CLASP2 facilitates dynamic actin filament organization along the microtubule lattice
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (Arcadia Science)
Abstract
CLASP2α directly cross-links multiple actin filaments to the microtubule lattice. CLASP2α-mediated cross-linking facilitates dynamic organization of actin filaments templated by microtubules. Furthermore, CLASP depletion results in disorganized actin fibers and reduced coalignment of actin fibers with microtubules, suggesting that microtubule–CLASP–actin interaction may influence overall cytoskeletal organization in cells.
Article activity feed
-
Taken together, we conclude that CLASP2α directly crosslinks F-actin to microtubules.
If you add purified CLASP2a-GFP to Phalloidin-stabilized F-actin, will the CLASP bind & coat F-actin? and if so, could you then add the MTs and see if they coat the stable F-actin filament? Basically, does this go both ways? or does it need to be MT -> CLASP2a -> Actin?
-
Interestingly, co-alignment of actin fibers with microtubules in CLASP-depleted cells (Figure 4F,G) was reduced as compared to controls (Figure 4E), suggesting a diminished coordination of these filaments, consistent with our in vitro result
It might be cool to see the rearrangement of the MTs in a single channel (or w/ DAPI) like Actin is shown in A-C. It looks like the Tubulin in F-G is whacky compared to the control, but its hard to tell if this is true with the actin overlay. From the whole cell view, it looks like MTs are localized more centrally in the knockdowns, compared to the cortical localization in the Control, but i didn't see any mention of this in the text (but I might have missed it and I'm not up to date on the CLASP-literature so it might already be known).
-
Furthermore, we found that F-actin also robustly co-localized with microtubules in the presence of GFP-L-TOG2-S (PCC = 0.88 ± 0.04, SD, N = 9 FOVs over 3 independent experimental days) (Figure 1B,D).
Very cool!!!! Strangely, it looks like the signal in the MT and F-actin channels are comparable between Full Length and truncated CLASP2as, but it looks like there is less GFP signal in the truncated version, compared to full-length. This can be seen with the increased blue signal in the merged channel. While the colocalization quantifications are awesome and needed, it could be nice to have some intensity quantifications / linescans as well, like you did for F-actin signal in Figure 2
This is weird, right? Is truncated CLASP2a better at binding F-actin (and maybe worse at binding MTs) than FL? Even though you cite a paper saying that …
Furthermore, we found that F-actin also robustly co-localized with microtubules in the presence of GFP-L-TOG2-S (PCC = 0.88 ± 0.04, SD, N = 9 FOVs over 3 independent experimental days) (Figure 1B,D).
Very cool!!!! Strangely, it looks like the signal in the MT and F-actin channels are comparable between Full Length and truncated CLASP2as, but it looks like there is less GFP signal in the truncated version, compared to full-length. This can be seen with the increased blue signal in the merged channel. While the colocalization quantifications are awesome and needed, it could be nice to have some intensity quantifications / linescans as well, like you did for F-actin signal in Figure 2
This is weird, right? Is truncated CLASP2a better at binding F-actin (and maybe worse at binding MTs) than FL? Even though you cite a paper saying that TOG1 (which is deleted in this construct) also binds to F-actin?
-
However, using a low-speed co-sedimentation approach, we did not find any evidence that CLASP2α can bundle F-actin (Supplemental Figure S2B-D).
Is it possible that microtubules are required for CLASP2a's F-actin bundling abilities? You show it can bind F-actin in the absence of MTs (Sup Fig 1) but its binding & function could be different in the presence of MTs
-
silanized
Just curious, is silanization essential for these assays to work? I recently saw some literature that silica nanoparticles can mess up actin dynamics / localization (https://doi.org/10.1007/s00204-020-02694-6 ; doi:10.1021/nn103344k). I know that basically all actin dynamics have been measured in the presence of glass/silica, and the 30 min on silanized glass likely wouldn't impact anything substantially, and there really aren't any good alternatives to silica-based coverslips, so definitely not expecting anything else, but I was just curious what the extra silanization does?
-
Video 2
This is amazing!!! I'd love to see if this is happening with the truncated CLASP2a as well! Since the 2 actin binding TOGs are right next to each other, do you think both of those are both binding actin monomers helping to form stable dimers? helping to initiate non-spontaneous actin nucleation? Maybe even TOG3 binds actin?
-
-
silanized
Just curious, is silanization essential for these assays to work? I recently saw some literature that silica nanoparticles can mess up actin dynamics / localization (https://doi.org/10.1007/s00204-020-02694-6 ; doi:10.1021/nn103344k). I know that basically all actin dynamics have been measured in the presence of glass/silica, and the 30 min on silanized glass likely wouldn't impact anything substantially, and there really aren't any good alternatives to silica-based coverslips, so definitely not expecting anything else, but I was just curious what the extra silanization does?
-
Furthermore, we found that F-actin also robustly co-localized with microtubules in the presence of GFP-L-TOG2-S (PCC = 0.88 ± 0.04, SD, N = 9 FOVs over 3 independent experimental days) (Figure 1B,D).
Very cool!!!! Strangely, it looks like the signal in the MT and F-actin channels are comparable between Full Length and truncated CLASP2as, but it looks like there is less GFP signal in the truncated version, compared to full-length. This can be seen with the increased blue signal in the merged channel. While the colocalization quantifications are awesome and needed, it could be nice to have some intensity quantifications / linescans as well, like you did for F-actin signal in Figure 2
This is weird, right? Is truncated CLASP2a better at binding F-actin (and maybe worse at binding MTs) than FL? Even though you cite a paper saying that …
Furthermore, we found that F-actin also robustly co-localized with microtubules in the presence of GFP-L-TOG2-S (PCC = 0.88 ± 0.04, SD, N = 9 FOVs over 3 independent experimental days) (Figure 1B,D).
Very cool!!!! Strangely, it looks like the signal in the MT and F-actin channels are comparable between Full Length and truncated CLASP2as, but it looks like there is less GFP signal in the truncated version, compared to full-length. This can be seen with the increased blue signal in the merged channel. While the colocalization quantifications are awesome and needed, it could be nice to have some intensity quantifications / linescans as well, like you did for F-actin signal in Figure 2
This is weird, right? Is truncated CLASP2a better at binding F-actin (and maybe worse at binding MTs) than FL? Even though you cite a paper saying that TOG1 (which is deleted in this construct) also binds to F-actin?
-
Taken together, we conclude that CLASP2α directly crosslinks F-actin to microtubules.
If you add purified CLASP2a-GFP to Phalloidin-stabilized F-actin, will the CLASP bind & coat F-actin? and if so, could you then add the MTs and see if they coat the stable F-actin filament? Basically, does this go both ways? or does it need to be MT -> CLASP2a -> Actin?
-
Interestingly, co-alignment of actin fibers with microtubules in CLASP-depleted cells (Figure 4F,G) was reduced as compared to controls (Figure 4E), suggesting a diminished coordination of these filaments, consistent with our in vitro result
It might be cool to see the rearrangement of the MTs in a single channel (or w/ DAPI) like Actin is shown in A-C. It looks like the Tubulin in F-G is whacky compared to the control, but its hard to tell if this is true with the actin overlay. From the whole cell view, it looks like MTs are localized more centrally in the knockdowns, compared to the cortical localization in the Control, but i didn't see any mention of this in the text (but I might have missed it and I'm not up to date on the CLASP-literature so it might already be known).
-
Video 2
This is amazing!!! I'd love to see if this is happening with the truncated CLASP2a as well! Since the 2 actin binding TOGs are right next to each other, do you think both of those are both binding actin monomers helping to form stable dimers? helping to initiate non-spontaneous actin nucleation? Maybe even TOG3 binds actin?
-
However, using a low-speed co-sedimentation approach, we did not find any evidence that CLASP2α can bundle F-actin (Supplemental Figure S2B-D).
Is it possible that microtubules are required for CLASP2a's F-actin bundling abilities? You show it can bind F-actin in the absence of MTs (Sup Fig 1) but its binding & function could be different in the presence of MTs
-
