HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sullivan, R. Articles by Koffer, A. Search for Related Content PubMed PubMed Citation Articles by Sullivan, R. Articles by Koffer, A. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 53, 38140-38146, December 31, 1999

Rho Controls Cortical F-actin Disassembly in Addition to, but Independently of, Secretion in Mast Cells^*

Richard Sullivan, Leo S. Price^§, and Anna Koffer^¶

From the Physiology Department, University College London, University Street, London WC1E 6JJ, United Kingdom

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Localized disassembly of cortical F-actin has long been considered necessary for facilitation of exocytosis. Exposure of permeabilized mast cells to calcium/ATP induces cortical F-actin disassembly (calmodulin-dependent) and secretion (calmodulin-independent). The delay in the onset of secretion is characteristic for the calcium/ATP response and is abolished by GTP. Here we report that a constitutively active mutant of Rho (V14RhoA) enhanced both secretion and cortical F-actin disassembly. In addition, V14RhoA mimicked GTP by abolishing the delay in secretion. Inhibition of Rho by C3 transferase prevented both secretion (~80%) and F-actin disassembly (~20%). Thus, both Rho GTPase and calcium/calmodulin contribute to the control of cortical F-actin disassembly. Stabilization of actin filaments by high concentrations of phalloidin or by a calmodulin-inhibitory peptide (based on the calmodulin-binding domain of myosin light chain kinase) did not affect the extent of secretion or the secretion-enhancing effects of V14RhoA. These results further support the existence of divergent, Rho-dependent, pathways regulating actin and exocytosis. Furthermore, compound Y-27632, a specific inhibitor of Rho-associated protein kinase (p160^ROCK), attenuated the Rho-induced loss of cortical F-actin without affecting secretion. A model is presented in which Rho regulates secretion and cortical F-actin in a manner dependent on and/or synergistic with calcium.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The plasma membrane of eukaryotic cells is supported by an actin cortex, a network of actin filaments organized and regulated by a variety of actin-binding proteins. Changes in the rigidity of the cortical actin network constitute an important component of the cellular response to receptor activation and require localized disassembly of cortical actin filaments. Signals from cell surface receptors to the cytoskeleton are often mediated by small GTP-binding proteins of the Rho family, e.g. RhoA, -B, -C, -D, -E, and -G; Rac1 and -2; Cdc42; and TC10. Several targets for the members of this family have been identified, including a number of serine threonine kinases (PKN, Rho kinase, p65^PAK) and tyrosine kinases (p120^ACK, MLK3). In addition to cell morphology, Rho GTPases control many other functions such as neuronal development, cell cycle, and gene transcription (1-3). New evidence indicates participation of Rho GTPases in the regulation of exocytosis, endocytosis, and vesicular transport in general (4-9). To what extent this is accomplished via regulation of actin is still unclear.

Cortical actin disassembly is particularly important for secreting cells; cortical filaments have long been considered to act as a barrier preventing the access of secretory granules to the plasma membrane (10). Mast cells serve as a useful model for investigating the relationship between cytoskeletal rearrangements and exocytosis (11). Use of permeabilized cells allows control of the internal environment and thus investigation of specific effects induced by ions, nucleotides, or proteins. Such studies have established that calcium and GTP, acting in synergy, induce high levels of secretion (12). With respect to the cytoskeletal effects, GTPS,¹ a nonhydrolyzable analogue of GTP, was found to induce a selective loss and increase of F-actin in the cell cortex and interior, respectively. The former effect is not significantly inhibited by a Rho-specific inhibitor, C3 transferase, while the latter is entirely dependent on the activity of Rho (13). Secretion induced by calcium alone (i.e. in the absence of added GTP/GTPS) is associated only with a profound loss of the cortical F-actin (14). Thus, cortical F-actin disassembly is induced by both calcium-dependent and -independent pathways.

In this paper, we have focused on the calcium-dependent responses of permeabilized mast cells. Calcium-induced cortical F-actin disassembly requires calmodulin; it is promoted by the addition of exogenous calmodulin and strongly inhibited by depletion of the endogenous calmodulin. None of these effects, however, are accompanied by changes in the secretory responses. The main target for calmodulin is myosin light chain kinase; its inhibitor, ML-7, prevents cortical disassembly, again without effects on secretion.² Thus, myosin II-based contractility, activated by calcium-calmodulin, may be a prerequisite for cortical F-actin disassembly.

Calcium-induced secretion is greatly enhanced by constitutively active mutants of Rho-related GTPases, V14RhoA and V12Rac1, and inhibited by a specific Rho inhibitor, C3 transferase (4). It is relatively resistant to inhibition of Rac by a dominant-inhibitory Rac protein, N17Rac1 (4). Here, we have examined the role of Rho in calcium-dependent secretion in relation to its cytoskeletal effects. Rho was found to regulate both secretion and cortical F-actin disassembly in a manner that is synergistic with and/or dependent on calcium but is calmodulin-independent. The latter function of Rho is additional to and distinct from the Rho-induced de novo actin polymerization (which is calcium-independent). Stabilization of actin filaments by high concentration of phalloidin prevented neither secretion nor the secretion-enhancing effects of Rho, supporting the model for divergent regulation of actin and exocytosis by this GTPase. This model is further supported by a finding that compound Y-27632, a specific inhibitor of Rho-associated protein kinase, p160^ROCK, attenuated the loss of cortical F-actin without affecting secretion.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

GTPS and ATP were obtained from Roche Molecular Biochemicals. Streptolysin-O was from Murex Diagnostics (Dartford, UK). Glass "Multitest" slides (6-mm diameter wells) were from ICN Biomedicals (Aurora, OH). Compound Y-27632 was a generous gift from Yoshitomi Pharmaceutical Industries, Ltd. All other reagents were obtained from Sigma.

Proteins

Escherichia coli transfected with expression vectors containing GST-V14RhoA DNA, GST-V12Rac1 DNA or GST-C3 transferase DNA were kindly provided by Dr. Anne Ridley, and the glutathione S-transferase fusion proteins were prepared and cleaved as described (15) with minor modifications. V14RhoA was preactivated by loading with GTP (16). The activated protein was stabilized by the addition of MgCl₂ to give a final Mg²⁺ concentration of 2 mM, dialyzed, and concentrated using the Microsep 10-kDa cut-off centrifugal concentrators (Filtron, North Borough, MA). This final dialysate was used for control experiments. Active protein concentrations were determined by a filter binding assay using [³H]GDP (16). Concentrations of active proteins (usually 60-70% of total) are given for all of the experiments.

Calmodulin was purified from bovine brain according to the protocol in Ref. 17. The calmodulin-inhibitory peptide, MLCK peptide (MLCKP; Ac-RRKWQKTGHAVRAIGRL-CONH₂), and its control peptide (Ac-RRKEQKTGHAVRAIGRE-CONH₂) (17) were gifts from Dr. Kati Torok (School of Biological Sciences, Queen Mary and Westfield College, University of London, UK).

Cells

Rat peritoneal mast cells were prepared as described previously (18). The cells were resuspended in a solution containing 137 mM NaCl, 2.7 mM KCl, 1.0 mM CaCl₂, 2 mM MgCl₂, 5.6 mM glucose, 1 mg ml¹ bovine serum albumin, and 20 mM NaPIPES, pH 7.2 (chloride buffer). For confocal microscopy, cells in chloride buffer were allowed to attach to glass slides for 1 h at room temperature (~25,000 cells/well). Suspended cells were used for flow cytometry and for hexosaminidase assays.

Suspended cells (usually 1-2 × 10⁶ cells/ml) were permeabilized for 2 min at 30 °C in a buffer containing 137 mM sodium glutamate, 2 mM MgCl₂, 1 mg/ml bovine serum albumin, 20 mM NaPIPES, pH 6.8 (glutamate buffer (GB)) in the presence of 3 mM EGTA (GB-EGTA) and streptolysin-O (0.4 units/ml). To permeabilize glass-attached cells, the cells were washed with GB and then exposed for 8 min at room temperature to streptolysin-O at 0.4 units/ml in GB-EGTA. After permeabilization, both suspended and attached cells were washed free of soluble components and excess streptolysin-O with GB-EGTA. Cells were stimulated by the addition of the indicated combinations of calcium, EGTA, ATP, and GTPS. Free Ca²⁺ concentration was buffered by 3 mM Ca²⁺ EGTA buffer system, pH 6.8, with appropriate dissociation constants as given in Ref. 19. Control cells were exposed to 3 mM EGTA, 3 mM ATP in GB. Hexosaminidase release from at least duplicate samples was assayed as described previously (18).

Where indicated, attached or suspended rat peritoneal mast cells were permeabilized as above and pretreated with the specified agents (phalloidin, C3 together with NAD⁺, or V14RhoA in the GTP-bound form). While V14RhoA was left with the cells throughout the experiment, C3 together with 0.5 mM NAD⁺ was removed just prior to the addition of triggers.

Flow Cytometry

Identification of Secreting Cells-- Intact cells, suspended in chloride buffer, were treated with unconjugated succinyl-concanavalin A (100 µg/ml for 10 min at room temperature) to reduce the background staining. Cells were then washed, permeabilized as above, and exposed to 3 mM EGTA (control) or to the triggers (20 min at 30 °C). The incubation was stopped by the addition of cold GB, and the cells were pelletted by centrifugation (5 min at 1500 × g), resuspended in GB-EGTA containing fluorescein isothiocyanate-succinyl-concanavalin A (64 µg/ml). After 20 min on ice, the stain was washed off (twice with GB-EGTA), and the cells were finally resuspended (approximately 5 × 10⁵ ml¹) in GB-EGTA containing 0.5% formaldehyde (without any bovine serum albumin) and analyzed on an EPICS Elite flow cytometer (Coulter Electronics Inc., Hialeah, FL), equipped with an argon ion laser. Excitation was at 488 nm, and emission was at 525 nm. Forward angle (1.5-19°) and 90° light scatter and relative fluorescence intensity readings were collected. Freehand "gates" were drawn around subpopulations on two-parameter light scatter distributions using the Elite software, and the relative fluorescence intensity of cells falling within these gates was plotted on one parameter histograms.

Staining with Rhodamine Phalloidin-- Staining with rhodamine phalloidin (RP) was performed either before (prelabeling) or after (postlabeling) cell activation.

Prelabeling-- Where indicated, suspended mast cells were permeabilized and prelabeled with 0.1 µM RP in GB-EGTA (10 min on ice). Cells were then washed with GB and exposed to various triggering solutions for 20 min at 30 °C, quenched with cold GB, fixed, and analyzed as above.

Postlabeling-- This was performed after the triggering and the quenching of the cells but before their fixation: 0.5 µM RP in GB-EGTA, 20 min on ice.

Actin Filament Stabilization-- In some experiments (see Fig. 5B), cells were pretreated with 2 µM unlabeled phalloidin in order to stabilize actin filaments, and this was followed by postlabeling with 0.5 µM RP. After staining, cells were washed with GB-EGTA, resuspended (approximately 5 × 10⁵ ml¹) in GB-EGTA containing 0.5% formaldehyde (without any bovine serum albumin), and analyzed on an EPICS Elite flow cytometer (Coulter Electronics), equipped with an argon ion laser. Excitation was at 488 nm, and emission was at 575 nm. Fluorescence intensity of control cells (exposed to EGTA/ATP) was taken as 100%. Means obtained from triplicate samples, each containing at least 10,000 cells, are shown.

Confocal Microscopy and Image Analysis

Fluorescent images were obtained using a Leica confocal laser-scanning microscope attached to a Leitz Fluovert-FU microscope. Excitation was at 514 nm, and emission was collected with a >600-nm barrier filter. Digital images (equatorial optical slices) were displayed using the Leica "glowovun" look-up table. Each experiment (in duplicates or triplicates) was repeated, and the results were reproduced at least three times.

A quantitative estimate of F-actin distribution changes was obtained by radial line scan analysis of the images of the equatorial slices (13). Three radial scan lines were obtained per cell, each extending from the exterior to the center of the cell. To obtain the distribution plots, as displayed in Fig. 4, A and B, the lines obtained for each set of experimental conditions were averaged to form a mean profile ± S.E. (n > 50; the statistic is appropriate for the cell number and not the number of lines). For Fig. 4C, integration of areas under the traces from regions designated as cortical and internal (pixels 1-10 and 11-35, respectively, representing 0-2.45 and 2.45-8.575 µm from the cell edge) produced data that were then expressed in the form of a bar chart. Changes in the cortical and internal regions were calculated as percentages of the total F-actin content in both regions from appropriate control cells (the sum of intensities of pixels 1-35 for EGTA-treated cells). Thus, relative changes in distribution of F-actin induced by calcium/ATP and/or V14RhoA could be evaluated.

Statistics

All figures shown are representative of at least three separate experiments (levels of secretion vary greatly between individual experiments, but the effects are reproducible). Where error bars are included, these are S.E. Where indicated, data were analyzed by unpaired, one-tailed Student's t tests using Instat^® software. p values of <0.05 were taken as significant and those of <0.002 as extremely significant.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

After permeabilization, mast cells lose up to 70% of their total actin. The remaining F-actin is present mostly in the cell periphery, as a prominent subplasmalemmal network. Exposure of such permeabilized cells to calcium (pCa < 6) and ATP ([ATP] > 1 mM), induces a reduction of the cortical F-actin by 30-50%. Not all of the cells exposed to calcium/ATP are capable of the secretory response; two separate cell populations could be distinguished by flow cytometry (using light scatter characteristics and/or fluorescein isothiocyanate-succinyl-concanavalin A staining). Analysis of the cytoskeletal responses of these two populations revealed that the F-actin content of cells exposed to Ca/ATP is reduced in both secreting and nonsecreting cells (as reflected by the rhodamine-phalloidin fluorescence, not shown). Thus, F-actin loss occurs in both populations and is not sufficient to induce secretion. Calmodulin plays a crucial role in the calcium-induced loss of cortical F-actin; the latter is promoted by the addition of exogenous calmodulin and strongly inhibited by depletion of the endogenous calmodulin. None of these effects, however, are accompanied by any changes in the secretory responses, indicating that calmodulin is not required for the late steps in exocytosis.² The proportion of secreting cells is enhanced by the addition of constitutively active mutants of both Rho (V14RhoA) and Rac (V12Rac1) and inhibited by a Rho inhibitor, C3 transferase (4). The dominant inhibitory Rac protein, N17Rac1 does not have any significant effects on calcium-induced secretion (4). This paper investigates whether the Rho-induced enhancement of secretion is associated with any changes in the cytoskeletal responses.

Synergistic Effects of Calcium and Rho on Secretion-- Synergistic effects of calcium and GTP/GTPS on secretion from streptolysin-O-permeabilized rat peritoneal mast cells are well established (12). To investigate whether Rho mimics GTPS in this synergy, we have studied the effects of a constitutively active mutant of Rho, V14RhoA, and of a Rho inhibitor, C3 transferase, on the time course of and calcium requirements for secretion from suspended permeabilized cells. Fig. 1A shows that similarly to the effects of GTPS, V14RhoA eliminated the delay in the onset of secretion, characteristic for the secretory response of permeabilized mast cells to calcium (20). The extent of hexosaminidase release was also enhanced by both GTPS and V14RhoA. C3, on the other hand, was inhibitory. GTPS reduces calcium requirements of permeabilized mast cells for secretion, and again, similar effects were observed after the addition of V14RhoA (Fig. 1B). In contrast to GTPS, V14RhoA was unable to induce secretion in the absence of calcium.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   V14RhoA abolishes the delay in the onset of secretion and reduces calcium requirements. A, permeabilized mast cells were pretreated for 5 min on ice with control buffer, C3 (1 µg/ml) plus NAD+ (0.5 mM), or with V14RhoA (8 µg/ml) and then triggered with calcium (pCa 5) plus 3 mM ATP (, , ) or with calcium, 3 mM ATP, 30 µM GTPS () and incubated at 30 °C. Aliquots were removed at the indicated time intervals for assay of the released hexosaminidase. B, cells were treated as above and then triggered with the indicated concentrations of calcium. All figures are representative of three experiments. V14RhoA was loaded with GTP and dialyzed. The final dialysate was used for the control ().

Effects of Inhibiting Calmodulin and Rho-- To test whether the effects of exogenous calmodulin require or are affected by the activity of Rho, we have pretreated permeabilized mast cells with C3 transferase prior to the addition of the exogenous calmodulin. The results are shown in Fig. 2. The C3 pretreatment inhibited both F-actin disassembly (Fig. 2A) and secretion (Fig. 2B). The sensitivity of the secretory response to C3 was, however, much greater than that of the cytoskeletal response. Relative to the responses of control cells, C3 (at 1 µg/ml) caused 20 ± 3% inhibition of F-actin disassembly and 83 ± 5% inhibition of secretion (n = 9). On the other hand, exogenous calmodulin (at 12 µM, a concentration that approximates its cytosolic level) had a marked effect on F-actin (Fig. 2A) without affecting secretion (Fig. 2B). Calcium/calmodulin-induced F-actin disassembly was still inhibited by C3, a small but a significant effect (Fig. 2A).

View larger version (39K): [in this window] [in a new window]   Fig. 2.   A and B, the effects of calmodulin are apparent in the presence of C3, a Rho inhibitor. Permeabilized mast cells in suspension were incubated with or without C3 (1 µg/ml) plus 0.5 mM NAD+ (5 min on ice). Cells were further preincubated with or without calmodulin (12 µM, 5 min on ice) and subsequently exposed to 3 mM EGTA plus 3 mM ATP or to calcium (pCa5) plus 3 mM ATP. After 20 min at 30 °C, aliquots were taken for hexosaminidase release (B), and cells were stained (postlabeled) with 0.5 µM RP, fixed, and analyzed by flow cytometry for relative F-actin content (A). *, p = 0.05; **, p = 0.005, both significant. C and D, the effects of Rho are apparent in the presence of MLCKP, a calmodulin inhibitor. Permeabilized mast cells in suspension were preincubated (5 min on ice) with calcium (pCa 6.5) and MLCKP (10 µM). Control cells were treated with the control peptide (see "Experimental Procedures"). Cells were further preincubated with or without V14RhoA (8 µg/ml final concentration, 5 min at room temperature) and then exposed to 3 mM EGTA plus 3 mM ATP or to calcium (pCa 5) plus 3 mM ATP. After 20 min at 30 °C, secretion (D) and F-actin content (C) were analyzed as above. #, p = 0.002, extremely significant.

To test whether the Rho-induced enhancement of both calcium-dependent secretion and F-actin disassembly require the presence of calmodulin, we have used permeabilized cells depleted of this protein. Cells were pretreated with a calmodulin-inhibitory peptide (MLCKP), a specific 17-residue peptide based on the calmodulin-binding domain of myosin light chain kinase (17). Such treatment depletes most of the endogenous calmodulin from the permeabilized cells.² Responses of these calmodulin-depleted cells to calcium/ATP are shown in Fig. 2, C and D. MLCKP treatment prevented F-actin disassembly (Fig. 2C) but not secretion (Fig. 2D). Calmodulin-depleted cells were still responsive to the addition of V14RhoA (8 µg/ml); in both control and MLCKP-treated cells, V14RhoA enhanced calcium-induced secretion (Fig. 2D) as well as F-actin loss (Fig. 2C). V14RhoA-induced enhancement of F-actin disassembly was, however, much smaller than that induced by calmodulin (compare Fig. 2, compare A and C).

Synergistic Effects of Calcium and Rho on Cortical F-actin Disassembly-- Previous studies of mast cells exposed to GTPS/EGTA did not reveal any involvement of Rho in regulation of cortical F-actin disassembly. Calcium-independent (i.e. GTPS/EGTA-induced) cortical F-actin disassembly was neither inhibited by C3 transferase nor stimulated by V14RhoA (13). The above data suggest that Rho may control cortical filaments in a calcium-dependent manner. To further examine the effects of Rho on F-actin morphology, we have used confocal microscopy. Fig. 3 shows confocal micrographs of glass-attached permeabilized cells stained with RP after their exposure to EGTA or calcium (postlabeled). Fig. 4, A and B, show line scan analyses of such images. Control cells (exposed to EGTA) exhibit prominent cortical F-actin (Fig. 3A). Figs. 3B and 4B show the reduction of the cortical F-actin in cells exposed to calcium; the effect of C3 transferase (inhibition of this loss of F-actin) is shown in Fig. 3D. In the absence of calcium, V14RhoA induced de novo actin polymerization in the cell interior and only a very small reduction in the cortical F-actin (Figs. 3E and 4A). In the presence of calcium, however, Rho enhanced cortical disassembly in addition to its effect on actin polymerization in the cell interior. This is shown in Figs. 3F and 4B. Fig. 4C shows a quantitative evaluation of the V14RhoA-induced changes in distribution of F-actin, expressed in the form of a bar chart. Relative changes in the levels of F-actin in the cortical (full bars) and the interior (hollow bars) regions are expressed as percentages of the total F-actin content in both regions of the control cells. The effects of V14RhoA on cortical F-actin disassembly are thus synergistic with or dependent upon calcium, while those on actin polymerization are calcium-independent.

View larger version (46K): [in this window] [in a new window]   Fig. 3.   Effects of C3 and V14RhoA on F-actin morphology. Glass-attached mast cells were permeabilized and pretreated for 5 min on ice with control buffer (A and B), 1 µg/ml C3 plus 0.5 mM NAD+ (C and D), or with V14RhoA (8 µg/ml) (E and F). Cells were then stimulated with 3 mM EGTA plus 3 mM ATP (A, C, and E) or calcium (pCa 5) plus 3 mM ATP (B, D, and F) for 20 min at 30 °C, fixed, and stained (postlabeled) with 0.5 µM rhodamine phalloidin to visualize F-actin. Bar, 20 µm.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Changes in the F-actin distribution induced by V14RhoA in attached permeabilized mast cells. A and B, glass-attached mast cells were permeabilized, pretreated with () or without ( and ) V14RhoA as described for Fig. 3, E and F. Cells were then exposed to 3 mM EGTA plus 3 mM ATP (EA) or to calcium (pCa 5) plus 3 mM ATP (CA) as indicated. Densitometric line scan analyses were performed on confocal images (equatorial slices) of the RP-stained cells (postlabeled). Pixel distance is 0.245 µM. The intensity profiles of these lines were pooled to form a mean profile ± S.E. n = 50 cells. C, relative changes in distribution of F-actin induced by V14RhoA. Changes in the levels of F-actin in the cortical (solid bars) and the interior (open bars) regions expressed as percentages of the total F-actin content in both regions from control cells.

Synergistic effects of V14RhoA and calcium on cortical F-actin disassembly were confirmed by experiments with suspended cells, using flow cytometry (Fig. 5A). For this experiment, cells were prelabeled with a low concentration of RP (0.1 µM) to determine the F-actin loss. This protocol does not interfere with the cytoskeletal responses of activated cells and avoids detection of any new filaments polymerized after cell activation (13). The addition of V14RhoA to permeabilized mast cells prior to and during their exposure to Ca/ATP significantly enhanced the extent of cortical F-actin disassembly (left) as well as that of secretion (right). On the other hand, V14RhoA caused neither disassembly nor secretion in the absence of calcium (i.e. in 3 mM EGTA, not shown).

View larger version (28K): [in this window] [in a new window]   Fig. 5.   A, Rho enhances calcium-induced cortical F-actin disassembly and secretion. Permeabilized mast cells were suspended in GB and prelabeled for 5 min on ice with 0.1 µM RP, followed by a further pretreatment with or without V14RhoA (8 µg/ml). Cells were then exposed to 3 mM EGTA plus 3 mM ATP (open bars) or to calcium (pCa 5) plus 3 mM ATP (solid bars) and incubated for 20 min at 30 °C. Relative F-actin content was determined by flow cytometry (left panel), and secretion was determined by hexosaminidase assay (right panel). B, Rho induces de novo actin polymerization and enhances secretion in cells with stabilized actin cortex. Cells were pretreated as above with 2 µM unlabeled phalloidin with or without V14RhoA (8 µg/ml) and then exposed to 3 mM EGTA plus 3 mM ATP or to calcium (pCa 5) plus 3 mM ATP (20 min at 30 °C). Cells were then postlabeled with 0.5 µM RP and fixed. Secretion and F-actin were determined as above. Secretion from the cells with stabilized actin was the same as that from control cells (not pretreated with phalloidin).

Stabilization of Cortical F-actin by Phalloidin Does Not Prevent Secretion-enhancing Effects of Rho-- Experiments with permeabilized mast cells, pretreated with increasing concentrations of unlabeled phalloidin, have shown that calcium-induced cortical actin disassembly is prevented at concentrations of phalloidin >1.5 µM. Secretion, however, was unaffected by these higher concentrations of phalloidin (not shown). We have used a similar protocol to investigate whether the secretion-enhancing effects of V14Rho are dependent on its disassembly-enhancing effects (Fig. 5B). Cells were pretreated with 2 µM (unlabeled) phalloidin to stabilize the filaments and to prevent calcium-induced disassembly. After the exposure to calcium, cells were stained with RP to reveal any newly polymerized filaments. It is clear that F-actin cortex stabilization did not affect the ability of V14RhoA to enhance secretion (Fig. 5B, right). The increase in F-actin content, apparent after the addition of V14RhoA, is the result of V14RhoA-induced de novo polymerization of actin (left).

p160^ROCK Inhibitor Attenuates Rho-induced Enhancement of Cortical F-actin Disassembly-- One of the downstream targets of Rho is a Rho-associated protein kinase, p160^ROCK. This enzyme was found to phosphorylate and so inactivate the myosin-binding subunit of myosin light chain phosphatase, leading to an increase in the level of phosphorylated myosin II light chain (MLC) (21). To test whether cortical F-actin disassembly is regulated by Rho as a result of Rho controlling actomyosin-based contractility, we have used compound Y-27632, a specific p160^ROCK inhibitor (22). Fig. 6A shows that Y-27632 had no effect on calcium-induced secretion and did not prevent its enhancement by V14RhoA. On the other hand, the Y compound had a marginal inhibitory effect on the calcium-induced cortical F-actin disassembly and caused a significant reduction of its V14RhoA-induced enhancement (Fig. 6B). Compound Y-27632 had no effects on de novo actin polymerization (not shown).

View larger version (39K): [in this window] [in a new window]   Fig. 6.   Y-27632 inhibits Rho-induced enhancement of cortical actin disassembly. Suspended mast cells were permeabilized, washed, and prelabeled with 0.1 µM RP for 5 min on ice. After washing, cells were further incubated for 5 min on ice with control buffer (con), 1 µg·ml1 C3 and 0.5 mM NAD+ (+C3), 30 µM compound Y-27632 (+Y), 8 µg·ml1 V14RhoA (+Rho), or 30 µM compound Y-27632 for 3 min followed by 8 µg·ml1 V14RhoA for 2 min (+Y/+Rho). Cells were then triggered with calcium (pCa 5) plus 3 mM ATP for 20 min at 30 °C, keeping the concentrations of added agents constant. Supernatant aliquots were taken for determination of released hexosaminidase (A), and cells were analyzed by flow cytometry for relative F-actin content (B). Basal cells (exposed to 3 mM EGTA plus 3 mM ATP) released 4% hexosaminidase, and their F-actin content was taken as 100%. The Y-compound, C3, and V14RhoA had no effect on secretion or total F-actin content of basal cells (i.e. in 3 mM EGTA plus 3 mM MgATP, not shown). **, p = 0.0002 (extremely significant).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Calcium-independent Effects of Rho: De Novo Actin Polymerization-- In the absence of calcium, GTPS induces low levels of secretion that can be partially inhibited by a Rho inhibitor, C3 transferase, and only slightly stimulated by the addition of the constitutively active mutant of Rho, V14RhoA (4). The addition of V14RhoA/EGTA to permeabilized mast cells does not, however, induce secretion (Fig. 1), indicating that in the absence of calcium, activation of Rho is not a sufficient trigger.

GTPS/EGTA also induces two cytoskeletal effects: de novo actin polymerization and cortical F-actin disassembly (4, 13). The former effect is completely inhibited by C3. Moreover, V14RhoA mimics this effect, inducing actin polymerization independently of calcium (Figs. 3E and 4A). Cytochalasin inhibits this localized increase in F-actin without affecting secretion (23). Although there is a clear correlation between actin polymerization and secretion, their control by Rho appears to proceed via two divergent signaling pathways. Since most of the soluble monomeric actin leaks out from the permeabilized cells, Rho seems to control recruitment of actin from a membrane-bound pool of actin monomers. Our evidence for the Rho-induced polymerization of actin contrasts with that of Machesky and Hall (24). These authors concluded that Rho bundles actin filaments but does not induce polymerization; monomeric Cy3-labeled actin, microinjected into Swiss 3T3 fibroblasts, was incorporated into lamellipodia, induced by activation of Rac, but not into stress fibers formed after Rho activation. This discrepancy could be explained by the existence of the suggested membrane-bound pool of monomeric actin from which Rho recruits monomers and in which Cy3-actin would not be included.

GTPS/EGTA-induced cortical F-actin disassembly is not significantly affected by C3 and thus seems to be independent of Rho activity. V14RhoA/EGTA does not induce F-actin loss (Figs. 3E and 4A). This effect of GTPS can be mimicked by AlF₄, an activator of heterotrimeric G-proteins (13). Our preliminary results indicate that GTPS/EGTA induces cortical F-actin disassembly via a G_i protein; the loss of F-actin is inhibited by pertussis toxin.³

Calcium-dependent Effects of Rho on Secretion-- Calcium/ATP-induced secretion is strongly inhibited by C3 transferase, implicating Rho in this response (4). While in the absence of calcium, V14RhoA was unable to mimic the effects of GTPS on secretion, it had a strong secretion-enhancing effect in its presence (Figs. 1, 2D, 5, and 6A). V14RhoA reduced the requirements of permeabilized cells for calcium (Fig. 1B) and abolished the delay in the onset of calcium-induced secretion. The rate-limiting step of calcium-induced secretion may therefore be a calcium-induced activation of the endogenous Rho. This could occur via calcium-sensitive Rho-GTPase-activating proteins (such as IQGAP (25, 26) or class IX myosins (27)) or via calcium-sensitive guanine nucleotide exchange factors (such as Vav (28) or those similar to guanine nucleotide-releasing factor for Ras, Ras-GRF (29)) (see X in the model, Fig. 7). Alternatively, calcium and ATP could work together to maintain GTP levels by transphosphorylating the endogenous GDP (as discussed in Ref. 20) and also included under X in the model, Fig. 7). Another explanation of the synergy with or the dependence upon calcium could be the existence of parallel signaling pathways from Rho and calcium that converge and act on a common target (see Y in the model, Fig. 7).

View larger version (22K): [in this window] [in a new window]   Fig. 7.   The model. Rho-related GTPases play a key role in mast cell exocytosis, converging onto a putative component Y, which may control fusion of the plasma and granular membranes (independently of calmodulin). Rho is activated by a calcium-sensitive protein X, also in a calmodulin-independent manner. Protein X may be the calcium-binding protein, Ce, previously postulated to activate a GTPase required for exocytosis, Ge (35). Calcium may also activate Y independently of Rho, but in a synergistic manner. Both calmodulin/MLCK and Rho/Rho kinase induce actomyosin II-based contractility, which is crucial for disassembly of the actin cortex but not for secretion. Disassembly of the contracting cortex is aided by calcium-dependent activation of filament-severing proteins (14) and inhibition of filament-cross-linkers (36). In addition, Rho and Rac (independently of secretion) promote an increase in F-actin within the cell interior: Rho by inducing de novo actin polymerization and Rac by retaining the disassembling cortical filaments (13).

Calcium-dependent Effects of Rho on Cortical F-actin Disassembly-- In addition to controlling secretion and actin polymerization, Rho was found to have an additional role, that of promoting cortical F-actin disassembly. This role became apparent only in the presence of calcium: C3 transferase inhibited and V14RhoA promoted calcium-induced F-actin loss. The control of F-actin disassembly by Rho is, however, only partial. At 1 µg/ml, C3 inhibited secretion by ~80%, while the disassembly was inhibited by only 20%. Another, Rho-independent, pathway to calcium-induced cortical F-actin disassembly must therefore exist. Our recent data point to the MLCK pathway: calcium-induced F-actin loss is greatly promoted by exogenous calmodulin and strongly inhibited by 1) calmodulin antagonists; 2) ML-7, a specific inhibitor of MLCK; and 3) 2,3-butanedione 2-monoxime, an uncompetitive, low affinity inhibitor of myosin MgATPase).² Rho and calmodulin seem to act independently of each other, since the calmodulin effect on cortical disassembly was still apparent in the presence of C3 transferase (Fig. 2A). Moreover, V14RhoA could induce a small but significant decrease in F-actin content in calmodulin-depleted cells (Fig. 2C).

Rho and calcium/calmodulin may control cortical actin disassembly by pathways that converge on myosin II light chain (see the model, Fig. 7). One of the targets of Rho, the Rho-associated protein kinase p160^ROCK, is involved in the control of myosin II light chain phosphorylation. This kinase phosphorylates the myosin-binding subunit of myosin light chain phosphatase, decreasing its activity (21). In addition, the myosin-binding subunit of MLC phosphatase binds to Rho directly (21). A further level of complexity is added by the finding that Rho kinase also phosphorylates MLC on Ser-19, the same site as that phosphorylated by MLCK (30). Compound Y-27632, a specific p160^ROCK inhibitor (22), had a negligible effect on calcium-induced F-actin disassembly but significantly reduced its enhancement by V14RhoA (Fig. 6B). This suggests that p160^ROCK participates in the regulation of actin cortex, but other targets of Rho may also be involved. Taken together, our data indicate that cortical F-actin disassembly is dependent on actomyosin-based contractility. Both calmodulin and Rho may induce an increase in MLC phosphorylation, leading to an increase in myosin activity and contractility. Calmodulin/MLCK-dependent, rather than Rho-dependent process, however, constitutes the major pathway to F-actin disassembly (see Fig. 7).

Relevance of the Cortical F-actin Disassembly to Secretion-- The relevance of the cortical F-actin disassembly to secretion is still somewhat controversial. Prevention of F-actin disassembly by phalloidin does not affect either the secretion or secretion-enhancing effects of Rho (Fig. 5B). Moreover, agents other than phalloidin (e.g. inhibitors of calmodulin and MLCK) also stabilize cortical filaments without affecting secretion (Fig. 2D).² F-actin disassembly itself is not sufficient for secretion to occur: it is induced by calcium/ATP in both secreting and nonsecreting cell populations. An additional (presumably Rho/Rac-dependent) signal must, therefore, be necessary to initiate exocytosis. Thus, our results support a model with divergent signaling pathways to the cytoskeleton and to the late steps of exocytosis (23). It is conceivable, however, that a great part of the observed cytoskeletal changes is physiologically relevant for responses of intact cells exposed to concentration gradients of stimuli: under such conditions, cell polarization and motility may play a crucial role before exocytosis is initiated. Moreover, stabilization of F-actin by jasplakinolide, a cell-permeant actin-specific agent (31), inhibits secretion in intact, but not in permeabilized, cells.⁴ This suggests actin involvement in the receptor activation-secretion coupling.

In permeabilized cells, manipulation of the actin cytoskeleton by actin-binding proteins such as thymosin (32) or scinderin (33) induces changes in secretory responses. In mast cells, late steps of exocytosis were affected as a consequence of filament severing by gelsolin (14). Whether this effect is related to the finding that gelsolin is a downstream effector of Rac (34) remains to be established. Different specific pools of actin may be involved, controlling distinct steps of exocytosis. For example, a pool of filaments, closely associated with the plasma membrane and severed by exogenous gelsolin, may constitute a barrier to secretion. In contrast, those filaments that are more peripheral to the plasma membrane and undergo disassembly as a consequence of contraction do not appear to interfere with secretion. Our results are compatible with a model in which myosin-based contractility, activated by Rho and/or calcium/calmodulin, is required for the cortical F-actin disassembly, but this type of disassembly is not required for the late steps of exocytosis. Fig. 7 summarizes the evidence and shows our current model.

	ACKNOWLEDGEMENTS

We thank Dr. Kati Torok for providing the calmodulin-inhibitory peptide and Yoshitomi Pharmaceutical Industries, Ltd. for supplying the compound Y-27632. We especially thank Dr. Anne Ridley for providing the E. coli, transfected with V14RhoA and C3 transferase, and for helpful discussions. We also acknowledge Prof. S. Bolsover who kindly allowed the use of his confocal microscope, purchased with funds from the Wellcome Trust; Arnold Pitzey for generous help with the EPICS Elite flow cytometer; and Prof. J. Judah for support and encouragement.

	FOOTNOTES

^* This work was supported by grants from the Medical Research Council, the Wellcome Trust, the National Asthma Campaign, and the Mason Medical Research Foundation.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Present address: Merck (UK) Research & Development, Harrier House, West Drayton, Middlesex UB7 7QG, United Kingdom.

^§ Present address: Div. of Cell Biology, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands.

^¶ To whom correspondence should be addressed: Physiology Dept., University College London, University St., London WC1E 6JJ, United Kingdom. Tel.: 44-171-2096094; Fax: 44-171-3876368; E-mail: a.koffer@ucl.ac.uk.

² R. Sullivan, M. Burnham, and A. Koffer, submitted for publication.

³ R. Sullivan and A. Koffer, unpublished observations.

⁴ R. Sullivan and A. Koffer, unpublished results.

	ABBREVIATIONS

The abbreviations used are: GTPS, guanosine 5'-3-O-(thio)triphosphate; MLC, myosin light chain; MLCK, myosin light chain kinase; MLCKP, MLCK peptide; piperazine-N, N'-bis(2-ethanesulfonic acid; GB, glutamate buffer; RP, rhodamine phalloidin.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				S. W. Moore, J. P. Correia, K. Lai Wing Sun, M. Pool, A. E. Fournier, and T. E. Kennedy Rho inhibition recruits DCC to the neuronal plasma membrane and enhances axon chemoattraction to netrin 1 Development, September 1, 2008; 135(17): 2855 - 2864. [Abstract] [Full Text] [PDF]

				Y. Bi and J. A. Williams A role for Rho and Rac in secretagogue-induced amylase release by pancreatic acini Am J Physiol Cell Physiol, July 1, 2005; 289(1): C22 - C32. [Abstract] [Full Text] [PDF]

				T. Oka, M. Hori, and H. Ozaki Microtubule Disruption Suppresses Allergic Response through the Inhibition of Calcium Influx in the Mast Cell Degranulation Pathway J. Immunol., April 15, 2005; 174(8): 4584 - 4589. [Abstract] [Full Text] [PDF]

				J. Li, K. L. O'Connor, M. R. Hellmich, G. H. Greeley Jr., C. M. Townsend Jr., and B. M. Evers The Role of Protein Kinase D in Neurotensin Secretion Mediated by Protein Kinase C-{alpha}/-{delta} and Rho/Rho Kinase J. Biol. Chem., July 2, 2004; 279(27): 28466 - 28474. [Abstract] [Full Text] [PDF]

				M. L. Vitale and M. E. Carbajal Involvement of Myosin II in Dopamine-induced Reorganization of the Lactotroph Cell's Actin Cytoskeleton J. Histochem. Cytochem., April 1, 2004; 52(4): 517 - 527. [Abstract] [Full Text] [PDF]

				T. Oka, M. Hori, A. Tanaka, H. Matsuda, H. Karaki, and H. Ozaki IgE alone-induced actin assembly modifies calcium signaling and degranulation in RBL-2H3 mast cells Am J Physiol Cell Physiol, February 1, 2004; 286(2): C256 - C263. [Abstract] [Full Text]

				S. Gasman, S. Chasserot-Golaz, M. Malacombe, M. Way, and M.-F. Bader Regulated Exocytosis in Neuroendocrine Cells: A Role for Subplasmalemmal Cdc42/N-WASP-induced Actin Filaments Mol. Biol. Cell, February 1, 2004; 15(2): 520 - 531. [Abstract] [Full Text] [PDF]

				E. Lecuona, K. Ridge, L. Pesce, D. Batlle, and J. I. Sznajder The GTP-binding Protein RhoA Mediates Na,K-ATPase Exocytosis in Alveolar Epithelial Cells Mol. Biol. Cell, September 1, 2003; 14(9): 3888 - 3897. [Abstract] [Full Text] [PDF]

				A. Pendleton, B. Pope, A. Weeds, and A. Koffer Latrunculin B or ATP Depletion Induces Cofilin-dependent Translocation of Actin into Nuclei of Mast Cells J. Biol. Chem., April 11, 2003; 278(16): 14394 - 14400. [Abstract] [Full Text] [PDF]

				S. W. Klarenbach, A. Chipiuk, R. C. Nelson, M. D. Hollenberg, and A. G. Murray Differential Actions of PAR2 and PAR1 in Stimulating Human Endothelial Cell Exocytosis and Permeability: The Role of Rho-GTPases Circ. Res., February 21, 2003; 92(3): 272 - 278. [Abstract] [Full Text] [PDF]

				K. W. Kang, S. J. Lee, J. W. Park, and S. G. Kim Phosphatidylinositol 3-Kinase Regulates Nuclear Translocation of NF-E2-Related Factor 2 through Actin Rearrangement in Response to Oxidative Stress Mol. Pharmacol., November 1, 2002; 62(5): 1001 - 1010. [Abstract] [Full Text] [PDF]

				H. van Leeuwen, G. Elliott, and P. O'Hare Evidence of a Role for Nonmuscle Myosin II in Herpes Simplex Virus Type 1 Egress J. Virol., March 7, 2002; 76(7): 3471 - 3481. [Abstract] [Full Text] [PDF]

				T. Adachi, R. Vita, S. Sannohe, S. Stafford, R. Alam, H. Kayaba, and J. Chihara The Functional Role of Rho and Rho-Associated Coiled-Coil Forming Protein Kinase in Eotaxin Signaling of Eosinophils J. Immunol., October 15, 2001; 167(8): 4609 - 4615. [Abstract] [Full Text] [PDF]

				T. Ishizuka, F. Okajima, M. Ishiwara, K. Iizuka, I. Ichimonji, T. Kawata, H. Tsukagoshi, K. Dobashi, T. Nakazawa, and M. Mori Sensitized Mast Cells Migrate Toward the Agen: A Response Regulated by p38 Mitogen-Activated Protein Kinase and Rho-Associated Coiled-Coil-Forming Protein Kinase J. Immunol., August 15, 2001; 167(4): 2298 - 2304. [Abstract] [Full Text] [PDF]

				T. Togo, T. B. Krasieva, and R. A. Steinhardt A Decrease in Membrane Tension Precedes Successful Cell-Membrane Repair Mol. Biol. Cell, December 1, 2000; 11(12): 4339 - 4346. [Abstract] [Full Text]

				M. Glogauer, J. Hartwig, and T. Stossel Two Pathways Through Cdc42 Couple the N-formyl Receptor to Actin Nucleation in Permeabilized Human Neutrophils J. Cell Biol., August 21, 2000; 150(4): 785 - 796. [Abstract] [Full Text] [PDF]

				D. B. Buxton and R. S. Adelstein Calcium-dependent Threonine Phosphorylation of Nonmuscle Myosin in Stimulated RBL-2H3 Mast Cells J. Biol. Chem., October 27, 2000; 275(44): 34772 - 34779. [Abstract] [Full Text] [PDF]

				E. Ales, N. H. Gabilan, M. F. Cano-Abad, A. G. Garcia, and M. G. Lopez The Sea Anemone Toxin Bc2 Induces Continuous or Transient Exocytosis, in the Presence of Sustained Levels of High Cytosolic Ca2+ in Chromaffin Cells J. Biol. Chem., November 22, 2000; 275(48): 37488 - 37495. [Abstract] [Full Text] [PDF]

				E. Klussmann, G. Tamma, D. Lorenz, B. Wiesner, K. Maric, F. Hofmann, K. Aktories, G. Valenti, and W. Rosenthal An Inhibitory Role of Rho in the Vasopressin-mediated Translocation of Aquaporin-2 into Cell Membranes of Renal Principal Cells J. Biol. Chem., June 1, 2001; 276(23): 20451 - 20457. [Abstract] [Full Text] [PDF]

				G. Tamma, E. Klussmann, K. Maric, K. Aktories, M. Svelto, W. Rosenthal, and G. Valenti Rho inhibits cAMP-induced translocation of aquaporin-2 into the apical membrane of renal cells Am J Physiol Renal Physiol, December 1, 2001; 281(6): F1092 - F1101. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sullivan, R. Articles by Koffer, A. Search for Related Content PubMed PubMed Citation Articles by Sullivan, R. Articles by Koffer, A. Social Bookmarking                      What's this?