Phorbol ester-induced actin assembly in neutrophils: role of protein kinase C

The shape changes and membrane ruffling that accompany neutrophil activation are dependent on the assembly and reorganization of the actin cytoskeleton, the molecular basis of which remains to be clarified. A role of protein kinase C (PKC) has been postulated because neutrophil activation, with the attendant shape and membrane ruffling changes, can be initiated by phorbol esters, known activators of PKC. It has become apparent, however, that multiple isoforms of PKC with differing substrate specificities exist. To reassess the role of PKC in cytoskeletal reorganization, we compared the effects of diacylglycerol analogs and of PKC antagonists on kinase activity and on actin assembly in human neutrophils. Ruffling of the plasma membrane was assessed by scanning EM, and spatial redistribution of filamentous (F)-actin was assessed by scanning confocal microscopy. Staining with NBD-phallacidin and incorporation of actin into the Triton X-100-insoluble ("cytoskeletal") fraction were used to quantify the formation of (F)-actin. [32P]ATP was used to detect protein phosphorylation in electroporated cells. Exposure of neutrophils to 4 beta-PMA (an activator of PKC) induced protein phosphorylation, membrane ruffling, and assembly and reorganization of the actin cytoskeleton, whereas the 4a-isomer, which is inactive towards PKC, failed to produce any of these changes. Moreover, 1,2-dioctanoylglycerol, mezerein, and 3-(N-acetylamino)-5-(N-decyl-N-methylamino)-benzyl alcohol, which are nonphorbol activators of PKC, also promoted actin assembly. Although these effects were consistent with a role of PKC, the following observations suggested that stimulation of conventional isoforms of the kinase were not directly responsible for actin assembly: (a) Okadaic acid, an inhibitor of phosphatases 1 and 2A, potentiated PMA-induced protein phosphorylation, but not actin assembly; and (b) PMA-induced actin assembly and membrane ruffling were not prevented by the conventional PKC inhibitors 1-(5-isoquinolinesulfonyl)-2-methylpiperazine, staurosporine, calphostin C, or sphingosine at concentrations that precluded PMA-induced protein phosphorylation and superoxide production. On the other hand, PMA-induced actin assembly was inhibited by long-chain fatty acid coenzyme A esters, known inhibitors of nuclear PKC (nPKC). We conclude that PMA-induced actin assembly is unlikely to be mediated by the conventional isoforms of PKC, but may be mediated by novel isoforms of the kinase such as nPKC.     

Role of actin binding protein phosphorylation in platelet cytoskeleton assembly

Actin binding protein from human blood platelets is shown to exist in the resting platelet as a phosphorylated protein and contains two residues of phosphate per 260,000 kd. Removal of one-half of these residues with E. coli alkaline phosphatase results in the loss of its ability to crosslink F-actin into a low speed sedimentable complex (its cytoskeleton) and to bind to an F-actin affinity column. Thus, phosphorylation-dephosphorylation of ABP may be an important regulatory mechanism by which the platelet regulates its shape via its cytoskeletal structure.     

Receptor-mediated actin assembly in electropermeabilized neutrophils: role of intracellular pH

Neutrophil activation by a variety of stimuli is accompanied by an intracellular acidification, which has been postulated to mediate actin polymerization (Yuli and Oplatka, Science 1987, 235, 340). This hypothesis was tested using 7-nitrobenz-2-oxa-1,3-diazole (NBD)-phallacidin staining and flow cytometry, or right angle light scattering to study actin assembly in intact and electrically permeabilized human neutrophils. Intracellular pH was measured fluorimetrically using a pH sensitive dye. In cells stimulated with N-formyl-methionyl-leucyl-phenylalanine (fMLP) at 21 degrees C, actin assembly clearly preceded the intracellular acidification in response to fMLP. Moreover, actin polymerization persisted in cells where intracellular pH was clamped near the resting (unstimulated) level using nigericin/K+. Finally, fMLP induced a significant increase in F-actin content in electropermeabilized neutrophils equilibrated with an extracellular medium containing up to 50 mM HEPES. These observations indicate that fMLP-stimulated F-actin assembly is not mediated by a decrease in intracellular pH and suggest that changes in transmembrane potential and ionic gradients are unlikely to mediate actin polymerization.     

Chemotactic peptide modulation of actin assembly and locomotion in neutrophils

To determine the relationship between the state of actin polymerization in neutrophils and the formyl-methionyl-leucyl-phenylalanine (fMLP)-induced changes in the locomotive behavior of neutrophils, the mean rate of locomotion (mROL), the percent G-actin, and the relative F-actin content of neutrophils were determined. The mROL was quantified by analysis of the locomotion of individual cells; the percentage of total actin as G-actin was measured by DNase I inhibition; and the F-actin was determined by fluorescence-activated cell sorter (FACS) analysis of nitrobenzoxadiazol (NBD)-phallacidin-stained neutrophils. Neutrophils stimulated with fMLP exhibit a change in their mROL that is biphasic and dose dependent. The mROL of neutrophils exposed to 10(-8) M fMLP, the KD, is 11.9 +/- 2.0 micron/min (baseline control 6.2 +/- 1.0 micron/min). At 10(-6) M fMLP, the mROL returns to baseline levels. Stimulation of neutrophils with fMLP also induces action polymerization. Evidence for actin polymerization includes a 26.5% reduction in G-actin and a twofold increase in the amount of NBD-phallacidin staining of cells as determined by FACS analysis. The NBD-phallacidin staining is not due to phagocytosis, is inhibited by phalloidin, requires cell permeabilization, and is saturable at NBD-phallacidin concentrations greater than 10(-7)M. The fMLP-induced increase in NBD-phallacidin staining occurs rapidly (less than 2 min), is temperature dependent, and is not due to cell aggregation. Since NBD-phallacidin binds specifically to F-actin, the increase in fluorescent staining of cells likely reflects an increase in the F-actin content of fMLP-stimulated cells. FACS analysis of NBD-phallacidin-stained cells shows that the relative F-actin content of neutrophils stimulated with 10(-11)-10(-8)M fMLP increases twofold and remains increased at concentrations greater than 10(-8)M fMLP. Therefore, the fMLP-induced increase in F-actin content of neutrophils as determined by FACS analysis of NBD-phallacidin-stained cells coincides with a decrease in G-actin and correlates with increased mROL of neutrophils under some (10(-11)-10(-8)M fMLP) but not all (greater than 10(-8)M fMLP) conditions of stimulation. Quantification of the F-actin content of nonmuscle cells by FACS analysis of NBD-phallacidin-stained cells may allow rapid assessment of the state of actin polymerization and correlation of that state with the motile behavior of nonmuscle cells.     

Role of protein phosphatases in hypoxic preconditioning

To find a protein kinase C (PKC)-independent preconditioning mechanism, hypoxic preconditioning (HP; i.e., 10-min anoxia and 10-min reoxygenation) was applied to isolated rat hearts before 60-min global ischemia. HP led to improved recovery of developed pressure and reduced end-diastolic pressure in the left ventricle during reperfusion. Protection was unaffected by the PKC inhibitor bisindolylmaleimide (BIM; 1 micromol/l). It was abolished by the inhibitor of protein phosphatases 1 and 2A cantharidin (20 or 5 micromol/l) and partially enhanced by the inhibitor of protein phosphatase 2A okadaic acid (5 nmol/l). In adult rat cardiomyocytes treated with BIM and exposed to 60-min simulated ischemia (anoxia, extracellular pH 6.4), HP led to attenuation of anoxic Na(+)/Ca(2+) overload and of hypercontracture, which developed on reoxygenation. This protection was prevented by treatment with cantharidin but not with okadaic acid. In conclusion, HP exerts PKC-independent protection on ischemic-reperfused rat hearts and cardiomyocytes. Protein phosphatase 1 seems a mediator of this protective mechanism.     

Role of gelsolin in actin depolymerization of adherent human neutrophils.

Human neutrophils generally function adherent to an extracellular matrix. We have previously reported that upon adhesion to laminin- or fibronectin-coated, but not uncoated, plastic there is a depolymerization of actin in neutrophils. This phenomenon was not affected by inhibitors of the more well-studied components of the signal transduction pathway, specifically, pertussis toxin, an inhibitor of G-proteins, H-7 or staurosporine, inhibitors of protein kinase C, or herbimycin A, an inhibitor of nonreceptor tyrosine kinase. We therefore focused our attention on actin-binding proteins and measured the changes in the partitioning of gelsolin between the Triton X-100-soluble and -insoluble cellular fractions which occur upon neutrophil adhesion by means of quantitating anti-gelsolin antibody binding to aliquots of these fractions. It was found that approximately 90% of the total cellular gelsolin was found in the Triton X-100-soluble fraction in suspended cells, but that upon adherence to either fibronectin- or laminin-coated plastic about 40% of the soluble gelsolin could be detected in the insoluble fraction. This effect was not observed in cells adherent to uncoated plastic, wherein more than 90% of the gelsolin was found in the soluble fraction. Results of immunofluorescence microscopy of these cell preparations was consistent with this data. A gelsolin translocation to the insoluble cellular actin network may account for a part of the observed actin depolymerization.     

Role of protein phosphatases in the cancer microenvironment

Cancer cells depend on a supportive niche (the tumor microenvironment) that promotes tumor cell survival while protecting the malignant cells from therapeutic challenges and the host's defense systems. Cancer cells and the support cells in the tumor microenvironment communicate via cytokines/chemokines, cell:cell contact, or alterations in the metabolic state of the niche (e.g. hypoxia) that promote growth and survival of the tumor cell, influence metastasis, and defeat immune surveillance. These signaling pathways involve dysregulation of not only protein kinases but also protein phosphatases as normal signal transduction processes require both activation and deactivation. For instance, aberrant receptor signaling can result from constitutive activation of a tyrosine kinase such as FLT3 or inactivation of a tyrosine protein phosphatase such as SHP-2 (PTPN11). Activation of serine/threonine kinases such as AKT and ERK are often observed during the development of drug resistance while genomic and non-genomic suppression of serine/threonine protein phosphatases such as PP2A achieve similar results. It is fairly clear that the various protein phosphatases will impact processes that support drug resistance. Of growing interest is the emerging model whereby the support cells in the tumor microenvironment actually serve as drivers of tumorigenesis. This phenomenon has been most prominently observed in osteoblast cells in leukemic niches. At least one protein phosphatase, PTPN11, has emerged as a critical driver of this process in juvenile myelomonocytic leukemia. This review will cover the role of various serine/threonine and tyrosine protein phosphatases in processes that are central to tumor microenvironment function.     

Role of the actin cytoskeleton during respiratory burst in chemoattractant-stimulated neutrophils

The aim of this study was to clarify the role of the actin cytoskeleton during chemotactic peptide fMet-Leu-Phe (fMLF)-stimulated respiratory burst in human neutrophil granulocytes. Reactive oxygen species (ROS) was measured as luminol-amplified chemiluminescence (CL) and F-actin content as bodipy phallacidin fluorescence in neutrophils treated with latrunculin B or jasplakinolide, an inhibitor and activator of actin polymerization, respectively. Latrunculin B markedly decreased, whereas jasplakinolide increased, the F-actin content in neutrophils, unstimulated or stimulated with fMLF. Latrunculin B enhanced the fMLF-triggered ROS-production more than tenfold. Jasplakinolide initially inhibited the fMLF-induced CL-response, however, caused a potent second sustained phase (>400% of control). Both actin drugs triggered a substantial CL-response when added 5-25 min after fMLF. This was also valid for chemotactic doses of fMLF, where latrunculin B and jasplakinolide amplified the ROS-production 5-10 times. By using specific signal transduction inhibitors, we found that the NADPH oxidase activation triggered by destabilization of the actin cytoskeleton occurs downstream of phospholipase C and protein kinase C but is mediated by Rho GTPases and tyrosine phosphorylation. In conclusion, rearrangements of the actin cytoskeleton are a prerequisite in connecting ligand/receptor activation, generation of second messengers and assembly of the NADPH oxidase in neutrophil granulocytes.     

Okadaic acid causes breakdown of self-incompatibility in Brassica oleracea: evidence for the involvement of protein phosphatases in the incompatible response

Summary Detached pistils from inbred lines of Brassica oleracea L. var alboglabra were fed with okadaic acid (OA), an inhibitor of serine/threonine protein phosphatases, via the transpiration stream. Following self-pollination, pollen tubes were observed to have grown into or through the styles of pistils treated with OA, but not those of untreated controls. Treatment with 1 M OA was sufficient to completely overcome self-incompatibility (SI) in an inbred line homozygous for the S63 allele, though an OA concentration of 5 M was required to cause breakdown of SI in an inbred line homozygous for the S29 allele. At the higher concentration used, pollen tube growth was arrested before the pollen tubes reached the ovary, but this effect was also noted in cross-pollinated styles treated in the same manner. These data provide evidence for the involvement of type 1 and/or type 2A protein phosphatases in the Brassica SI signal transduction mechanism. Present address after November 1993: Department of Biology, Colorado State University, Fort Collins, Colorado, USA     

Effects of antagonists of protein phosphatases on superoxide release by neutrophils.

Neutrophils stimulated with 4 beta-phorbol 12-myristate 13-acetate (PMA) release large quantities of superoxide (O2-) and exhibit phosphorylation of two proteins with molecular masses of 47(p47) and 49 kDa (p49). Addition of inhibitors of protein kinases (e.g. 1-(5-isoquinolinylsulfonyl)-2-methylpiperazine (H-7)) to these cells after stimulation with PMA results in the loss of 32P from these proteins and a rapid cessation of O2- release (e.g. Heyworth, P. G., and Badwey, J. A. (1990) Biochim. Biophys. Acta 1052, 299-305). In this paper we report that antagonists of type 1 and 2A protein phosphatases (okadaic acid, calyculin A) prevented both the loss of 32P from p47 and the termination of O2- release in stimulated neutrophils treated with H-7. Calyculin A also caused a remarkable hyperphosphorylation of a number of proteins in neutrophils and increased O2- release from these cells in response to a suboptimal amount of PMA. Enzymes present in both the soluble and particulate fractions of neutrophils catalyzed the near complete dephosphorylation of 32P-labeled p47 and p49 bound to Immobilon-P membranes. Dephosphorylation of these blotted phosphoproteins occurred at physiological rates and was inhibited by okadaic acid and calyculin A. These data strongly suggest that p47 undergoes a continual cycle of phosphorylation and dephosphorylation throughout the period of O2- release when PMA is the stimulus. Moreover, we show that antagonists of type 1 and 2A protein phosphatases block dephosphorylation of p47 both in vivo and in vitro, indicating that these enzymes may modulate O2- release under certain circumstances.     

Utility of immobilon-bound phosphoproteins as substrates for protein phosphatases from neutrophils.

Immobilon-bound phosphoproteins labeled with 32P were utilized as substrates to study the enzymes in neutrophils that are active against the major products of protein kinase C. The labeled proteins were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred electrophoretically to immobilon-P membranes. Both particulate and soluble phosphatases were found to be active against the blotted phosphoproteins. Reactions were followed by autoradiography as the loss of 32P from individual protein bands. The tumor promoter okadaic acid and the hepatoxin microcystin-LR inhibited these reactions in a manner consistent with the enzymes being type 1 and/or 2A protein phosphatases.     

Tec kinases regulate actin assembly and cytokine expression in LPS-stimulated human neutrophils via JNK activation

The acute inflammatory response involves neutrophils wherein recognition of bacterial products, such as lipopolysaccharide (LPS), activates intracellular signaling pathways. We have shown that the mitogen-activated protein kinase (MAPK) c-Jun NH₂ terminal kinase (JNK) is activated by LPS in neutrophils and plays a critical role in monocyte chemoattractant protein (MCP)-1 expression and actin assembly. As the Tec family kinases are expressed in neutrophils and regulate activation of the MAPKs in other cell systems, we hypothesized that the Tec kinases are an upstream component of the signaling pathway leading to LPS-induced MAPKs activation in neutrophils. Herein, we show that the Tec kinases are activated in LPS-stimulated human neutrophils and that inhibition of the Tec kinases, with leflunomide metabolite analog (LFM-A13), decreased LPS-induced JNK, but not p38, activity. Furthermore, LPS-induced actin polymerization as well as MCP-1, tumor necrosis factor-α, interleukin-6, and interleukin-1β expression are dependent on Tec kinase activity.     

Involvement of the conserved adaptor protein Alix in actin cytoskeleton assembly

The conserved adaptor protein Alix, also called AIP1 or Hp95, promotes flattening and alignment of cultured mammalian fibroblasts; however, the mechanism by which Alix regulates fibroblast morphology is not understood. Here we demonstrate that Alix in WI38 cells, which require Alix expression for maintaining typical fibroblast morphology, associates with filamentous actin (F-actin) and F-actin-based structures lamellipodia and stress fibers. Reducing Alix expression by small interfering RNA (siRNA) decreases F-actin content and inhibits stress fiber assembly. In cell-free systems, Alix directly interacts with F-actin at both the N-terminal Bro1 domain and the C-terminal proline-rich domain. In Alix immunoprecipitates from WI38 cell lysates, actin is the most abundant partner protein of Alix. In addition, the N-terminal half of the middle region of Alix binds cortactin, an activator of the ARP2/3 complex-mediated initiation of actin polymerization. Alix is required for lamellipodial localization of cortactin. The C-terminal half of the middle region of Alix interacts with alpha-actinin, a key factor that bundles F-actin in stress fibers. Alix knockdown decreases the amount of alpha-actinin that associates with F-actin. These findings establish crucial involvement of Alix in actin cytoskeleton assembly.     

Okadaic acid-induced transcriptional downregulation of type I collagen gene expression is mediated by protein phosphatase 2A

Summary In Saccharomyces cerevisiae, a small proportion of the glucose-6-P dehydrogenase activity is firmly associated with the mitochondrial fraction and is not removed by repeated washing or density-gradient centrifugation. However, the enzyme is released by sonic disruption. Mitochondrial glucose-6-P dehydrogenase that is released by sonication and partially purified has been found to be similar to cytosol glucose-6-P dehydrogenase with respect to electrophoretic mobility, isoelectric point, pH optimum, molecular size, and apparent K _m 's for NADP⁺ and glucose-6-P. These results indicate that a single species of glucose-6-P dehydrogenase is synthesized in S. cerevisiae and that the enzyme has more than one intracellular location. Mitochondrial glucose-6-P dehydrogenase may be a source of intramitochondrial NADPH and may function with hexokinase and transhydrogenase to provide a pathway for glucose oxidation that is coupled to the synthesis of mitochondrial ATP. A constant proportion of total glucose-6-P dehydrogenase activity remains compartmented in the mitochondrial fraction throughout the growth cycle.     

Role of gelsolin interaction with actin in regulation and creation of actin nuclei in chemotactic peptide activated polymorphonuclear neutrophils.

In vitro Ca++ activates gelsolin to sever F-actin and form a gelsolin-actin (GA) complex at the+end of F-actin that is not dissociated by ethylene glycol-bis(beta-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA) but is separated by EGTA+PIP/PIP2. The gelsolin blocks the+end on the actin filament, but the-end of the filament can still initiate actin polymerization. In thrombin activated platelets, evidence suggests that severing of F-actin by gelsolin increases GA complex, creates one-end actin nucleus and one cryptic+end actin nucleus per cut, and then dissociates to yield free+ends to nucleate rapid actin assembly. We examined the role of F-actin severing in creation and regulation of nuclei and polymerization in polymorphonuclear neutrophils (PMNs). At 2-s intervals after formyl peptide (FMLP) activation of endotoxin free (ETF) PMNs, change in GA complex was correlated with change in+end actin nuclei,-end actin nuclei, and F-actin content. GA complex was quantitated by electrophoretograms of proteins absorbed by antigelsolin from cells lysed in 10 mM EGTA,+end actin nuclei as cytochalasin (CD) sensitive and-end actin nuclei as CD insensitive increases in G-pyrenyl actin polymerization rates induced by the same PMNs, and F-actin content by NBDphallacidin binding to fixed cells. Thirty three percent of gelsolin was in GA complex in basal ETF PMNs; from 2-6 s, GA complexes dissociate (low = 15% at 10 s) and sequentially+end nuclei and F-actin content and then-end nuclei increase to a maximum at 10 s. At > s GA complex increase toward basal and + end nuclei and F-actin content returned toward basal. These kinetic data show gelsolin regulates availability of + end nuclei and actin polymerization in FMLP. However, absence of an initial increase in GA complex or - end nucleating activity shows FMLP activation does not cause gelsolin to sever F- or to bind G-actin to create cryptic + end nuclei in PMNs; the results suggest the + nucleus formation is gelsolin independent.     

Propionic acid-induced calcium mobilization in human neutrophils

The ability of propionic acid to elicit an increase in the level of cytoplasmic free calcium in human neutrophils was examined in detail. Propionic acid induced a rapid and dose-dependent mobilization of calcium that relied on both internal and external sources of calcium. The effects of propionic acid on the mobilization of calcium were inhibited by pertussis toxin, but not cholera toxin, implicating a guanine nucleotide binding protein. Furthermore, preincubation of the neutrophils with phorbol 12-myristate 13-acetate resulted in a decreased mobilization of calcium. This inhibitory activity of phorbol myristate acetate was antagonized by the protein kinase C inhibitor H-7. Preincubation of the cells with the synthetic chemotactic factor fMet-Leu-Phe caused a reduction in the magnitude of the calcium transient elicited by propionic acid. However, the calcium response to propionic acid was not affected by antagonists of fMet-Leu-Phe and platelet-activating factor binding or by an inhibitor of leukotriene synthesis. Propionic acid did not elicit a mobilization of calcium in monocytes, platelets, lymphocytes, or undifferentiated HL-60 cells. However, the treatment of the HL-60 cells with dimethylsulfoxide resulted in the appearance of a calcium response to propionic acid. The potential physiological significance of these findings are discussed.     

Role of protein phosphatases in regulation of cardiac inotropy and relaxation

We studied the effects of the protein phosphatase (PP) inhibitor cantharidin (Cant) on time parameters and force of contraction (FOC) in isometrically contracting electrically driven guinea pig papillary muscles. We correlated the mechanical parameters of contractility with phosphorylation of the inhibitory subunit of troponin (TnI-P) and with the site-specific phosphorylation of phospholamban (PLB) at serine-16 (PLB-Ser-16) and threonine-17 (PLB-Thr-17). Cant (after 30 min) started to increase FOC (112 +/- 4% of control, n = 10) and TnI-P and PLB-Thr-17 (120 +/- 5 and 128 +/- 7% of control) without any alteration of relaxation time. Cant (10 microM) started to increase PLB-Ser-16, but the relaxation was shortened at only 100 microM (from 140 +/- 9 to 116 +/- 12 ms, n = 9). Moreover, 100 microM Cant, 3 min after application, started to increase PLB-Thr-17, TnI-P, and FOC. Cant (100 microM) began to increase PLB-Ser-16 after 20 min. This was accompanied by shortening of relaxation time. Differences in protein kinase activation or different substrate specificities of PP may explain the difference in Cant-induced site-specific phosphorylation of PLB in isometrically contracting papillary muscles. Moreover, PLB-Thr-17 may be important for inotropy, whereas PLB-Ser-16 could be a major determinant of relaxation time.     

Cofilin phosphatases and regulation of actin dynamics

Cofilin is a ubiquitous actin-binding factor required for the reorganization of actin filaments in eukaryotes. The dephosphorylation of cofilin enables its actin severing and depolymerizing activity and drives directional cell motility, thus providing a simple phosphoregulatory mechanism for actin reorganization. To date, two cofilin-specific phosphatases have been identified: Slingshot and Chronophin. These cofilin phosphatases are unrelated in sequence and regulatory properties, each potentially providing a unique mechanism for cofilin activation under varying biological circumstances.     

Dia-interacting protein modulates formin-mediated actin assembly at the cell cortex

BACKGROUND: Mammalian Diaphanous (mDia)-related formins and the N-WASP-activated Arp2/3 complex initiate the assembly of filamentous actin. Dia-interacting protein (DIP) binds via its amino-terminal SH3 domain to the proline-rich formin homology 1 (FH1) domain of mDia1 and mDia2 and to the N-WASp proline-rich region. RESULTS: Here, we investigated an interaction between a conserved leucine-rich region (LRR) in DIP and the mDia FH2 domain that nucleates, processively elongates, and bundles actin filaments. DIP binding to mDia2 was regulated by the same Rho-GTPase-controlled autoinhibitory mechanism modulating formin-mediated actin assembly. DIP was previously shown to interact with and stimulate N-WASp-dependent branched filament assembly via Arp2/3. Despite direct binding to both mDia1 and mDia2 FH2 domains, DIP LRR inhibited only mDia2-dependent filament assembly and bundling in vitro. DIP expression interfered with filopodia formation, consistent with a role for mDia2 in assembly of these structures. After filopodia retraction into the cell body, DIP expression induced excessive nonapoptotic membrane blebbing, a physiological process involved in both cytokinesis and amoeboid cell movement. DIP-induced blebbing was dependent on mDia2 but did not require the activities of either mDia1 or Arp2/3. CONCLUSIONS: These observations point to a pivotal role for DIP in the control of nonbranched and branched actin-filament assembly that is mediated by Diaphanous-related formins and activators of Arp2/3, respectively. The ability of DIP to trigger blebbing also suggests a role for mDia2 in the assembly of cortical actin necessary for maintaining plasma-membrane integrity.     

Dynamics of capping protein and actin assembly in vitro: uncapping barbed ends by polyphosphoinositides

Bursts of actin polymerization in vivo involve the transient appearance of free barbed ends. To determine how rapidly barbed ends might appear and how long they might remain free in vivo, we studied the kinetics of capping protein, the major barbed end capper, binding to barbed ends in vitro. First, the off-rate constant for capping protein leaving a barbed end is slow, predicting a half-life for a capped barbed end of approximately 30 min. This half-life implies that cells cannot wait for capping protein to spontaneously dissociate from capped barbed ends in order to create free barbed ends. However, we find that phosphatidylinositol 4,5-bisphosphate (PIP2) and phosphatidylinositol 4- mono-phosphate (PIP) cause rapid and efficient dissociation of capping protein from capped filaments. PIP2 is a strong candidate for a second messenger regulating actin polymerization; therefore, the ability of PIP2 to remove capping protein from barbed ends is a potential mechanism for stimulating actin polymerization in vivo. Second, the on- rate constant for capping protein binding to free barbed ends predicts that actin filaments could grow to the length of filaments observed in vivo during one lifetime. Third, capping protein beta-subunit isoforms did not differ in their actin binding properties, even in tests with different actin isoforms. A major hypothesis for why capping protein beta-subunit isoforms exist is thereby excluded. Fourth, the proposed capping protein regulators, Hsc70 and S100, had no effect on capping protein binding to actin in vitro.