Limits...
Formin and capping protein together embrace the actin filament in a ménage à trois.

Shekhar S, Kerleau M, Kühn S, Pernier J, Romet-Lemonne G, Jégou A, Carlier MF - Nat Commun (2015)

Bottom Line: This is further confirmed using single-molecule imaging.We show that formin FMNL2 behaves similarly, thus suggesting that this is a general property of formins.Implications in filopodia regulation and barbed-end structural regulation are discussed.

View Article: PubMed Central - PubMed

Affiliation: Cytoskeleton Dynamics and Cell Motility, Department of Biochemistry, Biophysics and Structural Biology, I2BC, CNRS, 91198 Gif-sur-Yvette, France.

ABSTRACT
Proteins targeting actin filament barbed ends play a pivotal role in motile processes. While formins enhance filament assembly, capping protein (CP) blocks polymerization. On their own, they both bind barbed ends with high affinity and very slow dissociation. Their barbed-end binding is thought to be mutually exclusive. CP has recently been shown to be present in filopodia and controls their morphology and dynamics. Here we explore how CP and formins may functionally coregulate filament barbed-end assembly. We show, using kinetic analysis of individual filaments by microfluidics-assisted fluorescence microscopy, that CP and mDia1 formin are able to simultaneously bind barbed ends. This is further confirmed using single-molecule imaging. Their mutually weakened binding enables rapid displacement of one by the other. We show that formin FMNL2 behaves similarly, thus suggesting that this is a general property of formins. Implications in filopodia regulation and barbed-end structural regulation are discussed.

Show MeSH

Related in: MedlinePlus

Formin FMNL2 also binds to CP-capped filaments and uncaps via a transient BCF state.(a) Kymograph of a capped filament undergoing uncapping and fast processive growth on exposure to formin FMNL2. Filaments elongating from spectrin–actin seeds (set-up #1, Fig. 1a) were first exposed to 20 nM CP and PA for a couple of minutes (B+C→BC). Paused filaments were then exposed to 250 nM FMNL2 for 30 s (BC+F→BCF). Once formin was removed from the flow and PA was introduced, fast elongation was observed (BCF→BF+C). Note that, as expected, the FMNL2 elongation rate seen here is much slower compared with that of formin mDia1 as seen in Fig. 3b. (b) Fraction of CP-capped paused filaments that resume rapid elongation (BF state) during exposure to PA only, versus time, from an initial population of capped filaments exposed to 250 nM (black, n=56 filaments), 500 nM (red, n=55 filaments) and 750 nM (blue, n=43 filaments) FMNL2 for 30 s. Symbols represent the experimental data and the solid lines are the exponential fits. Only three representative CDFs are shown here for the ease of reading, see Supplementary Fig. 10 for details. (c) The maximum fraction of BF filaments (plateau values of curves such as the ones shown in) as a function the product of formin FMNL2 concentration [F] and exposure duration (Texpo). The solid line is an exponential fit corresponding to equation (3). Inset: kobs=k′−C+k′−F is independent of the experimental condition. Horizontal line represents the average (Error bars: s.e.m.).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4660058&req=5

f4: Formin FMNL2 also binds to CP-capped filaments and uncaps via a transient BCF state.(a) Kymograph of a capped filament undergoing uncapping and fast processive growth on exposure to formin FMNL2. Filaments elongating from spectrin–actin seeds (set-up #1, Fig. 1a) were first exposed to 20 nM CP and PA for a couple of minutes (B+C→BC). Paused filaments were then exposed to 250 nM FMNL2 for 30 s (BC+F→BCF). Once formin was removed from the flow and PA was introduced, fast elongation was observed (BCF→BF+C). Note that, as expected, the FMNL2 elongation rate seen here is much slower compared with that of formin mDia1 as seen in Fig. 3b. (b) Fraction of CP-capped paused filaments that resume rapid elongation (BF state) during exposure to PA only, versus time, from an initial population of capped filaments exposed to 250 nM (black, n=56 filaments), 500 nM (red, n=55 filaments) and 750 nM (blue, n=43 filaments) FMNL2 for 30 s. Symbols represent the experimental data and the solid lines are the exponential fits. Only three representative CDFs are shown here for the ease of reading, see Supplementary Fig. 10 for details. (c) The maximum fraction of BF filaments (plateau values of curves such as the ones shown in) as a function the product of formin FMNL2 concentration [F] and exposure duration (Texpo). The solid line is an exponential fit corresponding to equation (3). Inset: kobs=k′−C+k′−F is independent of the experimental condition. Horizontal line represents the average (Error bars: s.e.m.).

Mentions: Formation of the ternary BFC complex with the FMNL2 formin was tested using set-up #1. Fast processive growth of FMNL2-bound barbed ends was rapidly arrested on exposure to CP. However, no fast elongation was seen to resume when CP was removed from the flow, possibly due to low affinity of FMNL2 for the barbed ends causing very fast dissociation of FMNL2 on CP binding. We therefore tested the formation of the ternary BCF complex starting from capped BC filaments. Filaments growing from spectrin–actin seeds were capped by CP, and subsequently exposed to varying concentration of FMNL2 for different exposure times. The rate constant for formin FMNL2 association to a capped barbed end (k′+FMNL2) was calculated as described for formin mDia1 in Fig. 3c,d. A typical kymograph of a CP-capped paused filament switching to formin-based fast elongation on exposure to 250 nM FMNL2 is illustrated in Fig. 4a. As observed with mDia1, the capped filament, following 30 s exposure to FMNL2, started fast processive elongation only after being exposed to PA for 45 s. This result indicates that at the end of the 30 s incubation of the capped filament with FMNL2, the barbed end was in the BCF and not in the BF state. FMNL2 behaves like mDia1 in Fig. 3b. Thus the data rule out a simple competition scheme between CP and FMNL2. Expectedly, the number of BC filaments that switch to BF filaments changes as the formin concentration is varied at constant exposure time (Fig. 4b), or as both are varied (Supplementary Fig. 10 and Fig. 4c). The derived association rate constant of FMNL2 to BC filaments is k′+FMNL2=0.115 μM−1·s−1, 13 times lower than free barbed ends (k+FMNL2=1.54 μM−1·s−1, Supplementary Fig. 11). As expected, the value of kobs=k′−C+k′−FMNL2 was found to be independent of the experimental conditions (kobs=0.01407±0.00153, s−1). Using equations (1, 2, 3) (see Methods section) we calculate k′−C=4.7±0.51 × 10−3 s−1and k′−FMNL2=9.4±1.0 × 10−3 s−1 (NBF/B0=0.334±0.026). Note that CP dissociates at very similar rate from BCF state for both mDia1 and FMNL2.


Formin and capping protein together embrace the actin filament in a ménage à trois.

Shekhar S, Kerleau M, Kühn S, Pernier J, Romet-Lemonne G, Jégou A, Carlier MF - Nat Commun (2015)

Formin FMNL2 also binds to CP-capped filaments and uncaps via a transient BCF state.(a) Kymograph of a capped filament undergoing uncapping and fast processive growth on exposure to formin FMNL2. Filaments elongating from spectrin–actin seeds (set-up #1, Fig. 1a) were first exposed to 20 nM CP and PA for a couple of minutes (B+C→BC). Paused filaments were then exposed to 250 nM FMNL2 for 30 s (BC+F→BCF). Once formin was removed from the flow and PA was introduced, fast elongation was observed (BCF→BF+C). Note that, as expected, the FMNL2 elongation rate seen here is much slower compared with that of formin mDia1 as seen in Fig. 3b. (b) Fraction of CP-capped paused filaments that resume rapid elongation (BF state) during exposure to PA only, versus time, from an initial population of capped filaments exposed to 250 nM (black, n=56 filaments), 500 nM (red, n=55 filaments) and 750 nM (blue, n=43 filaments) FMNL2 for 30 s. Symbols represent the experimental data and the solid lines are the exponential fits. Only three representative CDFs are shown here for the ease of reading, see Supplementary Fig. 10 for details. (c) The maximum fraction of BF filaments (plateau values of curves such as the ones shown in) as a function the product of formin FMNL2 concentration [F] and exposure duration (Texpo). The solid line is an exponential fit corresponding to equation (3). Inset: kobs=k′−C+k′−F is independent of the experimental condition. Horizontal line represents the average (Error bars: s.e.m.).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4660058&req=5

f4: Formin FMNL2 also binds to CP-capped filaments and uncaps via a transient BCF state.(a) Kymograph of a capped filament undergoing uncapping and fast processive growth on exposure to formin FMNL2. Filaments elongating from spectrin–actin seeds (set-up #1, Fig. 1a) were first exposed to 20 nM CP and PA for a couple of minutes (B+C→BC). Paused filaments were then exposed to 250 nM FMNL2 for 30 s (BC+F→BCF). Once formin was removed from the flow and PA was introduced, fast elongation was observed (BCF→BF+C). Note that, as expected, the FMNL2 elongation rate seen here is much slower compared with that of formin mDia1 as seen in Fig. 3b. (b) Fraction of CP-capped paused filaments that resume rapid elongation (BF state) during exposure to PA only, versus time, from an initial population of capped filaments exposed to 250 nM (black, n=56 filaments), 500 nM (red, n=55 filaments) and 750 nM (blue, n=43 filaments) FMNL2 for 30 s. Symbols represent the experimental data and the solid lines are the exponential fits. Only three representative CDFs are shown here for the ease of reading, see Supplementary Fig. 10 for details. (c) The maximum fraction of BF filaments (plateau values of curves such as the ones shown in) as a function the product of formin FMNL2 concentration [F] and exposure duration (Texpo). The solid line is an exponential fit corresponding to equation (3). Inset: kobs=k′−C+k′−F is independent of the experimental condition. Horizontal line represents the average (Error bars: s.e.m.).
Mentions: Formation of the ternary BFC complex with the FMNL2 formin was tested using set-up #1. Fast processive growth of FMNL2-bound barbed ends was rapidly arrested on exposure to CP. However, no fast elongation was seen to resume when CP was removed from the flow, possibly due to low affinity of FMNL2 for the barbed ends causing very fast dissociation of FMNL2 on CP binding. We therefore tested the formation of the ternary BCF complex starting from capped BC filaments. Filaments growing from spectrin–actin seeds were capped by CP, and subsequently exposed to varying concentration of FMNL2 for different exposure times. The rate constant for formin FMNL2 association to a capped barbed end (k′+FMNL2) was calculated as described for formin mDia1 in Fig. 3c,d. A typical kymograph of a CP-capped paused filament switching to formin-based fast elongation on exposure to 250 nM FMNL2 is illustrated in Fig. 4a. As observed with mDia1, the capped filament, following 30 s exposure to FMNL2, started fast processive elongation only after being exposed to PA for 45 s. This result indicates that at the end of the 30 s incubation of the capped filament with FMNL2, the barbed end was in the BCF and not in the BF state. FMNL2 behaves like mDia1 in Fig. 3b. Thus the data rule out a simple competition scheme between CP and FMNL2. Expectedly, the number of BC filaments that switch to BF filaments changes as the formin concentration is varied at constant exposure time (Fig. 4b), or as both are varied (Supplementary Fig. 10 and Fig. 4c). The derived association rate constant of FMNL2 to BC filaments is k′+FMNL2=0.115 μM−1·s−1, 13 times lower than free barbed ends (k+FMNL2=1.54 μM−1·s−1, Supplementary Fig. 11). As expected, the value of kobs=k′−C+k′−FMNL2 was found to be independent of the experimental conditions (kobs=0.01407±0.00153, s−1). Using equations (1, 2, 3) (see Methods section) we calculate k′−C=4.7±0.51 × 10−3 s−1and k′−FMNL2=9.4±1.0 × 10−3 s−1 (NBF/B0=0.334±0.026). Note that CP dissociates at very similar rate from BCF state for both mDia1 and FMNL2.

Bottom Line: This is further confirmed using single-molecule imaging.We show that formin FMNL2 behaves similarly, thus suggesting that this is a general property of formins.Implications in filopodia regulation and barbed-end structural regulation are discussed.

View Article: PubMed Central - PubMed

Affiliation: Cytoskeleton Dynamics and Cell Motility, Department of Biochemistry, Biophysics and Structural Biology, I2BC, CNRS, 91198 Gif-sur-Yvette, France.

ABSTRACT
Proteins targeting actin filament barbed ends play a pivotal role in motile processes. While formins enhance filament assembly, capping protein (CP) blocks polymerization. On their own, they both bind barbed ends with high affinity and very slow dissociation. Their barbed-end binding is thought to be mutually exclusive. CP has recently been shown to be present in filopodia and controls their morphology and dynamics. Here we explore how CP and formins may functionally coregulate filament barbed-end assembly. We show, using kinetic analysis of individual filaments by microfluidics-assisted fluorescence microscopy, that CP and mDia1 formin are able to simultaneously bind barbed ends. This is further confirmed using single-molecule imaging. Their mutually weakened binding enables rapid displacement of one by the other. We show that formin FMNL2 behaves similarly, thus suggesting that this is a general property of formins. Implications in filopodia regulation and barbed-end structural regulation are discussed.

Show MeSH
Related in: MedlinePlus