The Arp2/3 complex, UNC-115/abLIM, and UNC-34/Enabled regulate axon guidance and growth cone filopodia formation in Caenorhabditis elegans
© Norris et al; licensee BioMed Central Ltd. 2009
Received: 21 July 2009
Accepted: 2 October 2009
Published: 2 October 2009
While many molecules involved in axon guidance have been identified, the cellular and molecular mechanisms by which these molecules regulate growth cone morphology during axon outgrowth remain to be elucidated. The actin cytoskeleton of the growth cone underlies the formation of lamellipodia and filopodia that control growth cone outgrowth and guidance. The role of the Arp2/3 complex in growth cone filopodia formation has been controversial, and other mechanisms of growth cone filopodia formation remain to be described.
Here we show that mutations in genes encoding the Arp2/3 complex (arx genes) caused defects in axon guidance. Analysis of developing growth cones in vivo showed that arx mutants displayed defects in filopodia and reduced growth cone size. Time-lapse analysis of growth cones in living animals indicated that arx mutants affected the rate of growth cone filopodia formation but not filopodia stability or length. Two other actin modulatory proteins, UNC-115/abLIM and UNC-34/Enabled, that had been shown previously to affect axon guidance had overlapping roles with Arp2/3 in axon guidance and also affected the rate of filopodia initiation but not stability or length.
Our results indicate that the Arp2/3 complex is required cell-autonomously for axon guidance and growth cone filopodia initiation. Furthermore, they show that two other actin-binding proteins, UNC-115/abLIM and UNC-34/Enabled, also control growth cone filopodia formation, possibly in parallel to Arp2/3. These studies indicate that, in vivo, multiple actin modulatory pathways including the Arp2/3 complex contribute to growth cone filopodia formation during growth cone outgrowth.
The growth cone of a developing axon senses and responds to extracellular cues, resulting in the migration of the growth cone and thus axon to its correct target region in the nervous system [1, 2]. Growth cones display dynamic, actin-based lamellipodial protrusions ringed by filopodia that together guide the growth cone to its target [3–5]. The Arp2/3 complex is a seven-member protein complex that nucleates actin filaments from the sides of pre-existing actin filaments. In cultured cells, the Arp2/3 complex is necessary to form the network of branched actin filaments underlying lamellipodia [6–8].
The Arp2/3 complex has been implicated in axon pathfinding in vivo. In Drosophila, the Arp2/3 complex and its regulators WAVE/Scar and Kette are required for proper axon pathfinding , and in Caenorhabditis elegans, the Arp2/3 regulators WAVE-1/WAVE, WASP-1/WASP, GEX-2/Sra-1, and GEX-3/Kette are required for proper axon guidance . However, in these studies, the effects of Arp2/3 on growth cone outgrowth were not described. The role of the Arp2/3 complex in growth cone morphology has been controversial. In two studies using cultured hippocampal neurons, inhibition of Arp2/3 activity resulted in increased axon length with little or no effect on growth cone filopodia or morphology [11, 12]. Another recent study using cultured hippocampal neurons and neuroblastoma cells showed that Arp2/3 knock-down resulted in defective actin organization and defective lamellipodia and filopodia formation in the growth cone . Studies reported here show that Arp2/3 is indeed required for growth cone morphology and filopodia formation in vivo, and that Arp2/3 is required specifically for the rate of filopodia initiation but not filopodial stability or length. Thus, our studies demonstrate that Arp2/3 is required for growth cone filopodia formation and extend these findings in vivo, explaining the axon pathfinding defects observed in Arp2/3 mutants. Furthermore, UNC-115/abLIM has been implicated in axon pathfinding in C. elegans [14–16] and Drosophila , but its role in growth cone dynamics and morphology during outgrowth is not understood. Our results here show that UNC-115/abLIM is also involved in the initiation of growth cone filopodia.
Experiments described here in C. elegans show that mutations in genes encoding subunits of the Arp2/3 complex (arx genes) cause defects in axon guidance, and that Arp2/3 is required cell-autonomously in neurons for proper axon guidance. Analysis of growth cones during outgrowth in vivo revealed that arx mutants displayed reduced numbers of growth cone filopodia and also reduced growth cone size. Time-lapse analysis of growth cones in living animals indicated that the Arp2/3 complex controlled the rate of growth cone filopodia initiation but not filopodia stability or length. These results indicate that, in C. elegans, the Arp2/3 complex is a key regulator of growth cone filopodia initiation during outgrowth.
The actin modulatory proteins UNC-34/Enabled and UNC-115/abLIM have been shown previously to affect axon guidance in C. elegans [15, 16, 18]. Enabled is a key regulator of filopodia formation in many systems, including neurons, and is thought to act by blocking actin-capping activity, thus allowing for long filament growth in filopodia [19, 20], although recent studies indicate that Enabled might have anti-capping independent roles  that are involved in filopodia formation .
The actin binding protein UNC-115/abLIM controls axon pathfinding and lamellipodia and filopodia formation in cultured cells and in C. elegans neurons [14, 15, 23]. We show here that unc-115 also caused defects in growth cone filopodia initiation and growth cone size, similar to arx.
The role of Enabled in growth cone filopodia formation has been described previously, and we found that UNC-34/Enabled affected growth cone filopodia initiation, but did not affect filopodial growth or stability once a filopodium formed. This is in contrast to previous studies that indicated that Enabled also affected filopodial elongation . In sum, our results indicate that three distinct actin modulatory molecules, Arp2/3, UNC-115/abLIM, and UNC-34/Enabled, control filopodia initiation in growth cones in vivo that are required for proper axon guidance
Materials and methods
All experiments were performed at 20°C using standard C. elegans techniques . The following mutations and transgenic constructs were used: X: unc-115(mn481 and ky275), lqIs2 [osm-6::gfp], mig-2(mu28); I: arx-7(ok1118), hT2 [bli-4(e937) let-?(q782) qIs48] (I;III), lqIs40 [gcy-32::gfp]; II: juIs76 [unc-25::gfp]; III: arx-4(ok1093), sC1 [dpy-1(s2170)]; IV: ced-10(n1993), lqIs3 [osm-6::gfp], nT1 [qIs51] (IV;V); V: arx-2(ok1269), unc-34(e951). The chromosomal locations of lqIs49 [gcy-32::gfp] and lqIs75 [gcy-32::arx-7::mCherry] were not determined. The arx-2(+), arx-4(+), and arx-7(+) whole-gene regions were maintained as extrachromosomal arrays. Extrachromosomal arrays were generated by germ line microinjection and integrated into the genome by standard techniques .
arx-2, arx-4, and arx-7 mutations were balanced by the rearrangements nT1, sC1, and hT2, respectively. nT1 and hT2 harbored transgenes that drove gfp expression in the pharynx, whereas sC1 had no gfp marker. arx-2 and arx-7 homozygotes were identified by lack of pharyngeal green fluorescent protein (GFP), whereas arx-4 homozygotes were identified by the protruding vulva (Pvl) phenotype in young adults. arx-4 L1 larvae were not imaged because they could not be unambiguously identified as could arx-2 and arx-7. Double mutants with ced-10 were confirmed by the gonad misrouting phenotype and by the persistent cell corpse phenotype , and double mutants with mig-2 were confirmed by the gonad misrouting phenotype . Double mutants with unc-34 and unc-115 were confirmed by the uncoordinated phenotype of these mutants.
arx transgene construction
For cell-specific rescue, the coding regions of arx-4 and arx-7 (from the initiator methionine to the stop codon) were amplified from N2 genomic DNA and placed behind the osm-6 promoter to drive PDE expression, or the gcy-32 promoter to drive PQR expression. For arx-7, the stop codon was not included, and the coding region was fused in frame to the coding region for GFP.
Analysis of PDE axon guidance defects
As previously described, PDE axons were visualized with an osm-6 promoter::gfp transgene . osm-6::gfp, expressed in all ciliated sensory neurons including PDE, allows unambiguous identification of the PDE axon from other axons . A PDE axon was considered defective if the axon failed to reach the ventral nerve cord or if it reached the ventral nerve cord at a position greater than a 45° angle from the PDE cell body. Young adult animals were scored, although arx mutants were often scored as L3 or L4 larvae. Percentages of defective PDE axon guidance in each strain were determined, and significance was judged by a two-sided t-test with unequal variance unless otherwise noted.
Imaging and analysis of PQR dendrite development
The PQR dendrite was visualized using gcy-32::gfp transgenes . Larvae were synchronized by washing adults and larvae from a plate with many eggs and allowing the eggs to hatch. Newly hatched larvae were washed from the plate at 0.5 h intervals and allowed to develop for the following times before imaging: 6 h, 6.5 h, 7 h, 7.5 h, and 8 h. Animals were imaged using a Leica DMR microscope with a Qimaging Retiga EXi camera. Z-stacks of each PQR were captured and were subject to nearest neighbor deconvolution using Openlab software (Improvision, Waltham, MA USA).
Filopodia were defined as growth cone protrusions that were less than 0.5 μm in width. The number of filopodial protrusions from the growth cone of PQR neurons at the 7 to 7.5 h timepoint was determined by scrolling through each Z-stack of images. An average number and standard deviation of filopodial protrusions per PQR dendrite was determined, and significance of difference between genotypes was determined by a two-sided t-test with unequal variance unless otherwise noted.
The lamellipodial perimeter of the growth cone was used as an indicator of growth cone size. At the 7 to 7.5 h timepoint, the distal 3 μm of the tip of the PQR dendritic protrusion was traced using ImageJ software, and a value representing the perimeter of this region was derived. An average was derived from multiple tracings, and significance of difference between genotypes was determined by a two-sided t-test with unequal variance unless otherwise noted.
Time lapse imaging of VD growth cones
VD growth cones were imaged using an unc-25::gfp transgene juIs76 . Early L2 stage larvae were isolated by washing hatched worms off of a plate and allowing eggs to hatch for one-half hour. Newly hatched larvae were washed off and allowed to develop at 20°C for 16 h. Larvae with visible VD growth cones were imaged. Animals were transferred to a 2% agarose pad with a drop of 10 mM muscimol (Sigma-Aldrich, St. Louis, MO USA) in M9 . The drop was allowed to evaporate for 4 to 5 minutes before a coverslip was placed over the animals. Growth cones that were just emerging from the ventral muscle quadrant were imaged with a Leica DMR microscope with a BD CARV II wide-field light source and QImaging Rolera mGi camera at intervals of 120 s. Total duration of time-lapse imaging ranged from 20 to 60 minutes.
Dynamic projections emanating from the growth cone less than 0.5 μm in width were scored as filopodia. Filopodia length was measured using ImageJ software. Growth cone translocation speed was determined by measuring the distance from the distal tip of the lamellipodial portion of the growth cone to a fixed reference point along the axon, often a point at which the axon has crossed a lateral axon tract, which can cause a slight bulge in the axon. Significance of difference was determined by a two-sided t-test with unequal variance.
Mutations in arx-2, arx-4, and arx-7 cause PDE axon guidance defects
The Arp2/3 complex is composed of seven molecules, Arp2, Arp3, and ArpC1-ArpC5 . The C. elegans genome contains genes encoding each of these subunits. Deletion mutations in genes encoding three subunits were analyzed; arx-2 Arp2, arx-4 ArpC2, and arx-7 ArpC5 (Figure 1). arx-2(ok1269) was a 1,330-bp deletion that removed part of exon 4 and all of exons 5 and 6. arx-4(ok1093) was a 772-bp deletion that removed the 3' splice site of intron 4 and part of exon 5. arx-7(ok1118) was a 1,352-bp deletion that removed most of exon 3 of arx-7 (almost half of the arx-7 coding region) and extended into the downstream gene Y54E10BR.1, removing exon 1 and most of intron 1. While arx-2(ok1269) and arx-4(ok1093) are predicted to affect only arx-2 and arx-4, respectively, arx-7(ok1118) might affect both arx-7 and Y54E10BR.1.
From mothers heterozygous for each mutation (wild-type maternal contribution, M+), arx-2(ok1269M+), arx-4(ok1093M+), and arx-7(ok1118M+) homozygotes grew slowly and arrested as larvae or young adults. Animals that survived to adulthood were sterile and had protruding vulvae (Pvl phenotype). Most arx-2(ok1269M+) animals arrested before reaching adulthood in the L3 or L4 larval stage. arx-7(ok1118) animals arrested in L4 or early adulthood, and arx-4(ok1093) animals survived to adulthood but were sterile and Pvl. RNA interference of arx-2 and arx-7 caused embryonic lethality due to defects in gastrulation as previously described , indicating that arx-2, arx-7, and likely arx-4 have strong maternal contributions that are sufficient for embryonic and larval development. The variation in arrest phenotype of arx mutants could be due to differences in maternal contribution and not distinct functions of each arx gene.
To ensure that the phenotype observed in arx mutants was due to disruption of the arx gene, the entire wild-type genomic region for each locus was amplified by PCR and used in transgenic rescue experiments (see Materials and methods). arx-4 is the downstream gene in a C. elegans operon, and the entire operon region composed of the genes K02F3.1, K02F3.12, and Y6D11A.1 was included in the fragment. The arx-2(+) and arx-4(+) transgenes rescued the slow growth, sterility and pVul phenotype of arx-2(ok1269) and arx-4(ok1093), respectively, and transgenic animals were viable and fertile. Furthermore, PDE axon defects were completely rescued (Figure 2E; 15% to 3% for arx-2(ok1269M+) and 34% to 1% for arx-4(ok1093); P < 0.0001 for each). The arx-7(+) fragment rescued the slow growth and pVul components of the arx-7(ok1118) phenotype as well as PDE axon pathfinding defects (30% to 3%; P < 0.0001; Figure 2E). However, arx-7(ok1118); Ex [arx-7(+)] animals were still sterile. The arx-7(ok1118) deletion also affected the downstream gene Y54E10BR.1, which was not included in the rescuing fragment. The sterility of arx-7(ok118); Ex [arx-7(+)] transgenic animals might be due to the effects of ok1118 on Y54E10BR.1. These data indicate that the PDE axon guidance defects observed in arx mutants are due to mutation of the arx genes.
arx-4 acts cell-autonomously in PDE axon guidance
To determine if arx-4 acts in neurons to control axon pathfinding, a transgene was constructed consisting of the osm-6 promoter upstream of the arx-4 coding region. The osm-6 promoter is active exclusively in ciliated sensory neurons including PDE and has no apparent activity in other tissues [15, 28]. arx-4(ok1093M+) animals harboring the Posm-6::arx-4(+) transgene were slow-growing, Pvl, and sterile, similar to arx-4(ok1093) homozygotes alone, confirming that the Posm-6::arx-4(+) transgene was not broadly active. However, PDE axon guidance defects were rescued in these animals (34% to 2%; P < 0.0001; Figure 2E), suggesting that expression of arx-4 in the PDE neurons rescued arx-4 axon guidance defects. These results suggest that arx-4 acts cell-autonomously in the PDE neurons in axon guidance.
arx mutations enhance mutations in ced-10/Rac, mig-2/RhoG
Previous results indicated that the Rac-like molecules CED-10/Rac  and MIG-2/RhoG  act redundantly in axon guidance  and that the Arp2/3 activators WVE-1/WAVE and WSP-1/WASP might act in the CED-10/Rac and MIG-2/RhoG pathways, respectively . The loss-of-function alleles ced-10(n1993) and mig-2(mu28) enhanced the PDE axon pathfinding defects of arx-2(ok1269M+), arx-4(ok1093M+), and arx-7(ok1118M+) (Figure 2F). For example, ced-10(n1993); arx-4(ok1093M+) displayed 77% PDE axon guidance defects compared to 34% for arx-4(ok1093M+) alone; and mig-2(mu28); arx-7(ok1118M+) displayed 68% compared to 30% for arx-7(ok1118M+) alone (P < 0.0001 for all comparisons). Double mutants did not exhibit ectopic axon formation as observed in ced-10; mig-2 double mutants . That each arx mutation was enhanced by both ced-10(n1993) and mig-2(mu28) suggests that the Arp2/3 complex might act in both the CED-10 Rac and MIG-2 RhoG pathways in PDE axon guidance. Possibly, both WVE-1/WAVE and WSP-1/WASP pathways converge on regulation of Arp2/3 in axon guidance.
The actin-binding proteins UNC-115/abLIM and UNC-34/Enabled have redundant function with Arp2/3 in axon guidance
Previous studies showed that the actin-binding protein UNC-115/abLIM acts downstream of Rac signaling in axon guidance , raising the possibility that Rac signaling employs multiple actin-modulating pathways in axon guidance (UNC-115/abLIM and Arp2/3). Indeed, unc-115 mutations enhanced PDE guidance defects of arx mutants (Figure 2G). For example, unc-115(ky275); arx-2(ok1269M+) displayed 83% PDE axon guidance defects compared to 15% for arx-2(ok1269M+) alone (P < 0.0001 for all differences). Double mutants did not display enhanced ectopic axon formation as observed in double mutants of unc-115 with ced-10 and mig-2. That unc-115 mutations enhanced PDE axon guidance defects of arx mutations indicates that UNC-115/abLIM acts in parallel to Arp2/3 in PDE axon guidance, a result consistent with the idea that Rac GTPases employ both Arp2/3 and UNC-115/abLIM.
Previous studies showed that UNC-115/abLIM and UNC-34/Enabled had redundant roles in PDE axon guidance . As predicted by this result, UNC-34/Enabled and Arp2/3 also had parallel function in PDE axon guidance (Figure 2G). Alone, loss-of-function unc-34(e951) mutants displayed 8% PDE guidance defects. unc-34(e951); arx-4(ok1093) and unc-34(e951); arx-7(ok1118M+) displayed 91% and 97% PDE guidance defects compared to 35% and 35% for the arx mutants alone (P < 0.0001 for each comparison). Together, these data indicate that Arp2/3, UNC-115/abLIM, and UNC-34/Enabled all act redundantly to control PDE axon guidance.
Development of the posterior PQR neurite involves a growth-cone-like structure with filopodia
Mutations in arx genes caused axon guidance defects, but the effect of arx mutations on the growth cone during neurite extension was unclear. Developing neurites in arx mutants were analyzed. Transgenes used to assay PDE axon morphology were not visibly expressed early in PDE development during axon extension, including osm-6::gfp  and cat-1::gfp . Therefore, the development of a distinct neurite was examined. A gcy-32::gfp reporter  was used to analyze outgrowth of the posterior PQR neurite, which develops into a ciliated sensory dendrite .
arx mutants affect PQR growth cone morphology and filopodia formation
Posterior PQR neurite extension was assessed in arx-2(ok1269M+) and arx-7(ok1118M+) living larvae expressing gcy-32::gfp. These larvae exhibited a general growth delay, and timing of PQR dendritic events were slowed by approximately one-half hour. At 7 h, initial wedge-like cell shape with filopodia appeared normal in arx-2 and arx-7 mutants (Figure 4E). From 7 to 7.5 hours, arx-2 and arx-7 extended dorsal-posterior foot-like structures. However, these protrusions displayed fewer filopodial extensions than wild type (Figure 4F). Neurite consolidation appeared normal in arx-2 and arx-7, but from 7.5 to 8.5 hours, the growth cones at the distal tip displayed fewer filopodia (Figure 4G,H). Furthermore, fewer filopodia were evident along the length of the neurites. The posterior protrusions and growth cones of arx-2 and arx-7 mutants often appeared thinner and less ramified than in wild type, possibly indicating a growth cone lamellipodial defect as well (see Figure 5 for additional representative images of wild-type and arx-7(ok1118M+) growth cones).
The number of filopodia on the growth cones and neurites of wild type (7-8 h), arx-2(ok1269M+), and arx-7(ok1118M+) (7.5-8.5 h) were counted (Figure 6A; see Materials and methods). Wild type displayed an average of 4.89 filopodia per neurite whereas arx-2(ok1269M+) displayed 2.27 and arx-7(ok1118M+) 1.78 (P < 0.0001 for each; Figure 6B).
As mentioned previously, the growth cones of arx mutants often appeared less robust and smaller in size than wild type. To quantify this defect, the perimeters of growth cones (excluding filopodia) from wild type and mutant backgrounds were traced (Figure 6A; see Materials and methods). The growth cone was defined as the widened region at the tip of the neurite, starting at 3 μm from the distal growth cone tip. Wild-type growth cones displayed an average perimeter of 7.96 ± 0.91 μm (Figure 6C). arx-7 displayed a significantly reduced average growth perimeter (6.77 ± 0.51 μm; P < 0.001).
arx-7 acts cell-autonomously in PQR filopodia regulation
A transgene consisting of the gcy-32 promoter driving arx-7 tagged with coding region for GFP was constructed. This transgene was predicted to express full-length ARX-7 with GFP at the carboxyl terminus. Animals harboring this transgene showed ARX-7::GFP fluorescence in the URX, AQR, and PQR neurons and nowhere else (data not shown). arx-7(ok1118M+) animals harboring this transgene displayed the gross phenotype of arx-7(ok1118M+) alone (arrested larvae and sterile, Pvl adults), indicating that arx-7(+) activity was not widely provided by the Pgcy-32::arx-7::gfp transgene. However, this transgene rescued filopodia formation to wild-type levels in arx-7(ok1118M+) PQR neurites (Figure 6B): an average of 5.1 filopodia per neurite compared to 1.78 for arx-7(ok1118M+) alone (P < 0.0001). arx-7 rescue was not significantly different than wild type (5.1 versus 4.89; P = 0.71). These data indicate that arx-7 acts cell-autonomously in the regulation of growth cone filopodia.
unc-115/abLIM mutation enhanced PQR filopodia defects of arx-2 and arx-7
As reported above, unc-115 mutations enhanced PDE axon guidance defects of arx-2, arx-4, and arx-7, indicating redundancy of function in PDE axon guidance. The effect of unc-115 on PQR posterior neurite extension was examined. Alone, unc-115(ky275), a null allele, displayed no defects in PQR posterior neurite filopodia formation (for example, 4.61 filopodia versus 4.89 for wild type, P = 0.51; Figure 6B). However, unc-115(ky275) enhanced filopodia defects of arx-2(ok1269M+) and arx-7(ok1118M+). unc-115(ky275); arx-2(ok1269M+) displayed 1.22 filopodia per neurite compared to 2.27 for arx-2(ok1269M+) alone (P = 0.003), and unc-115(ky275); arx-7(ok1118M+) displayed 1.04 compared to 1.78 for arx-7(ok1118M+) alone (P = 0.007).
unc-115(ky275) mutants also displayed a reduced growth cone perimeter (Figure 6C; 6.52 ± 0.57 μm compared to 7.96 ± 0.91 μm for wild type; P < 0.001), indicating that UNC-115 might be involved in growth cone lamellipodium regulation. unc-115(ky275) displayed 41% PQR dendrite guidance defects (the undulating dendrite), compared to 19% for wild type (Figure 7C). Alone, unc-115(ky275) had no affect on filopodia, suggesting that unc-115 PQR dendrite guidance defects were due to affects on lamellipodia and not filopodia.
unc-115; arx double mutants displayed a defect not seen in unc-115, arx-2, or arx-7 single mutants, a rudimentary or absent posterior PQR neurite (Figure 4J,K): 12 of 70 PQR neurons from unc-115(ky275); arx-2(ok1269M+) and 9 of 75 unc-115(ky275); arx-7(ok1118M+) animals lacked a significant posterior protrusion. Some animals showed a rudimentary posterior protrusion (Figure 4J) whereas other showed no posterior neurite (Figure 4K). These observations suggest that in addition to being required for filopodia formation, the Arp2/3 complex and UNC-115/abLIM might also redundantly control neurite initiation or the ability of a neurite to form and extend.
unc-34/Enabled affects filopodia but not growth cone perimeter
As previously observed in the HSN neuron , unc-34(e951) mutants displayed severely reduced filopodia-like protrusions on the PQR dendritic neurite (Figures 4I and 6; 1.48, P < 0.0001). That unc-34 affected these thin, filopodia-like protrusions supports the notion that they are bona fide filopodia, as Enabled molecules have been shown to control filopodia formation in many systems, including C. elegans neurons [40, 41]. arx-7(ok1118M+); unc-34(e951) displayed significantly fewer filopodia than each single alone, but an additive effect of the two mutations combined could not be excluded (Figure 6B). Some arx-7(ok1118M+); unc-34(e951) double mutants lacked a strong posterior neurite from PQR (2 of 17), similar to arx; unc-115 double mutants, suggesting that UNC-34/Enabled and the Arp2/3 complex might control the ability of a neurite to form or to extend. unc-34 mutants did not display a reduction in PQR growth cone perimeter (7.78 ± 0.9 μm compared to 7.96 ± 0.91 μm for wild type; P = 0.387; Figure 6C).
These data indicate that Arp2/3 and UNC-115 affect both growth cone filopodia and growth cone perimeter, which might reflect lamellipodial protrusion. In contrast, unc-34 affected only growth cone filopodia but not growth cone perimeter. Thus, growth cone perimeter and number of filopodia are at least partially independent processes. For example, unc-115 alone affected growth cone perimeter but not filopodia, and unc-34 affected filopodia but not growth cone size. These results indicate that the filopodial defects seen in arx mutants might be a specific defect in filopodia and are unlikely to simply be a secondary effect of reduced lamellipodial protrusion.
ARX-7, UNC-115/abLIM, and UNC-34 affect filopodia initiation but not duration or length
The effects of Arp2/3, UNC-115, and UNC-34 on PQR growth cone filopodia formation could be due to decreased filopodial stability or half-life, or could be due to a decrease in the rate of filopodial initiation. We imaged growth cones from wild type and mutants over time to assess the dynamic effects of these mutations of filopodia formation. The commissural growth cones of the VD motor axons have been used in previous time-lapse studies [31, 42]. The VD axons begin extension during the L1/L2 larval molt and continue to extend early in the second larval stage. As the VD commissural growth cones emerge dorsally after traversing the ventral muscle quadrant, they display a spread appearance with multiple dynamic filopodial extensions . Until the growth cones reach the dorsal muscle quadrant, they generally retain a spread morphology with multiple filopodial extensions ( and our observations).
Growth cone translocation rates in cytoskeletal mutants.
t-test P-value versus wild-type
Wild type (6)
7.38 ± 1.86
4.08 ± 1.74
7.02 ± 1.32
3.84 ± 2.88
The Arp2/3 complex controls growth cone size and filopodia formation in vivo
The results presented here show that the Arp2/3 complex is required for the initiation of growth cone filopodia of C. elegans neurons in vivo. The role of Arp2/3 in growth cone morphology has been controversial, based on studies in cultured neurons. Some studies concluded that Arp2/3 was not involved in growth cone filopodia formation [11, 12], while another demonstrated a role for Arp2/3 in growth cone filopodia formation . Our results presented here clearly demonstrate that Arp2/3 is required for growth cone filopodia formation in vivo. The previous study that found no role of Arp2/3 in cultured hippocampal growth cone filopodia formation used expression of a dominant negative WAVE construct to inhibit Arp2/3 , whereas the study that found an effect of Arp2/3 in hippocampal growth cones used small interfering RNA knockdown of the p34-Arc subunit of the Arp2/3 complex . Possibly, the dominant-negative WAVE construct did not inhibit Arp2/3 as well as the small interfering RNA. In any case, our studies of C. elegans growth cones in vivo demonstrate that Arp2/3 is a key regulator of growth cone filopodia formation.
A previous study using cultured hippocampal neurons  showed that Arp2/3 was present in the growth cone at branch points of actin filaments as observed in other cell types, and filaments with Arp2/3 at their base contributed to filopodial bundles, suggesting that Arp2/3 provides a source of actin filaments for filopodial formation . The nature of the growth cone cytoskeletal defects in arx mutant growth cones is difficult to discern, as the growth cones are very small compared to those of cultured hippocampal neurons (<5 μm in diameter compared to approximately 20 μm for cultured hippocampal neurons). However, it is possible that there is reduced nucleation of lamellipodial actin filaments in arx-7 mutant growth cones, resulting in less availability of lamellipodial actin filaments to contribute to filopodial bundles. This idea is consistent with the time-lapse analysis of growth cone filopodia, which showed that arx mutants had reduced filopodial initiation rates compared to wild type (Figure 9). arx-7 mutant PQR growth cones also displayed a reduction in the size of the growth cone (Figures 5 and 6), consistent with a role of Arp2/3 in the nucleation of growth cone lamellipodial actin filaments. Interestingly, once filopodia formed in arx- 7 mutants, they had characteristics similar to wild type, including time of persistence/stability and maximal length. Thus, the Arp2/3 complex might primarily affect the initiation of filopodia and have only a minor role in filopodial stability or extension once a filopodium has initiated. However, we cannot rule out the possibility that filopodia initiate and collapse before they are visible as filopodia in our imaging assays.
The Arp2/3 complex is required for axon guidance
As a growth cone navigates its environment, filopodia are thought to act as sensors of extracellular cues and have a role in guiding the growth cone [43, 44]. Consistent with this idea, we found that arx mutants have defects in the guidance of the PDE and VD/DD axons and of the PQR dendrite. Reduction of filopodia on arx mutant growth cones might lead to growth cone guidance defects. However, arx axon guidance defects were incompletely penetrant (for example, 15 to 34% for the PDEs). This could be due to the action of the remaining few filopodia on arx mutant growth cones or could be due to lamellipodial guidance. Indeed, unc-34/Enabled mutants lack most or all growth cone filopodia but have relatively weak effects on axon guidance ([16, 18, 40] and this work).
UNC-115/abLIM controls growth cone size and filopodia formation
As mentioned above, the weak axon guidance defects of arx mutants could be explained by the lamellipodial guidance or by the activity of the remaining few filopodia on mutant growth cones, presumably formed by Arp2/3-independent mechanisms. UNC-115/abLIM has previously been implicated in axon pathfinding, but the role of the molecule in regulation of growth cone dynamics and morphology during outgrowth had not been demonstrated. We found that UNC-115/abLIM also affected the formation of growth cone filopodia. While unc-115 displayed no significant reduction of filopodia on PQR growth cones, it did enhance filopodia defects of arx-7. Furthermore, unc-115 displayed reduced filopodia initiation rate on VD/DD growth cones. These data indicate that UNC-115/abLIM is also a regulator of filopodia initiation similar to Arp2/3. That unc-115/abLIM mutants alone have defects in VD filopodia but not PQR filopodia suggests that different neurons have distinct requirements for UNC-115/abLIM in growth cone filopodia formation. This could be due to different guidance signals detected by these neurons that trigger distinct downstream signal transduction pathways, resulting in different requirements for UNC-115/abLIM.
UNC-115/abLIM binds to actin filaments via the carboxy-terminal villin headpiece domain (VHD), and promotes the formation of lamellipodia and filopodia in C. elegans neurons and cultured fibroblasts [15, 23]. UNC-115/abLIM might act as a bundling protein in filopodial bundles, or it might produce new actin filaments for use in filopodia, similar to Arp2/3. Indeed, the villin headpiece domain from human villin can nucleate actin polymerization , but this activity has not been described for UNC-115/abLIM. Our results here suggest that UNC-115/abLIM might also have a role in lamellipodia organization similar to Arp2/3, as unc-115 mutants also displayed reduced PQR growth cone size.
UNC-34/Enabled affects filopodia formation without affecting growth cone morphology
Enabled has been extensively implicated in filopodia formation, including on growth cones, and our results that unc-34 mutant growth cones had reduced filopodia numbers confirm this role of Enabled. In time-lapse studies, we found that unc-34 affected the rate of filopodia formation, suggesting that Enabled controls filopodia initiation similar to Arp2/3 and UNC-115/abLIM. This is similar to previous results in cultured neurons in response to netrin . We found that unc-34 had no effect on filopodia stability or length once a rare filopodium formed, indicating that UNC-34 has no role in filopodia outgrowth or stability. This is in contrast to cultured neurons, where Ena/VASP inhibition also resulted in filopodia extension defects .
In contrast to arx-7 and unc-115 mutants, unc-34 mutant PQR growth cones, while nearly devoid of filopodia, displayed no significant reduction in size. While all three mutations caused reduced filopodia initiation, this difference might hint at distinct mechanisms of action of these molecules. Possibly, Arp2/3 and UNC-115/abLIM primarily affect lamellipodial actin organization such that filaments are less available for filopodia formation in mutants. In contrast, UNC-34/Enabled had a filopodia-specific effect, consistent with the role of Ena/VASP molecules as actin anti-capping proteins. It is surprising that unc-34 had little or no effect on stability or length of the few filopodia that did form in these mutants. Possibly, the anti-capping activity of UNC-34/Enabled is required only during initiation and other proteins can serve its function during elongation. It is also possible that filopodial initiation requires other actin-interacting properties of Enabled that have been described recently (for example, filament nucleation or bundling) [21, 22].
Arp2/3, UNC-115/abLIM, and UNC-34/Enabled act together in axon guidance
Our results indicate that Arp2/3, UNC-115/abLIM, and UNC-34/Enabled are required for growth cone filopodia formation. However, the effects of these mutations on PDE and VD/DD axon guidance and PQR dendritic guidance were incompletely penetrant. Strikingly, double mutants of arx with unc-115 and unc-34 displayed greatly enhanced PDE axon guidance defects, and previous studies showed that unc-34 and unc-115 had overlapping roles in PDE axon guidance. Possibly, perturbing multiple actin modulating molecules had general, nonspecific effects on the cytoskeleton, resulting in increased defects. We think this is not the case as these interactions were specific. erm-1 encodes the only member of the ezrin/radixin/moesin family of actin-membrane linking proteins in the C. elegans genome. erm-1 mutation caused weak guidance defects on its own, indicating that ERM-1 is required for axon guidance. However, erm-1 did not enhance unc-115 or unc-34, suggesting that the enhancement of arx, unc-115, and unc-34 is specific to these genes and is not the result of general cytoskeletal disruption.
Together, our data indicate that Arp2/3, UNC-115/abLIM, and UNC34/Enabled might act in a common process in axon guidance, namely in the initiation of growth cone filopodia (Figure 11). Indeed, unc-115 and unc-34 enhanced the PQR filopodia defects of arx-7, although the enhancement by unc-34 might simply be an additive effect because unc-34 mutants alone display severely reduced filopodia numbers. It is also possible that all three molecules affect filopodia formation and have redundant roles in other aspects of growth cone morphology. Indeed, this seems to be the case, as both unc-115; arx and unc-34; arx double mutants displayed severe PQR neurite outgrowth and formation defects. This is surprising for unc-34/Enabled, as Ena/VASP molecules were traditionally considered to be anti-capping proteins with specific effects on filopodia. However, recent studies have suggested that Ena/VASP molecules have other effects on actin, including bundling and nucleation , have anti-capping-independent roles in filopodia formation  and control the initiation of neurites in the mammalian central nervous system [48, 49]. Possibly these other roles of UNC-34/Enabled explain the severe defects in neurite formation in unc-34; arx double mutants.
Many molecules have been implicated in axon guidance through genetic, molecular, and biochemical studies. The effects of these molecules on the behavior and dynamics of the growth cone during axon guidance are unclear in many cases. Through a combination of genetic analysis and time-lapse imaging of growth cones during outgrowth, we show that the Arp2/3 complex acts cell autonomously in the initiation of new growth cone filopodia during growth cone outgrowth, possibly explaining the axon guidance defects observed in arx mutants. We show that two other actin modulatory molecules, UNC-115/abLIM and UNC-34/Enabled, also affect growth cone filopodia initiation, which might explain the genetic redundancy of Arp2/3, UNC-115/abLIM, and UNC-34/Enabled in axon guidance. Arp2/3 and UNC-115/abLIM, but not UNC-34/Enabled, also affect the size of the growth cone, possibly reflecting a role in the formation of growth cone lamellipodium. Thus, multiple actin regulatory molecules have shared roles (filopodia initiation) and distinct roles (lamellipodia formation) in the growth cone during axon outgrowth.
green fluorescent protein
ventral nerve cord.
The authors wish to thank the Caenorhabditis Genetics Center, sponsored by the NIH National Institute of Research Resources, for C. elegans strains, the C. elegans Gene Knockout Consortium (Bob Barstead, Gary Moulder, Mark Edgley, Don Moerman) for arx and other mutations, E Struckhoff for technical assistance, Brian Ackley for critical reading of the manuscript and helpful discussions, and members of the Lundquist lab for helpful discussions. This work was supported by NIH grant NS40945 and NSF grant IOS93192 to EAL and NIH grant P20 RR016475 from the INBRE Program of the National Center for Research Resources (J Hunt, PI).
- Tessier-Lavigne M, Goodman CS: The molecular biology of axon guidance. Science. 1996, 274: 1123-1133. 10.1126/science.274.5290.1123.View ArticlePubMedGoogle Scholar
- Mortimer D, Fothergill T, Pujic Z, Richards LJ, Goodhill GJ: Growth cone chemotaxis. Trends Neurosci. 2008, 31: 90-98. 10.1016/j.tins.2007.11.008.View ArticlePubMedGoogle Scholar
- Pak CW, Flynn KC, Bamburg JR: Actin-binding proteins take the reins in growth cones. Nat Rev Neurosci. 2008, 9: 136-147. 10.1038/nrn2236.View ArticlePubMedGoogle Scholar
- Gallo G, Letourneau PC: Regulation of growth cone actin filaments by guidance cues. J Neurobiol. 2004, 58: 92-102. 10.1002/neu.10282.View ArticlePubMedGoogle Scholar
- Zhou FQ, Cohan CS: How actin filaments and microtubules steer growth cones to their targets. J Neurobiol. 2004, 58: 84-91. 10.1002/neu.10278.View ArticlePubMedGoogle Scholar
- Borisy GG, Svitkina TM: Actin machinery: pushing the envelope. Curr Opin Cell Biol. 2000, 12: 104-112. 10.1016/S0955-0674(99)00063-0.View ArticlePubMedGoogle Scholar
- Svitkina TM, Borisy GG: Arp2/3 complex and actin depolymerizing factor/cofilin in dendritic organization and treadmilling of actin filament array in lamellipodia. J Cell Biol. 1999, 145: 1009-1026. 10.1083/jcb.145.5.1009.PubMed CentralView ArticlePubMedGoogle Scholar
- Beltzner CC, Pollard TD: Pathway of actin filament branch formation by Arp2/3 complex. J Biol Chem. 2007, 283: 7135-7144. 10.1074/jbc.M705894200.View ArticlePubMedGoogle Scholar
- Zallen JA, Cohen Y, Hudson AM, Cooley L, Wieschaus E, Schejter ED: SCAR is a primary regulator of Arp2/3-dependent morphological events in Drosophila. J Cell Biol. 2002, 156: 689-701. 10.1083/jcb.200109057.PubMed CentralView ArticlePubMedGoogle Scholar
- Shakir MA, Jiang K, Struckhoff EC, Demarco R, Patel FB, Soto MC, Lundquist EA: The Arp2/3 activators WAVE and WASP have distinct genetic interactions with Rac GTPases in C. elegans axon guidance. Genetics. 2008, 179: 1-15. 10.1534/genetics.108.088963.View ArticleGoogle Scholar
- Strasser GA, Rahim NA, VanderWaal KE, Gertler FB, Lanier LM: Arp2/3 is a negative regulator of growth cone translocation. Neuron. 2004, 43: 81-94. 10.1016/j.neuron.2004.05.015.View ArticlePubMedGoogle Scholar
- Pinyol R, Haeckel A, Ritter A, Qualmann B, Kessels MM: Regulation of N-WASP and the Arp2/3 complex by Abp1 controls neuronal morphology. PLoS ONE. 2007, 2: e400-10.1371/journal.pone.0000400.PubMed CentralView ArticlePubMedGoogle Scholar
- Korobova F, Svitkina T: Arp2/3 complex is important for filopodia formation, growth cone motility, and neuritogenesis in neuronal cells. Mol Biol Cell. 2008, 19: 1561-1574. 10.1091/mbc.E07-09-0964.PubMed CentralView ArticlePubMedGoogle Scholar
- Lundquist EA, Herman RK, Shaw JE, Bargmann CI: UNC-115, a conserved protein with predicted LIM and actin-binding domains, mediates axon guidance in C. elegans. Neuron. 1998, 21: 385-392. 10.1016/S0896-6273(00)80547-4.View ArticlePubMedGoogle Scholar
- Struckhoff EC, Lundquist EA: The actin-binding protein UNC-115 is an effector of Rac signaling during axon pathfinding in C. elegans. Development. 2003, 130: 693-704. 10.1242/dev.00300.View ArticlePubMedGoogle Scholar
- Shakir MA, Gill JS, Lundquist EA: Interactions of UNC-34 Enabled with Rac GTPases and the NIK kinase MIG-15 in Caenorhabditis elegans axon pathfinding and neuronal migration. Genetics. 2006, 172: 893-913. 10.1534/genetics.105.046359.PubMed CentralView ArticlePubMedGoogle Scholar
- Garcia MC, Abbasi M, Singh S, He Q: Role of Drosophila gene dunc-115 in nervous system. Invert Neurosci. 2007, 7: 119-128. 10.1007/s10158-007-0047-1.View ArticlePubMedGoogle Scholar
- Withee J, Galligan B, Hawkins N, Garriga G: Caenorhabditis elegans WASP and Ena/VASP proteins play compensatory roles in morphogenesis and neuronal cell migration. Genetics. 2004, 167: 1165-1176. 10.1534/genetics.103.025676.PubMed CentralView ArticlePubMedGoogle Scholar
- Lebrand C, Dent EW, Strasser GA, Lanier LM, Krause M, Svitkina TM, Borisy GG, Gertler FB: Critical role of Ena/VASP proteins for filopodia formation in neurons and in function downstream of netrin-1. Neuron. 2004, 42: 37-49. 10.1016/S0896-6273(04)00108-4.View ArticlePubMedGoogle Scholar
- Barzik M, Kotova TI, Higgs HN, Hazelwood L, Hanein D, Gertler FB, Schafer DA: Ena/VASP proteins enhance actin polymerization in the presence of barbed end capping proteins. J Biol Chem. 2005, 280: 28653-28662. 10.1074/jbc.M503957200.PubMed CentralView ArticlePubMedGoogle Scholar
- Bear JE, Gertler FB: Ena/VASP: towards resolving a pointed controversy at the barbed end. J Cell Sci. 2009, 122: 1947-1953. 10.1242/jcs.038125.PubMed CentralView ArticlePubMedGoogle Scholar
- Applewhite DA, Barzik M, Kojima S, Svitkina TM, Gertler FB, Borisy GG: Ena/VASP proteins have an anti-capping independent function in filopodia formation. Mol Biol Cell. 2007, 18: 2579-2591. 10.1091/mbc.E06-11-0990.PubMed CentralView ArticlePubMedGoogle Scholar
- Yang Y, Lundquist EA: The actin-binding protein UNC-115/abLIM controls formation of lamellipodia and filopodia and neuronal morphogenesis in Caenorhabditis elegans. Mol Cell Biol. 2005, 25: 5158-5170. 10.1128/MCB.25.12.5158-5170.2005.PubMed CentralView ArticlePubMedGoogle Scholar
- Brenner S: The genetics of Caenorhabditis elegans. Genetics. 1974, 77: 71-94.PubMed CentralPubMedGoogle Scholar
- Mello C, Fire A: DNA transformation. Methods Cell Biol. 1995, 48: 451-482. full_text.View ArticlePubMedGoogle Scholar
- Reddien PW, Horvitz HR: CED-2/CrkII and CED-10/Rac control phagocytosis and cell migration in Caenorhabditis elegans. Nat Cell Biol. 2000, 2: 131-136. 10.1038/35004000.View ArticlePubMedGoogle Scholar
- Zipkin ID, Kindt RM, Kenyon CJ: Role of a new Rho family member in cell migration and axon guidance in C. elegans. Cell. 1997, 90: 883-894. 10.1016/S0092-8674(00)80353-0.View ArticlePubMedGoogle Scholar
- Collet J, Spike CA, Lundquist EA, Shaw JE, Herman RK: Analysis of osm-6, a gene that affects sensory cilium structure and sensory neuron function in Caenorhabditis elegans. Genetics. 1998, 148: 187-200.PubMed CentralPubMedGoogle Scholar
- Yu S, Avery L, Baude E, Garbers DL: Guanylyl cyclase expression in specific sensory neurons: a new family of chemosensory receptors. Proc Natl Acad Sci USA. 1997, 94: 3384-3387. 10.1073/pnas.94.7.3384.PubMed CentralView ArticlePubMedGoogle Scholar
- Jin Y, Jorgensen E, Hartwieg E, Horvitz HR: The Caenorhabditis elegans gene unc-25 encodes glutamic acid decarboxylase and is required for synaptic transmission but not synaptic development. J Neurosci. 1999, 19: 539-548.PubMedGoogle Scholar
- Weinkove D, Bastiani M, Chessa TA, Joshi D, Hauth L, Cooke FT, Divecha N, Schuske K: Overexpression of PPK-1, the Caenorhabditis elegans type I PIP kinase, inhibits growth cone collapse in the developing nervous system and causes axonal degeneration in adults. Dev Biol. 2008, 313: 384-397. 10.1016/j.ydbio.2007.10.029.PubMed CentralView ArticlePubMedGoogle Scholar
- Sawa M, Suetsugu S, Sugimoto A, Miki H, Yamamoto M, Takenawa T: Essential role of the C. elegans Arp2/3 complex in cell migration during ventral enclosure. J Cell Sci. 2003, 116: 1505-1518. 10.1242/jcs.00362.View ArticlePubMedGoogle Scholar
- White JG, Southgate E, Thomson JN, Brenner S: The structure of the nervous system of the nematode Caenorhabditis elegans. Philos Trans R Soc Lond. 1986, 314: 1-340. 10.1098/rstb.1986.0056.View ArticleGoogle Scholar
- Lundquist EA, Reddien PW, Hartwieg E, Horvitz HR, Bargmann CI: Three C. elegans Rac proteins and several alternative Rac regulators control axon guidance, cell migration and apoptotic cell phagocytosis. Development. 2001, 128: 4475-4488.PubMedGoogle Scholar
- Gobel V, Barrett PL, Hall DH, Fleming JT: Lumen morphogenesis in C. elegans requires the membrane-cytoskeleton linker erm-1. Dev Cell. 2004, 6: 865-873. 10.1016/j.devcel.2004.05.018.View ArticlePubMedGoogle Scholar
- Van Furden D, Johnson K, Segbert C, Bossinger O: The C. elegans ezrin-radixin-moesin protein ERM-1 is necessary for apical junction remodelling and tubulogenesis in the intestine. Dev Biol. 2004, 272: 262-276. 10.1016/j.ydbio.2004.05.012.View ArticlePubMedGoogle Scholar
- Duerr JS, Frisby DL, Gaskin J, Duke A, Asermely K, Huddleston D, Eiden LE, Rand JB: The cat-1 gene of Caenorhabditis elegans encodes a vesicular monoamine transporter required for specific monoamine-dependent behaviors. J Neurosci. 1999, 19: 72-84.PubMedGoogle Scholar
- Sulston JE, Horvitz HR: Post-embryonic cell lineages of the nematode, Caenorhabditis elegans. Dev Biol. 1977, 56: 110-156. 10.1016/0012-1606(77)90158-0.View ArticlePubMedGoogle Scholar
- Goldberg DJ, Burmeister DW: Stages in axon formation: observations of growth of Aplysia axons in culture using video-enhanced contrast-differential interference contrast microscopy. J Cell Biol. 1986, 103: 1921-1931. 10.1083/jcb.103.5.1921.View ArticlePubMedGoogle Scholar
- Chang C, Adler CE, Krause M, Clark SG, Gertler FB, Tessier-Lavigne M, Bargmann CI: MIG-10/lamellipodin and AGE-1/PI3K promote axon guidance and outgrowth in response to slit and netrin. Curr Biol. 2006, 16: 854-862. 10.1016/j.cub.2006.03.083.View ArticlePubMedGoogle Scholar
- Drees F, Gertler FB: Ena/VASP: proteins at the tip of the nervous system. Curr Opin Neurobiol. 2008, 18: 53-59. 10.1016/j.conb.2008.05.007.PubMed CentralView ArticlePubMedGoogle Scholar
- Knobel KM, Jorgensen EM, Bastiani MJ: Growth cones stall and collapse during axon outgrowth in Caenorhabditis elegans. Development. 1999, 126: 4489-4498.PubMedGoogle Scholar
- Davenport RW, Dou P, Rehder V, Kater SB: A sensory role for neuronal growth cone filopodia. Nature. 1993, 361: 721-724. 10.1038/361721a0.View ArticlePubMedGoogle Scholar
- Dent EW, Tang F, Kalil K: Axon guidance by growth cones and branches: common cytoskeletal and signaling mechanisms. Neuroscientist. 2003, 9: 343-353. 10.1177/1073858403252683.View ArticlePubMedGoogle Scholar
- Eden S, Rohatgi R, Podtelejnikov AV, Mann M, Kirschner MW: Mechanism of regulation of WAVE1-induced actin nucleation by Rac1 and Nck. Nature. 2002, 418: 790-793. 10.1038/nature00859.View ArticlePubMedGoogle Scholar
- Innocenti M, Zucconi A, Disanza A, Frittoli E, Areces LB, Steffen A, Stradal TE, Di Fiore PP, Carlier MF, Scita G: Abi1 is essential for the formation and activation of a WAVE2 signalling complex. Nat Cell Biol. 2004, 6: 319-327. 10.1038/ncb1105.View ArticlePubMedGoogle Scholar
- Friederich E, Vancompernolle K, Huet C, Goethals M, Finidori J, Vandekerckhove J, Louvard D: An actin-binding site containing a conserved motif of charged amino acid residues is essential for the morphogenic effect of villin. Cell. 1992, 70: 81-92. 10.1016/0092-8674(92)90535-K.View ArticlePubMedGoogle Scholar
- Dent EW, Kwiatkowski AV, Mebane LM, Philippar U, Barzik M, Rubinson DA, Gupton S, Van Veen JE, Furman C, Zhang J, Alberts AS, Mori S, Gertler FB: Filopodia are required for cortical neurite initiation. Nat Cell Biol. 2007, 9: 1347-1359. 10.1038/ncb1654.View ArticlePubMedGoogle Scholar
- Kwiatkowski AV, Rubinson DA, Dent EW, Edward van Veen J, Leslie JD, Zhang J, Mebane LM, Philippar U, Pinheiro EM, Burds AA, Bronson RT, Mori S, Fässler R, Gertler FB: Ena/VASP is required for neuritogenesis in the developing cortex. Neuron. 2007, 56: 441-455. 10.1016/j.neuron.2007.09.008.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.