Cell Motility and the Cytoskeleton 16:80-87 (1990)
Kinetic Analysis of Chemotactic Peptide-Induced Act in Polymerization in Neutrophils Danher Wang, Keith Berry, and Thomas H. Howard Departments of Cell Biology (D.W., T.H.H.), Microbiology (K.B.), and Pediatrics (T.H.H.), University of Alabama Birmingham School of Medicine, Birmingham Definition of the kinetics of ligand-activated actin polymerization in the neutrophi1 is important for ultimately understanding the mechanisms utilized for regulation of actin polymerization in this non-muscle cell. To better define the kinetics of formyl peptide (fMLP) -induced actin polymerization in neutrophils we determined F-actin content at 5 second intervals after activation of human neutrophils with a range (10-"-10-9 M) of fMLP concentrations. The state of actin polymerization was monitored by quantifying F-actin content with NBD phallacidin binding in both flow cytometric and extraction assays. Results demonstrate three successive kinetic periods of fMLP-induced actin polymerization in neutrophils, a lag period, a 5 second period when rate of polymerization is maximal, and a period of declining rate of actin polymerization as F-actin content approaches a maximum. The duration of the lag period, the maximum rate of polymerization, and the maximum extent of polymerization all depend upon the fMLP concentration. The lag period varies from 0 to 12 seconds and is followed in 5-10 seconds by a 5 second burst of actin polymerization when the rate is as great as 9% increase in F-actin content per second. After the 5 second burst of polymerization, the rate of polymerization rapidly declines. The study defines three distinct kinetic periods of fMLP-induced actin polymerization during which important rate-limiting biochemical events occur. The mechanistic and motile implications of kinetic periods are discussed.
Key words: cytoskeleton, chemotaxis, polymerization, motility, nucleation
INTRODUCTION
An understanding of the molecular events during actin polymerization in vitro evolved from detailed kinetic studies of actin polymerization under varied conditions of ionic strength, nucleotide concentration, monomer concentration, and numbers/type of filament ends [Korn, 1982; Korn et al., 19871. Definition of molecular events of in situ actin polymerization through study of the kinetics of ligand-induced actin polymerization in non-muscle cells poses a more complex problem because of limitations inherent to the experimental systems. For example, free monomer concentration, filament end availability, and alterations in ligand-receptor interactions that occur during ligand-induced changes in the state of actin polymerization are difficult to control. 0 1990 Wiley-Liss, Inc.
However, despite the limitations inherent to the study of ligand-induced actin polymerization in non-muscle cells, several laboratories have made contributions that yield insight into the biochemical mechanism(s) for chemotactic peptide-induced actin polymerization in the human neutrophil. These contributions include 1) establishing methods that allow quantification of dynamic changes in the F-actin content of non-muscle cells [Howard and Meyer, 1984; Wallace et al., 1984; Howard and Oresajo, 1985a,b]; 2) identification of the barbed-end of actin Received October 13, 1989; accepted January 16, 1990. Address reprint requests to Thomas H. Howard, M.D., UAB Pediatric HematologyiOncology, The Children's Hospital, 1600 7th Avenue South, Birmingham, AL 35233.
Actin Polymerization in Neutrophils
filaments as the predominant end for filament growth [Wallace et al., 1984; Howard and Oresajo, 1985a; Fecheimer and Zigmond, 19831; 3) exclusion of increase in cytosolic Ca+ concentration as a requirement for polymerization [Howard and Wang, 1987; Bergtsson et al., 1986; Sklar et al., 1985; Sha'afi et al., 19861; 4) demonstration that a G-protein is involved in transmembrane signaling events associated with actin polymerization [Bergtsson et al., 19861; and, 5 ) documentation of increased amounts of barbed-end nucleating activity in ligand-activated neutrophils [Carson et al., 19871. As was the case for actin polymerization in vitro, a detailed kinetic analysis of chemotactic peptide-induced actin polymerization will advance initial efforts to understand the mechanism(s) of actin polymerization in neutrophils. Such studies could define critical times for the occurrence of biochemical events that limit the extent and rate of polymerization and yield insight into the mechanism for regulating polymerization. For example, if as reported [Howard and Oresajo, 1985a; Wallace et al., 19841 the maximum extent of chemotactic peptideinduced polymerization depends upon peptide concentration, it is essential to know whether the rate of polymerization also depends upon peptide concentration. Demonstrating that the maximum extent of actin polymerization depends upon peptide concentration and that the rate of polymerization is concentration independent would suggest that peptide regulates F-actin content of neutrophils by releasing a concentration-dependent amount of monomeric actin as previously suggested [Howard and Oresajo, 1985aI. Alternatively, finding that both rate and maximum extent of polymerization depend upon peptide concentration would suggest that actin polymerization is regulated by the creation of or unmasking of actin nuclei as previously suggested [Carson et al., 1987; Chapponier and Howard, 19861. This paper reports the kinetics of actin polymerization in neutrophils activated with a wide range of concentrations ( 10-'1-10-9 M) of the chemotactic peptide, formyl-methionyl-leucyl-phenylalanine(fMLP), and determines F-actin content at 5 second intervals by both NBD phallacidin extraction [Howard and Oresajo, 1985b] and flow cytometry [Howard and Oresajo, 1985al. The studies define three important kinetic periods during fMLP-induced actin polymerization in neutrophils: 1) a brief lag period, which precedes fMLPinduced actin polymerization and the duration of which depends upon fMLP concentration; 2) a period of maximum rate of actin polymerization, which follows the lag period of 5-10 seconds; and 3) a period of progressive decline in rate of actin polymerization as F-actin content approaches a maximum value. Both the maximum rate of actin polymerization in the 5-10 seconds after the lag period and the mean rate of actin polymerization (deter-
81
mined over the entire period when F-actin content increases) depend upon the activating concentration of fMLP.
+
MATERIALS AND METHODS Materials Materials were obtained from the following suppliers: formyl-met-leu-phe (Sigma, St. Louis, MO); Hypaque (Sterling Laboratories, Rensselaer, NY); NBD phallacidin (Molecular Probes, Junction City, OR): DMSO (Fisher Scientific); dextran in saline was Macrodex (Pharmacia, NJ). Preparation of Neutrophils After withdrawal of blood from human volunteers in EDTA anticoagulant human neutrophils were purified on ficoll-Hypaque gradients after dextran sedimentation as previously described [Howard and Oresajo, 1985a; Boyum, 19631. The resulting cells are 95-97% neutrophils, 2-3% eosinophils, and 5 1% mononuclear cells. Cells were maintained at 25°C throughout preparation and were resuspended in HankdHEPES balanced salt solution (H/HBSSG+ ) with composition as previously described [Howard and Meyer, 19841. Cells were studied within 5 hours of blood withdrawal. +
Quantification of F-Actin Content by NBD phallacidin Extraction Neutrophils (2 X lo6 celldassay) were exposed to DMSO (50.000 I/vol%) or the indicated concentrations of fMLP (10-1t-10-9M) for the indicated periods of time at 25°C. Cells were fixed with 3.7% formalin, permeabilized, and stained with lysophosphatidylcholine (80 pg/ml final concentration) and NBD phallacidin [Barak and Yocum, 19811 (3.3 X lo-' M, final concentration). Cells were then pelleted, overlaid with absolute methanol, and extracted for 1 hour. The methanol extract was removed, and the relative fluorescence intensity (RFI) of the sample was determined by spectrofluorometry (excitation 465 nm, emission 530 nm) on a Perkin-Elmer LS-3 spectrofluorometer, all as previously described (Howard and Oresajo, 1985 b). F-actin content is expressed as RFI of test sample/RFI of DMSO control at 0 time. Quantification of F-Actin Content by Flow Cytometry Neutrophils ( 1 X lo6 cells/assay) were exposed to DMSO (0.0001 ~ 0 1 % )or fMLP at the indicated final concentrations ( 10-1'-10-9 M) for the indicated times. Cells were fixed and stained as described above and analyzed with a FACStar (Becton-Dickinson, Mountain View, CA) flow cytometer at 488 nm excitation with a
82
Wang et al.
70 mW Argon laser and long pass (>530 nm) emission filter. Cells were gated for neutrophils to exclude debris and minor mononuclear cell contamination. Cells gated and analyzed for F-actin content were 98-100% neutrophils and 0-2% eosinophils based upon sorted samples of cells. Data were collected for 5,000 cells on a linear fluorescence scale with gain settings of 1.0 for forward scatter, 1.0 for side scatter with the fluorescence PMT settings of voltage 465, gain 8.0. The software/statistical programs of FACStar yield a distribution histogram of fluorescence, mean and mode fluorescence channel number, and variance of fluorescence in the 5,000 cell population. F-actin content is proportional to fluorescence and is expressed as either mean or mode fluorescence channel number as previously described [Howard and Meyer, 1984; Howard and Oresajo, 1985al. and as indicated in text. To determine rates of polymerization F-actin content is expressed as F-actin content in test sample divided by F-actin content of control (DMSO) cells at zero time. All rates were determined after excluding any lag time. To assure control for fixation time in determining rates of polymerization, the zero second time point reflects the immediate and simultaneous addition of fMLP and fixative. RESULTS The Time Course of fMLP-Induced Actin Polymerization Quantified by NBD phallacidin Extraction
In earlier studies [Howard and Oresajo, 1985a1, we suggested that the rate of actin polymerization in N L P activated neutrophils is independent of fMLP concentration, while the maximum extent of actin polymerization depends upon fMLP concentration. In the studies presented here, we activated neutrophils with varied fMLP concentrations at 25°C and utilized both the extraction assay [Howard and Oresajo, 1985bl and the flow cytometric assay with improved ( 5 second interval) time resolution to define the relationship between time course of actin polymerization and fMLP concentration. As shown in a representative experiment (see Fig. l), when ficollHypaque purified neutrophils are exposed to lo-' '-lop9 M fMLP and F-actin content is determined at 5 second intervals by NBD phallacidin extraction, a transient lag period during which F-actin content increases minimally, if at all, is observed (note N L P concentrations of
Kinetic Analysis of Chemotactic Peptide-Induced Act in Polymerization in Neutrophils Danher Wang, Keith Berry, and Thomas H. Howard Departments of Cell Biology (D.W., T.H.H.), Microbiology (K.B.), and Pediatrics (T.H.H.), University of Alabama Birmingham School of Medicine, Birmingham Definition of the kinetics of ligand-activated actin polymerization in the neutrophi1 is important for ultimately understanding the mechanisms utilized for regulation of actin polymerization in this non-muscle cell. To better define the kinetics of formyl peptide (fMLP) -induced actin polymerization in neutrophils we determined F-actin content at 5 second intervals after activation of human neutrophils with a range (10-"-10-9 M) of fMLP concentrations. The state of actin polymerization was monitored by quantifying F-actin content with NBD phallacidin binding in both flow cytometric and extraction assays. Results demonstrate three successive kinetic periods of fMLP-induced actin polymerization in neutrophils, a lag period, a 5 second period when rate of polymerization is maximal, and a period of declining rate of actin polymerization as F-actin content approaches a maximum. The duration of the lag period, the maximum rate of polymerization, and the maximum extent of polymerization all depend upon the fMLP concentration. The lag period varies from 0 to 12 seconds and is followed in 5-10 seconds by a 5 second burst of actin polymerization when the rate is as great as 9% increase in F-actin content per second. After the 5 second burst of polymerization, the rate of polymerization rapidly declines. The study defines three distinct kinetic periods of fMLP-induced actin polymerization during which important rate-limiting biochemical events occur. The mechanistic and motile implications of kinetic periods are discussed.
Key words: cytoskeleton, chemotaxis, polymerization, motility, nucleation
INTRODUCTION
An understanding of the molecular events during actin polymerization in vitro evolved from detailed kinetic studies of actin polymerization under varied conditions of ionic strength, nucleotide concentration, monomer concentration, and numbers/type of filament ends [Korn, 1982; Korn et al., 19871. Definition of molecular events of in situ actin polymerization through study of the kinetics of ligand-induced actin polymerization in non-muscle cells poses a more complex problem because of limitations inherent to the experimental systems. For example, free monomer concentration, filament end availability, and alterations in ligand-receptor interactions that occur during ligand-induced changes in the state of actin polymerization are difficult to control. 0 1990 Wiley-Liss, Inc.
However, despite the limitations inherent to the study of ligand-induced actin polymerization in non-muscle cells, several laboratories have made contributions that yield insight into the biochemical mechanism(s) for chemotactic peptide-induced actin polymerization in the human neutrophil. These contributions include 1) establishing methods that allow quantification of dynamic changes in the F-actin content of non-muscle cells [Howard and Meyer, 1984; Wallace et al., 1984; Howard and Oresajo, 1985a,b]; 2) identification of the barbed-end of actin Received October 13, 1989; accepted January 16, 1990. Address reprint requests to Thomas H. Howard, M.D., UAB Pediatric HematologyiOncology, The Children's Hospital, 1600 7th Avenue South, Birmingham, AL 35233.
Actin Polymerization in Neutrophils
filaments as the predominant end for filament growth [Wallace et al., 1984; Howard and Oresajo, 1985a; Fecheimer and Zigmond, 19831; 3) exclusion of increase in cytosolic Ca+ concentration as a requirement for polymerization [Howard and Wang, 1987; Bergtsson et al., 1986; Sklar et al., 1985; Sha'afi et al., 19861; 4) demonstration that a G-protein is involved in transmembrane signaling events associated with actin polymerization [Bergtsson et al., 19861; and, 5 ) documentation of increased amounts of barbed-end nucleating activity in ligand-activated neutrophils [Carson et al., 19871. As was the case for actin polymerization in vitro, a detailed kinetic analysis of chemotactic peptide-induced actin polymerization will advance initial efforts to understand the mechanism(s) of actin polymerization in neutrophils. Such studies could define critical times for the occurrence of biochemical events that limit the extent and rate of polymerization and yield insight into the mechanism for regulating polymerization. For example, if as reported [Howard and Oresajo, 1985a; Wallace et al., 19841 the maximum extent of chemotactic peptideinduced polymerization depends upon peptide concentration, it is essential to know whether the rate of polymerization also depends upon peptide concentration. Demonstrating that the maximum extent of actin polymerization depends upon peptide concentration and that the rate of polymerization is concentration independent would suggest that peptide regulates F-actin content of neutrophils by releasing a concentration-dependent amount of monomeric actin as previously suggested [Howard and Oresajo, 1985aI. Alternatively, finding that both rate and maximum extent of polymerization depend upon peptide concentration would suggest that actin polymerization is regulated by the creation of or unmasking of actin nuclei as previously suggested [Carson et al., 1987; Chapponier and Howard, 19861. This paper reports the kinetics of actin polymerization in neutrophils activated with a wide range of concentrations ( 10-'1-10-9 M) of the chemotactic peptide, formyl-methionyl-leucyl-phenylalanine(fMLP), and determines F-actin content at 5 second intervals by both NBD phallacidin extraction [Howard and Oresajo, 1985b] and flow cytometry [Howard and Oresajo, 1985al. The studies define three important kinetic periods during fMLP-induced actin polymerization in neutrophils: 1) a brief lag period, which precedes fMLPinduced actin polymerization and the duration of which depends upon fMLP concentration; 2) a period of maximum rate of actin polymerization, which follows the lag period of 5-10 seconds; and 3) a period of progressive decline in rate of actin polymerization as F-actin content approaches a maximum value. Both the maximum rate of actin polymerization in the 5-10 seconds after the lag period and the mean rate of actin polymerization (deter-
81
mined over the entire period when F-actin content increases) depend upon the activating concentration of fMLP.
+
MATERIALS AND METHODS Materials Materials were obtained from the following suppliers: formyl-met-leu-phe (Sigma, St. Louis, MO); Hypaque (Sterling Laboratories, Rensselaer, NY); NBD phallacidin (Molecular Probes, Junction City, OR): DMSO (Fisher Scientific); dextran in saline was Macrodex (Pharmacia, NJ). Preparation of Neutrophils After withdrawal of blood from human volunteers in EDTA anticoagulant human neutrophils were purified on ficoll-Hypaque gradients after dextran sedimentation as previously described [Howard and Oresajo, 1985a; Boyum, 19631. The resulting cells are 95-97% neutrophils, 2-3% eosinophils, and 5 1% mononuclear cells. Cells were maintained at 25°C throughout preparation and were resuspended in HankdHEPES balanced salt solution (H/HBSSG+ ) with composition as previously described [Howard and Meyer, 19841. Cells were studied within 5 hours of blood withdrawal. +
Quantification of F-Actin Content by NBD phallacidin Extraction Neutrophils (2 X lo6 celldassay) were exposed to DMSO (50.000 I/vol%) or the indicated concentrations of fMLP (10-1t-10-9M) for the indicated periods of time at 25°C. Cells were fixed with 3.7% formalin, permeabilized, and stained with lysophosphatidylcholine (80 pg/ml final concentration) and NBD phallacidin [Barak and Yocum, 19811 (3.3 X lo-' M, final concentration). Cells were then pelleted, overlaid with absolute methanol, and extracted for 1 hour. The methanol extract was removed, and the relative fluorescence intensity (RFI) of the sample was determined by spectrofluorometry (excitation 465 nm, emission 530 nm) on a Perkin-Elmer LS-3 spectrofluorometer, all as previously described (Howard and Oresajo, 1985 b). F-actin content is expressed as RFI of test sample/RFI of DMSO control at 0 time. Quantification of F-Actin Content by Flow Cytometry Neutrophils ( 1 X lo6 cells/assay) were exposed to DMSO (0.0001 ~ 0 1 % )or fMLP at the indicated final concentrations ( 10-1'-10-9 M) for the indicated times. Cells were fixed and stained as described above and analyzed with a FACStar (Becton-Dickinson, Mountain View, CA) flow cytometer at 488 nm excitation with a
82
Wang et al.
70 mW Argon laser and long pass (>530 nm) emission filter. Cells were gated for neutrophils to exclude debris and minor mononuclear cell contamination. Cells gated and analyzed for F-actin content were 98-100% neutrophils and 0-2% eosinophils based upon sorted samples of cells. Data were collected for 5,000 cells on a linear fluorescence scale with gain settings of 1.0 for forward scatter, 1.0 for side scatter with the fluorescence PMT settings of voltage 465, gain 8.0. The software/statistical programs of FACStar yield a distribution histogram of fluorescence, mean and mode fluorescence channel number, and variance of fluorescence in the 5,000 cell population. F-actin content is proportional to fluorescence and is expressed as either mean or mode fluorescence channel number as previously described [Howard and Meyer, 1984; Howard and Oresajo, 1985al. and as indicated in text. To determine rates of polymerization F-actin content is expressed as F-actin content in test sample divided by F-actin content of control (DMSO) cells at zero time. All rates were determined after excluding any lag time. To assure control for fixation time in determining rates of polymerization, the zero second time point reflects the immediate and simultaneous addition of fMLP and fixative. RESULTS The Time Course of fMLP-Induced Actin Polymerization Quantified by NBD phallacidin Extraction
In earlier studies [Howard and Oresajo, 1985a1, we suggested that the rate of actin polymerization in N L P activated neutrophils is independent of fMLP concentration, while the maximum extent of actin polymerization depends upon fMLP concentration. In the studies presented here, we activated neutrophils with varied fMLP concentrations at 25°C and utilized both the extraction assay [Howard and Oresajo, 1985bl and the flow cytometric assay with improved ( 5 second interval) time resolution to define the relationship between time course of actin polymerization and fMLP concentration. As shown in a representative experiment (see Fig. l), when ficollHypaque purified neutrophils are exposed to lo-' '-lop9 M fMLP and F-actin content is determined at 5 second intervals by NBD phallacidin extraction, a transient lag period during which F-actin content increases minimally, if at all, is observed (note N L P concentrations of
