There are three mammalian variants of dynamin that share 80% sequence identity. They differ in their expression (Schmid and Frolov, 2011; Ferguson and De Camilli, 2012): dynamin-1 is specific to neurons, dynamin-2 is ubiquitously expressed, dynamin-3 is located in the brain, lungs and testis (Vaid et al., 2007). Dynamins are constituted of 5 domains: the GTPase (G) domain, the stalk – a rigid coiled-coil, which mediates assembly into helical polymers through 3 structurally conserved interfaces (Faelber et al., 2011; Ford et al., 2011; Reubold et al., 2015), the Bundle Signaling Element (BSE), which connects the GTPase domain to the stalk (Chappie et al., 2010; Chappie et al., 2011), a Pleckstrin Homology (PH) domain, which binds to phosphatidylinositol 4,5-bisphosphate (PI(4,5)P₂) (Salim et al., 1996) and the proline-rich domain (PRD), which binds partners such as Bin Amphiphysin Rvs (BAR)-domain proteins (Raimondi et al., 2011).

Dynamin can assemble into a helical coil around membrane necks and tubes. The constriction of the coil is required for fission and the most constricted state of dynamin is observed in presence of GTP (Antonny et al., 2016). However, the use of GTP energy is still debated: either it is used in helix constriction through twisting, or in disassembly of the constricted polymer (Antonny et al., 2016). When GTP is added to pre-assembled dynamin tubes in vitro, rapid torsion and constriction are observed (Roux et al., 2006). When GTP is added along with dynamin to membranes in vitro, highly constricted helices are observed, suggesting that GTP-bound dynamin assembles directly into a highly constricted state (Sundborger et al., 2014). However, the K44A used in this study has residual GTPase activity, and thus the highly constricted state could result from limited GTP hydrolysis. Furthermore, a partial depolymerization is observed when GTP is added to dynamin tubules, which could play a role in fission (Bashkirov et al., 2008; Pucadyil and Schmid, 2008). This depolymerization may come from the fact that the PH domain of dynamin in the most-constricted state is titled, probably breaking the bond between dynamin and the membrane (Sundborger et al., 2014). It is thus difficult to infer from in vitro data how GTP hydrolysis, constriction and disassembly are coupled.

Structural studies have neither solved whether fission occurs by active constriction or active depolymerization of the dynamin helix. In the presence of GTP or GTP-analogs, GTPase domains form dimers (called G-G), explaining the higher propensity of dynamin to self-assemble with GTP. However, the strongest GTPase domain dimerization occurs using GDP·AlF₄^‾, that mimics the transition state of hydrolysis (Chappie et al., 2010; Chappie et al., 2009). In this state, dynamin can form helices independently of the membrane. Moreover, the BSE position relative to the GTPase domain rotates by 90°, showing that GTP hydrolysis induces a conformational change compatible with constriction (Chappie et al., 2010; Chappie et al., 2011; Sundborger et al., 2014; Chappie et al., 2009). But as the structure of this transition state does not fit into the cryo-EM structure of the most constricted state of the dynamin helix (Sundborger et al., 2014), it is possible that the GDP·AlF₄^‾-bound state of dynamin induces stresses within the helix that lead to its disassembly.

The dynamics of the fission reaction has been studied extensively in cells; it requires 5–10 s and is highly stochastic (Merrifield et al., 2005), consistent with in vitro fission rates in optimal conditions (Morlot et al., 2012). The minimal number of dynamin monomers for fission reaction is between 26 and 28, suggesting that at least one complete helical turn of dynamin is required for fission (Cocucci et al., 2014; Grassart et al., 2014). However, when GTP is hydrolyzed relative to fission in vivo is not known as in cells measuring or following the activity of a specific enzyme remains a challenge. Thus, how dynamin’s GTP hydrolysis is coupled to constriction or to disassembly in vivo has not been addressed yet.

In order to follow the GTPase activity of dynamin in vivo, we ought to isolate a conformation-specific nanobody that preferentially binds one of the specific nucleotidic states of dynamin. To isolate a conformation specific nanobody against GTP-loaded dynamin-2, we screened a phage-display library of synthetic nanobodies (Moutel et al., 2016) against recombinant human dynamin-2 loaded with GMPPCP. Further characterization of 192 selected clones, allowed us to exclude false positives (see Materials and methods, Supplementary methods and Figure 1—figure supplement 1), and isolate 33% of positive clones. Based on their affinities in dot-blot assays (see Supplementary methods), we chose five dynamin-binding nanobodies and tested them in pull-down assays against various nucleotidic states of dynamin. Pull-downs were performed with dynamin assembled onto SUPported bilayers with Excess membrane Reservoir (SUPER) templates (Pucadyil and Schmid, 2008; Neumann et al., 2013) adapted for pull-down assays: magnetic silica beads were coated with lipid membranes, from which dynamin-coated membrane tubules can be formed, as dynamin binding is lost when membranes are dissolved with saponin (Figure 1—figure supplement 2). Beads are further pulled down with a magnet. All five nanobodies bound dynamin-2 efficiently regardless of its nucleotide load. As GTP analogs substantially increased the amount of dynamin pulled down (Figure 1—figure supplement 3), consistent with their ability to promote dynamin oligomerization (Carr and Hinshaw, 1997), GMPPCP-loaded dynamin-2 was used as a reference to ensure that the amount of dynamin pulled down was the same in every nucleotide condition.

We then tested our nanobodies against dynamin-1. We found one nanobody, that we named dynab, that preferentially binds to GDP·AlF₄^‾ loaded dynamin-1 (Figure 1A). Overall, dynab has a lower affinity for dynamin-1 than for dynamin-2 (see Figure 1A). Dynab does not bind dynamin one through its PRD domain (Figure 1—figure supplement 5A). Since the GDP·AlF₄^‾ -bound state of dynamin mimics the GTP hydrolysis transition state, dynab could potentially track the GTPase activity of dynamin by binding dynamin only during GTP hydrolysis. To test this in vitro, we first pulled-down dynab with fission defective mutants K44A and K142A. K44A has a reduced affinity for nucleotides, and thus a lower overall GTPase activity (Damke et al., 1994), while K142A has almost normal GTPase activity, while its helix cannot constrict (Marks et al., 2001). The results were similar to the wild-type (Figure 1—figure supplement 5B), which may indicate that even if the mutants have deficient GTPase activity, they can be locked into the same transition state when loaded with GDP·AlF₄^‾.

Figure 1 with 6 supplements see all

Download asset Open asset

Figure 1—source data 1

(panel D) Malachite Green Assay.

https://doi.org/10.7554/eLife.25197.009

Download elife-25197-fig1-data1-v2.docx

As results of the pull-down with GTPase defective mutants were not conclusive, we used a membrane sheets assay, in which lipids are dried onto a coverslip and rehydrated to form stacks of lipid bilayers (Roux et al., 2006; Itoh et al., 2005). The addition of Alexa-488-dynamin-1 to the membrane sheets resulted in the formation of dynamin tubules, which can be visualized by confocal microscopy (Figure 1B). Atto-565-dynab weakly bound to dynamin-1 tubules. In contrast, in presence of GDP·AlF₄^‾ we observed a strong binding of dynab to dynamin-1 tubules (Figure 1B). This confirmed that dynab binds preferentially to the transition state of dynamin-1, consistently with pull-down results. As GDP·AlF₄^‾ can induce formation of membrane-free dynamin oligomers, we further checked that dynab was co-localizing with dynamin 1-coated membrane tubules only in presence of GDP·AlF₄‾ (Figure 1—figure supplement 6). Using the same approach, we confirmed that dynab binds dynamin-2 tubules regardless of the nucleotide load (Figure 1B).

To test if dynab could detect dynamin-1 GTPase activity, we performed time-lapsed imaging of dynamin-1 tubules following the addition of a GTP solution. A sudden recruitment of dynab to the dynamin-1 tubules appeared after GTP addition (Figure 1C). In contrast, neither GMPPCP nor GDP additions led to binding of dynab (Figure 1C). We concluded that dynab binds to dynamin-1 tubules hydrolyzing GTP and is thus able to detect the GTPase activity of dynamin. Importantly, we further verified dynamin-1 and 2 GTPase activities were unchanged in presence of dynab (Quan and Robinson, 2005) (Figure 1D). Our results indicate that dynab has a stronger affinity for GDP·AlF₄^‾ loaded dynamin-1 than for other states of dynamin-1 and thus detect the transient hydrolytic state of dynamin-1.

We then wondered if dynab may be used as an intrabody to track GTPase activity of dynamin-1 during endocytosis in living fibroblastic cells. Dynamin and dynab were expressed as fluorescently-tagged proteins in fibroblastic cells imaged by Total Internal Reflection Fluorescence (TIRF) microscopy which allow for visualization of single clathrin pits formation. To generate fibroblastic cells expressing solely dynamin-1 or dynamin-2, we used a triple conditional knockout cell line for the three dynamin genes (TKO cells) (Park et al., 2013). After removing the three dynamin genes (Figure 1—figure supplement 4), we co-expressed dynamin-1-mCherry or dynamin-2-mCherry and dynab-EGFP. In those cells, dynab co-localized with both 74 ± 4% of dynamin-1 and 88 ± 12% dynamin-2 punctae (Figure 2B).

Figure 2

Download asset Open asset

We also used HeLa cells overexpressing dynamin-1 or 2 with dynab, expecting that overexpression could overcome the presence of endogenous dynamin-2. Dynab also co-localized with 80 ± 9% of the dynamin-1 and 97 ± 3% two punctae in HeLa overexpressing cells (Figure 2B). In HeLa cells, dynab colocalized with 76 ± 11% of clathrin-EGFP punctae (Figure 2C), showing that in these cells, dynab mostly follows the activity of dynamin oligomerized at clathrin-coated pits CCPs.

We then analyzed the dynamics of dynab/dynamin co-localization while being recruited at CCPs. By TIRF, dynamins show a Gaussian profile of intensity with time (Merrifield et al., 2005; Merrifield et al., 2002; Doyon et al., 2011; Taylor et al., 2011), which duration is 10–20 s. Dynab intensity mirrored the dynamin-2 intensity profile (panel 6, Figure 3A). This confirms that in vivo, dynab binds dynamin-2 independently of the nucleotide state. This also indicates that dynab can be used as an intrabody and that it interacts with fast kinetics with its target in living cells.

Figure 3 with 1 supplement see all

Download asset Open asset

Interestingly, dynab intensity profile relative to dynamin-1 displayed a variety of patterns. We observed three categories of patterns: (i) concomitant dynamin-dynab peaks - events where dynab fluorescence intensity grew along with dynamin-1 intensity and peaked simultaneously (panel 1, Figure 3A); (ii) late dynab peaks - events where dynab peaked after dynamin-1 (panel 2, Figure 3A); and (iii) early dynab peaks - events where dynab peaked before dynamin-1 (panel 3, Figure 3A). We also observed double dynab peaks (panels 4 and 5, Figure 3A) as well as double dynamin peaks which in most cases were accompanied by dynab peaks (panel 5, Figure 3A).

These various profiles could indicate that dynab also detects dynamin GTPase activity in vivo. To test this, we blocked the fission activity of dynamin by incubating TKO and HeLa cells in a hypertonic media for 1’(Morlot et al., 2012; Heuser and Anderson, 1989). In both TKO and HeLa cells, dynamin-1 was blocked at clathrin-coated pits with no co-staining for dynab (Figure 3B and Figure 3—figure supplement 1) suggesting that dynab could not detect non-active, but assembled dynamin-1 in both TKO and HeLa cells. On the contrary, in cells expressing dynamin-2 and dynab, the hypertonic shock resulted in a complete co-localization of dynab and dynamin-2, which agrees with a non-conformation dependent interaction of dynab with dynamin-2 (Figure 3B and Figure 3—figure supplement 1). We further tested the ability of dynab to bind fission defective dynamin-1 K44A (Damke et al., 1994) and K142A mutants (Marks et al., 2001). We co-transfected HeLa cells with dynamin-1 K44A or dynamin-1 K142A tagged with mCherry along with dynab. Dynab did not co-localize with dynamin-1 K44A while it co-localized with dynamin-1 K142A (Figure 3C). These results confirmed that dynab detects the GTPase active forms of dynamin-1, as K44A is GTPase defective (Damke et al., 1994) while K142A is constriction defective but GTPase active (Marks et al., 2001).

From these controls, we concluded that dynab could detect dynamin-1 GTPase activity in fibroblastic cells. From the different binding profiles of dynab to dynamin-1 described above, we concluded that dynamin’s GTPase activity in vivo is discontinuous and stochastic: it occurs in burst that are randomly distributed in time, instead of being continuous and proportional to the amount of dynamin present at the pit. We then analyzed statistically the relevance of these observations to better understand how dynamin GTP hydrolysis and polymerization are coupled.

To analyze the coupling of dynamin and dynab fluorescence profiles over time, we analyzed the time-lapse images obtained with TIRF in HeLa and TKO cells co-transfected with dynamin1- or dynamin-2 -mCherry and with dynab-EGFP. As a control, we used profiles obtained in cells transfected with both dynamin-1-EGFP and dynamin-1-mCherry as we expected them to be simultaneously recruited and disassembled from coated pits (Cocucci et al., 2014; Grassart et al., 2014; Merrifield et al., 2002; Taylor et al., 2011), called DYNctrl in the following. A software was developed to post-process all intensity profiles (see Materials and methods). Each of the profiles of either dynab or dynamin was fitted with 4 Gaussians. For each peak, we obtained 3 values of the Gaussian fit: the width, multiplied by four, is an estimate of the duration d of the event, the peak (maximum) value p, and the time t of the peak.

In TKO cells, the duration of dynamin-1 or 2 events was not affected by dynab, while in HeLa cells, dynab slightly reduced the duration of dynamin-1 events and had no effect on dynamin-2 (Figure 4A). Neither it did change the duration of clathrin-coated pits (Figure 4—figure supplement 1). We therefore concluded that the expression of dynab did not cause significant impairment of clathrin-mediated endocytosis.

Figure 4 with 3 supplements see all

Download asset Open asset

Figure 4—source data 1

(panel A) Comparison of duration of dynamin 1–2 events with and without dynab (expressed in seconds), and statistical report.

https://doi.org/10.7554/eLife.25197.017

Download elife-25197-fig4-data1-v2.docx

Figure 4—source data 2

(panel B) Comparison of duration of dynamin 1–2 events with dynab events in TKO cells (expressed in seconds), and statistical report.

https://doi.org/10.7554/eLife.25197.018

Download elife-25197-fig4-data2-v2.docx

Figure 4—source data 3

(panel B) Comparison of duration of dynamin 1–2 events with dynab events in Hela cells (expressed in seconds), and statistical report.

https://doi.org/10.7554/eLife.25197.019

Download elife-25197-fig4-data3-v2.docx

Figure 4—source data 4

(panel C) Cumulative probability of time difference between t(dynab)-t(dyn1-2) and DYNctrl in TKO cells, and statistical report.

https://doi.org/10.7554/eLife.25197.020

Download elife-25197-fig4-data4-v2.docx

Figure 4—source data 5

(panel C) Cumulative probability of time difference between t(dynab)-t(dyn1-2) and DYNctrl in HeLa cells, and statistical report.

https://doi.org/10.7554/eLife.25197.021

Download elife-25197-fig4-data5-v2.docx

Figure 4—source data 6

(panel C) Relative frequency Δt values; Δt distribution expressed in percentage of relative dynamin duration.

https://doi.org/10.7554/eLife.25197.022

Download elife-25197-fig4-data6-v2.docx

Figure 4—source data 7

(panel D and F) Maximum intensity values of dynab peaks vs dynamin peaks.

https://doi.org/10.7554/eLife.25197.023

Download elife-25197-fig4-data7-v2.docx

Figure 4—source data 8

Data and statistical analysis of CCPs persistence in presence of dynab in HeLa cells.

https://doi.org/10.7554/eLife.25197.024

Download elife-25197-fig4-data8-v2.xlsx

We thus concluded that dynab did not significantly interfered with the dynamics of endocytic pits. In TKO and HeLa, dynab average duration was 70–80% of dynamin-1’s average duration (Figure 4B). This shows that dynamin-1 is not permanently active when oligomerized at the neck of endocytic pits. As expected, dynab and dynamin-2 durations are equal in HeLa cells, while in TKO cells, a small but significant difference between dynamin-2 and dynab durations was observed (Figure 4B), which may indicate a slight conformation-preference of dynab for dynamin-2 in those cells.

We then analyzed how the peak of dynamin-1 GTPase activity (dynab’s peak time) correlated with the dynamin-1 peak in time. We compared the distribution of time differences Δt between dynab and dynamin-1 peaks to the ones of dynab and dynamin-2 and DYNctrl. In both TKO and HeLa cells transfected with dynamin-1, we observed that the dynab-dynamin-1 Δt distribution is more spread than dynab-dynamin-2 or DYNctrl (Figure 4C). The comparison of Δt distributions in TKO highlight that about 47% of the dynab peaks occur before dynamin-1’s peak (negative values of Δt), while for dynamin-2, less than 35% of the dynab peaks occurred before. For DYNctrl, the distribution of Δt is very narrow, and only 25% of the Δt are negative. For HeLa cells, 46% of events have a negative Δt for dyn1-dynab, 44% for dynamin-2-dynab and 33% for DYNctrl.

Since the duration of dynamin presence at CCPs is very variable, we calculated the time difference Δt between peaks of dynamin and the peaks dynab, and normalized it by the duration of the dynamin peak: we obtained Δt expressed in percentage of dynamin duration and binned the values in 5% clusters (Figure 4C). In TKO and HeLa cells, the distribution of Δt for dynamin-2-dynab resembles the distribution of DYNctrl, where no time difference is expected. In contrast, the Δt distribution between dynab and dynamin-1 revealed a more spread distribution, suggesting that the kinetics of dynamin-1 recruitment and release is not fully coupled with its hydrolysis cycle. Uncoupling is clearer considering basic statistics (Figure 4D): in TKO cells, if Δt values less than ±10% are taken as simultaneous events, the majority of dynab peaks happened before or after the dynamin peak (29% peaks before and 24% of peaks after), while 47% of the events are simultaneous. Note that these 47% are not the same population than the 47% of negative Δt values described above. For dynab-dynamin-2 and the DYNctrl, the majority of peaks (>60%) are simultaneous. In HeLa, the same statistics also show a complete uncoupling of dynab and dynamin-1 peak, with 33% of dynab peaks occurring before and 32% after dynamin-1 peak, while only 35% are simultaneous. For dynab-dynamin-2 and DYNctrl, the majority (>50%) of events are simultaneous.

Dynab also displays a double peak behavior (Figure 3A). In TKO cells, 8% of dynamin-1 peaks are coupled with two dynab peaks, while it represented only 5.3% and 1.6% of dynamin-2 and DYNctrl respectively. In HeLa cells, 5.8% of dynamin-1 events are coupled to double dynab peaks, 2.8% of dynamin-2 events and 2.5% of DYNctrl events. In most cases, the first peak occurs before the dynamin-1 peak, the second after (Figure 4—figure supplement 2). This supports the possibility that in double-peak events, the first burst of GTPase activity may happen too early, and that dynamin is not polymerized enough to perform its function (less than one turn), and that the second peak is the efficient activity peak that causes fission.

Considering the fact that TKO and HeLa cells gave very similar results, we concluded that overexpression of dynamin-1 in standard cell lines allows for the use of dynab without the necessity of removing endogenous dynamins. Our results indicate that the GTPase activity of dynamin occurs in bursts during polymerization of dynamin at the neck of endocytic pits, and that these bursts are not temporally correlated with the peak of dynamin presence at the pits.

We next tested if the intensity of the GTPase activity was correlated to the amount of dynamin polymerized around the neck. We found a strong correlation between the amount of dynamin-1 and dynab (Figure 4—figure supplement 3) present at endocytic pits. Although temporally uncoupled, it seemed that the GTPase activity is proportional to the amount of dynamin present at the pit. This supports a concerted hydrolysis of GTP by all dynamins at the neck of endocytic pits, rather than a fixed number of turns of the dynamin helix being active while other turns being inactive. In the control experiments (DYNctrl) where dynamin-1-EGFP and dynamin-1-mCherry are co-expressed, low correlation was found, whereas a strong correlation is found with dynamin-2, as expected from the high affinity of dynab for dynamin-2.

In order to check if dynab could interfere with clathrin-mediated cargo uptake, we assayed transferrin and EGF internalization in HeLa cells while overexpressing dynamin-1 or −2 in presence or absence of dynab. Although overexpression of either dynamin-1 and −2 slightly lowers transferrin and EGF endocytosis, we observed that the expression of dynab alone or together with dynamin-2 dramatically impairs transferrin internalization (Figure 5), while it has a minor effect on clathrin-mediated EGF internalization (low concentration of EGF are internalized mostly by CME, (Figure 6) (Sigismund et al., 2008). When dynab is co-expressed with dynamin-1, transferrin internalization is similar to internalization of transferrin in cells overexpressing dynamin-1 or dynamin-2 alone (Figure 5). Thus, dynab does not interfere with dynamin-1 mediated endocytosis. We however wondered how dynab could specifically interfere with the transferrin uptake mediated by dynamin-2, and not dynamin-1. As discussed above, the overall higher affinity of dynab for dynamin-2 may cause dynab to bind stronger to dynamin, potentially interfering sterically with the required interaction of the transferrin receptor (TfR) with the clathrin coat.

Figure 5

Download asset Open asset

Figure 5—source data 1

(panel C) Data and statistical analysis of transferrin internalization in HeLa cells.

https://doi.org/10.7554/eLife.25197.026

Download elife-25197-fig5-data1-v2.docx

Figure 6

Download asset Open asset

Figure 6—source data 1

(panel C) Data and statistical analysis of EGF internalization in HeLa cells.

https://doi.org/10.7554/eLife.25197.028

Download elife-25197-fig6-data1-v2.docx

We thus looked the localization of the TfR in cells overexpressing dynab. In these cells, whether dynamin two was co-expressed with dynab or not, dynab expression specifically blocked the TfR at the membrane, as no endosomal pool of the TfR was visible (Figure 5). The endosomal pool was still visible when dynab is co-expressed with dynamin-1. Moreover, EGF receptor staining was not altered by dynab or dynamin-2 +dynab expression (Figure 6).

In this study, we visualized dynamin’s GTPase activity in cells with the conformation specific nanobody dynab which interacts more strongly with the GDP·AlF₄^‾ bound form of dynamin-1. To our knowledge, dynab is a unique conformation-specific nanobody in that it binds preferentially the intermediate hydrolytic state of an enzyme, allowing to follow in real time the enzymatic activity of dynamin in vivo. By using dynab in live cell imaging experiments, we showed that dynamin’s GTP hydrolysis occurs in bursts and is not time-correlated with dynamin assembly.

There is a large consensus that dynamin-mediated membrane fission reaction requires the energy of GTP hydrolysis (Antonny et al., 2016). Two models of dynamin’s mechanism differ by how the GTP hydrolytic energy is spent (Antonny et al., 2016). In the constriction model, this energy is proposed to be used to constrict the polymer, in which GTPase domains could act as myosins, sliding adjacent helical turns, tightening the dynamin helix. In the depolymerization model, dynamin is proposed to assemble into an already tightened helix. Fission occurs when depolymerization is triggered by GTP hydrolysis, which energy would be used for breaking non-covalent bonds between monomers. Our finding that GTPase activity of dynamin is not correlated with dynamin polymer disassembly challenges the second model.

A surprising finding is that dynamin GTPase activity occurs in burst, which suggest that it is strongly regulated. This is consistent with the requirement of a coordinated GTPase hydrolysis for efficient fission (Liu et al., 2013). However, assembly triggers immediate increase in dynamin GTPase activity in vitro (Marks et al., 2001; Warnock et al., 1996), which should lead to a perfect co-localization of dynab with dynamin-1 over time. Our finding of bursts of GTP hydrolysis uncoupled with the assembly/disassembly of dynamin suggests that in cells, a molecular trigger is required to allow for GTPase activation. It is even more striking that in approx. 30% of the cases, the burst of GTPase activity occurs while dynamin starts depolymerizing, suggesting that the molecular regulation of the GTPase activity is fully uncoupled from the assembly process.

Although the nature of this trigger is unknown, we can envisage several hypotheses. Various BAR (Bin-Amphiphysin-Rvs) proteins have been proposed to modulate dynamin’s GTPase and fission activities (Farsad et al., 2001; Yoshida et al., 2004; Neumann and Schmid, 2013). One of these BAR proteins, syndapin, sees its interaction with dynamin blocked when dynamin is phosphorylated (Anggono et al., 2006). Furthermore, bursts of dynamin dephosphorylation occur at the synapse during active phases, suggesting that the activation of endocytosis could be triggered by dynamin-syndapin interactions (Armbruster et al., 2013). Finally, another trigger may come from the specific activation of dynamin GTPase activity provided by fueling of GTP by the nucleotide diphosphate kinase NM23 that binds to dynamin (Boissan et al., 2014). How the NM23-Dynamin interaction is regulated is still unknown, but these three triggers may all participate in the control of dynamin GTPase activity.

1. 1. Anggono V
   2. Smillie KJ
   3. Graham ME
   4. Valova VA
   5. Cousin MA
   6. Robinson PJ

   (2006) Syndapin I is the phosphorylation-regulated dynamin I partner in synaptic vesicle endocytosis

   Nature Neuroscience 9:752–760.

   https://doi.org/10.1038/nn1695

   • PubMed
   • Google Scholar
2. 1. Antonny B
   2. Burd C
   3. De Camilli P
   4. Chen E
   5. Daumke O
   6. Faelber K
   7. Ford M
   8. Frolov VA
   9. Frost A
   10. Hinshaw JE
   11. Kirchhausen T
   12. Kozlov MM
   13. Lenz M
   14. Low HH
   15. McMahon H
   16. Merrifield C
   17. Pollard TD
   18. Robinson PJ
   19. Roux A
   20. Schmid S

   (2016) Membrane fission by dynamin: what we know and what we need to know

   The EMBO Journal 35:2270–2284.

   https://doi.org/10.15252/embj.201694613

   • PubMed
   • Google Scholar
3. 1. Armbruster M
   2. Messa M
   3. Ferguson SM
   4. De Camilli P
   5. Ryan TA

   (2013) Dynamin phosphorylation controls optimization of endocytosis for brief action potential bursts

   eLife 2:e00845.

   https://doi.org/10.7554/eLife.00845

   • PubMed
   • Google Scholar
4. 1. Bashkirov PV
   2. Akimov SA
   3. Evseev AI
   4. Schmid SL
   5. Zimmerberg J
   6. Frolov VA

   (2008) GTPase cycle of dynamin is coupled to membrane squeeze and release, leading to spontaneous fission

   Cell 135:1276–1286.

   https://doi.org/10.1016/j.cell.2008.11.028

   • PubMed
   • Google Scholar
5. 1. Boissan M
   2. Montagnac G
   3. Shen Q
   4. Griparic L
   5. Guitton J
   6. Romao M
   7. Sauvonnet N
   8. Lagache T
   9. Lascu I
   10. Raposo G
   11. Desbourdes C
   12. Schlattner U
   13. Lacombe ML
   14. Polo S
   15. van der Bliek AM
   16. Roux A
   17. Chavrier P

   (2014) Membrane trafficking. Nucleoside diphosphate kinases fuel dynamin superfamily proteins with GTP for membrane remodeling

   Science 344:1510–1515.

   https://doi.org/10.1126/science.1253768

   • PubMed
   • Google Scholar
6. 1. Carr JF
   2. Hinshaw JE

   (1997) Dynamin assembles into spirals under physiological salt conditions upon the addition of GDP and gamma-phosphate analogues

   Journal of Biological Chemistry 272:28030–28035.

   https://doi.org/10.1074/jbc.272.44.28030

   • PubMed
   • Google Scholar
7. 1. Chappie JS
   2. Acharya S
   3. Liu YW
   4. Leonard M
   5. Pucadyil TJ
   6. Schmid SL

   (2009) An intramolecular signaling element that modulates dynamin function in vitro and in vivo

   Molecular Biology of the Cell 20:3561–3571.

   https://doi.org/10.1091/mbc.E09-04-0318

   • PubMed
   • Google Scholar
8. 1. Chappie JS
   2. Acharya S
   3. Leonard M
   4. Schmid SL
   5. Dyda F

   (2010) G domain dimerization controls dynamin's assembly-stimulated GTPase activity

   Nature 465:435–440.

   https://doi.org/10.1038/nature09032

   • PubMed
   • Google Scholar
9. 1. Chappie JS
   2. Mears JA
   3. Fang S
   4. Leonard M
   5. Schmid SL
   6. Milligan RA
   7. Hinshaw JE
   8. Dyda F

   (2011) A pseudoatomic model of the dynamin polymer identifies a hydrolysis-dependent powerstroke

   Cell 147:209–222.

   https://doi.org/10.1016/j.cell.2011.09.003

   • PubMed
   • Google Scholar
10. 1. Cocucci E
    2. Gaudin R
    3. Kirchhausen T

    (2014) Dynamin recruitment and membrane scission at the neck of a clathrin-coated pit

    Molecular Biology of the Cell 25:3595–3609.

    https://doi.org/10.1091/mbc.E14-07-1240

    • PubMed
    • Google Scholar
11. 1. Damke H
    2. Baba T
    3. Warnock DE
    4. Schmid SL

    (1994) Induction of mutant dynamin specifically blocks endocytic coated vesicle formation

    The Journal of Cell Biology 127:915–934.

    https://doi.org/10.1083/jcb.127.4.915

    • PubMed
    • Google Scholar
12. 1. Dimitrov A
    2. Quesnoit M
    3. Moutel S
    4. Cantaloube I
    5. Poüs C
    6. Perez F

    (2008) Detection of GTP-tubulin conformation in vivo reveals a role for GTP remnants in microtubule rescues

    Science 322:1353–1356.

    https://doi.org/10.1126/science.1165401

    • PubMed
    • Google Scholar
13. 1. Doyon JB
    2. Zeitler B
    3. Cheng J
    4. Cheng AT
    5. Cherone JM
    6. Santiago Y
    7. Lee AH
    8. Vo TD
    9. Doyon Y
    10. Miller JC
    11. Paschon DE
    12. Zhang L
    13. Rebar EJ
    14. Gregory PD
    15. Urnov FD
    16. Drubin DG

    (2011) Rapid and efficient clathrin-mediated endocytosis revealed in genome-edited mammalian cells

    Nature Cell Biology 13:331–337.

    https://doi.org/10.1038/ncb2175

    • PubMed
    • Google Scholar
14. 1. Faelber K
    2. Posor Y
    3. Gao S
    4. Held M
    5. Roske Y
    6. Schulze D
    7. Haucke V
    8. Noé F
    9. Daumke O

    (2011) Crystal structure of nucleotide-free dynamin

    Nature 477:556–560.

    https://doi.org/10.1038/nature10369

    • PubMed
    • Google Scholar
15. 1. Farsad K
    2. Ringstad N
    3. Takei K
    4. Floyd SR
    5. Rose K
    6. De Camilli P

    (2001) Generation of high curvature membranes mediated by direct endophilin bilayer interactions

    The Journal of Cell Biology 155:193–200.

    https://doi.org/10.1083/jcb.200107075

    • PubMed
    • Google Scholar
16. 1. Ferguson SM
    2. De Camilli P

    (2012) Dynamin, a membrane-remodelling GTPase

    Nature Reviews Molecular Cell Biology 13:75–88.

    https://doi.org/10.1038/nrm3266

    • PubMed
    • Google Scholar
17. 1. Ford MG
    2. Jenni S
    3. Nunnari J

    (2011) The crystal structure of dynamin

    Nature 477:561–566.

    https://doi.org/10.1038/nature10441

    • PubMed
    • Google Scholar
18. 1. Grassart A
    2. Cheng AT
    3. Hong SH
    4. Zhang F
    5. Zenzer N
    6. Feng Y
    7. Briner DM
    8. Davis GD
    9. Malkov D
    10. Drubin DG

    (2014) Actin and dynamin2 dynamics and interplay during clathrin-mediated endocytosis

    The Journal of Cell Biology 205:721–735.

    https://doi.org/10.1083/jcb.201403041

    • PubMed
    • Google Scholar
19. 1. Heuser JE
    2. Anderson RG

    (1989) Hypertonic media inhibit receptor-mediated endocytosis by blocking clathrin-coated pit formation

    The Journal of Cell Biology 108:389–400.

    https://doi.org/10.1083/jcb.108.2.389

    • PubMed
    • Google Scholar
20. 1. Itoh T
    2. Erdmann KS
    3. Roux A
    4. Habermann B
    5. Werner H
    6. De Camilli P

    (2005) Dynamin and the actin cytoskeleton cooperatively regulate plasma membrane invagination by BAR and F-BAR proteins

    Developmental Cell 9:791–804.

    https://doi.org/10.1016/j.devcel.2005.11.005

    • PubMed
    • Google Scholar
21. 1. Jaqaman K
    2. Loerke D
    3. Mettlen M
    4. Kuwata H
    5. Grinstein S
    6. Schmid SL
    7. Danuser G

    (2008) Robust single-particle tracking in live-cell time-lapse sequences

    Nature Methods 5:695–702.

    https://doi.org/10.1038/nmeth.1237

    • PubMed
    • Google Scholar
22. 1. Liu YW
    2. Mattila JP
    3. Schmid SL

    (2013) Dynamin-catalyzed membrane fission requires coordinated GTP hydrolysis

    PLoS One 8:e55691.

    https://doi.org/10.1371/journal.pone.0055691

    • PubMed
    • Google Scholar
23. 1. Marks B
    2. Stowell MH
    3. Vallis Y
    4. Mills IG
    5. Gibson A
    6. Hopkins CR
    7. McMahon HT

    (2001) GTPase activity of dynamin and resulting conformation change are essential for endocytosis

    Nature 410:231–235.

    https://doi.org/10.1038/35065645

    • PubMed
    • Google Scholar
24. 1. Merrifield CJ
    2. Feldman ME
    3. Wan L
    4. Almers W

    (2002) Imaging actin and dynamin recruitment during invagination of single clathrin-coated pits

    Nature Cell Biology 4:691–698.

    https://doi.org/10.1038/ncb837

    • PubMed
    • Google Scholar
25. 1. Merrifield CJ
    2. Perrais D
    3. Zenisek D

    (2005) Coupling between clathrin-coated-pit invagination, cortactin recruitment, and membrane scission observed in live cells

    Cell 121:593–606.

    https://doi.org/10.1016/j.cell.2005.03.015

    • PubMed
    • Google Scholar
26. 1. Morlot S
    2. Galli V
    3. Klein M
    4. Chiaruttini N
    5. Manzi J
    6. Humbert F
    7. Dinis L
    8. Lenz M
    9. Cappello G
    10. Roux A

    (2012) Membrane shape at the edge of the dynamin helix sets location and duration of the fission reaction

    Cell 151:619–629.

    https://doi.org/10.1016/j.cell.2012.09.017

    • PubMed
    • Google Scholar
27. 1. Moutel S
    2. Bery N
    3. Bernard V
    4. Keller L
    5. Lemesre E
    6. de Marco A
    7. Ligat L
    8. Rain JC
    9. Favre G
    10. Olichon A
    11. Perez F

    (2016) NaLi-H1: A universal synthetic library of humanized nanobodies providing highly functional antibodies and intrabodies

    eLife 5:e16228.

    https://doi.org/10.7554/eLife.16228

    • PubMed
    • Google Scholar
28. 1. Neumann S
    2. Pucadyil TJ
    3. Schmid SL

    (2013) Analyzing membrane remodeling and fission using supported bilayers with excess membrane reservoir

    Nature Protocols 8:213–222.

    https://doi.org/10.1038/nprot.2012.152

    • PubMed
    • Google Scholar
29. 1. Neumann S
    2. Schmid SL

    (2013) Dual role of BAR domain-containing proteins in regulating vesicle release catalyzed by the GTPase, dynamin-2

    Journal of Biological Chemistry 288:25119–25128.

    https://doi.org/10.1074/jbc.M113.490474

    • PubMed
    • Google Scholar
30. 1. Nizak C
    2. Monier S
    3. del Nery E
    4. Moutel S
    5. Goud B
    6. Perez F

    (2003) Recombinant antibodies to the small GTPase Rab6 as conformation sensors

    Science 300:984–987.

    https://doi.org/10.1126/science.1083911

    • PubMed
    • Google Scholar
31. 1. Park RJ
    2. Shen H
    3. Liu L
    4. Liu X
    5. Ferguson SM
    6. De Camilli P

    (2013) Dynamin triple knockout cells reveal off target effects of commonly used dynamin inhibitors

    Journal of Cell Science 126:5305–5312.

    https://doi.org/10.1242/jcs.138578

    • PubMed
    • Google Scholar
32. 1. Pelkmans L
    2. Fava E
    3. Grabner H
    4. Hannus M
    5. Habermann B
    6. Krausz E
    7. Zerial M

    (2005) Genome-wide analysis of human kinases in clathrin- and caveolae/raft-mediated endocytosis

    Nature 436:78–86.

    https://doi.org/10.1038/nature03571

    • PubMed
    • Google Scholar
33. 1. Pucadyil TJ
    2. Schmid SL

    (2008) Real-time visualization of dynamin-catalyzed membrane fission and vesicle release

    Cell 135:1263–1275.

    https://doi.org/10.1016/j.cell.2008.11.020

    • PubMed
    • Google Scholar
34. 1. Quan A
    2. Robinson PJ

    (2005) Rapid purification of native dynamin I and colorimetric GTPase assay

    Methods in enzymology 404:556–569.

    https://doi.org/10.1016/S0076-6879(05)04049-8

    • PubMed
    • Google Scholar
35. 1. Raimondi A
    2. Ferguson SM
    3. Lou X
    4. Armbruster M
    5. Paradise S
    6. Giovedi S
    7. Messa M
    8. Kono N
    9. Takasaki J
    10. Cappello V
    11. O'Toole E
    12. Ryan TA
    13. De Camilli P

    (2011) Overlapping role of dynamin isoforms in synaptic vesicle endocytosis

    Neuron 70:1100–1114.

    https://doi.org/10.1016/j.neuron.2011.04.031

    • PubMed
    • Google Scholar
36. 1. Reubold TF
    2. Faelber K
    3. Plattner N
    4. Posor Y
    5. Ketel K
    6. Curth U
    7. Schlegel J
    8. Anand R
    9. Manstein DJ
    10. Noé F
    11. Haucke V
    12. Daumke O
    13. Eschenburg S

    (2015) Crystal structure of the dynamin tetramer

    Nature 525:404–408.

    https://doi.org/10.1038/nature14880

    • PubMed
    • Google Scholar
37. 1. Roux A
    2. Uyhazi K
    3. Frost A
    4. De Camilli P

    (2006) GTP-dependent twisting of dynamin implicates constriction and tension in membrane fission

    Nature 441:528–531.

    https://doi.org/10.1038/nature04718

    • PubMed
    • Google Scholar
38. 1. Salim K
    2. Bottomley MJ
    3. Querfurth E
    4. Zvelebil MJ
    5. Gout I
    6. Scaife R
    7. Margolis RL
    8. Gigg R
    9. Smith CI
    10. Driscoll PC
    11. Waterfield MD
    12. Panayotou G

    (1996)

    Distinct specificity in the recognition of phosphoinositides by the pleckstrin homology domains of dynamin and Bruton's tyrosine kinase

    The EMBO Journal 15:6241–6250.

    • PubMed
    • Google Scholar
39. 1. Schmid SL
    2. Frolov VA

    (2011) Dynamin: functional design of a membrane fission catalyst

    Annual Review of Cell and Developmental Biology 27:79–105.

    https://doi.org/10.1146/annurev-cellbio-100109-104016

    • PubMed
    • Google Scholar
40. 1. Sigismund S
    2. Argenzio E
    3. Tosoni D
    4. Cavallaro E
    5. Polo S
    6. Di Fiore PP

    (2008) Clathrin-mediated internalization is essential for sustained EGFR signaling but dispensable for degradation

    Developmental Cell 15:209–219.

    https://doi.org/10.1016/j.devcel.2008.06.012

    • PubMed
    • Google Scholar
41. 1. Sundborger AC
    2. Fang S
    3. Heymann JA
    4. Ray P
    5. Chappie JS
    6. Hinshaw JE

    (2014) A dynamin mutant defines a superconstricted prefission state

    Cell Reports 8:734–742.

    https://doi.org/10.1016/j.celrep.2014.06.054

    • PubMed
    • Google Scholar
42. 1. Taylor MJ
    2. Perrais D
    3. Merrifield CJ

    (2011) A high precision survey of the molecular dynamics of mammalian clathrin-mediated endocytosis

    PLoS Biology 9:e1000604.

    https://doi.org/10.1371/journal.pbio.1000604

    • PubMed
    • Google Scholar
43. 1. Vaid KS
    2. Guttman JA
    3. Babyak N
    4. Deng W
    5. McNiven MA
    6. Mochizuki N
    7. Finlay BB
    8. Vogl AW

    (2007) The role of dynamin 3 in the testis

    Journal of Cellular Physiology 210:644–654.

    https://doi.org/10.1002/jcp.20855

    • PubMed
    • Google Scholar
44. 1. Warnock DE
    2. Hinshaw JE
    3. Schmid SL

    (1996) Dynamin self-assembly stimulates its GTPase activity

    Journal of Biological Chemistry 271:22310–22314.

    https://doi.org/10.1074/jbc.271.37.22310

    • PubMed
    • Google Scholar
45. 1. Yoshida Y
    2. Kinuta M
    3. Abe T
    4. Liang S
    5. Araki K
    6. Cremona O
    7. Di Paolo G
    8. Moriyama Y
    9. Yasuda T
    10. De Camilli P
    11. Takei K

    (2004) The stimulatory action of amphiphysin on dynamin function is dependent on lipid bilayer curvature

    The EMBO Journal 23:3483–3491.

    https://doi.org/10.1038/sj.emboj.7600355

    • PubMed
    • Google Scholar

Author details

1. Valentina Galli

   Department of Biochemistry, University of Geneva, Geneva, Switzerland

   Contribution

   Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing—original draft, Writing—review and editing

   Competing interests

   No competing interests declared

2. Rafael Sebastian

   Department of Computer Sciences, Universidad de Valencia, Valencia, Spain

   Contribution

   Software, Writing—original draft

   Competing interests

   No competing interests declared

3. Sandrine Moutel

   1. Institut Curie, PSL Research University, Paris, France
   2. Translational Department, Institut Curie, Paris, France

   Contribution

   Resources, Validation, Investigation, Methodology

   Competing interests

   No competing interests declared

4. Jason Ecard

   Institut Curie, PSL Research University, Paris, France

   Contribution

   Formal analysis, Investigation, Methodology

   Competing interests

   No competing interests declared

5. Franck Perez

   Institut Curie, PSL Research University, Paris, France

   Contribution

   Conceptualization, Funding acquisition, Methodology, Writing—original draft, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-9129-9401

6. Aurélien Roux

   1. Department of Biochemistry, University of Geneva, Geneva, Switzerland
   2. Swiss National Centre for Competence in Research Programme Chemical Biology, Geneva, Switzerland

   Contribution

   Conceptualization, Supervision, Funding acquisition, Writing—original draft, Project administration, Writing—review and editing

   For correspondence

   aurelien.roux@unige.ch

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-6088-0711

• 2,269

  views

• 410

  downloads

• 18

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.