Cell Motility and the Cytoskeleton 21:5&64 (1992)

Cytochalasins Induce Actin Polymerization in Human Leukocytes K. Murali Krishna Rao, Jaya Padmanabhan, and Harvey J. Cohen Geriatric Research, Education and Clinical Center, VA Medical Center; Center for the Study of Aging and Human Development, Duke University Medical Center, Durham, North Carolina

We studied the effect of cytochalasins (B, D, and E) on the F-actin content in human neutrophils and lymphocytes using NBD-phallacidin labeling followed by flow cytometry. All three cytochalasins induced a concentration- and time-dependent increase in the F-actin content in both cell types. The order of potency was cytochalasin D > E > B. The increase in F-actin content was accompanied by a decrease in the G-actin content as measured by DNase I inhibition assay. These observations suggest that in intact cells cytochalasins may function differently compared to purified and semipurified systems, and their effects may be modified through other actin-binding or sequestering proteins. 2-deoxyglucose (20 mM) caused a decrease in the basal F-actin content and significantly reduced the change induced by the cytochalasins. These results suggest that the state of actin in intact cells is regulated by cytosolic ATP levels, primarily by the integrity of the glycolytic pathway. Based on these observations, we conclude that the mechanism of action of cytochalasins in intact cells is more complex than current models suggest. Key words: cytoskeleton, neutrophils, lymphocytes, metabolic inhibitors, F-actin

INTRODUCTION

Cytochalasins are fungal products that induce profound changes in cell morphology by affecting the actin network of the cytoskeleton [Tannenbaum, 19781. Although their specificity towards actin seems to be well established, the actual mechanism of action of cytochalasins is not well understood. In vitro studies using purified actin show that cytochalasins inhibit the rate of actin polymerization [Brenner and Korn, 1979; Brown and Spudich, 1979; Flanagan and Lin, 1980; Lin et al., 19801 by blocking the addition of actin subunits to the fast-growing or the barbed end of the actin filaments [Brown and Spudich, 1979; Flanagan and Lin, 1980; Maruyama et al., 19841. In addition to preventing elongation of the actin filaments, cytochalasins have been shown to have profound effects on gels of cytoplasmic extracts [Pollard, 1976; Weihing, 19761 and on mixtures

Fletcher and Pollard, 19801. They have also been shown to reduce low-shear viscosities of actin filaments alone 0 1992 Wiley-Liss, Inc.

[Lin et al., 1980; MacLean-Fletcher and Pollard, 19801. On the basis of such in vitro studies, cytochalasins have been deduced to have the following properties: I ) “capping” of actin filaments by binding to the barbed (faster growing) end [Brenner and Korn, 1979; Brown and Spudich, 1979; Flanagan and Lin, 1980; Lin et al., 1980; Maruyama et al., 19841; 2) cleavage of actin filaments [Hartwig and Stossel, 1979; Selden et al., 1980; Mozo-Villaria and Ware, 1984; Maruyama et al., 19891; 3 ) acceleration of the kinetics of actin assembly [Brenner and Korn, 1980; Lanni et al., 1981; Tellam and Frieden, 19821; and 4) stimulation of ATPase activity of actin [Brenner and Korn, 1980; Brenner and Korn, 1981; Dancker and Kliche, 198 11. In intact cells, cytochalasin D (CD) treatment disrupts network organization, increases the number of ac-

Address reprint requests to K. Murali Krishna Rao, M.D., Box 182 A , V.A. Medical Center, Durham, NC 27705.

Cytochalasin-Induced Actin Polymerization

59

tin filament ends, and leads to the formation of filamen- teer donors after informed consent. The blood was subtous aggregates or foci composed mainly of actin jected to the Ficoll-Hypaque density gradient technique filaments. Metabolic inhibitors prevent filament redistri- to obtain the mononuclear cells [Boyum, 19681, and the bution, foci formation, and cell arborization, but not dis- red cell pellet was used to isolate neutrophils by dextran organization of the three-dimensional filament network sedimentation. The red cells in both preparations were [Schliwa, 19821. It is proposed that the effects of CD on lysed by hypotonic shock for 40 sec. The cells were cells results from both a direct interaction of the drug washed twice in Hanks’ balanced salt solution without with the actin filament network, which causes disruption calcium and magnesium. of the actin network, and a secondary, energy-depenDetermination of F-Actin Content dent, cellular response that leads to the formation of The F-actin content was quantitated by NBD-phalfilament foci [Schliwa, 19821. lacidin labeling as described previously [Wallace et al. , Immunofluorescence studies show that the treat1984; Rao et al., 19881. Briefly, 100 pl of the cell susment of cells with cytochalasins causes the disintegration pension (20 million cells/ml) was incubated with various of stress fibers [Weber et al., 19761. This observation, agents, and fixed and stained in a single step by adding combined with the electron microscopic studies showing 100 pl of a staining cocktail, consisting of 6.4% a decrease in the actin filament network, led to the wideparaformaldehyde, 200 pg/ml of lysophosphatidylchospread belief that cytochalasins inhibit actin polymerizaline, and 0.6 pM NBD-phallacidin in phosphate buffered tion in intact cells. The immunofluorescence studies and saline, pH 7.4 (PBS), for 30 min at room temperature. electron microscopic studies are usually conducted on The cells were spun at 12,000 g for 1 min in a microfuge cells growing in monolayers, and the measurements are 1% BSA using (Beckman) and washed twice with PBS/O. semiquantitative at best. Over the past few years, it has 1-min microfuge spins. The cells were finally suspended become feasible to measure the filamentous actin (F1 ml of the wash buffer and stored at 4°C until analysis in actin) content in intact cells using NBD-phallacidin laby flow cytometry. The F-actin content was expressed as beling and flow cytometry [Howard and Meyer, 1984; the mean channel fluorescence (MCF). The flow cytomWallace et al., 1984; Phatak et al., 19881. Using this technique, recently it was reported that CD increases eter was calibrated each time using the Quantitative FluF-actin content in murine lymphocytes [Wilder and Ash- orescent Bead Standards (Flow Cytometry Standards man, 19911. In this report we demonstrate that a similar Corp., Research Triangle Park, NC). In some experiments the F-actin content was deterincrease in F-actin content occurs in human neutrophils and lymphocytes when treated with three different cy- mined by a two-step staining procedure, as described by tochalasins (cytochalasins B, D, and E). Further, we Wallace et al. [1984]. The cells were fixed in paraforinvestigated the role of metabolic inhibitors in maintain- maldehyde for 48-72 h (without lysolecithin) and ing the F-actin levels in intact cells and in cytochalasin- stained with NBD-phallacidin, with essentially the same induced changes. The observations reported here provide results as obtained with the one-step procedure described important insights into the mechanism of action of cy- above. tochalasins in intact cells. Determination of G-Actin Content MATERIALS AND METHODS Reagents

L-a-lysophosphatidyl-choline,type I (L-a-lysolecithin); DNA (type 1); 2-deox yglucose; iodoacetic acid (sodium salt); 2,4-dinitrophenol; Na azide; cytochalasins B, D, and E; and homidium (ethidium) bromide were obtained from Sigma Chemical Co. (St. Louis, MO). 7-nitrobenz-2-oxa- 1,3 diazole (NBD)-phallacidin was from Molecular Probes (Junction City, OR). DNAse I was obtained from Calbiochem-Behring Corp. (La Jolla, CA). Cytochalasins were dissolved in ethanol at a concentration of 2 mM and stored at -20°C.

The G-actin content in the cells was determined by DNase I inhibition assay, as described previously [Rao and Varani, 1982; Varani et al., 19831. The G-actin content was expressed as the percent change in inhibition, compared to control untreated cells whose activity was taken as 100%. The decrease in inhibitory activity reflects the conversion of G-actin to F-actin. Statistical Analyses The paired t-test was used to estimate the differences between the different treatments.

Isolation of Neutrophils and Lymphocytes

RESULTS Effects of Cytochalasins on F-Actin Content in Neutrophils

Human peripheral blood was obtained by venipuncture and collected into heparin from healthy volun-

Cells were incubated with various concentrations of cytochalasins for 10 min at 37°C and the F-actin con-

60

Rao et al.

300

T

240

t

401 0 1 2

I

I

I

5

10

15

Cyto D

0-0

' '

0 1

I

5

I

I

I

10

30

60

Time (Min) Cytachalasin D

(uU)

Fig. 1. Cytochalasin D dose-response curve in human neutrophils. The cells were incubated for 10 rnin at 37"C, and the F-actin content was determined by NBD-phallacidin labeling. The data represent mean t SEM of 4 separate experiments. The error bars were smaller than the symbol in some cases.

tent was determined. Cytochalasins induced a concentration-dependent increase in the F-actin content in neutrophils, as shown in Fig. 1, for CD. The maximum response with CD was obtained at 10 p M and above. Cytochalasin B (CB) and cytochalasin E (CE) showed similar concentration optima (data not shown). The response to cytochalasin D was seen as early as 1 rnin and gradually increased up to 1 h (Fig. 2). Further, it is evident from the figure that CD is the most potent drug, CB the least potent, and CE has an activity in between. CB shows great variability in response, and when cells were incubated with the drug for 30-60 min the increase in the F-actin became negligible. Several control experiments were done to exclude nonspecific effects. Treatment of cells with up to 0.5% ethanol (used for dissolving cytochalasins) had no effect on the F-actin content. Further, adding CD after fixing the cells did not produce any change in fluorescence, indicating that cytochalasins were not simply modifying the binding of NBD-phallacidin with F-actin. This was also confirmed by adding a 1000-fold excess of unlabeled phallacidin after the staining cocktail, which completely abolished binding of NBD-phallacidin in both control and CD-treated cells. When excess unlabeled phallacidin was added 30 min after the staining cocktail, there was partial displacement from both control and CD-treated cells. Effects of Varying the Concentration of NBD-Phallacidin for Staining In order to study if the NBD-phallacidin concentration used has any effect on the F-actin content in the presence of cytochalasins, we used three different con-

Fig. 2 . The effects of different cytochalasins on the F-actin content in neutrophils. The drug concentration was 10 p M in each case. The data represent mean 2 SEM of 3-5 separate experiments. In experiments with cytochalasin B, the error bars were smaller than the data symbols.

centrations of NBD-phallacidin. The baseline fluorescence intensity was directly proportional to the concentration of the NBD-phallacidin used. Although there were some differences in the magnitude of the response, CD caused an increase in F-actin at all concentrations of NBD-phallacidin tested (Table I). Effects of Cytochalasins on the F-Actin Content in Lymphocytes The treatment of lymphocytes with different cytochalasins yielded results similar to those with neutrophils, both with respect to the time course and the concentration optima (data not shown). Figure 3 shows one representative example of four separate experiments with lymphocytes. The histograms were gated on the basis of light scatter to exclude monocytes from the fluorescence histograms shown in the figure. An increase in the Factin content was noted with all three cytochalasins. The histogram with CE is more heterogeneous, yielding a long tail; the reason for this heterogeneity is not known. Effects of Metabolic Inhibitors

We determined the effect of several metabolic inhibitors, 2-deoxyglucose7 iodoacetic acid, Na azide, and 2,4-dinitrophenol on the basal F-actin content and on changes induced by cytochalasin D (Table 11). Cells were preincubated for 10 min at 37°C with the metabolic inhibitors at the indicated concentrations and then were incubated with cytochalasin D for 10 additional minutes. Control cells were incubated for 20 min at 37°C with metabolic inhibitors alone. Treatment with deoxyglucose or iodoacetic acid alone reduced the amount of F-actin in unstimulated neutrophils, whereas 2,4-dinitrophenol ( 5 mM) and Na azide (10-30 mM) had no effect on the basal level of F-actin. Pretreatment with deoxyglu-

Cytochalasin-InducedActin Polymerization TABLE I. Effect of Varying the Concentration of NBD-Phallacidin on Cytochalasin D-Induced Actin Polymerization

61

Control

MCF - 74

F-actin content (MCF)

NBD-phallacidin OLM)

Control

CD (10 pM)

% change

64

111 74 41

73 45 46

0.6 0.3 0.15

51 28

The cells were incubated with CD for 10 min at 37°C and labeled with a staining cocktail containing various amounts of NBD-phallacidin as indicated. MCF = mean channel fluorescence.

Cyto. B MCF - 89

TABLE 11. Effects of Metabolic Inhibitors on F-Actin Content in Human Neutrophils F-actin content (MCF)

Treatment Control cells Deoxyglucose (20 mM) Iodoacetic acid (2 mM) 2,4-dinitrophenol (5 mM) Na azide (10 mM) Cytochalasin D (10 pM) Deoxyglucose + CD Iodoacetic acid + CD 2,4-dinitrophenol CD CD Na azide

+

+

56 37 42 55 56 199 113 210 245 209

P value

C to E &F- 123

2 7

*2 k

*

* 2 * 2 2 2

2 9 5 32 5 50 38 20

Cytochalasins induce actin polymerization in human leukocytes.

We studied the effect of cytochalasins (B, D, and E) on the F-actin content in human neutrophils and lymphocytes using NBD-phallacidin labeling follow...
654KB Sizes 0 Downloads 0 Views