Inflammation, Vol. 16, No. 4, 1992

STIMULUS-DEPENDENT ACTIN POLYMERIZATION IN BOVINE NEUTROPHILS P H I L I P N. B O C H S L E R , D A V I D F. D E A N ,

NANCY

R. N E I L S E N ,

and D A V I D O. S L A U S O N

Department of Pathobiology College of Veterinary Medicine The University of Tennessee Knoxville, Tennessee 37901

Abstract--Polymorphonuclear leukocytes (PMNs) are responsible for much of the first wave of leukocyte-mediated host defense against microbial pathogens. In order to migrate through the endothelium of vessel walls, undergo chemotaxis, and phagocytize microbes, PMNs must modulate their cytoskeletal elements and undergo change of cellular shape. We have used fluorescence flow cytometric analysis and cellular microscopic observations to demonstrate actin polymerization in bovine PMNs and to examine the kinetics of PMN actin polymerization utilizing different PMN stimuli. In addition, we compared temporal relationships between cellular shape and actin polymerization. Actin polymerization occurred rapidly, and the kinetics of actin polymerization were similar for each of the three PMN agonists used, ZAS (10%), PAF (10 -6 M), and rhC5a (10 -7 M). Actin polymerization was near-maximal by 10 sec poststimulation (95.4 % of maximal F-actin content attained by 10 sec poststimulation with ZAS stimulation), and reached peak values by 30 sec. The maximal increase in F-actin content of agonist-stimulated cells as compared to resting cells was 2.8-fold with ZAS; 2.3-fold with PAF; and 2.3-fold with rhC5a. PMN shape change (pseudopodia, membrane ruffles) was not as rapid, with only 22.4% of ceils attaining visible membrane deformation by 10 sec and requiring 120 sec to reach peak shape-change values. After attaining peak values, the two events also differed. Whereas the percent of shape-changed PMNs remained plateaned up to 5 min poststimulation, the F-actin content gradually decreased after 30 sec, approaching F-actin values of unstimulated PMNs.

INTRODUCTION M u c h o f initial c e l l - m e d i a t e d host d e f e n s e directed against bacterial invaders and o t h e r m i c r o b i a l p a t h o g e n s is p r o v i d e d by p o l y m o r p h o n u c l e a r leukocytes ( P M N s ) (1, 2). C o n t i n u o u s l y circulating in sentrylike fashion in the peripheral b l o o d , P M N s are d y n a m i c cells that respond to minute quantities o f inflam383 0360-3997/92/0800-0383506.50/0 9 1992 Plenum Publishing Corporation

384

Bochsler et al.

matory stimuli (3, 4), which sensitize, activate, or otherwise alert PMN to infectious agents or inflammatory events. When appropriately stimulated, PMNs display a repertoire of defense-oriented activity, including up-regulation of adhesion-related surface molecules, such as CD1 lb/CD 18, adhesion to and migration through the vascular wall, chemotaxis toward inflammatory foci, and phagocytosis and destruction of inflammatory material utilizing enzymatic and oxygenradical based systems (5). Remodeling of the PMN cytoskeleton is intrinsic to many PMN functional events, including transendothelial migration, chemotaxis in tissue, secretion, and phagocytosis. Actin is a major and ubiquitous cytoskeletal protein in eukaryotic cells and contributes to both structural and functional cellular activity (6, 7). Modulation of the state of actin polymerization within the cell is responsible, in part, for restructuring of the cytoskeleton, and cytoskeletal changes correlate with motile behavior of cells (8). Functional change in cellular activity related to shape change or movement is often associated with a net increased presence of F-actin (filamentous or polymeric form), and a decrease in G-actin (monomeric form) within the cytosol (9). Actin polymerization is indicated by increased cellular concentration of F-actin and occurs rapidly as a result of minute quantities of chemotactic signal (10), and as such, is a sensitive, early indicator of PMN stimulation. PMN shape change, with rapid development of membrane ruffles, elongate pseudopodia, and an overall change from round to polarized conformation, is similarly an early activation-associated event in PMNs (11). PMN shape change is also used by leukocyte biologists as an indicator of PMN activation and appears to occur concomitantly with increased F-actin content. Utilizing fluorescence flow cytometric quantitation of total cellular F-actin content over time, we demonstrate similarities of kinetics of PMN actin polymerization utilizing several PMN stimuli. We also demonstrate temporal relationships between actin polymerization and cellular shape in bovine PMN. Data resulting from quantitative morphologic observations of PMN shape change and F-actin content suggest that PMN actin polymerization precedes cellular shape change. F-actin content subsequently decreases after attaining an early peak value, but cells retain morphological evidence of shape change over time, even though the F-actin content has decreased.

MATERIALS AND METHODS Neutrophil Isolation. Blood (60-120 ml) was collected from clinically normal cows older than 2 years of age by jugular venipuncture using acid citrate dextrose (1 : 10) as the anticoagulant. Total and differential leukocyte counts and hematocrit values were performed on each blood sample using routine hematologic methods; only samples within established normal limits were used in these studies. Neutrophils were isolated by differential centrifugation combined with flash hypotonic

Stimulus-Dependent Actin Polymerization

385

lysis of the red blood cells as previously described (11, 12). Blood was centrifuged (730 g for 15 min), and the plasma, buffy coat, and the top half of the red cell layer were aspirated and discarded. Red blood cells were lysed by addition of 20 ml of cold, sterile, pyrogen-free distilled water, and isotonicity restored with 10 ml of cold, sterile, pyrogen-free saline (2.7% NaC1). Neutrophils were pelleted by centrifugation (200 g for 10 min) and washed twice with cold Hanks' buffered salt solution (HBSS) without Ca 2§ and Mg 2+. The final cell suspension was resuspended in HBSS containing Ca 2+ and Mg 2+ (HBSS-CM). Final preparations of PMNs, as verified by Wright-Giemsa stained cytocentrifuge preparations, were comprised of a minimum of 80 % PMNs, and the remaining cells were primarily lymphocytes and eosinophils. Cell viability was always high ( > 98 %) as judged by trypan blue dye exclusion. Labeling of F-Actin and Fluorescence Flow Cytometry. Isolated PMNs (3.0 x 106/ml) were prewarmed for 5 rain at 37~ and subsequently stimulated with platelet-activating factor (PAF; 10 6 M), zymosan-activated bovine serum (ZAS; 10%), recombinant human C5a (rhC5a; 10 -7 M), or unstimulated control cells maintained in HBSS-CM. rhC5a has been demonstrated to have a stimulatory effect on bovine PMNs, similar to its effect on human PMNs (13). Some treatments included pretreatment of PMNs with 5/~g/ml cytochalasin B for 10 min at 4~ prior to stimulation. At time intervals of 0 (control cells), 5, 10, 30, 120, and 300 sec poststimulation, the PMNs were rapidly fixed with 1% paraformaldehyde at 22~ PMNs were subsequently centrifuged and washed with HBSS-CM, then resuspended in and incubated with a mixture of lysophosphatidylcholine (100 /zg/ml) and Bodipy-labeled phallacidin (10 -8 M) for 30 min at 4~ with constant agitation in the dark. After incubation, the PMNs were similarly centrifuged and washed and then resuspended in phosphate-buffered saline (pH 7.3) for fluorescence flow cytometric analysis. Quantitative fluorescence analysis was accomplished utilizing a Becton-Dickinson FACScan, with analysis gates set to collect and quantitate the Bodipy fluorescence (similar to fluorescein) associated with PMNs. Any cellular autofluorescence was subtracted from relative fluorescence values, and 10,000 cells were counted per treatment. Some cellular preparations were photographed using Zeiss epifluorescence microscopy. Neutrophil Shape Change. Shape change of PMNs was assessed (11), and data were expressed as percent of PMNs with shape change (membrane raffles and lamellipodia) at each time point. Preparations were photographed using an Olympus BH-2 equipped with Nomarsky optics. Zymosan-Activated Serum. Fresh blood was collected and allowed to clot, then centrifuged (200g for 15 rain); the serum was collected and pooled. The alternate complement pathway was activated by using zymosan A at 2.0 mg/ml of serum and was incubated for 30 rain at 37~ with constant agitation (11). After incubation, the serum was cooled to 4~ centrifuged for 10 miu at 2000g, and decanted from the zymosan pellet. The serum was then further clarified by ultracentrifugation at 20,000g for 10 min. The ZAS was tested for PMN stimulatory activity by aggregometry (14, 15) and frozen in 1.0-ml aliquots at - 7 0 ~ until use. Laboratory Reagents. Materials were obtained as follows: zymosan A, lysophosphatidylcholine, PAF, and rhC5a were purchased from Sigma Chemical Co. (St. Louis, Missouri). Bodipy phallacidin was obtained from Molecular Probes, Inc. (Eugene, Oregon). Bodipy phallacidin is detected with standard fluorescein filters and may be more photostable than standard fluorescein or NBD-phallotoxin preparations.

RESULTS

PMN

s h a p e c h a n g e o c c u r r e d in r a p i d a n d p r e d i c t a b l e f a s h i o n , a n d P M N s p r o -

g r e s s i v e l y d e v e l o p e d v i s i b l e m e m b r a n e ruffles a n d p s e u d o p o d i a ( F i g u r e 1). N o t all P M N s in t h e p o p u l a t i o n u n d e r w e n t s h a p e c h a n g e that w a s d e t e c t a b l e u t i l i z i n g

386

Bochsler et al.

b

Fig. 1. Shape change of bovine PMNs inducedby a PMN agonist. (A) Most of the untreated PMNs held in buffer alone (without agonist) have a round cellular shape. After stimulation (B) by ZAS (10%) for 5 min, most PMNs exhibit varying degrees of membrane perturbation, evidenced by membrane ruffles and pseudopodia. Wet-mountpreparation of PMN using Nomarskyoptics, •

standard light microscopy. F-actin staining was easily detected using standard fluorescence microscopic technique on Bodipy phallacidin-labeled PMNs (Figure 2). Fluorescence was most intense in the plasma membrane and subjacent cytosol, correlating well with the predominant cellular location of F-actin. The kinetics of P M N shape change were noticeably different from the kinetics of actin polymerization (Figure 3). Actin polymerization occurred rapidly, with 95.4% of maximal F-actin content attained by 10 sec poststimulation and peak values by 30 sec. Shape change o f PMNs was not as rapid, with only 22.4% of total cells attaining visible membrane deformation by 10 sec and requiring 120 sec to reach peak shape-change values. After attaining peak values, the two events also differed. Whereas the percent of shape-changed PMNs remained

Stimulus-Dependent Actin Polymerization

387

Fig. 2. F-actin in stimulated PMNs. Cells display bright fluorescence associated with binding of Bodipy phallacidin label to F-actin. Fluorescence is most intense in pseudopodia and subplasma membrane locations. PMN were stimulated with ZAS (10%) for 5 min. • 9~

SO.

I'~

9 800 "0

70,

T

T

"6

9500

vZR

80.

o

r

50,

~.

4o,

" 9~ T

Shape Change - 400

.

O"-

0

~

F-Actin

f,

o

9

~o. ~= Z ~E n

0-- 9 ~

Z ..~ Q,.

........

T

.#

9300

j-

20

9 200

-~

10 84 0

I

I

I

[

0

30

60

90

I 120

I 150

I 180

I 210

I 240

I 270

100 300

Seconds

Fig. 3. Comparison of kinetics of PMN actin polymerization and membrane shape change9Actin polymerization and maximal F-actin content of ZAS-stimulated PMNs preceded PMN shape change, and F-actin content subsequently decreased with time. Relative fluorescence of PMNs (right ordinate axis) indicates relative amount of F-actin. All values are + SEM; N = 9 for F-actin; N = 12 for shape change.

plateaued up to 5 m i n poststimulation, the F-actin content gradually decreased, approaching F-actin values o f u n s t i m u l a t e d P M N (Figures 3-5). T h e kinetics o f actin p o l y m e r i z a t i o n were similar for each of the three P M N agonists used: Z A S , P A F , and rhC5a (Figures 4 and 5). In all cases, actin p o l y m e r i z a t i o n was n e a r - m a x i m a ! b y 1 0 - 3 0 sec poststimulation, and subsequently declined, eventually approaching levels o f F-actin in unstimulated cells. P M N s held in buffer alone; without a P M N agonist, showed a limited increase

388

Bochsler et al. 600.

z -5 13_

.4 500.

*6

'"'t'\

C

400"

O

~O.

~

I

T

o - - o Control

""

T

~g -~

200.

100

I

I

I

I

0

30

60

90

I

I

I

I

I

I

I

I

120 150 180 210 240 270 300 Seconds

Fig. 4. Kinetics of actin polymerization in PMNs. The kinetic profile of actin polymerization was similar for both PMN agonists (ZAS, 10%; PAF, 10 -6 M). After reaching peak values, F-actin content in both agonist-stimulated groups decreased toward levels in control P M N (buffer alone). All values are _+ SEM; N = 9.

800~ o--e

Z

0-

500-

T !-3

= t~ II)

400-

o

~00-

GSa

m---a csa + C'~ochalas|nB o . - o co.~,

Lt.

o "~

200.

100

I

I

I

I

n

30

60

II0

I

I

I

I

I

I

I

I

120 150 180 210 240 270 300 Seconds

Fig. 5. Actin polymerization in C5a-stimulated PMNs. The kinetic profile of actin polymerization was similar to observed with other stimuli (PAF, ZAS). Cytochalasin B abrogated actin polymerization in C5a-stimulated PMNs but did not reduce F-actin content to levels below those in control cell (buffer alone). Cells treated with cytochalasin B alone (no C5a) had F-actin content similar to those treated with both cytochalasin B and C5a. All values are + SEM; N = 9.

o f F-actin within the first 10 sec of incubation (25.8% as compared to basal levels), and this may have been due to handling (pipetting) of PMNs. This increase was small compared to the maximal increase in F-actin content o f agonist-stimulated cells (283.8% with ZAS; 233.9% with P A F ; 226.2% with rhC5a). Cytochalasin B, a fungal metabolite that binds to the fast-growing barbed

Stimulus-Dependent Actin Polymerization

389

ends of actin filaments and prevents the addition of actin molecules, effectively inhibited the rapid actin polymerization associated with C5a stimulation of PMN (Figure 5). However, preincubation of PMNs with cytochalasin B did not prevent the limited actin polymerization in unstimulated PMNs (buffer alone) that occurred within the first 10 sec (data not shown), nor did cytochalasin B pretreatment reduce F-actin content to levels below that of control PMNs (Figure 5).

DISCUSSION

Polymerization and depolymerization of actin in PMNs is associated with stimulus-mediated plasma membrane deformation, i.e., surface membrane rufties, blebs, and extension of pseudopodia/lamellipodia. Monomeric actin rapidly assembles into filaments referred to as F-actin, a process that correlates with change in cellular morphology and locomotion in PMNs (8, 16) and that can be inhibited by fungal agents such as the cytochalasins. Numerous actin-binding proteins exist that choreograph the dynamic flux of actin within cells and may either promote or inhibit reversible assembly of actin into filaments and also influence the three-dimensional shape of actin networks (7). Because cell-surface morphology and movement of PMNs are linked to important functions in host defense, such as chemotaxis and phagocytosis, understanding the relationship of actin polymerization to the cellular shape of PMNs is significant. Actin polymerization induced by the chemotactic factor f-Met-Leu-Phe (FMLP) in human PMNs occurs rapidly and has been variably reported to require 10 (17, 18), 30 (19), or 45 see (16) to attain peak F-actin values in human PMNs with maximal stimulatory concentrations of FMLP (10 -7 o r 10 - 6 M). Although the total amount of F-actin polymerized depends on the concentration of the stimulus, i.e., more stimulus results in more F-actin, the kinetics vary little as a result of stimulus concentration (6, 16, 19). Peak F-actin content occurs at a similar time point, regardless of concentration. Bovine PMNs do not respond to the formyl peptide, FMLP, but our data with bovine PMNs similarly indicates that peak actin polymerization occurs within a time frame of 10-30 sec after excitation with an appropriate agonist. The time-dependent profile of total F-actin content in bovine PMN using either rhC5a, PAF, or ZAS as stimulus was comparable. This may suggest that these stimuli induce actin polymerization through a common pathway or that multiple pathways with similar kinetics exist. It is of note that a relatively specific inhibitor of protein kinase C, CGP 41 251 (Ciba-Geigy Ltd, Basel, Switzerland) has been reported to have no effect on actin polymerization or depolymerization in human fNLPNTLstimulated PMNs (20), indicating that PKC activation is not required for cellular

390

Bochsler et al.

flux of actin. GTP-binding proteins have been reported to play a positive role in transducing membrane signals into actin polymerization in human PMNs (21, 22). Staurosporine, which inhibits several kinases with high potency (PKC, PKA, PKG, tyrosine kinases, phosphorylase kinase), actually induced an increase in cytoskeleton-associated actin (20). Actin polymerization in PMNs is quickly followed by a phase of depolymerization, and bovine PMNs in our study are clearly in a phase of net actin depolymerization by 120-300 sec poststimulation. This is consistent with reported observations of human PMNs stimulated by FMLP (16, 17, 23). PMN membrane conformational changes and actin polymerization may appear to occur concomitantly, and this might suggest that the two events are directly related, i.e., maximal PMN shape change occurs simultaneously with maximal F-actin content. Investigators have determined that the extent of actin polymerization in human PMNs does have a positive correlation with the extent of PMN shape change. Limited actin polymerization is associated with a ruffled and/or blebbed surface, and greater actin polymerization with polar conformation (23). The events are not simultaneous, however. In the same study, it was noted that maximal PMN conformational change to polar shape occurred at 300 sec poststimulation, whereas maximal F-actin content occurred at 30 sec. Further, the polar shape of PMNs occurred at a time point (300 sec) when most actin had depolymerized, and F-actin content was reduced to levels near that of unstimulated control PMN. Our results utilizing bovine PMN were similar to those reported for human PMN, and peak F-actin levels were reached by 30 sec poststimulation. The bovine PMN data do not differentiate between ruffled and/ or blebbed PMN forms and polarized configuration but clearly show that maximal numbers of bovine PMNs exhibit membrane conformational changes during a poststimulatory period (60-300 sec) of net actin depolymerization. These data, based on averages of values derived from populations of PMNs, indicate that maintenance of PMN polar conformation does not depend on continued maximal actin polymerization. Further, depolymerization of actin in PMNs does not immediately dictate return to a round smooth conformation of the entire cell. Other factors than F-actin content must influence the maintenance of altered PMN surface conformation. Other experimental evidence derived from observations of stimulated human PMNs indicate that smaller, short-lived fluctuations in F-actin content of PMNs may occur. Typical data of F-actin content derived from time-measurement intervals at 5, 10, or 30 sec or greater (poststimulation) yields a smooth curve, with a rapid, progressive increase, and a slower, steady decline. If sampled at 2-sec intervals, however, the F-actin content of PMNs oscillates as a function of time (18). In this study, the turbidity of PMN suspensions was used as an indicator of development of lamellipodia, and results suggested that rapid, cyclic variations in F-actin content correlated with periodic size fluctuations of lamellipodia, i.e., extension and retraction of lamellipods.

Stimulus-Dependent Actin Polymerization

391

Actin polymerization is a normal event of PMNs in response to an appropriate stimulus and is involved in many leukocyte functions. If standard events in PMN actin polymerization deviate from normal, the potential exists for compromised host defense. Baseline F-actin content of human neonatal PMNs may be higher than that of adult PMNs, and actin polymerization in neonatal PMNs as a result of stimulation may be less (24, 25). A genetically determined disorder of actin function has also been reported (26). Defects or alterations in PMN actin function may, therefore, underlie problems of the cellular immune system in some cases, and further study of the relationship between the PMN cytoskeleton and host defense is necessary. Acknowledgments--This work was supported by U.S.D.A. competitive research grant 90-372655611, and the University of Tennessee Centers of Excellence.

REFERENCES

1. SAWYER,D. W., G. R. DONOWITZ, and G. L. MANDELL. 1989. Polymorphonuclear neutrophils: An effective antimicrobial force. Rev. Infect. Dis. ll(Suppl. 7):$1532-$1544. 2. COLDITZ, I. G., R. L. KERLIN, and D. L. WATSON. 1988. Migration of neutrophils and their role in elaboration of host defense. In Migration and Homing of Lymphoid Cells. A. J. Husband, editor. CRC Press, Boca Raton, Florida. 135-165. 3. FERRANTE,A., M. NANDOSKAR,A. WALZ, D. H. B. GOH, and I. C. KOWANKO. 1988. Effects of tumor necrosis factor alpha and interleukin-1 alpha and beta on human neutrophil migration, respiratory burst and degranulation. Int. Arch. Allergy Appl. Immunol. 86:82-91. 4. WEBSTER,R. O., S. R. HONG, R. B. JOHNSTONJR., and P. M. HENSON. 1980. Biologic effects of the human complement fragments C5a and C5aae~A~gon neutrophil function. Imrnunopharmacology 2:201-219. 5. VAN KESSEL, K. P. M., and J. VERHOEF. 1990. A view to a kill: Cytotoxic mechanisms of human polymorphonuclear leukocytes compared with monocytes and natural killer cells. Pathobiology 58:249-264. 6. PACKMAN, C. H., and M. A. LICHTMAN. 1990. Activation of neutrophils: Measurement of actin conformational changes by flow cytometry. Blood Cells 16:193-207. 7. STOSSEL,T. P. 1989. From signal to pseudopod. How cells control cytoplasmic actin assembly. J. Biol. Chem. 264:18261-18264. 8. HOWARD,T. H., and W. H. MEYER. 1984. Chemotactic peptide modulation of actin assembly and locomotion in neutrophils. J. Cell Biol. 98:1265-1271. 9. KORN, E. D. 1982. Actin polymerization and its regulation by proteins from nonmuscle cells. Physiol. Rev. 62:672-737. 10. WALLACE,P. J., R. P. WERSTO, C. H. PACKMAN,and M. A. LICI-ITMAN. 1984. Chemotactic pept!de-induced changes in neutrophil actin conformation. J. Cell Biol. 99:1060-1065. 11. HOLDEN, W., D. O. SLAUSON, R. D. ZWAHLEN, M. M. SUYEMOTO, M. DORE, and N. R. NEILSEN. 1989. Alterations in complement-induced shape change and stimulus-specific superoxide anion generation by neonatal calf neutrophils. Inflammation 13:607-620. 12. CLIFFORD, C. B., D. O. SLAUSON, N~ R. NEILSEN, M. M. SUYEMOTO, R. D. ZWAHLEN, and D. H. SCHLAFER. 1989. Ontogeny of inflammatory cell responsiveness: Superoxide anion gen-

392

13. 14.

15. 16.

17. 18.

19. 20.

21.

22.

23. 24. 25.

26.

Bochsler et al.

eration by phorbol ester-stimulated fetal, neonatal, and adult bovine neutrophils. Inflammation 13:221-231. DORE, M., D. O. SLAUSON,M. M. SUYEMOTO,and N. R. NEILSEN. 1990. Calcium mobilization in C5a-stimulated adult and newborn bovine neutrophils. Inflammation 14:71-82. HAMMERSCHMIDT,D. E., T. K. BOWERS, C. J. LAMMI-KEEFE,H. S. JACOB, and P. R. CRADDOCK. 1980. Granulocyte aggregometry: A sensitive technique for the detection of C5a and complement activation. Blood 55:898-902. SLAUSON,D. O., D. S. SKRABALAK,N. R. NEILSEN, and R. D. ZWAHLEN. 1987. Complementinduced equine neutrophil adhesiveness and aggregation. Vet. Pathol. 24:239-249. HOWARD,T. H., andC. O. ORESAJO. 1985. The kinetics ofchemotactic peptide-induced change in F-actin content, F-actin distribution, and the shape of neutrophils. J. Cell Biol. 101:10781085. HOWARD, T., C. CHAPONNIER, H. YIN, and T. STOSSEL. 1990. Gelsolin-actin interaction and actin polymerization in human neutrophils. J. Cell Biol. 110:1983-1991. WYMANN, M. P., P. KERNEN, T. BENGTSSON, T. ANDERSSON, M. BAGGIOLIN1, and D. A. DERANLEAU. 1990. Corresponding oscillations in neutropbil shape and filamentous actin content. J. Biol. Chem. 265:619-622. WANG, D. H., K. BERRY, and T. H. HOWARD. 1990. Kinetic analysis of chemotactic peptideinduced actin polymerization in neutropbils. Cell Motil. Cytoskeleton 16:80-87. NIGGLI, V., and H. KELLER. 1991. On the role of protein kinases in regulating neutrophil actin association with the cytoskeleton. J. Biol. Chem. 266:7927-7932. BENGTSSON,Z., T. STENDAHL, and T. ANDERSSON. 1986. The role of cytosolic free calcium transient for fMet-Leu-Phe induced actin polymerization in human neutrophils. Eur. J. Cell Biol. 42:338-343. BENGTSSON,T., E. SARNDAHL,O. STENDAHL, and T. ANDERSSON. 1990. Involvement of GTPbinding proteins in actin polymerization in human neutrophils. Proc. Natl. Acad. Sci. U.S.A. 87:2921-2925. WATTS, R. G., M. A. CRISPENS, and T. H. HOWARD. 1991. A quantitative study of the role of F-actin in producing neutrophil shape. Cell. Motil. Cytoskeleton 19:159-168. HILMO, A., and T. H. HOWARD. 1987. F-actin content of neonate and adult neutrophils. Blood 69:945-949. SACCHI, F., N. H. AUGUSTINE, M. M. COELLO, E. Z. MORRIS, and H. R. HILL. 1987. Abnormality in actin polymerization associated with defective chemotaxis in neutrophils from neonates. Int. Arch. Allergy Appl. lmmunol. 84:32-39. SOUTHWICK,F. S., G. A. DABIRI, and T. P. STOSSEL. 1988. Neutrophil actin dysfunction is a genetic disorder associated with'partial impairment of neutrophil actin assembly in three family members. J. Clin. Invest. 82:1525-1531.