Chemo Taxis

7/31/2019 Chemo Taxis

1/10

Chemotaxis: moving forward and holding on to the past

Anna Bagorda1, Vassil A. Mihaylov1,2, Carole A. Parent11Laboratory of Cellular and Molecular Biology, Centre for Cancer Research, National Cancer Institute, NIH, Bethesda, Maryland, USA2Laboratory of Molecular Biology and Genetics, Medical University of Sofia, Sofia, Bulgaria

SummaryThe ability of cells to sense external chemical cues and respondby directionally migrating towards them is a fundamental pro-cess called chemotaxis. This phenomenon is essential for manybiological responses in the human body, including the invasion ofneutrophils to sites of inflammation. Remarkably, many of themolecular mechanisms involved in controlling neutrophils che-motaxis arose millions of years ago in the simple eukaryotic or-

ganism Dictyostelium discoideum. Both neutrophils and Dictyoste-

KeywordsDictyostelium, neutrophils, chemotactic signaling

lium use G protein-coupled signaling cascades to mediate che-motactic responses, which are responsible for transducing ex-ternal cues into highly organized cytoskeletal rearrangementsthat ultimately lead to directed migration. By using the geneti-cally and biochemically tractable organism Dictyostelium as amodel system, it has been possible to decipher many of the sig-nal transduction events that are involved in chemotaxis.

Thromb Haemost 2006; 95: 1221

Theme Issue Article

Correspondence to:Carole A. ParentLaboratory of Cellular and Molecular BiologyNational Cancer Institute, National Institutes of Health37 Convent Drive, Bldg.37/Room 2066Bethesda, Maryland208924256 USATel.: +1 301 435 3701, Fax: +1 301 496 8479E-mail: [email protected]

Received July 11, 2005Accepted after revision October 14, 2005

Financial support:A.B. is a recipient of a fellowship from F.I.R.C. (Italian Federation Cancer Research).

Prepublished online December 12, 2005 DOI: 10.1160/TH05070483

2006 Schattauer GmbH, Stuttgart

12

Dictyostelium discoideum:A powerful model for the study of chemotaxis

The capacity of a cell to sense gradients of extracellular chemi-cal signals (chemoattractants) and respond by migrating direc-

tionally towards their source is a fundamental process called che-motaxis. This process is remarkably conserved between mam-malian leukocytes and the simple eukaryoticamoeboid organism

Dictyostelium discoideum (1). The strong similarities that thesetwo systems share during chemotaxis and the ease by whichDic-tyostelium cells can be manipulated in the laboratory have ledscientists to use this system to investigate the molecular mech-anisms that regulate chemotaxis. Dictyostelium, is a versatilemodel for genetic, biochemical and cell biological analyses. Itssmall (34 Mbp) haploid genome has been completely sequenced(2) and harbors many genes homologous to those found in highereukaryotes (3). Endogenous genes can be easily disrupted byhomologous recombination, novel genes can be identified using

insertional mutagenesis, and structure-function analyses can beperformed by random mutagenesis approaches. Moreover,strains that can grow in liquid culture are available, making iteasy to obtain up to 1012 clonal cells in just few days. The amoe-

bae are easy to lyse and process for a multitude of biochemical

assays and for protein purification as well as for subcellular frac-tionations. Furthermore, these cells can be easily transfected andthe use of fluorescent probes has allowed in-depth cell biologicalanalyses. For all these reasons, Dictyostelium is perfectly poisedfor studies of fundamental cellular processessuch as cytokinesis,motility, chemotaxis, signal transduction and development.

A life cycle driven by chemotaxisThe amoeba Dictyostelium was first isolated from the soil of a

North Carolina forest by Kenneth Raper during a camping trip in1933 (4). Since then, this system has been used extensively as anexperimental model to study the developmental biology ofhigher organisms. In fact, the amoeba shares many of the proper-ties of more complex creatures, including an extraordinary abil-ity to escape a solitary life in order to initiate a multicellular de-velopmental program (57).

Chemotaxis is essential during the entire life cycle of Dic-tyostelium, which consists of two distinct phases:growth andde-velopment(Fig. 1). The natural habitat of the organism is the soil

where, during thegrowth (vegetative) phase,Dictyostelium existas single, free-living amoeboid cells that feed on bacteria and di-vide by binary f ission (5). During this phase the cells can hunt

bacteria due to their ability to chemotax towards folic acid, a by-product of bacterial metabolism (4). When challenged with ad-

VascularCellS

ignalling


2/1013

Bagorda et al.: Chemotactic signaling in Dictyostelium and neutrophils

verse conditions, such as starvation, Dictyostelium cells stop di-

viding and enter a differentiation program that leads the organ-ism into the developmentalphase of its life cycle where a newlyexpressed set of proteins shifts the chemotactic sensitivity of thecells from folic acid to cAMP (810). Although cAMP acts as aubiquitous intracellular second messenger, in Dictyostelium italso functions extracellularly as a chemoattractant. Indeed, che-motaxis towards cAMP is required for the aggregation of~ 100,000 single cells that later differentiate into a true multicel-lular organism. The end goal of this process is the formation of afruiting body that contains resistant spores that allow the organ-ism to survive the harsh conditions. Upon reappearance of nu-trients these spores germinate to produce single celled amoebaeand the cycle is re-initiated (Fig. 1).

During early development, within 4 hours of starvation, cellsspontaneously begin to produce and release molecules of cAMP.Cyclic AMP binds to specific surface receptors and activates a

plethora of signal transduction pathways that lead to chemotaxis,the synthesis and secretion of additional cAMP, and gene ex-

pression (11). Through an adaptation mechanism, the adenylylcyclase A (ACA), that generates cAMP, shuts down and a se-creted form of phosphodiesterase rapidly degrades the extracel-lular cAMP. This process results in oscillations of cAMP every 6min. This exquisitely regulated series of events generates propa-gating waves of cAMP that attract neighboring cells. Intri-guingly, during their movement toward the aggregation center,the cells align themselves in a head to tail fashion and form char-

acteristic streams (12, 13). After 10 hours, as many as 105 cells

aggregate into a mound, where the cells differentiate into eitherprestalk or prespore cells (14). During this later stage of devel-opment, the aggregate behaves as a multicellular organism: themound gradually elongates into a slug structure that migratesalong heat or light gradients in search of the perfect environmentfor further development. The prestalk cells move forward to thefront of the slug, while the prespore cells stay behind in the back.Once in an optimal environment, the slug ceases migration androunds up: the prestalk cells push their way down from the frontof the stalk, giving the appearance of a Mexican hat (Fig. 1).Then the stalk grows, by successive addition of prestalk cells,and the prespore cells are gradually elevated to the top of thestalk. During this process, known as culmination, prespore and

prestalk cells terminally differentiate into either viable spores, ordead stalk cells (15). During slug migration and fruiting bodyformation cAMP is continuously secreted in a pulsatile fashion a process that is essential to coordinate cell movement and shapechanges necessary to build the multicellular fruiting body (16).The final act of development consists of encapsulation of thespores: they lose water and build a thick wall, producing a barrierfor molecules the size of proteins, butnot for water or other smallmolecules (15). The amoebae contained in this coat are not dor-mant; they are sensitive to food availability and osmolaritywhere high concentrations of osmolytes keep them as spores and

prevent germination. Although little is known about spore ger-mination, it is clear that spores continuously sample their en-

Figure 1: A schematic representation of the life cycle ofDic-tyostelium. Morphological changes occurring during the growth and de-velopmental phases are depicted. Next to each stage of development,the corresponding time (in hours) after the beginning of starvation is

shown. The end goal of the differentiation process is the formation of amulticellular fruiting body containing resistant spores. Under favorableconditions, the spores germinate to produce single celled amoebae andre-initiate the cycle (see text for details).


3/1014


vironment through cAMP-dependent signals (17), until theyfind the appropriate conditions to re-initiate their life cycle as in-dependent single cells.

A life cycle driven by conserved signalsDictyostelium and neutrophil chemotaxis is controlled by signaltransduction pathways mediated through the activation of G-pro-teins coupled receptors (GPCRs). GPCRs constitute a large

family of seven transmembrane receptor proteins that mediatesignals by activating bound intracellular heterotrimeric G pro-teins. GPCR-mediated signaling is highly conserved (18) andrepresents a primary mechanism by which cells sense changes inthe external environment and convey this information to the in-tracellular compartment.

The initial step of chemotactic signaling in Dictyosteliumcells is binding of the chemoattractant, folic acid or cAMP, to its

Figure 2: A schematic representation of chemotactic signalingin Dictyostelium and mammalian neutrophils. The middle panelshows a cell chemotaxing towards a source of chemoattractant. Thechemotaxing cell has two morphologically and biochemically distinctcompartments, the front and the back, regulated by distinct signalingevents. Localized polymerization of F-actin at the side of the cell facingthe highest concentration of chemoattractant leads to the extension of aleading pseudopod. F-actin polymerization is suppressed at the sides andthe back of the cell, ensuring efficient migration towards the source of

the signal. The back of the cell is enriched in actomyosin that contractsfor retraction.The upper and lower panels depict in greater detail themajor signaling events that take place at the front and at the back of che-motaxing Dictyostelium (2A, B) and neutrophils (2C, D). The dashedstimulatory and inhibitory arrows between the front and back boxesrepresent the dynamic signaling interactions between the front and theback of chemotaxing cells (e.g. CRAC ACA,Akt PAKa in Dictyoste-lium; Cdc42 RhoA in neutrophils) (see text for details).

VascularCellS

ignalling


4/1015


cognate GPCR (Fig. 2B). This activates intracellular signalingpathways that mediate, among other responses, the directionalmigration toward the source of the chemoattractant. A similarsystem operates in mammalian neutrophils, where the formy-lated tripeptide, fMLP (N-formyl-L-methionyl-L-leucyl-L-phenylalanine), which is released by bacteria, the complement

factor C5a, the platelet-activating-factor (PAF), and chemokinesbind GPCRs and activate highly conserved downstream intracel-lular pathways (1) (Fig. 2D). TheDictyostelium GPCRs that bindcAMP are known as cARs (cAMP receptors). Four differentmembers of this family have been cloned: cAR1, cAR2, cAR3,cAR4 (1921). They are differentially expressed during devel-opment and display distinct binding affinities for cAMP, withcAR1 and cAR3 possessing the highest affinity binding sites(22).The first cAR to be expressed during development is cAR1;its function is partially overlapped by cAR3, which is expressedslightly later. cAR1 deficient mutants arrest in early devel-opment because of their inability to carry out chemotaxis. Cellslacking both cAR1 and cAR3 are completely insensitive to

cAMP (23). Cells that lack either cAR2 (the third expressed) orcAR4 (the fourth cAR to be expressed), despite their ability toaggregate, arrest during the multicellular stage of development(24, 25).

GPCRs are coupled to heterotrimeric G proteins composedof, andsubunits. In response to stimulation, the GPCRs ac-tivate the exchange of GDP for GTP on the subunit leading tothe dissociation of the GTP-bound-subunit from the dimer,which can both modulate downstream effectors. In mammaliancells, there are 4 major families of heterotrimeric G proteins dis-tinguishable by their a subunit isoforms: Gs, Gi/o, Gq/11,G12/13 (26). Gs activates all mammalian isoforms of adenylylcyclase (AC) resulting in increased levels of cAMP, while Gi is

also linked to AC, inhibiting a subset of AC isoforms. Gq acti-vates phospholipase C (PLC) isoform, which catalyzes theconversion of phosphatidylinositol 4,5-bisphosphate to diacylg-lycerol (DAG) and inositol trisphosphate (IP3). The most re-cently identified class of heterotrimeric G proteins includes theG12/13 family, which activates the small GTPases Ras and Rho.

Dictyostelium cells differentially express 14 distinct G subun-its, all closely related to the mammalian Gi family (27). Gen-etic studies (28) and FRET analyses (29) have shown that themain isoform linked to cAR1 is G2. The function of the otherG subunits is not clear, except for G4 and G9. It has beenshown that G4 is linked to the folic acid receptor (30) and thatG9 acts as a regulator of chemotactic response, attenuating the

activities of multiple pathways downstream of the chemoattract-ant receptor activation (31).

The mammalian GPCRs for chemoattractants and chemo-kines are mostly coupled to Gi, although neutrophils also exhibitG12/13-mediated chemotactic processes (26, 32). With the excep-tion of G12/13, which directly mediates signaling, Gi-regulatedchemotactic responses are mainly controlled through the Gsubunits, released following receptor activation (33). Mamma-lian cells have 5 different subunits and 14 distinct subunits,which do not pair indiscriminately, but combine selectively (34).The agonist-inducedactivation of mammalian chemokine recep-tors promotes the release of G, which is responsible for the ac-tivation of PLCb2 and phosphatidylinositol-3 kinase (PI3K) .

Gene deletion studies in mice revealed that PI3Kplays a crucialrole in the activation of downstream phagocyte function, includ-ing chemotaxis and oxygen radical formation (35, 36). PLC2,on the other hand, is important in chemoattractant-mediatedsuperoxide production but does not appearto be required to regu-late chemotaxis (37, 38). However, Ca++ signals may contribute

to chemotaxis as cyclicADP-ribose-induced Ca

++

mobilizationshave been reported to positively regulate neutrophil chemotaxistoward a discreet subset of chemokines (39, 40).

In Dictyostelium, the major effectors controlling chemotaxisare also predominantly regulated by G. In the early stage of

Dictyostelium development, cAMP interacts with cAR1 (the keyreceptor involved in chemotaxis at this stage) and stimulates thedissociation of G2 from G, where two and one isoformsare present in the genome. While PI3K activation, plays a criticalrole in the chemotactic response (4143), the involvement ofPLC activity appears to be dispensable (44). Intriguingly, Dic-tyostelium cells use alternative pathways to generate IP3 and thechemotactic behavior of cells lacking the putative IP3 receptor

and the role of Ca++

influx remains to be clarified (4548).Although heterotrimeric G proteins are essential for chemo-

taxis, the binding of cAMP to cAR1 induces responses that areindependent of G proteins (49, 50). It has been shown thatcAMP-mediated calcium influx (51), in Dictyostelium, still oc-curs in the absence of functional G proteins. In addition, severalother cAR1-mediated responses, including G2 phosphory-lation (52), loss of ligand-binding (53), ERK2 activation (54)and various gene expression events take place in a G protein-in-dependent manner. It has been shown that differentiated HL-60cells (a neutrophil-like cell line) also exhibit Gi-independentresponses to chemoattractant stimulation. Under pertussis toxin(PTX) treatment, which specifically ADP ribosylates the Gi

subunit thereby preventing the dissociation of the Gcomplex,these cells lose several chemokine responses, such as actin poly-merization and PKB phosphorylation (32) but still display awell-defined uropod-like structure at their back, which is a fun-damental component of cell polarity and chemotaxis (see

below).

Chemotaxis requires signaling events to bespatially confined

Seeing is believingDictyostelium and neutrophils possess the extraordinary ability

to precisely detect and respond to very shallow chemoattractantgradients (as low as 2% difference across their length) (55, 56).They do so by amplifying very small receptor occupancy differ-ences into highly polarized intracellular events that give rise to adramatic redistribution of cytoskeletal components. F-actin islocally polymerized at the front and actomyosin is localized tothe back of the cells (Fig. 2) (57). This asymmetricenrichment ofcytoskeletal components leads to cellular polarity, a prerequisitefor migration. The exposure of Dictyostelium cells and neutro-

phils to gradients of chemoattractants induces a rapid change inpolarity through the extension of anterior pseudopods. Pseudo-pod extension occurs through increased F-actin polymerizationand is mediated by theArp 2/3 complex, a sevensubunit complex


5/1016


that binds the sides of pre-existing actin filaments and inducesthe formation of branched polymers (58). As cells require asingle pseudopod for optimal migration, an equally importantcomponent of chemotaxis is the suppression of lateral pseudo-

pods. This event is promoted by myosin II filaments, which areconcentrated at the rear of migrating cell and provide the power

to retract the back. When considering this chemoattractant-me-diated acquisition of cellular polarity and migration, a funda-mental question arises: how do cells transduce shallow externalchemical gradients into an highly polarized cytoskeletal reorgan-ization?

Deciphering the signaling and cytoskeletal dynamics thatoccur during chemotaxis has been possible with the introductionof an imaging technique based on a natural fluorescent protein,the green fluorescent protein (GFP), produced by the biolumi-nescent jellyfishAequoreaVictoria. The cloning of GFP in 1992(59) marked a new era, as it quickly became a convenient andversatilereporter for use in cell imaging. GFP can be used asa re-

porter by itself or, most interestingly, can be fused in-frame to a

gene of interest, therefore permitting visualization of the dy-namic localization of proteins in live migrating cells. GFP wasthe only representative of the family of GFP-like fluorescent

proteins until 1998, when blue, cyan, yellow (60) and, very re-cently, red emissions (61), have been successfully generatedfrom theAequorea GFP to significantly expand the range of co-lors available for cell biological applications. The GFP mutantsallow studies with living cells to be extended to more than one in-tracellular protein, in multi-color labeling experiments. Thisgave rise to more sophisticated imaging techniques, such as Flu-orescence Resonance Energy Transfer (FRET) (62) to study thedynamics of protein-protein interactions. Moreover, newly de-veloped techniques, such as Fluorescence RecoveryAfter Photo-

bleaching (FRAP), provide tools for studying the trafficking anddiffusion kinetics of proteins (63).

Signals are amplified downstream of receptorsand G proteinsThe binding of chemoattractants to their surface receptors repre-sents the first event in a series of steps that eventually give rise toa polarized morphology and directed migration. The fact thatcells become highly polarized when exposed to shallow gradi-ents of chemoattractants, implies that the external signals must,at some point, be amplified to allow for distinct front and back

behaviors. As receptors represent the link between external sig-nals and the internal signaling machinery, it had been proposed

that this amplification could arise from the asymmetric distribu-tion of the receptors themselves. However, studies in both Dic-tyostelium and neutrophils, have established that chemoattract-ant receptors are uniformly distributed on the surface of chemo-taxing cells. Indeed, using a cAR1-GFP fusion construct, it wasshown that, in a chemoattractant gradient, cAMP receptors re-main uniformly distributed around the cell periphery (64). Simi-larly, Bourne et al. (65) later demonstrated that C5a receptorsfused to GFP are uniformly distributed showing no apparent in-creases anywhere on the plasma membrane of polarized differ-entiated PLB-985 cells, a neutrophil-like leukemia cell line. Al-though this uniform distribution, which allows cells to quicklyrespond to highly dynamic chemoattractant gradients is con-

served, it does not appear to apply to all GPCR-mediated che-motactic events. In some cases the receptors redistribute to theleading edge, as has been reported for CXCR4, in hematopoietic

progenitor cells (66), and CCR5, in motile Jurkat cells (67).These results suggest that the mechanisms regulating chemotac-tic responsiveness may vary among different cell types and dif-

ferent receptors.A similar line of investigation established that the amplifi-cation of the signal occurs downstream of heterotrimeric G pro-teins. By studying the cellular distribution of G-GFP in Dic-tyostelium, Jin and collaborators (68) showed that Gsubunits,like chemoattractant receptors, remain mostly uniformly dis-tributed on the plasma membrane during chemotaxis. Consider-ing that actin-binding proteins and F-actin accumulate in extend-ing pseudopods at the leading edge ofDictyostelium and neutro-

phils, the uniform distribution of both receptors and G proteinssuggest that the signals that lead to actin polymerization mustoccur by the selective activation of signaling pathways down-stream of receptors and G proteins and upstream of actin (57,

69).

PI3K/PTEN: Key components that spatially confinesignals in chemotaxing cells

The involvement of PI3K/PTEN as regulators of chemotaxiscame from studies of two PH domain-containing proteins:CRAC and PKBIn Dictyostelium binding of cAMP to cAR1 leads to the produc-tion of cAMP via the activation of the adenylyl cyclase ACA, a12 transmembrane protein related to the mammalian G protein-coupled adenylyl cyclases. While part of the synthesized cAMPremains inside the cell and activates PKA to regulate gene ex-

pression, the majority of the produced cAMP is secreted and asignal relay loop is initiated. The activation of ACA requires, inaddition to the Gsubunits, a regulator called CRAC (cytosolicregulator of adenylyl cyclase), which contains a pleckstrinhomology (PH) domain at its N-terminus (70) cells lackingCRAC show no receptor-mediated ACA activation. Under un-stimulated conditions CRAC is found in the cytoplasm. How-ever, following a uniform chemoattractant addition, CRACrapidly and transiently associates with the plasma membrane.Remarkably, when CRAC-GFP expressing cells are placed in agradient of chemoattractant, CRAC is specifically and persist-ently recruited to the leading edge (71).This redistribution is me-diated by the binding of the PH domain of CRAC to PtdIns (3, 4)

P2 and PtdIns (35) P3 (3-PI) (41, 72). In addition to its role inACA activation, it was recently determined that CRAC is in-volved in regulating chemotaxis (72). Cells lacking CRAC movevery slowly in gradients of chemoattractants and display broadfronts. However, as they are still capable of polymerizing corticalactin in response to chemoattractant stimulation, it was deter-mined that the chemotaxis defect is not due to alterations in themotility machinery but to defects in translating the external che-moattractant gradient into directed pseudopod extension. Inter-estingly, using mutagenesis analyses, it was established thatCRAC regulates ACA activation and chemotaxis independently.Mutants harboring C-terminal deletions, although defective intheir ability to activate adenylyl cyclase, are still able to chemo-

VascularCellS

ignalling


6/1017


tax, similarly to cells lacking ACA (72). Conversely, a CRACpoint mutant that can no longer interact with 3-PI shows defectsin both ACA activation and chemotaxis. These observationsunderline a key role for the spatial localization of CRAC in twodistinct, yet related, responses: chemotaxis and ACA activation.Searches of themammalian genomedatabases have not led to the

identification of CRAC homologues.Another PH-domain containing protein, Akt/PKB, associ-ates with the leading edge of bothDictyostelium (73) and neutro-

phils (74), again underlining the evolutionary conservation ofthe chemotactic signaling cascade. Studies performed with Dic-tyostelium have established that Akt/PKB is involved in main-taining cell polarity as well as proper chemotaxis. Cells lackingAkt/PKB are less polar and show significant defects in their abil-ity to localize PAKa (a homologue of mammalian p21-activatedkinases) and myosin II to their posterior (75). As active PAKa isrequired for myosin II assembly to the rear of moving cells, it was

proposed that the role of Akt/PKB, once translocated and acti-vated at the leading edge, is to phosporylate PAKa, which moves

to the posterior of chemotaxing cells where it co-localizes andpromotes myosin II assembly (Fig. 2A). The role of Akt/PKB inneutrophil chemotaxis has yet to be determined.

PI3K and PTEN regulate chemotaxisThe chemoattractant-mediated stimulation of PI3K, togetherwith the enzymatic activity of the phosphatidylinositol 3-phos-

phatase PTEN (Phosphatase and Tensin Homolog Deleted onChromosome 10), regulate the production of 3-PI. While PI3K

phosphorylates the 3 phosphate, PTEN removes the 3 phos-phate thereby counteracting the action of PI3K. Pharmacologicalinhibition of PI3Ks blocks chemotaxis and migration to variousdegrees in Dictyostelium, as well as in a variety of mammalian

cells (reviewed in [69]). Moreover, Dictyostelium cells lackingPI3K1 and PI3K2, two of the three class I PI3K expressed in thisorganism, and neutrophils harvested from mice lacking PI3K,

which is the principal isoform responsible for 3-PI production inthese cells, have defects in polarity and show reduced efficiencyof chemotaxis (3537, 76). As expected, Dictyostelium cellslacking PI3K activity have strongly diminished levels of 3-PIfollowing chemoattractant addition, as measured by a dramati-cally reduced recruitment of CRAC to the plasma membrane

(76). Conversely, cells lacking PTEN show a persistent associ-ation of CRAC to the plasmamembrane following chemoattract-ant stimulation and have strong polarity defects. These cells

polymerize F-actin robustly in response to chemoattractantstimulation and, when placed in a chemoattractant gradient,move slowly and with less direction, compared to wild type cells(77). Taken together, these studies suggest an involvement of3-PI in directional migration.

The cellular distribution of PI3K and PTEN in chemotaxingcells is mutually exclusive. Studies in chemotaxing Dictyoste-lium cells have shown that GFP-tagged PI3K1 and PI3K2 arespecifically recruited to the leading edge (42). Simultaneously,PTEN-GFP is selectively removed from the leading edge, and re-

mains associated with the plasma membrane at the sides andback of cells (42, 77). These responses seem to be conserved inmammalian leukocytes, as Li et al. have shown that PTEN is re-stricted to the rear of chemotactic neutrophils (78). This targetedcellular distribution of PI3K and PTEN therefore promotes therestricted production of 3-PI at the leading edge of chemotaxingcells and serves to amplify external chemical gradients intosharply polarized internal signals (Fig. 2).

Thepolarizing signal responsible for the recruitment of PI3Kand PTEN remains to be fully understood, although recentstudies have shown that the actin cytoskeleton might be involved(79, 80). Work performed with Dictyostelium has shown that theactivation of PI3K requires an input from a small Ras GTPase

and it was recently demonstrated that RasG, one of five RasGTPases expressed in Dictyostelium, is activated exclusively atthe leading edge of chemotaxing cells (42, 79, 81). It was pro-

Figure 3: Dictyostelium a paradigm for signal relay. The imagerepresents the current model for chemotactic signal propagation in Dic-tyostelium. Upon stimulation with chemoattractant (depicted as reddots), the PH-domain containing cytosolic protein CRAC is recruited tothe leading edge (CRAC-GFP localization is shown on the far top right

panel). By a mechanism that is still not elucidated CRAC activates ACA,which is enriched at the back (ACA-YFP localization is shown on the farbottom right panel). Produced cAMP is proposed to be released fromthe back to locally recruit neighboring cells (see text for details).


7/1018


posed that Ras activation promotes leading edge formation bystimulating basal PI3K activity, which leads to F-actin polymer-ization, and subsequently, a feedback loop, mediated through lo-calized F-actin, recruits more cytosolic PI3K to the leading edgeto amplify the signal (79). Similarly, an actin-dependent feed-

back loop has previously been proposed to regulate PH-domain

recruitment in neutrophils, where treatment with latrunculin, aninhibitor of actin polymerization, diminishes the recruitment ofthe PH-domain containing protein Akt/PKB to the plasma mem-

brane, thereby decreasing further downstream activation (80).

Chemoattractant signaling to actin and myosin II

The cytoskeletal rearrangements induced by chemoattractantsare mediated by Rho familyGTPases, including Rac, Cdc42, andRho. Thus, the chemoattractant-mediated spatiotemporal acti-vation of Rho GTPases is a key regulatory event during chemo-taxis. GTPases are GTP binding proteins that cycle between in-active GDP-bound and active GTP-bound forms. GTPase-acti-

vating proteins (GAPs) keep the enzymes in their inactive formby stimulating their low intrinsic GTP hydrolytic activity. Con-versely, guanine nucleotide-exchange factors (GEFs), whichcatalyze the release of the bound GDP for GTP, are essential toactivate the GTPases. It has been demonstrated that Rac is re-sponsible for mediating leading edge formation, stimulatingF-actin assembly at the front of both neutrophils andDictyoste-lium (8284). In neutrophils, Cdc42 is proposed to be requiredexclusively for directional migration (32) and for maintainingleading edge stability (83) in chemoattractantgradients. Interest-ingly, no Cdc42 homologues are found in Dictyostelium cells,thereby suggesting that alternative mechanisms should exist tostabilize the leading edge during directional migration.

F-actin polymerization at the frontActin polymerization at the leading edge of chemotaxing cells isinitiated through the generation of newly formed barbed ends bythe Arp2/3 complex, whose activity is controlled by the adaptor

proteins WASP (Wiskott-Aldrich Syndrome Protein) and SCAR(Suppressor of cAR).Through their CRIB domains (Cdc42/Rac-interacting binding domains),WASP/SCAR family members actas linkers between Rac/Cdc42 and actin filament formation (85)(Fig. 2B, D). GEFs for Rac and Cdc42 harbor PH domains andare therefore perfectly positioned to spatially control actin poly-merization during chemotaxis (Fig. 2) (86).

In neutrophils, the RacGEFs P-Rex1 (PtdIns(3,4,5)P3-de-

pendent Rac exchanger), whose activity is directly activated by3-PI and G (87), and Vav1, which is regulated by PtdIns(3,4,5)P3, seem to be responsible for Rac activation during che-motaxis (88, 89). Another GEF, PIXa (PAK-interacting ex-change factor), specific for Cdc42, has also been proposed tospatially control actin polymerization during chemotaxis. Li etal. (78) have described a pathway where Cdc42 is activated at thefront of moving cells, through the formation of a complex withPAK1 and PIXa. They propose that chemoattractants induce therelease of Gsubunits (from Gi), which binds to PAK1-PIXa,thereby recruiting the ternary complex to the leading edge. PIXathen activates Cdc42, which in turns activates PAK1 to mediateCdc42 activation in a feedback positive loop. Here again, 3-PI

appear to have a role in controlling the spatial activation ofCdc42, albeit unclear at this point, as Yoshii et al. (90) haveshown that PI3K products activate PIXa.

InDictyostelium, among the 18 isoforms of RhoGTPases ex-pressed, only the role of RacB has been specifically assessed dur-ing chemotaxis (84). It is required for proper chemotaxis and

upon chemoattractant stimulation it shows two peaks of acti-vation that correspond to the previously observed two peaks ofchemoattractant mediated F-actin polymerization (91). The sec-ond peak of RacB activation, correlating to F-actin assemblyduring pseudopod extension, is dependent on the PI3K pathway,linking PI3K to F-actin polymerization at the leading edge. Fur-thermore, RacB binds with strong affinity to the CRIB domainoftheDictyosteliumWASP. WASP has been recently shown to belocalized to the leading edge of chemotaxingDictyostelium cellsvia PI3K products a mechanism which is also observed inmammalian fibroblasts (92, 93). Taken together these studies

provide evidence that RacB is a possible candidate to stabilizethe leading edge formation in Dictyostelium cells.

Myosin II assembly at the backThe signals that control the retraction of the back of a chemo-taxing cell inDictyostelium andneutrophils appear to be distinct.During neutrophil migration, Rho activity is required to regulatemyosin-II-mediated contractility necessary for tail retraction(94), through a pathway that involves the action of its effectorROCK (Rho kinase) (95) (Fig. 2C). In differentiated HL-60cells, stimulated with fMLP, RhoA is distributed predominantlyto the sides and rear of stimulated cells, where it is thought to beactivated by G12/13 (32, 96). Interestingly, the cytoskeleton alsoappears to regulate the back of migrating cells, as inhibitors ofmicrotubule filament formation have recently been shown to en-

hance RhoA activity, leading to enhanced polarity and impairedchemotactic efficiency under chemoattractant stimulation (97,98). Moreover, Li et al. presented evidence that activated Cdc42,at the front of chemotactic neutrophils, is required for the local-ization of activated RhoA at the back, where, through ROCK, itmediates the phosphorylation of PTEN. This pathway is then in-volved in controlling the localization and activation of PTEN tothe back of chemotaxing cells (99).

InDictyostelium Rho homologues are not present and two al-ternativepathwayshave beenimplicated to spatiallycontrol backretraction. Oneof these involves the phosphorylation of PAKabyAkt/PKB, which, as already mentioned, is required for myosin IIassembly at theback of thecells (75).The other pathwayinvolves

cGMP, which has also been shown to be critical for myosin II as-sembly in Dictyostelium (Fig. 2A). The involvement of this sol-uble nucleotide in chemotaxis regulation emerged from studiesusing cell lines possessing either high or low cGMP levels. It wasshown that high cGMP levels give rise to increased myosin II

phosphorylation and cytoskeletal association as well as im-proved chemotaxis. On the other hand, low levels of cGMP gen-erally produce the opposite effect (100, 101). Interestingly, thesoluble guanylyl cyclase, the enzyme responsible for cGMP pro-duction following chemoattractant stimulation, has recently

been shown to be enriched at the leading edge of cells (102), al-though its product mediates myosin filament formation at the

back.

VascularCellS

ignalling


8/1019


References1. Parent CA.Making allthe right moves: chemotaxisin neutrophils and Dictyostelium. Curr Opin Cell Biol2004; 16: 413.2. EichingerL, PachebatJA,GlocknerG et al.Thege-nome of the social amoeba Dictyostelium discoideum.Nature 2005; 435: 4357.3. Kreppel L, Fey P, Gaudet P et al. dictyBase: a newDictyostelium discoideum genome database. NucleicAcids Res 2004; 32(Database issue): D3323.

4. Kessin RH. Dictyostelium: evolution, cell biology,and the development of multicellularity. Cambridge:Cambridge University Press; 2001.5. Escalante R, Vicente JJ. Dictyostelium discoideum:a model system for differentiation and patterning. Int JDev Biol 2000; 44: 81935.6. Chisholm RL, Firtel RA. Insights into morphogen-esis from a simpledevelopmentalsystem.Nat Rev MolCell Biol 2004; 5: 53141.

7. Kimmel AR, Firtel RA. Breaking symmetries:regulation of Dictyostelium development through che-moattractant and morphogen signal-response. CurrOpin Genet Dev 2004; 14: 5409.8. Iranfar N, Fuller D, Loomis WF. Genome-wide ex-pression analysesof gene regulation during early devel-opment of Dictyostelium discoideum. Eukaryot Cell2003; 2: 66470.

These results highlight an emerging theme duringDictyoste-lium and neutrophil chemotaxis where a requirement for dy-namic signaling cross-talk between the front and back of cells isessential to properly orchestrate the complex cytoskeletal rear-rangements that are required for migration.

Signal relayAs discussed above, the binding of chemoattractants to GPCRstriggers an exquisitely regulated series of signaling events thatlead to the spatial activation of the cytoskeletal apparatus and di-rected migration. Chemotactic cells possess the ability to relaythe chemotactic signal to surrounding cells through productionand release of more attractants, which act in an autocrine and

paracrine fashion to spread the chemotactic response. As withthe signals that lead to cell polarity and motility, this signal relayresponse is shared by Dictyostelium and neutrophils.

Signal relay is essential for the development of Dictyoste-lium. Upon starvation, these cells spontaneously aggregate by

producing, secreting, and detecting cAMP. The binding of cAMPto cAR1 leads to the synthesis of additional cAMP through theactivation of ACA. As the cells are chemotaxing to the secretedcAMP signal, they recruit neighboring cells to the aggregate fol-lowing each other and forming head-to-tail chains, calledstreams. As expected, cells lacking ACA do not form streamswhen subjected to an external cAMP gradient. Remarkably, itwas shown that ACA-YFP is highly enriched at the back of che-motaxing cells (Fig. 3).This finding, coupled with other mutantanalyses data, led Kriebel et al. (103) to propose that streamingdepends not only on the presence of ACA, but also on its propercellular distribution at the back of cells. This enrichment would

provide a compartment from which cAMP is secreted to locally

act as a chemoattractant. Interestingly, as mentioned, the essen-tial regulator of ACA, CRAC, is recruited to the front of chemo-taxing cells. Therefore the activation of ACA, initiated at thefront of cells, may reach a maximum at the back where it is per-fectly localized to relay the chemotactic signal to neighboringcells.

The ability to relay signals is conserved in mammalian neut-rophils. IL-8 (104, 105) and leukotriene B4 (LTB4) (106) repre-sent good examples of chemoattractants released by neutrophilsfollowing chemotactic activation. Several chemoattractants, in-cluding fMLP, C5a, PAF and LTB4, stimulate neutrophil produc-tion and secretion of IL-8 (107).Although the secretory pathwayof IL-8 remains to be established, there is evidence that this pro-

cess could represent a mechanism used by neutrophils to attracta greater number of leukocytes to sites of inflammation (108)and to prime the superoxide production triggered by other che-

mokines (109). Furthermore, it has been demonstrated that thegreater part of newly synthesized cytosolic IL-8 remains in thecells (110, 111), where it may constitute a source of IL-8 releasedwhen apoptotic neutrophils are digested by macrophages. Thesynthesis of LTB4 in neutrophils involves a complex series ofenzymatic reactions that are initiated at the plasma membrane

(112). Under fMLP and C5a stimulation, a cytoplasmic phosp-holipase A, cPLA2, translocates to the nuclear membrane ofneutrophils, where it hydrolizes membrane bound lipids to formarachidonic acid. Simultaneously, a 5-lipoxygenase (5-LO) is re-cruited to the nuclear membrane where it associates with the5-LO activating protein (FLAP) and acts on arachidonic acid togenerate LTA4. LTA4, by means of LTA4 hydrolase, is finallytransformed into LTB4, which is then secreted from the cells.Thesecreted LTB4binds to a specific class of Gi-coupled GPCRs andis part of an auto-amplification loop that mediates adhesion, mi-gration and further accumulation of activated neutrophils to sitesof inflammation (113, 114). Whether LTB4 secretion is spatiallylocalized during neutrophil chemotaxis remains to be deter-

mined.

Perspectives

Dictyostelium and mammalian neutrophils, although separatedby millions of years of evolution, show remarkable similarities inthe molecular mechanisms used to regulate the cytoskeletal rear-rangements involved in directional sensing and migration. In-deed, Dictyostelium provides a simple model system whereidentical single cells respond to one major chemoattractant.

Neutrophils, on the other hand, respond to a multitude of attrac-tants that are generated from a wide variety of sources, including

bacterially secreted formylated peptides, products of the com-

plement cascade, and a plethora of chemokines derived fromhost endothelial, epithelial, and stromal cells. While signalcross-talk and integration is more complex in these cells, it isclear that the molecular mechanisms involved in generatingasymmetric signals during neutrophil chemotaxis originatedmany years ago when Dictyostelium cells developed mech-anisms to resist harsh environmental conditions. Further studieswill continue to decipher how multiple signals are orchestratedinto persistent and directed chemotaxis responses.

Acknowledgements

The authors would like to thank the members of the Parent laboratory fordynamic and insightful discussions and for very useful comments on themanuscript. We are also grateful to the editorial assistance of the NCI CCRFellows Editorial Board.


9/1020


9. Saran S, Meima ME, Alvarez-Curto E et al. cAMPsignaling in Dictyostelium. Complexity of cAMP syn-thesis, degradation and detection. J Muscle Res CellMotil 2002; 23: 793802.10. Kriebel PW, Parent CA. Adenylyl cyclase ex-pression and regulation during the differentiation ofDictyostelium discoideum. IUBMB Life 2004; 56:5416.11.

KimmelAR, Parent CA.Thesignal to move: D. dis-coideum go orienteering. Science 2003; 300: 15257.12. Kessin RH. Cell motility: Making streams. Nature2003; 422: 4812.13. Shaffer B. Primitive Motile Systems in Cell Biol-ogy. Academic, NewYork 1964: 387405.14. Kay RR. Chemotaxis and cell differentiation inDictyostelium. Curr Opin Microbiol 2002; 5: 5759.15. Thomason P, Traynor D, Kay R. Taking the plunge.Terminal differentiation in Dictyostelium. TrendsGenet 1999; 15: 159.16. Meima M, Schaap P. Dictyostelium development-socializing through cAMP. Semin Cell Dev Biol 1999;10: 56776.17. Virdy KJ, Sands TW, Kopko SH et al. High cAMPin spores of Dictyostelium discoideum: associationwith spore dormancy and inhibition of germination.

Microbiology 1999; 145: 188390.18. Karnik SS, Gogonea C, Patil S et al. Activation ofG-protein-coupled receptors:a common molecularme-chanism. Trends Endocrinol Metab 2003; 14: 4317.19. Klein PS, Sun TJ, Saxe CL, 3 rd et al. A chemoat-tractant receptor controls development in Dictyoste-lium discoideum. Science 1988; 241: 146772.20. Saxe CL, 3rd, Johnson R, Devreotes PN et al.Multiple genes forcell surface cAMP receptorsin Dic-tyostelium discoideum. Dev Genet 1991; 12: 613.21. Saxe CL, 3rd, Johnson RL, Devreotes PN et al. Ex-pression of a cAMP receptor gene of Dictyostelium andevidence for a multigene family. Genes Dev 1991; 5:18.22. Johnson RL,Van Haastert PJ, KimmelAR et al.Thecyclic nucleotide specificity of three cAMP receptors

in Dictyostelium. J Biol Chem 1992; 267: 46007.23. Insall RH, Schaap P, Devreotes PN. Two cAMP re-ceptors activate common signaling pathways in Dic-tyostelium. Mol Biol Cell 1994; 5: 70311.24. Saxe CL, 3rd, Ginsburg GT, Louis JM et al. CAR2,a prestalk cAMP receptor required for normal tipformation and late development of Dictyostelium dis-coideum. Genes Dev 1993; 7: 26272.25. Louis JM, Ginsburg GT, Kimmel AR. The cAMPreceptor cAR4 regulates axial patterning and cellulardifferentiation during late development of Dictyoste-lium. Genes Dev 1994; 8: 208696.26. Wettschureck N,Offermanns S. Mammalian G pro-teins and their cell type specif ic functions. Physiol Rev2005; 85: 1159204.27. Brzostowski JA, Johnson C, Kimmel AR. Galpha-mediated inhibition of developmental signal response.Curr Biol 2002; 12: 1199208.28. Kumagai A, Hadwiger A, Pupillo M et al. Molecu-lar genetic analysis of two G protein subunits in Dic-tyostelium. J Biol Chem 1991; 266: 12208.29. Janetopoulos C, JinT, Devreotes PN.Receptor-me-diated activation of heterotrimeric G-proteins in livingcells. Science 2001; 291: 240811.30. Hadwiger JA, Lee S, Firtel RA. The G alpha sub-unit G alpha 4 couples to pterinreceptors andidentifiesa signaling pathway that is essential for multicellulardevelopment in Dictyostelium. Proc NatlAcad Sci U SA 1994; 91: 1056670.31. Brzostowski JA, Parent CA, Kimmel AR. A Galpha-dependent pathway that antagonizes multiplechemoattractantresponses that regulate directionalcellmovement. Genes Dev 2004; 18: 80515.

32. Xu J, Wang F, Van Keymeulen A et al. Divergentsignals and cytoskeletal assemblies regulate self-or-ganizing polarity in neutrophils. Cell 2003; 114:20114.33. Neptune ER, Bourne HR. Receptors induce che-motaxis by releasing the subunit of Gi, not by acti-vating Gq or Gs. Proc Natl Acad Sci USA 1997; 94:1448994.34.

Kehrl JH. G-protein-coupled receptor signaling,RGS proteins, and lymphocyte function. Crit Rev Im-munol 2004; 24: 40923.35. Hirsch E, Katanaev VL, Garlanda C et al. Centralrole for G protein-coupled phosphoinositide 3-kinasegamma in inflammation.Science2000; 287:104953.36. Sasaki T, Irie-Sasaki J, Jones RG et al. Function ofPI3Kgamma in thymocyte development, T cell acti-vation, and neutrophil migration. Science 2000; 287:10406.37. Li Z, Jiang H, Xie W et al. Roles of PLC-beta2 and-beta3 and PI3Kgamma in chemoattractant-mediatedsignal transduction. Science 2000; 287: 10469.38. Ferretti ME, Nalli M, BiondiC et al.Modulation ofneutrophilphospholipaseC activityand cyclicAMP le-vels by fMLP-OMe analogues. Cell Signal 2001; 13:23340.

39. Partida-Sanchez S, Cockayne DA, Monard S et al.Cyclic ADP-ribose production by CD38 regulates in-tracellularcalciumrelease,extracellular calcium influxand chemotaxis in neutrophils and is required for bac-terial clearance in vivo. Nat Med 2001; 7: 120916.40. Partida-SanchezS, Iribarren P, Moreno-Garcia MEet al. Chemotaxis and calcium responses of phagocytesto formyl peptide receptor ligands is differentiallyregulated by cyclic ADP ribose. J Immunol 2004; 172:1896906.41. HuangYE,Iijima M, Parent CA et al. Receptor-me-diated regulation of PI3Ks confines PI(3,4,5)P3 to theleading edge of chemotaxing cells.Mol Biol Cell 2003;14: 191322.42. Funamoto S, Meili R, Lee S et al. Spatial and tem-poral regulation of 3-phosphoinositides by PI 3-kinase

and PTEN mediates chemotaxis. Cell 2002; 109:61123.43. Comer FI,Parent CA.PI 3-kinasesand PTEN:howopposites chemoattract. Cell 2002; 109: 5414.44. DrayerAL, Van der Kaay J, Mayr GW et al. Role ofphospholipase C in Dictyostelium: formation of inosi-tol 1,4,5-trisphosphate and normal development incells lacking phospholipase C activity. Embo J 1994;13: 16019.45. van Haastert PJ, van Dijken P. Biochemistry andgenetics of inositol phosphate metabolism in Dictyos-telium. FEBS Lett 1997; 410: 3943.46. Van Dijken P, Bergsma JC, Van Haaster t PJ. Phosp-holipase-C-independent inositol 1,4,5-trisphosphateformation in Dictyostelium cells. Activation of a plas-ma-membrane-bound phosphatase by receptor-stimu-lated Ca2+ influx. Eur J Biochem 1997; 244: 1139.47. TraynorD, Milne JL, InsallRH et al.Ca(2+) signal-ling is not required for chemotaxis in Dictyostelium.Embo J 2000; 19: 484654.48. Schaloske RH, Lusche DF, Bezares-Roder K et al.Ca2+ regulationin theabsenceof theiplAgeneproductin Dictyostelium discoideum. BMC Cell Biol 2005; 6:13.49. Milne JL, Kim JY, Devreotes PN. Chemoattractantreceptor signaling, G protein-dependent and -indepen-dent pathways. In: Corbin J, Francis S, editors. SignalTransduction in Health and Disease, Advances in sec-ond messenger and phosphoprotein research. Philad-elphia: Lippincott-Raven 1997: 83103.50. Brzostowski JA, Kimmel AR. Signaling at zero G:G-protein-independent functions for 7-TM receptors.Trends Biochem Sci 2001; 26: 2917.

51. Milne JL, Wu L, Caterina MJ et al. Seven helixcAMP receptorsstimulate Ca2+ entry in theabsenceoffunctional G proteins in Dictyostelium. J Biol Chem1995; 270: 592631.52. Chen MY, Devreotes PN, Gundersen RE. Serine113 is the site of receptor-mediated phosphorylation ofthe Dictyostelium G protein alpha-subunit G alpha 2. JBiol Chem 1994; 269: 2092530.53.

Xiao Z, Yao Y, Long Y et al. Desensitization ofG-protein-coupled receptors. agonist-induced phos-phorylation of the chemoattractant receptor cAR1lowers its intrinsic affinity for cAMP. J Biol Chem1999; 274: 14408.54. Maeda M, Aubry L, Insall R et al. Seven helix che-moattractant receptors transiently stimulate mitogen-activated protein kinase in Dictyostelium. Role of het-erotrimeric G proteins. J Biol Chem 1996; 271:33514.55. Mato JM, Losada A, Nanjundiah V et al. Signalinput for a chemotactic response in the cellular slimemold Dictyostelium discoideum. Proc Natl Acad SciUSA 1975; 72: 49913.56. Tranquillo RT, Lauffenburger DA, Zigmond SH. Astochastic model for leukocyte random motility andchemotaxis based on receptor binding fluctuations. J

Cell Biol 1988; 106: 3039.57. Parent CA, Devreotes PN. A cell's sense of direc-tion. Science 1999; 284: 76570.58. Pollard TD, Borisy GG. Cellular motility driven byassembly and disassembly of actin f ilaments. Cell2003; 112: 45365.59. Prasher DC, Eckenrode VK, Ward WW et al. Pri-mary structure of the Aequorea victoria green-fluor-escent protein. Gene 1992; 111: 22933.60. Tsien RY. The green fluorescent protein. Annu RevBiochem 1998; 67: 50944.61. Shaner NC, Campbell RE, Steinbach PA et al. Im-proved monomeric red, orange and yellow fluorescentproteins derived from Discosoma sp. red fluorescentprotein. Nat Biotechnol 2004; 22: 156772.62. Zhang J, Campbell RE,TingAY et al.Creatingnew

fluorescent probes for cell biology. Nat Rev Mol CellBiol 2002; 3: 90618.63. Reits EA,NeefjesJJ.Fromfixedto FRAP: measur-ing protein mobility and activity in living cells. NatCell Biol 2001; 3: E1457.64. Xiao Z, Zhang N, Murphy DB et al. Dynamic dis-tribution of chemoattractant receptors in living cellsduring chemotaxis and persistent stimulation. J CellBiol 1997; 139: 365374.65. Servant G, Weiner OD, Neptune ER et al. Dy-namics of a chemoattractant receptor in living neutro-phils during chemotaxis. Mol Biol Cell 1999; 10:116378.66. van Buul JD, Voermans C, van Gelderen J et al.Leukocyte-endothelium interaction promotes SDF-1-dependent polarization of CXCR4. J Biol Chem2003; 278: 3030210.67. Gomez-Mouton C, Lacalle RA, Mira E et al. Dy-namic redistribution of raft domains as an organizingplatform for signaling during cell chemotaxis. J CellBiol 2004; 164: 75968.68. Jin T, Zhang N, LongY et al. Localization of the Gprotein complex in living cells during chemotaxis.Science 2000; 287: 10346.69. Iijima M, Huang YE, Devreotes P. Temporal andspatial regulation of chemotaxis. Dev Cell 2002; 3:46978.70. Insall R, Kuspa A, Lilly PJ et al. CRAC, a cytosolicprotein containing a pleckstrin homology domain, isrequired for receptor andG protein-mediated activationof adenylyl cyclase in Dictyostelium. J Cell Biol 1994;126: 153745.

VascularCellS

ignalling


10/1021


71. Parent CA, Blacklock BJ, Froehlich WM et al. Gprotein signaling events are activated at the leadingedge of chemotactic cells. Cell 1998; 95: 8191.72. Comer FI, Lippincott CK, Masbad JJ et al. ThePI3K-mediated activation of CRAC independentlyregulates adenylyl cyclase activation and chemotaxis.Curr Biol 2005; 15: 1349.73. Meili R, Ellsworth C, Lee S et al. Chemoattractant-

mediated transient activation and membrane localiz-ation of Akt/PKB is required for efficient chemotaxisto cAMP in Dictyostelium. EMBO J 1999; 18:2092105.74. Servant G, Weiner OD, Herzmark P et al. Polar-ization of chemoattractant receptor signaling duringneutrophil chemotaxis. Science 2000; 287: 103740.75. Chung CY, Potikyan G, Firtel RA. Control of cellpolarity and chemotaxis by Akt/PKB and PI3 kinasethrough the regulation of PAKa. Mol Cell 2001; 7:93747.76. Funamoto S, Milan K, Meili R et al. Role of phos-phatidylinositol 3' kinase and a downstream pleckstrinhomology domain-containing protein in controllingchemotaxis in dictyostelium. J Cell Biol 2001; 153:795810.77. Iijima M, Devreotes P. Tumor suppressor PTEN

mediates sensing of chemoattractant gradients. Cell2002; 109: 599610.78. Li Z, Hannigan M, Mo Z et al. Directional sensingrequiresG betagamma-mediated PAK1and PIXalpha-dependent activation of Cdc42. Cell 2003; 114:21527.79. Sasaki AT, Chun C, Takeda K et al. Localized Rassignaling at the leading edgeregulatesPI3K, cell polar-ity, and directional cell movement. J Cell Biol 2004;167: 50518.80. Wang F, Herzmark P, Weiner OD et al. Lipid prod-ucts of PI(3)Ks maintainpersistent cell polarity anddi-rected motility in neutrophils. Nat Cell Biol 2002; 4:5138.81. Kae H, Lim CJ, Spiegelman GB et al. Chemoat-tractant-induced Ras activation during Dictyostelium

aggregation. EMBO Rep 2004; 5: 6026.82. vanHaastert PJ, devreotes PN.Chemotaxis: signal-ing the way forward. Nature Review Mol Cell Biol2004; 5: 62634.83. SrinivasanS, Wang F, Glavas S et al.Rac and Cdc42play distinct roles in regulating PI(3,4,5)P3 and polar-ity during neutrophil chemotaxis. J Cell Biol 2003;160: 37585.84. Park KC, Rivero F, Meili R et al. Rac regulation ofchemotaxis and morphogenesis in Dictyostelium.Embo J 2004; 23: 417789.85. Vartiainen MK, MacheskyLM. TheWASP-Arp2/3pathway: genetic insights. Curr Opin Cell Biol 2004;16: 17481.

86. Raftopoulou M, Hall A. Cell migration: RhoGTPases lead the way. Dev Biol 2004; 265: 2332.87. Hill K, Krugmann S, Andrews SR et al. Regulationof P-Rex1 by phosphatidylinositol (3,4,5)-trisphos-phate and Gbetagamma subunits. J Biol Chem 2005;280: 416673.88. Welch HC, Coadwell WJ, Ellson CD et al. P-Rex1,a PtdIns(3,4,5)P3 and Gbetagamma-regulated gua-

nine-nucleotide exchange factor for Rac. Cell 2002;108: 80921.89. Kim C, Marchal CC, Penninger J et al. The hemo-poietic Rho/Rac guanine nucleotide exchange factorVav1 regulates N-formyl-methionyl-leucyl-phenylal-anine-activated neutrophil functions. J Immunol 2003;171: 442530.90. Yoshii S, Tanaka M, Otsuki Y et al. alphaPIX nu-cleotide exchange factor is activated by interactionwith phosphatidylinositol 3-kinase. Oncogene 1999;18: 568090.91. Chen L, Janetopoulos C, Huang YE et al. Twophases of actin polymerization display different depen-dencies on PI(3,4,5)P3 accumulation and have uniqueroles during chemotaxis. Mol Biol Cell 2003; 14:502837.92. Myers SA, Han JW, Lee Y et al. A Dictyostelium

homologue of WASP is required for polarized F-actinassembly during chemotaxis. Mol Biol Cell 2005; 16:2191206.93. Oikawa T, Yamaguchi H, Itoh T et al.PtdIns(3,4,5)P3 binding is necessary for WAVE2-in-ducedformationof lamellipodia.Nat CellBiol 2004;6:4206.94. Worthylake RA, Lemoine S, Watson JM et al.RhoA is required for monocyte tail retraction duringtransendothelial migration. J Cell Biol 2001; 154:14760.95. Niggli V. Rho-kinase in human neutrophils: a rolein signalling for myosin light chain phosphorylationand cell migration. FEBS Lett 1999; 445: 6972.96. Johnson EN, Seasholtz TM, Waheed AA et al.RGS16 inhibits signalling through the G alpha 13-Rho

axis. Nat Cell Biol 2003; 5: 1095103.97. Xu J, Wang F, Van Keymeulen A et al. Neutrophilmicrotubules suppresspolarity and enhance directionalmigration. Proc Natl Acad Sci USA 2005; 102:68849.98. Niggli V. Microtubule-disruption-induced and che-motactic-peptide-induced migration of human neutro-phils: implications for differential sets of signallingpathways. J Cell Sci 2003; 116: 81322.99. Li Z, Dong X, Wang Z et al. Regulation of PTENby Rho small GTPases. Nat Cell Biol 2005; 7:399404.100.Bosgraaf L, Russcher H, Smith JL et al. A novelcGMP signalling pathway mediating myosin phos-

phorylation and chemotaxis in Dictyostelium. Embo J2002; 21: 456070.101.Bosgraaf L, Van Haastert PJ. A model for cGMPsignal transduction in Dictyostelium in perspective of25 years of cGMP research. J Muscle Res Cell Motil2002; 23: 78191.102.Veltman DM, Roelofs J, Engel R et al. Activationof soluble guanylyl cyclase at the leading edge during

Dictyostelium chemotaxis. Mol Biol Cell 2005; 16:97683.103.Kriebel PW, BarrVA, Parent CA.Adenylylcyclaselocalization regulates streaming during chemotaxis.Cell 2003; 112: 54960.104.Bazzoni F, Cassatella MA, Rossi F et al. Phagocy-tosing neutrophilsproduceand release highamounts ofthe neutrophil-activating peptide 1/interleukin 8. J ExpMed 1991; 173: 7714.105.Baggiolini M, Moser B, Clark-Lewis I. Interleu-kin-8 and related chemotactic cytokines. The Giles Fil-ley Lecture. Chest 1994; 105(3 Suppl): 95S-98S.106.Kannan S. Amplification of extracellular nucleo-tide-induced leukocyte(s) degranulation by contingentautocrine and paracrine mode of leukotriene-mediatedchemokine receptor activation. Med Hypotheses 2002;59: 2615.

107.Cassatella MA. Neutrophil-derived proteins: sell-ing cytokines by the pound. Adv Immunol 1999; 73:369509.108.Matsukawa A, Yoshinaga M. Sequential gener-ation of cytokinesduring theinitiative phase of inflam-mation, with reference to neutrophils. Inflamm Res1998; 47 Suppl 3: S13744.109.Brechard S, Bueb JL, Tschirhart EJ. Interleukin-8primes oxidative burst in neutrophil-like HL-60through changes in cytosolic calcium. Cell Calcium2005; 37: 53140.110.Kuhns DB, Gallin JI. Increased cell-associatedIL-8 in human exudative and A23187-treated periph-eralblood neutrophils. J Immunol 1995;154: 655662.111.Marie C, Fitting C, Cheval C et al. Presence ofhigh levels of leukocyte-associated interleukin-8 upon

cell activation and in patients with sepsis syndrome.In-fect Immun 1997; 65: 86571.112.Soberman RJ, Christmas P. The organization andconsequences of eicosanoid signaling. J Clin Invest2003; 111: 110713.113.Ford-Hutchinson AW, Bray MA, Doig MV et al.Leukotriene B, a potent chemokinetic and aggregatingsubstance released from polymorphonuclear leuko-cytes. Nature 1980; 286: 2645.114.PalmbladJ, MalmstenCL, Udenet al. LeukotrieneB4 is a potent and stereospecific stimulator of neutro-phil chemotaxis and adherence. Blood 1981; 58:65861.

Documents

Chemo Taxis