethylpiperidine-1-oxyl (pH 5.6; both from Sigma-Aldrich) (32). The methanolic medium suspension was stored at −80°C for at least 24 h and then spun down at 5,000 × g for 20 min at −10°C. The supernatants were concentrated to dryness in a centrifuge under a reduced atmosphere and dissolved in 30% aqueous methanol for HPLC analysis. Immediately after collection of the conditioned medium, cells were harvested with trypsin plus EDTA and counted. Equal amounts of nonconditioned assay medium with or without known amounts of LTC₄, LTD₄, and LTE₄ run under identical conditions were used as negative control or internal standard, respectively.

To assess the effect of hematopoietic cells on LT production, stromal cultures depleted of hematopoiesis were overlaid or not with 5 × 10⁴ bone marrow-derived CD34⁺ hematopoietic cells per 12.5-cm² culture flask in 5 ml of MyeloCult H 5100 supplemented with gentamicin (100 mg/l) and hydrocortisone (10⁻⁶ M). Both stromal cells and CD34⁺ progenitors were isolated from the same donor source. After 4 wk of culture with a weekly semidepopulation of nonadherent cells, layers were rinsed twice with PBS and cultured for an additional 18 h in X-Vivo 15 medium without phenol red for collection of conditioned medium as described above.

LT profiles were determined by combined use of HPLC and competitive enzyme immunoassay (EIA). HPLC was performed as described (32) using a Hypersil ODS column (4.6 × 100 mm, 5 μm particles; Agilent Technologies) with a Hypersil ODS precolumn (4.0 × 20 mm, 5 μm particles; Agilent) that was run at a flow rate of 1 ml/min on an Agilent 1100 system. 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 hydroxide). Cysteinyl leukotrienes were monitored at 280 nm, and HPLC fractions containing LTC₄, LTD₄, and LTE₄ were isolated and collected for EIA according to the retention times of known standards run under identical conditions. EIA analysis of endogenously produced leukotrienes was performed using cysteinyl leukotriene EIA kits (Cayman Chemical) according to the manufacturer’s instructions. The prior separation by HPLC allowed assessment for the amount of the respective LT metabolites within each fraction.

RT-PCR analysis

Total RNA from BMEC-1 endothelial cells and primary human stroma samples (treated or not with MPA for three cycles as described above) was isolated using an RNeasy mini kit (Qiagen), including digestion with DNase I. First-strand cDNA was synthesized with the use of murine leukemia virus reverse transcriptase and oligo(dT)₁₆ primers (both Applied Biosystems). Subsequent PCR analysis for 5-lipoxygenase (5-LO) and GAPDH (used as internal standard) was performed with gene-specific primers based on the human receptor sequences as follows: 5-LO, 5′-ACTGGAAACACGGCAAAAAC-3′ (forward) and 5′-GTGCAGGGGTCTGTTTTGTT-3′ (reverse; product size 501 bp); GAPDH, 5′-CGGGAAGCTTGTCATCAATGG-3′ (forward) and 5′-GGCAGTGATGGCATGGACTG-3′ (reverse; product size 358 bp). A pre-PCR heat step (5 min at 95°C) provided a PCR Hot-Start by partially activating AmpliTaq Gold DNA polymerase (Applied Biosystems). For analysis of 5-LO and GAPDH expression, 36 and 17 amplification cycles, respectively, were run, each consisting of denaturation for 15 s at 95°C, annealing for 30 s at 58°C, and elongation for 90 s at 72°C, followed by a final extension step of 10 min at 72°C. PCR products were separated by horizontal agarose gel electrophoresis with subsequent ethidium bromide staining. To detect possible contamination of genomic DNA in the RNA sample, each reaction was performed without previous reverse transcription.

Cell lysate preparation and Western blotting

Primary peripheral blood CD34⁺ HPC and cell lines were serum starved for 3 h in X-Vivo 20 medium, resuspended at 2 × 10⁷/ml in X-Vivo 20, and incubated with 100 nM LTD₄ (final concentration) for the indicated time periods at 37°C. After stimulation, whole cell lysates were prepared in buffer containing 50 mM HEPES (pH 7.5), 10% glycerol, 1% Triton X-100, 1.5 mM MgCl₂, 150 mM NaCl, protease inhibitors (Complete Mini; Roche) and the phosphatase inhibitors NaF (100 mM), Na₄P₂O₇ (10 mM), and activated Na₃VO₄ (1 mM; all from Sigma-Aldrich). Equal amounts of protein (20 μg/lane) were separated on a 10% SDS-polyacrylamide gel and transferred onto nitrocellulose membrane. The blots were probed with phospho-specific polyclonal Abs against p44/42 MAPK (Thr²⁰²/Tyr²⁰⁴), Pyk2 (Tyr⁴⁰²), and AKT (Thr³⁰⁸) or with control Abs against the nonphosphorylated forms, respectively (all from Cell Signaling Technology). Bands were visualized with ECL (GE Healthcare).

In vivo effects of the specific CysLT₁ antagonist montelukast on HPC mobilization.

After informed consent as approved by the ethical committee of the University of Tübingen (project number 147/2006V), blood samples for research purposes were obtained from otherwise healthy human subjects receiving treatment with montelukast (independently of the measurements described here) as a prophylaxis of exercise-induced asthma. All subjects had normal blood cell counts and normal findings from a physical examination and were taking no other medication. Patients with allergic asthma or active asthmatic disease or any other abnormal symptoms were excluded. The study design was adopted from studies with the CXCR4 antagonist AMD3100 for stem cell mobilization, which demonstrated an 8-fold increase in circulating CD34⁺ HPC 4 h after administration of the antagonist. In our study, venous blood samples were obtained before and at 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 count and CD34⁺ cell number by flow cytometry.

Statistical analysis

All results are expressed as the mean ± SEM of 3–5 independent experiments, except where indicated. Data analysis was performed with SigmaStat 2.03 software (SPSS). Differences between groups were compared by Student’s t test or paired t test and considered significant at p ≤ 0.05.

LTD₄-induced adhesion of human CD34⁺ progenitors

Preincubation with LTD₄ rapidly up-regulated adhesion of HPCs to endothelium in a time- and dose-dependent manner by up to 80%, with a maximum effect in cells treated with 1 μM LTD₄ for 15 min (Fig. 1⇓A). Under these conditions, 22.17 ± 2.45% of input cells adhered to HUVEC, compared with 12.53 ± 2.08% without LTD₄ preincubation. Increased HPC adhesion was already detectable at LTD₄ concentrations of <1 nM (Fig. 1⇓B). In contrast, LTC₄ weakly up-regulated HPC adhesion to HUVEC at 1 μM, whereas addition of LTE₄ did not increase adhesion at concentrations up to 1 μM (Fig. 1⇓C). LTD₄-triggered adhesion of CD34⁺ cells to endothelium could be reduced by mAbs against α₄ integrin, but not by α_L and β₂ integrin mAbs, indicating an effect of LTD₄ on VLA-4-dependent mechanisms in this system (Fig. 1⇓D). When CD34⁺ cells pretreated with LTD₄ under optimum conditions (1 μM/15 min) were brought in contact with the VLA-4 ligand VCAM-1 immobilized on a plastic surface, adhesion was >2-fold increased compared with that of untreated cells (Fig. 1⇓E). Adhesion could be completely blocked by the α₄ integrin mAb (p < 0.001 vs isotype plus LTD₄), demonstrating its specificity. The effect of LTD₄ on HPC adhesion to HUVECs and VCAM-1 could be significantly blocked with the CysLT₁ antagonists MK-571 (p < 0.01 vs LTD₄ in Fig. 1⇓D and p < 0.05 vs isotype plus LTD₄ in Fig. 1⇓E, respectively) and LY171883 (p < 0.01 vs LTD₄), respectively, demonstrating the involvement of CysLT₁ (Fig. 1⇓, D and E). The data also ruled out the potential participation of multidrug resistance protein 1 in LTD₄-induced up-regulation of HPC adhesion, because only MK-571 but not LY171883 recognizes multidrug resistance protein 1. PTX only partially blocked LTD₄-induced adhesion of HPC to HUVEC, indicating that CysLT₁ signals via both PTX-sensitive (i.e., G_i) and PTX-insensitive (e.g., G_q) G proteins (Fig. 1⇓D). Adhesion of HPC to fibronectin, an extracellular matrix component serving as a β₁ ligand, was also significantly increased after pretreatment with LTD₄ (Fig. 1⇓E), similar to the experiments examining adhesion to endothelium or immobilized VCAM-1. Again, MK-571 abrogated the triggering effect of LTD₄ (p < 0.05 vs isotype plus LTD₄). Adhesion was partially inhibited by Abs against α₄ integrin (p < 0.05 vs isotype plus LTD₄) and α₅ integrin (p < 0.05 vs isotype plus LTD₄) in this model system. Considering that the already spontaneous adhesion to fibronectin is mediated by integrins, complete inhibition would result in adhesion below that of an unstimulated control and reach the value of BSA-coated wells, as observed for HPC adhesion to VCAM, which is solely mediated by α₄ integrin. The data point out an effect of LTD₄ not only on VLA-4 (α₄β₁ integrin), but also on VLA-5 (α₅β₁ integrin)-mediated mechanisms. Both integrins are expressed on HPCs and serve as receptors for fibronectin.

• Download figure
• Open in new tab
• Download powerpoint

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 1 μM LTD₄. Adhesion is expressed as relative adhesion (rel.) compared with the respective control (carrier solution only). ∗, p < 0.05; ∗∗, p < 0.01 vs control (Ctr). B, Dose dependence of HPC adhesion to HUVEC monolayers. CD34⁺ cells were incubated with the indicated concentrations of LTD₄ for 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⁺ cells were incubated in the presence (+) or absence (−) of LTD₄, LTC₄, or LTE₄ (all at 1 μM) for 15 min, followed by exposure to activated HUVECs. ∗, p < 0.05 vs, control without leukotrienes. D, Effect of blocking Abs to VLA-4 and β₂ integrins or receptor antagonists on LTD₄-induced adhesion to HUVECs. CD34⁺ cells were incubated with mAb, isotype-matched IgG, CysLT₁ antagonists (MK-571, LY17183), or PTX and stimulated with 1 μM LTD₄ 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 μM/15 min). The specificity of the effects was assessed with MK-571 and Abs against VLA-4 (α₄) and VLA-5 (α₅). Wells coated with BSA served as control for unspecific binding. ∗∗, p < 0.01; and ∗∗∗, p < 0.001 vs isotype (Iso.).

VLA-4-mediated adhesion of both primitive and committed progenitors was up-regulated by LTD₄

HPCs harvested from the adhesion experiments on VCAM-1 were further characterized. The percentage of the more immature CD34⁺CD38⁻ cells among the CD34⁺ progenitors that adhered to VCAM-1 in response to LTD₄ was similar to that of spontaneously adhering cells and tended to be greater than the percentage of immature progenitors among the initially isolated CD34⁺ cells before adhesion, but the difference was not statistically significant (Fig. 2⇓A). Analogous results were seen for CD34⁺CD90⁺ cells, a subpopulation that is also enriched for more primitive progenitors, indicating that the effect of LTD₄ on VLA-4-mediated adhesion was not dependent on progenitor maturation. With regard to lineage-committed progenitors, preincubation of HPCs with LTD₄ (1 μM/15 min) also resulted in a significant increase in the total numbers of attached BFU-E (erythroid burst-forming unit) and CFU-GM cells as well as in the overall number of attached clonogenic cells, whereas the amount of adhered CFU-M and CFU-Mix cells was only moderately enhanced (Fig. 2⇓B). Similarly, preincubation with LTD₄ consistently resulted in up-regulation of CAFC adhesion to a donor-specific extent (up to 238%; Fig. 2⇓C), which underscored that LTD₄ also affected VLA-4-mediated adhesion of more primitive progenitor cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

LTD₄-induced adhesion of primitive and more committed CD34⁺ HPCs to VCAM-1. A, HPC phenotype in the starting population and after adhesion. More primitive, immature CD34^high/CD38^low and CD34^high/CD90⁺ progenitor cells (rectangular gates) were quantified in the starting population before adhesion (left), in spontaneously adhering cells (middle; Medium), and in cells adhering after stimulation with 1 μM LTD₄ for 15 min (right). Data are expressed as percentage of gated events. B, Adhesion of lineage-committed HPCs (CFU). CD34⁺ cells were stimulated with 1 μM LTD₄ for 15 min before exposure to rhVCAM. Adhering cells were harvested and plated in a semisolid colony-assay. ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001 vs 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 with HPC harvested from adhesion experiments on VCAM-1 by a limiting dilution setup. Data are presented as number of CAFCs per 10⁴ initial cells brought in contact with VCAM-1 (cells from three different donors were analyzed).

LTD₄ up-regulated proliferation of CD34⁺ HPC in vitro

Analysis of cell numbers after ex vivo expansion of CD34⁺ HPC in serum-free, cytokine-supplemented medium for 1 wk in the presence or absence of the indicated concentrations of LTD₄ demonstrated a dose-dependent effect on cell proliferation in cytokine-supplemented liquid cultures (Fig. 3⇓A). The presence of 100 nM LTD₄ resulted in a 2-fold increase in the total cell number (Fig. 3⇓B). In contrast, the presence of LTC₄ and LTE₄ augmented HPC proliferation only weakly (18 and 12% increase, respectively), which was not statistically significant compared with proliferation without LT addition. Proliferation experiments using varying cytokine combinations revealed an effect of LTD₄ on cell expansion predominantly in samples containing IL-3 (Fig. 3⇓C). After 1 wk of cytokine-supplemented liquid culture, lineage-committed and more primitive progenitors were increased in the presence of 100 nM LTD₄, as demonstrated by the expansion of clonogenic progenitors (Fig. 3⇓D) and the concentration of primitive CAFCs present in the culture medium, as determined by limiting dilution analysis (Fig. 3⇓E). As CAFCs represent only a small fraction of the CD34⁺ progenitors with a distinct proliferation capacity, we also chose the 1 μM concentration, which was highly active in the adhesion experiments, for analysis of CAFC proliferation.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

LT-induced proliferation of CD34⁺ HPC. A, Effect of LTD₄ on ex vivo expansion of HPC. CD34⁺ cells were cultured in cytokine-supplemented serum-free medium (control (Ctr), ∼10-fold increase after 1 wk) and compared with those additionally supplemented with the indicated concentrations of LTD₄. ∗, p < 0.05; ∗∗, p < 0.01 vs control. rel., Relative adhesion (compared with control) B, LT-induced expansion of HPCs. CD34⁺ cells were incubated in cytokine-supplemented, serum-free medium in the presence (+) or absence (−) of 1 μM LTD₄, LTC₄, or LTE₄, respectively, before the amount of cells was assessed. ∗, p < 0.05 vs control. C, Effect of cytokines on expansion of HPCs. The indicated cytokines were added (100 ng/ml each) in the presence or absence of 100 nM LTD₄, and the cell number was assessed after 1 wk. ∗, p < 0.05; ∗∗, p < 0.01 vs control. SCF, Stem cell factor. D, Effect of LTD₄ on HPC expansion of committed progenitors (CFU). HPCs before and after 1 wk of culture in cytokine-supplemented, serum-free medium with or without 100 nM LTD₄ were analyzed for their clonogenic potential. Plating efficiency was assessed by plating equal numbers of cells in methylcellulose. By multiplying with the corresponding total cell number, the CFU expansion was calculated. ∗, p < 0.05 vs control. E, Effect of LTD₄ on primitive 5-wk CAFCs in cytokine-supplemented liquid culture. Long-term cultures on FBMD-1 stromal feeders were set up with cells cultured for 1 wk in cytokine supplemented serum-free medium by a limiting dilution setup (12 replicate wells/condition; 12 dilution steps). ∗, p < 0.05 vs control.

FACS analysis of cells expanded under the above mentioned conditions in the presence or absence of 100 nM LTD₄ showed no marked differences in the expression patterns of the differentiation markers CD14, CD15, CD38, CD41, and glycophorin A, demonstrating an equal proliferative effect of LTD₄ on the monocytic, granulocytic, megakaryocytic, and erythropoietic lineages (Table I⇓).

LTD₄ activated major signaling pathways in CD34⁺ cells

To understand possible molecular mechanisms involved in the effects of LTD₄ on progenitor cell adhesion and proliferation, CD34⁺ cells were stimulated with 100 nM LTD₄ and analyzed by Western blotting for the activation of signaling molecules using phospho-specific Abs. As shown in Fig. 4⇓A, LTD₄ rapidly induced the activation of p44/42 MAPK (ERK1/2) in both CD34⁺ cell lines as well as in the primary mobilized CD34⁺ HPCs, whereas LTD₄ had no effect on the Jurkat T cell line that does not express CysLT₁ and therefore served as a negative control. An effect of LTD₄ on ERK p44/42 MAPK signaling has been reported in other cells (including differentiated epithelial and hematopoietic cells) (33, 34), but not in hematopoietic progenitor and stem cells to date. Additionally, the nonreceptor tyrosine kinase Pyk2 was phosphorylated after LTD₄ treatment (Fig. 4⇓B). Pyk2 represents a homologue of the focal adhesion kinase and is involved in the integrin-signaling pathway, particularly in CD34⁺ hematopoietic progenitor cells that do not express the focal adhesion kinase (35). The effects of LTD₄ on p44/42 MAPK and Pyk2 support our observations that LTD₄ triggers not only chemotaxis (17) but also integrin-mediated adhesion and proliferation of CD34⁺ HPCs. Interestingly, protein kinase B (AKT), a major signal molecule that has been implicated in progenitor cell survival and proliferation (36), was less affected by LTD₄ treatment under the conditions used in our studies (Fig. 4⇓C). 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 pathway (p < 0.05 and p < 0.001 vs LTD₄, respectively, shown in Fig. 4⇓D). 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, as they completely block any proliferation in the assay systems used.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

LTD₄-induced signal transduction in CD34⁺ cell lines and HPCs. A, Phosphorylation of p44/42 MAPK (P-44/42 MAPK) in cell lines and primary CD34⁺ HPC in response to LTD₄. KG1a and Kasumi-1 were analyzed by Western blotting after stimulation with LTD₄ (100 nM) for up to 20 min. Jurkat cells not expressing CysLT₁ served as a negative control. Primary CD34⁺ cells were analyzed before (0) and after stimulation with LTD₄ for 1 min. PB, Peripheral blood. B, Phosphorylation of Pyk2 (P-Pyk2) in KG1a cells and in primary CD34⁺ HPCs. The conditions were similar to those in the experiments described in A. C, Phosphorylation of AKT (P-AKT) in KG1a cells and in primary CD34⁺ HPCs. A positive control for phosphorylated AKT was added (Pos.). D, Effect of tyrosine kinase inhibitors on LTD₄-induced adherence of HPCs. CD34⁺ cells were incubated with LTD₄ (1 μM) and genistein (10 μM) or LTD₄ and PD98059 (10 μM) for 15 min before exposure to activated HUVECs. ∗, p < 0.05 vs control. rel., Relative adhesion (compared with control).

Cysteinyl leukotrienes are produced by bone marrow stroma and endothelium

By RT-PCR, stromal and immortalized BMEC1 cells expressed 5-LO, which is the key enzyme of the leukotriene biosynthesis pathway (4, 37) (Fig. 5⇓A). As measured by HPLC and immunoassay (Fig. 5⇓B), biologically significant amounts of LTC₄, LTD₄, and LTE₄ were found in the supernatants of confluent layers of these cells (Fig. 5⇓C). Summarizing all three LTs (LTC₄, LTD₄, and LTE₄), concentrations were as high as 21.35 ± 5.25 nM (endothelium) and 18.27 ± 4.13 nM (stroma) in the culture medium, respectively. As mentioned above, significant effects on HPC proliferation in vitro were observed already at a final LTD₄ concentration of 1 nM, demonstrating the physiological relevance of the data observed. Interestingly, in a conditioned medium of stromal cells not deprived of hematopoietic cells by treatment with MPA, almost no cysteinyl leukotrienes could be detected (Fig. 5⇓C). To confirm that the absence of hematopoiesis is required for LT synthesis in these cells, pure stromal layers were cocultured with bone marrow CD34⁺ cells (derived from the same donor source). Indeed, LT levels were reduced by 89.7–91.3% (Fig. 5⇓D). This suggests LT metabolism and/or modulation of LT production by bone marrow hematopoietic cells and supports previous findings that the quantity and profile of LTs produced both in vitro and in vivo critically depend on cellular interactions and the composition of mixed cell populations (4).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Production of cysteinyl leukotrienes by BMEC-1 and stromal cells (Stroma). A, Expression of 5-LO mRNA in BMEC-1 and primary stromal cells. RT-PCR was used for this analysis (positive control, 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. Confluent 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 quantified by EIA. C, Production of cysteinyl leukotrienes by human BMEC-1 and primary human stromal cells. LTC₄, LTD₄, and LTE₄ were quantified as described in B. D, Cysteinyl leukotriene production of stromal cells without and with establishment of hematopoiesis in vitro. Bone marrow (BM)-derived CD34⁺ cells originating from the same donor were added to a MPA-treated stromal layer and cultured for 4 wk. The amount of cysteinyl leukotrienes was analyzed as described in B and compared with the stromal cells without hematopoiesis.

Effect of a CysLT₁ antagonist on human HPC mobilization

We investigated whether montelukast-sodium (Singulair), a selective CysLT₁ antagonist used in asthma therapy, could mobilize hematopoietic progenitor cells from bone marrow to peripheral blood in human subjects with effects similar to those described for the CXCR4 antagonist AMD3100 (21). In these previous studies, an 8-fold increase of circulating CD34⁺ cells was observed 4 h after a single dose of the CXCR4 antagonist, suggesting that HPC mobilization is due to prevention and disruption of homing. Table II⇓ demonstrates that administration of Singulair (10 mg) had no effect on the white blood count and on the number of CD34⁺ HPCs in the peripheral blood at any of the time points studied. In asthmatic patients, a single dose of 10 mg of montelukast causes physiological effects (bronchodilation) within several hours after oral administration, with the mean peak plasma concentration of the drug being achieved 3 h after oral intake (38, 39). Based on this, we conclude that inhibition of CysLT₁ does not result in HPC mobilization during steady-state hematopoiesis in vivo.