Axon Guidance: Proteins Turnover inTurning Growth Cones

Dispatch

Gianluca Gallo and Paul C. Letourneau

Accurate navigation by a neuronal growth conerequires the modulation of the growth cone’s respon-siveness to spatial and temporal changes in expres-sion of guidance cues. These adaptations involvelocal protein synthesis and turnover in growth conesand distal axons.

During the development of a nervous system, patternsof axonal connections are formed as motile growthcones of developing axonal terminals detect andrespond to characteristically distributed extracellularguidance cues [1,2]. Growth cones protrude finger-likefilopodia and veil-like lamellipodia (Figure 1A), whichdetect guidance molecules and typically respondeither ‘positively’ by moving toward the source of acue (Figure 1B) or ‘negatively’ by avoiding the sourceof a cue (Figure 1C). Guidance cues direct growthcone migration by regulating cytoskeletal functions[3–5]. Filopodial and lamellipodial movements resultfrom the dynamics and organization of actin filaments:‘positive’ cues promote actin filament polymerization,while ‘negative’ cues cause actin depolymerizationand reorganization. The local balance of actin filamentdynamics and organization within a growth conedetermines the direction of axon growth: for example,contact with a negative guidance cue results in theinhibition of lamellipodial and filopodial production onthe side of the growth cone making the contact [6].

The path of a growth cone from its neuronal site oforigin to its synaptic target is divided into segments,in which spatial and temporal variations in guidancecues accompany changes in developing tissues.Growth cone responsiveness to guidance cues alsochanges as growth cones navigate along their path-ways. For example, to maintain a chemotropic responseover a long distance, a growth cone must be able todetect small, local concentration differences over a range of several orders of magnitude [3]. Whenascending a gradient of a positive cue, a growth conemust turn away from a concentration that had earlierelicited actin polymerization. Does this adaptationinvolve adjustments in the sensitivity or number ofguidance cue receptors, or in signaling triggered byreceptor–ligand binding? Is the signal triggered by a guidance cue at the high end of a gradient greaterthan when the growth cone is at the low end of thegradient or does the strength of the cytoplasmic signalremain constant? In other locations, growth conesdevelop new sensitivities to a guidance cue that earlierwas ignored. What signals trigger the expression of a

guidance cue receptor? Here we review three recentpapers [7–9] on the regulation of growth cone behav-iors in response to guidance cues. These paperspresent evidence that localized protein synthesis andproteolysis are required for growth cone responses toguidance cues.

Resetting the Growth ConeIn order to examine how growth cones remain sen-sitive to a range of guidance cue concentrations, Minget al. [7] investigated adaptation to guidance cues.Spinal neuron growth cones of the frog Xenopus turntoward a source of either brain derived growth factor(BDNF) or netrin-1. Exposure to a uniform concentra-tion of a cue renders growth cones unable to respondto a gradient of that cue, a process termed adaptation.After 60–90 minutes exposure to the cue, however,growth cones regain the ability to respond to a gradi-ent. This resumption of responsiveness to a gradientof the cue is termed resensitization. Thus, growth conesfirst adapt to a guidance cue but then become resen-sitized to it.

Ming et al. [7] investigated the signaling required toadapt and resensitize to guidance cues. Cytosolic[Ca2+] is an important regulator of axon growth [2–4].Ming et al. reported that adaptation correlated withincreased growth cone cytosolic [Ca2+], but directelevation of [Ca2+] alone did not produce adaptationto guidance cues. Gradients of the guidance cuecaused elevated cytosolic [Ca2+] in growth cones.During the adaptation period, however, gradients ofguidance cues failed to increase cytosolic [Ca2+]levels. Following resensitization, gradients could againelicit [Ca2+] elevation, although the base-line [Ca2+]was elevated relative to naïve growth cones. Thesedata indicate an initial failure and subsequent restora-tion of [Ca2+] signaling by guidance cues during adap-tation and resensitization, respectively.

Ming et al. [7] went on to demonstrate that resensi-tization requires axonal protein synthesis (Figure 1D).They found that protein synthesis inhibitors blockedresensitization, and axons severed from the cell bodyunderwent protein synthesis-dependent resensitiza-tion. Biochemical studies revealed that netrin-1 andBDNF both activate the mitogen-associated protein(MAP) kinase. Pharmacological evidence indicatedthat MAP kinase signaling is required for resensitiza-tion. The MAP kinase pathway signals to the nucleusand alters gene transcription. Ming et al. [7] suggestthat activation of MAP kinase signaling by guidancecues may modulate local protein synthesis in axons,or alternatively it may be required as additionalsignaling in conjunction with protein synthesis toresensitize growth cones.

Local Protein Synthesis at an Intermediate TargetSpinal cord commissural axons are attracted to theventral midline, a source of the attractant netrin. Once

Current Biology, Vol. 12, R560–R562, August 20, 2002, ©2002 Elsevier Science Ltd. All rights reserved. PII S0960-9822(02)01054-0

Department of Neuroscience, University of Minnesota , 6-145Jackson Hall, 321 Church St SE, Minneapolis, Minnesota55455, USA. E-mail: letou001@umn.edu

at the midline, the axons cross, turn and extend intolongitudinal tracts, never re-crossing the midline. Thischange in growth cone behavior involves loss of res-ponsiveness to netrin [10]. Brittis et al. [8] have recentlyprovided evidence that the commissural axons alsoupregulate their expression of the tyrosine kinasereceptor EphA2 on the distal axon segments. Axonallylocated EphA2 mRNA is translated in the contralateralgrowth cones, and the receptors are expressed ontheir surfaces, giving the distal growth cones newresponsivity along the next segment of their path.Brittis et al. [8] found that the EphA2 mRNA in thedistal axons contains highly conserved translationalcontrol sequences, which may be activated when theaxons reach the contralateral side of the spinal cord.

Protein Synthesis and Turnover in Growth ConesCampbell and Holt [9] investigated the role of proteinsynthesis in responses of Xenopus retinal growth conesto the negative cue semaphorin 3A and the positivecue netrin-1. By immunocytochemistry, mRNA, ribo-somes and the translation factor eIF-4E were found ingrowth cones. Campbell and Holt [9] next investigatedwhether the translational machinery in growth conescontributes to guidance by testing whether proteinsynthesis is required for guidance by gradients ofcues (Figure 1B,C). Inhibition of protein synthesisabolished growth cone responses to gradients of bothsemaphorin 3A and netrin. Inhibition of transcriptiondid not alter growth cone responses. Importantly, bysevering axons near the cell body, Campbell and Holt[9] demonstrated that the required protein synthesisoccurs in the axon. Thus, protein synthesis in the axon,but not mRNA transcription, is required for growthcones to respond to guidance cues.

Do guidance cues activate protein synthesis inaxons? By quantifying the signal from growth conesstained for the phosphorylated, inactive form of elon-gation factor eIF-4E, Campbell and Holt [9] demon-strated eIF-4E is activated in response to guidancecues. Additionally, 3H-leucine incorporation into proteinwas stimulated by guidance cues in axons that hadbeen separated from their cell bodies, providing directevidence for guidance cue-induced protein synthesisin axons.

The observation that growth cone guidance requireslocal protein synthesis suggests that protein degrada-tion could also be involved in guidance. The additionof ubiquitin to proteins targets them for proteasome-mediated degradation, so Campbell and Holt [9]investigated whether components of the ubiquitin–proteasome system are present in growth cones.Immunocytochemistry revealed the presence of thisproteolysis machinery in growth cones. They foundthat inhibitors of proteasome activity blocked guid-ance by positive and negative guidance cues. Impor-tantly, the intensity of staining with an antibody againstubiquitinated proteins revealed a large increase inubiquitination in response to guidance cues.

The results of Campbell and Holt [9] demonstratethat protein synthesis and turnover are required torespond to guidance cues (Figure 1D). Interestingly,although both semaphorin 3A and lysophosphatidicacid act as negative guidance cues, semaphorin 3Arequires only protein synthesis while lysophosphatidicacid requires only proteasome activity. The responseto netrin-1 requires both protein synthesis and prote-olysis. Thus, although protein turnover is affected byguidance cues, the branch of the turnover pathwayinvolved differs according to the specific cue involved,and guidance cues activate multiple pathways tocontrol protein turnover in growth cones.

A New Dimension to Growth ConesCollectively, the studies by Campbell and Holt [9],Brittis et al. [8] and Ming et al. [7] provide new func-tions for protein synthesis in axons and growth cones[11]. These observations demonstrate that proteinsynthesis and degradation in axonal growth cones is important during axon guidance and open newavenues. What are the mRNA species that contributeto resensitization and axon guidance? How are thesemRNAs targeted to the growth cone? What triggerstranslation of EphA2 mRNA after commissural axonscross the midline? What pathways regulate proteinsynthesis at the growth cone? What is the mechanismby which proteins synthesized in response to sema-phorin 3A mediate growth cone collapse? What arethe differences in the signaling of MAP kinase to thenucleus versus the axon? What targets proteins for

Current BiologyR561

Figure 1.

(A) Example of a chick retinal ganglion cell axonal growth cone. Growth cones extend (B) towards the source of a chemoattractant(upper right corner) and (C) away from the source of a chemorepellent guidance cue (upper right corner). (D) A diagram illustrating thesuggested roles of protein synthesis and degradation in mediating the effects of guidance cues on axon navigation.

Current Biology

A B C D

Filopodium

Lamellipodium

Chemoattraction Chemorepulsion

Guidance cues

Growth coneprotein content

Growth conebehavior

Axon growth

Positive Negative

Axonal proteinsynthesis

Proteindegradation

DispatchR562

proteasome-mediated proteolysis? In proteolysis-mediated growth cone collapse, are specific proteinstargeted for destruction?

References1. Yu, T.W. and Bargmann, C.I. (2001). Dynamic regulation of axon

guidance. Nat. Neurosci. 4, 1169–1176.2. Muller, B.K. (1999). Growth cone guidance: first steps towards a

deeper understanding. Annu. Rev. Neurosci. 22, 351–388.3. Song, H-j. and Poo, M-m. (2001). The cell biology of neuronal navi-

gation. Nat. Cell Biol. 3, E81–E88.4. Letourneau, P.C. (1996). The cytoskeleton in nerve growth cone

motility and axonal pathfinding. Perspect. Dev. Neurobiol. 4,111–123.

5. Gallo, G. and Letourneau, P.C. (2000). Neurotrophins and thedynamic regulation of the neuronal cytoskeleton. J. Neurobiol. 44,159–173.

6. Fan, J. and Raper, J.A. (1995). Localized collapsing cues can steergrowth cones without inducing their full collapse. Neuron 14,263–274.

7. Ming, G-l., Wong, S.T., Henley, J., Yuan, X-b., Song, H-j., Spitzer,N.C. and Poo, M-m. (2002). Adaptation in the chemotactic guidanceof nerve growth cones. Nature 417, 411-418.

8. Brittis, P.A., Lu, Q. and Flanagan, J.G. (2002). Axonal protein syn-thesis provides a mechanism for localized regulation at an interme-diate target. Cell 110, 223-235.

9. Campbell, D.S. and Holt, C.E. (2001). Chemotropic responses ofretinal growth cones mediated by rapid local protein synthesis anddegradation. Neuron 32, 1023–1026.

10. Stein, E. and Tessier-Lavigne, M. (2001). Hierarchical organizationof guidance receptors: silencing of netrin attraction by Slit througha Robo/DCC receptor complex. Science 291, 1928–1938.

11. Zhang, H.L., Eom, T., Oleynikov, Y., Shenoy, S.M., Liebelt, D.A.,Dictenberg, J.B., Singer, R.H. and Bassell, G.J. (2001). Neu-rotrophin-induced transport of a ß-actin mRNP complex increasesß-actin levels and stimulates growth cone motility. Neuron 31,261–275.