Published online before print August 7, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Department of Clinical Sciences and
Center for Comparative Medicine and Translational Research, College of Veterinary Medicine, North Carolina State University, Raleigh, North Carolina, USA
1 Correspondence: CVM Research Building, Room 244, North Carolina State University, 4700 Hillsborough St., Raleigh, NC 27606, USA. E-mail: sam_jones{at}ncsu.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
did not affect Ser 157 phosphorylation. The kinetics of fMLF-induced VASP Ser 157 phosphorylation levels closely matched the kinetics of the fold-change in F-actin levels in fMLF-stimulated neutrophils. fMLF-induced Ser 157 phosphorylation was abolished by pretreatment with the PKA inhibitor H89 and the adenylyl cyclase inhibitor SQ22536. In contrast, fMLF-induced Ser 157 phosphorylation was unaffected by the PKC inhibitors calphostin and staurosporine, the PKG inhibitors Rp-8-pCPT-cGMP and KT5823, and the calmodulin-dependent protein kinase II inhibitor KN-62. Inhibition of adhesion with EDTA or the anti-β2-integrin antibody IB4 did not alter fMLF-induced VASP phosphorylation or dephosphorylation. These data show that chemoattractant stimulation of human neutrophils induces a rapid and transient PKA-dependent VASP Ser 157 phosphorylation. Adhesion does not appear to be an important regulator of the state of VASP Ser 157 phosphorylation in chemoattractant-stimulated neutrophils.
Key Words: PKA • chemotaxis • fMLF
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The PKA substrate vasodilator-stimulated phosphoprotein (VASP) is a member of the enabled (Ena)/VASP family of F-actin-associated proteins, a key effector of PKA in cytoskeletal organization, integrin adhesion, and migration [6 , 7 ]. In fibroblasts, VASP localizes to the actin cytoskeleton at adhesion sites by binding the focal adhesion proteins zyxin and vinculin [8 9 10 11 ] and at the leading edge of lamellipodium by binding the scaffolding protein lamellipodin [12 ]. VASP binds to its partners at these sites via an N-terminal pleckstrin homology-like Ena/VASP homology 1 (EVH1) domain, which binds to FPPPP motifs in zyxin, vinculin, and lamellipodin [6 , 12 13 14 ]. The C-terminal EVH2 domain contains an F-actin binding site and a coiled-coil motif important for VASP oligomerization [13 , 15 ]. VASP also contains a central proline-rich region with binding sites for the actin-monomer binding protein profilin [16 ] and Src homology 3 (SH3) and WW domain-containing proteins [6 , 13 ], suggesting that VASP is a key element of signaling module formation at sites of cytoskeletal organization. VASP promotes actin polymerization by nucleating actin filaments and by delivering profilin-G-actin complexes to the barbed end. VASP is phosphorylated by PKA, PKG, and PKC on serine 157 (Ser 157) and by PKA and PKG on Ser 239 and threonine 278 [17 , 18 ]. Ser 157 is the preferred site for PKA-induced phosphorylation [18 , 19 ]. Phosphorylation by PKA reduces the capacity of VASP to bind, bundle, and nucleate actin filaments [6 , 20 , 21 ], and thus is a crucial, post-translational modification.
VASP is essential for normal fibroblast lamellipodial dynamics and cellular migration [22 23 24 ]. The ability of the Ena/VASP family member Mena to regulate migration requires Ser 236, which corresponds to VASP Ser 157 [25 ]. VASP Ser 157 is also required for the ability of cAMP to inhibit platelet integrin-mediated adhesion [26 27 28 ]. Together, these data demonstrate that VASP Ser 157 and its phosphorylation are important for regulating cellular adhesion and migration. However, little is known about VASP and its phosphorylation in neutrophil migration. We hypothesized that PKA mediates its effects on neutrophil migration by regulating VASP Ser 157 phosphorylation. We began testing this hypothesis by characterizing VASP Ser 157 phosphorylation in neutrophils stimulated by chemoattractants.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
was from R&D Systems, Inc. (Minneapolis, MN, USA). Leukotriene B4 (LTB4) and 1-O-hexadecyl-(7,7,8,8-d4)-2-O-acetyl-sn-glycerol-3-phosphorylcholine (PAF) were from Cayman Chemical (Ann Arbor, MI, USA). KT5720, H-89, staurosporine, and 3-isobutyl-1-methylxanthine (IBMX) were from Alexis (San Diego, CA, USA). Rp-8-pCPT-cGMP was from Biolog Life Science Institute (Bremen, Germany). EDTA was from Fisher Scientific (Atlanta, GA, USA). KN-62 was from Biomol International (Plymouth Meeting, PA, USA). KT5823, SQ22536 [9-(tetrahydro-2'-furyl)adenine], Ro-20-1724 [4-(3-butoxy-4-methoxybenzyl)-2-imidazolidinone], and 4-{[3',4'-(methylenedioxy)benzyl]amino}-6-methoxyquinazoline (4-MB) were from Calbiochem/EMD Chemicals (San Diego, CA, USA). FCS was obtained from Hyclone (Logan, UT, USA). Calcein was obtained from Molecular Probes (Eugene, OR, USA). Alexa Fluor 546 phalloidin was from Invitrogen (Carlsbad, CA, USA). HRP-conjugated goat anti-mouse IgG1(
) secondary antibody was from Caltag Laboratories (Burlingame, CA, USA). HRP-conjugated rabbit anti-mouse IgG secondary antibody was obtained from Pierce (Rockford, IL, USA). Mouse IgG1 anti-VASP mAb, diisopropylfluorophosphate (DFP), and rhIL-8 were from BD Biosciences (San Diego, CA, USA). mAb IB4 (anti-β2, CD18) [29
] and W6/32 (anti-class I HLA) [30
] were prepared as described. Phospho-VASP (Ser 157) mAb and anti-rabbit, HRP-conjugated secondary antibody were obtained from Cell Signaling (Beverly, MA, USA). Anti-L-plastin (Lpl) mAb Lpl4A.1 (IgG1) was prepared as described previously [31
] and used as purified IgG. Antiphospho-Lpl (p-Lpl)I4E11 was a kind gift of Dr. Eric Brown (University of California, San Francisco, CA, USA) and used as a tissue-culture supernatant.
Preparation of neutrophils
Human leukocyte-rich plasma was separated from whole blood using dextran sedimentation. Neutrophils were isolated using a Ficoll gradient. In brief, 
6 ml plasma was layered on 5 ml sterile, endotoxin-free Ficoll-Paque solution and spun at 1800 rpm for 20 min. Neutrophils were used if they demonstrated greater than 98% viability, as determined by exclusion of trypan blue dye incorporation. RBCs were lysed by hypotonic lysis, and remaining neutrophils were washed once with HBSS. Cells were resuspended in HBSS with 20 mM HEPES, 8.9 mM sodium bicarbonate, 1 mM Ca2+, and 1 mM Mg2+ prior to assays (HBSS++). Cells pretreated with EDTA were suspended in HBSS, rather than HBSS++, to prevent Ca2+- and Mg2+-dependent integrin ligand binding.
Western blotting
For preparation of lysates, purified neutrophils (25x106/mL) were treated with inhibitors in microcentrifuge tubes and incubated at 37°C. Cells (2.5x106) were then added to 5% FCS-coated wells of a 24-well tissue-culture plate (Costar, Cambridge, MA, USA) and allowed to adhere for 5 min at 37°C. Cells were then stimulated with 100 nM fMLF, 100 ng/mL TNF-, 10 ng/mL PMA, or appropriate vehicle control (VC) for the indicated time periods during incubation at 37°C. After 20 total min, 2x Triton lysis buffer (2% Triton-X, 50 mM Na fluoride, 2.5 mM Na pyrophosphate, 100 mM Tris, 5 mM DFP, 100 µg/mL pepstatin, 1 mM iodoacetamide, 1 mM PMSF, and 10 µg/mL aprotinin/leupeptin) was added to wells, and plates were incubated on ice with agitation for 20 min. After lysis, cell solutions were transferred to microcentrifuge tubes and spun at 14,000 rpm for 8 min. Supernatants were collected, and protein concentrations were measured using bicinchoninic acid protein assay reagent (Pierce). Samples were mixed with 2x sample buffer (125 mM Tris, pH 6.7, 5% SDS w/v, 20% glycerol v/v, 0.01% bromophenol blue w/v, 0.01% 2-ME v/v in diH2O) and boiled for 5 min. Equal protein concentrations were analyzed by 10% SDS-PAGE. Resolved samples were transferred to an Immobilon-P polyvinylidene fluoride transfer membrane (Millipore, Billerica, MA, USA) and blocked for 1 h with Superblock blocking buffer (Pierce) before overnight incubation with the indicated primary antibody at 4°C. Membranes were washed for 30 min before 1 h incubation with appropriate HRP-conjugated secondary antibody. Membranes were again washed before development with ECL (WB substrate, Pierce) and radiography (Biomax scientific imaging film, Kodak). Films were scanned, and the density of the bands was measured with densitometric software (Scanalytics, Fairfax, VA, USA).
Chemotaxis assay
Human neutrophils were isolated and labeled with the fluorescent dye calcein for 30 min at room temperature. Cells were then washed and resuspended in a chemotaxis buffer containing HBSS++ with 2% FCS. Cells were treated with the indicated inhibitors for 30 min at 37°C prior to the placement of 1 x 104 cells on a 2-µm pore-size membrane on a ChemoTx® plate (Neuro Probe, Inc., Gaithersburg, MD, USA). Lower wells of the plate were filled with chemotaxis buffer, with or without 100 nM fMLF. Standard wells contained 1 x 104-labeled cells, which were allowed to migrate for 1 h at 37°C. After incubation, cells on the top of the filter were washed away with PBS. EDTA (0.5 mM) was added to the top of the filter for 5 min to detach adherent cells. The plate was then centrifuged at 1000 rpm for 1 min. The filter was removed, and the fluorescence was measured in the lower wells (485 nm excitation, 530 nm emission wavelengths) using an fMax fluorescence plate reader (Molecular Devices, Sunnyvale, CA, USA). Percent migration was determined by dividing the fluorescence of each well by the fluorescence of the standard wells containing 1 x 104-labeled cells.
Actin polymerization assay
Human neutrophils were isolated as described previously, washed once in buffer A (140 mM NaCl, 1 mM KH2PO4, 5 mM Na2PO4, 1.5 mM CaCl2, 0.3 mM MgSO4, 1 mM MgCl2, 10 mM HEPES, pH 7.4) before final resuspension in buffer A at a concentration of 5 x 106/ml. Cell suspension (98 µl) was placed in sterile BD FalconTM, 5 ml polystyrene round-bottom tubes (BD Biosciences). Cells were stimulated with 100 nM fMLF for indicated lengths of time before fixation with 100 µl 3% formaldehyde and 0.1% BSA in PBS for 20 min. Cells were then washed once in buffer A before permeabilization with 0.1% Triton and 1 unit per reaction of Alexa Fluor 546 phalloidin in 100 µl buffer A for 30 min on ice. PBS (400 µl) was added to the sample before transfer to a fresh tube for FACS analysis.
Actin staining
Freshly isolated human neutrophils (0.5x106) in HBSS++ with 1% BSA were added to FCS-coated glass coverslips in a 24-well tissue-culture plate. Cells were allowed to settle onto the glass before 100 nM fMLF or VC was added for indicated periods of time of incubation at 37°C. The supernatant was removed, and cells were fixed with 0.5 ml 3% paraformaldehyde solution (25 mM PIPES, pH 7, 50 mM KCl, 3 mM MgCl, 3% paraformaldehyde w/v, 10 mM EGTA in sterile water) for 20 min. Coverslips were then washed with PBS three times before extraction with 0.5 ml cold Triton lysis buffer (10 mM PIPES, pH 6.8, 0.5% Triton X-100 v/v, 300 mM sucrose, 100 mM KCl, 3 mM MgCl2 in sterile water). Cells were washed again before staining with Alexa-phalloidin 546, diluted 1:20 in PBS for 30 min at room temperature. Fluorescence of Alexa-phalloidin-546-stained F-actin was viewed with a Nikon TE-200 inverted epifluorescence microscope using a digital camera (SPOT, Diagnostics Instruments Inc., Sterling Heights, MI, USA) and associated software for image capture.
Coating of plates with immune complexes (ICs)
Each well of a 24-well, sterile, tissue-culture plate was coated with 200 µl 100 µg/ml poly-l-lysine in PBS and incubated at room temperature for 45 min. Wells were then washed thee times with PBS before adding 200 µl fresh 1% gluteraldehyde in PBS at room temperature for 15 min. Wells were again washed three times with PBS before the addition of 200 µl 100 µg/ml BSA in PBS for 4 h at room temperature. Wells were dumped and then blocked with 200 µl 1% human albumin serum w/v and 0.1 M glycine, pH 6.8, in PBS overnight at 4°C. Wells were washed again three times with PBS before addition of 200 µl 1:50 -BSA antiserum for 2 h at room temperature. Wells were washed three final times with PBS before use in experiments.
Statistical analysis
Data are reported as mean ± SE. Data were analyzed by Student’s two-sample t-test assuming equal variance or paired two-sample for means. A P < 0.05 was considered statistically significant.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Approximately 40% of VASP was phosphorylated on Ser 157 in unstimulated neutrophils (Fig. 1
). Stimulation with fMLF induced rapid and transient Ser 157 phosphorylation. Maximal phosphorylation was observed by 30 s, with 60% of VASP phosphorylated (Fig. 1)
. VASP phosphorylation then decreased, returning to baseline levels by 10 min following stimulation. Stimulation of cells with the chemoattractants PAF, LTB4, or IL-8 resulted in a similar pattern of rapid phosphorylation, which peaked by 30 s and waned within 10 min (Fig. 2
). To determine whether the fMLF-induced phosphorylation pattern was specific for VASP, we evaluated the fMLF-induced phosphorylation response of another important leukocyte-specific cytoskeletal protein, the actin-bundling protein L-plastin (Lpl). Lpl phosphorylation increased after 1 min of stimulation with fMLF and sustained for at least 20 min (Fig. 1)
. The prolonged fMLF-induced phosphorylation of Lpl is in stark contrast to the rapid phosphorylation and dephosphorylation event of VASP. To determine whether the characteristic pattern of VASP phosphorylation was unique to chemoattractant stimulation, we also stimulated cells with two nonchemoattractant activators of neutrophil adhesion: the proinflammatory cytokine TNF- and the PKC activator PMA. Stimulation of cells with TNF- at concentrations sufficient to induce strong adhesion (data not shown) did not result in a change in the phosphorylation state of VASP Ser 157 as compared with VC at any time-point we examined (Fig. 3
). Alternatively, stimulation of cells with PMA (10 ng/ml) resulted in rapid and sustained VASP Ser 157 phosphorylation for the incubation period of 30 min (Fig. 3)
. As fMLF induces a dependable and rapid phosphorylation and dephosphorylation event of Ser 157 of VASP, we used it as a representative chemoattractant for the remainder of our studies.
 View larger version (24K): [in a new window] |
Figure 1. VASP Ser 157 and Lpl phosphorylation in response to fMLF. (A) Purified neutrophils were stimulated with 100 nM fMLF or VC (Me2SO) for the indicated time periods during a 20-min incubation at 37°C after adherence to FCS-coated wells as described in Materials and Methods. Equal protein amounts at each time-point were analyzed for VASP by Western blot analysis. Densitometry was performed, and the density of the upper 50-kD and lower 46-kD bands was measured. "% VASP Phosphorylated" was calculated by dividing the upper phosphorylated band by the total amount of VASP (upper 50 kD band+lower 46 kD band) and presented as the mean ± SE. Stimulation of cells resulted in rapid phosphorylation and dephosphorylation of Ser 157 of VASP. The percent of phosphorylated VASP was significantly greater than VC at the time-points from 0.5 to 10 min (*, P<0.05) after fMLF stimulation. Data are representative of at least four separate experiments (n=6 fMLF; n=4 VC) using neutrophils from different donors. (B) Purified neutrophils were stimulated as above. Equal protein amounts at each time-point were analyzed for Lpl and p-Lpl by Western blot. Stimulation with 100 nM fMLF did not alter the amount of total Lpl in the cells. fMLF stimulation resulted in increased levels of p-Lpl by 1 min, and this level of phosphorylation was sustained for at least 20 min.
|
 View larger version (39K): [in a new window] |
Figure 2. VASP Ser 157 phosphorylation in response to chemoattractants PAF, LTB4, and IL-8. Purified neutrophils were stimulated with (A) 100 nM PAF, (B) 100 nM LTB4, or (C) 100 nM IL-8 or VC (Me2SO) for the indicated time periods during a 20-min incubation at 37°C after adherence to FCS-coated wells as described in Materials and Methods. Equal protein amounts at each time-point were analyzed for VASP by Western blot analysis. Stimulation with PAF, LTB4, or IL-8 resulted in a similar pattern of rapid phosphorylation and dephosphorylation of Ser 157 of VASP as seen after fMLF stimulation. Data are representative of at least three separate experiments using neutrophils from different donors.
|
 View larger version (44K): [in a new window] |
Figure 3. Neutrophil activators TNF- and PMA do not induce the rapid and transient Ser 157 phosphorylation of VASP, which is seen with chemoattractants. (A) Purified neutrophils were stimulated with 100 ng/ml TNF-, 100 nM fMLF, or VC (Me2SO) for the indicated time periods during a 20-min incubation at 37°C after adherence to FCS-coated wells as described in Materials and Methods. Equal protein amounts at each time-point were analyzed for VASP by Western blot analysis. Stimulation of cells with TNF- resulted in no significant change in the phosphorylation state of VASP as compared with VC. (B) Purified neutrophils were stimulated with 10 ng/ml PMA or VC (Me2SO) for the indicated time periods during a 30-min incubation at 37°C after adherence to FCS-coated wells as described in Materials and Methods. Stimulation of cells with the PKC activator PMA resulted in phosphorylation of Ser 157 of VASP, which was sustained for at least 30 min. Data are representative of least three separate experiments using neutrophils from different donors.
|
 View larger version (28K): [in a new window] |
Figure 4. The rapid phosphorylation and dephosphorylation of Ser 157 of VASP after fMLF stimulation corresponds with polarization and F-actin levels in neutrophils. (A) Purified neutrophils were allowed to adhere to FCS-coated glass coverslips prior to 100 nM fMLF or VC (Me2SO) stimulation for the indicated periods of time before permeabilization, fixation, and staining of F-actin with phalloidin. At 30 s, fMLF-stimulated neutrophils began to polarize and develop an F-actin-rich leading-edge and a trailing-edge uropod. In contrast, VC cells remained spherical with diffuse distribution of F-actin. From 60 s to 20 min of fMLF stimulation, cells continued to polarize, as evidenced by membrane ruffles and further development of the leading edge in preparation for migration. (B) Purified neutrophils were stimulated with VC (Me2SO) or 100 nM fMLF for 0, 30 s, 60 s, 180 s, or 300 s before fixation, permeabilization, and staining with Alexa Fluor 546 phalloidin, as described in Materials and Methods, and FACS analysis. The reported mean fluorescence intensity for each sample was divided by a "time zero", unstimulated control to determine the fold change in F-actin levels for each treatment group. VC-treated cells had similar levels of F-actin at all time-points, except a significantly reduced amount (0.78) at 20 min as compared with time zero (*, P<0.05). In contrast, cells stimulated with fMLF had a 2.5-fold increase in F-actin levels within 30 s. By 60 s, F-actin levels decreased to twofold over time zero and continued to decline toward basal levels. Cells stimulated with fMLF for 30 s, 60 s, and 180 s had significantly increased levels of F-actin when compared with VC cells at the same time-points ( , P<0.05). Data are representative of three independent experiments.
|
Fig. 9
). As H89 also potently inhibits CamKII, which could account for the discrepancy between H89 and KT5720 to inhibit fMLF-induced VASP phosphorylation, we pretreated cells with the CamKII inhibitor KN-62 (25 µM) but did not see an affect on fMLF-induced VASP Ser 157 phosphorylation (Fig. 5)
. VASP Ser 157 phosphorylation was abolished by pretreatment with 3 mM SQ22536, a cell-permeable adenylyl cyclase inhibitor (Fig. 5)
, further supporting a role for PKA in the mechanism of VASP Ser 157 phosphorylation in response to fMLF stimulation. fMLF-induced VASP Ser 157 phosphorylation was not affected by the PKG inhibitors Rp-8-pCPT-cGMP (50 µM) or KT5823 (2.5 µM), nor the PKC inhibitors staurosporine (100 nM) or calphostin C (10 µM; Fig. 5
), ruling out a role for PKG or PKC in this mechanism.
 View larger version (18K): [in a new window] |
Figure 5. VASP fMLF-induced Ser 157 phosphorylation is PKA-dependent. (A) Purified neutrophils were pretreated with a range of concentrations of the PKA inhibitors H-89 or KT5720, the adenylyl cyclase inhibitor SQ22536, or VC (Me2SO) before stimulation with 100 nM fMLF for 60 s at 37°C after adherence to FCS-coated wells as described in Materials and Methods. Equal protein amounts were analyzed by Western blot analysis using an anti-whole, VASP-specific antibody. Densitomety was performed as described previously for the maximum concentrations of each inhibitor. KT5720 (25 µM) pretreatment did not reduce levels of fMLF-induced VASP Ser157 phosphorylation as compared with VC. In contrast, H89 (25 µM) or SQ22536 (3 mM) pretreatment resulted in significantly reduced levels of fMLF-induced VASP Ser 157 phosphorylation, similar to those of untreated cells (*, P<0.05). (B) Cells were pretreated as in A with a range of concentrations of the indicated inhibitors before stimulation with 100 nM fMLF. Inhibition of PKC (staurosporine and calphostin C), PKG (KT5823 and Rp-8-pCPT-cGMP), or calmodulin-dependent protein kinase II (CamKII; KN-62) did not reduce fMLF-induced VASP Ser 157 phosphorylation levels. Data are representative of at least three separate experiments using neutrophils from different donors.
|
 View larger version (33K): [in a new window] |
Figure 6. Dephosphorylation of VASP requires down-regulation of PKA. Purified neutrophils were pretreated with the indicated concentrations of the PDE-IV-specific inhibitor Ro-20-1724 (Ro) to reduce degradation of cAMP or the PDE-V-specific inhibitor 4-MB to reduce cGMP degradation. Cells were alternatively treated with IBMX to reduce degradation of cAMP and cGMP. Cells were then stimulated with 100 nM fMLF for 20 min at 37°C after adherence to FCS-coated wells as described in Materials and Methods. Equal protein amounts at each time-point were analyzed for VASP by Western blotting, and densitometric analysis was performed as described previously. IBMX and Ro treatment resulted in a significant increase in the amount of phosphorylated VASP when compared with VC-treated cells (*, P<0.05). Treatment with 4-MB did not result in increased phosphorylation levels of VASP when compared with VC. Data are representative of at least three separate experiments using neutrophils from different donors.
|
 View larger version (16K): [in a new window] |
Figure 7. Effect of altered PKA activity on fMLF-induced migration. Purified, calcein-labeled neutrophils were pretreated with the indicated inhibitor prior to placement of 1 x 104 cells on a 2-µM membrane of a 96-well ChemoTx® plate. The lower wells contained chemotaxis buffer, with or without 100 nM fMLF, and cells on the top of the filter were in HBSS++ or HBSS++ with 100 nM fMLF suspension. After 1 h of incubation at 37°C, nonmigrated cells were scraped off the top of the filter. EDTA (0.5 mM) was added prior to centrifugation to release the cells from the lower side of the filter. The fluorescence was measured in the lower wells. Percent migration was calculated by dividing the fluorescence of the wells by the fluorescence of standard wells containing 1 x 104-labeled cells and is presented as the mean ± SE. Less than 10% of untreated cells exhibited random migration into wells without chemoattractant (HBSS/HBSS). In contrast, 100 nM fMLF induced a significant increase in directed migration (HBSS/fMLF) of untreated cells into the lower chamber. Pretreatment with H89 resulted in a significant reduction in directed migration toward fMLF when compared with untreated control cells. When PKA activity was prolonged with IBMX, there was also a significant reduction in fMLF-induced migration and in chemokinesis after H89 and IBMX treatment (fMLF/HBSS and fMLF/fMLF; *, P<0.05).
|
 View larger version (26K): [in a new window] |
Figure 8. Adhesion does not regulate fMLF-induced Ser 157 VASP phosphorylation. (A) Purified neutrophils were pretreated with 1 mM EDTA to prevent adhesion. Cells were then unstimulated or stimulated with 100 nM fMLF for the indicated time periods during a 20-min incubation at 37°C after plating on FCS-coated wells as described in Materials and Methods. Equal protein amounts at each time-point were analyzed for VASP by Western blot analysis. Densitometry was performed, and the density of the upper 50-kD and lower 46-kD bands was measured. % VASP Phosphorylated was calculated by dividing the upper phosphorylated band by the total amount of VASP (upper 50 kD band+lower 46 kD band) and presented as the mean ± SE. EDTA pretreatment significantly reduced the amount of fMLF-induced VASP phosphorylation at the 30 s, 60 s, and 3 min time-points (*, P<0.05) but did not affect the overall pattern of rapid phosphorylation followed by rapid dephosphorylation. (B) Purified neutrophils were pretreated with 100 µg/ml IB4 or W6/32 (W6) antibody prior to stimulation with fMLF as in A. There was no difference in Ser 157 phosphorylation after IB4 blockade of β2 integrins when compared with cells treated with the control antibody W6/32. Data are representative of at least three separate experiments using neutrophils from different donors.
|
 View larger version (19K): [in a new window] |
Figure 9. IC activation of neutrophils results in PKA-dependent Ser 157 VASP phosphorylation. Purified neutrophils were preincubated with 25 µM KT5720, 25 µM H89, or VC for 30 min at 37°C. Cells (2x106) cells were then added to FCS- or IC-coated wells of a 24-well tissue-culture plate (as described in Materials and Methods). Cells were allowed to adhere for 30 min before Triton lysis. Equal protein amounts were evaluated by Western blot analysis. p-Lpl blots were stripped and reprobed for total Lpl. There was no appreciable phosphorylation of Lpl or VASP in cells, which were allowed to adhere to FCS. In contrast, IC-induced adhesion resulted in phosphorylation of Lpl and VASP, and this phosphorylation was abolished by pretreatment with KT5720 or H89. Blots are representative of three separate experiments using neutrophils from three separate donors.
|
Alteration of PKA activity decreases fMLF-induced migration
Next, we examined the requirement for VASP Ser 157 phosphorylation and dephosphorylation in the mechanism of neutrophil migration in response to fMLF stimulation, assessing chemotaxis and chemokinesis. Our approach was to inhibit phosphorylation with 25 µM H89 or inhibit the dephosphorylation event with 400 µM IBMX and determine the effect on fMLF-induced migration. Pretreatment with H89 resulted in a significant reduction in directed migration toward fMLF when compared with untreated control cells ("HBSS/fMLF", Fig. 7
). When PKA activity was prolonged with IBMX, there was a significant reduction in fMLF-induced migration. In addition, there was a significant reduction in chemokinesis after H89 and IBMX treatment ("fMLF/HBSS" and "fMLF/fMLF", Fig. 7
). We conclude from these results that alteration in PKA activity, which corresponds to altered VASP Ser 157 phosphorylation/dephosphorylation, results in reduced neutrophil chemokinesis and chemotaxis.
Adhesion does not regulate VASP Ser 157 phosphorylation in neutrophils stimulated by fMLF
Adhesion is an important regulator of VASP Ser 157 phosphorylation in fibroblasts [39
]. Moreover, many neutrophil functions are dependent on adhesion for optimal activation. To determine whether adhesion regulates neutrophil activation through modification of VASP Ser 157 phosphorylation, we examined whether inhibition of adhesion affects fMLF-induced VASP Ser 157 phosphorylation. We found no difference in the kinetics or extent of VASP phosphorylation after fMLF stimulation in cells attached to a surface as compared with those in suspension (unpublished results). Treatment with 1 mM EDTA to block adhesion by inhibiting Ca2+- and Mg2+-dependent integrin ligand binding slightly reduced the magnitude of fMLF-induced VASP phosphorylation for up to 180 s (Fig. 7)
. By 5 min, EDTA-treated cells maintained similar levels of phospho-VASP as control cells. To alternatively block adhesion, we pretreated cells with the β2-integrin (CD18)-specific mAb IB4 [29
] or an isotype-matched control antibody, W6/32, specific for class I HLA [30
]. Pretreatment with IB4 at a concentration that maximally inhibited adhesion to β2-integrin substrates (unpublished results) did not affect fMLF-induced phosphorylation or dephosphorylation of VASP as compared with cells treated with control antibody at any time-point (Fig. 8
). We conclude that neither fMLF-induced phosphorylation nor subsequent dephosphorylation of VASP Ser 157 is β2-integrin adhesion-dependent.
IC-induced Ser 157 phosphorylation of VASP is PKA-dependent
IC-induced activation of neutrophils, through ligation of FcRs, results in integrin activation and initiates intracellular signaling necessary for the induction of an inflammatory response. IC-induced activation is dependent on PKA-dependent signaling and PI-3K pathways [40
, 41
]. Ligation of FcRs results in a transient increase in intracellular cAMP with a peak at 5 min and return to baseline by 15 min, and this increase is blocked by inhibition of PI-3K [40
]. Sustained adhesion to ICs, over 30 min, results in PKA-dependent phosphorylation of Lpl [40
, 42
]. To determine whether VASP potentially has a role in IC-induced integrin activation in neutrophils, we examined levels of phospho-VASP in IC-activated neutrophils. After adherence to FCS-coated wells, there was no appreciable phosphorylation of Lpl or VASP (Fig. 9
). In contrast, IC-induced adhesion resulted in phosphorylation of Lpl and VASP, and this phosphorylation was abolished by pretreatment with the PKA inhibitors KT5720 or H89. We can conclude from these data that IC stimulation of VASP Ser 157 phosphorylation and Lpl phosphorylation is PKA-dependent.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
It is attractive to speculate that VASP may regulate neutrophil migration by affecting the organization of the actin cytoskeleton. Similar to the role of VASP in other cells, VASP phosphorylation may have a negative effect on actin polymerization in neutrophils. The phosphorylation of VASP by PKA has been shown to inhibit F-actin binding and reduce actin nucleation [20 , 43 ]. Our data show that after fMLF stimulation of neutrophils, there is rapid polymerization of actin, which is maximal by 30 s, and the cells begin to exhibit a shape change with focal distribution of F-actin to the leading edge in preparation for directed migration (Fig. 4) [44 ]. Our data show that VASP is coordinately, maximally phosphorylated at 30 s after fMLF stimulation during this phase of rapid actin polymerization and maximal increase in F-actin content, with a return to baseline phosphorylation levels by 10 min. The beginning of VASP Ser 157 dephosphorylation coincides with a decrease in F-actin content and reorganization of F-actin in the cell as the cell body condenses (Fig. 4) . Our data suggest that transient phosphorylation of VASP may be important for initiating actin polymerization and the shape change required for adhesion and ultimately, polarization of chemoattractant-stimulated neutrophils. As Ser 157 phosphorylation reduces actin binding, it is likely that this rapid phosphorylation of VASP releases it from the cytoskeleton. This may in turn help destabilize the somewhat rigid cortical actin cytoskeleton of resting neutrophils, which may be necessary to ultimately enable polymerization of new F-actin and reorganization of the existing cytoskeletal structure.
This proposed model for the role of VASP Ser 157 phosphorylation in the regulation of F-actin dynamics and shape changes in migrating neutrophils is consistent with the observation that actin polymerization is inhibited by PKA-mediated phosphorylation of VASP Ser 157 [20 ]. Moreover, studies in platelets have shown that polymerization cannot occur until dephosphorylation occurs, most likely by protein phosphatases PP2A, PP2B, and PP2C (in vitro) [18 , 27 ]. Global activation of PKA with cAMP-elevating agents in neutrophils inhibits F-actin polymerization and organization [45 , 46 ], adhesion [4 , 47 ], and migration [3 , 48 , 49 ]. Our data suggest that this may at least in part be a result of reduced dephosphorylation of VASP.
Another way that Ser 157 phosphorylation could be regulating cell motility is through effects on the affinity of VASP for binding partners. VASP binds via its EVH1 domain to several proteins including vinculin, zyxin, and lamellipodin. Although it has been shown that Ser 157 phosphorylation does not affect VASP EVH1-mediated binding to vinculin or zyxin [50 , 51 ], there are likely alternate EVH1 domain-binding proteins, whose binding is affected by VASP Ser 157 phosphorylation levels. Ser 157 is located in the proline-rich region near the (GP5)3 motif, which binds the actin-binding protein profilin, promoting the addition of actin monomers to barbed ends of actin filaments [52 , 53 ]. It is surprising that VASP Ser 157 phosphorylation does not affect profilin binding directly. However, profilin competes with the signaling molecule Abl tyrosine kinase for binding to the praline-rich motif of VASP via their shared SH3 domains, and the interaction of Abl with VASP is reduced by Ser 157 phosphorylation [39 ]. It is interesting that the SH3 domains of the tyrosine kinases Fyn, Lyn, and Src bind the Ena/VASP family member, Ena/VASP-like (EVL), although phosphorylation of EVL Ser 157 inhibits only Src-SH3 binding [54 ]. Perhaps Ser 157 phosphorylation uncovers profiling-binding sites on VASP, thus indirectly increasing the association of profilin-actin monomer complexes with VASP. In this scenario, VASP dephosphorylation may then be necessary for relocalization of VASP-profilin-actin monomer complexes to sites of F-actin polymerization. Uncovering how the phosphorylation of Ser 157 of VASP in neutrophils affects assembly of binding-partner complexes will help clarify its role in cytoskeletal polarity and chemotaxis.
The association between VASP and integrins in neutrophils remains elusive. On one hand, VASP appears to be critical for integrin function, at least in platelets. Indeed, the ability of VASP to regulate platelet integrins is dependent on Ser 157 [26 ]. Conversely, integrin engagement has a profound effect on VASP Ser 157 phosphorylation in fibroblasts, and fibroblast detachment increases PKA-dependent VASP Ser 157 phosphorylation; reattachment decreases phosphorylation rapidly, yet Ser 157 becomes rephosphorylated to an intermediate level during cell-spreading [39 ]. Previous work has shown that β2 integrins down-regulate PKA activity during neutrophil adhesion [55 ], suggesting that a similar mechanism to that in fibroblasts for regulating VASP phosphorylation may exist. In light of this data, we sought to determine whether integrin-mediated adhesion was required for dephosphorylation of VASP in chemoattractant-stimulated neutrophils. We found no role for adhesion in the pattern of VASP phosphorylation in response to fMLF stimulation. We found no difference in fMLF-induced VASP phosphorylation or dephosphorylation when comparing cells in suspension to cells adherent to β2-integrin substrate FCS. Inhibition of β2-integrin engagement, through depletion of Ca2+ and Mg2+ or through antibody blockade, did not affect the rapid phosphorylation and dephosphorylation of VASP in response to fMLF. Conversely, activation of FcR and β2-integrin ligation with ICs resulted in VASP Ser 157 phosphorylation, and this response was PKA-dependent. Our data suggest that VASP Ser 157 phosphorylation and subsequent dephosphorylation occur prior to or independently of β2-integrin signaling and that stable, IC-induced adhesion alone is sufficient to induce phosphorylation. Thus, VASP Ser 157 phosphorylation serves an important yet incompletely understood role in integrin-mediated cell adhesion and migration, and further studies need to be done to dissect the sequence of events that occur during neutrophil adhesion and integrin engagement.
At the center of the regulatory control of neutrophil migration is PKA, which regulates β2-integrin avidity and adhesion in neutrophils [4 ] and directs migration through activation of its many effectors, including VASP. Our lab has shown recently that asymmetrical PKA activity is required for polarization of the actin cytoskeleton and neutrophil migration [5 ]. Indeed, a gradient of the PKA inhibitor expected to asymmetrically decrease PKA activity, resulting in low PKA activity at the leading edge and high PKA activity at the opposite pole, is sufficient to induce migration and polarize the actin cytoskeleton. This suggests that PKA activity must be regulated precisely and spatially to enable polarization and migration. When we presumably destroyed the spatial control of PKA activity with the use of H89, thereby inhibiting all activity, or IBMX and flooding the cell with cyclic nucleotides and destroying the ability of the cell to maintain local pockets of activity through localized PDE activity, there was a significant reduction in chemokinesis and chemotaxis (Fig. 7) . We can correlate this data with the knowledge that PKA induces VASP phosphorylation and that the down-regulation of PKA is required for the dephosphorylation and therefore, conclude that altered VASP phosphorylation results in reduction in neutrophil migration. Until we are able to genetically modify VASP phosphorylation levels in primary neutrophils, as previously accomplished in fibroblasts [23 ], we cannot fully define the impact and function of VASP Ser 157 phosphorylation on neutrophil migration.
We have shown that VASP Ser 157 phosphorylation in response to fMLF is PKA-dependent. In support, the kinetics of VASP Ser 157 phosphorylation closely matches that of adenylyl cyclase activity. In neutrophils, chemoattractants fMLF, LTB4, C5a, and IL-8 induce a rapid and transient stimulation of adenylyl cyclase activity, which peaks by 1 min and returns to baseline by 8 min [56 ]. Preincubation of neutrophils with H89 or the adenylyl cyclase inhibitor SQ22536 resulted in abolishment of fMLF-induced VASP phosphorylation. Yet, preincubation with KT5720 did not affect phosphorylation. Although KT5720 is regarded as a PKA inhibitor, it in fact inhibits many other protein kinases more potently [38 ]. As a promiscuous inhibitor, perhaps KT5720 is also inhibiting an alternate kinase responsible for the down-regulation of fMLF-induced PKA activity, resulting in a net intermediate level in the cell and no apparent reduction in VASP phosphorylation levels as visualized by Western blotting. Alternatively, preincubation with KT5720 before IC stimulation resulted in the abolishment of VASP phosphorylation and PKA-dependent Lpl phosphorylation. One explanation for the discrepancy between H89 and KT5720 to inhibit VASP Ser 157 phosphorylation after fMLF stimulation could be explained by differences in their ability to penetrate the membrane or once cytosolic, various subcellular regions of the cell. KT5720 has been shown to inhibit PKA-dependent, endothelin-stimulated neutrophil migration but only after cells were electroporated [57 ], which suggests that membrane permeation of KT5720, structurally unrelated to H89, may be unpredictable in neutrophils. The various isoforms of PKA have a unique, spatial restriction and ability to be active in precise "micropockets" in the cytosol because of the assortment of specific A kinase anchoring proteins, which help localize the enzyme [58 ].
As neutrophils have impressive, 50,000 membrane fMLF receptors with a dissociation constant of only 20 nmol/L [59 ], we can hypothesize that potent and geographically specific inhibition of PKA would be necessary prior to stimulation to see an appreciable reduction in cAMP-dependent VASP phosphorylation levels. Perhaps different isoforms of PKA are responsible for IC- versus fMLF-induced VASP phosphorylation to fMLF. The relative potencies of H89 and KT5720 for the two different isoforms and their splice variants of the PKA catalytic subunits are unknown [58 , 60 ]. As increases in cAMP, using a PDE-IV-specific inhibitor, prolonged fMLF-induced VASP Ser 157 phosphorylation, and increases in cGMP, using a PDE-V-specific inhibitor, had no effect, we conclude that fMLF-induced VASP Ser 157 phosphorylation is indeed cAMP-dependent. We have considered the possibility that cAMP is signaling in a PKA-independent manner: One recently identified signaling target is an exchange protein, activated directly by cAMP (Epac). Epac is a signaling molecule that links cAMP to the small GTPase Rap1 and has been implicated in the pathways for cell adhesion [61 , 62 ], insulin secretion [63 , 64 ], exocytosis [63 , 65 ], and as an inhibitor of the ERK5 pathway in cardiomyocytes [63 , 66 ]. As H89 does not inhibit Epac, and Epac protein is not expressed in fMLF-stimulated neutrophils [67 ], we do not consider Epac a likely candidate for cAMP-activated VASP phosphorylation.
In conclusion, we have shown in this paper that pretreatment of cells with the PKA inhibitor H89, which abolishes VASP phosphorylation in response to fMLF, or a PKA activator IBMX, which prolongs VASP phosphorylation in response to fMLF, results in a decrease in fMLF-directed migration. VASP appears to be an important effector of PKA in the mechanism of neutrophil migration. In our model of neutrophil migration, we expect VASP to be phosphorylated predominately on Ser 157 at the trailing-edge and dephosphorylated predominately at the leading-edge pseudopod. This model is consistent with our proposed mechanism by which VASP control of F-actin polymerization and structure may be regulated by Ser 157 phosphorylation. However, our hypothesis is in contrast to recent studies in fibroblasts, which demonstrate that PKA activity and Ser 157 phospho-VASP are enriched in fibroblast lamellipodia [68
]. Moreover, spatial restriction of PKA to the fibroblast leading edge and subsequent phosphorylation of 4 integrins is required for migration, suggesting that the role for PKA in regulating fibroblast and neutrophil polarity and migration is fundamentally different. Hence, a unique role for PKA substrates such as VASP may exist for neutrophils and other cells, which must respond rapidly to chemoattractant signals and initiate migration more quickly than fibroblasts.
Received February 20, 2006; revised July 3, 2007; accepted July 11, 2007.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
M β2 avidity in polymorphonuclear neutrophils J. Biol. Chem. 273,10556-10566RII-mediated adhesion and phagocytosis induce L-plastin phosphorylation in human neutrophils J. Biol. Chem. 271,14623-14630This article has been cited by other articles:
 |
|
 |
|
  N. F. Neel, M. Barzik, D. Raman, T. Sobolik-Delmaire, J. Sai, A. J. Ham, R. L. Mernaugh, F. B. Gertler, and A. Richmond VASP is a CXCR2-interacting protein that regulates CXCR2-mediated polarization and chemotaxis J. Cell Sci., June 1, 2009; 122(11): 1882 - 1894. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. P. Lee, C. H. Serezani, A. I. Medeiros, M. N. Ballinger, and M. Peters-Golden Crosstalk between Prostaglandin E2 and Leukotriene B4 Regulates Phagocytosis in Alveolar Macrophages via Combinatorial Effects on Cyclic AMP J. Immunol., January 1, 2009; 182(1): 530 - 537. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
