Published online before print April 23, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Division of Hematology, Department of Medicine, University Hospital Groningen, The Netherlands;
Respiratory Medicine Division, Department of Medicine, University of Cambridge School of Clinical Medicine, Addenbrooke’s and Papworth Hospitals, United Kingdom; and
Sanquin Blood Bank North East Netherlands, Groningen, The Netherlands
1Correspondence: University Hospital Groningen, Department of Hematology Research, Hanzeplein 1, 9713 GZ Groningen, The Netherlands. E-mail: E.Vellenga{at}int.azg.nl
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: MDS • signal transduction • cellular activation • phosphatidylinositol 3-kinase
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The Myelodysplastic syndromes (MDS) are distinguished by a differentiation defect in the multipotent stem-cell compartment, resulting in disturbed proliferation and differentiation of the erythroid, myeloid, or megakaryocytic lineages [6 ]. The ensuing granulocytopenia and aberrant functioning of neutrophils make bacterial infections the most common cause of death in MDS patients [7 ]. We have recently shown that GM-CSF priming of the fMLP-triggered respiratory burst is severely impaired in neutrophils from MDS patients [8 ].
The molecular mechanisms underlying priming of ROS production are still unclear. Generation of ROS is mediated by the granulocyte nicotinamide adenine dinucleotide phosphate (NADPH) oxidase. This enzyme consists of a membrane-bound component, flavocytochrome b558,and the cytosolic factors p47phox, p40phox, p67phox, and Rac [9 ]. Several reports have suggested that priming of the neutrophil respiratory burst results from enhanced translocation of flavocytochrome b558, which is stored in the membranes of specific granules and secundary vesicles, to the plasma membrane [10 , 11 ]. However, we have previously shown that whereas GM-CSF priming of ROS production is decreased in neutrophils from MDS patients, GM-CSF-induced increases in flavocytochrome b558 expression at the plasma membrane are normal in these cells [12 ]. This suggests that an additional mechanism involved in GM-CSF priming is affected in MDS patients.
Several signal-transduction molecules have been implicated in the priming of NADPH oxidase activity [13
]. For example, the involvement of the lipid kinase phosphatidylinositol 3-kinase (PI-3K) in GM-CSF priming of fMLP-stimulated ROS production has been established by using specific inhibitors [8
, 14
]. A number of class I isoforms of PI-3K have been identified; class IA catalytic subunits p110
, -ß, and -
associate with the regulatory subunit p85, whereas the class IB catalytic subunit p110
associates with p101 [15
]. Upon stimulation of neutrophils, these PI-3Ks produce the phosphoinositide PI(3,4,5)trisphosphate (PIP3), which acts as an important second messenger for superoxide anion generation [16
]. Its immediate metabolic products, PI(3,4)P2 and PI(3)P1,have also been implicated in regulating ROS production through an interaction with p47phox and p40phox [17
, 18
].
Evidence of priming at a molecular level comes from studies showing increased fMLP-stimulated PI-3K activation in GM-CSF-primed neutrophils compared with unprimed cells [19 , 20 ]. Furthermore, phosphorylation of the PI-3K downstream target protein kinase B (PKB/Akt) can be enhanced by GM-CSF pretreatment of neutrophils. Previously, we have shown that PKB/Akt activation in neutrophils from MDS patients is defective, suggesting disturbed PI-3K signaling in these cells [8 ].
Rac, a member of the Rho family of small GTPases, plays an important role in ROS production by regulating the assembly of the NADPH oxidase complex [21 ]. In resting cells, Rac is bound to guanosine 5'-diphosphate (GDP) and hence, is inactive. Guanine nucleotide exchange factors (GEFs), stimulated upon neutrophil stimulation, activate Rac by allowing the exchange of GDP for guanosine 5'-triphosphate (reviewed in ref. [22 ]). Inactivation occurs as a result of intrinsic GTPase activity, which can be enhanced by GTPase-activating proteins (GAPs). Different GEFs for Rac have been described to possess a PI-3K-binding domain, resulting in PI-3K-dependent activation of Rac (reviewed by Welch et al. [23 ]). In return, Rac can activate PI-3K itself, although the mechanism for this is still unclear. Another GTPase capable of triggering PI-3K activity is Ras, a member of the well-studied Ras–Raf–mitogen-activated protein kinase kinase–extracellular-regulated kinase (ERK) signaling pathway [24 ]. Ras can also activate the GTPase Ral through PI-3K-dependent or independent pathways [25 ]. However, Ras-independent mechanisms of Ral activation have also been identified [26 ]. The cellular function of Ral is still unclear, although recent evidence suggests that a downstream target of Ral, Ral-binding protein 1 (RLIP1), may play a regulatory role in Rac activation [27 ].
In this study, we addressed whether the disturbed GM-CSF priming of fMLP-induced ROS production in MDS neutrophils might reflect a defect in the activation of the GTPases Rac, Ras, and Ral and the production of PIP3. Our results demonstrate that in contrast to healthy donor neutrophils, Rac activation and PIP3 mass accumulation in response to fMLP stimulation could not be enhanced by GM-CSF in neutrophils from MDS patients. These impaired responses did not appear to be a result of a general GM-CSF or fMLP receptor-signaling defect, as the fMLP-triggered activation of Ral and GM-CSF priming of Ras activity were normal in MDS neutrophils. Taken together, our data suggest that in neutrophils from MDS patients, an impaired priming of the PI-3K–Rac signaling pathway results in a decreased GM-CSF priming of ROS production and that this defect is located at the level of PI-3K.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Patients
Eleven patients with MDS were studied. MDS was classified as refractory anemia (RA) or RA with ringed sideroblasts, according to French-American-British Cooperative Group criteria [28
]. Informed consent was obtained from all patients. The Human Subject Review Board of the University Hospital Groningen (The Netherlands) approved the protocol.
Isolation of neutrophils
Peripheral blood, anticoagulated with 0.32% sodium citrate, was obtained from healthy volunteers and MDS patients. Neutrophils were isolated as described [8
, 29
]. In short, mononuclear cells were removed by centrifugation over Fycoll-Paque (Amersham, Upsala, Sweden), and erythrocytes were lysed with ice-cold NH4Cl solution. Neutrophils were allowed to recover for 30 min at 37°C in RPMI 1640 (BioWhittaker, Verviers, Belgium) supplemented with 0.5% human serum albumin (HSA; CLB, Amsterdam, The Netherlands). Before stimulation, cells were resuspended in incubation buffer (20 mM HEPES, 132 mM NaCl, 6 mM KCl, 1 mM MgSO4, 1.2 mM KH2PO4, 5 mM glucose, 1 mM CaCl2, 0.5% HSA). In all cases, the cell population isolated consisted of >95% neutrophils as determined by May-Grünwald Giemsa staining.
Rac, Ras, and Ral activation assays
Activated Rac was precipitated using bacterial lysate containing GST–PAK–CRIB, activated Ras was pulled down using bacterial lysate containing GST–Raf–RBD, and activated Ral was precipitated using bacterial lysate containing GST–RLIP–RBD as described previously [8
, 26
]. Briefly, neutrophils (1x106 cells/ml for Rac activation assay and 10x106 cells/ml for Ral and Ras activation assays) were stimulated with 1 µM fMLP for the indicated time and lysed for 10 min in lysis buffer [50 mM Tris, pH 7.4, 10% glycerol, 200 mM NaCl, 1% Nonidet P-40, 2 mM MgCl2, 2 mM sodiumorthovanadate, and protease inhibitors; one tablet Complete (Roche, Mannheim, Germany) per 50 ml buffer]. Lysates were cleared by centrifugation, and GST–PAK–CRIB, GST–Raf–RBD, or GST–RLIP–RBD protein, precoupled to glutathione-sepharose beads (Pharmacia, Uppsala, Sweden), was added for 30–45 min at 4°C. Beads were washed three times with 1x lysis buffer and boiled in Laemmli sample buffer. The bound proteins were separated by 15% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and were transferred to polyvinylidene difluoride membrane (Immobilon-P; Millipore, Bedford, MA). Activated Rac, Ras, or Ral was detected by probing the membrane with anti-Rac, anti-Ras, or anti-Ral antibodies, respectively, and rabbit anti-mouse peroxidase-conjugated antibodies (Dako, Denmark) followed by enhanced chemiluminescence (ECL). Equal amounts of glutathione-sepharose beads were loaded in all samples as determined by Ponceau S staining of the membranes. Quantification of the amount of precipitated, active GTPase was performed by densitometry of the films, using ImageMaster1D Elite (Pharmacia, Woerden, The Netherlands). Results are presented as normalized densitometry values [arbitrary units (AU)].
PIP3 mass assay
r-Ins(1,3,4,5)P4-binding protein (GAPIP4BP) was purified as decribed previously [30
], and sample purity was analyzed by SDS-PAGE. Total and nonspecific binding of [3H]Ins(1,3,4,5)P4 in each preparation of GAPIP4BP was assessed by serial dilution and inclusion of 0.1 nM InsP6, respectively. Thereafter the GAPIP4BP was used at a concentration to achieve a maximum binding of 20% of the total [3H]Ins(1,3,4,5)P4 input. Purified neutrophils (8x106) were stimulated with 1 µM fMLP or vehicle for the indicated times, with or without prior priming with GM-CSF (5 ng/ml) for 15 min. The reactions were stopped by the addition of methanol/chloroform (2:1, v/v), and lipid extractions were performed as described previously [20
, 31
]. After drying, the samples were processed as previously detailed and stored at –20°C until analyzed. Immediately before the assay, samples were resuspended in dilute acetic acid to a final pH of 5. The Ins(1,3,4,5)P4 radioreceptor assay was performed as described using 0.005 µCi [3H]Ins(1,3,4,5)P4 per sample [16
].
Statistical analysis
Differences between unstimulated and stimulated samples from the same donor and differences between samples treated with or without signal-transduction inhibitors were calculated using the Wilcoxon-nonparametric test for paired samples. Differences between healthy volunteers and MDS patients were calculated using the Mann-Whitney nonparamentric test for unpaired samples. Data were expressed as mean ± SEM. P values <0.05 were considered statistically significant.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (36K): [in a new window] |
Figure 1. GM-CSF priming of Rac activation is disturbed in neutrophils from MDS patients. Unprimed neutrophils and neutrophils that were primed with 5 ng/ml GM-CSF for 15 min were stimulated with 1 µM fMLP for the indicated times (s). Stimulation was stopped by adding 3x lysis buffer. Activated Rac was precipitated using GST–PAK–CRIB protein precoupled to glutathione-sepharose beads. Proteins were separated by 15% SDS-PAGE, and membranes were probed with Rac antibodies, followed by ECL. Rac activation was investigated in neutrophils from seven healthy controls and seven MDS patients. Two representative blots from healthy subjects (a) and three representative blots from MDS patients (c) are shown. Equal amounts of glutathione-sepharose beads were loaded in all samples, as determined by Ponceau S staining of the membranes (not shown). Protein levels of activated Rac were quantified by densitometry of the films. The means of the normalized values were calculated for the healthy donors (b) and the MDS patients (d). Significant differences between unprimed and GM-CSF-primed groups were calculated using the Wilcoxon-nonparametric test for paired samples and are indicated (*, P<0.05).
|
These results imply that in agreement with the decreased GM-CSF priming of fMLP-induced ROS production, priming of Rac activity by GM-CSF is also disturbed in neutrophils from MDS patients.
Rac activation is sensitive to the PI-3K inhibitor LY294002
To elucidate the sequence of activation of Rac and PI-3K upon fMLP stimulation, healthy neutrophils were preincubated for 15 min with the specific PI-3K inhibitor LY294002 before priming with GM-CSF for 15 min and/or stimulation with fMLP for 30 s. Figure 2a
demonstrates that LY294002 treatment resulted in a significant attenuation of Rac activation in response to fMLP stimulation in unprimed or GM-CSF-primed cells. Densitometry of the blots revealed that inhibition of PI-3K activity reduced the fMLP-triggered Rac activation from 14 ± 2 to 8 ± 2 (AU, n=5, P=0.04) in unprimed cells. In GM-CSF-primed neutrophils, the fMLP-triggered Rac activation was reduced from 21 ± 1 to 7 ± 1 (AU, n=5, P=0.04) as a result of LY294002 treatment (Fig. 2b)
. These results demonstrate that activation of the small GTPase Rac by fMLP could be a downstream event from PI-3K activation.
 View larger version (20K): [in a new window] |
Figure 2. Effect of LY294002 on Rac activation in unprimed and GM-CSF-primed, healthy neutrophils. Unprimed neutrophils and neutrophils that were primed with 5 ng/ml GM-CSF for 15 min were stimulated with 1 µM fMLP for 30 s. Where indicated, cells were pretreated with 10 µM LY294002 for 15 min. Activated Rac was precipitated as described in Figure 1
. Three representative blots are shown (a). Protein levels of activated Rac were quantified by densitometry of the films, and the means of the normalized values for five individual experiments are shown (b). Differences between LY294002 treated and untreated groups were calculated using the Wilcoxon-nonparametric test for paired samples, and significant reductions in Rac activation are indicated (*, P<0.05).
|
 View larger version (20K): [in a new window] |
Figure 3. Impaired GM-CSF priming of PIP3 mass accumulation in neutrophils from MDS patients. (a) Neutrophils were stimulated with 1 µM fMLP for the indicated amount of time, with or without prior treatment with 5 ng/ml GM-CSF for 15 min. Reactions were stopped by the addition of MeOH/CHCl3 (2:1 v/v), lipids were extracted, and the Ins(1,3,4,5)P4 isotope dilution assay was performed as described in Materials and Methods. Results represent the mean PIP3 mass accumulation in samples from five healthy volunteers and five MDS patients. Statistical differences between unstimulated and fMLP-stimulated groups were calculated using the Wilcoxon-nonparametric test for paired samples, and significant differences are indicated (*, P<0.05). Significant differences between unprimed and GM-CSF-primed groups are indicated (**, P<0.05). (b) Neutrophils from healthy donors were primed with 5 ng/ml GM-CSF for 15 min before stimulation with 1 µM fMLP for 45 s. Where indicated, cells were preincubated with LY294002 (10 µM for 30 min). PIP3 measurement was performed in three independent experiments as described. PIP3 mass accumulation of a representative experiment is shown.
|
In conclusion, these results indicate that as for ROS production and Rac activation, priming of PI-3K activity by GM-CSF is severely disturbed in neutrophils from MDS patients.
Priming of Ras activation in neutrophils from healthy donors and MDS patients
Activation of the Ras GTPase has been placed upstream of PI-3K activity. To establish whether the observed disturbance in PIP3 accumulation in MDS neutrophils could be attributed to disturbed Ras signaling, we investigated the activation of this small GTPase in neutrophils from healthy donors compared with MDS patients. Stimulation of healthy neutrophils with 1 µM fMLP resulted in a transient activation of Ras, with a maximum activation after 30 s (Fig. 4a
). After 10 min of fMLP stimulation, levels of active Ras were reduced to baseline values (not shown). Preincubation of neutrophils with 5 ng/ml GM-CSF before fMLP stimulation resulted in a significantly increased and prolonged activation of Ras when compared with fMLP stimulation without GM-CSF pretreatment. Densitometry of the blots showed that this priming effect was significantly higher than the accumulative activation of Ras by fMLP and GM-CSF alone (Fig. 4b)
. These data are consistent with the increased and prolonged fMLP-induced phosphorylation of the Ras downstream target ERK1/2, which we previously found in healthy GM-CSF-primed neutrophils [8
].
 View larger version (36K): [in a new window] |
Figure 4. fMLP-induced Ras activation can be primed by GM-CSF in neutrophils from healthy donors and MDS patients. Unprimed neutrophils and neutrophils that were primed with 5 ng/ml GM-CSF for 15 min were stimulated with 1 µM fMLP for the indicated time (s). Activated Ras was precipitated using GST–Raf–RBD protein precoupled to glutathione-sepharose beads as described. Ras activation was investigated in neutrophils from five healthy controls and seven MDS patients. Two representative blots from healthy subjects (a) and three representative blots from MDS patients (c) are shown. Protein levels of activated Ras were quantified by densitometry of the films. The means of the normalized values were calculated for the healthy donors (b) and the MDS patients (d). Significant differences between unprimed and GM-CSF-primed groups were calculated using the Wilcoxon-nonparametric test for paired samples and are indicated (*, P<0.05).
|
We subsequently set out to investigate the position of Ras relative to PI-3K in signal transduction (Fig. 5a ). Incubation of neutrophils with 10 µM of the PI-3K inhibitor LY294002 before stimulation of the cells with fMLP did not reduce the amount of activated Ras present in the cells [14±4 vs. 12±4 (AU), P=0.7]. Likewise, priming of fMLP-stimulated Ras activation by GM-CSF pretreatment of neutrophils was also not affected by the PI-3K inhibitor [31±3 vs. 39±3 (AU), P=0.08; Fig. 5b ]. These results indicate that fMLP-induced Ras activation can be primed by GM-CSF in a PI-3K-independent manner and that this priming is normal in MDS neutrophils.
 View larger version (35K): [in a new window] |
Figure 5. Effect of LY294002 on Ras activation in unprimed and GM-CSF-primed, healthy neutrophils. Unprimed neutrophils and neutrophils that were primed with 5 ng/ml GM-CSF for 15 min were stimulated with 1 µM fMLP for 2 min. Where indicated, cells were pretreated with 10 µM LY294002 for 30 min. Activated Ras was precipitated as described previously. Two representative blots are shown (a). Protein levels of activated Ras were quantified by densitometry of the films, and the means of the normalized values for five individual experiments are shown (b).
|
 View larger version (27K): [in a new window] |
Figure 6. fMLP-induced Ral activation cannot be primed by GM-CSF and is not inhibited by LY294002 in neutrophils from healthy donors. (a) Unprimed, healthy neutrophils and neutrophils that were primed with 5 ng/ml GM-CSF for 15 min were stimulated with 1 µM fMLP for the indicated time (s). Activated Ral was precipitated using GST–RLIP–RBD protein precoupled to glutathione-sepharose beads as described. Two representative blots are shown. (b) Protein levels of activated Ral were quantified by densitometry of the films. The means of the normalized values from three individual experiments were calculated. (c) Neutrophils were stimulated with 1 µM fMLP for 30 s, with or without prior incubation with 10 µM LY294002 for 30 min. Activated Ral was precipitated as described. Blots from two individual experiment are shown. (d) Active Ral precipitated in c was quantified by densitometry. The means of the normalized values from three individual experiments were calculated.
|
We next compared the fMLP-stimulated activation of Ral in neutrophils from healthy donors (n=7) and MDS patients (n=3). Figure 7a shows that as in healthy donors, Ral activation in response to fMLP was rapid and transient in neutrophils from MDS patients. Densitometry of the blots indicated that the degree of Ral activation was identical in MDS patients and healthy volunteers (Fig. 7b) .
 View larger version (34K): [in a new window] |
Figure 7. Ral activation by fMLP is normal in neutrophils from MDS patients. Neutrophils were stimulated with 1 µM fMLP for the indicated time (s), and activated Ral was precipitated using GST–RLIP–RBD protein precoupled to glutathione-sepharose beads as described. Two representative blots from healthy subjects (a) and three representative blots from MDS patients (b) are shown. Protein levels of activated Ral were quantified by densitometry of the films. The means of the normalized values were calculated for seven healthy donors and three MDS patients (c).
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Several studies have shown that the GTPase Ras can activate PI-3K [24 ]. However, we now show that although priming of PIP3 accumulation, a result of PI-3K activity, was absent in neutrophils from MDS patients, Ras actvation was normal. This suggests that another signal transducer capable of activating PI-3K might be affected in MDS neutrophils. Alternatively, the defect is located at the level of PI-3K itself, a suggestion corroborated by our finding that protein levels of PI-3K subunits are disturbed in MDS neutrophils [8 ]. Other studies have revealed that the level of PIP3 within cells is carefully balanced by different lipid kinases and phosphatases [36 ]. It can therefore not be formally excluded that one or more of these enzymes are responsible for the disturbed PIP3 accumulation in neutrophils from MDS patients.
Neutrophils from PI-3K–/– mice, which do not produce PIP3 and ROS in response to fMLP, show normal activation of the small GTPase Rac after fMLP stimulation [37
]. This would indicate a PI-3K-independent Rac activation, which is not involved in the neutrophil respiratory burst. In contrast, studies describing a patient with an inhibitory Rac2 mutation have clearly established the importance of this GTPase in ROS production [38
]. Furthermore, other studies indicate that Rac activation in response to fMLP is PI-3K-dependent in unprimed and GM-CSF-primed neutrophils [39
]. In the present study, attenuation of Rac activation, although significant, was not complete. This suggests that Rac activation might occur through multiple signal-transduction routes, including a PI-3K-independent pathway.
Early reports suggested that the diversity in PI-3Ks (PI-3K, -ß, -, and -) contributes to the distinct PI-3K-dependent, cellular functions activated upon stimulation of cells with different stimuli (reviewed in ref. [15
]). For example, G protein-coupled receptors (GPCRs), e.g., the fMLP receptor, were described to exclusively activate p110, whereas p110, -ß, and - activation was attributed to receptor tyrosine kinases and cytokine receptors such as the GM-CSF receptor [40
41
42
]. Recent studies, however, have shown that GPCRs can activate class IA PI-3Ks, in addition to PI-3K [43
44
45
]. Another recent study demonstrated that in p85–/– mice, GM-CSF priming of fMLP-stimulated ROS production is severely decreased, indicating the involvement of class IA PI-3Ks in the GM-CSF-priming response [46
]. However, equally convincing studies have shown that p110 is the primary subunit responsible for the enhanced PIP3 production seen in primed neutrophils [20
]. It is conceivable that full activation of class IA and class IB PI-3Ks is neccessary in GM-CSF-primed neutrophils to get a full fMLP-induced response, as was recently described for stromal cell-derived factor-1-induced migration of leukemic T cells [47
]. To what extent activation of the different PI-3K classes is affected in MDS neutrophils remains to be elucidated.
In conclusion, we show a disturbed GM-CSF priming of fMLP-induced PIP3 accumulation and Rac activation in neutrophils from MDS patients. Activation of Ras could be primed normally by GM-CSF in MDS neutrophils, and fMLP-triggered activity of the Ras target Ral was likewise undisturbed in these cells. Taken together, our data suggest that in neutrophils from MDS patients, a defect in priming of the PI-3K–Rac signaling pathway results in a decreased GM-CSF priming of ROS production and that this defect is located at the level of PI-3K activity.
Received February 5, 2004; revised March 12, 2004; accepted March 13, 2004.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and GM-CSF J. Leukoc. Biol. 64,537-545[Abstract] by Ras Curr. Biol. 12,1068-1075[CrossRef][Medline] is signalled by enhanced phosphatidylinositol 3,4,5-triphosphate but not inositol 1,4,5-triphosphate accumulation FEBS Lett. 439,147-151[CrossRef][Medline] in chemoattractant-mediated signal transduction Science 287,1046-1049 J. Biol. Chem. 275,23636-23641 is a mediator of Gß-dependent Jun kinase activation J. Biol. Chem. 273,2505-2508 are substrates, in vitro, for a constitutively associated protein tyrosine kinase in platelets Blood 91,930-939 subunits and class Ia phosphoinositide 3-kinases J. Biol. Chem. 275,26225-26232This article has been cited by other articles:
 |
|
 |
|
  G. M. Fuhler, N. R. Blom, P. J. Coffer, A. L. Drayer, and E. Vellenga The reduced GM-CSF priming of ROS production in granulocytes from patients with myelodysplasia is associated with an impaired lipid raft formation J. Leukoc. Biol., February 1, 2007; 81(2): 449 - 457. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. M. Fuhler, G. J. Knol, A. L. Drayer, and E. Vellenga Impaired interleukin-8- and GRO{alpha}-induced phosphorylation of extracellular signal-regulated kinase result in decreased migration of neutrophils from patients with myelodysplasia J. Leukoc. Biol., February 1, 2005; 77(2): 257 - 266. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
