micro-scale chromophore-assisted laser inactivation of nerve growth cone proteins

Micro-Scale Chromophore-Assisted Laser Inactivationof Nerve Growth Cone ProteinsANDREA BUCHSTALLER AND DANIEL G. JAY*Department of Physiology, Tufts University School of Medicine, Boston, Massachusetts 02111

KEY]tWORDS CALI; loss of function; growth cone motility; filopodia; lamellipodia

ABSTRACT Directed growth cone movement is crucial for the correct wiring of the nervoussystem. This movement is governed by the concerted actions of cell surface receptors, signalingproteins, cytoskeleton-associated molecules, and molecular motors. In order to investigate themolecular basis of growth cone motility, we applied a new technique to functionally inactivateproteins: micro-scale Chromophore-Assisted Laser Inactivation [Diamond et al. (1993) Neuron11:409–421]. Micro-CALI uses laser light of 620 nm, focused through microscope optics into a 10-µmspot. The laser energy is targeted via specific Malachite green-labeled, non-function-blockingantibodies, that generate short-lived protein-damaging hydroxyl radicals [Liao et al. (1994) ProcNatl Acad Sci USA 91:2659–2663]. Micro-CALI mediates specific loss of protein function withunachieved spatial and temporal resolution. Combined with time-lapse video microscopy, it offersthe possibility to induce and observe changes in growth cone dynamics on a real time base. Wepresent here the effects of the acute and localized inactivation of selected growth cone molecules ongrowth cone behavior and morphology. Based on our observations, we propose specific roles for theseproteins in growth cone motility and neurite outgrowth. Microsc. Res. Tech. 41:97–106,2000. r 2000 Wiley-Liss, Inc.

INTRODUCTIONAn important event during the formation of neurocir-

cuitry is the outgrowth of neurites along specific path-ways to their appropriate targets. At the tips of growingneurites, growth cones move in response to chemicalcues present in the embryonic environment (reviewedby Jay, 1996; Mueller, 1999). It is the sensory andmotile machinery within the growth cones that controlstheir explorative behavior, and it is of fundamentalinterest to understand the mechanisms of this machin-ery. Processes that govern neurite outgrowth includethe binding of receptor molecules to guidance cues inthe extracellular environment, the activation of signaltransduction pathways, and the reorganization of thelocal cytoskeleton inside the growth cone (Letourneauet al., 1994; Lin and Forscher, 1993; Tanaka and Sabry,1995). These processes are highly integrated and mustoccur at precise times and locations within the growthcone. The proteins that act in these processes form acomplex dynamic network. These properties have madeit difficult to understand how growth cones work.Recently a new technology called micro-scale Chromo-phore-Assisted Laser Inactivation (micro-CALI) hasprovided new insight into the molecular mechanismsthat control growth cone navigation.

The growth cone is a dynamic motile structure withhand-like morphology. At its periphery, filopodia formlong protrusions that act as sensory antennae. Lamelli-podia expand between the filopodia in the forwardmovement of the growth cone (Kater and Rehder, 1995).Filopodia and lamellipodia extend and retract, based onthe activity of the underlying F-actin cytoskeleton(Bray, 1989; Yamada et al., 1971). F-actin dynamicsmediate changes in growth cone shape and motility,

and dictate the subsequent direction of neurite out-growth (Letourneau, 1996). In the growth cone’s centraldomain, growing microtubules engorge regions of theperiphery and consolidate to extend the nascent neurite(Forscher and Smith, 1988; Goldberg and Burmeister1986; Mitchison and Kirschner, 1988).

The coordinated action of signaling molecules, cyto-skeleton-associated molecules, and molecular motorspromotes F actin and microtubule dynamics, and thegeneration of forces inside of the growth cone necessaryfor directed movement. Based on localization studiesand in vitro biochemistry, many candidate moleculeshave been implicated in growth cone motility andguidance. However, a direct demonstration of the impor-tance of a specific protein in these processes requirestheir functional inactivation in the moving growth cone(Letourneau, 1996).

Functional inactivation subtracts a specific proteinfrom its natural environment to address its potentialrole in a cellular process. A resulting disruption of thatprocess supports the hypothesis that the targeted pro-tein is required. Studies of neurite outgrowth andpathfinding have included the use of function-blockingantibodies (Rutishauser et al., 1978), pharmacologicalinhibitors (Williams et al., 1994), antisense oligonucleo-tides (McFarlane et al., 1996), dominant negative inhibi-tors (Aigner and Caroni, 1993), and genetic knockouts

Contract grant sponsor: National Institute of Health; Contract grant numbers:NS34699, EY11992, and CA81668; Contract grant sponsor: ‘‘OesterreichischeNationalfonds zur Foerderung Wissenschaftlicher Forschung’’; Contract grantnumber: J1673-MOB.

*Correspondence to: Daniel Jay, Department of Physiology, Tufts UniversitySchool of Medicine, Boston MA 02111. E-mail: [email protected]

Received 9 July 1999; accepted in revised form 16 September 1999

MICROSCOPY RESEARCH AND TECHNIQUE 48:97–106 (2000)

r 2000 WILEY-LISS, INC.

(reviewed by Holtmaat et al., 1998) to mediate func-tional inactivation. However, these approaches all gen-erate a chronic loss of protein function. In recent years,a complementary method called Chromophore-AssistedLaser Inactivation (CALI) has been added to this list.CALI causes the acute and localized inactivation ofspecific proteins in an organism (Jay, 1988; Jay andKeshishian, 1990) and micro-CALI, an extension of thisapproach, inactivates proteins within a single cell orspecific subcellular domains (Diamond et al., 1993).Together with time-lapse video microscopy, micro-CALIhas been used to analyze protein function on real timein single moving growth cones (Castelo and Jay, 1999;Chang et al., 1995; Liu et al., 1999; Sydor et al., 1996;Wang et al., 1996).

The aim of this paper is to provide an overview ofmicro-CALI and to show its use in inactivating specificproteins in the neuronal growth cone. First a compari-son between micro-CALI and other perturbation meth-ods will be made. Then a description of micro-CALI andhow it is performed will be provided. Finally, a sectionon how micro-CALI has contributed to our understand-ing of growth cone motility and neurite outgrowthconcludes the paper.

COMPARISON OF CALI WITH OTHERFUNCTIONAL INACTIVCATION STRATEGIESAn array of functional inactivation strategies has

been used to understand the molecular mechanisms ofgrowth cone motility and axon guidance. Table 1 sum-marizes these approaches and cites selected referencesfrom the literature that describe their use to addressneurite outgrowth and pathfinding. Distinguishing fea-tures in comparing these various methods are theirselectivity, as well as the spatial specificity and tempo-ral discrimination provided.

Function-blocking antibodies are valuable tools totest for the participation of cell surface molecules in

growth cone guidance (Rutishauser et al., 1978). Theyhave been used to investigate neurite outgrowth andaxon pathfinding in vitro and in vivo, e.g., in the retina(Brittis et al., 1995; Pollerberg et al., 1993), in the limb(Honig et al., 1996, 1998a,b), and at the embryonicmidline (Stoeckli et al., 1995). In experiments usingfunction-blocking antibodies to study neurite out-growth, the behavior of entire neuronal populations isgenerally assessed, but single growth cones have alsobeen observed by time-lapse microscopy (Brittis et al.,1995; Stoeckli et al., 1997). For the study of intracellu-lar proteins, various methods are available to loadantibodies into neurons, such as microinjection, lipofec-tion, and electroporation. Generally, function-blockingantibodies work at very high concentrations, that aresometimes difficult to bring into cells and growth cones.The effect exerted depends on their affinity and chemi-cal specificity, and discrimination between closely re-lated proteins may be difficult.

Pharmacological inhibitors have been applied success-fully to investigate receptor-mediated signaling path-ways involved in axon outgrowth on defined substrata(Doherty and Walsh, 1996; Williams et al., 1994) andneurotransmitter-induced growth cone turning (Songet al., 1997). Pharmacological inactivation has alsobeen used to study axon outgrowth and pathfinding inthe Xenopus retinotectal system in vivo (Worley andHolt, 1996). Pharmacological approaches are often ham-pered by the fact that most of these inhibitors have poorspecificity as well as poor temporal and spatial control.They sometimes inactivate a broad range of structur-ally related molecules, and are not suited to study rapidgrowth cone dynamics. Recently, higher specificity hasbeen achieved by using synthetic peptides that targetthe interactions of the protein of interest with itssubstrates or cofactors (Mochly-Rosen and Kauvar,1998).

TABLE 1. Methods to functionally inactivate proteins involved in growth cone motility

Method Mode of actionTemporal, spatial and chemical

specificity of inhibition Selected references

Micro-CALI Laser induced free radical damage to a pro-tein bound with a dye-labeled antibodywithin a single cell

Acute and local, high specificity,affects single cells or subcellularregions direct inactivation

Takei et al., 1999Chang et al., 1995Takei et al., 1998aSydor et al., 1996Castelo and Jay, 1999Wang et al., 1996Liu et al., 1999

Function-blockingantibodies

Induction of conformational changes throughdirect antibody-antigen binding

Chronic and global, high specificitydirect inactivation

Rutishauser et al., 1978Stoeckli et al., 1997Honig et al., 1998

Pharmacologicalinhibitors

Antagonize an enzymatic activity by directlybinding the protein’s catalytic or regula-tory sites

Chronic, if bath applied acute andlocal if pipette-applied low speci-ficity direct inactivation

Williams et al., 1994Worley and Holt, 1996Song et al., 1997

Dominant negativeinhibitors

Abolition of wild-type protein function byectopic expression of catalytically inactivemutants

Chronic and global, high specificitydirect or indirect inactivation

McFarlane et al., 1996; 1998Kuhn et al., 1998; 1999Luo et al., 1995

Antisense oligo-nucleotides

Inhibition of gene expression by formationand subsequent degradation of double-stranded RNA-DNA complexes

Chronic, global, high specificity, indi-rect inactivation

Aigner and Caroni, 1993; 1995Wylie et al., 1998

Genetic knockout Gene disruption based on homologous recom-bination of input DNA with genomic DNAsequences

Chronic, global (whole organism),high specificity, indirect inactiva-tion

Strittmatter et al., 1995Kruger et al., 1998Dahme et al., 1997Cohen et al., 1998Henkemeyer et al., 1996Harada et al., 1994

98 A. BUCHSTALLER AND D.G. JAY

The expression of dominant-negative forms of signal-ing molecules has been used to study the role of growthfactor signaling during nervous system development. Adominant-negative form of the FGF-receptor causederrors in Xenopus retinal ganglion cell axonal targetrecognition (McFarlane et al., 1996). Expression ofdominant negative mutants of Rac and other smallGTPases desensitized growth cones to CNS myelin orcollapsin-1 in vitro (Kuhn et al., 1999) and causedguidance errors when expressed in Drosophila neuronsin vivo (Luo et al., 1994). In conjunction with activators,dominant-negative inhibitors may serve to study simul-taneously the effects of loss and gain of function of thesame protein (Kuhn et al., 1998). Ectopic expression ofcatalytically inactive mutants depends on the possibil-ity of constructing such mutants, and on the availabil-ity of systems for which expression of recombinantproteins is efficient.

Antisense oligonucleotides have been primarily usedin studies concerning molecular mechanisms of growthcone motility in vitro. Antisense oligodeoxyribonucleo-tides, complementary to myosin IIB, attenuated filopo-dial extension and neurite outgrowth in (Wylie et al.,1998). Depletion of GAP-43 in primary sensory neuronsaltered the growth cones’ response to external stimulias well as its motile behavior (Aigner and Caroni,1995). The success of antisense inhibition of proteinexpression is dependent on the hybridization efficiencyof the nucleotide to the target RNA and on the stabilityof the oligonucleotide. An antisense molecule is typi-cally taken to be ‘‘specific’’ if there is no gross loss of cellviability, and the levels of the target RNA and itsassociated protein fall much more than those of thecontrol RNAs. These reasons have occasionally limitedthe success of antisense approaches.

Genetic knockout technology is an excellent approachto assess the function of a protein in the context of anentire organism where the neuroanatomy and behaviorcan be studied. Neurons from genetic knockout micehave also been cultured in vitro in order to test growthcone behavior under defined conditions. For example, ithas been shown that the retinal ganglion cell axons ofGAP-43 knockout mice remain trapped in the opticchiasm unable to navigate past this midline decisionpoint, while their in vitro growth rates remain un-changed (Kruger et al., 1998; Strittmatter et al., 1995).Genetic knockouts provide high specificity and, whenconditional gene knockouts are used, can generateanimals with regionally specific loss of function atdifferent developmental stages (Feil et al., 1996). How-ever, it still remains difficult to use genetic knockouts tostudy growth cone proteins that have important func-tions early in development. Their loss of functioncauses embryonic lethality so that their specific role ingrowth cones cannot be assessed. Additionally, compen-sation for the loss of the gene product (Harada et al.,1994) or changes in the regulation of other genes mayyield an apparently unaltered or misleading phenotype.

CALI provides a complementary inactivation strat-egy with some distinct advantages over these otherapproaches. CALI uses non function-blocking antibod-ies that bind specifically to the targeted protein (Jay,1988). These antibodies are multiple-conjugated withmalachite green (MG), a chromophore that absorbs 620nm light, a wavelength not absorbed by most cellular

components. Upon irradiation with 620 nm pulsed,high-powered laser light, MG generates short-lived freehydroxyl radicals that inactivate the bound antigen,whereas other proteins, even close neighbors in aprotein complex, remain largely unaffected. In fact,experiments with free radical quenchers demonstratedthat the hydroxyl radicals reach a half-maximal inacti-vation distance of 15 A from the dye moiety (Liao et al.,1994). Investigations conducted on the multi-subunitT-cell receptor complex (TCR) showed that targeteddisruption of one subunit does not affect the functionalintegrity of the other subunits present (Liao et al.,1995). It has also been shown that CALI of myosin Vusing an antibody directed against the tail domainselectively blocked the motor domain activity, withoutaffecting its ATPase activity, situated about 100 A awayfrom the motor domain (Wang et al., 1996).

CALI converts non function-blocking antibodies intostrongly inactivating reagents. Because inactivation isdetermined by the time and location of laser irradia-tion, CALI is suitable to study proteins required later indevelopment. The loss of function beeing acute andlocal, the opportunity for functional compensation re-mains low.

CALI has been applied in many different biologicalsystems. Specific inactivation was achieved in 90% ofthe 40 cases examined, as assessed by in vitro activitytests or by analyzing in vivo phenotypes (reviewed inWang and Jay, 1996; Jay, 1999). CALI of proteinsencoded by segment-polarity genes even skipped andpatched in Drosophila embryos generated phenotypesequivalent to genetic loss of function mutants(Schmucker et al., 1994; Schroeder et al., 1996). Itprovides a means of functional inactivation in systemsthat lack techniques for targeted gene disruption suchas grasshoppers (Jay and Keshishian, 1990), Tribolium(Schroder et al., 1999) and chicken (Chang et al., 1995).

While CALI inactivates target proteins inside a 2-mmlaser spot, micro-CALI focuses a laser beam throughthe optics of an inverted microscope to a 10-µm spot.Micro-CALI was first used to study the function of theneural cell adhesion molecules fasciclin I and fasciclinII in vivo (Diamond et al., 1993). By laser-irradiatingthe growth cone and the cell body of grasshopper Ti1neurons separately, it could be demonstrated, thatfasciclin I mediates axon fasciculation, whereas fascic-lin II is involved in axonogenesis. Since then, micro-CALI has been applied repeatedly to investigate therole of cell surface receptors, signaling proteins, cytoskel-eton-associated molecules, and molecular motors ingrowth cone dynamics and neurite outgrowth (Casteloand Jay, 1999; Chang et al., 1995; Sydor et al., 1996;Wang et al. 1996). With its unique spatial resolution,micro-CALI provides us with a means to induce asym-metric distributions of target protein activities inside ofa single growth cone (Chang et al., 1995; Liu et al.,1999; Mack et al., 1999; Wang et al., unpublished data).

It is important to keep the limitations in mind, whenusing micro-CALI to study in situ protein functions.Like several other inactivation techniques, micro-CALIcan provide a correlation, but no direct link between theinduced loss of function and growth cone behavior. A setof rigorous controls is necessary to confirm that thecellular effects observed are due to the specific laser-mediated inactivation of the target bound by MG-

99MICRO-CALI OF GROWTH CONE PROTEINS

labeled antibody. If possible, CALI in vitro data shouldbe used to sustain the observed effect of micro-CALI insitu. In this case, it is important to consider that thetwo techniques may address different functions on thesame protein, depending on whether the inactivatedprotein is present in its cellular environment or not.Within the cellular environment, the effects of CALImay also be more complex than a simple loss of function(e.g., generation of constitutively active proteins; dam-aging neighboring proteins). However in the best stud-ied cases, these effects have not been observed thus far(Liao et al., 1995; Schmucker et al., 1994; Schroder etal., 1999). When interpreting the data, one should beaware that, the effects of micro-CALI are highly local-ized, and recovery may occur by diffusion of activeproteins from unirradiated regions (within minutes) orby de novo synthesis (with hours or days).

MICRO-CALI METHODOLOGYAntibody Selection, Labeling, and Loading

For micro-CALI to be effective, a specific antibodywith high affinity to the target protein is selected. Itshould not block function or do so only at high concentra-tions. Monoclonal or polyclonal antibodies that fit thesecriteria may be used. Antibodies need not to be com-pletely purified, if Western blot analysis and immunocy-tochemistry can exclude the binding of serum compo-nents to the proteins present in neurons. When lowconcentrations of purified antibodies (5 2 mg/ml) areused for MG-labeling, an equal weight of bovine serumalbumin is added to the reaction mixture in order tostabilize the antibody and to reduce its aggregationduring labeling. In numerous studies, laser irradiationof MG-labeled BSA has not affected the cellular pro-cesses studied (Chang et al., 1995; Sydor et al., 1996).

MG is bound covalently to the antibody by methodsdescribed in Jay (1988) and Beermann and Jay (1994).MG isothiocyanate (MGITC, Molecular Probes Inc.,Eugene, OR) is prepared by dissolving the dry reagentin dimethylsulfoxide at a concentration of 10 mg/ml.The dye is then added to a solution of 400 µg of totalprotein in 500 µl of 0.5M NaHCO3, pH 9.5, in four 2-µlaliquots added once every 5 minutes with continuousrocking. The mixture is then incubated for an addi-tional 15 minutes. The mixture is centrifuged at maxi-mum speed for 30 seconds on a microfuge (to sedimentparticulate material). The MG-labeled reagent is sepa-rated from free label and labeling buffer by gel filtrationwith a prepacked Sepharose G25 M column (PD10column, Pharmacia), and eluted with Hanks balancedsalt solution (HBSS, Gibco BRL) with calcium andmagnesium. If necessary, the eluted MG-labeled anti-body solution can be concentrated with a micro-concentrator. The final concentration of the proteinsolution depends on the subsequent application andusually lies between 0.2 and 2.0 mg/ml. The labelingratio is obtained by measuring the optical density at620 nm (molar absorptivity of MGITC5 150,000 M-1

cm-1) to determine the concentration of the dye anddividing it by the starting concentration of the proteinsolution. A labeling ratio of 6–8 dyes per protein mol-ecule is optimal (Linden et al., 1992).

MGITC reacts with the amino groups of the antibodyto form a stable thioester in a high pH environment.The reaction mixture should not contain other free

amino groups. As a concurrent reaction, hydrolysis ofthe isothiocyanate occurs, which depletes the reagentand causes a purple insoluble precipitate to form. Mostof the precipitate is removed by centrifugation and gelfiltration but some remains hydrophobically associatedwith the surface of the protein, and will dissociate witha half time of many weeks at 280°C and days at 4°C.This free dye can be toxic to cells and samples with alarge quantity of dye aggregate should not be used forCALI. The protein solution is aliquoted quickly afterlabeling, quick frozen, stored at 280°C and used for lessthan 6 months.

MG-labeled antibodies directed against intracellularproteins have been loaded into chick dorsal root gan-glion (DRG) neurons by microinjection (Wang et al.,1996), lipofection (Mack et al., 1999), or by a modifiedtrituration method (Sydor et al., 1996). Triturationbreaks up DRG tissue by multiple passes through amicropipette tip and is thought to create temporaryholes in the cell membrane that enable antibodies toenter the cell. To visualize loading, fluorescein-conju-gated non-immune IgG (1 mg/ml) is added to theMG-labeled antibody solution. After trituration, cellsare centrifuged, gently resuspended in culture medium,and plated on laminin-coated glass coverslips. They areincubated until they develop growth cones and neurites(,2 hours). MG-labeled antibodies are retained ingrowth cones with a half-life of ,12 hours (unpublisheddata). Figure 1 shows successful loading and retentionof anti-talin antibodies in DRG growth cones at thetime-point of laser irradiation. In contrast, nonimmuneIgG is loaded efficiently but extraction removes theantibody from fixed neurons.

Micro-Cali, Controls, and Time-LapseVideo Microscopy

For experiments utilizing micro-CALI, neurons arefirst observed by epifluorescence to verify antibody-loading. Samples are then irradiated by directing a620-nm pulsed laser beam through the fluorescenceoptics of an inverted microscope (Fig. 2). The pulsedlaser beam is generated by a nitrogen-driven dye laser(model 337, Laser Sciences Inc., Newton MA) using thefluorescent dye DCM. The parameters used are 30-µJpulses with a pulse width of 3.5 nsec at a frequency of20 Hz for 5 minutes (6,000 pulses).

The determination of the in situ role of a candidateprotein depends on comparing the effects of micro-CALIof a target protein with a matched set of controls. Thesecontrols may include neurons loaded with MG-labelednon-immune antibody (of the same Ig class as thespecific antibody) or MG-labeled BSA, and subjected tolaser irradiation. Also is it crucial to analyze thebehavior of unloaded growth cones under laser irradia-tion, in order to rule out any unspecific damage causedby the laser light. Furthermore, to determine whetherthe antibody used is non-function blocking, unlabeledspecific antibody should be introduced into the cells,and its effect should be compared to the effect generatedby micro-CALI. Finally, a comparison of behavior of theirradiated and unirradiated side of growth cones loadedwith the MG-labeled specific antibody is particularlyinformative. Growth cone behavior is quite variableand the direct comparison with the unirradiated half ofthe same growth cone is an ideal internal control for


micro-CALI experiments. Experiments of this typeallow the investigator to discern subtle changes ingrowth cone behavior after the localized loss of acandidate protein. On occasion, we have had the oppor-tunity to demonstrate the specificity of the proteininactivation by testing the roles of protein isoforms ingrowth cones. For example, could we demonstrate thatinactivation of myosin V in DRG growth cones causedchanges in filopodial extension rates, whereas micro-CALI of the related protein myosin I-b induced lamelli-podial expansion (Wang et al., 1996)?

For all these experimental and control treatments,growth cone behavior is recorded before, during, andafter irradiation by time-lapse video microscopy withScion Image or custom written software. Images arestored either directly on the computer hard drive orusing an optical memory disc recorder. Image analysisand morphometry is performed with the NIH image orother morphometry software that allow measurementsto be made and stored in a spread sheet. Growth coneparameters that have been measured include bending/buckling of filopodia, filopodial number, the rates offilopodial motility, the rates of lamellipodial expansionor retraction, the growth cone surface area, the rates of

neurite outgrowth, and the angle of deviation from theoriginal trajectory. Quantitative data are analyzed us-ing Cricket Graph software (Malvern, PA) and MinitabStatistics Software (Pennsylvania State University).Generally micro-CALI and control data sets are com-pared using Student’s t-test or ANOVA, but whencomparing the irradiated and unirradiated regions ofthe same growth cone, the paired t-test is used. Whenthe percentages of neurons affected for experimentaland control data sets are compared, appropriate testssuch as the binomial test or Poisson test are employed.

MICRO-CALI OF PROTEINS INVOLVED INGROWTH CONE MOTILITY

We and others have used micro-CALI to study thefunction of growth cone receptors, signaling molecules,actin-associated proteins, molecular motors, and micro-tubule-associated proteins that are thought to functionin growth cone motility. Table 2 summarizes the effectsof CALI on some of these proteins. The proteins wereinactivated in subregions of chick dorsal root ganglianeuronal growth cones extending over a uniform lami-nin substratum. Subsequent behavior was observed byvideo-enhanced microscopy and analyzed by quantita-tive morphometry. Micro-CALI of these proteins eachcaused specific changes in growth cone behavior orneurite extension. Applied at the leading edge, micro-CALI of these different proteins has resulted in lamelli-podial expansion or retraction, cessation of filopodialmotility, filopodial retraction, bending, and buckling.Micro-CALI has caused growth cones to turn, split, orcollapse and neurites to retract or extend depending onthe protein inactivated. Thus, different behaviors canbe attributed to the acute loss of specific proteins in thegrowth cone and allows one to propose functions ofthese proteins in growth cone motility. These studiesattest to the utility of micro-CALI in addressing growthcone mechanisms. The following sections illustratesome recent examples.

Membrane ReceptorsL1 and NCAM-180 are two well-characterized neural

cell adhesion molecules of the immunoglobulin super-family (Moos et al., 1988; Murray et al., 1986; Williamsand Barclay, 1988). They are neurite promoting sub-strates (Doherty et al., 1990; Lemmon et al., 1989),thought to act in axon growth and guidance in themajor fiber tracts of the nervous system (Cremer et al,1994; Honig et al., 1997; Kamiguchi et al., 1998;

Fig. 1. Antibody loading into growth cones.Chicken DRG neurons were trituration loadedwith MG-labeled non-immune IgG or anti-talin antibody. The cells were either fixedbefore (loaded) or after (extracted) detergentextraction and then probed with FITC-labeledsecondary antibody. The anti-talin antibodies,in contrast to the non-immune IgG, wereretained after extraction, showing specificbinding of the anti-talin antibody towards itsantigen expressed in neuronal growth cones.The lower panel shows rhodamine-phalloidinstaining of the actin cytoskeleton.

Fig. 2. Micro-CALI setup. The beam of a nitrogen-driven dye laseris directed through the fluorescence optics of an inverted microscope.For a correct alignment of the sample, a glass-bottomed Petri dish,colored with blue ink, is first brought into the light path. Activatingthe laser bleaches a small spot, the size of the beam, which can bemarked on the monitor screen. The sample is then aligned with themark on the screen, and treated according to the parameters describedin the text.


Landmesser et al., 1988). They also seem involved inneuronal plasticity and memory formation (reviewed byHoffman, 1998). Interactions with the actin cytoskel-eton and second messenger molecules assessed underdiverse experimental conditions have been extensivelydescribed (reviewed by Burden-Gulley et al., 1997;Doherty and Walsh, 1996), but the precise roles of L1and NCAM-180 in the growth cone have not yet beendetermined.

Takei et al. (1999) used micro-CALI to study thefunction of L1 and NCAM-180 in chick DRG growthcones by inactivating their intracellular domain, thoughtto mediate interactions with the cytoskeleton (Davisand Bennett, 1994; Kramer et al., 1997). Micro-CALI ofL1 caused a slow neurite retraction after 10 minutesbut did not affect growth cone motility. In contrast,micro-CALI of NCAM-180 caused localized filopodialand lamellipodial retraction but did not affect neuriteextension. These findings suggested that L1 and NCAM-180 act in distinct steps of neurite outgrowth; L1 mayact in neurite extension while NCAM-180 may act inprotrusion of the growth cone leading edge.

Signal Transduction MoleculesThe activation of surface receptors in response to

extracellular cues causes a diverse array of signaltransduction events that are critical for axon guidance(reviewed in Bixby and Bookman, 1996). Micro-CALIhas been applied to signal transduction molecules toaddress their roles in axon growth and guidance.

Calcineurin is a calcium/calmodulin-dependent ser-ine threonine phosphatase that is found in growthcones and may play a key role in growth cone motility.Calcineurin has in vitro substrates that are potentialeffectors of growth cone motility including GAP-43 andthe microtubule associated protein tau. Chang et al.(1995) used micro-CALI to show that calcineurin isrequired for filopodial and lamellipodial motility. Theyfirst demonstrated that CALI of purified calcineurininactivated its phosphatase activity and, then, thatinactivation of calcineurin inside DRG neurons de-creased the phosphorylation state of tau within theseneurons. When micro-CALI of calcineurin was per-formed on chick DRG growth cones, the filopodia and

lamellipodia within the irradiated area retracted. Thisdid not occur outside of the laser spot or for a variety ofcontrol treatments, regardless of laser light. The local-ized retraction caused by CALI of calcineurin resultedin a net growth cone turning away from the laser spot.This suggested that growth cone turning is caused by alocal retraction at the leading edge that is, in turn,correlated with an asymmetry of calcineurin function inthe growth cone.

Many studies have implicated protein tyrosine ki-nases in neurite outgrowth (Atashi et al., 1992; Gold-berg and Wu, 1996; Worley and Holt, 1996) but showingwhich of the many kinases found in growth cones isrequired has been difficult. pp60-c-src is a well-characterized protein tyrosine kinase that is a majorcomponent of the developing brain and nervous system(Cotton and Brugge, 1983). pp60-c-src has been impli-cated in many signal transduction pathways and manyin vitro kinase substrates have been identified. How-ever, the endogenous targets of pp60-c-src are notknown and its in situ function in cells has also not beenestablished. A genetic knockout of pp60-c-src had verylittle effect on brain development (Soriano et al., 1991),but others have shown that these mice have a twofoldincrease in the activity of homologous tyrosine kinasesin the brain (Grant et al., 1995).

We have investigated the role of pp60-c-src in develop-ing chick dorsal root ganglion (DRG) neurons (Hoffman-Kim et al., unpublished data). CALI directed againstpurified pp60-c-src inactivated .85% of its tyrosinekinase activity in vitro and when CALI of pp60-c-srcwas performed on DRG neurons in culture, tyrosinephosphorylation was specifically reduced by 60% withinthese cells. We analyzed growth cone motility aftermicro-CALI of pp60-c-src. We observed a significantenhancement on the rate of neurite extension and didnot see other morphological effects. Micro-CALI of thehomologous tyrosine kinase pp59-fyn or a variety ofcontrol treatments did not affect neurite extension. Asneurite extension is based on microtubule dynamics(Tanaka and Kirschner,1995) and previous studies haveshown that tyrosine phosphorylation of tubulin de-creases microtubule formation in vitro (Matten et al.1990), we suggest that an important role of pp60-c-src

TABLE 2. Summary of results obtained with micro-CALI of growth cone proteins*

Growth conemolecule

Growth conespeed

Lamellipodialbehavior

Filopodialbehavior Overall growth cone and neurite behavior References

L1 Decreased n.d. Unchanged Neurite retraction, growth cone morphologyunchanged

Takei et al., 1999

NCAM Unchanged Retraction Retraction Growth cone turning away from laser spot Takei et al., 1999Calcineurin Unchanged Retraction Retraction Growth cone turning away from laser spot Chang et al., 1995src Increased Expansion Unchanged Enhanced growth cone advancement, neurite

extensionHoffman-Kim et al.,

unpublished datafyn Unchanged n.d. Unchanged Unchanged Hoffman-Kim et al.,

unpublished dataIP3R Decreased n.d. Unchanged Growth arrest and neurite retraction Takei et al., 1998Talin Unchanged n.d. Stalling Unchanged Sydor et al., 1996Vinculin Unchanged n.d. Increased bending

and bucklingUnchanged Sydor et al., 1996

Radixin Unchanged Retraction Unchanged Growth cone splitting Castelo and Jay, 1999Myosin Ib Unchanged Expansion Unchanged Growth cone turning towards laser spot Wang et al., unpublished dataMyosin V n.d. Unchanged Retraction Unchanged Wang et al., 1996tau Decreased Retraction Unchanged Growth cone turning away from laser spot Liu et al., 1999MAP 1B n.d. Retraction n.d. Growth cone turning away from laser spot Mack et al., 1999

*Generally, parameters analyzed were growth cone speed, lamellipodial and filopodial behavior and overall growth cone behavior. Growth cone turning occurredmainly, when the laser spot was applied asymmetrically onto the growth cone. n.d. not determined.


in developing DRG neurons is a control of microtubuledynamics via its tyrosine kinase activity.

Actin-Associated ProteinsThe regulation of growth cone motility occurs by the

action of actin-associated proteins that control thedynamics of polymerization and the contacts that theactin cytoskeleton makes with the membrane and othercellular structures. Talin and vinculin are structuralproteins found in focal adhesion complexes in fibro-blasts but also associated with focal contacts in growthcones (Letourneau and Shattuck, 1989). Talin andvinculin bind to each other and associate with the actincytoskeleton and integrins (reviewed in Burridge andFath, 1989).

We investigated talin and vinculin function by inacti-vating them in subregions of chick dorsal root ganglianeuronal growth cones and by observing subsequentbehavior by video-enhanced microscopy and quantita-tive morphometry (Sydor et al., 1996). Micro-CALI oftalin resulted in the temporary cessation of filopodialextension and retraction. Inactivation of vinculin causedan increased incidence of filopodial bending and buck-ling within the laser spot but had no effect on extensionor retraction. These findings showed that talin acts infilopodial motility and may couple both extension andretraction to actin dynamics. The results also suggestedthat vinculin is not required for filopodial extension andretraction but plays a role in the structural integrity offilopodia. The distinct filopodial behaviors in responseto micro-CALI of talin and vinculin (which bind to eachother) were a good demonstration of the specificity ofmicro-CALI for different growth cone proteins.

During growth cone movement, F-actin at the leadingedge is also connected by actin crosslinking proteins.One group of proteins that may act as crosslinkers arethe ERM (ezrin, radixin, moesin) family (Tsukita andYonemura, 1997). It has been suggested that radixinmaintains the stability of lamellipodia by performingthis role at the leading edge of moving cells (Algrain etal., 1993: Tsukita et al., 1994). Antisense oligonucleo-tides directed to all three ERM family members perturbcell adhesion and microvilli formation in thymoma cellsbut it was necessary to add them together for a signifi-cant effect (Tsukita et al., 1994). These data indicatethat ezrin, radixin, and moesin can functionally substi-tute for each other.

When micro-CALI of radixin was performed, growthcones split away from the laser spot and form two smallgrowth cones (Fig. 3). The growth cone splitting fre-quency after micro-CALI of radixin, with two differentmonoclonal antibodies, was 80–90%, while the fre-quency for a variety of controls, including MG-labeledIgG and unlabeled anti-radixin antibodies, was indistin-guishable from that for untreated neurons. These find-ings support the hypothesis that radixin is required tolamellipodial stability at the leading edge. When ra-dixin is focally inactivated, this may cause a localweakening of the cytoskeleton and subsequent move-ment splits the growth cone.

Molecular MotorsMyosins are required for the force generation and

movement of material during growth cone motility.Critical roles for the myosins have been shown for

growth cones (Lin et al., 1996) but as there are severaldifferent isoforms found in neurons (Bridgman andDailey, 1989), it has been difficult to attribute specificfunctions to each one of these. We have investigated therole of two unconventional myosins: myosin V and Ibwith micro-CALI time-lapse video microscopy (Wang et

Fig. 3. Micro-CALI of radixin causes growth cone splitting. ChickenDRG neurons were loaded with MG-labeled anti-radixin antibody andFITC-IgG, then plated on a laminin-coated dish. After neurites hadstarted to grow, antibody-loading was verified by epifluorescence. Thecenters of emerging and antibody-loaded growth cones were then laserirradiated for 5 min. Growth cone splitting always occurred preciselyduring laser treatment, suggesting that splitting is a direct conse-quence of micro-CALI of radixin. Bar 5 10 µm.


al., 1996). Myosin V is concentrated in organelle-richregions of the growth cone in rodent superior cervicalganglion cells (Evans et al., 1997) and has been associ-ated with fast transport of ER vesicles on actin fila-ments purified from giant squid axons (Tabb et al.,1998) and movement of melanosomes (Wu et al., 1997).Myosin Ib is found at the leading edge of growth conesand could serve to bind cortical F-actin to the mem-brane (Wagner et al., 1992).

During micro-CALI of myosin V, when filopodia ex-tend, they do so at 1/3 their normal rate; when theyretract, they do so normally. These findings showedthat myosin V is specifically involved in filopodialextension but not retraction (Wang et al., 1996). Thestochastic nature of filopodial collapse observed reflectsa potential role in myosin V supplying material tofilopodia during extension but we cannot rule out a rolein force generation. Micro-CALI of myosin Ib had noeffect on filopodial motility but instead induced lamelli-podial expansion (Wang et al., 1996). Repeated asym-metric inactivation of myosin Ib induced growth coneturning towards the irradiated side (Fig. 4). Theseexperiments demonstrated distinct roles for two myo-sin isoforms that act on different subcellular F-actin-based structures at the leading edge of growth cones.

Microtubule-Associated ProteinsMicrotubule engorgement into the growth cone pe-

riphery and bundling of these microtubules to form thenascent neurite are critical processes for axon exten-sion and must involve the action of many microtubule-associated proteins (MAPs) (Mitchison and Kirschner,1988). Tau and MAP 1B are MAPs that can bundlemicrotubules in vitro (Cleveland et al., 1977). Both arefound in axons and growth cones, and seem to beimportant for neurite outgrowth. The genetic knockoutof tau (Harada et al., 1994) showed very slight defectsin neurocircuitry and targeted disruption of the firstexon of MAP1B (Takei et al., 1997) only resulted indelays in nervous system development. However disrup-tion of MAP1B exon 5 generated more severe effects

(Edelmann et al., 1996), including embryonic lethalityin the homozygote mice and impaired outgrowth as wellas immunohistochemical changes in different regions ofthe brain of heterozygous animals. The roles of tau (Liuet al.,1999) and a phosphoisoform of MAP 1B (Mack etal., 1999) have been recently studied using micro-CALIin DRG neurons and retinal ganglion cells, respectively.Micro-CALI of tau showed a reduction of neurite exten-sion rates and also a significant lamellipodial retractionin the irradiated region. Micro-CALI of MAP 1B causeda marked lamellipodial retraction followed by growthcone turning away from the irradiated spot. Thesefindings confirm requirements for tau in neurite exten-sion and growth cone morphology. The results alsosuggest a novel function for both tau and MAP 1B inlamellipodial structure that may be related to the factthat these MAPs can also bind to F-actin (Sattilaro etal., 1981; Selden and Pollard, 1983). Interestingly, thedouble knockout of tau and MAP 1B has been generatedin mice and showed severe neurological defects anddiminished neurite elongation of hippocampal neuronsin culture suggesting that these proteins compensatedfor each other in the single knockout experiments(Takei et al., 1998b).

CONCLUSIONS AND FUTURE DIRECTIONSThe number of proteins involved in growth cone

motility and their complicated pattern of spatial andtemporal relations makes it difficult to investigatecorrelation between molecular events and subsequentgrowth cone behavior. It is of great interest to establishpathways of interaction of these proteins that controlgrowth cone movement. Because micro-CALI offers aunique spatial and temporal resolution, it can be usedto analyze subtle changes in growth cone parametersdue to focal and specific inactivation of growth coneproteins. The combination of micro-CALI with high-resolution time-lapse video microscopy allows exactmeasurements of growth cone characteristics (e.g.,filopodial and lamellipodial length) before and afterlaser treatment and a detailed subsequent mathemati-cal analysis.

In the future, advances in image acquisition, in situlabeling of specific proteins (by fusion with greenfluorescent protein), and improved data processingsystems may be combined with micro-CALI. Use ofmicro-CALI in conjunction with these techniques shouldallow for the detailed study of the generation of motilitywithin the growth cone.

ACKNOWLEDGMENTSWe thank Elisabeth Pollerberg for sharing unpub-

lished data and Canwen Liu and Tom Diefenbach forvaluable comments on the manuscript. The work pre-sented has been supported by the National Instituteof Health Grants NS34699, EY11992, and CA81668,and the ‘‘Oesterreichische Nationalfonds zur Foer-derung Wissenschaftlicher Forschung’’ grant J1673-MOB (to A. B.).

REFERENCESAigner L, Caroni P. 1993. Depletion of 43-kD growth-associated

protein in primary sensory neurons leads to diminished formationand spreading of growth cones. J Cell Biol 123:417–429.

Fig. 4. Micro-CALI of myosin Ib causes growth cone turning.Chicken DRG neurons were microinjected with MG-labeled anti-myosin Ib monoclonal antibody and their growth cones were laserirradiated on one side, three times over a period of 40 minutes. Thecamera lucida drawing shows that inactivation of myosin Ib results ingrowth cone turning towards the laser spot. Bar 5 10 µm.


Aigner L, Caroni P. 1995. Absence of persistent spreading, branching,adhesion in GAP–43- depleted growth cones. J Cell Biol 128:647–660.

Aigner L, Arber S, Kapfhammer JP, Laux T, Schneider C, Botteri F,Brenner HR, Caroni P. 1995. Overexpression of the neural growth-associated protein GAP–43 induces nerve sprouting in the adultnervous system of transgenic mice. Cell 83:269–278.

Algrain M, Turunen O, Vaheri A, Louvard D, Arpin M. 1993. Ezrincontains cytoskeleton and membrane binding domains accountingfor its proposed role as a membrane-cytoskeletal linker. J Cell Biol120:129–139.

Atashi JR, Klinz SG, Ingraham CA, Matten WT, Schachner M, ManessPF. 1992. Neural cell adhesion molecules modulate tyrosine phos-phorylation of tubulin in nerve growth cone membranes. Neuron8:831–842.

Beermann AE, Jay DG. 1994. Chromophore-assisted laser inactiva-tion of cellular proteins. Methods Cell Biol 44:715–732.

Bixby JL, Bookman RJ. 1996. Intracellular mechanisms of axongrowth induction by CAMs and integrins: some unresolved issues.Perspect Dev Neurobiol 4:147–156.

Bray D. 1989. Growth cone formation and navigation: axonal growth.Curr Opin Cell Biol 1:87–90.

Bridgman PC, Dailey ME. 1989. The organization of myosin and actinin rapid frozen nerve growth cones. J Cell Biol 108:95–109.

Brittis PA, Lemmon V, Rutishauser U, Silver J. 1995. Unique changesof ganglion cell growth cone behavior following cell adhesion mol-ecule perturbations: a time-lapse study of the living retina. Mol CellNeurosci 6:433–449.

Brugge JS, Cotton PC, Queral AE, Barrett JN, Nonner D, Keane RW.1985. Neurones express high levels of a structurally modified,activated form of pp60c-src. Nature 316:554–557.

Burden-Gulley SM, Pendergast M, Lemmon V. 1997. The role of celladhesion molecule L1 in axonal extension, growth cone motility,signal transduction. Cell Tissue Res 290:415–422.

Burridge K, Fath K. 1989. Focal contacts: transmembrane linksbetween the extracellular matrix and the cytoskeleton. Bioessays10:104–108.

Castelo L, Jay DG. 1999. Radixin is involved in lamellipodial stabilityduring nerve growth cone motility. Mol Biol Cell 10:1511–1520.

Chang HY, Takei K, Sydor AM, Born T, Rusnak F, Jay DG. 1995.Asymmetric retraction of growth cone filopodia following focalinactivation of calcineurin. Nature 376:686–690.

Cleveland DW, Hwo SY, Kirschner MW. 1977. Physical and chemicalproperties of purified tau factor and the role of tau in microtubuleassembly. J Mol Biol 116:227–247.

Cohen NR, Taylor JS, Scott LB, Guillery RW, Soriano P, Furley AJ.1998. Errors in corticospinal axon guidance in mice lacking theneural cell adhesion molecule L1. Curr Biol 8:26–33.

Cotton PC, Brugge JS. 1983. Neural tissues express high levels of thecellular src gene product pp60c-src. Mol Cell Biol 3:1157–1162.

Cremer H, Lange R, Christoph A, Plomann M, Vopper G, Roes J,Brown R, Baldwin S, Kraemer P, Scheff S, et al. 1994. Inactivation ofthe N-CAM gene in mice results in size reduction of the olfactorybulb and deficits in spatial learning. Nature 367:455–459.

Dahme M, Bartsch U, Martini R, Anliker B, Schachner M, Mantei N.1997. Disruption of the mouse L1 gene leads to malformations of thenervous system. Nat Genet 17:346–349.

Davis JQ, Bennett V. 1994. Ankyrin binding activity shared by theneurofascin/L1 /NrCAM family of nervous system cell adhesionmolecules. J Biol Chem 69:27163–27166.

Diamond P, Mallavarapu A, Schnipper J, Booth J, Park L, O’ConnorTP, Jay DG. 1993. Fasciclin I and II have distinct roles in thedevelopment of grasshopper pioneer neurons. Neuron 11:409–421.

Doherty P, Walsh FS. 1994. Signal transduction events underlyingneurite outgrowth stimulated by cell adhesion molecules. Curr OpinNeurobiol 4:49–55.

Doherty P, Walsh FS. 1996. CAM-FGF receptor interactions: a modelfor axonal growth. Mol Cell Neurosci 8:99–111.

Doherty P, Fruns M, Seaton P, Dickson G, Barton CH, Sears TA, WalshFS. 1990. A threshold effect of the major isoforms of NCAM onneurite outgrowth. Nature 343:464–466.

Edelmann W, Zervas M, Costello P, Roback L, Fisher I, HammarbackJA, Cowan N, Davies P, Wainer B, Kucherlapati R. 1996. Neuronalabnormalities in microtubule-associated protein 1B mutant mice.Proc Natl Acad Sci U S A 93:1270–1275.

Evans LL, Hammer J, Bridgman PC. 1997. Subcellular localization ofmyosin V in nerve growth cones and outgrowth from dilute-lethalneurons. J Cell Sci 110:439–449.

Feil R, Brocard J, Mascrez B, LeMeur M, Metzger D, Chambon P.1996. Ligand-activated site-specific recombination in mice. ProcNatl Acad Sci U S A 93:10887–10890.

Forscher P, Smith SJ. 1988. Actions of cytochalasins on the organiza-tion of actin filaments and microtubules in a neuronal growth cone.J Cell Biol 107:1505–1516.

Goldberg DJ, Burmeister DW. 1986. Stages in axon formation: observa-tions of growth of Aplysia axons in culture using video-enhancedcontrast-differential interference contrast microscopy. J Cell Biol103:1921–1931.

Goldberg DJ, Wu DY. 1996. Tyrosine phosphorylation and protrusivestructures of the growth cone. Perspect Dev Neurobiol 4:183–192.

Grant SG, Karl KA, Kiebler MA, Kandel ER. 1995. Focal adhesionkinase in the brain: novel subcellular localization and specificregulation by Fyn tyrosine kinase in mutant mice. Genes Dev9:1909–1921.

Harada A, Oguchi K, Okabe S, Kuno J, Terada S, Ohshima T,Sato-Yoshitake R, Takei Y, Noda T, Hirokawa N. 1994. Alteredmicrotubule organization in small-calibre axons of mice lacking tauprotein. Nature 369:488–491.

Henkemeyer M, Orioli D, Henderson JT, Saxton TM, Roder J, PawsonT, Klein R. 1996. Nuk controls pathfinding of commissural axons inthe mammalian central nervous system. Cell 86:35–46.

Hoffman KB. 1998. The relationship between adhesion molecules andneuronal plasticity. Cell Mol Neurobiol 18:461–475.

Hoffman-Kim D, Chen A, Xu A, Wang TF, Jay DG. 1999. pp60c-src butnot p59 is a negative regulator of neurite outgrowth in chick sensoryneurons. J Neurosci (in press).

HoltmaatAJ, OestreicherAB, Gispen WH, Verhaagen J. 1998. Manipu-lation of gene expression in the mammalian nervous system:application in the study of neurite outgrowth and neuroregenera-tion- related proteins. Brain Res Brain Res Rev 26:43–71.

Honig MG, Rutishauser US. 1996. Changes in the segmental patternof sensory neuron projections in the chick hindlimb under conditionsof altered cell adhesion molecule function. Dev Biol 175:325–337.

Honig MG, Petersen GG, Rutishauser US, Camilli SJ. 1998a. In vitrostudies of growth cone behavior support a role for fasciculationmediated by cell adhesion molecules in sensory axon guidanceduring development. Dev Biol 204:317–326.

Honig MG, Frase PA, Camilli SJ. 1998b. The spatial relationshipsamong cutaneous, muscle sensory and motoneuron axons duringdevelopment of the chick hindlimb. Development 125:995–1004.

Ignelzi MA Jr, Miller DR, Soriano P, Maness PF. 1994. Impairedneurite outgrowth of src-minus cerebellar neurons on the celladhesion molecule L1. Neuron 12:873–84.

Jay DG. 1988. Selective destruction of protein function by chromophore-assisted laser inactivation. Proc Natl Acad Sci USA 85:5454–5458.

Jay DG. 1996. Molecular mechanisms of directed growth cone motility.Perspect Dev Neurobiol 4:137–145.

Jay DG. 1999. Chromophore-assisted laser inactivation (CALI) toelucidate cellular mechanisms of cancer. Biochem Biophys Acta1424:1139–1148.

Jay DG, Keshishian H. 1990. Laser inactivation of fasciclin I disruptsaxon adhesion of grasshopper pioneer neurons. Nature 348:548–550.

Kamiguchi H, Hlavin ML, Lemmon V. 1998. Role of L1 in neuraldevelopment: what the knockouts tell us. Mol Cell Neurosci 12:48–55.

Kater SB, Rehder V. 1995. The sensory-motor role of growth conefilopodia. Curr Opin Neurobiol 5:68–74.

Kramer I, Hall H, Bleistein U, Schachner M. 1997. Developmentallyregulated masking of an intracellular epitope of the 180 kDa isoformof the neural cell adhesion molecule NCAM. J Neurosci Res 49:161–175

Kruger K, Tam AS, Lu C, Sretavan DW. 1998. Retinal ganglion cellaxon progression from the optic chiasm to initiate optic tractdevelopment requires cell autonomous function of GAP–43. J Neuro-sci 18:5692–5705.

Kuhn TB, Brown MD, Bamburg JR. 1998. Rac1-dependent actinfilament organization in growth cones is necessary for beta1-integrin-mediated advance but not for growth on poly-D-lysine. J Neurobiol37:524–540.

Kuhn TB, Brown MD, Wilcox CL, Raper JA, Bamburg JR. 1999.Myelin and collapsin–1 induce motor neuron growth cone collapsethrough different pathways: inhibition of collapse by opposingmutants of rac1. J Neurosci 19:1965–1975.

Lagenaur C, Lemmon V. 1987. An L1-like molecule, the 8D9 antigen, isa potent substrate for neurite extension. Proc Natl Acad Sci U S A84:7753–7757.

Lamb RF, Ozanne BW, Roy C, McGarry L, Stipp C, Mangeat P, Jay DG.1997. Essential functions of ezrin in maintenance of cell shape andlamellipodial extension in normal and transformed fibroblasts. CurrBiol 7:682–688.

Landmesser L, Dahm L, Schultz K, Rutishauser U. 1988. Distinct


roles for adhesion molecules during innervation of embryonic chickmuscle. Dev Biol 130:645–670.

Lemmon V, Farr KL, Lagenaur C. 1989. L1-mediated axon outgrowthoccurs via homophilic binding mechanism. Neuron 6;1597–1603.

Letourneau PC. 1996. The cytoskeleton in nerve growth cone motilityand axonal pathfinding. Perspect Dev Neurobiol 4:111–123.

Letourneau PC, Shattuck TA. 1989. Distribution and possible interac-tions of actin-associated proteins and cell adhesion molecules ofnerve growth cones. Development 105:505–519.

Letourneau PC, Snow DM, Gomez TM. 1994. Regulation of growthcone motility by substratum bound molecules and cytoplasmic[Ca21]. Prog Brain Res 103:85–98.

Liao JC, Roider J, Jay DG. 1994. Chromophore-assisted laser inactiva-tion of proteins is mediated by the photogeneration of free radicals.Proc Natl Acad Sci USA 91:2659–2663.

Liao JC, Berg LJ, Jay DG. 1995. Chromophore-assisted laser inactiva-tion of subunits of the T-cell receptor in living cells is spatiallyrestricted. Photochem Photobiol 62:923–929.

Lin CH, Forscher P. 1993. Cytoskeletal remodeling during growthcone-target interactions. J Cell Biol 121:1369–1383.

Lin CH, Espreafico EM, Mooseker MS, Forscher P. 1996. Myosindrives retrograde F-actin flow in neuronal growth cones. Neuron16:769–782.

Linden KG, Liao JC, Jay DG. 1992. Spatial specificity of chromophoreassisted laser inactivation of protein function. Biophys J 61:956–962.

Lipsich LA, Lewis AJ, Brugge JS. 1983. Isolation of monoclonalantibodies that recognize the transforming proteins of avian sar-coma viruses. J Virol 48:352–360.

Liu CA, Lee G, Jay DG. 1999. Tau is required for neurite outgrowthand growth cone motility of chick sensory neurons. Cell MotilCytoskeleton (in press).

Luo L, Liao YJ, Jan LY, Jan YN. 1994. Distinct morphogeneticfunctions of similar small GTPases: Drosophila Drac1 is involved inaxonal outgrowth and myoblast fusion. Genes Dev 8:1787–802.

Mack TGA, Koester, M, Pollerberg GE. 1999. The microtubule-associated protein MAP1B is involved in local stabilization ofturning growth cones. Mol Cell Neurosci (in press).

Matten WT, Aubry M, West J, Maness PF. 1990. Tubulin is phosphory-lated at tyrosine by pp60c-src in nerve growth cone membranes. JCell Biol 111:1959–1970.

McFarlane S, Cornel E, Amaya E, Holt CE. 1996. Inhibition of FGFreceptor activity in retinal ganglion cell axons causes errors intarget recognition. Neuron 17:245–254.

Mitchison T, Kirschner M. 1988. Cytoskeletal dynamics and nervegrowth. Neuron 1:761–772.

Mochly-Rosen, D, L.M. Kauvar. 1998. Modulating protein kinase Csignal transduction. Adv Pharmacol 44:91–145.

Moos M, Tacke R, Scherer H, Teplow D, Fruh K, Schachner M. 1988.Neural adhesion molecule L1 as a member of the immunoglobulinsuperfamily with binding domains similar to fibronectin. Nature334:701–703.

Mueller BK. 1999. Growth cone guidance: first steps towards a deeperunderstanding. Annu Rev Neurosci 22:351–388.

Muller BK, Bonhoeffer F. 1995. Molecular inactivation. Spatially andtemporally defined molecular knockouts. Curr Biol 5:1255–1256.

Muller BK, Jay DG, Bonhoeffer F. 1996. Chromophore-assisted laserinactivation of a repulsive axonal guidance molecule. Curr Biol6:1497–1502.

Murray BA, Hemperly JJ, Prediger EA, Edelman GM, CunninghamBA. 1986. Alternatively spliced mRNAs code for different polypep-tide chains of the chicken neural cell adhesion molecule (N-CAM). JCell Biol 102:189–193.

Pollerberg GE, Beck-Sickingen A. 1993. A functional role for themiddle extracellular region of the neural cell adhesion molecule(NCAM) in axonal fasciculation and orientation. Dev Biol 56:324–340.

Rutishauser U, Gall WE, Edelman GM. 1978. Adhesion among neuralcells of the chick embryo. IV. Role of the cell surface molecule CAMin the formation of neurite bundles in cultures of spinal ganglia. JCell Biol 79:382–393.

Sattilaro RF, Dentler WL, LeCluyse EL. 1981. Microtubule-associated-proteins (MAPs) and the organization of actin filaments in vitro. JCell Biol 90:467–473.

Schmucker D, Su AL, Beermann A, Jackle H, Jay DG. 1994. Chromo-phore-assisted laser inactivation of patched protein switches cellfate in the larval visual system of Drosophila. Proc Natl Acad SciUSA 91:2664–2668.

Schroder R, Jay DG, Tautz D. 1999. Elimination of EVE protein by

CALI in the short germ band insect Tribolium suggests a conservedpair-rule function for even skipped. Mech Dev 80:191–195.

Schroeder R, Tautz D, Jay DG. 1996. Chromophore assisted laserinactivation of even skipped in Drosophila phenocopies genetic lossof function. Dev Genes Eval 206:86–88.

Selden SC, Pollard TD. 1983. Phosphorylation of microtubule-associated proteins regulates their interaction with actin filaments.J Biol Chem 258:7064–7071.

Shah EM, Jay DG. 1993. Methods for ablating neurons. Curr OpinNeurobiol 3:738–742.

Song HJ, Ming GL, Poo MM. 1997. C-AMP-induced switching inturning direction of nerve growth cones. Nature 388:275–279.

Soriano P, Montgomery C, Geske R, Bradley A. 1991. Targeteddisruption of the c-src proto-oncogene leads to osteopetrosis in mice.Cell 64:693–702.

Stoeckli ET, Landmesser LT. 1995. Axonin–1, Nr-CAM, Ng-CAM playdifferent roles in the in vivo guidance of chick commissural neurons.Neuron 14:1165–1179.

Stoeckli ET, Sonderegger P, Pollerberg GE, Landmesser LT. 1997.Interference with axonin–1 and NrCAM interactions unmasks afloor-plate activity inhibitory for commissural axons. Neuron 18:209–221.

Strittmatter SM, Fankhauser C, Huang PL, Mashimo H, FishmanMC. 1995. Neuronal pathfinding is abnormal in mice lacking theneuronal growth cone protein GAP–43. Cell 80:445–452.

Sydor AM, Su AL, Wang FS, Xu A, Jay DG. 1996. Talin and vinculinplay distinct roles in filopodial motility in the neuronal growth cone.J Cell Biol 134:1197–1207.

Tabb JS, Molyneaux BJ, Cohen DL, Kuznetsov SA, Langford GM.1998. Transport of ER vesicles on actin filaments in neurons bymyosin V. J Cell Sci 111:3221–3234.

Takei Y, Kondo S, Kanada A, Thomata S, Noda T, Hirokawa N. 1997.Delayed development of nervous system in mice homozygous fordisrupted microtubule-associated protein 1B (MAP1B) gene. J CellBiol 137:1615–1626.

Takei K, Shin RM, Inoue T, Kato K, Mikoshiba K. 1998a. Regulation ofnerve growth mediated by inositol 1,4,5-trisphosphate receptors ingrowth cones. Science 282:1705–1708.

Takei Y, Junlin T, Harada A, Inomata-Terada S, Hirokawa N. 1998b.Abnormal development of nervous system in mice lacking tau andMAP1B. Mol Biol Cell 9:394a.

Takei K, Chan TA, Wang FS, Deng H, Rutishauser U, Jay DG. 1999.The neural cell adhesion molecules L1 and NCAM–180 act indifferent steps of neurite outgrowth. J Neurosci (in press). TanakaE, Sabry J. 1995. Making the connection: cytoskeletal rearrange-ments during growth cone guidance. Cell 83:171–176.

Tanaka E, Kirschner MW. 1995. The role of microtubules in growthcone turning at substrate boundaries. J Cell Biol 128:127–137.

Tsukita S, Yonemura S. 1997. ERM proteins: head-to-tail regulation ofactin-plasma membrane interaction. Trends Biochem Sci 22:53–58.

Tsukita S, Oishi K, Sato N, Sagara J, Kawai A. 1994. ERM familymembers as molecular linkers between the cell surface glycoproteinCD44 and actin-based cytoskeletons. J Cell Biol 126:391–401.

Wagner MC, Barylko B, Albanesi JP. 1992. Tissue distribution andsubcellular localization of mammalian myosin I. J Cell Biol 119:163–170.

Wang FS, Jay DG. 1996. Chromophore-assisted laser inactivation(CALI): probing protein function in situ with a high degree of spatialand temporal resultion. Trends Cell Biol 6:442–445.

Wang FS, Wolenski JS, Cheney RE, Mooseker MS, Jay DG. 1996.Function of myosin-V in filopodial extension of neuronal growthcones. Science 273:660–663.

Williams AF, Barclay AN. 1988. The immunoglobulin superfamily:domains for cell surface recognition. Annu Rev Immunol 6:381–405.

Williams EJ, Furness J, Walsh FS, Doherty P. 1994. Characterizationof the second messenger pathway underlying neurite outgrowthstimulated by FGF. Development 120:1685–1693.

Worley TL, Holt CE. 1996. Expression and herbimycin A-sensitivelocalization of pp125FAK in retinal growth cones. Neuroreport7:1133–1137.

Wu X, Bowers B, Wei Q, Kocher B, Hammer JA 3rd. 1997. Myosin Vassociates with melanosomes in mouse melanocytes: evidence thatmyosin V is an organelle motor. J Cell Sci 110:847–859.

Wylie SR, Wu PJ, Patel H, Chantler PD. 1998. A conventional myosinmotor drives neurite outgrowth. Proc Natl Acad Sci U S A 95:12967–12972.

Yamada, K.M., B.S. Spooner, N.K. Wessells. 1971. Ultrastructure andfunction of growth cones and axons of cultured nerve cells. J CellBiol 49:614–635.


micro-scale chromophore-assisted laser inactivation of nerve growth cone proteins

Documents