Journal of Leukocyte Biology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published online as doi:10.1189/jlb.1001894 on May 22, 2003

Published online before print May 22, 2003
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
jlb.1001894v1
73/6/830    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ratner, S.
Right arrow Articles by Galy, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ratner, S.
Right arrow Articles by Galy, A.
(Journal of Leukocyte Biology. 2003;73:830-840.)
© 2003 by Society for Leukocyte Biology

Role of Rho-family GTPase Cdc42 in polarized expression of lymphocyte appendages

Stuart Ratner*,{dagger}, Marie P. Piechocki* and Anne Galy{ddagger}

* Breast Cancer Research Program and
{ddagger} Stem Cell Transplantation Program, Barbara Ann Karmanos Cancer Institute, and
{dagger} Department of Immunology and Microbiology, Wayne State University School of Medicine, Detroit, Michigan

Correspondence: Dr. Stuart Ratner, Breast Cancer Research Program, Barbara Ann Karmanos Cancer Institute, 110 E. Warren Ave., Detroit, MI 48201. E-mail: ratners{at}kci.wayne.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Lymphocytes polarize for motility by developing a broad anterior, where lamellipodia arise, and a simple stalk-like posterior appendage, the uropod. Through time-lapse analysis of normal and leukemic human T cells, it was found that this polarized form is maintained by a mechanism that excludes lamellipodia from the uropod. Lamellipodia regularly traveled rearward to encroach upon the uropod but disassembled abruptly at the uropod border. This exclusion of lamellipodia from the uropod required the Rho-family guanosine triphosphatase Cdc42. Reduction of Cdc42 activity by expression of dominant-negative Cdc42 resulted in "two headed" cells in which lamellipodia persisted at the distal end of the uropod. Random and chemotactic motility were impaired. Increased Cdc42 activity, induced by expression of activated, mutant Cdc42, was accompanied by a general loss of lamellipodia. The results suggest that one role of Cdc42 in lymphocyte motility is to preserve polarity by concentrating lamellipodial disassembly signals in the uropod.

Key Words: human • T cell • leukemia • motility • chemotaxis • cytoskeleton


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Lymphocytes undergo a metamorphosis as they extravasate into inflamed tissue. In response to appropriate combinations of cytokines and adhesion ligands, they change from a spherical, nonadhesive form adapted for transport through vasculature into a bipolar, motile form adapted for rapid migration through stroma and parenchyma [1 2 3 ]. At the anterior pole, f-actin is organized into sheet-like lamellipodia interspersed with spiky filopodia. In time-lapse analyses, the lamellipodia have proven to show continuous motility and great lability [4 ]. The posterior of the lymphocyte is drawn out to form the uropod, a stalk-like appendage that is devoid of lamellipodia [5 ].

Unlike the posterior projections, which trail behind most types of motile cell, the lymphocyte uropod is a stable and structurally specialized appendage. Always forming at the microtubule (MT)-organizing center (MTOC), the uropod sequesters the cell’s array of MTs, MT-associated organelles, and intermediate filaments, forcing them to condense into a slender, sheaf-like configuration [6 , 7 ]. Myosins, ezrin, moesin, and spectrin also localize in the uropod [8 , 9 ]. The only other mammalian cell type to show a similar, specialized uropod is the polymorphonuclear leukocyte, another rapid penetrator of solid tissues [2 ]. It has been suggested that the uropod is an adaptation for efficient infiltration of inflammatory sites. Lymphocytes arresting on endothelial walls can recruit additional leukocytes from circulation by snagging them with the highly adhesive tip of the uropod, which displays densities of intercellular adhesion molecule-1 (ICAM-1), ICAM-3, CD43, and CD44 [10 ]. The withdrawal of the cage-like assembly of MT and intermediate filaments into the uropod increases cellular deformability and may thereby facilitate migration through fine intercellular and stromal channels [7 , 11 ].

This study is part of an effort to understand the mechanisms through which T cells establish and maintain their highly polarized form. We report that lamellipodia are restricted to the anterior pole by a mechanism which causes them to disassemble when they encroach on the forward boundary of the uropod. We also found evidence that this polarity enforcement mechanism depends on the activation of the Rho-family guanosine triphosphatase (GTPase) Cdc42. The findings provide a possible explanation for the previously reported requirement for Cdc42 activation in directional lymphocyte migration [1 , 12 , 13 ].


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
T cells and T cell leukemia line
Buffy coats from healthy donors (American Red Cross, Washington, DC) were enriched for T cells by centrifugation into Ficoll-Hypaque and passage through anti-immunoglobulin (Ig) columns (R & D Systems, Minneapolis, MN). They were activated by 48-h culture in flasks coated with anti-CD3 (UCHT1, PharMingen, San Diego, CA) and were allowed to expand with daily thinning to 1.5 x 106/ml in Dulbecco’s modified Eagle’s medium (DMEM) containing 5% calf serum, 5% fetal bovine serum, and recombinant interleukin-2 at 10 U/ml. Cells were used between days 6 and 12 of culture, as proliferation diminished but viability remained near 100%. During this period, more than 80% of the cells were polarized at a given moment. The phenotype varied from 80% to 95% CD3+. Of the CD3+ cells, 71– 85% were CD8+, and >95% were CD45RA-RO+.

The motile and intrinsically polarized clonal cell line 18-1.61 was derived from the human CD4+ T cell leukemia CCRF–CEM (American Type Culture Collection, Manassas, VA). The parental line was selected three times for high motility in type I collagen gel followed by three rounds of limiting-dilution cloning. Line 18-1.61 closely mimics anti-CD3-activated, normal human T cells in terms of polarized cytoskeletal structure. (Examples will be seen in Results.) They also show intrinsic substrate-independent polarity, the expression of polarized form while suspended over a nonadhesive surface. This is a behavior provoked in normal lymphocytes by chronic exposure to mitogenic or chemokinetic signals [7 , 14 ].

Expression of mutant GTPases
pEXV3 vectors containing cDNA constructs coding for N-terminal myc-tagged dominant-negative mutant N17Cdc42 and N-terminal myc-tagged activated mutant V12Cdc42 were generous gifts of Dr. Alan Hall (University of London, UK) [15 ]. The 637-bp myc-tagged N17Cdc42 fusion construct was cloned into the EcoRI site of the multiple cloning site of the LZRS–internal ribosomal entry site (IRES) enhanced green fluorescent protein (EGFP) vector (kind gift of Dr. Hergen Spits, Netherlands Cancer Institute, Amsterdam) using standard techniques (Fig. 1 ). In this vector, a transcript separated by an IRES sequence coded the myc-tagged N17Cdc42 and the EGFP reporter gene, resulting in two separate proteins expressed in a tightly linked manner. The orientation, integrity, and identity of the fragment inserted in the retroviral vector were confirmed by restriction endonuclease digestion and sequence analysis. Virus-producing cells were prepared by transfecting the LZRS-(N17Cdc42)-IRES EGFP plasmid into Ampho-{Phi}NX cells (kind gift of Dr. Garry Nolan, Stanford University, Palo Alto, CA) using Lipofectamine Plus (Gibco, Grand Island, NY). Transfected cells were selected in the presence of 2 µg/ml puromycin for 3 days, and virus batches were harvested from the supernatant of puromycin-resistant confluent cultures. Control viruses were produced by transfection of Ampho-{Phi}NX cells with the parent LZRS–IRES EGFP plasmid. Cells transformed with this virus express only EGFP and will be referred to as "EGFP control".



View larger version (24K):
[in this window]
[in a new window]
  		Figure 1. Representation of the bicistronic retroviral vectors used to transduce 18-1.61 leukemic T cells with myc-tagged N17Cdc42 and EGFP (upper) or EGFP alone (lower). LTR, long terminal repeat.

 
The 18-1.61 cell line was infected by spinning 4 h at 2000 g with LZRS-(N17Cdc42)-IRES EGFP virus or control LZRS–IRES EGFP virus (1x106 cells/ml undiluted virus, 25°C, 8 µg/ml protamine sulfate). After 24–48 h, EGFP-positive cells were enriched by cell sorting for green fluorescence and subcloned by three rounds of limiting-dilution culture. Six stable, clonal N17Cdc42-expressing cell lines were obtained ("18-1.61-N17Cdc42"), as confirmed by expression of EGFP fluorescence and of the myc-tagged protein (see Results). Three stable subclones of 18-1.61 expressing EGFP-control vector were also obtained ("18-1.61-EGFP control"). All EGFP- and N17Cdc42-expressing subclones displayed similar doubling times of 36–48 h and viabilities consistently >95%. All experiments were performed with cultures which were at less than eight passages after selection.

For unknown reasons, the retroviral vector strategy proved unsuitable for the transduction of the activated mutant V12Cdc42, but transient transfection could be achieved at a frequency of 20–30% with a plasmid vector. For V12Cdc42, the cDNA encoding the N-terminal myc-tag fusion protein was isolated from pEXV3V12Cdc42 as a 624-bp EcoRI fragment and cloned into the EcoRI site of the pTracer mammalian expression vector (Invitrogen, Carlsbad, CA). The orientation and integrity of the inserted fragment were confirmed as described above. Expression of this plasmid in mammalian cells results in expression of the myc-tag V12Cdc42 fusion protein under the cytomegalovirus promoter and of the Zeocin–EGFP fusion protein under the elongation factor-{alpha} promoter. Zeocin–EGFP fusion protein allowed for the rapid detection of transfected cells by EGFP fluorescence and selection of transfected cells with the antibiotic Zeocin. Cells transfected with a control pTracer plasmid expressing only Zeocin–EGFP were used as controls.

Western blotting and pull-down assay of GTPase activation
For Western blotting of native and mutant Cdc42, 1–2 x 107 cells were lysed in 1 ml 50 mM Tris/HCl, pH 7.4, containing 5 mM MgCl2, 1 mM EDTA, 1.0% Nonidet P-40 (NP-40), 0.25% sodium deoxycholate, 0.75 µM aprotinin, 10 µM leupeptin, 2.8 µM pepstatin, 15 µM bestatin, and 0.80 mM 4-(2-aminoethyl)-benzenesulfonyl fluoride for 30 min at 4°C. After centrifugation at 10,000 g, protein in postnuclear supernatant was quantified by colorimetric assay (DC, BioRad, Hercules, CA). Protein solutions were diluted in 3x Laemmli buffer, heated at 95°C for 3 min, and applied to a 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis mini-gel, and all lanes received equal protein loads (usually 20 µg). Proteins were blotted onto polyvinylidene difluoride membrane (Millipore, Bedford, MA), blocked with phosphate-buffered saline (PBS) with 1% low-fat dry milk and 0.1% Tween-20, and agitated overnight at 4°C. Blots were incubated with monoclonal anti-Cdc42 (Upstate Biotechnology, Lake Placid, NY), anti-myc-tag, or normal mouse Ig in blocking buffer, 1–3 h, then washed and incubated 1 h with horseradish peroxidase-conjugated goat anti-mouse Ig (Transduction Laboratories, Lexington, KY). After extensive washing with PBS with 0.1% Tween-20, proteins were visualized on X-ray films by enhanced chemiluminescence (Amersham/Pharmacia, Piscataway, NJ).

For assay of GTPase activity, an affinity precipitation "pull-down" technique was used [16 ]. It was based on the principle that only the activated [guanosine 5'-triphosphate (GTP)-bound] form of Cdc42 binds specifically to the p21-binding domain (PBD) of human p21-activated kinase-1 (PAK-1). Cells (2x107) cells were lysed in 1 ml magnesium-containing lysis buffer (MLB) consisting of 25 mM HEPES, pH 7.5, containing 150 mM NaCl, 1% NP-40, 0.25% Na deoxycholate, 10% glycerol, 25 mM NaF, 10 mM MgCl2, 10 mM Na orthovanadate, 10 µg/ml leupeptin, and 10 ug/ml aprotinin. Protein content, which ranged from 850 to 1300 µg/ml, was verified to differ by less than 20 µg/ml between samples within each experiment. To each ml lysate was added 10 µg glutathione-agarose beads coupled to a glutathione S-transferase fusion protein containing human PAK-1 PBD (Upstate Biotechnology). After rocking 60 min at 4°C, beads were harvested by centrifuge pulse and washed 3x with MLB. Proteins were eluted by heating beads 95°C for 3 min in 20 µl 2x Laemmli buffer. Eluates were electrophoresed, blotted, and probed with anti-Cdc-42 or control Ig as above. Negative controls consisted of PAK-1 PBD beads incubated in lysis buffer only. To estimate the size of the pool of Cdc42 available for activation, an aliquot of every lysate was pretreated with the nonhydrolyzable GTP analog GTPyS (100 µM, 30 min, 37°C) and then subjected to pull-down assay.

Analysis of morphology and migratory activity
Culture chambers were constructed by mounting an 18-mm Teflon ring with surgical sealant onto a 22 x 60 mm No. 1 cover glass (Corning, Corning, NY). Cells (5x105) were added in 700 µl culture medium supplemented with 20 mM HEPES, and chambers were covered with another No. 1 glass. Chambers were placed on the thermostatic 37°C stage of an inverted microscope (Zeiss Axiovert 35, Zeiss, Thornwood, NY) for time-lapse recording with a Dage chilled CCD digital camera. Time-lapse output was stored on tape by a Panasonic TL time-lapse video recorder or saved to a hard disc with the time-lapse routine of the MCID5 image analysis system (Imaging Research, St. Catherine’s, Ontario).

In some experiments, the chamber surface was rendered nonadhesive to lymphocytes by preincubation with 2% bovine serum albumin (BSA). Morphological changes were observed as the cells rested on the nonadhesive surface or drifted over it. In other experiments, surfaces were fibronectin-coated by incubation in 20 mM Tris buffer, pH 8.0, containing bovine fibronectin (Gibco) at 20 ug/ml overnight at 4°C, followed by blocking of nonspecific cell-binding sites, 1 h, with 2% BSA (fraction V, Sigma Chemical Co., St. Louis, MO). After 1 h incubation at 37°C, nonadherent cells were washed off by gentle pipetting with 37°C medium. Usually 40–60% of cells remained adherent. Antibody inhibition experiments proved this adhesion to be dependent mainly on {alpha}4ß1 integrin (not shown).

For observations of the formation and movement of appendages, cells were videotaped at a time-lapse factor of 12x normal speed and a magnification of 1000x. Recordings were made of cells in suspension and those migrating over fibronectin-coated surfaces. Cells often exhibited transient reversions to nonpolar form or on fibronectin, to a stationary outspread configuration in which the uropod could not be distinguished. Only cells that maintained clear polarity for the duration of observation periods (30–120 min) were included in the datasets. For each experimental condition, at least 20 observations were recorded. Statistical analysis of lamellipodial lifespans was by ANOVA (Instat, Graphpad Software, San Diego, CA).

For characterization of migratory paths on fibronectin, fields of adherent cells were videotaped at 400x and a time-lapse factor of 24x for 30- to 45-min periods. During playback, the movements of individual cells were traced onto plastic overlays. Path length of each tracing was calculated by the MCID5 image analysis system. Most cells displayed alternating periods of migration on the surface and detachment into suspension. Only cells that remained adherent for at least 20 min were traced.

Quantitation of spontaneous and chemotactic migration was performed in modified, 10-well Boyden chambers (Neuroprobe, Gaithersburg, MD) with 3-µm pore polycarbonate filters coated with fibronectin, as described for coverslips above. Lymphocytes were added to upper compartments at 5 x 104/well in DMEM, 2% BSA. They were allowed to migrate 3.5 h in a range of positive and negative gradients of chemoattractant, which were created by adding stromal cell-derived factor-1{alpha} (SDF-1{alpha}) to upper or lower compartments in concentrations from 0 to 25 ng/ml. Most migrating cells detached from the filter and dropped into the lower compartment. Cells collected from five wells for each condition were pooled and counted. Residual cells still adhering to the underside of the filter were also counted after Giemsa staining. Each experiment was repeated three times. Statistical analysis was by ANOVA followed by paired t-tests with Bonferroni correction, and each pair consisted of EGFP-control cells versus N17Cdc42 expressors at matching concentration gradients (Instat, Graphpad Software.)

Immunocytochemical staining
Cells were fixed at 37°C while in suspension by the addition of 33% paraformaldehyde (PF) to a final concentration of 3%. For some stains, noted below, cells were pelleted and fixed by suspension in methanol. For intracellular staining of PF-fixed cells, all buffers and stains contained 0.2% Triton X-100. Primary stains included monoclonal antibodies against Cdc42 (Transduction Laboratories), myc-tag (Invitrogen), {alpha}-tubulin (clone B512, Sigma Chemical Co.), vimentin (clone V9, Sigma Chemical Co.), ICAM-3 (Chemicon, Temecula, CA), rabbit polyclonal antisera against myosin II (BTI, Stoughton, MA), and moesin (gift of Dr. A. Bretscher, Cornell University, Ithaca, NY). Moesin stains were performed on methanol-fixed cells. Normal Ig, ascites, or serum were used as controls. F-actin was stained with rhodamine B-isothiocyanate (RITC)-phalloidin (Molecular Probes, Eugene, OR). All secondary antibodies were RITC-conjugated to allow simultaneous verification of EGFP fluorescence. Images were obtained with a Zeiss LSM-310 confocal microscope (Confocal Microscopy Core Facility, Karmanos Cancer Institute, Detroit, MI).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Enforcement of polarized expression of appendages in activated T cells
The behavior of activated T cells in suspension over nonadhesive BSA-coated coverslips was recorded by time-lapse videomicroscopy. Suspension culture was chosen as the initial technique in this study, as it allows the form and motion of lymphocyte appendages to be assessed without the complicating factors of spreading and distortion on an adhesive surface. The cells exhibited periods of spontaneous polarization as defined by elongated shape and distinct uropod. These periods were variable, lasting a mean of 37.8 min (SD=10.7, n=25 60-min observations).

During periods of polarization, the uropod was always recognizable as a stable, stalk-like structure that never bore lamellipodia. Although the suspended cells could not migrate, the uropod was taken to mark the posterior pole, as it does during migration. At the anterior pole, lamellipodia continually developed and in contrast to the stable uropod, never lasted more than 5 min. Furthermore, they displayed a spatially restricted life cycle, which kept them confined to the anterior pole. This life cycle, an example of which is shown in Figure 2 , operated as follows: Cycles began at intervals of 3–16 min, when the anterior subdivided into two lobes, each bearing a lamellipodium. One lobe quickly dominated, becoming the new anterior. The other traveled rearward until it reached the constricted region which defines the anterior border of the uropod. At that point, the lobe and its lamellipodia abruptly shrank and disappeared. Presumably, this disappearance represented the disassembly of the actin-myosin structures which defined the lobe and supported the lamellipodium. The mean lifespan of a lamellipodial lobe in 25 such observations was 2.0 ± 1.3 min. No lobes were observed to pass into the uropod. The lamellipodial life cycle was not significantly altered in spatial distribution or duration by the introduction of SDF-1{alpha} into the medium at 25 ng/ml, a concentration that is chemotactic and chemokinetic to these cells (as will be seen below).



View larger version (86K):
[in this window]
[in a new window]
  		Figure 2. In polarized, normal T cells, lamellipodia followed a spatially restricted life cycle. Images selected from time-lapse sequence recorded by a digital camera of a typical cell suspended over a nonadhesive BSA-coated surface. Lobe-bearing lamellipodia (arrows) separate from the anterior and travel rearward but disassemble abruptly upon reaching the anterior portion of the uropod (U). Elapsed time in seconds accompanies each frame. Note that to capture the rapid disassembly process, the last four frames are separated by shorter intervals than the first four frames. Original bar = 10 µm.

 
A similar lamellipodial life cycle was seen in activated T cells, which migrated spontaneously over fibronectin-coated surfaces. Cycles began at intervals of 2–11 min. The mean lifespan of a lamellipodial lobe was 3.1 ± 1.7 min (n=30) or ~50% longer than that seen in suspended cells. As in suspended cells, 100% of lobes disassembled as they crossed the uropod boundary. When SDF-1{alpha} was dispersed in the culture medium at 25 ng/ml, velocity of random migration approximately doubled (not shown), and the half-life of lamellipodial lobes decreased by 32% (P<0.05), but no lamellipodial lobe was observed to persist after crossing into the uropod (not shown). It was not possible to generate a concentration gradient of SDF-1{alpha} within these chambers.

It is well known that the anterior of the lymphocyte is a region that favors the presence of lamellipodia [5 ]. Our results indicate that this anterior restriction of lamellipodia reflects the division of the polarized cell into two sharply defined zones, one in which lamellipodia are produced and one in which lamellipodia are rapidly disassembled, and the borderline coincides with the forward boundary of the uropod. We next determined whether the Rho-family GTPase Cdc42 plays a role in defining these zones.

Expression of mutant protein
Cdc42 has been implicated as a transducer of polarity-regulating signals in the development and motility of many cell types. In lymphocytes, inhibition of Cdc42 has been found to disrupt the normal orientation of the T cell MTOC toward targets and antigen-presenting cells (APC) during synapse formation [17 ] and to inhibit directed motility [1 , 13 ]. To determine whether Cdc42 has a role in the polarized expression of lymphocyte appendages, mutant forms of this GTPase were expressed in a model cell line 18-1.61, which was derived from CCRF–CEM human T cell leukemia.

Cdc42 activity was specifically inhibited by expression of myc-tagged dominant-negative mutant N17Cdc42, which lacks the motif for binding downstream effectors via the CDC42/Rac1 interactive-binding motif [18 ]. Transduction was performed with the LZRS–IRES retroviral vector. In this vector, the gene for EGFP fluorescent reporter protein shared a common IRES with the GTPase so that EGFP and GTPase expression was tightly linked. Control vectors encoded only EGFP. Three different clonal sublines of N17Cdc42 expressors and EGFP controls were used in this study.

In N17Cdc42-transduced clones of 18-1.61, two bands of 22 and 23 kD stained specifically for Cdc42. (Results for typical clone B5 are seen in Fig. 3A , left, lane 2.) The 22-kD band represented native Cdc42, and the 23-kD band represented the myc-tagged N17 mutant [19 ], as it also stained specifically with anti-myc-tag (Fig. 3A , lane 4). From the relative densities of the 22- and 23-kD bands, the ratio of mutant-to-normal protein was estimated to range from 1.3: 1 to 1.6:1 across the three clonal lines. In EGFP-control clones, only the native 22-kD form was seen (result for typical clone G9 is shown in Fig. 3A , lane 1), and this band did not react with anti-myc-tag (Fig. 3A , lane 3).



View larger version (57K):
[in this window]
[in a new window]
  		Figure 3. Retroviral transduction of myc-tagged N17Cdc42 into 18-1.61 leukemia cells resulted in reduction of endogenous Cdc42 activity; N17 and wild-type proteins showed similar subcellular distributions. (A) Western blots of lysates of typical clonal cell lines stably transduced with control EGFP vector (lanes 1 and 3) or N17Cdc42-myc (lanes 2 and 4) and probed with anti-Cdc42 (left panel) or anti-myc-tag (right panel). Native Cdc42 appears as a 22-kD band and myc-tagged mutant at 23 kD. Note that only native Cdc42 was expressed in EGFP-control cells. (B) Results of pull-down assay of activated Cdc42 in same cell lines as in 3A. PAK-1 PBD beads were incubated with each lysate, and the eluates were Western blotted with anti-Cdc42. Lanes 1 and 3, Eluates from lysates spiked with GTPyS to activate irreversibly all GTP-binding proteins. Equivalent densities of the two bands indicate that the EGFP control (lane 1) and N17Cdc42 expressor (lane 3) contained equivalent pools of Cdc42 available for activation. Lanes 2 and 4, Eluates from unmodified lysates. Suppression of Cdc42 activity is indicated by the lighter density of band in lane 4 (N17Cdc42) relative to that in lane 2 (EGFP control). (C–E) Immunofluorescence analysis of gene expression in fixed and permeabilized cells. (C, D) Typical field of fixed and permeabilized 18-1.61 expressing N17Cdc42. (C) EGFP fluorescence; (D) anti-myc-tag RITC. (E) Distribution pattern of native Cdc42 in a typical cell of nontransduced, parental 18-1.61. Normal Ig stains yielded negligible fluorescence (not shown). Original bars = 10 µm.

 
Pull-down assays verified that this degree of N17Cdc42 expression was sufficient to reduce Cdc42 activity. Only the activated (GTP-bound) form of Cdc42 binds to beads coupled to PAK-1 PBD [16 ]. Western blots demonstrated that beads exposed to lysates of N17Cdc42-transduced cells (Fig. 3B , lane 2) bound markedly less Cdc42 than those exposed to control-cell lysates (Fig. 3B , lane 1), although Cdc42 activity was not entirely abolished. (Beads exposed only to lysis buffer produced no detectable signal; not shown). The size of the pool of Cdc42 available for activation was estimated by pull-down assays of lysates treated with the nonhydrolyzable GTP analog GTPyS. From densitometric comparisons of pull-downs from untreated and GTPyS-treated lysates, it was determined that 13% of available Cdc42 was activated in the typical EGFP-control clone (Fig. 3B , lanes 1 vs. 2) and 3% in the typical N17Cdc42-expressing clone (Fig. 3B , lanes 3 vs. 4). Similar results were obtained with the other two EGFP- and N17Cdc42-expressing clones (not shown). The results confirm that Cdc42 activity was reduced by the expression of N17Cdc42, presumably as the mutant competed with the normal protein for activation-inducing guanine nucleotide exchange factors [20 ].

Frequency of expression of N17Cdc42 in transduced cells was >95%, as determined by fluorescence microscopy for EGFP and myc-tag (Fig. 3C and 3D) and by flow cytometry (not shown). The subcellular distribution of myc-tagged protein in transduced clones was similar to that of native Cdc42 in controls, i.e., diffuse with slight densities in lamellipodia and uropod (Fig. 3D and 3E) .

Disruption of polarized morphology induced by N17Cdc42
Clones of 18-1.61 transduced with the control EGFP construct were PF-fixed while in suspension. In these preparations, 70–90% of cells showed polarized morphology essentially identical to that of untreated 18-1.61 (not shown) and to that of normal human T cells polarized by exposure to anti-CD3 (above). That is, there was a broad anterior pole, which contained the nucleus and bore lamellipodia, and a posterior, stalk-like uropod (e.g., Fig. 4A ). Often, the anterior was subdivided into two or three lamellipodial lobes.



View larger version (68K):
[in this window]
[in a new window]
  		Figure 4. Disruption of polarized morphology of 18-1.61 cells by expression of N17Cdc42 in 18-1.61. (A) Phase-contrast micrograph of typical expressor of EGFP-control vector fixed while in suspension. L, Lamellipodial lobes, i.e., subdivisions of the anterior that bear lamellipodia; U, uropod at posterior. Original bar = 10 µm. (B) Typical expressor of N17Cdc42. EL, Ectopic lamellipodial lobe projecting from uropod. (C) Normal lamellipodial life cycle in typical EGFP-control cell, seen in time-lapse frames of cell suspended over nonadhesive BSA-coated surface. Lamellipodial lobe (arrows) generated at anterior traveled rearward and disassembled as it approached the anterior boundary of the uropod (U). Note that total elapsed real time (upper left of each panel) was approximately 2 min. Original bar = 10 µm. (D) Ectopic lamellipodia of N17Cdc42 expressors resulted from persistence of rearward-traveling lamellipodial lobe (arrows) after it moved past the uropod boundary. Time-lapse sequence recorded as in C above. Stars denote uropod. From real-time clock (upper left of each panel), note that the ectopic lamellipodial lobe persisted for ~20 min after passing into the uropod region. (E) Ectopic lamellipodial lobes occurred only in N17Cdc42 expressors and showed temporal heterogeneity. Randomly chosen cells were recorded by time-lapse microscopy over a period of 40 min. When a lamellipodial lobe began its rearward procession, the time it persisted after passing the anterior boundary of the uropod was measured. Results are presented as frequency distributions, and each point indicates the persistence time of an individual lobe. Percentages listed over columns give percentage of cells with persistence greater than 0 min. Culture conditions included: 1., Cells in suspension over BSA coating; 2., cells adhering to fibronectin; and 3., cells adhering to fibronectin (Fnctn.) in medium containing SDF-1{alpha} (25 ng/ml). n = 23–35 cells per condition.

 
In clones of 18-1.61 which expressed N17Cdc42 the frequency of polarized cells was not significantly different from that of controls, but 14.5% of the cells had a "two-headed" appearance (SD=2.4, n=600 cells of three different clonal populations). This two-headedness resulted from the presence of one or more lamellipodial lobes located ectopically at the still-recognizable uropod (e.g., EL in Fig. 4B ). Such cells were absent from 18-1.61 EGFP-control cultures, in which the uropod was invariably free of lobes or lamellipodia. It was next found that by time-lapse analysis, the ectopic lobes resulted from a disruption of the normal life cycle of lymphocyte lamellipodia.

EGFP-expressing control cells exhibited a lamellipodial life cycle essentially identical to that described above for activated, normal T cells. An example is seen in Figure 4C . Lamellipodial lobes departed from the anterior at intervals of 4–20 min. They traveled rearward but abruptly diminished and disappeared at the anterior boundary of the uropod. The mean lifespan of a lamellipodial lobe was 2.2 ± 0.9 min (n=25).

In cells expressing N17Cdc42, lamellipodial lobes developed at approximately the same frequency as in controls. They always formed at the anterior and traveled rearward in the normal manner. The difference from control cells was that the lamellipodial lobes failed to diminish immediately upon reaching the uropod, persisting at the rear of the cell for periods of 4–39 min (Fig. 4D) . The ectopic lobes grew and shrank periodically but eventually always diminished and disappeared within 45 min. Thus, inactivation of Cdc42 did not affect the generation of lamellipodia-bearing structures but rather altered their fates, allowing them to exist in a region from which they are normally forbidden.

As described above, only ~15% of fixed cells displayed ectopic lamellipodial lobes, although nearly >95% expressed N17Cdc42. Extended time-lapse analysis indicated that this discrepancy was attributable to temporal heterogeneity in the effect of the mutant GTPase and that most expressors developed an ectopic lobe if observed long enough. When randomly chosen cells were each followed over a time-lapse period of 40 min, 38% (n=24) generated at least one ectopic lamellipodial lobe, whereas none of the controls did (n=23; Fig. 4E , 1.) In 120-min observations, incidence increased to 67% for N17Cdc42 expressors and remained 0% for controls (n=12; not shown). Normal lamellipodial cycles often alternated with ectopic ones in the same cell. The temporal heterogeneity might reflect cycles of exhaustion and regeneration of a limited quantity of N17Cdc42. Lamellipodial life cycles of EGFP and N17Cdc42 expressors were not significantly altered in spatial distribution or duration by the introduction of SDF-1{alpha} into the medium at 25 ng/ml (not shown), a concentration we have found to be chemotactic and chemokinetic to these cells (described below).

The effects of N17Cdc42 expression reported thus far were observed in suspended lymphocytes. We next determined whether N17Cdc42 produced the same effects in lymphocytes engaged in integrin-mediated adhesion to a substrate. During spontaneous migration on fibronectin-coated surfaces, parental 18-1.61 cells and EGFP-transduced controls exhibited lamellipodial life cycles similar to those described above for activated T cells on the same substrate. Lamellipodial lobes were generated at the anterior at intervals of 3–25 min. The mean lifespan of a lobe was 3.5 ± 1.0 min (n=23). Lobes (100%) disassembled as they crossed the uropod boundary (Fig. 4E , 2.). In cells expressing N17Cdc42, lamellipodial lobes arose at the anterior at a frequency comparable with that of controls and traveled rearward at a similar speed. As was the case in suspended cells, however, these lamellipodial lobes often failed to diminish at the uropod boundary. In 42% of cells, the lobes persisted at the rear of the cell for periods of 1–38 min (see Fig. 4E , 2., for frequency distribution).

Addition of SDF-1{alpha} to the culture medium at 25 µg/ml caused an 80% increase in the velocity of random migration of EGFP controls on fibronectin (not shown). Still, the lifespans of lamellipodial lobes did not change significantly (not shown), and no lobes persisted after crossing into the uropod (Fig. 4E , 3.). N17Cdc42 expressors exposed to the same concentration of SDF-1{alpha} developed ectopic lamellipodia at approximately the same frequency as those in control medium (cf., Fig. 4E , 2. and 3.; note that SDF-1 did not increase Cdc42 activity in control or N17Cdc42 expressors during 30- or 120-min incubations; not shown). Taken together, these results show that the polarity disruption caused by inactivation of Cdc42 was not overridden by integrin-mediated signaling from the substrate or by the presence of a chemokinetic factor. It was not possible to generate a concentration gradient of SDF-1{alpha} within these chambers.

Impairment of motility by expression of N17Cdc42
In Boyden chamber assays with fibronectin-coated, 3-µm-pore filters, EGFP-expressing, control cells displayed chemokinesis and chemotaxis toward the chemokine SDF-1{alpha}. Chemokinesis was indicated by a threefold increase in migration over the basal level when SDF-1{alpha} was presented in equal concentrations in the upper and lower compartments (shaded diagonal of squares in Fig. 5A ) or as negative concentration gradient (area above shaded diagonal in Fig. 5A ). A chemotactic component was indicated by the further increase in migration, up to five times basal level, which occurred when SDF-1{alpha} was presented as a positive gradient (area below shaded diagonal). N17Cdc42 expressors also exhibited chemokinesis and chemotaxis, but both responses were significantly reduced relative to EGFP controls. At all concentration ranges, migration was 50–65% lower than those of controls (cf., Fig. 5A and 5B ). This difference did not reflect differential adhesion to fibronectin, as there was no treatment-related difference in the percentage of cells remaining attached to the top or bottom of the filters (<1% in all conditions; not shown). Taken together, these findings indicate that inhibition of Cdc42 caused a reduction in random and chemotactic migration rather than a selective defect in chemotactic navigation.



View larger version (36K):
[in this window]
[in a new window]
  		Figure 5. Reduction in chemokinetic and chemotactic responses of 18-1.61 cells to SDF-1{alpha} by expression of N17Cdc42. Results of Boyden chamber assays of migration of cells through fibronectin-coated polycarbonate filters (3 µM pore), expressed as percentage of cells that moved through filter in a 3.5-h period. (A) EGFP-control cells; (B) N17-Cdc42 expressors. Values are means (±SD) from three experiments, each with a different clonal population, with five replicate migration wells run for each condition. (B) All values are significantly different from matching values in A, P < 0.01, as determined by ANOVA followed by paired t-tests with Bonferroni correction.

 
A likely explanation for Boyden chamber results is the competing locomotory activity of normal and ectopic lamellipodial lobes, which was observed in time-lapse recordings. Despite their posterior and posterio-lateral positions, the ectopic lamellipodia intermittently adhered to the substrate and generated tractive force in the same manner as the normal anterior lamellipodia. It was often observed that a period of forward migration mediated by normal lamellipodia was interrupted when the ectopic lamellipodia gripped the substrate, causing a sudden halt, lateral shift, or complete reversal of direction. Examples can be seen in a series of frame captures from a representative recording (Fig. 6A ) and in tracings of typical paths of random migration on fibronectin (Fig. 6B , in which course reversals, defined as abrupt course deviations of 150–210°, are indicated by arrows). In N17Cdc42 expressors, course reversals occurred at a mean rate of 5.4 ± 2.6/h in contrast to 0.2 ± 0.6/h for controls (n=30 tracings per condition; P< 0.01; Student’s t-test). All cases of course reversal in the N17Cdc42 expressors occurred when an ectopic lamellipodium adhered to the substrate (E arrows in Fig. 6B ), or when an adherent ectopic lamellipodium released from the substrate, allowing the normal lamellipdium to resume dominance (N arrows in Fig. 6B ). Thus, although N17Cdc42 expressors traveled approximately the same total distance, they were less able than controls to move persistently in a single direction. This interfering action of ectopic lamellipodia is consistent with the reduction in random and chemotactic motility seen in the Boyden chamber experiments, although a cause-and-effect relationship could not be tested.



View larger version (50K):
[in this window]
[in a new window]
  		Figure 6. Ectopic lamellipodial lobes of N17Cdc42 expressors interfered with forward locomotion by acting in conflict with normal lamellipodia. (A) Competing locomotory activities of normal and ectopic lamellipodial lobes on fibronectin-coated substrate, depicted in frames captured from time-lapse recording. Normal lamellipodial lobe of the subject cell is that closest to the top of the frame. Asterisks mark lobe(s) actually in contact with substrate in each frame. Segmented line tracks position of center of uropod of subject cell in each frame. Elapsed time is shown on clock readouts at upper left of each frame, and total = 3 min. (B) Tracings of typical, random, migratory paths of cells expressing EGFP-control vector (upper) or N17Cdc42 (lower) over 30-min time periods. In N17Cdc42 paths, E = point at which an ectopic lamellipodial lobe adhered to substrate and produced a sharp reversal in course (150–210°); N = point at which a sharp course reversal was caused by a normal lobe reasserting its dominance.

 
Ectopic lamellipodial lobes induced by N17Cdc42 were expressed against a background of otherwise normal polarity
EGFP-expressing control cells closely modeled the sharp anterior–posterior differentiation of normal, polarized T cells. The broad anterior bore lamellipodia and filopodia rich in f-actin (Fig. 7 A and B ). The posterior was drawn out to form the stalk-like uropod, and the MTOC was at the distal tip (Fig. 7E and 7F) . MT and vimentin-intermediate filaments were contained almost entirely within the uropod, where they were compressed into a compact bundle (Fig. 7E 7F 7I and 7J) . Myosin II formed a collar-like concentration around the proximal portion of the uropod but was present only as discrete patches at the lamellipodia (Fig. 7M and 7N) . The ezrin, radixin, and moesin family protein, moesin, massed heavily at the tip of the uropod and formed isolated clusters at the lamellipodia (Fig. 7Q and 7R) . CD43 (not shown) and ICAM-3 also localized at the uropod tip but were undetectable at the anterior (Fig. 7U and 7V) . ß1 Integrin formed punctate densities, which were heaviest at the uropod and relatively sparse at the lamellipodia (not shown).



View larger version (106K):
[in this window]
[in a new window]
  		Figure 7. Ectopic lamellipodial lobes induced by N17Cdc42 were structurally similar to normal lobes and were expressed against a background of otherwise normal polarity. Immunocytochemical comparisons of 18-1.61 cells transduced with EGFP-control vector with those transduced with N17Cdc42. Cells were fixed while in suspension. Matched, normal Ig, ascites, or serum controls yielded negligible fluorescence (not shown). u, Uropod; el, ectopic lamellipodial lobe. (A) Original bar = 10 µm.

 
Although 18-1.61-N17Cdc42 displayed ectopic lamellipodial lobes at the uropod, the underlying features of lymphocyte polarity remained normal. MT, vimentin-intermediate filaments, myosin II, moesin, and ICAM-3 all concentrated in a well-defined uropod (Fig. 7G 7H 7K 7L 7O 7P 7S 7T 7W and 7X) as did CD43 and ß1 integrin (not shown). These patterns are essentially identical to those seen in 18-1.61-EGFP-control cells. The ectopic lamellipodial lobes retained the properties typical of normal lobes and did not acquire properties of the uropod. The ectopic lobes bore lamellipodia, which were rich in f-actin and were interspersed with filopodia (Fig. 7C and 7D) . They did not receive MT or vimentin-intermediate filaments from the uropod (Fig. 7G 7H 7K and 7L) . Myosin II (Fig. 7O and 7P) , ß1 integrin (not shown), and moesin (Fig. 7S and 7T) were present in light, sparse concentrations comparable with those of normal lamellipodia. CD43 (not shown) and ICAM-3 were nearly undetectable in ectopic lobes (Fig. 7W and 7X) , just as they were in normal lobes. These results suggest that Cdc42-mediated regulation of lamellipodium location operates downstream of signals which specify the overall anterior–posterior polarity of the lymphocyte.

Effects of V12Cdc42
It was next determined whether the expression of ectopic lamellpodia resulted specifically from the inhibition of Cdc42 or simply from disturbance of normal levels of Cdc42 activity. Stable cell lines expressing activated, mutant myc-tagged V12Cdc42 could not be produced, but transient expression was achieved via the pTracer plasmid vector, which confers an EGFP-positive phenotype. It was confirmed by immunofluorescence analysis of fixed and permeabilized cells that EGFP positivity correlated reliably with expression of myc-tagged V12Cdc42 (not shown). Morphological effects were measured by comparison of EGFP-positive V12Cdc42 expressors versus EGFP-positive control cells, which had received the vector alone.

Expression of V12Cdc42 did disrupt the polarity of 18-1.61 but in a manner different from N17Cdc42. Rather than ectopic lamellipodia superimposed on polarized morphology, the cells showed a significant loss of polarity, i.e., elimination of uropod, lamellipodia, and elongated shape. Without these obvious "landmarks", it could not be determined whether the tubulin or vimentin systems retained any residual polarity, as they did in the case of N17Cdc42. In three independent transfection experiments, a mean of 46.3% ± 8.3% of V12Cdc42 expressors was in nonpolar form at a given moment as opposed to 13.7% ± 3.9% of control cells (P<0.05; Table 1 ).


View this table:
[in this window]
[in a new window]
  		Table 1. Morphological and Behavioral Consequences of Transient Expression of V12Cdc42 in CEM 18-1.61

 
V12Cdc42 expressors also showed significantly greater adhesion to fibronectin as well as a lower rate of random migration over a fibronectin-coated surface (Table 1) . The reduction in appendage formation and increased adhesion may have contributed to the reduced migration. V12Cdc42 has been reported to produce similar changes in other leukemia cell lines [1 , 21 ]. The results indicate that any alteration of normal Cdc42 activity can disturb lymphocyte polarity but that the nature of the disturbance depends on the type of alteration, and reduced activity permits ectopic lamellipodia, and increased activity suppresses the expression of all appendages.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
It has been proposed that the anterior and posterior poles of the motile lymphocyte are created by compartmentalized regulation of the actin-myosin cytoskeleton [5 ]. Our results support and further elucidate this concept. We found evidence that the anterior and uropod represent sharply segregated, regulatory compartments. In the anterior compartment, the generation and persistence of lamellipodia are favored. In the posterior compartment, which coincides with the uropod, lamellipodial dissolution is promoted. In addition, we have found that the maintenance of this compartmentalization requires activation of the Rho-family GTPase Cdc42.

These findings add to the growing body of information about the roles of Rho-family GTPases in determining shape change and polarity in lymphoid cells. Some functions are similar to those previously determined to operate in fibroblasts. In lymphoid and fibroblastic cells, lamellipodium formation is promoted by activation of Rac and filipodium formation by activation of Cdc42 [1 , 21 , 22 ]. In both cell types, RhoA regulates actin-myosin-mediated contractility and integrin-mediated adhesion [22 23 24 25 26 ]. Features unique to lymphocytes are also proving to be under Rho-GTPase regulation. Localization of moesin and ICAM-3 in the uropod is permitted by low levels of activity of Rac, Rho, and Cdc42 and conversely inhibited by activated mutants of these enzymes [1 ]. Rac1 and Cdc42, through interactions with Vav1 and Wiskott-Aldrich syndrome protein (WASP), regulate the molecular motors which cluster antigen receptor complexes during antigen stimulation. [27 ]

Cdc42 appears to be especially important in lymphocyte functions which require cellular polarity. Immunological synapses with target cells or APC normally form at the lymphocyte MTOC, but their placement becomes random after experimental perturbation of Cdc42 activity [17 ]. Polarization is necessary for motility, and inhibition of Cdc42 activity reduces random and chemotactic lymphocyte motility [13 , 28 ]. In this report, we have defined a mechanism that may explain why Cdc42 is necessary for these polarized lymphocyte functions. Lamellipodia arise at the anterior and move rearward in lobe-like assemblies. Cdc42 governs a mechanism which causes these lamellipodial lobes to disassemble before they intrude into the uropod. When Cdc42 activity was reduced in migrating lymphocytes, lamellipodial lobes failed to disassemble, and their continued motility at the posterior interfered with forward movement. This ectopic lamellipodial expression was not corrected by integrin-mediated adhesion to fibronectin, alone or in combination with a chemokinetic stimulus (Fig. 4E , 2. and 3.). We could not observe directly whether a chemotactic gradient could correct the polarity defect, but the fact that motility was not restored in Boyden chamber experiments suggests that it could not (Fig. 5) . Taken together, the results suggest that the polarity-enforcing action of Cdc42 is intrinsic to the polarized lymphocyte phenotype and is not induced only after encounter with adhesive or chemotactic signals.

Activated Cdc42 can localize appendage-regulating signals to particular cellular regions [29 ]. It is therefore possible that activated Cdc42 excludes lamellipodial lobes from the uropod by causing the posterior localization of proteins that favor actin severing and capping and actin-myosin disassociation. Conversely, activated Cdc42 might induce the anterior localization of proteins which promote lamellipodial growth and persistence, i.e., those that nucleate actin or favor actin-myosin assembly. Possible effectors that might be localized or locally activated in this manner include WASP, whose association with Cdc42 is necessary for chemotaxis of T cells to SDF-1 [13 ]. Cdc42-bound WASP activates the Arp2/3 complex, starting a process that leads to the formation of branched actin filaments. Additional actin-myosin regulators and scaffolding proteins known to lie downstream from Cdc42 include LIM kinase, phosphatidyl inositol-3 kinase, myotonic dystrophy kinase-related Cdc42-binding kinase, and Par6/protein kinase C-{zeta} complexes [30 31 32 33 34 35 ]. Activated Cdc42 complexes with IQGAP and cytoplasmic linker protein-170 to govern the linking of MT to the cortical cytoskeleton [36 ]. These downstream effectors could account for the promotion or inhibition of lamellipodium formation, depending on the background of other signal-transduction events. It will be of interest to determine which of these effectors shows disrupted subcellular distribution when Cdc42 activity is experimentally altered.

It is also conceivable that Cdc42 mediates spatial restriction of lymphocyte lamellipodia by directing the traffic of structural or regulatory molecules to specific poles. This idea is suggested by the recent finding that Cdc42 maintains apical–basal polarity of epithelial cells by regulating the polarized traffic of endosomes [37 , 38 ].

Cdc42 activation did not appear to be the primary determinant of polarity but rather an enforcer of the distinct properties of poles that were predetermined by other, still uncharacterized, signals. Even with Cdc42 inhibited, the uropod invariably formed at the MTOC, the usual landmark of the lymphocyte posterior [1 , 5 , 39 ], and the lamellipodia arose at the opposite pole. Localization of myosin and of moesin and its associated adhesion receptors at the uropod also occurred, indicating that these polarization events are controlled by signals upstream of or parallel to Cdc42. Also consistent with the concept that Cdc42 is not the primary specifier of anterior–posterior polarity is our finding that transient overexpression of constitutively activated mutant V12Cdc42 was not sufficient to induce polarization in NS4R1-10, a spherical human leukemia cell line selected from CCRF–CEM (not shown). It is conceivable that the primary determinants of lymphocyte polarity are signals carried by the MT array, as posterior placement of the MTOC is a constant feature of the polarized lymphocyte.

The generation of lamellipodia in lymphocytes has been found to be under the regulation of Rac [21 , 22 ]. This is also the case for the cell line used in the present study. Expression of dominant-negative Rac1 in CCRF–CEM resulted in populations completely devoid of lamellipodia (unpublished results). A plausible scenario that emerges from our study is one in which Cdc42 regulates the persistence of lamellipodia, which were previously produced under the influence of activated Rac. Does Cdc42 activity encourage the survival of lamellipodia at the anterior or promote their disassembly at the posterior? The latter possibility is more likely, in light of our finding that activated, mutant V12Cdc42 caused a small but significant suppression of lamellipodium formation. The experiments were not exactly comparable with those with N17Cdc42, as the mutant protein could only be transiently expressed; therefore, the matter remains to be resolved definitively in our experimental system. Note, however, that Cdc42 activation has been found to suppress lamellipodium formation in other lymphocyte systems [1 , 21 ].

One mechanism of polarity enforcement that may be ruled out is cross-regulation of Rac by Cdc42. If this were the case, expression of N17Cdc42 would have altered Rac activity, and we have found this not to be the case (unpublished result).

We also found that RhoA is not down-regulated by the expression of N17Cdc42 (unpublished result). The role of RhoA in restricting lamellipodia to the lymphocyte anterior still requires investigation. RhoA tends to inhibit the protrusion of lamellipodia through its contraction-promoting action [40 ]. This raises the possibility that prohibition of lamellipodia from the uropod is mediated by local activation of RhoA. Down-regulation of RhoA has been shown to prevent retraction of the monocyte tail during migration [41 ], and a similar phenomenon has been reported for T leukemia cells [1 ], but the effect of this manipulation on the spatial restriction of lamellipodia is yet to be determined.

A polarity enforcement mechanism similar to the one we found in lymphocytes might also operate in other motile cell types. In fibroblasts, inactivation of Cdc42 disabled the normal restriction of lamellipodia to the leading edge of the cell [42 ]. In macrophages, inactivation of Cdc42 was found to reduce polarity by permitting lamellipodia to spread more broadly than in controls [12 ]. The underlying mechanisms of these phenomena were not characterized. It would be interesting to determine whether they represented failures in a posterior lamellipodial-disassembly system, as was the case in our lymphocyte system.

Cdc42 activity is also required for the formation of immunological synapses at the proper point, at the lymphocyte MTOC [17 ]. This may prove to be another expression of the Cdc42 function reported in this paper, the exclusion of labile lamellipodial structures from the MTOC region.


   ACKNOWLEDGEMENTS
 
This work was supported in part by NIH Grant CA66985 and in part by Environmental Health Sciences Center Grants P30ES06639 from the National Institute of Environmental Health Sciences and P30CA22453 from the National Cancer Institute. We thank Drs. Alan Hall, Anthony Bretscher, Hergen Spits, Garry Nolan, and Anthony Bretscher for generous gifts of plasmids, vectors, cell lines, and antibodies. We also thank Amanda Labron and Jeffrey Oliver for excellent technical support and Dr. Kamiar Moin and Linda Mayernik for assistance with confocal microscopy.

Received October 20, 2001; revised January 24, 2003; accepted January 27, 2003.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. del Pozo, M. A., Vicente-Manzanares, M., Tejedor, R., Serrador, J. M., Sanchez-Madrid, F. (1999) Rho GTPases control migration and polarization of adhesion molecules and cytoskeletal ERM components in T lymphocytes Eur. J. Immunol. 29,3609-3620[CrossRef][Medline]
  2. Friedl, P., Brocker, E. B., Zanker, K. S. (1998) Integrins, cell matrix interactions and cell migration strategies: fundamental differences in leukocytes and tumor cells Cell Adhes. Commun. 6,225-236[Medline]
  3. von Andrian, U. H., Mackay, C. R. (2000) T-cell function and migration. Two sides of the same coin N. Engl. J. Med. 343,1020-1034[Free Full Text]
  4. Haston, W. S., Shields, J. M. (1984) Contraction waves in lymphocyte locomotion J. Cell Sci. 68,227-241[Abstract]
  5. Sanchez-Madrid, F., del Pozo, M. (1999) Leukocyte polarization in cell migration and immune interactions EMBO J. 18,501-511[CrossRef][Medline]
  6. Brown, M. J., Hallam, J. A., Liu, Y., Yamada, K. M., Shaw, S. (2001) Cutting edge: integration of human T lymphocyte cytoskeleton by the cytolinker plectin J. Immunol. 167,641-645[Abstract/Free Full Text]
  7. Ratner, S., Sherrod, W. S., Lichlyter, D. L. (1997) Microtubule retraction into the uropod and its role in T cell polarization and motility J. Immunol. 159,1063-1067[Abstract]
  8. Serrador, J. M., Nieto, M., Sanchez-Madrid, F. (1999) Cytoskeletal rearrangement during migration and activation of T lymphocytes Trends Cell Biol. 9,228-233[CrossRef][Medline]
  9. Wang, X. Y., Ostberg, J. R., Repasky, E. A. (1999) Effect of fever-like whole-body hyperthermia on lymphocyte spectrin distribution, protein kinase C activity, and uropod formation J. Immunol. 162,3378-3387[Abstract/Free Full Text]
  10. del Pozo, M. A., Sanchez-Mateos, P., Sanchez-Madrid, F. (1996) Cellular polarization induced by chemokines: a mechanism for leukocyte recruitment? Immunol. Today 17,127-131[CrossRef][Medline]
  11. Brown, M. J., Hallam, J. A., Colucci-Guyon, E., Shaw, S. (2001) Rigidity of circulating lymphocytes is primarily conferred by vimentin intermediate filaments J. Immunol. 166,6640-6646[Abstract/Free Full Text]
  12. Allen, W. E., Zicha, D., Ridley, A. J., Jones, G. E. (1998) A role for Cdc42 in macrophage chemotaxis J. Cell Biol. 141,1147-1157[Abstract/Free Full Text]
  13. Haddad, E., Zugaza, J. L., Louache, F., Debili, N., Crouin, C., Schwarz, K., Fischer, A., Vainchenker, W., Bertoglio, J. (2001) The interaction between Cdc42 and WASP is required for SDF-1-induced T lymphocyte chemotaxis Blood 97,33-38[Abstract/Free Full Text]
  14. Wilkinson, P. C. (1986) The locomotor capacity of human lymphocytes and its enhancement by cell growth Immunology 57,281-289[Medline]
  15. Ridley, A. J., Hall, A. (1992) The small GTP-binding protein Rho regulates the assembly of focal adhesions and actin stress fibers in response to growth factors Cell 70,389-399[CrossRef][Medline]
  16. Sander, E. E., van Delft, S., ten Klooster, J. P., Reid, T., van der Kammen, R. A., Michiels, F., Collard, J. G. (1998) Matrix-dependent Tiam1/Rac signaling in epithelial cells promotes either cell-cell adhesion or cell migration and is regulated by phosphatidylinositol 3-kinase J. Cell Biol. 143,1385-1398[Abstract/Free Full Text]
  17. Stowers, L., Yelon, D., Berg, L. J., Chant, J. (1995) Regulation of the polarization of T cells toward antigen-presenting cells by Ras-related GTPase CDC42 Proc. Natl. Acad. Sci. USA 92,5027-5031[Abstract/Free Full Text]
  18. Burbelo, P. D., Drechsel, D., Hall, A. (1995) A conserved binding motif defines numerous candidate target proteins for both Cdc42 and Rac GTPases J. Biol. Chem. 270,29071-29074[Abstract/Free Full Text]
  19. Qiu, R. G., Abo, A., McCormick, F., Symons, M. (1997) Cdc42 regulates anchorage-independent growth and is necessary for Ras transformation Mol. Cell. Biol. 17,3449-3458[Abstract]
  20. Feig, L. A. (1999) Tools of the trade: use of dominant-inhibitory mutants of Ras-family GTPases Nat. Cell Biol. 1,E25-E27[CrossRef][Medline]
  21. D’Souza-Schorey, C., Boettner, B., Van Aelst, A. L. (1998) Rac regulates integrin-mediated spreading and increased adhesion of T lymphocytes Mol. Cell. Biol. 18,3936-3946[Abstract/Free Full Text]
  22. Stam, J. C., Michiels, F., van der Kammen, R. A., Moolenaar, W. H., Collard, J. G. (1998) Invasion of T-lymphoma cells: cooperation between Rho family GTPases and lysophospholipid receptor signaling EMBO J. 17,4066-4074[CrossRef][Medline]
  23. Soede, R. D., Zeelenberg, I. S., Wijnands, Y. M., Kamp, M., Roos, E. (2001) Stromal cell-derived factor-1-induced LFA-1 activation during in vivo migration of T cell hybridoma cells requires Gq/11, RhoA, and myosin, as well as Gi and Cdc42 J. Immunol. 166,4293-4301[Abstract/Free Full Text]
  24. Woodside, D. G., Wooten, D. K., McIntyre, B. W. (1998) Adenosine diphosphate (ADP)-ribosylation of the guanosine triphosphatase (GTPase) Rho in resting peripheral blood human T lymphocytes results in pseudopodial extension and the inhibition of T cell activation J. Exp. Med. 188,1211-1221[Abstract/Free Full Text]
  25. Verschueren, H., De, B. P., De, B. J., Dewit, J., Aktories, K., Just, I. (1997) ADP-ribosylation of Rho-proteins with botulinum C3 exoenzyme inhibits invasion and shape changes of T-lymphoma cells Eur. J. Cell Biol. 73,182-187[Medline]
  26. Tominaga, T., Sugie, K., Hirata, M., Morii, N., Fukata, J. (1993) Inhibition of PMA-induced LFA-1-dependent lymphocyte aggregation by ADP ribosylation of the small molecular weight GTP-binding protein, Rho J. Cell Biol. 120,1529-1537[Abstract/Free Full Text]
  27. Krawczyk, C., Penninger, J. M. (2001) Molecular motors involved in T cell receptor clusterings J. Leukoc. Biol. 69,317-330[Abstract/Free Full Text]
  28. Yoshida, H., Tomiyama, Y., Ishikawa, J., Oritani, K., Matsumura, I., Shiraga, M., Yokota, T., Okajima, Y., Ogawa, M., Miyagawa, J., Nishiura, T., Matsuzawa, Y. (2000) Integrin-associated protein/CD47 regulates motile activity in human B-cell lines through CDC42 Blood 96,234-241[Abstract/Free Full Text]
  29. Schmitz, A. A., Govek, E. E., Bottner, B., Van Aelst, L. (2000) Rho GTPases: signaling, migration, and invasion Exp. Cell Res. 261,1-12[CrossRef][Medline]
  30. Chen, X. Q., Tan, I., Leung, T., Lim, L. (1999) The myotonic dystrophy kinase-related Cdc42-binding kinase is involved in the regulation of neurite outgrowth in PC12 cells J. Biol. Chem. 274,19901-19905[Abstract/Free Full Text]
  31. Edwards, D. C., Sanders, L. C., Bokoch, G. M., Gill, G. N. (1999) Activation of LIM-kinase by Pak1 couples Rac/Cdc42 GTPase signalling to actin cytoskeletal dynamics Nat. Cell Biol. 1,253-259[CrossRef][Medline]
  32. Osman, M. A., Cerione, R. A. (1998) Iqg1p, a yeast homologue of the mammalian IQGAPs, mediates cdc42p effects on the actin cytoskeleton J. Cell Biol. 142,443-455[Abstract/Free Full Text]
  33. Qiu, R. G., Abo, A., Martin, G. S. (2000) A human homolog of the C. elegans polarity determinant Par-6 links Rac and Cdc42 to PKC zeta signaling and cell transformation Curr. Biol. 10,697-707[CrossRef][Medline]
  34. Etienne-Manneville, S., Hall, A. (2001) Integrin-mediated activation of Cdc42 controls cell polarity in migrating astrocytes through PKCzeta Cell 106,489-498[CrossRef][Medline]
  35. Wang, F., Herzmark, P., Weiner, O. D., Srinivasan, S., Servant, G., Bourne, H. R. (2002) Lipid products of PI(3)Ks maintain persistent cell polarity and directed motility in neutrophils Nat. Cell Biol. 4,513-518[CrossRef][Medline]
  36. Fukata, M., Watanabe, T., Noritake, J., Nakagawa, M., Yamaga, M., Kuroda, S., Matsuura, Y., Iwamatsu, A., Perez, F., Kaibuchi, K. (2002) Rac1 and Cdc42 capture microtubules through IQGAP1 and CLIP-170 Cell 109,873-885[CrossRef][Medline]
  37. Kroschewski, R., Hall, A., Mellman, I. (1999) Cdc42 controls secretory and endocytic transport to the basolateral plasma membrane of MDCK cells Nat. Cell Biol. 1,8-13[CrossRef][Medline]
  38. Wu, W. J., Erickson, J. W., Lin, R., Cerione, R. A. (2000) The gamma-subunit of the coatomer complex binds Cdc42 to mediate transformation Nature 405,800-804[CrossRef][Medline]
  39. Friedl, P., Zanker, K. S., Brocker, E. B. (1998) Cell migration strategies in 3-D extracellular matrix: differences in morphology, cell matrix interactions, and integrin function Microsc. Res. Tech. 43,369-378[CrossRef][Medline]
  40. Arthur, W. T., Petch, L. A., Burridge, K. (2000) Integrin engagement suppresses RhoA activity via a c-Src-dependent mechanism Curr. Biol. 10,719-722[CrossRef][Medline]
  41. Worthylake, R. A., Lemoine, S., Watson, J. M., Burridge, K. (2001) RhoA is required for monocyte tail retraction during transendothelial migration J. Cell Biol. 154,147-160[Abstract/Free Full Text]
  42. Nobes, C. D., Hall, A. (1999) Rho GTPases control polarity, protrusion, and adhesion during cell movement J. Cell Biol. 144,1235-1244[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Virol.Home page
P. Chen, W. Hubner, M. A. Spinelli, and B. K. Chen
Predominant Mode of Human Immunodeficiency Virus Transfer between T Cells Is Mediated by Sustained Env-Dependent Neutralization-Resistant Virological Synapses
J. Virol., November 15, 2007; 81(22): 12582 - 12595.
[Abstract] [Full Text] [PDF]

Home page
 J. Cell Biol.Home page
A. Gerard, A. E.E. Mertens, R. A. van der Kammen, and J. G. Collard
The Par polarity complex regulates Rap1- and chemokine-induced T cell polarization
J. Cell Biol., March 12, 2007; 176(6): 863 - 875.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
jlb.1001894v1
73/6/830    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow