Threshold and Graded Response Behavior in Human Neutrophils

Jul 29, 1998 - Anna Waller, Karyn L. Sutton, Tamara L. Kinzer-Ursem, Afaf Absood, John R. Traynor, Jennifer J. Linderman, and Geneva M. Omann...
0 downloads 0 Views 289KB Size
11534

Biochemistry 1998, 37, 11534-11543

Threshold and Graded Response Behavior in Human Neutrophils: Effect of Varying G-Protein or Ligand Concentrations† Todd A. Riccobene,‡ Anna Waller,‡ Julie F. Hoffman,‡,§ Jennifer J. Linderman,‡ and Geneva M. Omann*,| Department of Chemical Engineering, UniVersity of Michigan, Ann Arbor, Michigan 48109, and Departments of Biological Chemistry and Surgery, UniVersity of Michigan Medical School and Veteran’s Administration Medical Center, Ann Arbor, Michigan 48105 ReceiVed December 22, 1997; ReVised Manuscript ReceiVed April 13, 1998

ABSTRACT: Observing the qualitative characteristics of response behavior as key variables in the signal transduction cascade are changed can provide insight into the fundamental roles of these interactions in producing cellular responses. Using flow cytometric assays and pertussis toxin (PT) treatment of human neutrophils, we have shown that actin polymerization stimulated with the chemoattractants N-formylMet-Leu-Phe, leukotriene B4, and interleukin-8 exhibits threshold behavior in terms of G-protein number. Partial PT treatment resulted in both responding and nonresponding populations of cells upon stimulation. As PT treatment was increased, the responding population of cells continued to respond maximally, while the number of cells responding decreased. We also showed that N-formyl peptide-stimulated oxidant production exhibits threshold behavior in terms of G-protein number, and the threshold for oxidant production is significantly greater than that for actin polymerization. The threshold behavior observed with PT treatment contrasted with the graded response behavior seen when cells were stimulated with different doses of ligand. For actin polymerization, only one population of cells was observed at submaximal ligand concentrations, and as ligand concentration was decreased the whole population responded submaximally. For oxidant production, as ligand concentration was decreased there were two populations of cells, but the responding cells responded submaximally. A mathematical model incorporating receptor/ligand binding and G-protein activation was developed to account for these differences in response behavior. Our results predict that an early signal transduction event in addition to, and not initiated by G-protein activation, is necessary to account for actin polymerization and oxidant production in neutrophils.

Cellular responses are governed to a large extent by the binding of ligands to cell surface receptors and the signal transduction events that follow. Observing the qualitative characteristics of response behavior as key variables in the signal transduction cascade are changed can provide insight into the fundamental roles of these interactions in producing cellular responses. For example, as the concentration of a signaling molecule is increased in a graded manner, the response may also increase in a graded manner, or the response may increase slowly until a particular concentration is reached and then increase very sharply to a maximum. This latter type of response behavior is often referred to as threshold or all-or-none behavior, and is observed for many types of responses [e.g., actin polymerization in neutrophils (1-3), Ca2+ elevation in helper T-cells (4), BC3H1 cells (5), and bovine smooth muscle cells (6)]. † This work was supported by National Science Foundation Grants BES-9410403 and BES-9713856, the Office of Research and Development, Medical Research Service, Department of Veteran’s Affairs, and the Cellular Biotechnology Training Program, NIH Training Grant GM08353. * Author to whom correspondence should be directed at Research Service (11R), Veteran’s Administration Medical Center, 2215 Fuller Road, Ann Arbor, MI 48105. Phone: (734) 769-7100, ext. 5238. Fax: (734) 761-7693. E-mail: [email protected]. ‡ Department of Chemical Engineering, University of Michigan. § Current address: Johnson & Johnson, Raritan, NJ 08869. | Departments of Biological Chemistry and Surgery, University of Michigan Medical School and Veteran’s Administration Medical Center.

The type of response behavior observed can suggest what type of molecular mechanisms are involved in producing the response. An example of this is given by Huang and Ferrell (7) and Kholodenko et al. (8), who show that threshold behavior can be achieved by having a number of enzymes that follow Michaelis-Menten type kinetics in series. Here we focus on varying the number of G-proteins and ligand/ receptor complexes participating in signal transduction in human neutrophils. Using a combination of experimental data and mathematical modeling, we provide insight into the mechanisms involved in producing the observed response behavior. Neutrophils, or polymorphonuclear leukocytes, are a critical component of the host defense mechanism and inflammatory response. They circulate in the blood until activated by the binding of chemoattractants to specific cell surface receptors. These chemoattractants include N-formyl peptides, leukotriene B4 (LTB4),1 and interleukin-8 (IL-8), each of which has its own receptor. The binding of chemoattractants leads to a complex series of responses including chemotaxis, phagocytosis, oxidant production, and 1 Abbreviations: CHO-MLF, N-formyl-methionyl-leucyl-phenylalanine; LTB4, leukotriene B4; IL-8, interleukin-8; PT, pertussis toxin; DHR-123, dihydrorhodamine-123; NBD, N-(7-nitrobenz-2-oxa-1,3diazol-4-yl); PHPA, parahydroxyphenylacetic acid; CHO-NLFNTKfl, N-formyl-norleucyl-leucyl-phenylalanyl-norleucyl-tyrosyl-lysinefluorescein.

S0006-2960(97)03133-4 CCC: $15.00 © 1998 American Chemical Society Published on Web 07/29/1998

Threshold and Graded Behavior in Neutrophil Responses

FIGURE 1: Signal transduction pathways leading to actin polymerization and oxidant production. Molecules thought to play a role include the β2 isoform of phospholipase C (PLCβ2), phosphatidylinositol-4,5 bisphosphate (PIP2), diacylglycerol (DAG), inositol1,4,5 trisphosphate (IP3), and protein kinase C (PKC). Conversion of receptors to a high-affinity form and interaction of receptors with the cytoskeleton also occur, although the role of these events in signaling is not well-understood.

degranulation. Uncontrolled activation of neutrophil responses can lead to damage of host tissue, and has been implicated in a number of acute and chronic inflammatory diseases (9-11). A quantitative understanding of the signaling mechanisms which lead to these responses is therefore of great importance. The neutrophil chemoattractant receptors couple to and activate a pool of pertussis toxin (PT) sensitive G-proteins upon binding ligand (12-14). To a large extent, the ensuing responses are mediated by activation of these G-proteins. Figure 1 shows a number of the intracellular events involved in the signal transduction cascade leading to oxidant production and actin polymerization including an increase in intracellular free calcium concentration, and activation of phospholipase C and protein kinase C (12, 15). Many other molecules, such as low molecular weight G-proteins and protein kinases and phosphatases, are known to be involved in oxidant production and actin polymerization, but the details of how they participate in the signal transduction cascade have not yet been clearly defined (14, 16). Because much of the signaling pathway is not known, we chose to focus on the early events in signal transduction up to G-protein activation. Some of these early events, such as the conversion of receptors to a high-affinity form (17-19) and the interaction of receptors with the cytoskeleton (2022), are G-protein independent but may still influence signaling. Using flow cytometry in conjunction with PT treatment to examine N-formyl-methionyl-leucyl-phenylalanine (CHOMLF) stimulated actin polymerization and intracellular Ca2+ elevation, Omann et al. (1-3) showed that decreasing the number of functional G-proteins resulted in an increased number of nonresponding cells, while the responding cells continued to respond at a maximal level. Omann et al. (13) also showed that the division of cells into responding and nonresponding populations was not caused by differential ADP-ribosylation of G-proteins by PT. These results imply that a threshold number of receptor-coupled G-proteins is required to initiate neutrophil responses. If the number of active G-proteins is reduced below this threshold, the cells

Biochemistry, Vol. 37, No. 33, 1998 11535 do not respond at all, while cells with greater than the threshold number of G-proteins respond at the maximal level. In contrast, in cells which have not undergone PT treatment, submaximal doses of ligand caused submaximal responses in all cells (graded behavior). It is not known if this threshold behavior is a general phenomena for G-protein coupled chemoattractant receptors and will also be seen with other neutrophil responses. To test the generality of this threshold phenomenon, we investigated the effect of stimulating neutrophils with the chemoattractants LTB4 and IL-8 on the actin polymerization response. We also used a flow cytometric assay to show that threshold behavior is seen with CHO-MLF stimulated oxidant production. In addition, to better understand the implications of our experimental results, we have proposed a simple mathematical model of the response behavior which makes the novel prediction that G-protein independent events are important for regulating responses. MATERIALS AND METHODS Neutrophil Isolation and PT Treatment. Human neutrophils were partially separated from citrated blood by gelatin sedimentation, and further isolated to >98% purity by counterflow elutriation (23). Following isolation, the cells were resuspended in HSB (5 mM KCl, 147 mM NaCl, 1.9 mM KH2PO4, 0.22 mM Na2HPO4, 5.5 mM glucose, 0.3 mM MgSO4, 1 mM MgCl2, 10 mM HEPES, pH 7.4) at 108/mL. Cells that were to undergo PT treatment were resuspended at 5 × 107/mL in Krebs-Ringer buffer with no Ca2+ and with 5.5 mM glucose, 25 mM HEPES, and 6.3 mg/mL cytochrome C added. PT from List Biologicals (Campbell, CA) was reconstituted to 100 µg/mL in sodium phosphate buffer (0.1 M sodium phosphate, 0.5 M NaCl, pH 7.0), giving a final concentration of 0.11 M sodium phosphate and 0.55 M NaCl, and added by dilution to cell suspensions to give the desired final PT concentration. An equivalent amount of PT vehicle (0.11 M sodium phosphate, 0.55 M NaCl, pH 7.0) was added to control cells. Cells were incubated at 37 °C for 1.0-1.5 h with constant rocking. After the incubation, cells were washed in HSB, strained through nylon mesh to remove aggregates, and resuspended in HSB at 108 cells/ mL. Reagents. Catalase, bovine serum albumin (BSA), dimethyl sulfoxide (DMSO), parahydroxyphenylacetic acid (PHPA), superoxide dismutase (SOD), horseradish peroxidase (HRP), cytochrome C, lysophosphatidylcholine, and CHO-MLF were from Sigma Chemical Co. (St. Louis, MO). LTB4 was from Biomol (Plymouth Meeting, PA). IL-8 was from R&D Systems (Minneapolis, MN). Dihydrorhodamine-123 (DHR123) and NBD-phallacidin were from Molecular Probes (Eugene, OR). All chemoattractant ligands were dissolved in DMSO and stored in aliquots at -20 °C. Prior to use the chemoattractants were diluted to