Microtubules coordinate protrusion-retraction dynamics in migrating dendritic cells

Institute of Science and Technology (IST) Austria, 3400 Klosterneuburg, Austria. Biomedical Center (BMC), Institute of Cardiovascular Physiology and Pathophysiology, Ludwig Maximilian University (LMU) Munich, 81377 Munich, Germany. Department of Immunology, State Research Institute Centre for Innovative Medicine, LT-08409 Vilnius, Lithuania. Institute of Biochemistry and Cell Biology, Otto von Guericke University, 39120 Magdeburg, Germany. Department of Pharmacy, Ludwig Maximilian University (LMU) Munich, 81377 Munich, Germany. Department of Chemistry, New York University, New York, NY 10003, USA. Department of Infectious Diseases, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA. Life and Medical Science (LIMES) Institute, University of Bonn, 53113 Bonn, Germany.


54
Here, we investigate MT-mediated processes that govern DC shape and motility. We 55 describe a mechanism by which local modulation of MT dynamics controls both, de-56 adhesion of cell attachment sites and retraction of entangled protrusions, thereby ensuring 57 that the advancing cell body maintains its morphological coherence.

58
To evaluate whether MTs are required for DC migration in physiological environments, we 61 employed mouse ear explants (Supplementary figure 1A). In this system, endogenous DCs 62 undergo maturation, up-regulate the chemokine-receptor 7 (CCR7) and subsequently 63 migrate along an interstitial gradient of the CCR7 ligand CCL21 towards afferent lymphatic 64 vessels (Ohl et al., 2004;Weber et al., 2013). We treated split ears with the MT-destabilizing The rounded cell morphologies and directional oscillations of Nocodazole-treated cells 96 prompted us to measure parameters of cellular contractility. We found that Nocodazole 97 treatment led to increased activation of RhoA ( Figure 2A) and phosphorylation of MLC 98 ( Figure 2B). This effect was reversed when Rho-associated protein kinase (ROCK) activity 99 was pharmacologically blocked using Y27632, suggesting that MTs control cell polarity by 100 regulating contractility.

101
Therefore, we tested whether ROCK inhibition on Nocodazole-treated cells would rescue 102 Nocodazole-induced migratory defects. We again employed the previously used succession 103 of reductionist experimental setups, which allowed us to separate the impact on locomotion 104 versus polarity. In 3D collagen gels, double-treatment with Nocodazole and Y27632 did not 105 rescue the migratory defect and cells slowed down even more than upon Nocodazole-only

146
To directly test for a potential causal link between MT depolymerization and local activation 147 of the contractile module, we devised a photo-pharmacological approach to depolymerize 148 MTs in migratory cells with spatiotemporal control. We used Photostatin-1 (PST-1), a 149 reversibly photo-switchable analog of combretastatin A-4, one of the most prominent MT 150 inhibitors. This compound can be functionally toggled between the active and inactive state 151 by blue and red light, respectively (Borowiak et al., 2015). To validate the approach, we 152 locally activated the drug and simultaneously visualized MT dynamics using EB3-mCherry.

153
We found that local photo-activation triggered an almost instantaneous disappearance of the 154 EB3 signal in the presence but not in the absence of Photostatin ( Figure 3F), indicating 155 immediate stalling of MT polymerization. Local photoactivation in protruding areas of the cell 156 consistently led to the collapse of the respective protrusion and subsequent re-polarization One potential link between MT depolymerization and cell contractility is the RhoA specific 165 GEF Lfc (the murine homolog of GEF-H1). Lfc is inactive when bound to MTs, whereas MT 166 depolymerization triggers its release and subsequent activation of the contractile module via 167 RhoA and its effectors ROCK and MLC kinase in other cell types (Chang et al., 2008; 168 Graessl et al., 2017;Krendel et al., 2002;Ren, 1998).

227
Inhibition of ROCK led to a further drop in migratory speed in both control and lfc -/cells,

228
arguing for additional -Lfc-independent -modes of ROCK activation. When we measured 229 cell length in 3D collagen gels ( Figure 6C, D) and under agarose ( Figure 6E), migrating lfc -/-230 DCs were significantly elongated compared to control cells, indicating retraction defects.

231
In cells that employ an amoeboid mode of migration, trailing edge retraction is essential in 232 two non-exclusive ways: i) under conditions where cells adhere to surfaces, retraction forces are required to disassemble integrin adhesion sites and ii) in complex 3D environments the 234 cell has to eventually retract all but one of its exploratory protrusions in order to not get 235 stalled by entanglement ((Lämmermann et al., 2009), Renkawitz et al., under revision).

236
We first tested the role of adhesion-resolution in under agarose assays, where, depending 237 on the surface conditions, DCs can flexibly shift between adhesion-dependent and 238 adhesion-independent locomotion (Renkawitz et al., 2009). Under adhesive conditions lfc -/-239 DCs were elongated compared to wild-type cells ( Figure 6E) and this elongation was lost 240 when the migratory substrate at the bottom was passivated with polyethylene glycol (PEG) 241 ( Figure 6F). When cells on adhesive surfaces were treated with Nocodazole, wild-type cells 242 shortened, as expected due to hypercontractility. Notably, lfc -/-DCs elongated even more 243 upon treatment with Nocodazole (Supplementary movie 9), indicating that elimination of Lfc-244 mediated hypercontractility unmasked additional modes of MT-mediated length control.

245
Elongation of lfc -/cells by Nocodazole was also largely absent on PEG-coated surfaces.

246
Importantly, not only morphological, but also migratory parameters were restored on 247 passivated surfaces (Figure 6 G, H). Together, these data demonstrate that whenever DCs 248 migrate in an adhesion-mediated manner, MTs control de-adhesion and this is partially 249 mediated via Lfc and myosin.

252
In 3D collagen gels adhesion plays a minor role (Friedl et al., 2012;Lämmermann et al., 253 2008;van Helvert et al., 2018). Therefore, elongation of lfc -/-DCs suggested that cells might 254 entangle rather than fail to de-adhere. To directly address this option, we used a microfluidic 255 setup, in which DCs migrate in a straight channel towards a junction where the channel 256 splits into four paths. In this setup DCs initially insert protrusions into all four channels before

271
To test for possible effects of Lfc beyond adhesion resolution and protrusion coordination,

272
we placed DCs in straight 1D channels, where neither trailing edge adhesion nor 273 entanglement are rate-limiting. Here, the hypercontractility-caused oscillations triggered by 274 Nocodazole ( Figure 1H and 7F) were substantially rescued by knockout of lfc ( Figure 7F-H).

275
Together, our data indicate that the major role of MT dynamics in migrating DCs is to specify 276 sites of protrusion-retraction and that this is partially regulated by the RhoA GEF Lfc and 277 actomyosin contraction.

279
Here we report that MT depolymerization in peripheral regions of migrating DCs locally 282 triggers actomyosin-mediated retraction via the RhoA GEF Lfc. Thereby MTs coordinate 283 protrusion-retraction dynamics and prevent that the cell gets too long or arborized.

284
How different cell types maintain their typical shape and how cell types with a dynamic 285 shape prevent losing physical coherence is poorly understood. This issue becomes

301
Based on our findings we propose a third model of shape control, in which MTs take the role 302 of the shape-sensor that signals to actin dynamics. This pathway might be particularly 303 relevant for leukocytes, as they do not develop stress fibers due to low adhesive forces and

306
Despite being therapeutically targeted, the role of MTs in leukocytes is poorly studied. In 307 neutrophil granulocytes and T cells, it was shown that pharmacological MT depolymerization 308 leads to enhanced cellular polarization, owing to a hypercontractility-induced symmetry 309 break that triggers locomotion but at the same time impairs directional persistence and 310 chemotactic prowess (Redd et al., 2006;Takesono et al., 2010;Xu et al., 2005;Yoo et al., drugs like Colchicine in the treatment of neutrophilic hyperinflammation, excessive 313 hypercontractility overwrites any morphodynamic subtleties and leaves the question of if is an important mediator between MT dynamics and actomyosin driven retraction.

317
Importantly, we show that DCs lacking both Lfc and MTs had even more severe cell shape 318 defects than the ones lacking Lfc only. This demonstrates that Lfc and myosin II are not the 319 only pathways and that MT depolymerization induces cell retraction via additional modes 320 that remain to be identified.

321
Although it is likely that multiple feedbacks signal between actin and MTs, we show that 322 there is a strong causal link between local MT catastrophes and cellular retraction, with MTs 323 acting upstream. This raises the key question how MT stability is locally regulated in DCs.

MTOC position
Back Front Side