News & Views | Published:

Cell biology

Barbed ends rule

Nature volume 430, pages 734735 (12 August 2004) | Download Citation


To explore their surroundings, cells use probes of various shapes. Whether the probes are broad and flat, or long and thin, seems to be regulated by proteins at the growing ends of actin filaments.

Cells reach out with various structures as they crawl and interact with their environment. Broad, flat protrusions called lamellipodia surge forward and adhere to surfaces, allowing cells to gain traction and move. Long, thin protrusions called filopodia extend several micrometres ahead of the cells, appearing to explore extracellular surfaces, sense guiding cues and direct the rest of the cell. Extension of both lamellipodia and filopodia is powered by polymerization of the structural protein actin into filaments. As the filaments lengthen, the fast-growing ‘barbed’ ends push the cell's boundary membrane outward. Lamellipodia protrude when the actin filaments at the leading edge are short and highly branched; by contrast, filopodia arise when the filaments are long, unbranched, and arranged in tight, parallel bundles. But given the similarities in the actin polymerization machinery that drives the two types of protrusion, what determines whether lamellipodia or filopodia form? Marisan Mejillano and colleagues, writing in Cell1, report that the answer lies with proteins that control actin polymerization at the filament barbed ends.

Two models have been proposed to explain the formation of these actin-based protrusions — one dealing with lamellipodia and one with filopodia (Fig. 1). The dendritic-nucleation model2 suggests that lamellipodia arise when new actin filaments form as branches on existing actin filaments, under the direction of a protein complex called Arp2/3. The filaments push the membrane outward as they elongate by polymerization, until the growing ends are ‘capped’. Capping protein binds tightly to the barbed ends, blocking filament growth3. The result is a lamellipodial actin network of short, stiff, branched filaments. According to the convergent-elongation model4 of filopodium development, a ‘tip complex’ of proteins recruits filaments of the lamellipodial network to become the filopodium. The tip complex includes Ena/VASP proteins, which encourage the growth of long, unbranched filaments by inhibiting the capping process5. Ena/VASP proteins might also recruit other proteins to further stabilize and organize the actin filaments into bundles as the filopodia lengthen. Both models suggest that whether or not capping occurs at the growing end of actin filaments is key to which protrusive structure forms.

Figure 1: Model for how the growing ends of actin filaments regulate cellular protrusions.
Figure 1

Actin filaments polymerize at their fast-growing ‘barbed’ ends near the plasma membrane. They stop growing when they are capped by capping protein. If barbed-end capping activity is high, filaments get capped, stop growing and remain short. This favours an actin filament network that is most efficient for lamellipodium expansion. Filopodia result when filaments of the lamellipodial actin network are recruited by a filopodial tip complex that includes Ena/VASP proteins; filaments grow long because Ena/VASP proteins also inhibit capping locally.

So Mejillano and colleagues1 studied the effects of reducing the amount of capping protein on the formation of protrusions, using cells that normally form both types. They found that depleting the cells of capping protein dramatically changes both lamellipodia and filopodia. Most striking was a massive induction of filopodium-like protrusions all over the cell surface. The protrusions were bona fide filopodia: they exhibited the motile behaviour of filopodia, contained parallel bundles of long actin filaments, and recruited two molecular hallmarks of filopodia, Ena/VASP proteins and fascin, a protein that helps to bundle the actin filaments.

In contrast to the explosion of filopodia, the normally smooth, coordinated lamellipodial protrusions became slow and erratic following the reduction in levels of capping protein. This behaviour would be expected if the cellular pool of actin monomers is used up in fuelling the rampant polymerization of filopodial actin filaments, rather than contributing to the lamellipodia. In addition, lamellipodial filamentous actin and Arp2/3 complex, which are normally enriched at the cell edge, were lost. The cytoplasm further away from the cell periphery became dense with actin filaments and Arp2/3 complex, suggesting that the machinery that disassembles actin filaments away from the cell edge is disrupted when barbed-end capping activity is low.

So, lowering barbed-end capping activity seems to tweak the actin machinery at the cell's leading edge towards polymerization of actin filaments in filopodia. Densely packed actin filaments that are inefficient at pushing the membrane forward thus replace the lamellipodial actin network. When capping protein was reinstated in the depleted cells, excess filopodia disappeared and lamellipodia were restored, indicating that capping protein is entirely responsible for the effect of barbed-end capping activity on cellular protrusions.

But why do filopodia, and not lamellipodia, grow explosively when barbed ends remain uncapped? Mejillano et al.1 discovered that Ena/VASP proteins also influence which type of protrusion the cell extends. Because Ena/VASP proteins inhibit barbed-end capping activity in vitro5, the team predicted that reducing capping protein in cells lacking these capping antagonists would further enhance filopodia at the expense of lamellipodia. But when they tried this experiment, no filopodia formed; instead, ruffles were produced along the cell edge. Most probably, these result when actin filaments at the membrane become so long and flexible that they buckle and collapse rearward as they fail to sustain a pushing force. So loss of barbed-end capping activity clearly enhances actin polymerization at the cell periphery, but is not sufficient to specify whether filopodia or lamellipodia protrude when Ena/VASP proteins are absent. However, when the Ena/VASP protein known as Mena was restored in previously depleted cells, the formation of abundant filopodia returned, highlighting a special role for Ena/VASP proteins in initiating filopodia, even if barbed-end capping activity is low.

Mejillano and colleagues' findings provide evidence that supports key features of the two models of actin-filament-based protrusion. First, as proposed in the dendritic-nucleation hypothesis2, regulated barbed-end capping is essential to maintain productive lamellipodia. Lamellipodia extend persistently when barbed ends are capped because the unpolymerized actin pool is maintained, and actin filaments of the network remain short and, hence, sufficiently stiff to push the membrane forward. Second, as predicted by the convergent-elongation hypothesis4, the formation of filopodia requires Ena/VASP proteins. These proteins seem to be the glue of the filopodial tip complex that recruits lamellipodial actin filaments and promotes their growth to form nascent filopodia.

So, when it comes to actin-filament-based protrusions, regulated polymerization at filament barbed ends is essential, and both capping protein and Ena/VASP proteins contribute by regulating barbed-end capping activity. Ena/VASP proteins are also essential for a functional filopodial tip complex. With Mejillano and colleagues' findings, therefore, we have a clearer picture of how barbed ends point the way.


  1. 1.

    et al. Cell 118, 363–373 (2004).

  2. 2.

    , & Annu. Rev. Biophys. Biomol. Struct. 29, 545–576 (2000).

  3. 3.

    & Curr. Opin. Cell Biol. 12, 97–103 (2000).

  4. 4.

    et al. J. Cell Biol. 160, 409–421 (2003).

  5. 5.

    et al. Cell 109, 509–521 (2002).

Download references

Author information


  1. Dorothy A. Schafer is in the Departments of Biology and Cell Biology, University of Virginia, Charlottesville, Virginia 22903, USA.

    • Dorothy A. Schafer


  1. Search for Dorothy A. Schafer in:

About this article

Publication history



Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing