Different Gi-coupled chemoattractant receptors signal qualitatively different functions in human neutrophils Miles Berger, Sadna Budhu, Emily Lu, Yongmei Li, Devora Loike, Samuel C. Silverstein, and John D. Loike Department of Physiology and Cellular Biophysics, Columbia University College of Physicians and Surgeons, New York, New York
Abstract: fMLP- or TNF-␣-stimulated neutrophils produced H2O2 when they adhered to fibrinogen-coated surfaces but not when they adhered to collagen I-, collagen IV-, or Matrigel-coated surfaces. In contrast, LTB4- or IL-8-stimulated neutrophils did not produce H2O2 when they adhered to any of these surfaces. fMLP and TNF-␣ were much more potent than LTB4 and IL-8 in stimulating neutrophils to up-regulate and to activate their ␣M2 integrins, as measured by the binding of specific monoclonal antibodies. Pretreatment of neutrophils with pertussis toxin completely blocked their production of H2O2 on fibrinogen-coated surfaces in response to fMLP and their migration through Matrigel in response to fMLP, LTB4, and IL-8. These data show that although the fMLP, LTB4, and IL-8 receptors are coupled to pertussis toxin-sensitive G␣ proteins, they signal neutrophils to initiate qualitatively different effector functions. We propose that the qualitative differences in effector functions signaled by different chemoattractants reflect qualitative differences in using G-protein  and/or ␥ subunits or other factors by their cognate receptors. J. Leukoc. Biol. 71: 798 – 806; 2002. Key Words: H2O2 䡠 fMLP 䡠 LTB4 䡠 integrins 䡠 fibrin
INTRODUCTION Leukocyte adhesion and chemotaxis are highly regulated functions that involve multiple signaling pathways [1]. In neutrophils, these functions are regulated differentially by specific chemoattractants and extracellular matrix proteins [2, 3]. For example, leukotriene B4 (LTB4), interleukin (IL)-8, formylMet-Leu-Phe (fMLP), and tumor necrosis factor ␣ (TNF-␣) stimulate neutrophils to adhere loosely to and migrate through three-dimensional gels composed of collagen I, collagen IV (CIV), or Matrigel [2, 3]. In contrast, only LTB4 and IL-8 stimulate neutrophils to migrate through fibrin gels [2, 3]. fMLP and TNF-␣ stimulate neutrophils to bind tightly to fibrin gels, thereby blocking chemotaxis [2]. Chemoattractants stimulate neutrophil adhesion and chemotaxis differentially by affecting expression and activation of integrins [3, 4]. For example, fMLP and TNF-␣, but not LTB4 or IL-8, activate neutrophil 1 integrins as measured by bind798
Journal of Leukocyte Biology Volume 71, May 2002
ing of 15/7, a monoclonal antibody (mAb) that recognizes an activation epitope on 1 integrins [3]. Interactions of these activated 1 integrins with matrix-associated ligands generate outside-in signals that increase the tightness of neutrophil adhesion to specific matrix proteins (i.e., fibrin) and block chemotaxis. Consistent with this, anti-1 integrin antibodies block tight adhesion between fMLP- or TNF-␣-stimulated neutrophils and fibrin-coated surfaces and allow these cells to migrate through fibrin gels [3]. Taken together, these findings strongly suggest that specific chemoattractants inhibit neutrophil chemotaxis by activating specific integrins. Chemoattractant-stimulated integrin activation also plays a pivotal role in neutrophil microbicidal responses such as H2O2 production. Studies have shown that TNF-␣ stimulates neutrophils to produce H2O2 when they adhere to specific matrix proteins such as vitronectin, fibronectin, and fibrinogen (Fg), but not when they are in suspension or adherent to stearic acid-coated surfaces [5, 6]. Interactions between specific 2 integrins and these matrix proteins are required for TNF-␣stimulated H2O2 secretion, because anti-2 integrin antibodies block this response [6], and 2 integrin-deficient neutrophils from patients with leukocyte-adhesion deficiency fail to produce H2O2 in response to TNF-␣ [6]. The studies demonstrated here were designed to test three hypotheses. First, do LTB4, IL-8, fMLP, and TNF-␣ exert differential effects on the expression and activation of 2 integrins? Second, is there any relationship between the effects of chemoattractants on the expression and/or activation of 2 integrins and their capacity to stimulate neutrophils to secrete H2O2? Third, do the chemoattractant receptors that mediate these responses do so through pertussis toxin-sensitive heterotrimeric G proteins? The data presented support an affirmative answer to all three questions. Intriguingly, they suggest that proteins other than pertussis toxin-sensitive G␣i subunits are responsible for the differential stimulation of neutrophil-effector functions that follow binding of LTB4, IL-8, and fMLP to their cognate seven-transmembrane spanning receptors.
Correspondence: John D. Loike, Department of Physiology and Cellular Biophysics, Columbia University College of Physicians and Surgeons, 650 W. 168th St., New York, NY 10027. E-mail:
[email protected] Received May 12, 2001; revised November 19, 2001; accepted December 3, 2001.
http://www.jleukbio.org
MATERIALS AND METHODS Reagents Mouse anti-human ␣M (MAC-1) and TNF-␣ were obtained from Upstate Biotechnology (Lake Placid, NY). CBRM1/5 was a generous gift of Dr. Timothy Springer (Harvard Medical School, Boston, MA), and mAb24 was kindly provided by Dr. Nancy Hogg (Imperial Cancer Research Fund, London, UK). IB4 was used as described [7]. 22F12C [8] was obtained from ICOS (Seattle, OR). fMLP, LTB4, and Ficoll-Hypaque were from Sigma Chemical Co. (St. Louis, MO). IL-8 was from Chemicon (Temecula, CA). Unless otherwise noted, fMLP was used at 10⫺6 M, LTB4 was used at 10⫺7 M, TNF-␣ was used at 10⫺8 M, and IL-8 was used at 10⫺7 M. CIV was from Fluka (Milwaukee, WI), collagen I was from Upstate Biotechnology, Matrigel was from Becton-Dickinson (Franklin Lakes, NJ), and Fg was from American Diagnostics (Greenwich, CT). Control isotype-matched antibodies [immunoglobulin IgG1 and IgG2b] were from Sigma Chemical Co., as were all other reagents.
Cells Neutrophils were isolated as described [9] from fresh, heparinized blood of healthy adults after informed consent. Cells (⬎95%) in each preparation were neutrophils, as determined by Wright-Giemsa staining [9]. Neutrophils were suspended in phosphate-buffered saline (PBS) with 5.5 mM glucose, 0.1% albumin, 0.5 mM Mg⫹⫹, and 1 mM Ca⫹⫹ (neutrophil buffer) and were maintained at 4°C until use.
H2O2 production Wells were coated with matrix proteins by the addition of either 50 l PBS containing 1 mg/ml Fg, 60 l PBS containing 1 mg/ml rat-tail collagen I, 45 l PBS containing 0.5 mg/ml Matrigel, or 50 l PBS containing 50 g CIV to each well of 96-well, 5-mm diameter-well, flat-bottom, polystyrene, tissueculture plates (Becton-Dickinson, Lincoln Park, NJ). The plates were incubated for 60 min at 37°C and PBS, and its content of unbound proteins was aspirated. Neutrophil buffer (175 l) containing 10 nmoles Amplex Red (Molecular Probes, Eugene OR), 0.21 units of horseradish peroxidase (HRP; Molecular Probes), 100 nmoles sodium azide (NaN3), and the appropriate stimulant(s) and inhibitor(s) was placed in each well, followed by the addition of 25 l neutrophil buffer containing 75,000 neutrophils. Thus, each well (final volume, 200 l) contained 50 M Amplex Red, 1.05 units/ml HRP, and 500 M sodium azide. The plates then were incubated at 37°C for the times indicated. Fluorescence in each well was measured in a Cytofluor II fluorometer using an excitation wavelength of 530 nm and an emission-detection wavelength of 590 nm as described [10]. For each experiment, sufficient H2O2 was placed into wells containing Amplex Red, HRP, and NaN3 to generate standard curves, which then were used to convert the arbitrary fluorescence units into nanomoles H2O2 produced by neutrophils under various conditions.
Fluorescein-activated cell-sorter (FACS) analysis FACS analysis was performed as described [3]. Briefly, neutrophils (105 cells/200 l neutrophil buffer) were incubated in suspension at 37°C for 30 min in the presence or absence of fMLP (10⫺7 M) or LTB4 (10⫺7 M), transferred to 96-well polystyrene tissue-culture microtiter plates (Corning, Corning, NY), incubated for 30 min at 4°C in 200 l neutrophil buffer containing the indicated primary antibody (2 g/ml), washed three times with neutrophil buffer at 4°C, incubated further for 30 min at 4°C with Alexa-488conjugated or phycoerythrin-conjugated rabbit anti-mouse F(ab⬘)2 in 200 l neutrophil buffer, washed three times again with neutrophil buffer at 4°C, and resuspended at 4°C in 300 l PBS containing 2% bovine serum albumin and 0.3 mg/ml propidium iodide to determine cell viability. The contribution of dead cells (usually ⬍2%) was removed from the final data analysis. The mean fluorescence intensity of 3–5 ⫻ 103 cells was determined using a BectonDickinson FACSCalibur.
Chemotaxis assays Chemotaxis of neutrophils through fibrin gels was performed as described [2]. Briefly, cell-culture inserts (pore size, 8 m; Becton-Dickinson, Mountain View, CA) were overlaid with 100 l PBS containing 1 mg/ml Fg in the presence of 0.1 unit thrombin. The inserts were incubated at 37°C to allow
fibrin-gel formation. One unit of PPACK (Calbiochem-Novabiochem, San Diego, CA) was added to inhibit thrombin, and the gels were washed with 250 l PBS to remove PPACK and inactivated thrombin. Neutrophils (106) in 100 l neutrophil buffer were placed in the upper compartment of each Fg-coated insert and incubated for 0 – 6 h at 37°C in a humidified atmosphere containing 95% air/5% CO2. At the times and concentrations specified, chemoattractants, antibodies, and/or peptides were added to the top and/or bottom compartments in 500 l neutrophil buffer. At the end of incubations, chambers were shaken to dislodge neutrophils from the lower surface of the inserts. The medium in each lower compartment was collected, and the number of neutrophils was determined using a Coulter Counter.
Adhesion assays Adhesion assays were performed as described previously [11, 12]. Briefly, 25 l Fg (2 mg/ml in PBS) was spotted onto the center of 6-cm bacterial Petri dishes (Fisher 1007) for 90 min at room temperature. Plates were blocked with Tween 20 as described previously [11]. Neutrophils (4⫻106) were suspended in Hanks’ balanced saline solution containing 10 mM Hepes and 1 mM MgCl2 (pH 7.3) and were preincubated with CBRM1/5 for 10 min at room temperature. Then, cells were added to the dishes, treated with fMLP or control buffer (such that the final volume in each dish was 3 ml), and allowed to adhere for the indicated times. Nonadherent cells were removed as described [12], and the number of adherent cells per 40⫻ ocular field was counted.
Statistical analysis Student’s t-test was used where appropriate when comparing the effects of various antibodies or pertussis toxin on the indicated neutrophil-effector functions.
RESULTS Specific combinations of chemoattractants and matrix proteins promote H2O2 production by adherent neutrophils Neutrophils stimulated with LTB4, IL-8, or TNF-␣ in suspension produce no detectable H2O2 above background (refs. [5, 6], and unpublished results). Neutrophils plated on collagen I-, CIV-, Matrigel-, or Fg-coated surfaces or uncoated plastic produced little H2O2 in response to 10⫺9–10⫺6 M LTB4, 10⫺8–2.5 ⫻ 10⫺7 M IL-8, or buffer alone (Fig. 1, A and B, and unpublished results). fMLP- or TNF-␣-stimulated neutrophils produced 0.8 –1.5 and 0.5–1.0 nmoles H2O2, respectively, in 3 h when plated on Fg-coated surfaces (Fig. 1A) and in a dose-dependent manner (Fig. 1B). Thus, H2O2 production by neutrophils depends on the nature of the stimulant and the specific extracellular matrix protein to which they have adhered. In accordance with the findings of Nathan [5], we observed a 30-min delay in H2O2 production by TNF-␣- or fMLPstimulated neutrophils adherent to Fg-coated surfaces (Fig. 1C). Most of the H2O2 produced by TNF-␣- or fMLP-stimulated neurophils occurred within 60 min of plating these cells on Fg-coated surfaces. No detectable amounts of H2O2 above background were produced by LTB4-stimulated neutrophils (Fig. 1, A–C). Numerous studies have shown that Ras, Erk, p38, p90rsk and other signaling proteins whose activity is regulated by formyl peptide receptors become activated within seconds or minutes of binding of formyl peptides to their cognate receptors [13– 21]. Yet, chemoattractant-stimulated neutrophils do not begin H2O2 production until 30 min later. Therefore, we examined Berger et al. Different Gi-coupled chemoattractant
799
Fig. 1. H2O2 production by LTB4-, TNF␣-, or fMLP-stimulated neutrophils adherent to plastic or to CIV-, Matrigel-, or Fg-coated surfaces. (A) Neutrophils (7.5⫻104) were incubated in buffer containing HRP, Amplex Red, and azide and the indicated stimuli in collagen I, CIV, Matrigel, Fg, or uncoated wells for 3 h at 37°C, at which time H2O2 production was measured as described in Materials and Methods. (B) Neutrophils were plated on Fg, treated at T ⫽ 0 with the indicated concentrations of fMLP, TNF-␣, LTB4, or control buffer, and were maintained at 37°C in buffer containing HRP, Amplex Red, and azide. H2O2 production was measured at 3 h as described in Materials and Methods. (C) Neutrophils were plated on Fg-coated wells, treated at T ⫽ 0 with fMLP (10⫺6 M), TNF-␣ (10⫺8 M), LTB4 (10⫺6 M), or control buffer, and were maintained at 37°C. H2O2 production was measured every 30 min. (D) Neutrophils were pretreated with buffer alone or buffer containing fMLP for 20 min at 37°C and then plated on Fg-coated wells. fMLP was then added to samples preincubated in control buffer. The data in A–D are from separate, representative experiments. Values are the means ⫾ SEM of triplicate wells (n⫽3).
whether this delay was a result of chemoattractant-receptor signaling or cell-matrix interactions. Neutrophils preincubated with fMLP (Fig. 1D) for 20 min in suspension before plating on Fg also exhibited a 30-min delay before initiating H2O2 secretion. This indicates that the delay in onset of H2O2 secretion does not reflect signals or processes activated by ligation of fMLP receptors alone. Moreover, neutrophils plated on Fgcoated surfaces for 20 min before the addition of fMLP still exhibited an ⬃30-min delay before the onset of H2O2 production (unpublished results). These findings indicate that the delay in initiation of an oxidative burst does not depend on the settling time or initial adherence of neutrophils onto Fg-coated surfaces [11, 12] or the time required for signals to be activated by chemoattractant receptors [13–21]. Rather, the delay correlates with secondary, postadhesive events such as cell spreading [6] and chloride-ion efflux [22]. Our findings suggest that the delay in onset of H2O2 production depends on a convergence of signals arising from fMLP receptors and from adhesion-promoting receptors. 800
Journal of Leukocyte Biology Volume 71, May 2002
Interactions between neutrophil 2 integrins and Fg are required for fMLP- or TNF-␣-stimulated H2O2 production Nathan et al. demonstrated [5, 6] that antibodies that block 2 integrins inhibit H2O2 production by TNF-␣-stimulated neutrophils adherent to Fg-coated surfaces. We confirmed this result (Fig. 2). Similarly, 2 integrin-blocking antibodies such as IB4 [7] and 22F12C [8] inhibited H2O2 production by fMLP-stimulated neutrophils on Fg-coated surfaces (Fig. 2). Isotype-matched, control antibody had no effect on H2O2 production by fMLP- or TNF-␣-stimulated neutrophils, and IB4 (or 22F12C) had no effect on H2O2 production by phorbol 12-myristate 13-acetate (PMA)-stimulated neutrophils (Fig. 2). Control experiments showed that IB4 also blocked fMLP- or TNF-␣-stimulated H2O2 production, even in the presence of the protein kinase A (PKA) inhibitor H-89 (Fig. 2). This rules out the possibility [23] that this mAb blocks neutrophil functions in our system by stimulating increases in intracellular http://www.jleukbio.org
no effect on the length of delay between the addition of fMLP or TNF-␣ and the initiation of H2O2 production by neutrophils adherent to Fg-coated surfaces (unpublished results). CBRM1/5 slightly increased (P⬍0.05) the maximal amount of H2O2 produced by fMLP-stimulated neutrophils adherent to these surfaces after 3 h (Fig. 4B); the mechanism by which CBRM1/5 exerts this effect is unknown.
1 Integrins do not play a role in neutrophil H2O2 production
Fig. 2. mAb IB4 and 22F12C block fMLP- or TNF-␣-stimulated H2O2 production. Neutrophils (7.5⫻104) were preincubated with 5 g/ml IB4, 22F12C, or IB4 ⫹ H-89 (PKA inhibitor) at 4°C for 1 h; plated on Fg-coated wells in buffer containing HRP, Amplex Red, azide, and fMLP (10⫺6 M), TNF-␣ (10⫺8 M), PMA (25 nM), or control buffer with or without antibody as indicated; and incubated at 37°C for 3 h. H2O2 production then was measured as described in Materials and Methods. Isotype-matched antibodies (IgG2b) had no effect on H2O2 production. The data are the mean ⫾ SEM of triplicate wells, representative of three separate experiments. Student’s t-test analysis determined that IB4 and 22F12C (*, P⬍0.05) reduced H2O2 production significantly compared with the respective conditions in the absence of antibody.
1 Integrins send outside-in signals that arrest migration of fMLP- or TNF-␣-stimulated neutrophils through Fg-containing matrices [3]. The migration-inhibitory effect of fMLP and TNF-␣ is reversed by anti-1 integrin antibodies [3]. Because fMLP and TNF-␣ also stimulated H2O2 production on fibrin(ogen), we examined whether 1 integrins also send outside-in signals to stimulate H2O2 secretion. However, neither 5 nor 30 g/ml AiiB2 [25] or P4C10 [26] anti-1blocking mAb blocked H2O2 production by fMLP or TNF-
cAMP levels and inducing PKA-mediated protein phosphorylation.
LTB4, IL-8, TNF-␣, and fMLP stimulate expression and activation of 2 integrins differentially fMLP and LTB4 activate 1 integrins differentially [3]. Therefore, we compared the effects of chemoattractants that stimulated H2O2 production (e.g., TNF-␣ and fMLP) with those that do not (e.g., LTB4 and IL-8) on 2 integrin expression and activation. Neutrophils treated with LTB4, IL-8, TNF-␣, or fMLP exhibited increased surface expression of ␣M2 integrins (Fig. 3, A–C and F, and refs. [2, 4]). At the concentrations used, TNF-␣ and fMLP were nearly equally potent in stimulating ␣M2 integrin expression and did so to a much greater extent than LTB4 or IL-8. In addition to promoting increased surface expression of 2 integrins, chemoattractants stimulate conformational changes in these proteins [12] and their aggregation in the plane of the membrane. The former effect, termed activation, increases the affinity of 2 integrins for cognate ligands [12]. ␣M2 activation can be detected using antibodies (e.g., mAb24 [24] and CBRM1/5 [12]) that bind to conformation-dependent neoepitopes on 2 integrins. By this measure, fMLP and TNF-␣ activated ␣M2 integrins (Fig. 3, E and F). In contrast, LTB4 (Fig. 3, D and F) or IL-8 (Fig. 3F) did not. CBRM1/5 has been shown to block the short-term adhesion (4-min incubation) of fMLP-stimulated neutrophils to Fgcoated surfaces (ref. [12], and Fig. 4A). However, CBRM1/5 did not block long-term adhesion (30-min incubation) of fMLPstimulated neutrophils to Fg-coated surfaces (Fig. 4A) and had
Fig. 3. Effects of fMLP, TNF, LTB4, and IL-8 on CD11b surface expression and activation. Fluorescence intensity of neutrophils incubated as described in Materials and Methods with anti-CD11b mAb (A–C) or with CBRM1/5 that recognizes an activation epitope on CD11b (D and E), followed by Alexa 488-conjugated rabbit F(ab⬘)2 anti-mouse IgG. Fluorescence intensity of neutrophils incubated as described in Materials and Methods. (F) Summarizes the fluorescence intensity of neutrophils incubated with anti-CD11b mAb, CBRM1/5, or mAb24 (which recognizes an activation epitope on CD11b) followed by Alexa 488-conjugated F(ab⬘)2 anti-mouse IgG. Note that the y-axis scales of D and E differ from those in A–C. Data shown are from a single experiment done in duplicate, representative of three.
Berger et al. Different Gi-coupled chemoattractant
801
Fig. 4. Effect of CBRM1/5 on neutrophil adhesion and H2O2 production. (A) neutrophils (5⫻104) were preincubated with 10 g/ml CBRM1/5 at 4°C for 1 h, suspended in medium containing the same concentration of CBRM1/5 and fMLP (10⫺6 M) as indicated, plated on Fg-coated wells, and incubated at 37°C for 4, 30, or 60 min as indicated. Adhesion was measured as described in Materials and Methods. (B) Neutrophils (7.5⫻104) were preincubated with 20 g/ml CBRM1/5 at 4°C for 30 min, suspended in buffer containing HRP, Amplex Red, azide, and the indicated chemoattractant, and incubated in Fg-coated wells at 37°C for 3 h. H2O2 production was assayed as described in Figure 1. Data shown (A) are from a single experiment representative of three and are the means ⫾ SEM of quadruplicate samples. Data shown (B) are from a single experiment representative of three and are the means ⫾ SE of triplicate samples. Student’s t-test analysis determined that CBRM1/5 antibody reduced adhesion significantly (*, P⬍0.05) only at the 4-min time course.
␣-stimulated neutrophils on Fg-coated surfaces (Fig. 5, and unpublished results).
Role of Gi/o activation in neutrophil H2O2 production and chemotaxis
Fig. 5. Anti-1 integrin mAb do not block fMLP- or TNF-␣-stimulated H2O2 production. Neutrophils (7.5⫻104) were preincubated with 5 g/ml mAbA iiB2 (directed against 1 integrins) and plated on Fg-coated wells in buffer containing AiiB2 and fMLP, TNF-␣, or control buffer as indicated. Data shown are from a single experiment representative of three and are the means ⫾ SEM of triplicate samples.
802
Journal of Leukocyte Biology Volume 71, May 2002
Chemoattractant-stimulated activation and increases in expression of 1 and 2 integrins are thought to be mediated by heptahelical Gi-coupled receptors [27–30]. Numerous studies document that pertussis toxin, which adenosine 5⬘-diphosphate ribosylates and thereby inactivates G␣i/o, blocks virtually all signals initiated by the fMLP receptor [29, 31, 32]. However, all of these studies focused on the effects of pertussis toxin on fMLP-stimulated neutrophils in suspension. We therefore examined the effects of pertussis toxin on fMLP-, LTB4-, or IL-8-stimulated neutrophil adhesion to Fgcoated surfaces and on the up-regulation of integrins that neutrophils use to adhere to these surfaces. Treatment of neutrophils with pertussis toxin (0.5–5 g/ml) for 1 h did not inhibit fMLP-stimulated, adhesion-dependent H2O2 production (Fig. 6A). In contrast, pretreatment of neutrophils with 0.5 g/ml pertussis toxin for ⱖ3 h inhibited H2O2 production completely by fMLP-stimulated neutrophils adherent to Fg (Fig. 6A); it did not affect the ability of PMA or TNF-␣ to stimulate H2O2 production or the inability of LTB4 and IL-8 to stimulate H2O2 production (Fig. 6A, and unpublished results). Pertussis toxin also completely blocked LTB4- and IL-8-stimulated migration through fibrin and CIV (Fig. 6B, and unpublished results) and fMLP-stimulated migration through Matrigel (Fig. 6B). Finally, pertussis toxin significantly reduced LTB-4, IL-8-, and fMLP-stimulated 2 integrin up-regulation http://www.jleukbio.org
Fig. 6. Pertussis toxin blocks fMLP-stimulated H2O2 production by neutrophils plated on Fg-coated wells (A) and LTB4- or fMLP-stimulated neutrophil chemotaxis (B). (A) Neutrophils were preincubated in neutrophil buffer without or with 0.5 g/ml pertussis toxin for 1–3 h at 37°C. Then, 105 neutrophils were added to each Fg-coated well containing HRP, Amplex Red, and azide and were incubated for 3 h at 37°C with or without fMLP (10⫺7 M) or PMA (0.25⫻10⫺7 M). The medium then was assayed for conversion of Amplex Red to resorufin, and H2O2 production was calculated as described in Materials and Methods. (B) Neutrophils (4⫻106) were preincubated with buffer alone or with buffer containing 0.5 g/ml pertussis toxin for 1 h at 37°C, and 106 neutrophils from pertussis toxin-containing or control buffer medium were placed on fibrin gels or Matrigel in chemotaxis chambers. LTB4 (10⫺7 M) or fMLP (10⫺7 M) was placed in the bottom compartment, and chemotaxis was assayed after 6 h as described [3]. The data (A and B) are the means ⫾ SEM of one experiment representative of three performed using duplicate or triplicate samples, respectively. Student’s t-test analysis (B) determined that incubating neutrophils with pertussis toxin for 3 h (*, P⬍0.05) reduced H2O2 production significantly in response to fMLP and neutrophil chemotaxis in response to LTB4 or fMLP.
(Fig. 7A, and unpublished results), and fMLP stimulated 2 activation (Fig. 7B). As expected, pertussis toxin did not affect TNF-␣-stimulated 2 integrin up-regulation (Fig. 7A). Thus, Gi activation is necessary for all LTB4-, IL-8-, and fMLP-stimulated neutrophil functions examined here.
DISCUSSION We demonstrate that fMLP, like TNF-␣ [6, 33], stimulates neutrophils adherent to Fg-containing matrices to secrete H2O2 in a 2 integrin-dependent manner. LTB4 and IL-8 did not stimulate neutrophils adherent to Fg-containing matrices to secrete H2O2, and none of these chemoattractants stimulated neutrophils adherent to CIV or Matrigel to secrete H2O2. These data suggest that two signals are necessary to stimulate neutrophils to produce H2O2: ligation of specific chemoattractant receptors and interaction of 2 integrins with cognate ligands on extracellular matrix proteins. This paradigm is reminiscent of the two-signal hypothesis of adaptive immunity: Two different classes of receptors must be ligated on B or T cells for efficient activation of these cells [34]. The most striking and important finding shown here is that fMLP, LTB4, and IL-8 all bind to Gi-coupled cell-surface receptors [27–32], yet they activate different intracellular sig-
nal cascades [15] and stimulate different cellular functions (Figs. 1 and 3, and ref. [3]). Although the receptors for fMLP, LTB4, and IL-8 are all heptahelical Gi-coupled receptors, the signal cascades and cell functions stimulated by ligation of fMLP receptors are much more similar to those stimulated by ligation of TNF-␣ trimeric receptors than those stimulated by ligation of LTB4 and IL-8 receptors. For example, fMLP and TNF-␣ stimulate neutrophils to produce H2O2 on Fg-coated surfaces (Fig. 1), activate 1 and 2 integrins (Figs. 3 and 7), cause cessation of neutrophil chemotaxis through fibrin gels [2, 3], stimulate neutrophils to form tight zones of apposition with fibrin matrices [2, 3], and activate p38 mitogen-activated protein kinase (MAPK) [15]. LTB4 and IL-8 do not stimulate any of these effects in neutrophils. Yet, pertussis toxin blocked all LTB4- and fMLP-stimulated neutrophil functions studied here. That is, it blocked the capacity of LTB4 and fMLP to stimulate expression of 2 integrins (Fig. 7A), of LTB4 and fMLP to stimulate neutrophils to migrate through Matrigel (Fig. 6B), of LTB4 to stimulate neutrophils to migrate through fibrin gels (Fig. 6B), of fMLP to activate 2 integrins (Fig. 7B), and of fMLP to stimulate neutrophils to produce H2O2 on fibrin(ogen)-containing matrices (Fig. 6A). How could the activation of the same G protein by different stimuli result in different functional responses? Berger et al. Different Gi-coupled chemoattractant
803
Fig. 7. Pretreatment of neutrophils with pertussis toxin blocks fMLP- and LTB4-stimulated 2 integrin up-regulation (A) and fMLP-stimulated, 2 integrin activation (B). Neutrophils (4⫻106) were preincubated in suspension with or without pertusssis toxin (0.5 g/ml) for 2 h at 37°C as described in Figure 6. The indicated chemoattractants [fMLP (10⫺7 M), LTB4 (10⫺7 M), or TNF (10⫺7 M)] were added to aliquots of 105 cells, and they were incubated for 30 min at 37°C, at which time CD11b up-regulation and activation were measured by FACS as described in Figure 3. Data are the average arbitrary fluorescence units per cell for each treatment from triplicate samples from one experiment representative of three. Student’s t-test analysis determined that incubating neutrophils with pertussis toxin for 3 h significantly (P⬍0.05) reduced the average binding per cell of mAb CD11b or CBRM1/5 in response to fMLP and LTB4.
First, there may be different thresholds of Gi activation for chemotaxis and H2O2 production, similar to the T-cell-receptor activation thresholds that are thought to govern thymic T-cell selection [35]. For example, at any given concentration, fMLP stimulates a much higher level of Gi activation in neutrophils than LTB4 [36]. At concentrations ranging from nM to M, fMLP and LTB4 stimulate neutrophil chemotaxis through gels formed of collagen I, CIV, or Matrigel [2]. However, concentrations of 0.1–1 M were required for fMLP to stimulate neutrophils to produce H2O2 (Fig. 1B). This suggests that the threshold of Gi activation required for H2O2 production is higher than for chemotaxis. Assuming that this is correct and that fMLP and LTB4 receptors activate the same Gi subunits, then very high LTB4 concentrations should activate Gi to about the same extent as 0.1 M fMLP and thereby stimulate H2O2 production. This model also predicts that low fMLP concentrations should activate Gi to the same extent as LTB4 concentrations that are sufficient to promote chemotaxis through fibrin matrices. Yet, even micromolar concentrations of LTB4 did not stimulate H2O2 production (Fig. 1B), and subnanomolar (0.1 nM) fMLP did not stimulate chemotaxis through fibrin gels [2]. Therefore, differences between the functional effects of fMLP versus LTB4 on neutrophils cannot be ascribed simply to different activation thresholds of the same Gi heterotrimer. Instead, there must be something qualitatively different about the signals initiated by fMLP receptors and those initiated by LTB4 receptors. Second, it is possible that different seven-transmembranespanning chemoattractant receptors couple to different G␣i subunits. The sole instance of such selectivity was described by Senogles [37], who demonstrated that alternative splice forms of the D2 dopamine receptor selectively couple to different G␣i subunits. Neutrophils express at least two different G␣i subunits, G␣i2 and G␣i3 [38, 39]. However, to our knowledge, there are no instances shown in which G␣i2 and G␣i3 804
Journal of Leukocyte Biology Volume 71, May 2002
activate different signal cascades or cellular functions. Because both of these subunits are sensitive to pertussis toxin, there are no specific pharmacological inhibitors to determine whether these receptors couple to different G␣i subunits in neutrophils. Third, the fMLP and LTB4 receptors might couple with similar affinity to each of the G␣i subunits but preferentially couple to different ␥ subunits. At least 5 different G subunits and 12 different G␥ subunits have been identified and characterized [40]. In fact, experiments with antisense oligonucleotides have shown that the muscarinic M4 and somatostatin receptors couple to different G␥ subunits, although both receptors couple to G␣o [41, 42]. Azpiazu et al. [43] have shown that molecular contacts between G␥ subunits and heptahelical receptors are necessary for G-protein activation, indicating that heptahelical receptors could have structural specificity for particular G␥ subunits. Moreover, different G␥ subunits have been shown to activate certain effector enzymes preferentially (e.g., phosphatidylinositol-3 kinase and phospholipase C [44]). Consistent with this concept, a G3 mutant has been found that enhances chemotaxis of human neutrophils in response to IL-8 and fMLP [45, 46] but does not affect fMLP-stimulated superoxide production. Taken together, these findings suggest that different receptors that activate the same G␣ subunits could nonetheless stimulate different signal cascades and cellular functions by preferentially coupling to and activating different ␥ subunits. Heptahelical receptors can couple directly to effector proteins other than heterotrimeric G proteins [47, 48]. In many instances, G proteins must be activated before these receptors can bind and/or activate other effectors. For example, Luttrell et al. [47] and McDonald et al. [48] have shown that following G-protein activation, specific -arrestins bind to -adrenergic receptors, thereby activating Src and MAPK. Furthermore, the studies of Neptune et al. [49] show that it is the heptahelical http://www.jleukbio.org
receptor and not the G␣ subunit that determines the cellular function that is stimulated by binding ligands to G-proteincoupled receptors. Although we favor the possibility that chemoattractant receptors couple selectively to different ␥ subunits, presently, our data do not allow us to distinguish between these possibilities. Nonetheless, they clearly show that there are qualitative differences between the effector functions activated by ligation of fMLP receptors versus those activated by ligation of LTB4 and IL-8 receptors. The simplest explanation for these differences is that these heptahelical Gi-coupled chemoattractant receptors initiate qualitatively different signals.
ACKNOWLEDGMENTS We thank Drs. Jens Huseman, Joan Brugge, James Smolen, Melvin Berger, and Steve Greenberg for helpful discussions, Nancy Hogg for the generous gift of mAb24, and Timothy Springer for the generous gift of CBRM1/5. This research was supported by a Columbia University Summer Undergraduate Research Fellowship (to M. B.) and National Institutes of Health Grant AI20516 (to S. C. S.).
REFERENCES 1. Bokoch, G. M. (1995) Chemoattractant signaling and leukocyte activation. Blood 86, 1649 –1660. 2. Loike, J. D., el Khoury, J., Cao, L., Richards, C. P., Rascoff, H., Mandeville, J. T., Maxfield, F. R., Silverstein, S. C. (1995) Fibrin regulates neutrophil migration in response to interleukin 8, leukotriene B4, tumor necrosis factor, and formyl-methionyl-leucyl-phenylalanine. J. Exp. Med. 181, 1763–1772. 3. Loike, J. D., Cao, L., Budhu, S., Marcantonio, E. E., El Khoury, J., Hoffman, S., Yednock, T. A., Silverstein, S. C. (1999) Differential regulation of beta1 integrins by chemoattractants regulates neutrophil migration through fibrin. J. Cell Biol. 144, 1047–1056. 4. Weber, C., Katayama, J., Springer, T. A. (1996) Differential regulation of beta 1 and beta 2 integrin avidity by chemoattractants in eosinophils. Proc. Natl. Acad. Sci. USA 93, 10939 –10944. 5. Nathan, C. F. (1987) Neutrophil activation on biological surfaces. Massive secretion of hydrogen peroxide in response to products of macrophages and lymphocytes. J. Clin. Investig. 80, 1550 –1560. 6. Nathan, C., Srimal, S., Farber, C., Sanchez, E., Kabbash, L., Asch, A., Gailit, J., Wright, S. D. (1989) Cytokine-induced respiratory burst of human neutrophils: dependence on extracellular matrix proteins and CD11/CD18 integrins. J. Cell Biol. 109, 1341–1349. 7. Wright, S. D., Rao, P. E., Van Voorhis, W. C., Craigmyle, L. S., Iida, K., Talle, M. A., Westberg, E. F., Goldstein, G., Silverstein, S. C. (1983) Identification of the C3bi receptor of human monocytes and macrophages by using monoclonal antibodies. Proc. Natl. Acad. Sci. USA 80, 5699 – 5703. 8. Sadhu, C., Masinovsky, B., Staunton, D. E. (1998) Differential regulation of chemoattractant-stimulated beta 2, beta 3, and beta 7 integrin activity. J. Immunol. 160, 5622–5628. 9. Wright, S. D., Weitz, J. I., Huang, A. J., Levin, S. M., Silverstein, S. C., Loike, J. D. (1988) Complement receptor type three (CD11b/CD18) of human polymorphonuclear leukocytes recognizes fibrinogen. Proc. Natl. Acad. Sci. USA 85, 7734 –7738. 10. Zhou, M., Diwu, Z., Panchuk-Voloshina, N., Haugland, R. P. (1997) A stable nonfluorescent derivative of resorufin for the fluorometric determination of trace hydrogen peroxide: applications in detecting the activity of phagocyte NADPH oxidase and other oxidases. Anal. Biochem. 253, 162–168. 11. Diamond, M. S., Staunton, D. E., Marlin, S. D., Springer, T. A. (1991) Binding of the integrin Mac-1 (CD11b/CD18) to the third immunoglobulin-like domain of ICAM-1 (CD54) and its regulation by glycosylation. Cell 65, 961–971.
12. Diamond, M. S., Springer, T. A. (1993) A subpopulation of Mac-1 (CD11b/ CD18) molecules mediates neutrophil adhesion to ICAM-1 and fibrinogen. J. Cell Biol. 120, 545–556. 13. Avdi, N. J., Winston, B. W., Russel, M., Young, S. K., Johnson, G. L., Worthen, G. S. (1996) Activation of MEKK by formyl-methionyl-leucylphenylalanine in human neutrophils. Mapping pathways for mitogenactivated protein kinase activation. J. Biol. Chem. 271, 33598 –33606. 14. Downey, G. P., Butler, J. R., Brumell, J., Borregaard, N., Kjeldsen, L., Sue, A. Q. A. K., Grinstein, S. (1996) Chemotactic peptide-induced activation of MEK-2, the predominant isoform in human neutrophils. Inhibition by wortmannin. J. Biol. Chem. 271, 21005–21011. 15. Huang, R., Lian, J. P., Robinson, D., Badwey, J. A. (1998) Neutrophils stimulated with a variety of chemoattractants exhibit rapid activation of p21-activated kinases (Paks): separate signals are required for activation and inactivation of paks. Mol. Cell. Biol. 18, 7130 –7138. 16. Lian, J. P., Huang, R., Robinson, D., Badwey, J. A. (1999) Activation of p90RSK and cAMP response element binding protein in stimulated neutrophils: novel effects of the pyridinyl imidazole SB 203580 on activation of the extracellular signal-regulated kinase cascade. J. Immunol. 163, 4527– 4536. 17. Pillinger, M. H., Feoktistov, A. S., Capodici, C., Solitar, B., Levy, J., Oei, T. T., Philips, M. R. (1996) Mitogen-activated protein kinase in neutrophils and enucleate neutrophil cytoplasts: evidence for regulation of cell-cell adhesion. J. Biol. Chem. 271, 12049 –12056. 18. Thompson, H. L., Marshall, C. J., Saklatvala, J. (1994) Characterization of two different forms of mitogen-activated protein kinase kinase induced in polymorphonuclear leukocytes following stimulation by N-formylmethionyl-leucyl-phenylalanine or granulocyte-macrophage colony-stimulating factor. J. Biol. Chem. 269, 9486 –9492. 19. Torres, M., Hall, F. L., O’Neill, K. (1993) Stimulation of human neutrophils with formyl-methionyl-leucyl-phenylalanine induces tyrosine phosphorylation and activation of two distinct mitogen-activated protein-kinases. J. Immunol. 150, 1563–1577. 20. Worthen, G. S., Avdi, N., Buhl, A. M., Suzuki, N., Johnson, G. L. (1994) FMLP activates Ras and Raf in human neutrophils. Potential role in activation of MAP kinase. J. Clin. Investig. 94, 815– 823. 21. Zu, Y. L., Qi, J., Gilchrist, A., Fernandez, G. A., Vazquez-Abad, D., Kreutzer, D. L., Huang, C. K., Sha’afi, R. I. (1998) p38 Mitogen-activated protein kinase activation is required for human neutrophil function triggered by TNF-alpha or FMLP stimulation. J. Immunol. 160, 1982–1989. 22. Menegazzi, R., Busetto, S., Decleva, E., Cramer, R., Dri, P., Patriarca, P. (1999) Triggering of chloride ion efflux from human neutrophils as a novel function of leukocyte beta 2 integrins: relationship with spreading and activation of the respiratory burst. J. Immunol. 162, 423– 434. 23. Gresham, H. D., Graham, I. L., Anderson, D. C., Brown, E. J. (1991) Leukocyte adhesion-deficient neutrophils fail to amplify phagocytic function in response to stimulation. Evidence for CD11b/CD18-dependent and -independent mechanisms of phagocytosis. J. Clin. Investig. 88, 588 –597. 24. Dougherty, G. J., Dransfield, I., Hogg, N. (1988) Identification of a novel monocyte cell surface molecule involved in the generation of antigeninduced proliferative responses. Eur. J. Immunol. 18, 2067–2071. 25. Bohnsack, J. F., Akiyama, S. K., Damsky, C. H., Knape, W. A., Zimmerman, G. A. (1990) Human neutrophil adherence to laminin in vitro. Evidence for a distinct neutrophil integrin receptor for laminin. J. Exp. Med. 171, 1221–1237. 26. Carter, W. G., Wayner, E. A., Bouchard, T. S., Kaur, P. (1990) The role of integrins alpha 2 beta 1 and alpha 3 beta 1 in cell-cell and cellsubstrate adhesion of human epidermal cells. J. Cell Biol. 110, 1387– 1404. 27. Damaj, B. B., McColl, S. R., Mahana, W., Crouch, M. F., Naccache, P. H. (1996) Physical association of Gi2alpha with interleukin-8 receptors. J. Biol. Chem. 271, 12783–12789. 28. Igarashi, T., Yokomizo, T., Tsutsumi, O., Taketani, Y., Shimizu, T., Izumi, T. (1999) Characterization of the leukotriene B4 receptor in porcine leukocytes. Separation and reconstitution with heterotrimeric GTP-binding proteins. Eur. J. Biochem. 259, 419 – 425. 29. Wenzel-Seifert, K., Hurt, C. M., Seifert, R. (1998) High constitutive activity of the human formyl peptide receptor. J. Biol. Chem. 273, 24181– 24189. 30. Yokomizo, T., Izumi, T., Chang, K., Takuwa, Y., Shimizu, T. (1997) A G-protein-coupled receptor for leukotriene B4 that mediates chemotaxis. Nature 387, 620 – 624. 31. Lad, P. M., Olson, C. V., Smiley, P. A. (1985) Association of the N-formylMet-Leu-Phe receptor in human neutrophils with a GTP-binding protein sensitive to pertussis toxin. Proc. Natl. Acad. Sci. USA 82, 869 – 873. 32. Klinker, J. F., Wenzel-Seifert, K., Seifert, R. (1996) G-protein-coupled receptors in HL-60 human leukemia cells. Gen. Pharmacol. 27, 33–54.
Berger et al. Different Gi-coupled chemoattractant
805
33. Shappell, S. B., Toman, C., Anderson, D. C., Taylor, A. A., Entman, M. L., Smith, C. W. (1990) Mac-1 (CD11b/CD18) mediates adherence-dependent hydrogen peroxide production by human and canine neutrophils. J. Immunol. 144, 2702–2711. 34. Liu, Y., Janeway Jr., C. A. (1992) Cells that present both specific ligand and costimulatory activity are the most efficient inducers of clonal expansion of normal CD4 T cells. Proc. Natl. Acad. Sci. USA 89, 3845–3849. 35. Ashton-Rickardt, P. G., Bandeira, A., Delaney, J. R., Van Kaer, L., Pircher, H. P., Zinkernagel, R. M., Tonegawa, S. (1994) Evidence for a differential avidity model of T cell selection in the thymus. Cell 76, 651– 663. 36. Schepers, T. M., Brier, M. E., McLeish, K. R. (1992) Quantitative and qualitative differences in guanine nucleotide binding protein activation by formyl peptide and leukotriene B4 receptors. J. Biol. Chem. 267, 159–165. 37. Senogles, S. E. (1994) The D2 dopamine receptor isoforms signal through distinct Gi alpha proteins to inhibit adenylyl cyclase. A study with site-directed mutant Gi alpha proteins. J. Biol. Chem. 269, 23120 –23127. 38. Didsbury, J. R., Snyderman, R. (1987) Molecular cloning of a novel human GTP-binding protein and its potential role in chemoattractant stimulusresponse coupling. Trans. Assoc. Am. Physicians 100, 99 –108. 39. Beals, C. R., Wilson, C. B., Perlmutter, R. M. (1987) A small multigene family encodes Gi signal-transduction proteins. Proc. Natl. Acad. Sci. USA 84, 7886 –7890. 40. Downes, G. B., Gautam, N. (1999) The G protein subunit gene families. Genomics 62, 544 –552. 41. Kleuss, C., Scherubl, H., Hescheler, J., Schultz, G., Wittig, B. (1992) Different beta-subunits determine G-protein interaction with transmembrane receptors. Nature 358, 424 – 426.
806
Journal of Leukocyte Biology Volume 71, May 2002
42. Kleuss, C., Scherubl, H., Hescheler, J., Schultz, G., Wittig, B. (1993) Selectivity in signal transduction determined by gamma subunits of heterotrimeric G proteins. Science 259, 832– 834. 43. Azpiazu, I., Cruzblanca, H., Li, P., Linder, M., Zhuo, M., Gautam, N. (1999) A G protein gamma subunit-specific peptide inhibits muscarinic receptor signaling. J. Biol. Chem. 274, 35305–35308. 44. Maier, U., Babich, A., Macrez, N., Leopoldt, D., Gierschik, P., Illenberger, D., Nurnberg, B. (2000) Gbeta 5gamma 2 is a highly selective activator of phospholipid-dependent enzymes. J. Biol. Chem. 275, 13746 –13754. 45. Virchow, S., Ansorge, N., Rubben, H., Siffert, G., Siffert, W. (1998) Enhanced fMLP-stimulated chemotaxis in human neutrophils from individuals carrying the G protein beta3 subunit 825 T-allele. FEBS Lett. 436, 155–158. 46. Virchow, S., Ansorge, N., Rosskopf, D., Rubben, H., Siffert, W. (1999) The G protein beta3 subunit splice variant Gbeta3-s causes enhanced chemotaxis of human neutrophils in response to interleukin-8. NaunynSchmiedeberg’s Arch. Pharmacol. 360, 27–32. 47. Luttrell, L. M., Ferguson, S. S., Daaka, Y., Miller, W. E., Maudsley, S., Della Rocca, G. J., Lin, F., Kawakatsu, H., Owada, K., Luttrell, D. K., Caron, M. G., Lefkowitz, R. J. (1999) Beta-arrestin-dependent formation of beta2 adrenergic receptor-Src protein kinase complexes. Science 283, 655– 661. 48. McDonald, P. H., Chow, C. W., Miller, W. E., Laporte, S. A., Field, M. E., Lin, F. T., Davis, R. J., Lefkowitz, R. J. (2000) Beta-arrestin 2: a receptorregulated MAPK scaffold for the activation of JNK3. Science 290, 1574 – 1577. 49. Neptune, E. R., Iiri, T., Bourne, H. R. (1999) Galphai is not required for chemotaxis mediated by Gi-coupled receptors. J. Biol. Chem. 274, 2824 – 2828.
http://www.jleukbio.org
