Journal
of Leukocyte
Biology
48:353-358
Flow Cytometric Study of the Activation Polymorphonuclear Cells Francis
Belloc,
Philippe
Laboratoire (G.F.),
Vincendeau, and Michel
d’Hematologie,
Pessac,
and
Genevieve Freyburger, R. Boisseau
H#{244}pital Cardiologique
Laboratoire
d’Immunologie,
(F.B.,
M.R.B.),
PD.,
H#{244}pitalPellegrin,
of
Patrice
Unite
Bordeaux
(1990)
Dumain,
8, INSERM (P.V.),
France
The activation of human polymorphonuciear cells (PMN) by the chemotactic peptide, N-formyl-methionyl-leucyl-phenylalanine (FMLP), or the protein kinase C activator, phorbol myristate acetate (PMA), was studied using flow cytometry. Two probes were used to evaluate PMN activation: 1) a monoclonal antibody (M0F1 1) directed against an antigen (Ag) expressed on the membrane of monocytes and of activated PMN; 2) rhodamine phalloidln was used at the cytoplasmic level to measure the F-actin content. The expression of M0F1 1 antigen was found to be 3 to 5 times greater on the membrane of PMN activated by either FMLP or PMN as compared with membrane expression of the same Ag on resting PMN. This increase was found to be dose dependent for the two activators. Kinetic studies showed that a maximum response was observed in 1 to 2 mm at 37#{176}C when FMLP was used, whilst a similar response required 10 mm when PMA was used. The same discrepancy with activators was observed when actin polymerization was measured by labellIng with rhodamine phalloidin. However, pretreatment of PMN with cytochalasin B inhibited actin polymerization whilst M0F1 1 antigen expression was increased, suggesting that the M0F1 1 antigen could be stored in granules of resting PMN. The study of actin polymerization and of M0F1 1 antigen expression, separately or in combination, could be a useful tool for the detection of activated PMN in biological samples. Key words:
actin
polymerization,
cytoskeleton,
INTRODUCTION When
exposed
to chemoattractant
agents,
polymor-
phonuclear leukocytes (PMN) respond rapidly with a highly coordinated series of responses. In response to N-formyl-methionyl-leucyl-phenylalanine (FMLP) for example, activation occurs as a consequence of the stimulus binding to the receptors on the cell membrane [6]. Immediate, secondary changes in PMN metabolism occur; these include the activation of phospholipase C, the release and phosphorylation of phosphatidyl inositol,the increase in free cytoplasmic Ca and the activation of protein kinase C [12]. These metabolic events lead to physiologic responses by the PMN such as the release of inflammatory mediators and chemotaxis. Among biochemical events contributing to chemotaxis are actin polymerization [7,16] and the exposure of antigens on the surface membrane [2,4,5,8,18]. Another stimulus that induces PMN response is phorbol myristate acetate (PMA), which acts by the activation of protein kinase C. To date, flow cytometry has been used as a tool to follow PMN activation using monoclonal antibodies to detect membrane antigen response [2. 18] and actin polymerization [7,16], cytosolic calcium [10], and oxida,
© 1990
Wiley-Liss,
Inc.
membrane
antigen,
monoclonal
antibody
tive products [17] amongst other parameters. These studies were performed on isolated PMN and focused on only one of these parameters. In this paper, we have investigated the responses of PMN to two stimuli (FMLP and PMA) using flow cytometry. Two different probes were used on the same samples: one was a monoclonal antibody (MoFI 1) directed against a membrane antigen whose surface expression on PMN is increased by FMLP activation [1,5], and the other was rhodamine-phalloidin, which binds to cytoplasmic F actin and is frequently used to quantitate the rate of actin polymerization [16,19]. The kinetic and dose-dependent responses of membrane antigen expression and cytoplasmic actin polymerization were studied and compared. It appeared that the increase in F actin paralleled the increase in MoF1 I antigen expression suggesting that they are concomitant responses following
Received Reprint
March requests:
Cardiologique.
10.
1989;
Francis avenue
accepted Belloc,
Magellan.
March Laboratoire 33604
19,
1990.
d’Hematologie. Pessac-France.
H#{244}pital
354
Belioc
et al.
PMN activation. However the partial inhibition of F actin increase by cytochalasine B which was accompanied by an increase of MoF1 1 antigen expression suggested that the two responses were not completely interdependent functions. These results prove the usefulness of MoF1 1 and fluorescent phalloidin for flow cytometric analysis of the phenotype of activated PMN.
MATERIALS
AND METHODS
Antibodies 0KM 1 (directed against the C3bi receptor) was purchased from Ortho Diagnostic (France). Unrelated IgG (MsIgG2a and MsIgG2b) as well as fluoresceinated goat anti mouse IgG (GAM-FITC) were obtained from Coultronics (Margency; France). The monoclonal antibody MoF1 1 was obtained by immunizing BALB/c mice with human normal monocytes. The murine spleen cells were fused with murine myeloma cells X63 Ag8653. Positive clones were selected by indirect immunofluorescence on the basis of their ability to detect monocyte membrane antigens (P. Vincendeau, F. Belloc, et al.;manuscript in preparation). MoFI 1 is an IgG2a and has been submitted to the third and fourth International Workshop of Immunology for clusterization.
Cell Preparation
for Flow Cytometry
Blood was obtained by venipuncture with EDTA as anticoagulant. Whole blood leukocytes were enriched by gelatine sedimentation of red cells (1 volume of Plasmion to 4 volumes of blood) for 20 mm at 20#{176}C. The leukocytes were then pelleted by centrifugation, the remaining red cells were lysed in 0.85% ammonium chloride for 7 mm at 20#{176}C, and the leukocytes were washed and resuspended in RPM! 1640 at a concentration of 4.106 cells/mI and incubated at 37#{176}C for 5 mm. PMA or FMLP was then added to the leukocyte suspension at the indicated concentrations. The suspension was incubated at 37#{176}C to allow PMN activation, and samples were then processed as described below after different times of incubation.
lmmunofluorescence Leukocytes (106) were fixed with 1% paraformaldehyde (PFA) and 0.05% glutaraldehyde (final concentration) for 10 mm at 20#{176}C, washed with PBS, and incubated for 30 mm at 20#{176}C with 0.2 ml of PBS containing 10 pg of MoF1 1 or OKM1 IgG. The cells were then washed and incubated with 0.2 ml of a 1/40 dilution of GAM-FITC for 30 mm at 20#{176}C, washed again, and resuspended in PBS for flow cytometry. Non-specific labeling was measured by incubating 106 cells with unrelated IgG.
Rhodamine-Phalloidin
Binding
Leukocytes (5 x l0) were fixed, permeabilized, and stained at the same time by 2% PFA, 30 g/ml lysophosphatidyl choline, 0.1 xm of rhodamine-phalloidine (Molecular Probes, Eugene, OR). After 30 mm at 37#{176}C, the cells were washed in PBS and resuspended at 106 cells! ml in PBS for flow cytometry.
PMN Cytoskeletons Gel Electrophoresis
Analysis (PAGE)
by Polyacrylamide
PMN purification was required for cytoskeleton analysis. This was obtained by a two-step preparation procedure. Red blood cells from EDTA anticoagulated blood were sedimented on a 9.6% metrizoate and 5% Ficoll solution [3]; the plateletand leukocyte-rich plasma was then centrifuged on a discontinuous Percoll gradient [14]. Then 6 x 106 purified PMN in 500 l were activated with l0M FMLP or with 20 ng/ml PMA. At different times, 500 iii of lysing buffer [13] (20 mM EGTA, 20 mM imidazole HCI, 80 mM KC1, 2% Triton X-l00, pH 7.15) were added; the lysis was performed at 4#{176}C for 5 mm and centrifuged at 12 000g for 4 mm. The pellet was rinsed with washing buffer (lysing buffer without Triton X- 100) and solubilized in 150 pA of 10% SDS, 4 M urea, 20% glycerol, 2% 3-mercaptoethanol, pH 8). Proteins were then analyzed by SDS polyacrylamide gel electrophoresis (SDS-PAGE) on a 7.5% acrylamide slab gel using the Laemmli [9] discontinuous buffer system.
Flow Cytometry Samples prepared for flow cytometry were analyzed using an ATC 3000 (ODAM, Wissembourg, France) cell sorter equipped with a 2025 Spectraphysics argon ion laser. For immunofluorescence labeling, the laser was tuned on the 488 nm argon line and fluorescence was collected through a 515 nm long pass filter. For rhodamine-phalloidine experiments, the laser was tuned on the 514 nm argon line and fluorescence was collected through a 600 nm long pass filter. The fluorescence of PMN was analyzed by gating fluorescence of the PMN on the basis of their wide angle light scatter. Mean fluorescence channel was calculated by the ASPECT 3000 computer and the results were expressed as the activation ratio: mean fluorescence channel of activated PMN/ mean fluorescence channel of non-activated PMN.
RESULTS The blood
scatter antigen
fluorescence leukocyte
of PMN suspensions
can as
be measured a result
properties (Fig. 1). The expression (as measured by immunofluorescence)
of
in whole their
light
of MoFI 1 and the
Flow Cytometry
and PMN Activation
355
-J 4
FALS N
C
B
IL
I
of blood leukocytes labelled M0F1 1 antIbody. The fluoresPMN on a double scatter cythe x axis and the wide angle
light scatter on they axis. The fluorescence histograms of PMN after IndIrect labeling with M0F11 (B) or direct labeling with rhodamine-phalloidin (C) are shown for non actIvated (shadowed area) and iO M FMLP activated leukocytes.
F-actin content (as measured by rhodamine-phalloidin binding) were increased by activation of PMN. The increase in antigen expression was measured on PMN which were incubated for 5 mm with iM FMLP after treatment by cytochalasine B (5 xg/ml). Under these conditions, the activation ratio (AR, see “Materials and Methods”) was found to be 7 times higher for the MoF11 antigen (AR = 13.06 ± 2.26; n = 4) than for the OKM1 antigen (AR 1.89 ± 0.17; n = 4). This difference was essentially due to the relatively high level of C3bi receptor expression on the membrane of resting
PMN as compared with the basal level of MoFl 1 antigen expression. The kinetics of MoF 11 antigen expression and F-actin content during activation of PMN with FMLP or PMA are shown in Figure 2. Following activation by FMLP, the two parameters were rapidly increased and reached maximum levels after 1 mm for F-actin content and after 2 mm for MoF1 1 antigen expression (Fig. 2A). When activation was performed using PMA, the response kinetics were different. MoF1 1 antigen membrane expression increased progressively for 10 to 15 mm while the
FIg. 1. Flow cytometry analysis with rhodamine phalloldin and/or canoe analysis was gated on the togram (A). The axial scatter is on
Belloc
356
et al.
IMLI’
PMA
14
6
20
ny/mI
6 MaF
MoE
11
11
C
C
13
A
14
Phailoidin
:3
:2, C
C
1
2
3
Activation
4 time
5
mm.
5 Activation
10
15
zU
m!r.
time
Fig. 2. Kinetics of activation by FMLP and PMA on M0F11 membrane antigen expression and rhodamine-phalloidin binding in the cytoplasm of PMN. Whole blood leukocytes were activated by either iO- M FMLP (A) or 20 ng/mi PMA (B) for the Indicated times, fixed, and labeled using either M0F11 immuno fluorescence (._..) or rhodamine phailoidin (0-0). The fluorescence of the PMN was then analyzed by flow cytometry and the results are expressed as the activation ratio in PMN as a function of activation time. The data on this figure are from one representative experiment.
cytoplasmic F-actin increase was delayed and appeared only after 10 mm of activation (Fig. 2B). When isolated PMN were activated by FMLP, rapid actin incorporation into the cytoskeletal framework was observed using PAGE analysis (Fig. 3A). When the PMN were activated by PMA, the incorporation of actin into the cytoskeleton lasted during 5 mm and then decreased (Fig. 3B) while F-actin content continued to increase (Fig. 2B). The dose-response curves of MoFI 1 antigen expression and F-actin content after FMLP or PMA activation are shown in Figure 4. There was a dose-dependent increase in the F-actin content after activation with FMLP (10_to to l0 M FMLP). A similar pattern was observed with MoFI I Ag expression (Fig. 4A). The Factin increased and reached a plateau at a concentration of 2 ng/ml of PMA. Higher concentrations (one log) of PMA were required to induce a similar activation ratio with membrane Ag expression (Fig. 4B). Figure 5 demonstrates that 20 ng of PMA and l0 M FMLP induce similar activation ratios for antigen expression and actin polymerization. However, when leukocytes were pretreated with cytochalasmn B and then activated with FMLP, the MoFl 1 Ag expression was significantly increased fivefold whilst the actin polymerization response was significantly inhibited in PMN.
DISCUSSION Treatment of PMN by the chemoattractant FMLP or by protein kinase C activation (PMA) induces a motile response in PMN with extensive changes in the cytoplas-
mic contractile apparatus and in the membrane proteins. Actin polymerization is known to occur during PMN activation [12]. This polymerization was measured either by phalloidin binding and flow cytometry, or by Triton X-lOO cytoskeleton extraction and SDS-PAGE. These measurements were performed on isolated PMN [7,16]. In the present study, activation and measurements were performed on whole blood leukocyte suspensions contaminated by platelets but with similar results to previous reports. The PMN specific responses were assessed by electronic gating of the fluorescence analysis on the scattergram. The cell preparation was greatly simplified and allowed the use of flow cytometric measurement of actin polymerization by means of a routine assay. Rapid actin polymerization was found after PMN activation by FMLP, whilst, in response to PMA treatment, the F actin content increased only after 10 mm. This was in agreement with previous results which found that PMA induced an actin polymerization response that was slower and of less magnitude than the chemoattractant-stimulated response [20]. A similar discrepancy in the kinetic response to FMLP and PMA activation was found when the respiratory burst of PMN was measured [11]. Moreover, actin polymerization was accompanied by an incorporation of actin into the cytoskeletal framework, and we found similar results using flow cytometric analysis of phalloidin binding and electrophoretic analysis of Triton-insoluble cytoskeleton when PMN were activated by FMLP. In the contrary, these two parameters seem to be unrelated when PMA was used as an inducer. This result is of importance because it has been previously described that actin polymerization does not necessarily induce actin incorporation into the cytoskeletal Triton X-l00-insoluble framework [19]. At the membrane level, we measured the expression of an antigen which is recognized by the monoclonal antibody MoFl 1. The expression of this antigen was described to be greatly increased by cytochalasine B treatment prior to FMLP activation [1,5]. In this study, we confirm that this antibody produces a very high activation ratio as compared with other extensively used antibodies such as those directed against the C3bi receptor. When PMN were activated by FMLP without cytochalasine B the response was in the same order of magnitude with FMLP and PMA. For MoFI 1 antigen expression as for actin polymerization, the response was very rapid (less than 1 mm) with FMLP and more progressive with PMA. The two types of activation led to the same extent of expression of MoFI I antigen, but the pathways of activation leading to that response must be different. When the dose effect of these activators were studied at the cytoplasmic and at the membrane level, it appeared that actin polymerization was induced by lower concentrations of PMA than MoF1 I antigen expression. Low
Fiow Cytometry
to
A 12
and PMN Activation
357
B
FMLP
PMA
3456
123456
kO
91-,
.m4
4-
43-0 2
5 111 30
II
S
Fig. 3. Polyacryiamlde gel electrophoresls analysis of the cytoskeletons of PMN activated by FMLP (1 0’ M) or PMA (20 ng/ml). A: Lane 1: Molecular weight markers. Lane 2: Nonactlvated PMN. Lanes 3-6: PMN activated by FMLP durIng 2, 5, 10, or 30 s, respectively. B: Lane 1: Molecular weight markers. Lane
1
tCTII
5 111 15 mm.
2: Nonactlvated PMN. Lanes 3 to 6: PMN activated by PMA during 2, 5, 10, or 15 mm, respectively. The arrows Indicate the actin band. The numbers on the left Indicate molecular weight markers.
B
A 4-
4
MoF
ii
3-
3 Phalloidin
2.
2
Phal
loi din
MoF V
V
io
io
io8
io
io
N/h
io_2
10_i
ii -
100
10
io2
ng/ml
FIg. 4. The effect of the dose of FMLP or PMA on M0F1 1 membrane antigen expression (.-.) and phalloidin binding in the cytoplasm (0-c) of PMN. Whole blood leukocytes were incubated with Increasing concentrations of FMLP for 1 mm (A) or of PMA for 15 mm (B) and were processed as described In FIgure 2. The values are plotted as the mean of at least 3 experIments. The bars represent the standard deviatIon.
concentrations of activator can initiate the contractile response without completing the expression of intracellular antigens on the surface of PMN. This raised a question of whether actin polymerization would necessarily precede MoF 11 antigen expression. When PMN were pretreated with cytochalasin B prior to FMLP activation, actin polymerization was inhibited by 60%. At the same time, MoF1 I antigen expression was increased fivefold. This demonstrated that actin polymerization is not a necessary event for the expression of MoF1 1 antigen. Moreover, this may suggest a granular localization of MoF1 1 anti-
gen as it is well known that cytochalasin B promotes degranulation of PMN. The measurement of actin polymerization and expression of an activation-dependent membrane antigen may bring useful complementary information in either pathologic of pharmacologic studies concerning PMN activation. The important activation ratio produced by the MoFl 1 antigen may be a useful marker to detect in vivo activated PMN in patient blood samples. Until now, this was not possible because the limited activation ratio of previously known antigens. Labeling of actin polymer-
358
Belloc
et al.
A
(6)j
4.
phil
function
9.42,
1987.
Fearon,
D.T.,
teractions 20
-
MoF
nol.
11
5.
B
10 (8)
(9)
1,234,
Fischer,
cytometry.
Wong,
mediate
Anal.
Quant.
W.W.
Complement
biological
responses.
Cytol.
Histol.
ligand-receptor Annu.
in-
Rev.
Immu-
1983. G.F.,
Majdic,
0.,
surface antigens
after
cocyte
II. White
Typing
A.J..
and
Knapp,
stimulation Cell
Ed.)
Oxford:
Foster,
D.W.,
N.
of human
Altered
expression
granulocytes.
Differentiation
Oxford
of
In Leu-
Antigens,
University
Press,
(Mcp. 698,
1987.
PHALLOIDIN
6. Goetzl,
5-
(6)
and
that
Michael,
RHODAMINE
by flow
E.J.,
partial
(ii)
man
characterization neutrophil
and Goldman,
of
membrane
receptors
for
D.W.
protein
chemotactic
Isolation
constituents
and of
formylmethionyl
hupep-
tides. Biochemistry 20,5717, 1981. Howard, T.H., and Oresajo. CO. The kinetics of chemotactic peptide-induced change in F-actin content, F-actin distribution. and the shape of neutrophils. J. Cell Biol. 101,1078, 1985. 8. King, C.H., Haimes Goralnik, C., Kleinhenz, P.J., Marino, J.A., Sedor, J.R., and Mahmoud, A.F. Monoclonal antibody characterization of a chymotrypsin-like molecule on neutrophil 7.
C
+___
PMA
FMLP
CRc
FMLP
FMLP
1’
1’
+
p
of Leukocyte
Biology
48:353-358
Flow Cytometric Study of the Activation Polymorphonuclear Cells Francis
Belloc,
Philippe
Laboratoire (G.F.),
Vincendeau, and Michel
d’Hematologie,
Pessac,
and
Genevieve Freyburger, R. Boisseau
H#{244}pital Cardiologique
Laboratoire
d’Immunologie,
(F.B.,
M.R.B.),
PD.,
H#{244}pitalPellegrin,
of
Patrice
Unite
Bordeaux
(1990)
Dumain,
8, INSERM (P.V.),
France
The activation of human polymorphonuciear cells (PMN) by the chemotactic peptide, N-formyl-methionyl-leucyl-phenylalanine (FMLP), or the protein kinase C activator, phorbol myristate acetate (PMA), was studied using flow cytometry. Two probes were used to evaluate PMN activation: 1) a monoclonal antibody (M0F1 1) directed against an antigen (Ag) expressed on the membrane of monocytes and of activated PMN; 2) rhodamine phalloidln was used at the cytoplasmic level to measure the F-actin content. The expression of M0F1 1 antigen was found to be 3 to 5 times greater on the membrane of PMN activated by either FMLP or PMN as compared with membrane expression of the same Ag on resting PMN. This increase was found to be dose dependent for the two activators. Kinetic studies showed that a maximum response was observed in 1 to 2 mm at 37#{176}C when FMLP was used, whilst a similar response required 10 mm when PMA was used. The same discrepancy with activators was observed when actin polymerization was measured by labellIng with rhodamine phalloidin. However, pretreatment of PMN with cytochalasin B inhibited actin polymerization whilst M0F1 1 antigen expression was increased, suggesting that the M0F1 1 antigen could be stored in granules of resting PMN. The study of actin polymerization and of M0F1 1 antigen expression, separately or in combination, could be a useful tool for the detection of activated PMN in biological samples. Key words:
actin
polymerization,
cytoskeleton,
INTRODUCTION When
exposed
to chemoattractant
agents,
polymor-
phonuclear leukocytes (PMN) respond rapidly with a highly coordinated series of responses. In response to N-formyl-methionyl-leucyl-phenylalanine (FMLP) for example, activation occurs as a consequence of the stimulus binding to the receptors on the cell membrane [6]. Immediate, secondary changes in PMN metabolism occur; these include the activation of phospholipase C, the release and phosphorylation of phosphatidyl inositol,the increase in free cytoplasmic Ca and the activation of protein kinase C [12]. These metabolic events lead to physiologic responses by the PMN such as the release of inflammatory mediators and chemotaxis. Among biochemical events contributing to chemotaxis are actin polymerization [7,16] and the exposure of antigens on the surface membrane [2,4,5,8,18]. Another stimulus that induces PMN response is phorbol myristate acetate (PMA), which acts by the activation of protein kinase C. To date, flow cytometry has been used as a tool to follow PMN activation using monoclonal antibodies to detect membrane antigen response [2. 18] and actin polymerization [7,16], cytosolic calcium [10], and oxida,
© 1990
Wiley-Liss,
Inc.
membrane
antigen,
monoclonal
antibody
tive products [17] amongst other parameters. These studies were performed on isolated PMN and focused on only one of these parameters. In this paper, we have investigated the responses of PMN to two stimuli (FMLP and PMA) using flow cytometry. Two different probes were used on the same samples: one was a monoclonal antibody (MoFI 1) directed against a membrane antigen whose surface expression on PMN is increased by FMLP activation [1,5], and the other was rhodamine-phalloidin, which binds to cytoplasmic F actin and is frequently used to quantitate the rate of actin polymerization [16,19]. The kinetic and dose-dependent responses of membrane antigen expression and cytoplasmic actin polymerization were studied and compared. It appeared that the increase in F actin paralleled the increase in MoF1 I antigen expression suggesting that they are concomitant responses following
Received Reprint
March requests:
Cardiologique.
10.
1989;
Francis avenue
accepted Belloc,
Magellan.
March Laboratoire 33604
19,
1990.
d’Hematologie. Pessac-France.
H#{244}pital
354
Belioc
et al.
PMN activation. However the partial inhibition of F actin increase by cytochalasine B which was accompanied by an increase of MoF1 1 antigen expression suggested that the two responses were not completely interdependent functions. These results prove the usefulness of MoF1 1 and fluorescent phalloidin for flow cytometric analysis of the phenotype of activated PMN.
MATERIALS
AND METHODS
Antibodies 0KM 1 (directed against the C3bi receptor) was purchased from Ortho Diagnostic (France). Unrelated IgG (MsIgG2a and MsIgG2b) as well as fluoresceinated goat anti mouse IgG (GAM-FITC) were obtained from Coultronics (Margency; France). The monoclonal antibody MoF1 1 was obtained by immunizing BALB/c mice with human normal monocytes. The murine spleen cells were fused with murine myeloma cells X63 Ag8653. Positive clones were selected by indirect immunofluorescence on the basis of their ability to detect monocyte membrane antigens (P. Vincendeau, F. Belloc, et al.;manuscript in preparation). MoFI 1 is an IgG2a and has been submitted to the third and fourth International Workshop of Immunology for clusterization.
Cell Preparation
for Flow Cytometry
Blood was obtained by venipuncture with EDTA as anticoagulant. Whole blood leukocytes were enriched by gelatine sedimentation of red cells (1 volume of Plasmion to 4 volumes of blood) for 20 mm at 20#{176}C. The leukocytes were then pelleted by centrifugation, the remaining red cells were lysed in 0.85% ammonium chloride for 7 mm at 20#{176}C, and the leukocytes were washed and resuspended in RPM! 1640 at a concentration of 4.106 cells/mI and incubated at 37#{176}C for 5 mm. PMA or FMLP was then added to the leukocyte suspension at the indicated concentrations. The suspension was incubated at 37#{176}C to allow PMN activation, and samples were then processed as described below after different times of incubation.
lmmunofluorescence Leukocytes (106) were fixed with 1% paraformaldehyde (PFA) and 0.05% glutaraldehyde (final concentration) for 10 mm at 20#{176}C, washed with PBS, and incubated for 30 mm at 20#{176}C with 0.2 ml of PBS containing 10 pg of MoF1 1 or OKM1 IgG. The cells were then washed and incubated with 0.2 ml of a 1/40 dilution of GAM-FITC for 30 mm at 20#{176}C, washed again, and resuspended in PBS for flow cytometry. Non-specific labeling was measured by incubating 106 cells with unrelated IgG.
Rhodamine-Phalloidin
Binding
Leukocytes (5 x l0) were fixed, permeabilized, and stained at the same time by 2% PFA, 30 g/ml lysophosphatidyl choline, 0.1 xm of rhodamine-phalloidine (Molecular Probes, Eugene, OR). After 30 mm at 37#{176}C, the cells were washed in PBS and resuspended at 106 cells! ml in PBS for flow cytometry.
PMN Cytoskeletons Gel Electrophoresis
Analysis (PAGE)
by Polyacrylamide
PMN purification was required for cytoskeleton analysis. This was obtained by a two-step preparation procedure. Red blood cells from EDTA anticoagulated blood were sedimented on a 9.6% metrizoate and 5% Ficoll solution [3]; the plateletand leukocyte-rich plasma was then centrifuged on a discontinuous Percoll gradient [14]. Then 6 x 106 purified PMN in 500 l were activated with l0M FMLP or with 20 ng/ml PMA. At different times, 500 iii of lysing buffer [13] (20 mM EGTA, 20 mM imidazole HCI, 80 mM KC1, 2% Triton X-l00, pH 7.15) were added; the lysis was performed at 4#{176}C for 5 mm and centrifuged at 12 000g for 4 mm. The pellet was rinsed with washing buffer (lysing buffer without Triton X- 100) and solubilized in 150 pA of 10% SDS, 4 M urea, 20% glycerol, 2% 3-mercaptoethanol, pH 8). Proteins were then analyzed by SDS polyacrylamide gel electrophoresis (SDS-PAGE) on a 7.5% acrylamide slab gel using the Laemmli [9] discontinuous buffer system.
Flow Cytometry Samples prepared for flow cytometry were analyzed using an ATC 3000 (ODAM, Wissembourg, France) cell sorter equipped with a 2025 Spectraphysics argon ion laser. For immunofluorescence labeling, the laser was tuned on the 488 nm argon line and fluorescence was collected through a 515 nm long pass filter. For rhodamine-phalloidine experiments, the laser was tuned on the 514 nm argon line and fluorescence was collected through a 600 nm long pass filter. The fluorescence of PMN was analyzed by gating fluorescence of the PMN on the basis of their wide angle light scatter. Mean fluorescence channel was calculated by the ASPECT 3000 computer and the results were expressed as the activation ratio: mean fluorescence channel of activated PMN/ mean fluorescence channel of non-activated PMN.
RESULTS The blood
scatter antigen
fluorescence leukocyte
of PMN suspensions
can as
be measured a result
properties (Fig. 1). The expression (as measured by immunofluorescence)
of
in whole their
light
of MoFI 1 and the
Flow Cytometry
and PMN Activation
355
-J 4
FALS N
C
B
IL
I
of blood leukocytes labelled M0F1 1 antIbody. The fluoresPMN on a double scatter cythe x axis and the wide angle
light scatter on they axis. The fluorescence histograms of PMN after IndIrect labeling with M0F11 (B) or direct labeling with rhodamine-phalloidin (C) are shown for non actIvated (shadowed area) and iO M FMLP activated leukocytes.
F-actin content (as measured by rhodamine-phalloidin binding) were increased by activation of PMN. The increase in antigen expression was measured on PMN which were incubated for 5 mm with iM FMLP after treatment by cytochalasine B (5 xg/ml). Under these conditions, the activation ratio (AR, see “Materials and Methods”) was found to be 7 times higher for the MoF11 antigen (AR = 13.06 ± 2.26; n = 4) than for the OKM1 antigen (AR 1.89 ± 0.17; n = 4). This difference was essentially due to the relatively high level of C3bi receptor expression on the membrane of resting
PMN as compared with the basal level of MoFl 1 antigen expression. The kinetics of MoF 11 antigen expression and F-actin content during activation of PMN with FMLP or PMA are shown in Figure 2. Following activation by FMLP, the two parameters were rapidly increased and reached maximum levels after 1 mm for F-actin content and after 2 mm for MoF1 1 antigen expression (Fig. 2A). When activation was performed using PMA, the response kinetics were different. MoF1 1 antigen membrane expression increased progressively for 10 to 15 mm while the
FIg. 1. Flow cytometry analysis with rhodamine phalloldin and/or canoe analysis was gated on the togram (A). The axial scatter is on
Belloc
356
et al.
IMLI’
PMA
14
6
20
ny/mI
6 MaF
MoE
11
11
C
C
13
A
14
Phailoidin
:3
:2, C
C
1
2
3
Activation
4 time
5
mm.
5 Activation
10
15
zU
m!r.
time
Fig. 2. Kinetics of activation by FMLP and PMA on M0F11 membrane antigen expression and rhodamine-phalloidin binding in the cytoplasm of PMN. Whole blood leukocytes were activated by either iO- M FMLP (A) or 20 ng/mi PMA (B) for the Indicated times, fixed, and labeled using either M0F11 immuno fluorescence (._..) or rhodamine phailoidin (0-0). The fluorescence of the PMN was then analyzed by flow cytometry and the results are expressed as the activation ratio in PMN as a function of activation time. The data on this figure are from one representative experiment.
cytoplasmic F-actin increase was delayed and appeared only after 10 mm of activation (Fig. 2B). When isolated PMN were activated by FMLP, rapid actin incorporation into the cytoskeletal framework was observed using PAGE analysis (Fig. 3A). When the PMN were activated by PMA, the incorporation of actin into the cytoskeleton lasted during 5 mm and then decreased (Fig. 3B) while F-actin content continued to increase (Fig. 2B). The dose-response curves of MoFI 1 antigen expression and F-actin content after FMLP or PMA activation are shown in Figure 4. There was a dose-dependent increase in the F-actin content after activation with FMLP (10_to to l0 M FMLP). A similar pattern was observed with MoFI I Ag expression (Fig. 4A). The Factin increased and reached a plateau at a concentration of 2 ng/ml of PMA. Higher concentrations (one log) of PMA were required to induce a similar activation ratio with membrane Ag expression (Fig. 4B). Figure 5 demonstrates that 20 ng of PMA and l0 M FMLP induce similar activation ratios for antigen expression and actin polymerization. However, when leukocytes were pretreated with cytochalasmn B and then activated with FMLP, the MoFl 1 Ag expression was significantly increased fivefold whilst the actin polymerization response was significantly inhibited in PMN.
DISCUSSION Treatment of PMN by the chemoattractant FMLP or by protein kinase C activation (PMA) induces a motile response in PMN with extensive changes in the cytoplas-
mic contractile apparatus and in the membrane proteins. Actin polymerization is known to occur during PMN activation [12]. This polymerization was measured either by phalloidin binding and flow cytometry, or by Triton X-lOO cytoskeleton extraction and SDS-PAGE. These measurements were performed on isolated PMN [7,16]. In the present study, activation and measurements were performed on whole blood leukocyte suspensions contaminated by platelets but with similar results to previous reports. The PMN specific responses were assessed by electronic gating of the fluorescence analysis on the scattergram. The cell preparation was greatly simplified and allowed the use of flow cytometric measurement of actin polymerization by means of a routine assay. Rapid actin polymerization was found after PMN activation by FMLP, whilst, in response to PMA treatment, the F actin content increased only after 10 mm. This was in agreement with previous results which found that PMA induced an actin polymerization response that was slower and of less magnitude than the chemoattractant-stimulated response [20]. A similar discrepancy in the kinetic response to FMLP and PMA activation was found when the respiratory burst of PMN was measured [11]. Moreover, actin polymerization was accompanied by an incorporation of actin into the cytoskeletal framework, and we found similar results using flow cytometric analysis of phalloidin binding and electrophoretic analysis of Triton-insoluble cytoskeleton when PMN were activated by FMLP. In the contrary, these two parameters seem to be unrelated when PMA was used as an inducer. This result is of importance because it has been previously described that actin polymerization does not necessarily induce actin incorporation into the cytoskeletal Triton X-l00-insoluble framework [19]. At the membrane level, we measured the expression of an antigen which is recognized by the monoclonal antibody MoFl 1. The expression of this antigen was described to be greatly increased by cytochalasine B treatment prior to FMLP activation [1,5]. In this study, we confirm that this antibody produces a very high activation ratio as compared with other extensively used antibodies such as those directed against the C3bi receptor. When PMN were activated by FMLP without cytochalasine B the response was in the same order of magnitude with FMLP and PMA. For MoFI 1 antigen expression as for actin polymerization, the response was very rapid (less than 1 mm) with FMLP and more progressive with PMA. The two types of activation led to the same extent of expression of MoFI I antigen, but the pathways of activation leading to that response must be different. When the dose effect of these activators were studied at the cytoplasmic and at the membrane level, it appeared that actin polymerization was induced by lower concentrations of PMA than MoF1 I antigen expression. Low
Fiow Cytometry
to
A 12
and PMN Activation
357
B
FMLP
PMA
3456
123456
kO
91-,
.m4
4-
43-0 2
5 111 30
II
S
Fig. 3. Polyacryiamlde gel electrophoresls analysis of the cytoskeletons of PMN activated by FMLP (1 0’ M) or PMA (20 ng/ml). A: Lane 1: Molecular weight markers. Lane 2: Nonactlvated PMN. Lanes 3-6: PMN activated by FMLP durIng 2, 5, 10, or 30 s, respectively. B: Lane 1: Molecular weight markers. Lane
1
tCTII
5 111 15 mm.
2: Nonactlvated PMN. Lanes 3 to 6: PMN activated by PMA during 2, 5, 10, or 15 mm, respectively. The arrows Indicate the actin band. The numbers on the left Indicate molecular weight markers.
B
A 4-
4
MoF
ii
3-
3 Phalloidin
2.
2
Phal
loi din
MoF V
V
io
io
io8
io
io
N/h
io_2
10_i
ii -
100
10
io2
ng/ml
FIg. 4. The effect of the dose of FMLP or PMA on M0F1 1 membrane antigen expression (.-.) and phalloidin binding in the cytoplasm (0-c) of PMN. Whole blood leukocytes were incubated with Increasing concentrations of FMLP for 1 mm (A) or of PMA for 15 mm (B) and were processed as described In FIgure 2. The values are plotted as the mean of at least 3 experIments. The bars represent the standard deviatIon.
concentrations of activator can initiate the contractile response without completing the expression of intracellular antigens on the surface of PMN. This raised a question of whether actin polymerization would necessarily precede MoF 11 antigen expression. When PMN were pretreated with cytochalasin B prior to FMLP activation, actin polymerization was inhibited by 60%. At the same time, MoF1 I antigen expression was increased fivefold. This demonstrated that actin polymerization is not a necessary event for the expression of MoF1 1 antigen. Moreover, this may suggest a granular localization of MoF1 1 anti-
gen as it is well known that cytochalasin B promotes degranulation of PMN. The measurement of actin polymerization and expression of an activation-dependent membrane antigen may bring useful complementary information in either pathologic of pharmacologic studies concerning PMN activation. The important activation ratio produced by the MoFl 1 antigen may be a useful marker to detect in vivo activated PMN in patient blood samples. Until now, this was not possible because the limited activation ratio of previously known antigens. Labeling of actin polymer-
358
Belloc
et al.
A
(6)j
4.
phil
function
9.42,
1987.
Fearon,
D.T.,
teractions 20
-
MoF
nol.
11
5.
B
10 (8)
(9)
1,234,
Fischer,
cytometry.
Wong,
mediate
Anal.
Quant.
W.W.
Complement
biological
responses.
Cytol.
Histol.
ligand-receptor Annu.
in-
Rev.
Immu-
1983. G.F.,
Majdic,
0.,
surface antigens
after
cocyte
II. White
Typing
A.J..
and
Knapp,
stimulation Cell
Ed.)
Oxford:
Foster,
D.W.,
N.
of human
Altered
expression
granulocytes.
Differentiation
Oxford
of
In Leu-
Antigens,
University
Press,
(Mcp. 698,
1987.
PHALLOIDIN
6. Goetzl,
5-
(6)
and
that
Michael,
RHODAMINE
by flow
E.J.,
partial
(ii)
man
characterization neutrophil
and Goldman,
of
membrane
receptors
for
D.W.
protein
chemotactic
Isolation
constituents
and of
formylmethionyl
hupep-
tides. Biochemistry 20,5717, 1981. Howard, T.H., and Oresajo. CO. The kinetics of chemotactic peptide-induced change in F-actin content, F-actin distribution. and the shape of neutrophils. J. Cell Biol. 101,1078, 1985. 8. King, C.H., Haimes Goralnik, C., Kleinhenz, P.J., Marino, J.A., Sedor, J.R., and Mahmoud, A.F. Monoclonal antibody characterization of a chymotrypsin-like molecule on neutrophil 7.
C
+___
PMA
FMLP
CRc
FMLP
FMLP
1’
1’
+
p
